At some point, I was going to get around to commenting on Tony Andragna's unjustified skepticism over our technical ability to defend ourselves against missiles, but Will Vehrs beat me to it in the same archive.

Many look at the test failures (and ignore the successes) so far and conclude that it's Just Too Hard.

That's extremely myopic. If we'd had the same attitude early in the missile program in the late fifties, we would have neither missiles nor a space program--launch failures were a regular and discouraging occurrence (remember post-Sputnik panic from the book, "The Right Stuff"? "Our rockets always blow up").

Most technical arguments against missile defense (by people like Dick Garwin, Kosta Tsipis, et al) are of the form:

Here is how I think a missile defense would work (i.e., set up strawman).

I'm really smart, and if I can't think up ways around these problems, neither can anyone else (thereby knocking the straw out of the man).

While there have been some sincere, and even good arguments against missile defense, most are of the disingenuous variety described above. Will's right--if we can land a man on the moon, we can (eventually) defend ourselves against missiles. We can quickly come up with a defense against North Korean missiles. It would take a little longer to come up with a defense against Chinese, or even Russian missiles, but we can do it if we need to. The Soviets knew this, which is one of the primary reasons that they threw in the towel. Ultimately, defense is favored in economic terms for two reasons. The first is that for any intelligently-designed system, the kill vehicle is cheaper than the offensive warhead, on the margin. The second is that, even if this is not the case, and the marginal costs of defense are greater than the marginal costs of offense, we're a lot richer than our adversaries, and will remain so for some time, so we can afford to outspend them--it's still cheaper for us in terms of percentage GDP. Again, the Soviets recognized this.

It should also be noted (as Don Rumsfeld did the other day) that while there are game-theoretical arguments to be made for both cases (that defense will encourage an offensive arms race, and that defense will suppress one), the empirical evidence is in. For the thirty years that we had the ABM treaty, missile inventories were growing like mushrooms after a spring rain, unchecked by ABM treaties or (on the Soviet side) even by arms-control treaties. But in the past few years, and particularly in the past year, despite, or more likely because of, all the talk about scrapping the ABM treaty, we are reducing inventories.

But whether or not it will defend against a Russian onslaught is not a relevant issue to current decisions. Regardless of one's view of its morality (mine is dim), an argument can be made that MAD is stable in a bi-polar world. Such an argument falls apart in a multi-polar world, and it just becomes a matter of time until some dictator with bin-Laden ethics and intelligence (re: intelligence--that's not a compliment) flunks Game Theory 101 and decides to lob something at us. In such an event, just as with the gun-control debate in general, as an engineer, I'll trust hardware over paper every time.

And Will, the actual expression, post Apollo, among space policy enthusiasts is "If we can put a man on the Moon, why can't we put a man on the Moon?" Realistically, right now it would take us longer to put a man on the Moon than it did in 1960 (at least as a government effort). It's not because we don't have the technology...

My take on missile defense is different: I am not opposed to building one if it can be shown to be relatively effective and reasonably priced, but it must not consume resources out of proportion to its benefit (principally, the ability to handle lunatics and accidents). There's no reason to downplay the feasibility of missile defense, just its overall applicability.

I am skeptical that a determined, rational attacker will choose to employ missiles knowing that we have a defense system, when there are so many non-airborne ways to deliver a weapon of mass destruction. To paraphrase an old joke, "we can whip them with one hand tied behind our backs, but they won't fight that way."

(Offtopic)

Of course we can put a man on the Moon again. The technology would be vastly improved, but there would be a few differences in process. Let's see how NASA would do it today:

2 years to let the RFP;

1 year to assemble the small disadvantaged business subcontracts;

2 years to develop the ISO 9000 implementation plan;

4 years to implement ISO 9000;

4 years to abandon it and move to whatever the next quality buzzword system is by then;

1 year to develop the earned-value tracking system;

2 years to figure out that it tracks the wrong things;

5 years to reinvent the avionics and onboard systems;

4 years to write the design specification;

5 years to reinvent the ground control systems;

2 years to write the design specification;

5 years to retrofit essential requirements that were waived earlier to save money;

2 years to actually build the damn thing;

1 year to test it;

2 years to train the Senators traveling as cargo;

14 days to fly it.

Well Rand, nothing like my argument being misstated, but that's probably because I wasn't clear. I don't doubt that we have the "technical ablity" to produce an anti-missile defense, but if you want to use the space program as an exemplar, then you need remember the failures.

Unless you can convince me that an anti-missile defense system would be 100% effective in operation, then I'm not comfortable with it's utility - even one missile getting through would result in deaths to numerous to write-off as an "acceptable failure rate".

As far as "stability" is concerned - let's talk about an "Asian arms race"

The problem with the space program analogy above is that the theoretical basis for the rocket launches was there. That is, you can work out the thrust equations, payload mass, fuel requirements, and figure out how it should be done (I did this in a frosh physics class, just like a lot of other people). Then it is *just* an engineering problem, if an incredibly challenging one. The same holds true for the simplified BMD problem, but there is no theoretical basis for distinguishing dummies from real warheads. Unless we can figure out a way to distinguish the decoys from the real thing, the offensive strategy is to put up 100 decoys and one real warhead, on the theory that we cant knock them all down, and the one that gets through is overwhelmingly likely to be the bad one. If there is a theoretical solution to this problem, I would like to hear it.

The decoy problem is vastly overstated. Decoys that can really fool a decent sensor cost almost as much (at least in mass) as a warhead, which means that the costs of the delivery system go up accordingly. Again, it's simply an economic tradeoff, but as the technology evolves, defense will be favored, because a warhead can be destroyed by something with much less mass (and possibly simply by extremely energetic photons).

I don't want to get into an extensive debate on the subject here, because, as I said, it has been argued to death in the literature, and in places like the sci.space.policy and alt.nuclear.war newsgroups. But suffice it to say that if the arguments against it were as straightforward and irrefutable as you seem to think, that few would be seriously proposing it. But many are...

With all due respect, the "people wouldn't propose it if it wouldn't work argument" doesn't fly with me. I don't think everyone who wants this wants it for the same reason, and it is hopelessly naive to think that all the proponents simply want to keep America save from incoming ICBMs. This constitutes corporate welfare writ large, and if the ulterior motives are too hard to spot, try looking at a list of campaign donors. I understand the urge to avoid the technical argument, and you are quite right to beg off on it (I probably couldn't follow most of it anyway). But the fact remains that the "imperfect system" analogies are weak. A flak jacket, to take the analogy from Mr. de Havilland, will protect a soldier from many projectile weapons, although not all. The difference is, that the soldier wearing the flak jacket has a reasonable guess as to what those other threats are, and can respond. We do not have that ability with "rogue states" and their various WMDs (see the anthrax scare if you don't believe me). In any event, it will proceed, and the physics and engineering challenge will be met, to the extent that it can be. As with my earlier post above (on the other BMD thread), the important thing is the cost benefit analysis, and whether it is too soon to commit to this particular technology in haste