Friday, February 13, 2004

Time on Target

Reader SN writes:

I think that Steven's main point of disagreement was lost in that little ellipsis in your first paragraph, namely, the scenario of a 30 city strike (or even a 10 city strike). The increase in logistics requirements to produce and transport that many weapons, especially while maintaining secrecy, appears to him (and to me) to be outside the bounds of all reason. Even if AQ stole 30 suitcase nukes from Russia, transporting them and carrying out a coordinated strike (especially the coordination with SECRECY) would be extraordinarily difficult.

Fair enough. I'll say now for the record that a 30 city strike is hard to pull off and retract anything I may have said to the contrary. But the reasons for it's difficulty are not as obvious as it might seem. The real constraint on terrorist nuclear capability is the availability of fissile material, not the tactical aspects of deploying the weapons to US cities. If bomb-making material is available there are good reasons for favoring simultaneous attacks over a series of individual strikes. The first reason was articulated by reader SR. Unless the terrorists are extremely stupid, they will have realized that striking a wounding blow to America is the wrong way to go. There are thousands of jihadis buried in Afghanistan and Iraq who are mute testimony to what a dumb idea that is. The advantage of a militarily weaker enemy vis a vis the US vanishes once America gets cranked up. The correct strategy is to kill America in the first blow. That was beyond the capability of the Japanese, but God knows they tried. And unless the Islamists are stupid they will try it too.

The second is tactical. If you compress an attack in time, you can lessen the chance of successful defense. If you had them, it would be better to set off thirty nukes at once than one a week for thirty weeks. This approach has various names, like 'saturating the defense' or 'time on target'. But the underlying idea is the same: you get temporary local superiority for the time you need it. Nor is a simultaneous attack necessarily that much easier to detect. Because clandestine groups are small relative to their milieu, they resemble convoys at sea. One would expect that convoys of a hundred ships are easier to detect than a hundred individual ship sailings, but WW 2 operations research found that given the "vastness of the sea" a convoy was just as hard to find as a single ship. If you have a "fixed cost" in safe houses, training personnel and comm exposure it may be better to risk them once rather than serially. Of course there are disadvantages to a simultaneous strike, but they are not as great as one would think.

To recap, the main thing is to prevent the fissile material from getting into the terrorist's hands. That's where most of the safety lies. If the terrorists can get the material, then relying on the logistical difficulties to protect targets will be a weak reed.