Trying to digest "what we've learned since 9/11," Francis Fukuyama makes two points worth repeating:
Conventional military power continues to be useful against nation-states, but it is much less useful against networked, transnational movements that are deeply embedded in local populations. The United States, which spends as much on its military as the rest of the world combined, has not been able to pacify a small country of some 24 million people after 3½ years of effort, nor was Israel able to militarily disarm Hezbollah. This is due both to the nature of the enemy, and to constraints on the use of force to which all democracies are and will continue to be subject.
The converse side of the previous proposition is that conventional military power, including nuclear deterrence, should continue to be effective against nation-states like Iran. Anyone who believes that Iran's Islamist ideology is so extreme that it will be willing to in effect commit national suicide to achieve its ideological goals needs to defend that argument explicitly. It is possibly true, but far from obvious either from the history of earlier ideological regimes, or from Iran's own behavior since 1978.
Right. Iran knows we can destroy them, but would be unable to occupy them. In that way, if America let it be known that occupations were simply too much of a pain in the ass, and from now on it would use its military strength punitively rather than "constructively," our deterrent effect would likely increase. So long as Tehran thinks we're deciding whether to invade and rebuild, they'll (rightly) assume that our likelihood of following through is low. If they think we're deciding whether to level the place, however, their calculus would be different. Now, I'm rather certain leveling the place would be a bad idea, but it's a relatively better idea than actually invading.
