I’m more favorably inclined toward humanitarian intervention than Matt Yglesias, so it isn’t too surprising that I disagree with his piece in the Prospect re-evaluating the success of the Kosovo intervention. But even given that, his argument seems pretty weak:
Now that Kosovo is formally separate from Serbia, it seems overwhelmingly likely that protecting the rights of the country’s geographically concentrated Serb minority will either require the indefinite presence of international troops or else a further round of secession in which Serb sections of Kosovo are carved out and allowed to re-integrate with Serbia. This still looks defensible compared to the alternative course of action of standing aside and letting Milosevic have his way with the province. But many hawks looked at Kosovo and saw not a boundary case for when the use of force might be legitimate, but a new baseline against which future interventions should be judged. If you were willing to use force against Milosevic, the thinking went, then why not Saddam? Why not Sudan? This line of thinking came to a bad end in Mesopotamia, but many harken back to the Balkans to try to make the case that Iraq should be considered an exception and not something that casts aspersions on the utility of unconstrained American power.
You could say the same thing about World War II in the 1950s, couldn’t you? “Yes, the policy of using overwhelming military force to combat totalitarianism worked in defeating Imperial Japan and Nazi Germany, but that just led crazy conservatives to advocate use of the same method against the Soviet Union. The latter policy is clearly crazy, meaning that World War II must be suspect as well.” That argument is clearly wrong, but is it not the same one Matt’s making, with the countries switched? Granted, there’s an several orders of magnitude difference in scale, but still.
I’ve been batting this around for the past couple weeks, and I keep concluding it all comes down to competence. Not the procedural, “did you look at a map of Cuba before ordering the Bay of Pigs invasion?” competence, but substantive competence: “is this policy the most effective means of attaining the policy goal we’re after?” Substantive competence, like procedural competence, is situational. Military force may be the most effective means to a moral end in some situations, like Kosovo, and not in others, like Iraq. Diplomacy might be the most effective means to a security end in some situations, like North Korea, and not in others, like Afghanistan. You can’t boil it down to a generalization like “military intervention is an effective humanitarian tool” or “military intervention isn’t an effective humanitarian tool”; it all depends on the circumstances.