HUMANITARIAN. I think Yglesias goes too easy on Eric Posner's Washington Post op-ed attack on humanitarian intervention. Posner invokes Somalia, Kosovo, and Iraq as evidence that "experience shows that humanitarian war is an oxymoron." This can fairly be argued of Iraq 2003, but I'm unaware of any compelling evidence that the intervention in Somalia in 1992-3 failed to increase living standards for Somalis, at least for as long as the United Nations forces stayed. Certainly, the intervention failed to establish a state or resolve the problems of Somalia in the long term, but this is a different thing than saying it failed. The success of an intervention must be measured against the likely course of events in the absence of action, not in reference to whether it permanently solves a problem.
In the case of Kosovo, Posner's argument is even weaker. Posner writes "the Kosovo intervention, although regarded as a success in some quarters, has cost billions of dollars, required a seven-year occupation and could turn out to be a slow-motion version of Iraq," an assessment which seems wholly indefensible. Unless you are of the sort who believes that the same Serbs who had carried out brutal massacres in Bosnia and Croatia in the 1990s were planning to take it easy on Kosovo, the consequences of non-intervention were almost sure to be worse than the consequences of action. Posner is again playing a bait and switch, evaluating the war on whether it created a cheerful, independent Kosovo rather than on whether it prevented a genocide.
Finally, Posner ignores the humanitarian intervention in Iraq that worked, namely the creation of no-fly zones in the north and south. Threatening to destroy Iraqi aircraft if they crossed a line in their own territory is, believe it or not, a form of humanitarian military intervention. The intervention saved thousands of Kurdish and Shiite lives, and allowed the construction of political institutions in the only part of Iraq that isn't currently fractured by violence.
Humanitarian intevention can work under certain circumstances to avoid outcomes that we reasonably believe will be horrible. This doesn't mean every humanitarian intervention is sensible, but disaster in Iraq (which was not intended as a humanitarian operation in the first place) should hardly serve as justification to distort the rest of the picture.