Bernard Finel continues the discussion on limited interventions and responds to this post:

I disagree that limited interventions rarely stay limited. Lebanon 1983, Libya 1986, Somalia 1992-3 are obvious examples. We used limited force, and then withdrew — for good or ill — when the marginal costs exceeded marginal benefits. You don’t always have to double down, and in practice we haven’t always done so. Indeed, it was precisely the point of my post that we need to talk about limited objectives as a way to avoid the pressure of all-or-nothing intervention. The all-or-nothing notion is very much a post-9/11, nation building, “never again” sort of construct. We need to push back against it.

If we consider U.S. interventions during the last thirty years, what we find is that even many of the relatively limited ones were multi-year efforts that suffered from mission creep, and two of the interventions Finel cites are excellent examples of efforts should never have been made. It was politically possible (and necessary) to cut losses in Lebanon and Somalia because those interventions were finally judged to be tangential to perceived U.S. interests, which they always were. This was a belated recognition that the original decisions to intervene were misguided. To this day, many hawkish interventionists incredibly regard the decisions to withdraw from Lebanon and Somalia to be mistakes, and unfortunately that has tilted later debates towards maintaining costly long-term commitments for years after the initial intervention. Many of the interventions that do stay relatively limited are those that never served U.S. interests in the first place. One of the dangers of initially limited interventions is that there will be constant pressure to deepen U.S. involvement once U.S. “credibility” is on the line. This is made worse if, as in Kosovo and Libya, the U.S. involves NATO as well. Though the original objectives might have been limited and not particularly important to U.S. or allied security, the prospect of failing to achieve even those limited objectives is never considered acceptable, which is why the U.S. and NATO began seriously contemplating the use of ground forces in 1999 before Milosevic yielded.

I acknowledge that the purpose of punitive strikes is to raise costs to the regime. Isn’t it possible that limited punitive strikes could also do just enough damage to drive many Syrians back towards the regime without doing enough to change its behavior? It seems to me that there is reason to believe that Assad has not unleashed as much destructive power on his opponents as he could, and it seems that the reason he has not is that he does not want to risk triggering outside intervention. Once punitive strikes against his regime begin, his forces could lash out even more violently against rebel population centers. The effort to punish Assad could result in effectively punishing the civilian population.

I agree with Finel’s suggestion about the need to offer Assad an incentive to give up power. If his only options are being put on trial or being killed, he will probably take his chances by trying to hold on to power and fight to the end. At present, there seems to be very little interest in any sort of negotiated solution, and almost all discussion of Western responses to Syria’s conflict has been framed in terms of military action or nothing. That is the sort of all-or-nothing thinking that most needs to be challenged.