Stephen M. Walt

Our slow-motion exit from Afghanistan

Our slow-motion exit from Afghanistan

Secretary of Defense Leon Panetta announced yesterday that the U.S. is going to step back from a combat role in Afghanistan by mid-2013, and shift over to an "advise and assist role" instead. Assuming he means it, we’ll be ending our combat role about a year before all U.S. troops are supposed to be out.

As regular readers know, I’ve favored a greatly reduced presence in Afghanistan for a long time, simply because I didn’t think a COIN/nation-building campaign there was worth the costs, and because I don’t think the outcome in Afghanistan makes much difference in the larger struggle against Al Qaeda. (In other words, I reject the "safe haven" justification for the war, largely because Al Qaeda has havens elsewhere and Afghanistan isn’t an especially desirable one from their point of view).

But by a strange coincidence, we were discussing an aspect of this problem in my graduate course the very same day that Panetta made his announcement, in the context of a broader discussion on international cooperation. As some of you know, one of the basic principles of the literature on cooperation is that it is facilitated when there is a lengthy "shadow of the future." States are more likely to cooperate today if they anticipate being able to reap the benefits of cooperation far into the future; they will be leery of stiffing potential partners and foregoing that stream of long-term benefits.

What does this insight have to do with Afghanistan? Although I favor getting out as rapidly as possible, we ought to do so with the full knowledge that announcing a certain date (or even an approximate date) will reduce Afghan incentives to cooperate with us now and in the interim, and their incentive to cooperate will decline more and more as the date of withdrawal nears. Once they know that the stream of benefits is finite, they will be less willing to make adjustments or concessions to us in order to keep us in the fight. So by announcing we’re leaving, Panetta was tacitly acknowledging that our leverage over the Afghan government is going to erode pretty quickly. Not that it was ever that great, of course.

Notice: This situation is different than trying to encourage greater Afghan cooperation by threatening to leave if they don’t shape up, coupled with a credible promise to stay if they do. In this case, continued U.S. help would be conditional on Afghan cooperation and reform. But that’s not what we’re saying: Instead, we’ve made an essentially unconditional pledge to end our combat role (and eventually leave completely). In short: We’ve had enough of this war and are heading home, if not exactly briskly.

As I said, I think this is the right course of action. But actions have consequences, and we should be under no illusions about what it means for our ability to determine outcomes there. Washington still has a few cards to play (i.e., we can still empower different contenders by providing them with money, arms and training), but our long-term influence over decisions there is going to decline rapidly. But unless you’re one of those people who thinks it’s a good idea for Americans to try to steer the politics of an impoverished, deeply-divided Islamic country in the middle of Central Asia, this development really isn’t so bad.