It's easy to understand why Obama is delaying a final decision on Afghanistan strategy. He's come to a major fork in the road -- and each potential choice calls for radically different actions that can't be easily reversed.
Eric Martin can correct me if I'm wrong, but Obama seems to be faced with two rival strategy approaches. The first is a counterterrorism approach focused on taking out bad guys. The second is a counterinsurgency approach (COIN, etc.) that focuses on winning the population.
The problem, though, is that each strategy requires dramatically different tactics and resource levels. A counterterrorism approach in a hostile country probably requires scaling down current troop levels. Counterinsurgency, by contrast, requires a dramatic increase in troops to be done right. Neither choice, once taken, can be easily undone. So deliberation seems right to me.
All that said, one of my concerns with the counterinsurgency approach is simply that too many things have to go right for it to work. I'm not saying it's impossible, or that it's a ridiculous argument. It just strikes me as a strategy that carries a very low probability of success given all the good things that have to happen.
First, we have to assume that we can pull this off with less troops than counterinsurgency theory suggests we need. Second, we have to assume that Karzai's government will not only get their act together, but will be perceived differently (and more favorably) by the actual population.
Third, we have to assume that we are capable of pacifying the population centers. Even if we are, though, it doesn't mean we've "won." The current counterinsurgency approach still assumes that the Taliban will control much of the countryside for the foreseeable future. The ultimate hope, then, is that our "ink spot" strategy will allow us to eventually expand control out into these areas.
And we have to achieve all of this without creating new high-profile civilian deaths that inflame the population. Plus, we have to stay there many, many years -- and spend a lot of money.
And even assuming all of that goes well, we're still not home. Andrew Bacevich writes:
If we could wave our magic wand today and transform Afghanistan into whatever it is the COIN [counterinsurgency] advocates think they can achieve there, the threat posed by jihadism would still exist and would not even be appreciably diminished.
It just seems hard to adopt a strategy that requires so many things to go right, which even then isn't assured to help our national security.