peacefare.net

Afghanistan decision time, again

Douglas Ollivant at foreignpolicy.com is asking the right questions:

What are our national interests in Afghanistan? Which of those are vital?

How much are we willing to pay for them (money, blood, institutional focus)?

What other costs does our policy in Afghanistan incur (e.g., reduced leverage in Pakistan and Kyrgyzstan due to reliance on supply lines through their territory)?

What are the opportunity costs? How might our goals be accomplished in other ways?

Is our policy sustainable to some sort of completion?

And what — if anything — do we owe to the people of Afghanistan who have sided with the NATO effort?

None of these questions is about numbers of troops or about the timeline for withdrawal, which is what the press focuses most attention on.  But the troop decision depends on the answers to these prior questions:  what do you want them to do, how much will it cost (important to include the opportunity costs, as Ollivant does), and how long will it take?

The New York Times offers at least some insight into possible answers to the first question.  If, as President Obama has said many times, our primary objective in Afghanistan is to rid it of al Qaeda, then the mission we need to continue is a counter-terrorism one (find them and kill them) that likely requires relatively few troops.  Vice President Biden and a goodly number of members of Congress will line up on that side.

The problem is “gone today, back tomorrow.”  If we leave Afghanistan a weak state unable to control its territory, there is every reason to think that al Qaeda will come back.  That’s why there has long been a state-building, counter-insurgency component to the mission, though the President has never made it clear what end-state he is seeking to achieve when it comes to governance in Afghanistan.  That’s not surprising considering the challenges, currently epitomized by failure of the country’s largest bank. But we could just as easily make reference to the country’s thriving drug trade and endemic corruption.  If we stay, and if building the Afghan state is part of the mission, we are going to need to get more precise about what we are trying to accomplish.

Why should we care how Afghanistan is governed?  The one-word answer is “Pakistan.”  If al Qaeda or other extremists re-establish themselves in Afghanistan, there is every reason to expect them to attack Pakistan, an already fragile state with a large and growing nuclear arsenal.  As Trudy Rubin explains in the Philadelphia Inquirer, this is a major reason for continuing the counter-insurgency and statebuilding effort in Afghanistan and presumably explains why outgoing Secretary of Defense Gates seems confident that the President will resist domestic political pressure to reduce troops rapidly.

None of this makes staying in Afghanistan look attractive.  As the American ambassador has made clear, President Karzai’s harsh criticism of the American and NATO efforts in his country is taking a toll, as are the mounting costs of the war.  I find it hard to fault people who would prefer to get out quickly (my wife has been on that side of the argument for a couple of years now), even if my brain tells me having to return to Afghanistan would be worse than staying.

The one thing I would ask is this:  if we are going to stay to stabilize Afghanistan and build its state to the point that it can fight al Qaeda and other militants on its own, we need to be honest about how long it is going to take and how much it is going to cost.  The projected date for turning over security in the whole country to the Afghans, 2014, is looking far too soon, even if Washington remains willing to pay Kabul’s security bills.  I’ve seen something of “stabilization” in Bosnia, Kosovo and Iraq, and I’ve studied it elsewhere.  My guess is that we’ll be there at least 10 more years in significant numbers if we want to get a half-decent job done.  That’s another trillion dollars, more or less.  Even among friends, that’s a lot of money, and a lot of other opportunities foregone.