Andrew Exum perceives some writing on the wall now that Hamid Karzai appears to have won an election at the expense of his legitimacy.

When people look back on the Afghanistan war, this might be the moment when historians will judge we should have cut the cord on the Afghan government. If we believe Generals McChrystal and Petraeus, and we believe a counterinsurgency campaign to represent our best chance of success in Afghanistan, then we have a big problem. Because if we believe what we ourselves have learned about counterinsurgency campaigns, we understand that we cannot be successful in one if the host nation government is seen as increasingly illegitimate — and that’s what the Karzai government is.

Legitimacy, as Lipsett writes, is a relative. Its root is the belief that existing institutions are those most appropriate for society. The deeply unpopular Taliban’s form of government is not seen as being a better alternative to what the Afghans currently have. But the Afghans are both losing faith and looking for something else. The United States and its allies are blamed not just for keeping Karzai in power but also for the excesses of his government, his relatives, and local officials. And for those who are calling for a withdrawal from Afghanistan, this election presents the best possible excuse to do so.

To take the observation further: the U.S. has interests in Afghanistan/Pakistan that exist independently of the legitimacy of the Karzai government. Specifically: the destruction of al-Qaeda across the border in Pakistan and the subsequent elimination of its supporting networks of fighters, financing and assorted other support in Afghanistan. All strategies in Afghanistan are meaningless and foolish if they don’t advance those interests at acceptable cost. If keeping U.S. troops in Afghanistan, for instance, defeats that purpose, then the troops need to come home. Similarly, the U.S. has an interest in the legitimacy of the Karzai government because and only because a legitimate Afghan government is perceived to have utility toward the ultimate anti-al-Qaeda goal.

As Ex indicates, the theft of the election renders the presumption behind that last perception inert. It doesn’t follow from the stolen election that U.S. forces need to withdraw from Afghanistan, since that leaves the core interest at stake in Afghanistan unaddressed. What does follow is that the U.S. needs to decouple its Afghanistan strategy from the Karzai government.

One option is to focus increasingly on military operations in south and eastern Afghanistan to erode the Taliban’s influence in its heartland and filter the civilian aid necessary for giving the population incentives for bandwagoning with the U.S. through local proxies or through direct administration, a far less desirable scenario. This option, in truth, probably means more kinetic military operations than it does civilian reconstruction efforts; and it also probably means a reduced reliance on training the Afghan security forces and a reduced expectation on them taking over, since I don’t know how you bankroll Afghan security forces without working through the government. In fairness, I don’t also know how you circumvent Kabul for aid efforts without risking a showdown with the government that makes continued operations against al-Qaeda support elements ultimately untenable. You may be able to do it somewhat, but probably not for very long. 

In short, there are certain facts that compel changes in strategy. An illegitimate host government is one of them. Counterinsurgency-as-nation-building needs to be, as the saying goes, rethought now, as does any strategy predicated on support for the host government. If the taint of illegitimacy spreads from Karzai to the U.S., we’ll be placed in a situation where the "strategically decisive" Afghan popular sentiment decides against us, and that would render the core interest out of the realm of advancement. 

Just to be clear: this is me spitballing. I don’t have a fully-thought out proposal; I’m trying to think through the options. Comment away.

Update: The Pentagon’s Shawn Brimley comes to a similar conclusion as my spitballing does. In Ex’s comments (at 1:29 p.m., I can’t figure out how to get the distinct URL) he writes:

I think the best option is to go forward by focusing on local and regional approaches to security capacity and governance. I mean look, this is ultimately what we did in Iraq. Yes the election outcome is a worst case scenario but to me suggests simply shifting our focus toward a bottom-up approach, something that both the history of Afghanistan and our own recent operational experience suggests is the best option in a world of bad ones.

Because I learned in Liberal School never to take my own side in an argument: How would Karzai react if we tell him that all of this money we promised for reconstruction is going to go around his office? His Axl-esque paranoia will be rather borne out…