Matthew Yglesias wonders:

At the intersection of these two arguments is the idea of “no blank check” for Hamid Karzai. How, one naturally wonders, is the check not blank if the President of the United States has defined the mission as serving a vital American interest? If you made the case that the mission is a good idea differently—if you just said we’re obligated to the Afghan people and government to give it a try—then your check has real limits. We’re obligated, but they’re obligated too, and if they don’t meet their obligations we can meet ours so we’ll have to walk away. But that’s not what he said. What he said was “I make this decision because I am convinced that our security is at stake in Afghanistan and Pakistan . . . [w]e must keep the pressure on al Qaeda, and to do that, we must increase the stability and capacity of our partners in the region.” Insofar as that’s true, then it’s true completely independently of how we feel about the Afghan government, so Afghan government actions have a limited influence on our policy, so whatever checks we write to them are pretty much blank.

I think it’s fair to say that the problem Yglesias identifies can’t be solved within the logic of Obama’s strategy, but it can be mitigated. As my piece reports, the administration is aiming to — well, if not exactly circumvent the Karzai government, say that it has more Afghan partners than just Karzai: “it would aim its military and development assistance down to Afghanistan’s provinces and districts, where Karzai’s influence is relatively tenuous.” Hence the reorientation of development aid from big reconstruction projects — and if you’ve been to Afghanistan and seen the destitution there, you have a sense of how little sense those make — to “immediate impact” projects like agriculture, and particularly irrigation*.  And that’s of a piece with the decision to partner with local or tribal militias.

And guess what: that’s extremely problematic. I do not know, and no administration official has credibly explained to me, why Karzai won’t perceive the U.S.’s double game and proceed with commensurate intransigence. But I do know that they’re aware of the problem, and looking toward mitigation. They think that “immediate impact” agriculture happens so rapidly and is so rapidly beneficial that the Afghan government isn’t going to waste time obstructing it, and will instead take credit for partnering with the U.S. to produce it, since that’s in a weak government’s interest. And that in turn creates healthy local connections between outposts of the government in the provinces and district with the populace, with the prospect of a virtuous circle rolling out.

I do not know if it will work. But it strikes me as at least a plausible way to broaden the U.S.-Afghan relationship from just a U.S.-Karzai relationship. And unlike previous administrations, I do not get the sense that the Obama administration or McChrystal’s command sees the approach as stronger or less problematic than it is.

* For an account of how smart Army officers learned not to screw around with Afghan agriculture, see this post.