How COIN Generalists Fail Afghanistan

by Joshua Foust on 10/6/2008

This is far from a scientific pronouncement, but a recent op-ed by two fellows at the Center for a New American Security reveals some interesting thinking from establishment counterinsurgency theorists I think helps to explain why we seem to be understanding Afghanistan so poorly.

It’s not that the op-ed is necessarily bad or defficient in any way (though it is in many), but rather where they make leaps of imagination. For example: a staple of COIN is that increasing the legitimacy of the government, and connecting people to that government, is vital. So what’s with this?

The timing of two coming events, however, give cause for hope: the American election next month and the Afghan presidential election next year…

One sign of the current government’s unpopularity is that nearly all the prominent Afghans we met on our recent trip hinted at being presidential candidates in 2009. Still, when asked who will win that election, they responded unanimously, “Whichever candidate the United States supports.” Washington should send a message to every candidate that even tacit support depends on a serious commitment on three fronts: combating corruption; decentralizing governance; and negotiating political reconciliation with Taliban members who renounce violence.

How will any of those things increase the government’s legitimacy? To a large degree, Hamid Karzai lacks legitimacy precisely because he has been trying to do all of those things, however imperfectly—he is seen by many as actually too close to the U.S. Some refer to him as “Bush’s lapdog”—an Obama or McCain presidency will fare little better on that front. Karzai has sacked, or tried to sack, incompetent governors and sub-governors—which have then been reinstated under U.S. pressure (Antonio Giustozzi documented this extensively around the south). Or non-corrupt governors get sacked for criticizing the U.S.’s mistakes too much, and then are tarred as corrupt by the embassy so his replacement will have an easier time shaking things up.

Is decentralizing governance really the way to go? This sounds like a coded plug for the Community Development Councils, a project of the National Solidarity Programme which aims to decentralize government functions and in some cases even usurp traditionally centralized government activities. It shows promise, but is entrenching the occasionally violent autonomy of some of the more tribal areas like Loya Paktia really a sound counterinsurgency strategy? How does this “connect the people to the government,” or increase its legitimacy?

Lastly, negotiation with the Taliban is… problematic, at best. For starters I’d wonder “which Taliban,” knowing that didn’t even begin to get at the complexity of the problem (which is ultimately found in Afghanistan’s wickedly complex social relationships—something the U.S. policy establishment has taken precious little time to understand).

Worse still, many of these ideas are quite contradictory. The government cannot curb corruption—stamping down on police thievery, bribes, and opium smuggling—without a stronger presence in the districts. It cannot establish a stronger presence while becoming more decentralized—perhaps in theory, but in the real world the fundamentals of how Afghanistan works means the two will not coincide. So it’s a devil’s choice: decentralize, or push the government into the woleswalis, and in either case have a devil of a time connecting the people to the government.

In short, this piece makes for a nice microcosm of how quick country tours bracketed with little study of society tell even very smart people almost nothing about the country. All Singh and Fick got from their whirlwind tour of government officials or selected elders (under what conditions I wonder? Was there military escort for high-profile think tank fellows, or were they off on their own without body armor?) was a bunch of contradictory suggestions for how the next president can move forward, with no indication for how competing objectives could be balanced, ordered, or even navigated. This is because, unless there was a serious word-count limitation, they don’t seem to understand the fundamental forces driving conflict in Afghanistan. It isn’t government legitimacy, and it’s not even necessarily corruption (though anti-corruption is a highly effective COIN tool): the problem, the big problem they did not mention once, is security.

Oh sure, like all Americans writing about Afghanistan, they mentioned the 280 soldiers we’ve lost this year. They didn’t mention the (approximately) 720 policemen, or the (approximately) 680 ANA troops Afghanistan has lost (we don’t really know how many died because as best I can tell there is no official monitoring system for local counterinsurgent losses). While losing 280 soldiers is indeed tragic, many thousand civilians have died this year—and a not-insignificant number of those have been at U.S. or ISAF hands. The biggest reason more villages and villagers don’t support Karzai or the U.S. is fear, plain and simple. And Singh and Fick don’t seem to consider that an issue because, it seems, to them, and to far too many Big Thinkers in DC, it’s all about us—and not them. The “them” is the critical missing piece of the fight, and until we start to learn how we can help “them,” we won’t win.

Subscribe to receive updates from Registan

This post was written by...

– author of 1848 posts on 17_PersonNotFound.

Joshua Foust is a Fellow at the American Security Project and the author of Afghanistan Journal: Selections from His research focuses primarily on Central and South Asia. Joshua is a correspondent for The Atlantic and a columnist for PBS Need to Know. Joshua appears regularly on the BBC World News, Aljazeera, and international public radio. Joshua's writing has appeared in the Columbia Journalism Review, Foreign Policy’s AfPak Channel, the New York Times, Reuters, and the Christian Science Monitor. Follow him on twitter: @joshuafoust

For information on reproducing this article, see our Terms of Use

Previous post:

Next post: