Monday, January 3, 2011

My Skepticism and Optimism Over Reports of Progress in Afghanistan

One of the tough parts about an anti-insurgency campaign is that it's a pretty tough operation to measure with hard data. That drives people like myself quite crazy, at times. We depend largely on reports from mid- and high-level commanders, who have a bias to over-report success and under-report failure. But we do what we can.

There are a few reasons to be optimistic from some recent news. The first is that an (apparently) relatively large and relatively influential Pashtun tribe in Helmand, the Alikozai, have agreed to deny support to the Taliban in exchange for international aid and development.

It might be a key advance in an attempt to break a vicious cycle in the area, in which the local population doesn't support the Afghan government due to lack of services & security, and the lack of support from the local population bars the Afghan government from taking a foothold. The long-term strategy is to wash the Taliban out of stronghold areas (like Sangin), build government services and security, and thus not allow the Taliban to effectively move back in.

I'm mostly cautiously optimistic, though it'll depend entirely on whether the US is able to capitalize on this (and get more local allies), about which I'm rather skeptical. The US' only serious hope is in the brutality and unpalatability of the Taliban alternative.

I'm more skeptical about fact-lacking reports from senior US military leaders about the success of the North Waziristan campaign. A recent press release about "many, many" fighters/leaders killed or sent into hiding due to drone attacks brings only raised eyebrows. How many dead; what have the effects been? Where do we see relief of attacks? How are we entrenching the gains? None of it's there, and the risk is that, if out leadership is using these reports as signs of success, then we'll fall into the same traps as we did in Vietnam: mis-use our resources, mis-count victories, and ultimately lose.

No comments: