What is state failure? See my conceptualisation of state failure on the right flank below.

Saturday, June 6, 2009

The new CNAS report many are blogging about

Authored by a number of CNAS fellows, including fellow-blogger Andrew "Abu" Exum, there is a new report out on how to do triage in Afghanistan and Pakistan (I mean, if somebody reading this hasn't yet noticed). Here are my critical remarks - some in the form of questions.

Some remarks:
- The civilian surge sells and works better if it is done in order to prepare the ground for channeling more aid through the Afghan government. If there is no commitment to that... (Wait, is that state-building we are doing there?)
- Also, take into account that corruption in Afghanistan is not really an independent variable (as much as you can objectively measure it). Prospects and pay matter at least as much as people. By the way the "people" in this equation are generally the B team that are not working as officers, drivers or interpreters for the IGO/NGO crowd.

Some questions:
- How can it be guaranteed that the Pakistani police get the aid that is meant for them?
- How can it be expected of them, even with better equipment, to take on seasoned fighters of the Tehrik-e-Taliban, who are aided at times by elements within Pakistani intelligence, as well as other state institutions?
- How could Pakistani police become a "bolstered" force in under a year?
- What is the relationship between the two key metrics named for Afghanistan? Can territory be ceded to avoid air strikes? Does territory have to be defended at all costs (thinking of Chora, Arghandab, Zhari, Panjway, Musa Qala etc.)? Is there a readiness to take more casualties among troops to avoid ceding territory, without air strikes?
- Do Pakistani civilians all crave for an end to the Taliban when many perceive the U.S. more of an enemy? Consequently, is it Pakistani "civilian control" that is tested when we measure cooperation with the U.S. "in Pakistan"? You know, a test of this could as well be how many totally anti-American articles appear in Pakistani dailies (not just the English-language ones of course).

Some more remarks still. If Afghan elections are to be held, a truly disruptive surge of insurgent activity needs to be avoided. Any attempt at that may need to be contained. For such an attempt, the resources would have to come in across the border. So a key metric in these upcoming months is "any kind of cross-border insurgent activity." And of course not just in the east, adjacent to the FATA, but on the Baluchistani border as well.
Caveat - for a little break in the midst of all this seriousness, with much more seriousness coming still. Here is the sort of cross-border movement you don't need to pay attention to - movements of a natural insurgent named Bayad, pictured in the right/lower corner...
Some more possible metrics - for Pakistan: number of registered IDPs; number of operational training camps in Pakistani territory (intel).
For Afghanistan: number of compounds destroyed by coalition fire, CAS, artillery, else (a negative metric, just like civilian casualties - as Jari says, screw up less); number of pro-government mullahs assassinated (again, a negative metric, something Alex Strick van Linschoten keeps emphasising a lot from his Kandahar vantage point, understandably); number of Afghan police killed (again, a negative metric); number of schools actually working (positive); number of rural health clinics working (positive); IED tip-offs; weapons cache tip-offs; durably IED-free districts/provinces.

Of course, I'm not trying to suggest the report doesn't have its merits. For example, we're in full agreement over the need for population-centric counterinsurgency in Afghanistan. Those who think ISAF should just imitate Sri Lanka's example: look at how well that works in Swat against this kind of insurgency - not at all the LTTE kind of insurgency, you know...

Counterinsurgency in Afghanistan can only work in the long run, however. So all the search for metrics may be in vain. You need some indicators of progress that may look good only ten years from now. Or later.

When it comes to Pakistan, one key to change there would be stopping treating that country like a 5-year old. "You're in terrible danger! Can't you see? I'll make you see it! Here, I will feed you my aid! Even more aid than before! EVER!" You know what I'm getting at... Key actors there might know there are dangers. Their job is made more difficult if other key actors are offended by being told what Pakistani interests are and that they have to act accordingly, on the double; if others want to wait till the cash-floodgates open before doing anything; and if some may be just looking to "privatise" some of the money. And so on... Places like the Swat are their responsibility.

Overall, regarding both countries: this is partly a ghost war, partly one in which difficult trade-offs may need to be made. One cannot really talk very openly about strategy in this sort of context.

No comments: