Thursday, August 6, 2009

Candide's take at COIN in Afghanistan (UPDATED)

Interesting talk today at USIP on counterinsurgency in Afghanistan with David Kilcullen and Andrew Wilder. Kilcullen was luminous as always, but although everything he said rings true, I am not sure where this takes us in terms of COIN strategy. Let me try to summarize his intervention in a nutshell and a (at least half-fictional) dialogue:

Kilcullen: So far we have attempted to defeat the insurgents rather than the insurgency. We need to change this and move from an enemy-centric approach to a population-centric one.

Candide: OK, that’s Galula 101, I get it. So that will require lots of boots on the ground, which is why we keep sending more and more guys.

Kilcullen: And since the Taliban’s strategy is to wait us out and intimidate the population rather than seek military engagement, this means that this commitment needs to be extended in time.

Candide: Uh oh, not sure this is going to fly with lots of NATO leaders who are hoping to get reelected.

Kilcullen: This is why this extended commitment can only be done realistically (and effectively) by the Afghans themselves, so we need to beef-up the ANA and ANP.

Candide: OK, so let’s train more people.

Kilcullen: Not that simple. Afghans will only resist the Talibans if they think that another form of government can bring them more security, justice and basic services (to put it simply). And it is worth training more soldiers and policemen (especially policemen) if you can make them accountable and ensure that they do not turn into security threats for the people they are meant to protect. Training more bad cops sounds exactly like what could make the life of the average Afghan more miserable.

Candide: Does that mean we should put everything on hold until the governance/corruption problem is solved? So what do we do now?

Seriously, if this is where the counterinsurgency is to be won, why isn’t there more thinking and effort put into good governance and accountability? Is it for fear of undermining Karzai? Or because the task is already deemed hopeless?

UPDATE: Here's the link to the AP story about Kilcullen's remarks. The lede:

An incoming adviser to the top U.S. general in Afghanistan predicted Thursday that the United States will see about two more years of heavy fighting and then either hand off to a much improved Afghan fighting force or "lose and go home."

David Kilcullen, a counterinsurgency expert who will assume a role as a senior adviser to Gen. Stanley McChrystal, has been highly critical of the war's management to date. He outlined a "best-case scenario" for a decade of further U.S. and NATO involvement in Afghanistan during an appearance at the U.S. Institute of Peace.

Under that timeline, the allied forces would turn the corner in those two years, followed by about three years of transition to a newly capable Afghan force and about five years of "overwatch."

"We'll fight for two years and then a successful transition, or we'll fight for two years and we'll lose and go home," Kilcullen said.

"I think we need to persist," he said, but with "some pretty significant limits on how much we're prepared to spend, how many troops we're prepared to send, how long we can do this for."


  1. For those (like me) who absorb information more by blunt-force trauma than Gallic subtlety, the full title of Candide was "Candide: Or, Optimism."

    Two things on this: The first is that I too was struck by the futility of expanding ANSF who are not competent, but also by the impossibility of delaying the expansion of the very same ANSF until such time as they are competent. It seems now that we need approach this, again, with an "ink blot/oil spot" strategy: form small cadres, train them to competence, and then spread those men throughout a number of units to be stood up later with less-trained volunteers. Yes, we eventually need something like 480K ANSF, but it's not going to happen overnight. And even if it did, those 480K wouldn't be close to effective; you'd need double that when factoring in their level of competence.

    On another note (which you didn't really touch on), it strikes me that the USG might ought to consider putting out some negative statements about Karzai as part of an IO plan to bolster the independence and legitimacy of whatever government emerges from the election. If we indicate now that Karzai is disappointing in X ways, the election is less likely to be viewed as rigged by Washington.

    Of course, this doesn't solve the problem of widespread corruption and poor governance, but it at least gives an opportunity for what looks like Karzai's inevitable victory to not be ALL bad.

    Or something like that. I had a mind to consider this further and write something thoughtful, but so much for that.

  2. But isn't the mere fact that we are still in Afghanistan evidence of optimism?..

  3. Or cynicism, depending on how you look at it.

  4. I actually thought that Wilder was more lucid on accountability, with the major take home that we are just putting too much money into Afghanistan to be able to actually track it and provide oversight, so we should be spending less and having more analysis done about what works, where it does and why.

    He also pointed to this 1972 RAND report (which I've only read the summary but looks fascinating) that also points to what I really think the key point of the discussion was- that the US can identify the challenges it aces but hasn't been good at accurately seeing the limits of what we can do about it.

    The subtext I got from both men is that they see governance/troop levels as a cyclical problem, so the best we can do is try to fight both problems simultaneously (through things like the Partnership model) to try to meet fairly modest objectives. Thats not quiet seeing it as hopeless, but its certainly a dark view.

  5. el-Belle -- You'll be pleased to know that 80% of the Ink Spots crew was at that event, the insurgents swimming amongst the sea of USIP peaceniks. None of us were wearing black pajamas, though, so you probably missed us.

  6. With all pardons to Voltaire, if anyone would have been akin to Candide in a picaresque, it would be Kilcullen.


  7. I think that people are actually putting some thought into governance and anti-corruption, we just haven't been doing it enough and those who do it aren't sharing enough.

    This discussion goes straight to Kilcullen's point about being overly focused on a top down approach. I also think though that we've also focused on bribery and overall petty corruption (I don't like the term petty but in the corruption world, that's what you use to refer to small bribes, theft of small things etc). In short, we've provided ministers with mentors and we're trying to train police and ANA to not solicit bribes etc.

    As far as I know though, we're not doing much for mid-level public finance officials and managers of key ministries and administrations.

    In short, we haven't seen discussion of a GEMAP type of program in Afghanistan (Governance and Economic Management Program, look at In Liberia, this involved giving international officials co-signing authority in the public finance and resource ministries (timber, diamonds and mines etc). This helped to create mechanisms for accountability.