January 18, 2011

Talking Tactics With Andrew Exum

I’ve been struggling in this space with how we can really gauge the success or failure of our actions in Afghanistan. I can’t say I’m any closer to an answer now, but after much wrangling over Twitter, Andrew Exum and I had this productive exchange via email. We’re posting this to both our blogs in the interest of fostering more conversation about the way we are fighting the war, and how we can ever hope to know if we’re winning… or not. So with that in mind, I’d love to see a conversation, below, about how we can interpret the confusing mishmash of often contradictory data that emerges from the war.

Joshua Foust: Recently, Paula Broadwell recounted on Tom Ricks’ blog some operations in the Arghandab Valley, in Kandahar province. I found some of the events she described, like razing entire villages to the ground, appalling. At least in terms of tone, you seemed to agree: on Twitter, you referred to some passages as “cringe-inducing.” I saw that as an example of questionable tactics in service of a non-existent strategy. But it also made me think back to a report you filed when you returned from a tour of the Arghandab. “Counterinsurgency,” you wrote, “as practiced at the tactical level, is the best I have ever seen it practiced.” Clearly, I’m missing something between the two accounts of this valley. So, what are the indicators you use to evaluate tactical counterinsurgency as the best you’ve ever seen?

Andrew Exum: Yeah, the main problem I had with Paula’s post concerned the inability to see how ISAF actions might — while making perfect sense to ISAF military officers (and a West Point graduate like Paula predisposed to see things from the perspective of a military officer) — be perceived from the Afghan perspective. One of the things you often hear older military officers tell younger military officers is to “turn the map around”: how might the battlefield look to the enemy? I think that in counterinsurgency operations, where the population might matter more than in conventional, maneuver warfare, we have an obligation to turn the map around and see how our actions might be perceived by the local population.

Like Paula, though, I was impressed with a U.S. unit I visited in the northern Arghandab River Valley (ARV) last month. I have not had the chance to visit or observe the ARV over a long period of time and cannot say whether or not improved tactics will have a strategic effect, but I have observed U.S. military units struggle with the conflict in Afghanistan since 2001. I myself served there as a young platoon leader in 2002 and again as a Ranger platoon leader in 2004. I only mention that because I often compare and contrast units and small-unit leaders today with myself and the units I led in 2002 and 2004. I returned again in 2009 after several years spent wandering around the Arabic-speaking world.

The way one evaluates the tactical performance of a unit in combat depends a lot on how one perceives the conflict and what is important for victory. When it comes to maneuver warfare, the U.S. military has reached something approaching consensus on how we evaluate the tactical performance of leaders. U.S. Army Field Manual 7-8, Infantry Rifle Platoon and Squad (pdf), for example, is a commonly accepted reference used to teach small unit leaders how to fight maneuver warfare at the tactical level in an infantry unit. It is based on both recent historical experiences as well as practical lessons learned. It contains loads of assumptions, most of which have been pretty rigorously tested. (With often painful results for those testing them!)

U.S. Army Field Manual 3-24, Counterinsurgency (pdf) and U.S. Army Field Manual 3-24.2, Tactics in Counterinsurgency (pdf) offer similar standards for how we can teach and then evaluate units in combat in counterinsurgency operations. I should add, though, that I do not think the U.S. military and the scholarly community has reached anything approaching consensus with respect to counterinsurgency. I also do not think we have as rigorously tested the assumptions in these manuals as we should. (To give but one example, I question the degree to which our provision of social services really matters for success.) That having been said, when it comes down to it, I feel both of our counterinsurgency field manuals get a lot right. The emphasis in 3-24.2 on leveraging and supporting host national security forces, for example, is spot on. So too is the appendix on intelligence preparation of the battlefield (IPB): you can’t just know who you are fighting; you also have to know about the environment in which you are fighting. And I agree with the considerations for both offensive and defensive operations. [Note: I welcome any scholars who would criticize the manuals. My own thoughts on the things I think each manual gets right have been influenced by a) historical studies, b) what I myself have been able to learn by fighting in Iraq and Afghanistan and c) spending a lot of time studying the conflicts in southern Lebanon and Afghanistan as a civilian scholar and researcher.]

Based on the doctrine, what I observed in the ARV was encouraging. I saw a unit conducting aggressive offensive operations, fully integrating special operations forces into their plans and operations, and taking local security forces really seriously. I also saw a very sophisticated IPB — the best I had myself ever seen at the company-grade level. The unit I spent an afternoon with, for example, really knew their neighborhood. They knew everyone who lived there and all the buildings in their area of operations. When something changed, I got the sense this unit would notice. And that’s really important. I use The Wire a lot to explain everything from Lebanese politics to counterinsurgency, and I would liken the U.S. Army to the character Ellis Carver: when we meet him in Season One, all he wants to do is kick ass and take names. By Season Five, though, he’s become a much smarter police officer. He’s taken the time to get to know the people he’s trying to protect and can thus better separate the bad guys from all the people just trying to get on with their lives.

Anyway, all of that led me to observe that U.S. counterinsurgency operations at the tactical level were some of the best I had ever seen. Caveat lector, I do not know whether or not these improved tactics will yield a strategic effect. There are too many phenomena — many of them exogenous, as @ndubaz pointed out on Twitter — that we cannot even observe much less measure. And we still have a lot of known pains in our asses (like Afghan governance and sanctuaries in Pakistan) that could render tactical gains ephemeral.

As one final caveat lector, my observations were based on a limited sample, and unit and leader performance should be assumed to be uneven across the country. Still, I was encouraged.

Joshua Foust: Okay, so I can summarize: the operations you saw last year in the Arghandab matched with your interpretation of how one would enact both tactical and counterinsurgency doctrine, yes? Aggressive operations, integrating SF, and taking local security forces seriously, all of which add up to good tactics? Is there any way to be more specific?

For example, in this Broadwell episode, the local unit was most certainly using aggressive operations, and they integrated SF, and they even worked through the ABP to develop local knowledge. The thing is, the aggression resulted in the destruction of an entire village (something General McChrystal strongly urged against in the 2009 COIN guidance for which you were a consultant), and the SF’s use of the ABP — Col. Raziq is not from the Arghandab (the ABP has no jurisdiction in the district) and his tribe has been in conflict with many communities in this part of the Arghandab — is, let us say, a bit questionable. How can we tell the difference between an appropriate use of these three aspects of good tactical activity, and inappropriate use of these three aspects of good tactical activity? For example, what makes aggression proper now, versus the restraint previous COIN strategies required?

Andrew Exum: Those are great questions, some of which I am hesitant to answer. I am reticent to pass judgment on operations I have not personally observed. I am especially reticent to comment from Washington, DC on operations in Afghanistan. My perch at 1301 Pennsylvania Avenue is a great place to think about strategy or policy, sure, but not so much operations and tactics. The best (only?) place to observe the latter is in Afghanistan itself. So instead of passing judgment on the aforementioned operations, let me ask some questions instead — questions that may be useful for both commanders on the ground as well as for analysts like Paula who have had the chance to directly observe the operations themselves:

1. What are we trying to do here?
2. What effect will these operations have on the enemy?
3. How will these operations affect or be perceived by the local population?
4. What are the trade-offs for using a character like Col. Raziq? (On the one hand, he is seen as being effective, but on the other hand … well, anyone who has not yet read the 2009 Matthieu Aikins profile of Raziq for Harper’s should.)
5. What are the likely second- and third-order effects of our operations?

The thing is, you can be, to quote one Stan McChrystal, “tactically brilliant but strategically stupid.” Are the operations that Paula describes tactically sound? Maybe — I don’t know. But I would hope that officers on the ground — as well as Paula herself — are thinking through whether or not these operations will have the strategic effect we hope they will have. Maybe they will. But I would hope we’re thinking through those five questions I listed above, which have more to do with strategy than tactics.

As far as tactics are concerned, I would again refer readers to FM 3-24.2 for what the U.S. Army considers to be good counterinsurgency tactics. I cannot myself reduce “good tactics” down to three or four things: I just picked out three or four things that I believed helped to illustrate why I left the ANV last month impressed.

Joshua Foust: Okay, so you don’t like to condemn events you didn’t personally witness. That’s… fine, I guess. I wonder why, though, an afternoon of briefings is sufficient to declare tactics good in one case but a few thousand words describing tactics is insufficient to question tactical decisions elsewhere. It’s kind of the crux of what started this whole discussion: at what point can we reasonably ask probing questions about conduct? The outlines of this village razing incident in the Arghandab, in my view, warrants probing questions precisely because it is such a drastic measure.

So, at best I can tell this leaves me with two remaining questions.

1) If tactics are good and adhere to theory, but either undermine or don’t advance our overall strategy, what’s the point of praising tactics? Isn’t that just wasted time, effort, money, and, most importantly, lives?

2) I can accept your view that it’s difficult to question too much from the U.S. But if no one sitting in Washington, DC, can really question the tactics we read about, in what way can we, in good faith, question and strive to understand the war? This, too, is at the heart of why I’m asking these questions. It’s not as if everyone who is interested in understanding the war can go embed with the troops (and there is, unfortunately, greater difficulty for war skeptics to get precious embed space, compared to non-skeptics). If personal accounts, even (as I called Broadwell’s latest) hagiographies, are not enough to prompt serious questions about our conduct, how can we reasonably evaluate what’s happening?

Andrew Exum: Okay, I’ll address your points one at a time, but before I do, let me just say that I have really enjoyed this. Compared with trying to explain this over Twitter, conventional prose is a joy. And your questions are good ones.

1. Oh, there is a lot of good in praising good tactics. Let me name two. First, improved tactics demonstrate a military organization that has learned — which big bureaucracies often have trouble doing! That’s very positive. Second, it is too early to tell whether or not the near-term security outlook for the ANV has changed for the better. But if it does, we will want to note the correlation between improved tactics and improved security for rather obvious reasons.

2. This is a great and legitimate question. I should be more careful and allow that we can, in fact, judge operations from afar when the documentary evidence is solid. I’m not trying to say I can’t second-guess or judge William Calley, for example, because I wasn’t personally at My Lai! But I would want a lot more documentation than Paula’s single blog post before weighing in on this particular example.

I think you are somewhat incorrect to say that skeptics do not get to visit Afghanistan. You write this because you’re thinking of people like me who travel there as part of our jobs as civilian researchers and have been outspoken in support (to varying degrees) of the current strategy. But plenty of other civilian researchers and journalists I know visit Afghanistan as guests of the command and return to write critical reports — and then visit again (see Hastings, Michael). Other journalists and civilian researchers write highly skeptical accounts without ever embedding (see Dorronsoro, Gilles). I mentioned earlier the journalist Matthieu Aikins, whose reporting I love. It’s worth pointing out that he has, in addition to observing the war as both an embedded and unembedded journalist, also been an outspoken skeptic of the current strategy and, together with fellow activist-journalists Nir Rosen, Gareth Porter and Ahmed Rashid, offered his own policy recommendations. (Along with some guy named Foust and a bunch of other non-journalists.) So if all we had to go on was a blog post from my friend Paula, I would agree with your point. But I linked to that great Aikins piece on Raziq from Harper’s that is required reading for many government analysts working on Afghanistan. There is a lot more of that kind of critical reporting and analysis out there — you and I link to it every day. I’m just hesitant to judge something after reading any one thing — and I think you would agree with me there.