The Strange Contradictions of Andrew Exum’s Afghanistan Trip

by Joshua Foust on 7/29/2009 · 22 comments

So, Andrew “Abu Muqawama” Exum is doing the interview circuit about his experience as a part of General McChrystal’s 60-Day re-review of the Afghan War. It’s interesting to try to make sense of what he said beforehand and what he’s saying afterward—I’ll be the first to admit that going there can significantly change one’s perspective (I, for one, came home convinced the Army is incapable of fighting the war properly)—but some of these changes in attitude, or temperament, or even just word choice are really interesting.

To kick things off, we have Exum’s appearance on the Charlie Rose Show, where he said something interesting. Fourteen minutes in, Ex is discussing how leadership and strategy have been largely absent in the first seven years of the war:

We haven’t had leadership in Afghanistan that’s really been able to take control of this situation [protecting communities from intimidation]. That’s not an indictment of General McKiernan, who by all accounts was a highly competent commander, and, just personally, speak as a former Army officer, I think it was a bit disheartening the way he was dismissed (as big a fan I am of General McChrystal). But again, it’s strategy, it’s resources, and it’s leadership, and we haven’t had all three in Afghanistan until quite recently, and there’s hope that we can turn this around but quite honestly it’s late in the game.

Back in May, when McChrystal’s nomination was first announced and McKiernan was summarily dismissed, Ex was singing a different tune:

This tells me that President Obama, Secretary Gates, and Gen. Petraeus are as serious as a heart attack about a shift in strategy in Afghanistan and Pakistan. This was ruthless, and they were not about to do the George Casey thing whereby a commander is left in the theater long after he is considered to have grown ineffective…

The sad truth of the matter is that people have been calling for McKiernan’s head for some time now. Many of the people with whom I have spoken do not think that McKiernan “gets” the war in Afghanistan — or counterinsurgency warfare in general. There was very little confidence that — with McKiernan in charge in Afghanistan — we the United States had the varsity squad on the field.

What changed, I wonder? Was it McKiernan publicly grumbling about the humiliation of his firing, going so far as to say—at his own retirement ceremony!—that he was so mad he almost skipped it? Or something else that he saw or hears at ISAF HQ?

A bit later on, Ex also gave an interview to Judah Grunstein of World Politics Review (transcript here). This time around, he also said something else that was quite interesting:

I think that the population is not being kinetically targeted in the same way it was in Iraq, but what that misses is a silent war of fear and intimidation. Let me sketch this out for you: The fall of Kandahar is not going to look like the Taliban rolling down the streets in tanks. The fall of Kandahar is going to look like the Taliban steadily making ground with a campaign of fear and intimidation, and creating an environment in which the Afghan government can’t operate in Kandahar, and Kandahar eventually becomes ideologically inhospitable to the government of Afghanistan, never mind Coalition forces. So first off, the population may not be targeted kinetically in the way that it was in Iraq, but it’s certainly being targeted.

Only, in the CNAS “Triage” paper, on which I was a consultant for some sections, Exum argued something else entirely:

To be sure, violence will rise in Afghanistan over the next year—no matter what the United States and its allies do. What matters, though, is who is dying. And here a particular lesson may be directly imported from the U.S. experience in Iraq. In 2007, during the Baghdad security operations commonly referred to as ‘the surge,’ U.S. casualties actually increased sharply. What U.S. planners were looking for, however, was not a drop in U.S. casualties—or even a drop in Iraqi security force casualties—but a drop in Iraqi civilian casualties.

And in a sidebar titled “Key Metrics over the Next 12 Months”:

Civilian Casualties: A decrease in civilian casualties—whether caused by the United States, coalition, Afghan forces, or the Taliban—will indicate a genuine improvement in security. Conversely, a rise in civilian deaths will imply deterioration in the security situation.

Jari Lindholm, who pointed out this contradiction, wonders, how does one measure intimidation as a metric of success?

There are some other strange inconsistencies in Exum’s interview with Grunstein. In one paragraph, he brags about how the U.S. Army officer corps has grown and developed “a keen understanding of the operating environment in Afghanistan.” He continues, “Whether or not we’re going to be able to translate our operational prowess into strategic success is very much a question that is yet to be answered. But there was reason for being encouraged.”

Only moments later, however, when discussing Kandahar, Ex says he was shocked—shocked!—at “the degree to which we still do not understand this country that we’ve been in for the past eight years.” Is that different from an “operating environment,” or is he talking about two different things? Ex went on to say that knowing the Taliban is great, but “we need to understand power brokers at every level, we need to understand how regional power brokers interact with the insurgency, with the government, what their business ties are.” I dunno, I seem to remember a program supposed to do just that being savaged repeatedly in the press then undermined from above.

That’s a sidenote. Exum has this to say about how the Army understands counterinsurgency:

Nowhere that I went was I able to get a really coherent definition of what it means to hold and what it means to build, and how you do that. And I don’t think we’ve cracked the nut operationally on how we do those things. So first off, I think there’s some confusion as far as what that means. Second off, without question, we do not have the resources to hold much terrain in Afghanistan. We’ve got very limited international forces in Afghanistan, and we’re actually not using them to their best effect if we’ve got them “holding.” So if the Marines in Helmand are holding terrain right now, that’s a waste of resources. The “hold” function should be executed by a robust Afghan national security force.

Well, okay, I really like tautologies too, but this is a meaningless passage. We don’t know what “hold” means, but whatever it does mean we sure don’t have the resources to do it, and if we are doing that thing we can’t do and can’t define, then we shouldn’t be doing it anyway? No offense but if I want to hear crap like that, I already have Bing West. Exum’s boss, John Nagl, has led the charge in talking about the need for hundreds of thousands of more Afghan troops, when they are, at best, many years and many more billions of dollars away from being able to do the job we demand of them (and their end strength is so large and expensive the entire country of Afghanistan does not generate sufficient economic activity to fund it, so such an end state is fundamentally unsustainable as well).

There are, of course, notable exceptions in areas Exum traveled to. While Khost has not been trending well lately, we absolutely know what we need to do to “hold” communities, and Ex defines it in his own talk: create a sufficient sense and presence of security such that communities no longer feel threatened or coerced into supporting the insurgency. That’s what a “hold” is, and even in Khost we have short-lived examples of successful “holding” (like embedding individual units at District Centers to create pockets of security) that Ex, I guess, just never heard about from his generals.

That’s the biggest problem in all this talk of strategy, and the heart of why I was so deeply skeptical of even the current review: no one talks to each other, and no one does their homework (and by that I mean “no one in charge”). According to Exum’s new model for how to use U.S. forces, they should be limited to air-assault type missions, except we need more of them, because the problem is the Taliban’s campaign of silent intimidation. No offense, Ex, but does that make even a jot of sense?

In reality, we have had many successes in which NATO forces have partnered successfully with Afghan forces to achieve surprising successes. In Kapisa, they did that in Alasay until Kabul’s politics ruined their force structure (and they’re trying again this year). In Khost, they did that until units rotated and plans shifted. In Helmand, they’ve never even tried—and they’ve only seen, at best, ephemeral success and mass population transfers.

So we know what works, and we know what doesn’t, and we know why what works never continues past the RIP/TOA (new units, new plans, no room in the OER for leaving well enough alone). Yet, it’s somehow this bizarro mystery at the top levels of the military leadership what’s going on, and their civilians advisers’ response seems to be an earnest wish for more unicorns in the form of suddenly competent and westernized security forces.

That, in a nutshell, is why I have turned so deeply pessimistic. It’s not that the war in Afghanistan is unwinnable. It is that the Army refuses to win it, refuses to do basic diligence before sending soldiers into harm’s way, and refuses to learn from its own mistakes. I’m poking at Exum here, but he’s not even close to the real problem. I just want to know why his ideas and analysis have changed so much, sometimes within a single interview. What’s his secret, that he seems to have such a handle on this?

Subscribe to receive updates from Registan

This post was written by...

– author of 1848 posts on

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


Fnord July 29, 2009 at 2:10 pm

Youre not hating Abu M, I think. Youre hating the last 8 years, and I agree with you. We fcked up the basics.

Spencer Ackerman July 29, 2009 at 2:49 pm

Agree w fnord here. Ex is just the conduit for the broader confusion you’re railing against, right?

BruceR July 29, 2009 at 3:08 pm

Exum: “…our ability to be successful in Afghanistan largely depends on the Afghans themselves. And we can not force the Afghan government to be responsible and to not be predatory towards its people. We can use our leverage, but it’s going to be an uphill struggle.”

So we’ll win if the Afghans want to win and any Western efforts are largely beside the point. Check. I’m certainly not seeing an argument for *more* troops in there.

Jimmy July 29, 2009 at 4:46 pm

Afghanistan is screwed up. We all know that. But do you have any ideas about how to make it better or are you just going to complain and nitpick Exum? None of your ideas seem any better- quite frankly your blog posts, at least recently, seem to be nothing more than whining and crying about how bad the situation is.

Joshua Foust July 29, 2009 at 4:48 pm

Yeah, okay I’m glad it came off that. Ex was just a convenient introduction to talking about all of this. But I’m also genuinely interested in his own evolution—he seems to have realized the fundamental weaknesses of our approach, but can’t quite bring himself to say so yet.

Jimmy, I spent the first five months of this year trying to catalogue and discuss how to improve the war, shift tactics and policies to be more effective, and how we’ve been able to do that successfully elsewhere. I’m whiny and pissed off now because I’m seeing the defense establishment is wholly uninterested in making an honest examination of how its own policies are contributing to this debacle of a war. They’re refusing to learn from themselves, hence I’m snippy about it.

Greyhawk July 29, 2009 at 4:59 pm

I’ve been listening for the new notes in the symphony, too – but I saw the Charlie Rose appearance as more an explanation of issues than a position statement.

As for metrics in general, I’m against ’em. In one regard you’re telling the enemy: “make these numbers look bad for us and you win”. In that light “dead civilians” seems a bad choice all around.

Metrics are a fine measurement of progress to a goal, but I’ve never seen them employed (in two+ decades of peace- and war-time military experience or in historical examples) in a way that they didn’t become the goal.

Jenson Daniel July 29, 2009 at 9:06 pm

Afghanistan doesn’t need organic economic activity does it?

I don’t know what natural resources Afghanistan has or what sectors the population developed before the wars. I’m going to assume there must have been some level of agricultural, metalurgical, services (banking) that supported the trading regimes. As I understand, Afghanistan served as a logistical conduit for the silk trade to the ‘weak underbelly of India’ and between Persia & China.

I do know that Reliance (which as a firm accounts for nearly 5% of India’s GDP) is building a refinery that will make it a market maker in oil (5% of global refinery capacity) with a focus on Asian markets. I do know that Iran & Saudi Arabia are aggressively looking to Asia as a growth sector in their investment portfolios as they shift away from the G8. I do know that the primary land route that bridges the Gulf sovereign funds & Iranian ambitions with Asian consumers is Afghanistan.

Could Afghanistan grow into this role? A broker for oil transport & maybe even a trading center? My questions reflect my hope I think.

P.S. Oil is an interesting resource. It’s still found in harsh underdeveloped regions. It’s still explored & cultivated by only a few supermajors who don’t use indigenous populations because of the engineering & finance learning curve. But it has recently become an asset class which means it has a variety of related financial products that ‘individuals’ can purchase in personal investment portfolios.

AJK July 29, 2009 at 11:41 pm

Re: Jensen. Afghanistan was mostly ag&mining pre-1979.
-Most of Afghanistan is mountains. Camels taking the long, dusty, trek from Isfahan to Kashgar in Orientalist novellas don’t like mountains. All the same, there is infrastructure. But a cynic could call it smuggling routes.
-Not a lot of oil. Some natural gas. Nothing Turkmenistan doesn’t have.
-Why on Earth would a company invest in the civil war morass that is Afghanistan c.2009 when Central Asia is secure (if a bit under the heel of a few different tyrants) and linked up to better infrastructure? Iran is trying to get a pipeline deal with Turkmenistan, which makes a lot more sense and is still fraught with difficulty.
-Supertankers. Whatever you want to trade between Oman and Mumbai can go by water, until Somali-trained al-Qaeda out of a Kaplanian nightmare show up on the Indian Ocean.
-Oil is interesting. Fascinating. has a lot of neat stuff about it. Oil doesn’t have much to do with Afghanistan though, especially when Kazakhstan is right up the (silk) road.

This is probably more snipe-y than it needs to be, and I apologize for that. But Asia is heeyuuge, and Afghanistan is probably the one place in it right now that there’s not a lot of money to be made for a lot of businesses. Micro hydropower(shameless “check out my blog!”), though…

Nathan July 30, 2009 at 7:54 am

Jari Lindholm, who pointed out this contradiction, wonders, how does one measure intimidation as a metric of success?

Night letters? They’re not necessarily a direct measure of intimidation — not all of them are clearly threats –and they’re not all going to be counted. However, they do indicate insurgents’ contact with and presence among the population.

Joshua Foust July 30, 2009 at 7:59 am

That’s one thought. But Exum is discussing a social current, for example when people react to beheadings or beatings. In an area where you cannot conduct statistically meaningful surveys, that’s a tough nut to crack.

Same with shambanah (or however it’s spelled). We only hear about a fraction of the letters that get posted. It’s a good trailing indicator of how America-friendly locals fare, but I’m just not sure how much real utility it has in a general sense. Have you read any good studies comparing the incidence of night letters in a given area with changes in Taliban activity? I think something like that would be a way of building a framework.

BruceR July 30, 2009 at 8:50 am

In Kandahar City itself, that silent fear and intimidation Exum’s talking about mostly involves a highly effective and targeted assassination campaign against government and Coalition supporters, combined with kinetic attacks against fixed installations. So yeah, I think it would be objectively measurable on that basis. It always has been. Kandaharis will start supporting the government when a prominent government supporter or two can be shown to have stayed alive and attack-free for a year or so.

Nathan July 30, 2009 at 8:53 am

You’re right. I don’t think you can measure “mood” the way he wants to. Shabnama provide a measure of insurgent contact with the population. However, that assumes the ability to get collective asses together to consistently collect such information.

I have no idea whether or not there is any relationship between Taliban activity and frequency of shabnama. There may not be. Just an idea of one physical thing that could be counted to indirectly measure intimidation.

David M July 30, 2009 at 12:38 pm

The Thunder Run has linked to this post in the blog post From the Front: 07/30/2009 News and Personal dispatches from the front and the home front.

el-belle July 30, 2009 at 1:06 pm

I’m also curious at the difference between Ex’s post-review comments, and those of Anthony Cordesman ( and, given that prior to the trip there didnt seem to be that much distance between them (for example looking at the recommendations in Triage and in while certainly not identical, I think there’s a clear parallel in the way the threat is articulated and the general areas that most need improvement). Where they further apart to begin with then I’m seeing, or did something about the trip change (either or both of) their views?

Toryalay Shirzay July 30, 2009 at 10:13 pm

Fnord is right on the money.I still don’t understand why the ultra-smart American superpower went into Afghanistan and let most Taleban and their leaders get away and go to their sanctuaries unscathed not to mention Alqaeda??!!Hey experts out there,come out and try to give some honest answers here,eh.
Joshua Foust is trying and he is mostly right ;the picture is not complete though,And BruceR points are well made.The Taleban are not very visible,but they are like the invisible man.Intimidation,fear,secret visits after hours and many other tactics are real threats Taleban use daily to defeat the coalition;on the other hand,counter-operations to above threats is hardly visible and one wonders why??

Schmedlap July 31, 2009 at 2:38 am

That post was a great discussion.

In regard to “measuring” fear and intimidation, Nathan and Bruce point out two of the factors that we tracked in Iraq: night letters and assassinations (as opposed to indiscriminate killing). Also, kidnappings, and VCDs of assassinations and torture distributed to the populace. Specific threats are another, though measuring these can be difficult. If the threats are effective enough, the intimidated parties will never tell you about the threat because they’re too scared. However, we found the fear and intimidation metrics in Iraq (in 2007) to be a good leading indicator. When the fear and intimidation dropped, the other “metrics” such as IEDs found/exploded, direct fire attacks, and indirect fire attacks soon dropped as well.

Unfortunately, fear and intimidation metrics cannot be measured very well without very good human intelligence collection, a solid understanding of the populace, and some historical data to put it in context. D’oh!

M Shannon August 1, 2009 at 7:15 am

I think the situation is obvious. Petreaus took credit for improvement in Iraq as a result of his very clever new doctrine. Logically it has to be applied in Afghanistan because it’s THE way to deal with insurgencies. The problem (even if you think it’s what worked in Iraq and I don’t) is that there are not enough troops to do pop-centric COIN in Afghanistan because NATO isn’t on board and the ANSF are largely rubbish.

As this unravels the public supporters of COIN will start to hedge so that they have skeptical comments on record so as to not go down with the ship.

Kung.Fu.Panda August 4, 2009 at 12:31 am

The “war” in Afghanistan was over a long time ago…about the time that Karzai was voted in. We achieved the core goals…overthrow the Taliban for supporting AQ, destroy the great bulk of in country insurgent infrastructure (i.e. destroying the cave networks we built for the mujahedeen) and fostered the establishment of an elected government. That’s all that we could have expected. From then on, Afghanistan should have been left to itself. “Nation-building” never works…when you have a corrupt and in some cases indefensible government (in the form of the Karzai administration), ethnic groups that are habituated to fighting each other regularly, and a non-existent national military and law enforcement structure there is nothing to reconstruct.

Recent efforts to portray the country as an Islamic “domino” where the fall of the Karzai government would result in a wave of regional destabilization and jihadi ascendance are a farcical re-hash of Vietnam-era propaganda.

FDChief August 5, 2009 at 3:56 pm

What I find depressing about all of the bloviating about COIN, CT, “wars on terror” ad infinitum is the total lack of geopolitical vision involved. Exum is a bobo, yes, but no bigger a bobo than most of the talking heads that pop up, like whack-a-moles, on the damn talk shows and Congressional hearings.

Bottom line: as a foreign occupier you have one way to win a guerilla war against a local adversary. The Romans figured it out 2,000 years ago: “Solitudenem facient et pacem appelant”. You make a wasteland and call it peace. It’s an indescribably vile, inhuman business and it works. You “pacify” the occupied land for a generation, or more.

Short of that, show me the Western occupation that EVER worked in the long run. Malaya is cheating, because you are depending on an enemy that consists of an ethnic minority making moron-level stupid decisions. The reality since 1945 is that the foreigner comes, the foreigner fights, the foreigner leaves. If the local government or what passes for a government can beat down, or buy out, or compromise with the insurgents the war ends. Otherwise, not.

Jesus wept. This isn’t rocket science, and yet the Exums and Kagans and Wests – not to mention the legion of uniformed policy and strategy wonks – keep circling this drain without ever starting from the simple bedrock hardscape I’ve just laid down.

So I get the whole whiny and petulant thing. If we haven’t figured out that there’s no there there yet, we never will.

It’s worth noting that Spain had the same problem with the Netherlands some three centuries ago. Didn’t work out well for them.

Previous post:

Next post: