Reviewing Obama’s War: Part I

This weekend I finally managed to have the time to sit down and watch this excellent PBS Frontline documentary called Obama’s War. Highly recommended and hats off to the folks at Frontline for a very good piece of work on an extremely important topic.  I’m going to do a number of posts all branching out of this doc this week. One of the key strengths of this film is that it gets some very big name folks on all sides of this issue.

As Andrew Exum (who by the way has the greatest avatar in the blogosphere), one of the ones interviewed put it:

John Nagl, Bill Mayville and Stan McChrystal make a good argument for a counterinsurgency campaign, while Andrew Bacevich and an especially pithy Celeste Ward make a good argument against pursuing such a campaign. All sides, in other words, acquit themselves rather well. All sides, that is, save for the Pakistani officials.

Digging deeper into the Frontline site, there is a page with transcripts from all the interviewers.  There are a whole mess of them, but the best ones in my opinion are Steve Coll, Andrew Bacevich, John Nagl-Andrew Exum, and Rory Stewart.

Nagl, Exum, and McChyrstal are on the side of a full counterinsurgency (COIN) operation in Afghanistan following their work in Iraq, predicated on clearing insurgents, holding territory, and building infrastructure policy so that a central government might come into take over, thereby allowing a natural exit of US/NATO/ISAF forces.

Stewart and Bacevich, for various differing reasons, stand opposed to such a position.

While Steve Coll represents something of an in-between point of view.

So I’ll start with Coll.  His interview is here.  Coll has the best understanding of the history of Afghanistan, and as a guy with a history degree, I think it’s the best place to start. 

Coll begins by describing how President Obama came into office basically with little to no choice but to have an initial troop increase in Afghanistan and (as of March of this year in fact) basically calling himself for a counterinsurgency campaign.  In the meantime, however, Obama has come to take a much more serious look at whether this option is viable.  The Obama Administration is using the Afghan election controversy to hold back on making a decision.

And then Coll makes an extremely important point about COIN relative to the larger framework of national security and foreign policy:

Counterinsurgency doctrine is now ascendant in the military and in civilian circles involved in national security thinking. It’s very widely distributed in the form of a new manual that the Army published several years ago, FM 3-24 (PDF) [Counterinsurgency Field Manual]. …

But the doctrine itself is not a one-size-fits-all manual. It can’t be applied by universal principles and succeed. It has to be adapted into local circumstances. And it has become, to some extent I fear, an end in itself.

The purpose of counterinsurgency doctrine is to serve strategic objectives, to meet and secure objectives that are subordinate to vital national security interests of the United States. That is, it’s a tool; it’s not an end. And I think one of the challenges that the president faces is to distinguish between his ends in Afghanistan — and in South Asia more broadly — and his means.

One important tool is counterinsurgency doctrine, but it’s not the only one that’s going to achieve these objectives.

In the case of John Nagl, I think Coll’s criticism is valid.  [A point Andrew Bacevich makes in some detail in his interview–more on that anon.]  Nagl says that the US is involved in a worldwide counterinsurgency campaign and that counterinsurgency is the best (only?) mechanism to properly deal with counterterrorism.  But counterterrorism is only one portion of the larger foreign policy goals/objectives of the United States.  Or should anyway.

In other words, it’s not just that Afghanistan is not Iraq and that the lessons learned via the surge in Iraq are not in a 1:1 correspondence to Afghanistan (although that is true and important to be sure), but that COIN is only one means to an end.  And when there is–as Coll argues there is and I agree with him–lack of clarity about what the goals/objectives are, then COIN can (and is) sold as some kind of panacea to all ills and foreign policy problems.

COIN proponents often cite the famed 20:1 rule.  1 counterinsurgent per every 20 people.  In the case of Afghanistan that means something like 500,000-600,000 counterinsurgents.  You read that number correctly.  5-6 with five zeros after  it.  Now all of those counterinsurgents need not be (and eventually should not be) foreign fighters.  The idea eventually is to create a national army and police force that will roughly approximate that number of people.

And here Afghan history–as opposed to theoretical constructs gained via generalizing from other contexts–is very important.

Coll:

The goal, I’m quite sure, of the troop deployments that will be recommended to President Obama is not to have American forces prosecute combat in Afghanistan for as long as it takes. The goal will be to have American forces in sufficient number to stabilize the country long enough so that Afghan security forces can be trained and deployed to carry the security mission forward as Americans move from the front lines into a supporting role.

That’s the narrative in Iraq; that’s the narrative that will be pursued in Afghanistan. The question is, how much time do you need? How much uncertainty do you have to bear in pursuit of that objective? And is the deployment of additional American forces itself destabilizing?

In other words, is this a case where the use of foreign expeditionary forces in a counterinsurgency role is not securing the population but provoking additional instability?

That’s a question that I think the president is going to have to debate before he decides to send these troops.

Gen. McChyrstal is asking for something like 40,000-60,000 more troops.  This will then be used to create an initial surge, attempting to buy time to accelerate the training of the Afghan army/police force.  Nagl (who at least is honest) thinks this will take at minimum five years.

And Afghan history is instructive.  They are not, to put it mildly, fans of occupation.  What Coll is asking is whether an obvious further embedding to that serious degree will provoke even more insurrection/insurgency within the country.  I tend to think so, but it’s an open question definitely worth asking.

Also consider this from Coll:

We know a couple of things that live in contradiction with each other.  One is, we know that the Taliban are not a popular movement, even among Pashtuns. Most Afghans, even in the Taliban strong lands, would prefer to live without Taliban governing them.  At the same time, we know that the Taliban are an element of indigenous social, tribal and cultural forces in Afghanistan, and they’re not outsiders arrived to wage a war on behalf of some neighboring government. They are partly that, but they are substantially local.

So the classic counterinsurgency task of separating the population from the insurgents is doubly complicated because of the role that the Taliban have developed as a parallel government, as a parallel source of teaching and religious instruction. And yet we know the Taliban are unpopular, so it’s not as if you’re fighting a lost cause.

The other problem is that Afghanistan has an obvious history of reaching a point of national hindsight on foreign troops in which a consensus develops that whatever good they announce they intend, whatever good we may have once thought they had to offer, we’ve had enough of them.  And there is no scientific way, I would assert, to know exactly how many foreign troops the Afghan body politic can digest without revulsion. And we are on the cusp, I think intuition would suggest, of forcing such a question into the forefront.

And more importantly, since the foundation of COIN (as currently practiced) is predicated on the notion of the creation of a strong central state, what about Afghan history relative to its state?

To be sure, between 1920 and 1975 there really was an Afghan state. It was a weak state, it was an unintrusive state, but it did describe and enforce a common culture, common borders.

The reasons many Afghans resist the Taliban is that they — I think quite correctly — have a memory of that state as their own. However much it was associated with poverty and underdevelopment, it was a coherent, peaceful, national state. And there is a desire among Afghans to resurrect that version of Afghanistan and then to try to move forward into a more modern political economy.

But that’s different than saying they want the government of New Jersey or Maryland or Colorado delivering services on every corner, ringing the bell when the school year begins and announcing holidays. That is not a tradition of state presence, formation or operation that has any roots in Afghanistan outside of a few cities.

So the first data point in deliberations on this policy is that the notion of a strong central government as the primary strategy for achieving the goal of defeating the Taliban should be thrown out.  There is not going to be a strong central state in Afghanistan.

Now there could be COIN projected to a different strategic mechanism of governance.  But unfortunately this (potentially rather creative and historically/culturally nuanced) view is not I think at the heart of the current COIN enthusiasm in US military circles.  In other words, the President should be very suspicious of the COIN idea in Afghanistan.  Not because of the military tactical side per se, but because it is linked to a very likely failed strategic end of strong national state-creation.  Notice that this problem continues even if there is a strong accelerated program of national army/police force training.  If you train all these guys, who are they fighting for if the central government is a failed enterprise?

Lastly Coll returns to the point he made at the beginning–what is the actual objective(s)?  Even if you could create a strong state function in Afghanistan–and this is as we see historically without precedent in Afghanistan–is the creation of a strong state the primary way to defeat al-Qaeda?  Are The Afghan Taliban and al-Qaeda synonymous (a question I’ll delve into more in relation to Rory Stewart’s interview).

Coll makes the argument for 2 objectives:

1. Preventing al-Qaeda from launching attacks on US and/or allies  (basically agreed upon by all sides.  Though there is disagreement as to best way to do that).

2. This one is more open and not one I’ve heard explicitly made before–or at least as clearly.

I think if you argue that a second vital American national security interest beyond Al Qaeda is the stability and perspective normalization of Pakistan in South Asia broadly, then you regard the Taliban as a serious obstacle to that vital interest.  Then you recognize that reversing the Taliban’s momentum, preventing them from taking power in either Afghanistan or in Pakistan, is a vital American interest. And you also recognize that this is not a conventional war. You’re not sending in more troops in order to get them, the Taliban, to a surrender ceremony.

You’re engaging in a complicated struggle in two countries, some of which involve direct military combat, some of which involves regional coercion, diplomacy, economic development, all kinds of political components. And it’s also eventually going to involve negotiations with elements of the Taliban.  So if that’s the goal, then you have to go back to this basic question: Are 20,000 or 30,000 more troops advancing that objective or not? And if you want to argue that they’re advancing the objective, then tell me why. This is the critical question. More troops have to serve strategic objectives.

It’s not just about stabilizing a plurality of Afghan provinces. That’s not the strategic objective. The strategic objective is to marginalize the Taliban to such an extent that threats to the integrity of the government of Afghanistan and the government of Pakistan recede and can no longer recur in any foreseeable future.

This objective is more controversial.  Does the US have an interest in the normalization of Southern Asia?  I think the answer to that is yes.  I’m not convinced that the Taliban are automatically going to A)take back over Afghanistan and B)invite al-Qaeda back in and basically re-live the 1990s in Afghanistan.

I do think however The Taliban (both Afghan and Pakistani versions) are a rogue force and do threaten–if not at least boxed in and somehow dealt with through various mechanisms (some military, some political amnesty/role sharing)–the normalization/political modernization of Southern Asia.  e.g. There are a number of Uzbek fighters in the Tribal lands of Af-Pak and could easily return home and start problems there.

But it is not at all clear to me that the prime way to deal with this problem is a massively poured in COIN state-building strategy.

A counterterrorism only strategy of just attacking al-Qaeda from the sky, minus some sort of counterinsurgency strategy, would I think leave Southern Asia in a serious problematic state.  Now just to be clear, since this idea is sometimes attributed to Vice President Joe Biden, Biden (as far as I understand) is recommending counterinsurgency focus (plus counterterrorism ops) in relation to Pakistan.  By counterinsurgency is there meant putting pressure on the Pakistan army to undertake offensives–like the current one–against the Pakistani Taliban while the US funnels money into the building up of civil society.

That isn’t the same as just abandoning the entire region except for aerial attacks on al-Qaeda.

Some conclusions from this round:

1. Afghanistan is not going to have a strong central state.  Certainly not within 5 years.  A COIN strategy predicated on and assuming such a state can be engineered is bound to hubristic neo-colonial failure.

2. Increasing troops levels could very well–given Afghan history–cause a larger revolt against foreign occupation.  Could.  The Taliban are not loved by the population, but neither are the foreign troops.

3. A larger discussion needs to be had about objectives with regard to the region.  (not al-Qaeda).  Attempting to use al-Qaeda as the means to sell a regional war leaves out important regional players like Russia and China.  This is separate from but related to the stupidity of trying to sell (absent any regional buy-in) strong sanctions against Iran based on potential nuclear weapons threat.  i.e. Trying to build diplomacy based on what are perceived to be US only interests/concerns.

4. Counterinsurgency could be done with A)the number of troops now in the field and B)without necessarily linking it to the Afghan State counter the common view of COIN advocates.