Center for a New American Security

Drop your COIN manuals!  Here are some more lessons we can learn from 19th-century Brits.  The link to the original text is there too.

Money quote (from page 10-11):

"Now no man of ordinary sense should for an instant attempt to make any sort of comparison between Regular and Irregular Cavalry.  There is no greater resemblance between them than between Macedon and Monmouth.  'Look you there is salmons in both,' but it goes no further."

There's actually an important lesson in there.  Apparently involving salmons.

Seriously, can we get more flowery analogies and alliteration into modern doctrinal publications?

Charlie tries to walk the fine line between defeatist, triumphalist, and realist in Iraq. (When she's really in a pinch she claims to be an academic and make some useless, but intriguing, pedantive distiction about what's going on there.) But, to her mind, handing over Anbar province to the Iraqis is a BFD. And while she's well aware the peace there is fragile, it was also unimaginable two years ago.

That said, she's pretty sure we still don't have a good grasp as to the underlying causes of that peace...nor of the Anbar Awakening upon which it rests. (The old adage about it being better to be lucky than good comes to mind.) The influx of "surge" troops let us take advantage of a fluid situation newly in our favor, but we didn't create it and we sure as hell can't RE-create it in Afghanistan (though tribal engagment should surely be part of our broader strategy there). Hopefully somebody will work that out before we redeploy all the soldiers and Marines from Anbar to that theater. There's still no secret decoder ring for counter-insurgency...

were greatly exaggerated. Apparently Jalaluddin Haqqani, Mujahideen leader, darling of Charlie Wilson and the CIA, and current insurgency leader, is still alive (or at least was recently).

A March 22 report by Al Jazeerah carries recent footage of the leader who has been replaced at least in the field by his son Sarajudin Haqqani. Most recent reporting out of Afghanistan has assumed Jalaluddin's death with many Afghans accepting a rumor that he had died on a visit to the United Arab Emirates.

The Haqqanis are involved primarily in a dispute within the Zadran tribe that has involved Pacha Khan Zadran (previous enemy of the Afghan government and US forces, current member of the Afghan Parliament, although with men still in Gitmo). The Zadran tribe falls outside of the much talked about Ghilzai-Durrani split, once again reinforcing the problem of conjuring the conflict in Afghanistan as a Pashto tribal dispute between the Ghilzais and Durranis. The main area of influence for the Zadran tribe is in Khost Province.

Khost, along with Paktya and Paktika, were part of a single province named Loya Paktya (greater Paktya). Unliked the dispersed tribal networks of the South and Southwest and far East of Afghanistan, tribes within these three provinces live in large contiguous areas. The jirga is a strong institution in these provinces as are tribal police/militias called arbakai. Also, unlike the other provinces with a strong insurgency, Paktya, Paktika, and Khost are for all intents and purposes poppy free. These dynamics make the three provinces the most likely area for an Anbar-like awakening in Afghanistan. Unfortunately, tribal frustration at both the government and Coalition has grown in recent months (neither the national government or ISAF could be bothered to show up in any force at a major tribal jirga held in November of 2007) while the main effort at developing local security solutions seems to be in the vicinity of Helmand, where a dispersed tribal network and much weaker jirga tradition make tribal security solutions far less likely to do anything but exacerbate conflict.

A week ago, Mike McConnell, US Director of National Intelligence, declared that the Afghan National government held control over 30% of Afghanistan while the Taliban held 10% and traditional elders administered the remaining territory.

The Afghan government was, shall we say, not sanguine about the assessment. Two days ago, Amrullah Saleh, the Afghan equivalent of Mike McConnell except that he actually has command authority, shot back.

Amrulleh Saleh said that eight of the country's 364 districts, which account for about five percent of the land and two percent of Afghanistan's population, are outside central government control.

"While in America, an administration fully backed by tribal chiefs or dominated by tribal chiefs may be seen as a liability," said Saleh. "But here we see it as a very strong asset."


Now, Kip has the utmost respect for Saleh and his organization (the NDS or National Directorate of Security), which has totally been overlooked by the military forces in Afghanistan. And, Kip understands the need to respond and the way traditional government has ruled in Kabul. Yet the truth is that the Afghan government under Karzai, unlike previous successful leaders such as Nadir Shah and Abdur Rahman, has failed to reach out to the tribes and other traditional leaders. The Ministry of Kuchi and Tribal Affairs is essentially one old building with one old bureaucrat established not for tribal outreach but instead to preserve the governments ability to reject the Durand line.

As an example of the ineptitude, a huge, multiple eastern province tribal jirga organized by the Tribal Liaison Office, an Afghan NGO, and covered by Tolo and Ariana TV in November 2007 was an indicator of the sentiments of the tribes (we hate the central government and ISAF for corruption and breaking their promises) and the lack of both government and international interest in the tribes as indicated by the low-level central government and non-existent ISAF participation in the event.

While it is normally Secretary Gates' job to get NATO Secretary-General Jaap de Hoop Scheffer's panties in a knot, Kip suspects that the future of Afghanistan will be determined by the 60% of Afghanistan's population not ruled by the Taliban or the Kabul regime.

Many of our readers likely came across Max Boot's article this past week, arguing that we're winning the war in Iraq but that victory is only possible if we maintain troop strength. And that kind of logic is pretty typical of most Iraq analysis:

current assessment + hand waving = conclusion you came to three years ago

Fortunately, one of Charlie's favorite people emailed in with his own thoughts* (forgive the length, the University of Chicago teaches many things, but brevity is not among them).

This is a good piece, but I would argue on a couple of points. For one, the political solution in Iraq is not about extending security to all corners of the country. That is what security operations are for. The political strategy should be about taking this brief pause in violence and using it to build institutions that will lessen the risk of future violence. Anyway, this article is all about how we've almost achieved a balance but we're not quite there yet... With the implicit argumentbeing that once we get to that balance we can declare victory. But a balance of power is a MEANS, NOT AN END. We have achieved a balance, now what are we going to do with it? In civil war settlements that last, institutions are formed that provide the credible commitment and which help lessen the security dilemma. This works because institutions are"sticky," and have independent power to affect behavior.

We have achieved a precarious equilibirum in Iraq. It is a balance of power, but balances of power are inherently unstable. They last as long as all parties believe time is on their side. When one side no longer believes time is on its side, the whole thing comes apart. So right now we should be institutionalizing the balance, rather than trying to figure out how to expand it to the far corners of the state. Provincial elections (WHICH WE SHOULD TAKE A STAKE IN) and accelerated con-federalism are good places to start. The structure of these provincial and regional governments are, institutionally, where we have the most room to manuever, because they weren't well defined in that mess of a constitution.

Internal balances of power in civil conflict require a credible outside commitment from an external balancer. This ameliorates the security dilemma and allows for compromises to be madeand institutions to be formed. INSTITUTIONALIZING OUR GAINS IS THE IMPORTANT PART (and I don't mean by forming more extra-governmental militias.) The external balancer can't stay forever, (and balances themselves don't last because they are all about expectations and perception of future relative balances.)

On a different note, I think it is important not to confuse concerned local citizens' groups with the Anbar Awakening. The Awakening is a tribal movement, instantiated in the Anbar Salvation Councils, while the CLCs are insurgent groups being turned into local militias. As Marc Lynch has pointed out elsewhere, there are huge tensions between the two, as CLCs claim that the tribal leaders are "reaping the fruits of the jihad," without having fought for it. This will only get worse as the tribal leaders make overtures to SCIRI/SIIC, which has angered the former insurgents in some of the original CLCs. We shouldn't conflate insurgent movements with tribal movements.

*Excerpted with permission; and with the hope it may convince a reader or two that PhDs are not always a complete waste of time (sadly, you won't get top notch training in causal inference anywhere else).

Seems Afghanistanica hit a nerve. Mac McCallister reponds to his criticism (via SWJ):

I take great pains not to advertise this model as a predictive tool in the sense that Afghanistanica’s comments seem to imply. Afghanistanica and I differ because I accept intuitively that effects resulting from all individual and group interactions are “determined not simply by preceding causes but are part of a continuous process of evolution. These complex interactions are too numerous to predict, identify and observe as they manifest themselves in their various end states along the historical timeline.”

[...]

I apologize for my primitive outlook on the human condition and readily accept that my position might be anathema to those that embrace the idea that reason and rational thinking is the only mechanism to create a social order. I therefore want to strongly reinforce the point that the model is not an attempt to predict individual or group behavior but an attempt to recognize patterns of behavior and why this pattern may exist.

[...]

On the other hand, I can’t disagree more with the insinuation that the lens “requires” one to see culture as static. It is as if I was told that a given professional football game is static because the players follow a set of simple rules. The game remains the same but the actions expressed on any given game day certainly do not.

Read the full post at SWJ. McCallister focuses mostly Afghanistanica's concern about "predicting" behavior, but that wasn't the sum total nature of the objection. For Charlie's money, the real heart of the argument lies in the opening salvo:

The majority of the Taliban and Jihadists are radically removed from the "older concepts of community and traditional codes of behavior." Of those that aren't foreign, most are products of refugee camps that have very little resemblance to traditional village life, and their behavior reflects this. And others have used "jihad" as an excuse to overthrow or marginalize the traditional leadership.

That sounds to this blogger like it would be difficult to discern any "pattern" of human behavior in this shifting cultural landscape. (This of course also dovetails with Kip's argument that there is no monolithic pashtunwali.) Now, no one would ever confuse Charlie with a Middle East or South Asian tribal expert, but would it be correct to say that the tribal dynamics in Pakistan are more fluid and in a greater state of "shock" than those found in Anbar in 2004-2007? And as a result, we're just a whole helluva lot less sure wtf is going there? Or is there a way to sufficiently tailor any tribal program to take these different levels of variation and disruption into account?

(Hope you're enjoying Space Mountain, Kip. Charlie's just going to continue with the WAGs until you get back.)

A quick note on the ongoing discussions about engaging the tribes in Pakistan and Afghanistan:

Much I have seen written in the press and by military authors on the tribal code known as pashtunwali suggests some sort of monolithic code upon which we can project group behavioral expectations.

However, pashtunwali is highly varied across Pakistan and Afghanistan and highly dependent on both tribe and location. The requirements of hospitality, the power of the tribal meetings, the empowerment of tribal militia, the extent, boundaries, and power of individual tribes (e.g., large contiguous tribal areas in the southeast versus small pockets of tribes running from Pakistan to Iran in the south and similarly dispersed tribes in the east), and the source of authority for elders (religion versus wealth versus military prowess versus hereditary) all vary.

Few things may prove as dangerous as generalized conceptions of pashtunwali in the hands of those who establish our strategy--especially given that two recents heads of CENTCOM's tribal engagement team in Afghanistan had no background whatsoever in either tribes or Afghanistan.