Choosing More Appropriate Tools to Understand the Population

Isn't everything that being discussed in this theme really all about "conflict ecosystems"?

How many intelligence analysts or for that matter Commanders can sit down and with paper and pencil work up the "conflict ecosystem" that is facing them in a particular operating area.

Instead we rely on HTS to do it and they are not that good either with their results being hit or miss.

Everyone stood up and appluded Kilcullen's article in 2004 when he spoke about "conflcit ecosystems", but really how has that has been translated into an effective intelligence tool at the ground level?

Gents, also see Bernt Glatzer's essay on Afghan Ethnic and Tribal groups in Maley's "Fundamentalism Reborn." He only cites two instances of tribal strength in Afghan history: the Ghilzai Pushtun uprising in the 1700s and the Durrani empire. The former was only able to maintain control for 2 years and the latter was heavily reliant on Persian elites. I think his work lends support to your argument.

Let me know next you're in London.

Best,
Steve

Readers of this blog and especially this post will certainly want to read this article:

http://www.aan-afghanistan.org/uploads/20100624TR-HowTribalAretheTaleba…

Anonymous:

Shapiro offers great critiques of current social science approaches to understanding local populations, but the alternatives he offers are for the most part just as full of bad assumptions as are the approaches he criticizes.

I guess I don't see how "traditional way of life" or "traditional society" is any more clearly defined than "culture" is. They're all assumming a body of people that is more-or-less internally homogenous. That simply isn't realistic.

Populations are never consistent. They change constantly. Trying to identify characteristics of populations themsevles is like looking at the clouds and findings the ones shaped like animals or houses or boats--you can identify the patterns all day long, but in the end the objects you point out will change and disappear as soon as you identify them. We're better off looking at local populations' operating environments, not the populations themelves. The environments don't change as quickly.

sir,

thank you for your answer regarding anthropological findings. This is a fruitfull discussion for my understanding of the subject.

By "traditional anthropological categories", I was not thinking about broad categories such as "culture" (that is indeed very difficult to define), it was more related to narrower concepts (in the sense of observed regularities in the environment).

One "narrower category in anthropology" of traditional societies would be, for instance, regularities mentionned in the litterature in the way young people try to access to power (by way of arms and by way of religion leadership). One could argue that this is a usefull pre existing category if one comes to study a particular traditional society (Sorry if this point is not clear, Ian Shapiro says it much more clearly than I do in "the flight from reality in social sciences").

That lead to the question: where, how and who is following a traditional way of living in Afghanistan ?

V/R

Anonymous:

I'm not sure what you mean by "traditional anthropological categories," so I'm not sure how to answer your questions.

If you mean categories traditionally employed within the academic discipline of anthropology, then I really don't know how to answer your questions, because the discipline of anthropology is too incoherent for any set of concepts to really be considered a consensus position with the field. Anthropologists have spent the last 200 years just trying to come up with a consensus definition of "culture." They still haven't found it. There's no reason to really believe anything anthropologists say over anything that anyone else says.

If you mean categories such as tribe that have been used for many years to describe various population, then yes, I do argue that those categories are unproductive.

It all comes down to this: to explain people's behavior by reference to their tribal affiliaiton (or their ethnic, political, class, religious, or any other group affiliation) is to assume that those people think in a way the no person at any time in any place has ever been shown to actually think. If you ignore all of the hype about various "social science" disciplines and just look at what we know about how the human brain works, you can see that those categories stand on more assumption than evidence.

We wouldn't want to fight a modern war with a Cold-War understanding of weaponry or logistics. Group-affiliation explanations of behavior rely on a Cold-War understanding of how the brain works. We can do better than that.

sir,

thank you for your interesting comments. In your different assertions, I do not understand a backfrop issue:

- do you consider that anthropoligical categories are not very productive and that we should frame our intelligence of LP with other analytical tools ?

You consider that using cognitive sciences insigths will make the job of population intelligence easier. Could you highlight for us the relations with traditonal antropological categories ?

If you say 'influences X, Y or Z cause behavior B', are you practicing anthropology or a new form ? In particular, if you generalize your findings in the causal relationship, do you consider that you are creating anthropological categories ?

V/R,

Ian:

Yes, I'm sorry for not being more clear. What I'm saying is that affiliation is not a variable, period. It doesn't matter what the groups is--ethnicity, tribe, or something more fluid and small-scale like qawm--the group affiliation is still a folk term that obscures the things that cause behavior. The affiliation never embodies those causal things.

Schaun,

I'm not sure I understand what your comment means.

To be clear I agree that a single-variable explanation, whatever the variable, for the behavior of people in Afghanistan is not going to yield good predictive results. I also share your frustration when analysts refuse to look beyond tribe. My comment was just to say that qawm doesn't necessarily equal tribe or ethnicity and so when people use those different terms they are referring to different things (I guess, different "influences" to follow your wording). Tribe and ethnicity are relatively fixed identity categories for Afghans whereas qawm has the potential to be an elective affinity.

Ian,

No, when said qawm, I meant qawm: groups of any shape and size are not cohesive--they are not capable of acting as powerbrokers or of empowering powerbrokers. Individual people within groups are capable of doing both, but for reasons that do not have anything to do with the group itself.

Influences X, Y, and Z cause behavior B. Some people use tribe to describe situations in which X, Y, and Z are present. Other people use qawm. Other people use ethnicity. Other people don't use any term at all. In all cases, the term isn't what's causing the behavior. No matter what the grouping is, the term is just an inconsistently accurate description of some of the things that matter.

"powerbrokers are never based on tribe (or qabila, Khel, Qawn, guild, or any of the other terms that are sometimes used in its place)"

I think you misunderstand the term qawm, Schaun. The word just means group, and if you want to see an excellent exposition of how leaders of qawms act as powerbrokers, check out "Buzkashi" by Azoy. It's a fantastic read.

That's the north, though. The problem is with identifying non-charlatan, non-criminal powerbrokers in the south and east rural Pashtun areas. Many here at SWJ would like to default to appointing men as leaders and arming them, and then post-factum saying, look at the tribal leader who has agreed to cooperate with the ISAF mission.

Ron:

Thanks for your comments.

The problem is, powerbrokers are never based on tribe (or qabila, Khel, Qawn, guild, or any of the other terms that are sometimes used in its place). Getting a local definition doesn't help any because you're still getting a definition of something that isn't really a decisive factor in people's behavior. They'll talk about tribes, and they'll probably even say that tribes are important, but the term is just a convenient codeword that masks the factors that actually influence behavior.

There's the stuff that actually matters, and then there are the ways people talk about that stuff. Tribes are the second of those two things. Knowing how people talk about stuff can be useful when we're speaking with them and interacting with them (which can sometimes be extremely important), but the focus on the article was on tribe's uselessness when it comes to assessing, explaining, or anticipating local population behavior.

Nice article Shaun.
I agree that tribal affiliation is is fluid and sometimes irrevelant..But in the Pushtun areas it is often critical. The problem with "tribe" is that it is non-operational, ill-defined and vague. "Tribe" is often translated from Qabila, Khayl/Khel and Qawn. A few questions can usually clear it up and give you the local definition or the definition the informant is using. Between the Soviet invasion, refugee time in Pakistan and the Taliban many of the old structures are gone or have shape-changed.
I found using the word "qawm" very useful as it also can mean "tribe, various communal groups, a village or a professional guild. In cities like Ghazni "guilds" are important. Power brokers may be based on any or all of these foundations.

strength and honor,
Ron

Anonymous:

1. Powerbrokers can be close relatives, but are not always close relatives.

2. Feuds can occur within the LP, but do not occur consistnetly between any specific subgroups. The feuding parties change over time.

3. Some local powerbrokers and LP members have incentives to fight the Taliban. Others do not.

4. LP prefer conflict solving period. Their choice of conflict resolution method depends on their familiarity with the method and the desirability of the expected outcome.

I think all of those elements are relevant for LP intelligence. Sometimes those things overlap with tribal affiliation. However, they do not consistently overlap with tribal affiliation, and even when they do, the affiliation has nothing to do with why those things are relevant.

Tribal affiliation is too fluid and ambiguous to be taken as some sort of immutable characteristic of the LP itself, and--partially for that very reason--it doesn't correspond to characteristics of the LP's operating environments consistently enough to be used as a placeholder for the constraints that actually affect LP behavior.

sir,

thank you for your answer on the description of the Shinwari case.

However, this description is a bit disturbing for my understanding of your point. The shinwari case may be described, in part, by the following elements:
1/ powerbrokers in the area are close relatives
2/ feuds routinely occurs between subgroups of LP
3/ local powerbrokers and LP perceive talibans as threats to their way of organizing their lives and therefore they have incentives to figth them
4/ LP tend to prefer conflict solving by ways of jirgas instead of formal justice (I added this one, but I surmised it is relevant)

If those elements are relevant for an intelligence of LP, my impression is that some people may argue that you are precisely manipulating traits related to a tribal description of LP.

Do you think those elements are relevant for an intelligence of LP ?
Do you think they are related to a tribe centric intelligence of LP ?

Your answer would certainly help me in the understanding of your point.

V/R

sir,

thank you for your answer on the description of the Shinwari case.

However, this description is a bit disturbing for my understanding of your point. The shinwari case may be described, in part, by the following elements:
1/ powerbrokers in the area are close relatives
2/ feuds routinely occurs between subgroups of LP
3/ local powerbrokers and LP perceive talibans as threats to their way of organizing their lives and therefore they have incentives to figth them
4/ LP tend to prefer conflict solving by ways of jirgas instead of formal justice (I added this one, but I surmised it is relevant)

If those elements are relevant for an intelligence of LP, my impression is that some people may argue that you are precisely manipulating traits related to a tribal description of LP.

Do you think those elements are relevant for an intelligence of LP ?
Do you think they are related to a tribe centric intelligence of LP ?

Your answer would certainly help me in the understanding of your point.

V/R

Anonymous:

The Shinwari example also demonstrates the very real potential for negative and counterproductive consequences of making decisions based on 'tribal analyses'. Based on what I've read in the media, the Shinwari "tribal pact' was assessed to be an agreement made with the 'shinwari tribe'. It was actually just an agreement made with a few key powerbrokers in one area of Nangarhar province. After it was completed, a conflict broke out between two 'Shinwari' villages that revealed the illusion that was the 'Shinwari Tribe'. That conflict has been widely reported on (Anand Gopal wrote a good piece on it a couple of months ago). Besides the fact that all the subtribes indicate the overall tribe is somewhat irrelevant, 'being Shinwari' doesn't explain anything about people's behavior in Nangarhar province. It doesn't explain why the original powerbroker began fighting the Taliban. It doesn't explain why the violent conflict broke out between the two villages that resulted in the death of over 10 locals. Assuming that it does consistently leads to ineffective decisions at best, and counterproductive ones at worst.

Anonymous:

The Shinwari tribe example is the same as all the rest of the examples of supposed tribal involvement, including those we cited in the article. It can be just as easily understood by taking out all references to tribe and instead employing the much better-understood concepts of region, powerbrokers, etc. In any case of "tribal" involvement, you have local powerbrokers who are able to coerce or negotiate the participation of others. Tribe isn't enough by itself to explain how this coercion or negotiation is possible, and the factors that do explain it--access to resources, imposition of costs, etc.--can do so without any reference to tribe at all. Tribe is at best an insufficient explanation of local behavior. At worst, it's fluff that distracts our attention from those things that are actually relevant.

Local people refer to tribes all the time. People often refer to things that don't really matter. In general, people are lousy at accurately identifying the reasons for their own behavior. The fact that people use tribe to describe or explain a situation is simply not a good enough reason to believe tribe is a useful or accurate.

COL Jones:

In order to "change the perceptions" of the insurgents, in the scenerio I have outlined above, one would think that the local government would have to:

a. Demand the immediate withdrawal of all foreigners -- military and other -- from the soil of their nation and society.

b. Return and, thereafter, refuse all aid, funds and other support received from these foreigners and from those that are like them.

c. Announce that (1) it was no longer the intention of the government to "transform" and "modernize" the society and (2) announce that the focus of the government, hereafter, would be on maintaining the traditional way of life that had sustained the society for all the past centuries.

d. Lastly, and to show the government's convictions in these matters, the government would abandon and destroy all "transforming"/"modernizing" projects and enterprises that had been done in partnership with the foreigners.

Undertaking the actions outlined in "a" - "d" above would certainly re-align the efforts of the government with the interests and desires of those who did not wish to be modernized and transformed (and who had rebelled accordingly). To this populace, government legitimacy has been restored.

The United States, likewise, could change the perceptions of this population group by announcing that it had made a grave mistake in trying to transform and modernize these people against their wishes.

To show its conviction (and to achieve a more-positive image in the eyes of this populace), the United States could announce that it (1) wholehearted supported the government's decisions re: "a"-"d" above, (2) would immediately withdraw its military and civilian forces and (3) would, henceforth, cease trying to effect the transformation and moderization of this society.

Now comes the rub.

a. In bowing to the wishes of those who DID NOT wish to transform (possibly the majority), the United States -- and the local government -- has abandonded that portion of the society (often the minority) that DID wish to join the modern world. This group, now agrieved, may take up an insurgency of its own.

b. Most critically: If the currrent and future interests of the United States are bound up in and dependent upon the successful transformation and modernization of this society (as some/many suggest that it is), then the United States, by supporting the wrong populace (those who do not wish to modernize and transform) will have destroyed itself.

Thus, given the choice of:

a. Commiting suicide or

b. Suppressing an insurgeny whose interests run counter to ours

Then the United States is compelled to make the intelligent choice.

One might suggest that there is an alternative approach that could be used, to wit: to "convince" (without the use of coercion or force) those who do not wish to be modernized and transformed that such (transformation/modernization) was really in their own best interests. But one might suggest that this has already been adequately tried and has proven to be ineffective.

sir,

your article expose interesting venues from cognitive sciences for population intelligence.

Your general argument is that tribe driven intelligence is not the most relevant activity to perform in Afghanistan.

There have been media reports (accuracy ?) about shinwari tribe involvement against Taliban in Momand valley with support from US special forces. How does this information fit with your article conclusions ?

V/R

COL Jones, et al.:

Let's say that we asked the population for their opinion.

And the population stated that they are not interested in and felt threatened by efforts being made by their government to "transform" and "modernize" their society; actions which, they felt, would dramatically alter or destroy their cherished way of life. This, they say, they do not want.

And this population also states that they believe that these efforts being made by their government -- to transform and modernize their society -- that these efforts are not being made at their request, in their behalf or in their best interests but, rather, they feel that these actions are being undertaken, instead:

a. At the behest of foreign powers,

b. On behalf of foreign peoples and

c. For the benefit of a foreign society.

It is for these reason, these populations state, that they have chosen to form an insurgency and rid themselves of their current government(s).

COL Jones: Do you think that representation I have made above also describes instances in which "the government has drifted off course, the population no longer agrees with the destination and the bus driver is no longer headed in the direction in which the people want to go."

If so, do you think that this distinctly different representation of what drives current and past insurgencies (1) may be accurate and (2) has merit in this discussion?

Dear Robert Jones:

Suggest that you should get beyond the angst of not being understood, then setting up a straw-man of your own by "attacking" those who disagree with you.

Remember what Emerson once said, "to be great is to be misunderstood."

With regard to your statement in the above post about your "four causal perceptions" and that they are "relatively timeless and universal" I would suggest to you that the biggest danger for any military organization is the arrogance of knowledge in the belief that its principles are timeless because if they are then why do they need changing, and more important for strategy, what other alternatives to them are there?

Suggest in this regard that you refer back to Clausewitz who says that genius is the one who knows the rules and principles but when necessary can move beyond them. The problem with the American military today in Afghanistan is that we are stuck in the mire of counterinsurgency principles and with no other alternative than to follow them blindly until they are hopefully actualized on the ground.

thanks

gian

An attempt to indicate possible difficulties with the "Jones Model" re: Afghanistan -- and certain other insurgencies -- and to address the idea of "understanding the population." Here goes:

Question #1: What segment(s) of the populace feel that they are excluded from "Good Governance" under the current regime(s) and, thus, are more likely to either support or participate in the insurgency?

Answer #1: Those portions of the population(s) that do not want to see their nations or societies "transformed" and incorporated into the "modern" world -- as per Western instructions and demands -- and as per Western ideas. (In this regard, let us consider such things as MAJ Gant's statement that "the more an alien force tries to change the way a tribe lives, the more that the tribe resists.")

Question #2: How do the above-described groups perceive of "Good Governance?"

Answer #2: For these groups (faced as they are with significant and unwanted foreign interference) "Good Governance" (and legitimacy, justice, respect and hope) are perceived in efforts made by the insurgency -- to defeat the foreigner and his puppet government -- and throw these entities out of their homeland.

Question #3: Are there other groups within these nations and societies that have opposing ideas and views?

Answer #3: Yes. Other segments of the population(s) (those in Afghanistan and elsewhere) believe that "transforming" and "modernizing" -- and accepting Western guidelines and help along these lines -- is in the very best interests of the their nations and their societies. For these groups, "Good Governance" is associated with efforts made the Western-supported government to achieve these goals ("transforming"/"modernizing") and in efforts made to defeat any insurgency that gets in the way.

Rick,

Thanks for your comments. You stated that the article did not show "that tribal affiliation was not an overwhelming constraint for some specific subsets." What more could we have done to show that? We cited several decades of research showing that tribal affiliation had never been shown to constrain behavior, and laid out three challenges that someone would have to answer in order to prove that that research was wrong or incomplete: show that the refuting evidence isn't just a restatement of someone else's opinion, show that the refuting evidence isn't just a restatement of what local people say, and show that the refuting evidence isn't just cherry-picking isolated examples.

If you can provide a counterargument that can withstand all three of those challenges, then we'd be happy to revise our argument.

To focus on tribal affiliation is to waste time and resources on something that doesn't really affect how local people act. There's no reason to continue using it.

I agree with the authors' conclusion that reliance on a single variable will not provide reliably predictable results when dealing with the "Afghan population"; I do not believe they showed that tribal affiliation was not an overwhelming constraint for some specific subsets. And is not that a way to reduce the workload of an overworked force- to identify those areas where with minimal support an exisitng political structure can withstand the insurgency? So to state that it should not ever be used seems to be throwing the baby out with the bath.