The Learning Curve
By Mark Safranski—aka “Zenpundit”
Steven Pressfield invited me to do a guest post here at “Tribes” and give my assessment of the vigorous debate that greeted the entry of “It’s the Tribes, Stupid: War & Reality in Afghanistan” into the blogosphere. Or, at least the corner of the blogosphere that is concerned with COIN, military affairs, foreign policy, terrorism, Afghanistan and Iraq. The following opinion is my own and does not necessarily reflect that of Mr. Pressfield.
This “Tribes” blog attracted an unusual amount of attention for a new blog primarily for three reasons:
The blogger, Steven Pressfield, is a celebrity, with an established audience who enjoy his novels.
The blog enunciates a general theme or meta-concept—“Tribalism”—and applies it to a complex and politically controversial war. Moreover, Pressfield applied the concept of “tribalism” in a way that contradicts academic usage as articulated by many subject matter experts.
The blog, in a technical sense, is very well done with the video blogging episodes constituting a powerful, psychological “hook” in terms of attention economy incentives.
Unsurprisingly, fireworks ensued.
With any activity, there is a “learning curve” and while bloggers usually make their maiden efforts in virtual anonymity, Steve jumped in to the deep end of the pool and consequently, took some lumps with criticism that, fair or foul, was entirely in keeping with the rough and tumble nature of blogging. As with every blogger before him who raised a ruckus, Steve reflected in the aftermath and moved up the learning curve.
That was process, but there is also an issue of substance: the question of “Tribalism” itself.
There was enthusiastic praise for ‘Tribes”, naturally, but the criticism was equally as strong because Pressfield’s theme of tribalism as a general explanatory model is a powerfully attractive one. Too attractive, in the view of subject matter experts (SME) who drill down to a very granular level of detail and see all of the particularistic caveats or limitations of tribalism that exist in a given society. Tribalism among the ancient Gauls was not a carbon copy of 21st century Afghanistan, the artificial kinship network of the Yakuza or Shaka Zulu’s Impi formations. Yet, some similarities or congruencies remain even among such historically diverse examples because a tribe is a durable social network. In terms of resilience, a tribe may be the most adaptive and secure social structure of all.
Social Science SME’s are, in my experience, far more uncomfortable with explanations that cut across their disciplinary boundaries than are their counterparts in the hard sciences. Furthermore, arguments that are predicated on psychosocial-cultural premises, like Ruth Benedict’s Chrysanthemum and the Sword or Raphael Patai’s The Arab Mind are especially suspect in today’s politically correct academic culture, not for their flaws (both works have serious flaws) but for their approach of constructing generalizations, which inevitably come laden with implied or explicit value-judgments. Because generalizations apply best at a panoramic level and become increasingly less relevant as you get down to the gritty details where other variables conflict or interact, SME seize on these caveats to justify throwing the baby out with the bathwater to avoid having to deal with the politically messy aspects that can easily derail academic careers.
Well, that’s not right, analytically speaking. The limitations, conflicts and contradictions are not usually categorical refutations of the proposed generalization, in this instance tribalism, but rather critically important feedback to understanding the complexity of the phenomena as it applies or fails to apply to a specific scenario. To reject either the exceptions and limitations or the generalization itself out of hand is to stop thinking about ideas and to begin chanting an ideology. Weighing the factors with as much intellectual honesty and analytical objectivity as you can muster, reorienting your views in light of empirical evidence and constructing a synthesis, is how you move up the learning curve.
Blogging is not a journal article or a book, formal and frozen in time. It should be a dynamic conversation, a learning curve for all involved with the understanding that all the participants engage and leave the conversation at different levels of understanding and views should change. One of Steve’s more controversial original points was the supremacy of the tribal mindset over Islamist radicalism or jihadist theology. There’s a serious conflict between the two, as the works of Olivier Roy and Gilles Kepel have detailed, but there is also, both in societies and in individuals, degrees of coexistence. Very seldom is something as complex as a social system reducible to an either-or equation. Steve went on to address the juxtaposition of tribalism and radicalism in his recent post, an example of how the blogosphere can move as a conversation instead of as an echo chamber.
Therefore, I welcome Steven Pressfield and “Tribes” to the conversation without end that is the blogosphere, and look forward to watching his blog evolve over time. I don’t expect to agree with him all of the time or him with me, but I know that I will learn something from the give-and-take as we all try to move up the learning curve
Zenpundit, you seem like a well educated individual. You are able to use a lot of words…that don’t seem to move anywhere, at least for me. I maybe more pragmatic in my approach or thought process than you (I do not mean this as a criticism of either of us just an observation). I strive for workable concepts…things you can apply, that get results. I believe Steven Pressfield has presented a workable concept that can be applied to folks with “boots on the ground”. In fact most of what I read from these brave individuals compliments what Steven has presented. I agree with your statement that most social sciences are uncomfortabel with discipline crossover, hence they want to nit pick any working theory due to some generalization it my present or politically incorrect implications it may carry. However, when your life is in danger, how much politically correctness do you care about? You need a working theory to at least get you in the door with the headman then figure out his unique situation.
I can be wordy, point taken. Sometimes clear language is better so let me try again.
From reading your comment, I do not think that we disagree very much about what Steven Pressfield is doing here. I think it is aimed at people, especially service personnel, who have a need for a practical rather than an academic understanding of tribalism and trying to provide that information is a good thing. My points were basically:
A) More academic critics are coming at Steve’s blog at a level of detailed knowledge that is not digestible to someone reading about tribalism for the first time.
B) Blogospheric debate, when civil, helps all of us learn more and refine our ideas – it is a valuable exchange.
Wisner or Mark , what exactly is the theory Steven is proposing? What are its testable claims? What predictions does it make about Afghans’ behaviors?
At best, I understand this to be a descriptive model, and one that is hopelessly broad. When we start making the argument that we’ve all got a little tribesman in us, we should probably realize that we’ve hit on a category with weak explanatory power — it’s far too broad a category — and that “tribe” probably describes informal networks all humans create to deal with insecurity and uncertainty and that there is probably an inverse relationship between security in society outside the netowrk and the strength of bonds in these networks. The implication we should take away about Afghans isn’t that we are civilized like Alexander’s shining armies and that they are base tribes like ancient Bactrians, but instead that Afghans are people dealing with insecurity the way that we Westerners might — very similar, to name one example, to the way that Missourians responded to violence and insecurity during the Civil War.
I’m very keen on the need for concepts and theories that can help us understand and more intelligently interact with culturally diffferent societies. But we need to be sure that these concepts and theories have actual utility. I’ll avoid being too granular, but suffice it to say that Afghan society does not work in the stereotypically classical tribal sense. (Mark, this is where I think you’re being dead wrong. If we say in COIN theory that we should know the population, we shouldn’t stop halfway with a nice theory that doesn’t have sufficient predictive or explanatory power because of an aversion to academic particularism.) Coincidentally, some colleagues and I were recently trying to turn up academics who specialize in Afghanistan who say that tribe is the critical or even very useful factor for understanding how Afghan society organizes and behaves. We can’t find a single one. (That’s not to reinforce my point; I’d love examples if people have them.)
As a final comment,
Mark, so what? This is a huge pet peeve of mine. I know I fall into that category, but from where I sit, I see neither interest nor inclination to engage or respond to these criticisms.
I should clarify regarding the citations for which I am looking. I am interested in finding the tribal argument being put forth by Afghan specialists with field research experience — people who put together a research plan and can describe their methodology. Again, this is not a rhetorical request. I really want to know if I’m missing something.
The rhetorical side of this is that the academic opinion is unanimous that tribe should not be the central unit of social analysis in Afghanistan. It is irresponsible for the generalist to dismiss this opinion for being too granular or because the theory is “neat.” This literature is either right or wrong. It is not even being discussed here.
We were just made aware that a few people have had problems posting to the comments section. The following is being posted at the request of Zenpundit:
Alexander’s armies had quite a few Persians…..but they were probably shiny, moreso than the Macedonians toward the end.
Good to have you here. For Steve’s readers who may not be familiar with Mr. Hamm or Registan, Nathan has been an important voice on Central Asian affairs in the blogosphere for years on a number of respected regional sites and has extensive experience living in the region.
Let me try to address your concerns in reverse order:
The latter statement has to be addressed by Steven Pressfield rather than me. On the other part, as a learning aspect, when SME are writing to the uninitiated, there’s often a too large assumption about what the laymen know and a tendency to bring an overloading amount of complexity to the discussion. I am guilty of this myself at times when teaching or writing about my research interests. Pressfield is probably not writing for a typical reader at Registan but his readers may become interested enough in Afghanistan or tribalism that they may start reading articles, books and sites like yours as a result. Where you see a static end-state, I see a gateway or a hook.
Richard Tapper has written on the negotiation of identity, with one of the major components being “qaum”, which if I recall has (or can have) a loose “tribal” meaning. I’m not qualified to rate experts in your field Nathan, but Nojumi describes the Parcham-Khalq Communists in Kabul thinking the tribes were important enough to warrant sending out the meddling Marxist officials to their villages ( incidentally, the Soviet advisers had cautioned the Taraki regime against it). Flipping through Ewans’ Afghanistan: A Short history, the tribes are present as at least a background political factor from Ahmed Shah Durrani to the fall of the Taliban. Here’s an analysis of warlordism and tribes in Afghanistan by Antonio Giustozzi and Noor Ullah (2006):
I suppose point in the argument hinges on what you mean by “critical” or “useful”. That Afghanistan (or any society) is far more complex than one variable, is something I’ll agree with but for an “unimportant” factor, tribal structures in Afghanistan seem to enjoy considerable longevity.
First I am not suggesting we stop halfway. I think that you and Josh fear that will happen with some readers. It will happen with some of them, you’re right. I’m more interested in those readers who are inspired to go further and keep learning.
I think also, on a methodological point regarding Social Science. “Predictive” is a high bar more suitable for hard science that can have appropriate experimental controls. For SS, I’d use “descriptive”, “speculative” and perhaps at best “probabilistic” analysis.
Tribes are a type of network structure and they can be artificial (social, legal, political) as well as being based on lineage. Most historical lineage tribes had provisions for adopting new members who were unrelated by means other than marriage ( though that was the most convenient device). Within sufficiently large tribes you can have both weak and strong ties or even other kinds of network structures present ( modular, hierarchy, scale-free etc). Network analysis is a useful tool for examining how people seek security and advantage within a group.
Being a long time advocate of horizontal thinking, I like broad comparisons and recognition of patterns and congruencies. They give us data that compartmentalizing, isolating and drilling down often does not ( those are useful tools as well. Granularity is a good thing -it is just not the only thing).
Thanks for the post. And I’m sympathetic to the idea of making some of these concepts—especially in a uniquely complex place like Afghanistan—understandable to normal people. Hell, I’m even a big fan of trying to craft large, cross-cutting theories onto the world.
That’s my problem here. There is substantive academic consensus that, though existent, tribe is not a useful analytical construct for Afghanistan. Which, to my discredit, I did not explain in my curiously unmentioned critiques of Pressfield’s project. No one would ever deny that there is such a thing as tribe in Afghanistan. But is it important?
Richard Tapper, one of the world’s foremost experts on tribalism, wrote specifically that “tribe” is not a useful analytical construct for understanding Afghanistan… in 1983. More recently, Berndt Glatzer in 2001 (but before 9/11!) wrote that tribe is only one of the many salient identities Afghans call upon depending on the context. He broke down most important identities to local ones, which he took pains to distinguish from the tribal identity.
In your own criticism of Pressfield’s fundamental thesis, you noted that he discounts religion as an organizing factor. Pressfied responded by saying religion is tribal. By that definition, all social groupings are tribal—as Nathan said, it becomes defined so broadly as to be meaningless.
Here’s where I scratch my head: rather than discussing how people reacted to the tenets of this theory, including those pesky experts with their “facts” and “details,” something which can be done, especially if you’re talking about the bloggy learing curve, you’ve decided to write that actual experts on the topic are wrong to note that the theory does not have explanatory power.
As a man who prides himself on such things, how does that wash?
Zenpundit, points taken. About point A… I guess we should make a distinction between pure academic exercise (which in any field tends to depart from practical, measurable results over time) and a workable, applicable, practical, measurable over time, evolving with more experience on the ground, “generalization”.
I find it interesting that academic opinion and expert analysis have found that tribalism should not be the central unit in social analysis in Afghanistan. Yet, in Pakistan the areas ungovernable by central authority because of the strong tribal order are called the Tribal Areas, Tribal Frontier etc… So, while in Pakistan its called tribalism yet when they cross the Hindu Kush it is something else…according to the experts. That is academic exercise at its finest!
Wisner, without using the words “tribe,” “tribal,” or “tribalism,” please describe the social and political structure of FATA and NWFP. This is the issue I’m trying to get at. We might as well say that Pashtun society is “hooflehopplist.” Without a clear defintion of the phenomenon, it’s just as useless. Also, please point me in the direction of “strong tribal order” in FATA or NWFP. There are plenty of places in which one finds plenty of insularity, conservatism, and hostility toward outside rule, but from where I sit, there seems to be as much intra-tribal disunity and factionalism as in Afghanistan.
Those areas on the other side of the Durand Line are governed, though differently and far more indirectly than the rest of Pakistan. The government of Pakistan adopted the British imperial approach to governance per the Frontier Crime Regulation. One of the results of the British and Pakistani approach to governance of these areas is the institutionalization and formalization of certain aspects of Pakistani Pashtun tribal society. Meanwhile, in Afghanistan, the last century, and especially the last several decades, these same aspects of Afghan Pashtun tribal society have been under assault. So, yes, there are some differences. State-tribe interactions and different histories matter.
I’ll clarify a little more. Tribalism — a very specific form of it — is an important part of the social fabric in Afghanistan and Pakistan. It is not, however, the alpha or omega, let alone both.
Nathan, your use of acronym and academic muscle flexing is impressive, sincerely. Yet it falls into the category of academic exercise or at least argument for agument’s sake (which isnt’ bad if both parties know what they are engaged in). Where does all of this get you? How does your definition further the sfety, security and effectiveness of our troops? Does it help the young Marine or seasoned SF vet get in the door to the headman in the village? Will it help the American at home understand the cultural and societal differences and difficulties we face? What I feel Steven Pressfield has done is provide a basic, working context from which to understand the Afghan (I hope this doesn’t offend our host). Once you have this in place you can then build your specific context to the specific area of operation. After all isn’t this what the experts would do if they were doing field research or were embedded with our forces? I think you will have to break down your second paragraph more because I think you may have just made my point (I could be wrong so feel free to clarify).
Again, I think there is a difference in where we are both coming from and how we are interpreting Steven’s blog/work. I’m taking a more utilitarian approach.
Wisner, my concern is entirely about utility. Mischaracterization of the situation leads directly to bad policy. I want to know what tribe actually means because it in turn defines how our Soldiers and Marines engage the Afghan population. The problem I’m seeing throughout these posts is that the goalposts move; tribe ends up explaining anything we want it to explain.
Do tribes have structure and leaders? If so, then we need to find out who they are so that we can engage them. To answer the question for Afghanistan, they usually don’t. At the local level there are influential men and a lot of the time, very influential men who are concealed from outsiders. In other areas, influence and authority resides with religious figures. Either way, local influentials open the door to engaging the population. We take a shortcut and go to someone akin to a sheikh to strike a deal with a large tribe. Instead, we have to go village by village.
If you want a competing understanding of Afghans, I propose it to be that Afghans organize locally and have local concerns regardless of ethnicity, language, or sect.
Wisener, that’s like complaining using “VA” for “Virginia” is elitist and academic. It’s a silly argument.
And Nathan’s point matters tremendously. If we cannot describe Afghan culture or social organization, then we cannot reliably influence or persuade it to win the war. It is a fundamental step: the Intelligence Preparation of the Battlefield. If your intel, the context on which you craft plans, is wrong, then your plans are wrong, or worse still: counterproductive.
The point both Nathan and I have been trying to explain is that reducing everything to “tribe” does not accurately describe Afghanistan or the tribal areas of Pakistan. Thus, making plans that affect the soldier based on a mischaracterization will lead to everyone—”tribesman” and solider alike—being worse off, with significantly more dead that would otherwise be necessary.
Let’s start with this:
That even assumes there is a headman in a village (and that there are discrete communities called “villages”). Our point is that a proper understanding of Afghanistan demonstrates these are the wrong questions to ask because they don’t actually tell you anything.
Oh, and the Hindu Kush, despite what Robert Kaplan says, are not between Afghanistan and Pakistan. Again, the part about knowing what you’re discussing before you start discussing it. Rather than writing off detailed knowledge of these places as being “academic,” “theoretical” or “heavily bogged in the details,” there COULD be an effort to actually understand them. As it is right now, complaining that expertise is hard does not actually help anyone.
“Does it help the young Marine or seasoned SF vet get in the door to the headman in the village?”
Bu that’s the whole point of the controversy, the good/bad tribe theory doesn’t significantly help understand what’s happening on the ground. We’re not helping a young marine by giving him broad theories devoid of practical use. Really, how does this help a marine?
Nathan, well written…”Afghans organize locally and have local concerns regardless of ethnicity, language, or sect.”…yes (although some will argue the part that comes after regardless), with the underlying Tribal Mindset (tribal mindset>which is what I was agreeing with or defending) . It seems tribalism is a term that needs to be categorized into the uses in which it is being proffered on this blog (mindset or structure).
Joshua Foust, whoa, so condescending. It is easy to critque but more difficult to present. I was attempting to get Nathan to present his idea..which he did in fine form. Again, I was upholding my belief that the Tribal Mindset Mr Pressfield presents is a fine working model. This model can be expanded upon where the specific area of operation reveals more detail. I still believe that his ideas on the Tribal Mindset help Americans et.al understand the Afghan. I believe that the Intelligence Preparation of the Battlefield begins with the mindset of the people and their social organization. Good luck with moving that mountain range.
I, a large part of the controversy is applying what Steven says as the Tribal Mindset and then arguing the Tribal Structure. Generalizations have become taboo in Western societies due to the emphisis on the individual. It does not mean they are not practical. How about presenting a competing opinion to Steven’s on the Tribal Mindset?
Wisener, I’ve been doing that. Pardon the frustration, but Mark’s entire post was to argue that people like me are too bogged down in “details,” when what we’re doing is explaining a more difficult, but more appropriate, framework.
You keep using the word “tribal mindset” and “tribesman.” Since Pressfield keeps changing the definition of what that means (most recently to include religion), could you tell me what it is? Because so far the discussion on his end hasn’t rested on any settled or established ontology.
As for the alternate models… I’m sorry, Pressfield has refused to link to my blog posts where I have previous laid out this model. Here they are.
The Myth of Taliban Tribalism
Breaking the Tribal Model
Misinterpreting ‘Tribal’ Sabotage
‘Tribal’ Engagement, or How We Lost Kapisa Are Are (Slowly) Regaining It
I’d also suggest reading these pieces, which are easily found either on the Internet or Amazon, which explain the idea much more comprehensively.
Glatzer, Bernt (2001). “War and Boundaries in Afghanistan: Significance and Relativity of Local and Social Boundaries.” Weld des Islams, 41, 3, pp. 379-99
Glatzer, Bernt (2002). “The Pashtun Tribal System.” in Pfeffer, G., and D.K. Behra (eds.), Concept of Tribal Society (Contemporary Society: Tribal Studies, Vol 5), pp. 265-282.
Giustozzi, Antonio (2007). Koran, Kalashnikov and Laptop: The Neo-Taliban Insurgency in Afghanistan 2002-2007.
Shahrani, Nazif (2002). “Factionalism, and the State in Afghanistan,” American Anthropologist, Vol. 104, No. 3.
That’s where you can see the specific reasons to drop the idea of a “tribal mindset” and a “tribal society” and start to see Afghanistan on its own terms, without any baggage we impose on it through these terms.
Part of the point I hope to communicate is what keeps getting called “the tribal mindset” is really a very fundamentally human mindset in situations of insecurity and disconnectedness. It is more or less “the local mindset.” Where I get concerned about the use of the word “tribe” is that it implies hierarchy, leadership, rigid social forms, and intense loyalties to one’s group that we don’t really see in Afghanistan. I understand the desire to have a general form that one can adjust during the IPB process for a specific AO. But, because features implied by the poorly defined concept of “the tribal mindset” are not present almost anywhere in Afghanistan, I suggest it has no utility.
Nathan, I can see your point. Do you concede this…Afghanistan and its neighbors to an extent have been living in situations of insecurity and disconectedness now for hundreds of years? If so would this not embed them with the “tribal mindset” or the “local mindset”? How does a Westerner relate to this? The word “tribe” certainly carries with it some baggage in some circles. But, while we argue over the proper term, may I ask how you feel about the body of ideas presented by Steven? If you were to throw out the word “tribe” would you agree with him?
Joshua, would it suprise you to know that difficult concepts do not thwart my grasping them and that I have read a couple of the refrences you have presented (albeit some time ago)? Since you have gone to the trouble of listing others I have not read, I will endeavor to read them. I will tell you this about my experience with academians…(I do not know you so I am not applying this to you although you seem to have willingly taken the offense from Mark’s comment…again, not applied to you personally). They have a lot at stake if they have come to a definition or concept first. If it gets proven wrong then they have wasted their time, maybe lost a bit of self esteem and someone else’s money. So, they will defend their definition and concept to the death. This seems to be more the case with the social sciences than with the physical sciences (although I know some physicists that are quite territorial). I will pose the same question to you as I have to Nathan. If it weren’t for the word “tribal” would you agree in sum or in part with Steven?
I don’t think I would agree because really, without the word “tribe,” I don’t think there really is much of an argument. If “tribalism” is no longer the enemy, what is? Small town insularism? In my opinion, a lot of the description of Afghan society is filled to the brim with inaccurate connotations that the word “tribe” carries.
Honestly, I don’t think that the way that Westerners deal with whatever this concept is needs to be too complicated. We need to keep in mind a few things: People care about local concerns; We aren’t trusted and can only build trust through repeated contacts and following through on promises; Leaders don’t dictate, they influence; Our actions can create winners and losers. And a really key way to avoiding a lot of problems is to arm oneself with lots and lots of local knowledge. I’m sure there are more, and I also think we could simplify this down to “Be nice, Be respectful, Be honest, and Drink a lot of tea.”
In regard to your other question, I agree and disagree. Certainly many generations of a condition leads to some kind of social embeddedness. However, I think cultural practices get more deeply rooted than social structures, which were already changing in some fairly drastic ways in Afghanistan prior to 1979.
I don’t know, I just find it strange that people who’ve read books (books written by folks with language training, training in understanding society, and years of living in Afghan villages) about Afghanistan think that tribes aren’t important; and then there are people like Zenpundit, Wisner, and Pressfield who think, oh yes, tribes are important, they just know it!
I know who I’d trust.
“Nathan, I can see your point. Do you concede this…Afghanistan and its neighbors to an extent have been living in situations of insecurity and disconnectedness now for hundreds of years?”
Concede Afghanistan and its neighbors have been living in situations of insecurity and disconnectedness now for hundreds of years? I suppose you would have to define that which they are disconnected from. We all tend to orient (internally connect and isolate as to an advantage) ourselves to that which we observe, and much that we observe is our local environment. At times in this modern world, I don’t feel very “connect” to my environment, local or otherwise, but it is hard for me to imagine that the Afghans and neighbors are disconnected from theirs. In fact, I imagine that their system of social “connective-ness” (tribes, religions, ethnic identity) is strong because of the environment they live in. On the other hand, they may be disconnected from other environments, which must be what you are getting at.
Even after reading “Kite runner”, it is also hard for me to think of Afghanistan in the terms of “insecure”. There are elements that are really messed-up inside the society, but that must be true for every society, even one so much connected as ours (USA). Go back and ask Alexander how insecure those Afghans are. What did the Afghans tell him, you have never dealt with our poverty before? I don’t know if the poverty is from the land or the people, but I imagine it is from the environment they live in. Bring an element willing to pay (Taliban, Nike) and perhaps you could start your own religion or shoe factory, but which would the environment support? Would the Afghans be willing to work in a shoe factory as entry into globalization or let the Taliban into a neighborhood with promises of support? The answer to both questions could be yes.
However, let’s talk environment. What does the environment support? Does the environment support entryway into globalization (by bringing in near slave-wages jobs) or does it support China’s effort in building an oil pipeline through the country, because that is the real connective-ness we are talking here.
The advantage of talking tribes is that the USA knows how to deal with tribes. I imagine that the US military has a highly documented and successful strategy in dealing with tribes; my guess is you start by building forts inside the environment of the tribes.
Instead of talking tribes we should talk orientations, at least on the academic level. The right question might be: how are they oriented to their environment? How have these people isolate their decision making to give themselves an advantage inside the environment they observe. Then we could ask ourselves: how is their orientation a threat to us? We know tribes.