Band of brothers at war

By Razib Khan | June 14, 2011 4:01 am

The fruits of human cooperation

ResearchBlogging.orgThe Pith: Human societies can solve the free rider problem, and generate social structure and complexity at a higher level than that of the band. That implies that much of human prehistory may have been characterized by supra-brand structures.

Why cooperation? Why social complexity? Why the ‘problem’ of altruism? These are issues which bubble up at the intersection of ethology and evolution. They also preoccupy thinkers in the social sciences who address fundamental questions. There are perhaps two major dimensions of the parameter space which are useful to consider here: the nature of the relationship between the cooperators, and the scale of the cooperation. An inclusive fitness framework tracks the relation between altruism and genetic relatedness. Reciprocal altruism and tit-for-tat don’t necessarily focus on the genetic relationship between the agents who exchange in mutually beneficial actions. But, in classical models they do tend to focus on dyadic relationships at a small scale.* That is, they’re methodologically individualistic at heart. So all complexity can be reduced to lower orders of organization. In economics a rational choice model of behavior is individualistic, as are the critiques out of behavioral economics.

There are other models which break out of this individualistic box, insofar as they make analogies between organisms at the individual scale to social entities which are aggregations of individuals (e.g., a colony or ethnic group). The society as an organism has an old intellectual pedigree, and was elaborated in great detail by Émile Durkheim. More recently David Sloan Wilson has attempted to resurrect this framework in an explicitly evolutionary sense. Wilson has also been the most vocal proponent of multi-level selection, which posits that the unit of selection can be above the level of the gene or individual. For example, selection operating upon distinctive ‘demes.’ Roughly, a breeding social unit.

There are major theoretical and practical issues with evaluating social units as ‘organisms.’ I will set those aside for now, and shift the focus to humans. I do so because some of those theoretical and practical issues abate when you put the spotlight on higher order cultural structure and variation. In a more technical sense it seems rather obvious that humans have the ability to throw up a large amount of between group ‘memetic’ variance, and maintain that variance, long enough that selection may be able to operate across the two different phenotypes which are homogeneous within group and utterly disjoint across group.

But even if such ‘cultural group selection’ is possible, that does not negate the power of kin, as well as other ‘lower level’ dynamics which may operate at cross-purposes with organismic social units. The biggest problem which comes to mind is the ‘free rider,’ the individual who takes from the benefits accrued to group harmony, but does not put anything into the system and so incur a cost. Over the long term evaluated on the individual scale the free rider is the fit, and therefore the group will become far less effective as its phenotype and genotype wax. This powerful logic is why individualist dynamics are so much more attractive. By simply optimizing fitness through invariant individual behavior you don’t have to confront the specter of the long term futility of the group strategy in the face of self-interested personal tactics.

Yet if you think about it the same problem confronts conventional biological organisms at the scale of the individual. We’re a coalition of disparate cells, some of which even retain their own distinctive genetic lineage (mitochondria). How is the problem of cooperation at this scale solved? If you want a book-length treatment, get Mark Ridely’s The Cooperative Gene: How Mendel’s Demon Explains the Evolution of Complex Beings. But we do have a variety of tactics to stall the ourselves from self-destructing via intra-organismic competition, though in many cases those tactics are futile by the end of your life. I’m referring here to the high probability that you’ll develop cancers, which are basically individual cells whose selfish replicative propensities destroy the useful equilibrium of tissues which help to maintain the integrity of the individual. Over the short to medium term cancerous lines of cells are highly fit, as they spread throughout your body. But over the long term they are self-defeating, insofar as the organism which they parasitize as free riders eventually comes crashing down due to the weight of the stresses which the selfish cells impose on the complex cooperative edifice that is the individual.

Many of these same dynamics have social applicability. In fact the metaphors at the level of cell and tissue derive from older social concepts. So let’s move back to humans. One extreme model of social complexity posits that all the baroque richness of human societies we see today are ad hoc extrapolations and reconfigurations of impulses and instincts which were shaped in an environment of evolutionary adaptedness (EEA) of the hunter-gatherer band. As an example, the idea of meta-ethnic spiritual brotherhood which is common to many ‘higher religions’ is simply an elaboration on our cognitive disposition to think in terms of kinship due to the evolutionary effect of inclusive fitness. Many individual selectionists, most radically George C. Williams, but also Richard Dawkins, seem to posit that human nature is at base positively evil in its selfish intent. Despite Dawkins’ atheism and anti-Christianity I have wondered on occasion if he didn’t have some similarities to a particular sort of reactionary Roman Catholic who took St. Augustine’s theories of original sin too much to heart. Be as that may be, these sorts of individual models generally either imply that social order and complexity are incidental, if valuable, byproducts of proximate instincts, or, social constructions emerge out of phenomena operating at cross-purposes with the stream of evolution (e.g., a complex ideological system constructed from our general intelligence).

This is of course one end of the spectrum. At the other end are a range of broad families of ideas which are group selectionist, or posit a more complex and nested array of dynamics and forces. Williams and his admirers were certainly right to point out the inchoate and woolly nature of much of the ‘survival of the species’ talk which was in the air in the mid-20th century. And, I think talking of taxon level biological selection is something we should do very cautiously if at all. In other words, I accept the general scale independence of evolution. But I do not believe that the 50,000 year experiment of human beings with social complexity is one long extended spandrel. Assuming infinite time for the human experiment to work itself out I can accept that social complexity is due to collapse because of its internal contradictions, but I am but a man alloted a mere few score years, and tend to assent to the proposition that phenomena which span millennia have some right to be accorded the due respect given to the ‘permanent things.’

A new paper in PNAS looks at a society of people who operate in the gray land between ‘small-scale hunter-gatherer bands’ and national entities with all the institutional accoutrements which that entails. The focus of the study are he Turkana. They are a group of Nilotic pastoralists who number between 500,000 and 1 million. They are subdivided into smaller patrilineal units, as well as territorial sections. But the major organizing force among the Turkana in terms of collective action seems to be ‘age group’ cohorts. Basically these are groups of men who come up together as peers. It seems that the Turkana lack institutional religion or formal hereditary leadership. So no kings or warlords of the Turkana who pass their charisma on to the next generation. And the Turkana fight. Or more precisely they raid. As pastoralists they raid for cattle, and they raid for vengeance. Finally, it seems that they do not as a rule raid each other, but rather direct their martial energies outward upon other ethnic groups.

Here’s the abstract, Punishment sustains large-scale cooperation in prestate warfare:

Understanding cooperation and punishment in small-scale societies is crucial for explaining the origins of human cooperation. We studied warfare among the Turkana, a politically uncentralized, egalitarian, nomadic pastoral society in East Africa. Based on a representative sample of 88 recent raids, we show that the Turkana sustain costly cooperation in combat at a remarkably large scale, at least in part, through punishment of free-riders. Raiding parties comprised several hundred warriors and participants are not kin or day-to-day interactants. Warriors incur substantial risk of death and produce collective benefits. Cowardice and desertions occur, and are punished by community-imposed sanctions, including collective corporal punishment and fines. Furthermore, Turkana norms governing warfare benefit the ethnolinguistic group, a population of a half-million people, at the expense of smaller social groupings. These results challenge current views that punishment is unimportant in small-scale societies and that human cooperation evolved in small groups of kin and familiar individuals. Instead, these results suggest that cooperation at the larger scale of ethnolinguistic units enforced by third-party sanctions could have a deep evolutionary history in the human species.

The raw numbers killed proportionally are rather high, but not atypical for many pre-state societies. There are two types of raids. Offensive mass attacks, which seem to be the closest the Turkana and their rivals come to “pitched battle,” and stealth raids with smaller complements of men. I couldn’t but help think of the Cattle Raid of Cooley. Material benefits are real and tangible in many cases, 3 cows per man if victory is theirs. But the costs are real too, the mortality rate is on the order of ~1% per raid. This explains how nearly ~20% of men are dying in their prime years due to violence. Assuming independent probabilities of death you only need 20 raids to have an expected outcome of survival of 0.80. Also, it must be noted that some raids are purely retaliatory and don’t entail any loot, or benefit, to the fighter. These raids of vengeance maintain the honor of the Turkana, and serve as deterrents to future attacks from their enemies. Mass action “tit-for-tat” if you will.

With all the costs and benefits as they are there is naturally free riding. Men beg off on fighting because they can’t find someone to watch their herds, or they’re ill. This might be especially tempting on vengeance raids, where the benefit is a public good which isn’t privately dispersed. Some men avoid being at the tip of the offensive spear during the conflict, and let others take risks so they might live another day. And of course there are stragglers who deviously catch the fleeing cattle first, and secure the best or only portions. If you’ve tread epic myths you know all the varieties of cowardly trickster behavior which might manifest when you are faced with temptations. These raiding parties are numerous, on the order of 250-300 men. They don’t consist of men who are closely related and from the same kin group, but rather a heterogeneous local lot of Turkana, albeit clustered by age group. It seems that the median number of age groups, settlements, and territorial sections, represented in these war parties are around 5 for all of these variables. These war parties are above Dunbar’s number, are not part of some unified group aside from ethnicity and local proximity.

Theory predicts that when you have a diverse lot that diverse interests are going to result in temptation to cheat and let those with whom you’re not close take the fall. How is the problem solved? I’ll quote:

Informally enforced norms allow the Turkana to partially solve the collective action problem in warfare. In 47% of the force raids in which desertions were reported, at least one of the deserters was sanctioned, and in 67% of the force raids in which cowardice was reported, at least one of the cowards was sanctioned (Fig. 7). There are two levels of sanctions. When a warrior’s behavior in a raid deviates from that of his comrades, he is subjected to informal verbal sanctions by his age-mates, women, and seniors. If there is consensus in the community that the act merits more serious sanctions, corporal punishment is initiated. Corporal punishment is severe: the coward or deserter is tied to a tree and beaten by his age-mates. One participant had scars on his torso from being whipped by his age group more than a decade earlier.

This is rather straightforward. In early modern European armies which were involved in set-piece battles there were dragoons stationed at the rear whose role was discourage desertion and retreat through intimidation and force. Obviously the incentive structure here was somewhat different, as defeat in war for a nation-state can have drastic consequences and punishment after the fact may be rendered moot. In the case of these raids documented in this paper it does not seem that the Turkana were involved in existential genocidal conflicts. This may be a function in part of modern norms and the constraining effect of African nation-states in which they’re embedded. Battles between regional warlords in late medieval Europe still occurred, and the monopoly of force accrued to the central government and the monarchy only over time. I would not be surprised if Turkana norms have shifted concomitantly, and non-capital punishment after the raid is an adjustment to the lack of existential urgency in this conflicts.

We know all of the results in this paper in the general verbal sense. How do you fix a free rider problem? You punish them! But the devil is in the details. Here the authors show quantitatively and descriptively that group level dynamics can manifest in a pre-state society above the level of the family band. In fact the unit of organization, the ethno-tribal group, scales up to 500,000 individuals or more! So the social norms were enforced across and beyond kinship groups. Rather it seems that among the Turkana the age groups have a particular power below the level of ethnicity. Presumably what in other contexts might be termed ‘fictive brothers.’ Interestingly these raiding parties were organized and led in an ad hoc and “crowd-sourced” fashion. They illustrate the power of spontaneous dynamics of structured order coming out of a less elaborated and simple social context. And importantly, the violence was directed outward. The rates of murder amongst the Turkana is rather low. Rather, the high risk of death is due to inter-group conflict.

But it seems that the authors are not presenting a simple inter-demic group selection argument. Much of the “action” here operates underneath the level of the group, insofar as group action and cohesiveness is mediated through the regulation of norms of collections of individuals and sub-group entities. This is why I personally find the “group” vs. “individual” dichotomy less than useful. Where do we draw the line from highly elaborated cultural structures built upon atomic units of individual human action to quasi-organismic societies? To a greater extent it seems a matter of taste and convenience, not substance.

One study on the Turkana proves nothing. It may just be part of the bigger puzzle though. For a generation evolutionary psychologists have focused on the model of the hunter-gatherer band during the Pleistocene. Anthropologists working within this tradition have attempted to show that successful hunters and warriors are fecund hunters and warriors. Individual level dynamics then would be validated, as social status is converted into biological currency. From what I have read in the literature (and mind you, I began one theoretically high committed to this hypothesis) the results have been somewhat mixed. This tells us perhaps that one dynamic to explain it all is not going to do the job.

Most of the world’s societies were and are not patrilineal pastoralists. But the Turkana are human, and so they give us a window into the intersection of human psychology and social context, and what that may produce. The intersection is multi-layered, and the product is difficult to distill down to a few broad characterizations. Human social complexity’s raw variety defies broadness of characterization with any economy. But it exists, and it needs explaining, bit by bit.

Citation: Sarah Mathew, & Robert Boyd (2011). Punishment sustains large-scale cooperation in prestate warfare PNAS : 10.1073/pnas.1105604108

* In theory inclusive fitness can obviously be generalized very broadly

Image credit: Wikimedia

CATEGORIZED UNDER: Anthroplogy, Evolution
  • bob sykes

    Very nice.

    There is a related blog over at gnxp reviewing “Braintrust. What Neuroscience Tells Us about Morality”, by Patricia S. Churchland. Could you provide a similar analysis of her book and how it relates to the game theoretic analyses?

  • AG

    By rewarding family of soldier who sacrificed his life, the genes of altruism gain edge over gene of selfishness when poverty is major life limiting factor in survival. Chin law (from terracotta soldiers kingdom) had very detailed rewarding system to promote bravery and punish desertion.

  • rick

    I’m not sure the way to read any of these studies is to be looking for one set of observations to repudiate looking at dynamics from one particular level of granularity or another. Depending on the process at hand, I think there will invariably be some levels of dynamics which are more useful than others.

    Even if, for instance, social status is converted directly into biological currency, this might validate individualistic dynamics, but it certainly doesn’t mean that a different higher-unit of selection couldn’t be a useful encapsulation of the individualistic machinery.

    Your observation that one dynamic to explain all is hopefully just common sense in etho-evolution: I can’t imagine that it’s actually contentious, is it?

  • Razib Khan

    Could you provide a similar analysis of her book and how it relates to the game theoretic analyses

    i don’t review books i haven’t read :-)

  • omar

    “i don’t review books i haven’t read”

    How White of you!

    (I just wanted to use that interesting phrase, no deeper meaning is intended). Nice post.

  • Diogenes

    Beliefs, Symbols may be more immediate mechanisms.
    Even freeloaders reconstruct their cheating as some form retribution. Otherwise they “can’t live with themselves”. Guilt is the other side of “Faith”. Appropriate punishment is often received with relief by the cheater. How irrational is that?
    There’s also the self-destructive behaviour (“Apoptosis”-like) of the self-accusing-unpunished guilt-laden or the belief-deprived; and some individuals propensity for idiosyncratic beliefs, becoming misadapted.

    Think we may have evolved already, as a species, well beyond purely rational cooperating independent agents.
    That’s maybe why we can’t expect reasonable behaviour from ourselves despite how much we try.
    Just aren’t making too many calls at that level anymore.
    Despite people’s beliefs about themselves.

  • Robert Hagedorn

    Adam and Eve, Saint Augustine, and original sin? Do a search: First Scandal.

  • chris y

    Despite Dawkins’ atheism and anti-Christianity I have wondered on occasion if he didn’t have some similarities to a particular sort of reactionary Roman Catholic who took St. Augustine’s theories of original sin too much to heart.

    I would have bet a substantial sum that Dawkins was a lapsed Catholic, but apparently he was raised as an Anglican, who tend to be pretty Augustine-lite.

  • Gav

    Interesting contrast with the Celtic cattle raiders you mentioned in passing. Here you had a strongly hierarchic society with raids (lluydd on this side of the Irish sea, usually translated as hostings) undertaken mainly by war bands (teulu) of high status males (uchelwyr). The contemporary literature, which may not be reliable, suggests that main reward for taking part was renown -”hwy clod na golud” and all that. Although the prospect of cows and other plunder could no doubt be a bonus this probably evened out in the longer term. Free-riders appear to have been punished by shame. (Diogenes mentions guilt. I’m not sure there’s any evidence of that in the literature.)

    Organised cattle raiding (as opposed to simple theft, which still goes on) persisted in parts of Britain probably until the early seventeenth century.

  • Jessica Pellien

    Would be happy to supply a review copy for you of Pat Churchland’s new book, Razib. Can you email me your mailing address?

    Publicist @ Princeton University Press


Discover's Newsletter

Sign up to get the latest science news delivered weekly right to your inbox!

Gene Expression

This blog is about evolution, genetics, genomics and their interstices. Please beware that comments are aggressively moderated. Uncivil or churlish comments will likely get you banned immediately, so make any contribution count!

About Razib Khan

I have degrees in biology and biochemistry, a passion for genetics, history, and philosophy, and shrimp is my favorite food. In relation to nationality I'm a American Northwesterner, in politics I'm a reactionary, and as for religion I have none (I'm an atheist). If you want to know more, see the links at


See More


RSS Razib’s Pinboard

Edifying books

Collapse bottom bar

Login to your Account

E-mail address:
Remember me
Forgot your password?
No problem. Click here to have it e-mailed to you.

Not Registered Yet?

Register now for FREE. Registration only takes a few minutes to complete. Register now »