The origin of war (summary of Cannibals and Kings by Marvin Harris, chapter 4)

Previously: Chapter 2.

I meant to write about Chapter 3 next, which is about the origin of agriculture. Basically, Harris thinks that the origin of agriculture is ultimately a result of the climate change that occured at the end of the Ice Age; but the exact causal sequence is complicated, and it varies depending on which region you’re talking about. I was struggling to write a post about it, so instead I’m going to write about Chapter 4, which is about the origin of warfare. I might write something on Chapter 3 later.

Why do people wage war against each other? In order to start answering this question we first have to understand that warfare is a phenomenon that has varied significantly in its nature across time and space. Different kinds of wars may happen for different reasons. We also have to understand that there is a difference between proximal and distal causes. The First World War happened, in one sense, because of the assassination of Archduke Franz Ferdinand, and yet most people who ask why the First World War started are not satisfied with this answer, because on its own it does not explain why the archduke was assassinated nor why his assassination caused a war. In general, when one asks “why does X happen?” and one receives the answer “because Y” there remain two questions about the cause of X—“why Y?” and “why does Y cause X?”. Y is the more proximal cause here, while the cause of Y and the cause of the fact that Y causes Y are the more distal causes. There is a temptation with questions of causation to think in terms of trying to find a first cause, but there is a sense in which first causes do not exist; everything is caused by something else. Yet it is possible to get to a stage where the remaining distal causes are already understood, or can be taken for granted; at that point you can say that the question has been answered. A lot of literature has been written by historians to try to get us to that point with the question of what caused the First World War. However, I’m not going to discuss this literature very much here, because even the causes identified in this literature are too proximal; they are the causes of one specific war, while the question at hand is the cause of war in general. And the First World War is a very atypical example of a war. In fact, all wars between states are atypical, in a certain sense, so the causes of these wars are not going to be the main focus of our discussion. States have only existed for the last few thousand years. But behaviorally modern humans have existed for at least 30,000 years. For most of that time, all human societies had a foraging mode of substance and were organised into bands1. It is only in the last 10,000 years that societies started to emerge that had a farming mode of subsistence and were organised into tribes, chiefdoms or states, and for much of this time these societies were a minority. So out of all the societies that have ever existed, most have been organised into foraging bands. Therefore, let’s start by examining the causes of warfare between foraging bands specifically.

At this point we have to note that there is some dispute over whether warfare is characteristic of forager societies. The dispute has a long history; Hobbes famously argued that humanity’s “state of nature” was a “war of all against all”, while Rousseau argued for the opposite. Nobody denies that there are forager societies today which wage war. But there are also some forager societies which have never been observed waging war, such as the Andaman Islanders, the Yahgan of Patagonia and the Semai of Malaysia. These are a minority of forager societies today, but some anthropologists believe that during the Palaeolithic, all forager societies were like this, and the modern forager societies which wage war acquired the practice due to contact with farmers. Marvin Harris is not one of those anthropologists. He thinks warfare is a very old phenomenon. The dispute is to some extent politically charged, because those with anti-war inclinations would like to believe that the prospensity towards warfare is not innate to the human species. Of course, even if humanity did have some innate prospensity towards warfare that wouldn’t necessarily mean it couldn’t be suppressed by culture. In any case, the question of how old warfare is as a phenomenon is a factual one, to be settled by facts.

Unfortunately, there is no strong evidence for any stance when it comes to Palaeolithic warfare. We certainly don’t find obvious giveaways, like the walls, towers and moats of Jericho, or the Talheim Death Pit; none of the cave paintings depict warfare. Plenty of skeletons have been found showing signs of trauma, but this is not unambiguous evidence; the trauma could have been caused by hunting accidents, or one-off incidents that were not part of an organised campaign of violence, or it could have been inflicted after death (the Manus people of New Guinea, for example, sever their dead relatives’ heads to use as keepsakes, while the Fore people, also of New Guinea, used to smash holes in their dead relatives’ skulls in order to eat their brains).

Harris admits that warfare was probably less frequent and less deadly in the Palaeolithic. Forager societies are mobile, so there is less need for conflict over territory (although it might still arise when there is no new land to escape to that is sufficiently fertile). There is also less sense of shared identity and hence less potential for xenophobia. Bands are small units, with less than 100 individuals each, so members of different bands have to intermarry to avoid inbreeding, and people often move between bands, perhaps accompanied by relatives within the band, to meet up with relatives outside of the band. Recorded instances of wars between forager bands are perhaps best described not as wars between bands, but as the sum of one-on-one fights between members of different bands who decide to resolve their individual disputes. In order to illustrate this, let’s have a look at one example of such a war that was recorded by C. W. Hart and Arnold Pilling in the late 1920s. This war took place between two bands of Australian aboriginals who lived on the Tiwi Islands in northern Australia, the Tiklauila-Rangwila and Mandiiumbula bands. The original account is not available to me, but here’s Harris’s retelling of it:

The Tiklauila-Rangwila men were the instigators. They painted themselves white, formed a war party and advised the Mandiiumbula of their intentions. A time was set for a meeting. When the two groups had gathered, both sides exchanged a few insults and agreed to meet formally in an open space where there was plenty of room. As night fell … individuals from the two groups exchanged visits, since the war parties included relatives on both sides and no one regarded every member of the other group as an enemy. At dawn the two groups lined up on opposite sides of the clearing. Hostilities began with some old men shouting out their grievances at one another. Two or three individuals were singled out for special attention.

Hence when spears began to be thrown, they were thrown by individuals for reasons based on individual disputes.

Since the old men did most of the spear throwing, marksmanship tended to be highly inaccurate.

Not infrequently the person hit was some innocent non-combatant or one of the screaming old women who weaved through the fighting men, yelling obscenities at everybody, and whose reflexes for dodging spears were not as fast as those of the men … As soon as somebody was wounded, even a seemingly irrelevant old crone, fighting stopped immediately until the implications of this new incident could be assessed by both sides.

Harris thinks that a typical war between forager bands in the Palaeolithic would have been much like this war among the Tiwi people. It would have arisen as a result of individual disputes when people who shared disputes against members of the same band decided to team up with each other. That would have been the proximal cause of these wars. Yet the question remains: why did these disputes need to be resolved by violence?

Don’t get me wrong; I’m not expecting that foragers wouldn’t have resorted to violence when it was advantageous for them. (Although there are forager societies which appear to have had a philosophy of complete non-violence, such as the Moriori of the Chatham Islands; unfortunately such societies tend to be selected against for obvious reasons—look up what happened to the Moriori.) But war between forager bands is immensely costly, perhaps more so for the societies involved than war between states. Remember, forager bands are small. And war disproportionately affects the stronger members of the band, the adult males, who make the biggest contribution to the band’s continued survival. Note also the mention in the above account of the Tiwi war of the fact that the Tiklauila-Rangwila and Mandiiumbula war parties included relatives on both sides. These people were trying to kill their own relatives, and other people who they regularly interacted with. It wasn’t like a war between states were the two parties can dehumanize each other.

Why couldn’t the Tiklauila-Rangwila men have resolved their grievances by engaging in some kind of ritualised mock combat? Of course, we don’t know how serious these grievances were, and killing has a finality which no other method of conflict resolution can rival—but it would be rash to assume that war was the only thing which would work. In fact, this was apparently how the Moriori resolved their conflicts.

Why might this kind of warfare happen? Let’s try and think of some possible reasons.

  1. The need to foster solidarity. By working together to fight an external enemy, groups that wage war increase their internal cohesion and hence their capability for survival.
  2. The need for entertainment. Groups might start wars just because their members enjoy it. (OK, this might sound a little ridiculous to people brought up in the modern Western cultural tradition which tends to hammer home the “war is hell” message, but in other cultures war is glorified. If you haven’t been in a war, how do you know what it’s really like? And if you have been in a war and not enjoyed it, might others have had a different experience?)
  3. The need to satisfy an innate “urge to kill”. Humans, or perhaps just human males specifically, have an instinct which compels them to kill others, which has evolved in the usual way via natural selection; for whatever reason, humans with an inclination towards violence have proved more reproductively successful. Of course it is possible to suppress and moderate this innate drive, just as is possible with other innate drives (people go on hunger strikes, and die from them, for example), but it remains present and manifests itself in the absence of mitigation.
  4. The desire to control more resources.

The last explanation here, that war occurs when political units compete over resources, might strike you as the most sensible one here, and it probably is the main explanation for wars between states. But does it work as an explanation of the Tiwi war? If wars between forager bands are waged for this reason, we would expect wars to often result in one band gaining resources at the expense of the other band. The ethnographic evidence, however, suggests otherwise. Victorious warriors certainly gain social status, and sometimes they gain women which they have kidnapped from the enemy, but often they return only with trophies, such as the severed heads of the men they have slain. After all, these societies do not have the capability to store food or other valuables in large quantities. Territorial expansion is out of the question; bands are mobile, so they lack territory in the first place, and besides they lack the organisational capability to subjugate a population and extract tribute from it. And as we have seen from Chapter 2, forager societies control their population growth, so they would have no need for more Lebensraum most of the time. Of course forager societies would not have always been completely successful at maintaining a constant population, so perhaps wars would occur for reasons of Lebensraum sometimes, but it would not be the main cause of war.

So, perhaps one of the other three explanations given above is more appropriate? Well, let’s have a look at them one-by-one.

Harris’s main problem with the solidarity explanation is that it needs to be accompanied with an explanation of why, out of all the ways a group could increase its sense of togetherness, warfare would be favoured over other methods which do not suffer from the rather significant costs associated with warfare. Is it so difficult to foster intragroup solidarity? In modern Western societies, sport seems to be able to do this to a considerable degree. Maybe it doesn’t work as well as warfare would, but it’s not clear—and warfare doesn’t just need to be better at fostering intragroup solidarity than sport, it needs to be so much better that it is worth the cost. Nobody has been able to show that this is the case, and without that the solidarity explanation has no explanatory power.

As for entertainment: well, I tried to convince you above that it’s not totally obvious that war isn’t fun, but after a careful look at the evidence it seems fair to conclude that war really isn’t fun. Of course societies tend to bombard their members with messages telling them that war is a blast (and not just literally), or, failing that, that killing people somehow makes you a better, more noble person—dulce et decorum est pro patria mori, right? But the very fact that the message has to be hammered home so much kind of gives away that this attitude towards war is far from natural. Humans don’t need these societal messages to get them to do things they enjoy doing; in fact there are some things that humans enjoy very much and do very frequently even when the society they live in is constantly telling them that it makes them a bad person. Everybody knows about the conscientious objectors during the two World Wars. What you might not know is that their equivalents existed even in many pre-state societies which waged war. The Crow Indians, for example, allowed adult males to avoid fighting as long as they dressed themselves in women’s clothing and worked as servants to the warriors. Even the notoriously warlike Yanomamo of the Amazon rainforest have to prepare themselves for battles by taking drugs and performing special rituals. So this explanation, too, is not convincing.

The “killer instinct” explanation suffers from the same problem. If humans have an innate urge to kill people, why is it so hard to get them to do it? In any case, why would such an innate urge be maintained by natural selection in the first place? The cause which the instinct explanation names is too proximal.

So if none of these explanations are satisfactory, what is Harris’s explanation?

I mentioned above how warfare could occur due to the need of bands for Lebensraum after population growth, but this would be a rare occurence. But a related reason why war would be beneficial would be the fact that war helps decrease the average population density in a region. Obviously, the prospect of war is good motivation for bands to try and stay away from each other. In fact, the threat of attack might encourage bands to stay away entirely from certain regions where they would be especially vulnerable. In this way “no man’s lands” are created, and these might be crucial to ensuring that foragers do not deplete the supply of the animals and plants that they consume.

But Harris doesn’t think these are the only benefits of war. He also claims that war helped foragers to control their population growth. On the face of it, this claim might seem strange because the direct effect of war is mainly to reduce the number of adult males in the population. But as long as the amount of adult women remains more or less constant, the rate of population growth will remain the same, especially in openly polygynous societies. (Not all forager societies are polygynous, but the more warlike societies tend to be more polygynous, according to Harris.)

Harris’s answer is that warfare is part of a kind of package of practices which mutually reinforce each other and serve to control population growth, so that forager populations remain within the carrying capacity of their environment. These three practices are female infanticide, warfare and male supremacy (i.e., patriarchy).

Female infanticide, of course, directly limits the rate of population growth. That was the topic of Chapter 2. But female infanticide is costly for the mother. Not necessarily in moral terms, because cultures vary greatly in how they view the morality of infanticide. But pregnancy isn’t easy, and it is not hard to see why mothers would be reluctant to kill their newborn babies after going through all that effort. And it’s not just mothers who suffer the costs, because the whole band has to work together to supply the extra food needed to feed the developing baby (or the newborn baby, since infanticide can occur by neglect, not just by direct killing).

Harris proposes (as we’ll see when we get to the next chapter) that foragers’ reliance on the strongest members of their societies, the adult males, for military purposes is the cause of the development of male supremacist institutions and practices. One of these male supremacist practises is the favouring of male babies over female ones, and this encourages female infanticide. Female infanticide would be favoured to some extent already because of its effect on population growth, but male supremacy makes it possible for it to be favoured to a greater extent, resulting in a greater effect on population growth.

Even if you don’t find this causal link illuminating (it is elaborated upon in the next chapter), Harris points to evidence that it holds up. William Divale studied a number of band and tribe societies for which census records were available covering the time when the society was pacified by the occupying state. He found that the ratio of boys less than 14 years old to girls in the same age bracket was significantly higher (128:100, on average) before pacification than after it (the average figure for societies less than 25 years post-pacification was 113:100, and after 25 years it dropped to 106:100, more or less the global average, which was 105:100 when Harris wrote the book but is 107:100 today). The figures are restricted to this age bracket for the obvious reason that when these boys grew up, many of them were killed in battle, so that the sex ratio among the adults in the unpacified societies was actually more or less exactly equal, and it actually became more skewed towards boys after pacification.

So, in summary: according to Marvin Harris, the practice of warfare facilitates the survival of forager societies because it encourages the dispersal of bands and the creation of no man’s lands where the animals and plants that they feed on can take refuge, and also because it facilitates the development of male supremacist institutions and practices; in particular, it encourages female infanticide, which limits population growth and thereby prevents the resources in the environment from being depleted.


  1. ^ Anthropologists classify societies via their extent of organisation into four types, going from least to most complex: bands, tribes, chiefdoms and states.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s