Saturday, 2 March 2013

Violence in Pre-Colonial Amazonia

     I've recently been reading Comparative Arawakan Histories, a collection of some excellent essays on Arawak societies and their cultural histories edited by Jonathan Hill and Fernando Santos-Granero. Arawak (or 'Maipurean') is a language family of South America and the Antilles, and Arawak languages produced many of the words we use in English for American products and ideas (maize, canoe, hurricane, cannibal, etc). Arawak-speaking groups appear to show remarkable consistencies in social structure and what we might term 'culture'.

      In particular, hierarchically-arranged chiefdoms and the deliberate disavowal of endo-warfare (that is to say, warfare within the group) and feuding, as well as the preference for creating strong alliances with other groups (especially other Arawak-speaking groups), mark out Arawak populations across the continent. Arawak languages are found from the Andean foothills in southern Amazonia through to the Caribbean, and these principles, with changes, are recurrent throughout the language area.
      Due to a common Arawak preference for limiting endo-warfare, many of the contributors in the volume draw attention to the differences in warfare among different South American indigenous societies.  Recently, there has been a lot of discussion about Napoleon Chagnon and the supposed misrepresentation of Amazonian societies as 'fierce' or 'warlike'.  Stephen Corry has made his traditional claim again that indigenous Amazonian societies are no more nor less violent than other societies.  Chagnon's view that the Yanomami (whose language, by the way, is in a family more-or-less of its own) are or were violent in comparison to modern British or American people has come into question on a number of grounds, few of them empirical.  I couldn't help but read the book in the context of these discussions.

        It is quite true that Amazonian societies of the past were made up of humans whose genetic material was/is largely the same as mine or yours, and that they were/are human beings in every way.  It is also true that Amazonian societies of the past were much more violent than they are today.  This change was due, in part, to cruel and vindictive 'pacification' programmes by colonial and state governments, I think, but this dubious origin shouldn't blind us to the phenomenon.  Amazonian societies were predominantly tribal or based on chiefdoms with relatively weak power structures.  This is not conducive to the suppression of violence.

       Amazonian warfare appears to have been endemic and violent, with several traditions of mutilation and cannibalism present throughout.  There was considerable diversity in this, as in every other sphere of human life in Amazonia, but one of the persistent features is the endemic nature of war.

       Panoan-speaking Conibo-Shipibo societies of the Ucayali region were so dependent on war that their women, traditionally weavers, abandoned the craft upon realising that theft and war were reliable producers of woven products.  The men acquired wives from other groups primarily through warfare, resulting in various sorts of polygamous arrangements.  They also dried the heads and disembodied hearts of their victims and hung them from the rafters of their houses as symbols of their strength.  It is also notable that the elderly, women, and children were all fair game, just as they were for the Piro, an Arawakan-speaking group whose pottery closely resembles the Conibo-Shipibo styles despite their speaking an Arawak language (indicating significant 'Panoanisation' of an Arawakan group).*  In northwest Amazonia, Baniwa warriors 'decapitated their vanquished enemies to prevent mystical attacks from the dead [...], whereas the Cubeo wore the smoked genitals of killed warriors over their own as a war trophy' (Comparative Arawak Histories, pg.36).**

        Both Arawakan and Panoan-speaking communities on the eastern slopes of the Andes, in the Peruvian Amazon, practiced cannibalism, which was also common among Tupi-Guarani-speaking groups (famously including Tupinamba people of the Brazilian coast, on the other side of the continent, whose cannibalistic proclivities were documented in the sixteenth century by Hans Staden).  Cannibalism was once present throughout Amazonia (albeit in circumscribed contexts), serving a variety of functions, almost none of them nutritive.  I believe it was primarily a vindictive act, although its origins in the region are obscured by its popularity throughout Amazonia.  Slave-raiding was also commonplace, seemingly long before Europeans arrived in force (although it is difficult to find evidence for this in the archaeological record, of course).

       These are not random examples of bizarre practices that prove that violence was all-pervasive.  They are simply graphic examples of a very common trend: endemic warfare and the rise of warriors whose experience gave them credibility and legitimacy.

       Warfare was important in all Arawak societies.  It was once common to argue that Arawak societies were inherently more peaceful than those of their neighbours, but this isn't accurate.  In reality, Arawak societies were good at suppressing feuds and other minor issues within the community, an attribute likely related to the hierarchical orientation of Arawak societies and the common acceptance of chiefly power (sometimes in divine form).  But warfare was part of the way in which peace was maintained internally; chiefs whose powers were respected within certain villages would predicate their claims to legitimacy on success in war.  Slave-raiding was a part of this as well.  The difference in orientation between Arawak groups and others with regard to warfare - noted long ago in the Caribbean by Europeans - wasn't about the total removal of violence among Arawaks, but rather the different social emphasis on killing people.  Arawaks did it outside the village and outside of the alliance.  Caribs (n.b., 'Carib' is a very confused term; wiki gives some indication of its problems), Panoans, and others, did it in a wider variety of social situations.

      The Yanomami, although linguistically unrelated to other groups in Amazonia, were not bucking a trend by conducting violent raids on one another.  They were not unique within Amazonia for the brutality of their activities.  Chagnon's claims that the Yanomami were 'fierce' should be uncontroversial in this context, and we shouldn't really be questioning whether he was right about the level of warfare in earlier Yanomami societies, unless we have very good reasons for believing that he was wrong.  Chagnon's views were/are flawed primarily in labelling the entire Yanomami community as 'fierce', and also in analysing Yanomami society as if it represented some earliest, primeval example of human society.  Jon Marks has written of Chagnon's scientific flaws in an excellent article here; it seems that Chagnon's attempts at explaining Yanomami violence were flawed right from the start.  He wasn't wrong about the presence of violence itself, however.

       None of this means, of course, that Amazonian societies deserve misery and punishment, or that pre-colonial violence in the region justifies punitive expeditions by colonial and modern governments.  I merely think we shouldn't be afraid of the truth: that most non-state, 'tribal', people around the world do very often live in a state of endemic warfare, and that this included, until very recently, many of the native human societies of the South American lowlands.  I no more believe that this should determine our attitudes and actions towards native Amazonians than should Tacitus's descriptions determine our attitudes towards the cultural descendants of Germanic peoples or the Romans, and the entire debate seems designed to obscure interesting facts about the world, and about Amazonia in particular.

*On a recent visit to the Pitt Rivers Museum, I saw a Conibo longbow made of palmwood.  It was about the height of a man and had a thick, twined string.  The bow itself was flat in cross-section and was beautifully tapered.  I would not be surprised if arrows shot from it were capable of utterly skewering a human being.

**Cubeo is a Tukanoan language. Panoan languages and cultures hybridised in southwestern Amazonia with Arawakan ones, and Tukanoan ones did with Arawakan in the north.


  1. Am reading up on Moriori (Michael King). You could say they were a weak chiefly society, pretty much egalitarian... they had one murder in about 50 years post-western-discovery... and the murderer was ostracized and committed suicide.

    I am of the opinion that many tribal societies were relatively peaceful (apart from the occasional raid or skirmish over a woman). The really nasty stuff came after societies surrendered to the rule of the Big Men and their incessant clamoring for more: more goods, more feasts, more people, more conflict... more power.

  2. The non-state societies with the least feuding and the lowest murder rates are typically those that are more hierarchically organised - ie, those that most closely resemble states. This is certainly true in South America. Arawakan-speaking societies are generally much more stratified than their non-Arawakan neighbours and they have no internal feuds. States make life less violent for everyone, but even so-called 'Big Men' have an interest in stopping feuding among the members of their communities.

    I don't think tribal horticulturalists are likely to be peaceful, whether within their societies or without. This is borne out by almost all of the archaeological, ethnographic, and ethnohistoric evidence that exists. The whole thing has to be thought of in terms of reasons for action for individuals. Horticulturalists need productive land, productive land can be acquired through fighting, and tribal societies have no states by definition, so there is no punishment for raiding or feuding. There has to be someone to stop people from fighting like this or else, when the situation favours raiding, they will do it. It's not just land that encourages fighting, either. In the absence of punishment from a state, minor sleights and insults can result in fighting. In Arawakan-speaking communities, these things were attenuated by the power of chiefs, so endo-feuding seldom arose.

    In the case of the Moriori, I think their communities were small enough that they could police themselves, and ostracism could be a severe punishment inflicted by the community. If the community were much bigger, ostracism would be difficult to achieve - some segment might have an interest in rehabilitating the person and their reputation. In the absence of a state, feuding could result. I think this is partly a numbers game - anarchy is possible in a tiny commune but completely impossible in a population of thousands, let alone millions.

    Inequality and its associated problems probably did result from the actions of 'Big Men', to some extent. To that extent, Rousseau seems to have been right. But inequality and violence are not the same thing, and a society without states is almost inherently violent. Only immediate-return foragers are immune from the development of raiding, but that is because they have little reason to engage in it, and it is notable that many immediate-return societies were/are internally violent, ie, most Inuit groups, etc.

    1. Yes, simple foragers do often have significant internal violence, but as I was trying to say, this is not always the case, and since we don't really know any that have not been affected by civ, it's hard to say how it once was.

      Anarchism would be possible if the organization began with small self-governing units, and then nested structures into larger configurations.

      Horticulturists can be peaceful too, as long as they keep caps on population. Viz the Tikopia.

      As for the state keeping things peaceful, of course it does not. The state cracks down on local self-governance and takes over the resolution of disputes. They enforce internal "peace" -- the cost of which is the destruction of autonomy and local culture. Then they go prey on others externally.

  3. As for the state keeping things peaceful, of course it does not. The state cracks down on local self-governance and takes over the resolution of disputes. They enforce internal "peace" -- the cost of which is the destruction of autonomy and local culture. Then they go prey on others externally.

    The state keeps the peace internally, meaning that humans kill one another much less when they live in a state. Whether they then feel compelled to conform to state values is a moot point; if you value cultural principles, even if they're self-destructive and violent, over human lives, then... I don't see how that wouldn't be an insane position.

    Anarchism is not possible and doesn't work in any community larger than a couple of hundred individuals (as found on Anuta and Tikopia). Pre-state societies are almost uniformly violent when compared with state societies - this is so true that it is easy to identify pre-state archaeological remains by the presence of palisades and other defensive features around domestic sites. This is why your statement,
    since we don't really know any that have not been affected by civ, it's hard to say how it once was
    is incorrect. We know of plenty of pre-state societies unaffected by 'civ' (civilisation? the state? modernity? industrialisation?). We know them through archaeology, linguistics, and ethnohistory. We know through the remains their violence left behind.

    It is unfortunate that simple empirical questions about archaeology and ethnography become so politicised. I have no political axe to grind - in fact, I lean as close to libertarianism as the facts will allow - but I find anarchism incommensurable with the way the world is. Not because of an a priori commitment to another political philosophy, but because anarchy is as good as empirically falsified.

    1. I did not come here to argue anarchism. My ideal would probably something on the order of Ostrom's commons' governance.

      Yes, humans murder one another less when they live within a state, while the state kills liberally outside, with the help of its own cannon fodder. Though of late, it seems that killing indirectly via economic weapons seems to have prevailed.

      I would be interested in knowing, for example, how many people died as a result of the English enclosures. I am betting the numbers were massive, but it would be hard to get the data since some died of city diseases, some starved, some turned to petty crime and were executed or died in prison, others died of horrific factory conditions into which they were forced, and so on. Murder by theft of the commons.

      I value culture and human lives, both. You are not welcome to paint me into a corner.

      As for ancient cultures or protocivilizations, there is plenty of evidence for peace. Catal Huyuk, for example. Or early Cucuteni. Or Norte Chico (Caral). I am told the same is true for the Harrapan people, but I have not verified it.

      You do seem to have an axe to grind... the defense of the state.

    2. What absolute piss. I have no axe to grind, but you clearly wish to avoid the mountain of evidence for pre-state violence - the Supe valley's early towns (probably not true states, by the way) were incredibly violent and even featured human sacrifice. It was earlier believed that the Norte Chico people were peaceful, but it now seems not to be the case (as you may know, if you continue reading The Origins of Inequality).

      Most states do not kill - certainly industrial and post-industrial states tend not to. Life expectancy has increased dramatically since the rise of industrial states across the globe. Violence of all kinds has decreased. The idea that states are inherently bad or militaristic is a load of libertarian bullshit and is bolstered by no evidence. Some states are militaristic, but almost none today are, and regardless of this fact, states are good for humans in terms of the reduction of violence. This is an empirical fact, not a piece of politicised speculation.

    3. You are bluffing. Show me the evidence that either Caral, Catal Huyuk, or early Cucuteni had warfare!

      I concede the two, possibly three human sacrifices were found in Caral and one other settlement. I did not claim they were angels. I claim there is no evidence of war (or of your claim of "incredible violence") -- not until a thousand years later. And a thousand years without war is no mean achievement.

    4. The sites you have chosen are famous for the exceptionally low frequency of burials. Almost none are known from early Cucuteni and those that come from Caral are often of sacrificial victims. I'd also point out that the main evidence we have for violence in preceramic Peru comes from the graphic depictions at Cerro Sechin, a place which otherwise affords little indication of violence. So I think the problem here is sampling bias.

      Also, neither Caral, Catalhoyuk, nor the Cucuteni towns were really urban settlements. Caral wasn't even the largest settlement in the Supe valley at the time, and the other larger ones nearby - Asia, Aspero, etc - do show evidence of violence. Catalhoyuk may be a real exception, though.

      Moreover, I consider human sacrifice - especially of a bound human being who clearly didn't want to be there - to be 'incredibly violent'.

      The main problem here is confirmation bias: you are looking only for the evidence that supports your position, a position you hold for fundamentally political reasons, instead of looking at all of the evidence for pre-state violence.

    5. The graphic descriptions of violence at Cerro Sechin come from a thousand years after Caral.

      Aspero shows a man sacrifice in the pyramid, and a baby in a basket burial. What else can you point to?

      I am looking for places that falsify the claim that pre-state societies are "almost uniformly violent" as you said.

      Whether or not you consider human sacrifice incredibly violent is not the point here -- every murder is. The point is that you claimed the Norte Chico *towns* as a whole were incredibly violent. There is zero indication for your claim.

    6. Headhunting took place at Asia, down the coast, but that was around the same time as at Cerro Sechin. So we know that there was violence at this time. But you're right - there's a lack of evidence for Caral. It's not evidence for peace, but absence of evidence for anything about war. They didn't carve it into monuments, but they also tied up and sacrificed an adult male, which is not a peaceful action. That is the sum total of our evidence: rising social complexity and the sacrifice of a warrior-age human. You are correct that we cannot infer that their entire civilisation was violent or even that they had any wars at all, but given the sheer weight of comparative evidence and the fact that we lack any burials of individuals who died from any causes, peaceful or otherwise, to deny the possibility that Caral's inhabitants engaged in frequent violent confrontation is an absurdity, and it is clearly politically motivated.

      Pre-state societies - especially sedentary or delayed-return societies - were violent according to all of the available evidence. The only places where this might be in doubt are places where we lack evidence, so... It doesn't exactly support your case.

    7. You've just argued yourself into a hole in the ground... and you know the rule of getting out of holes, right? Stop digging.

      As long as an excavation shows no signs of war for long periods of time, this is evidence. It may change, but until it does, it serves as evidence. And Catal Huyuk has the graves you seek. And no war, and rather pronounced equality. Even between the sexes.

      On the other hand, we have all the evidence we should ever want that state level societies are violent, predatory kleptocracies.

  4. So you can point to a couple of examples where either the evidence is non-existent or uninterpretable (like Caral and Cucuteni, both of which show evidence of human sacrifice, the deliberate and callous slaughter of a human being for religious purposes, hardly the act of peaceful people). The entirety of the rest of the archaeological and ethnohistoric record, however, points in the other direction: pre-state societies, especially those that are sedentary and horticultural/pastoral, are so much more violent, in terms of per capita death rate by human hands, than state societies that they aren't really comparable. All of the rest of the evidence points to this. States drastically reduce the death toll by human hands. Even the most militaristic state - even the Mongol empire - has an interest in preventing violence in its domain and among its people. This results in a generally improved life overall. And as this process has gone on for some time, it has resulted in the almost absurdly peaceable state in which we presently live. We are safer and freer now without a doubt.

    See also my post here:

    1. Haha. I am sure the Cypriots are feeling really free and safe right now.


You can post anonymously if you really want to, but I would appreciate it if you could provide some means of identifying who you are, if only for the purpose of knowing who has written what.