Great post. I guess as a non-expert, my sense is that the evidence of hunter-gatherer lifestyles is so weak and uncertain, the ideological biases are so strong and almost entirely in one direction (in Fry's cases the desired conclusion is pretty explicit if I remember correctly), and the space for social scientists to select, frame, interpret, etc, results in biased ways is so vast that I don't have much confidence in any conclusions in this area. Another question I have is what is really at stake in disputes about the rate of violence among hunger-gatherers. Even if it's true that the rate of violence is relatively low, it doesn't follow that humans aren't "naturally" violent, competitive, etc. It might just be that with extremely small scale mobile subsistence societies violence doesn't generally pay in terms of expected value. That story would predict that when the incentives change, people can quickly become much more violent, which the history seems to support. Anyway, just some thoughts that occurred to me whilst reading. Thanks!
Thanks! And I feel both of your observations are correct. My level of certainty for direct and even more so indirect inferences about any kinds of numeric rates for pre-historic, nomadic hunter-gatherers is relatively low. Additionally, it does seem fairly clear that whatever the precise rate of violence in the "state of nature," both violence and peace have evolutionary roots or were evolutionarily relevant for human beings.
Very interesting reply to Pinker whose book I've found engaging. If I remember correctly Pinker contextualized violence over time as a percentage of the estimated population. I terms of percent or per capita analysis wouldn't it be clear that by that metric violence has declined markedly overtime particularly over the past thousand years
Thanks for this post, which is both interesting and useful. I look forward to reading your two articles.
I will mention that, in some ways, Pinker can be his own worst enemy. If you read the preface and Intro to the book (which lots of people do), you get the impression that when it comes to warfare, he’s arguing for ultra-steep declines in rates, analogous to homicide and execution; but f you read the chapter on Great Power wars (which apparently few people do), the actual claim is much weaker (trendless fluctuation).
Great stuff as always. How do anthropologists differentiate between, say, a family feud that turns violent and warfare? Or the modern equivalent of a gang war and other warfare? I think one of the problems here is that, in pre-state societies, by definition human bands were small and I imagine evaluating the archaelogical evidence will make it harder to differentiate run of the mill social/moral violence and warfare.
Thanks! It's quite hard to do that, unfortunately.
For ethnography, you have to infer from the qualitative descriptions and context. When you do that (and there's interpretive ambiguity sometimes, of course), you come up with stuff like like this (from Fry and Söderberg, 2013):
"Of 135 lethal events with unambiguous perpetrator and victim information, 55% consisted of one killer and one victim. In 23% of the lethal events, more than one person participated in killing a single individual, and in 22% of the events, more than one person participated in killing more than one person (Fig. 1). In only one lethal event (0.7%), did a single killer dispatch more than one person (table S4, case 18), and the two victims were children. Tiwi society reflects a different pattern wherein 44% of the lethal events involved one killer and one victim, whereas the corresponding figure for the other 20 societies combined was 64% (supplementary text)."
You can then distinguish, minimally, between ordinary interpersonal events and coalitionary (war-like, intergroup) violence.
"At Jebel Sahaba, the co-occurrence of healed and unhealed lesions strongly supports sporadic and recurrent episodes of interpersonal violence between Nile valley groups at the end of the Late Pleistocene. The projectile nature of at least half of the lesions suggests inter-group attacks, rather than intra-group or domestic conflicts, and the frequency of healed wound confirm that these events were not always lethal and could occur several time during the life of an individual. While the number of parry fractures is higher among female individuals, and the blunt force trauma mostly present on immature individuals, the remaining pattern of lesions on female and immature individuals at Jebel Sahaba is inconsistent with domestic violence."
"A catastrophic single mass burial is highly unlikely and not supported by the archaeological evidence and the demographic analysis. With the exception of a higher percentage of parry fractures in females, there appears to be no patterning in the distribution of trauma or PIMs by rather age or sex. Based on the lesions, the projectile direction also reveals an equal number of posterior and anterior strikes that do not support face-to-face battles. Rather, the involvement of a range of ages and both sexes, with primary (n = 26), double (n = 4) and multiple (n = 4) burials, including some with evidence of disturbance due to the addition of later individuals, indicate small episodes of recurring violent events such as raids or ambushes against this community. This appears to have taken place on a short timescale given the homogeneity of the burial place and practices."
Wow, thank you so much for this post!! Ever since I read the work of Zimbardo and Pinker, I've been driving myself mad trying unsuccessfully to disentangle all the complexities and assess the various claims about human nature, our propensity for violence and war and our "natural" state. I haven't been able to work out anything coherent so it was such a pleasure to read your beautifully thorough exploration.
I have one question: have you read Rutger Bregman's "Humanity: A Hopeful History?" by any chance? I don't know what to make of some of his claims and data sources, which I am sceptical of (though I would really, really love to be proven wrong!), but most of the critiques I have read of his work seem too biased to fully trust too.
I did read a large chunk of Bregman's book. My impression is similar to yours. There are parts where he clearly oversteps or is one-sided, and I think some of his simplistic framing (the veneer theory) is biased. But generally, I agreed with quite a lot of what I read.
Yes, the veneer theory was one of the framings I found the least convincing too. But I really did enjoy the book overall, so that's great to know. Thanks so much for the response :).
I had high hopes for it, but the introductory chapter (especially the style of it but also some of the substance) really put me off. It doesn't come across as a dispassionate scientific inquiry at all. I felt one early chapter on homicide was okay. Then there was a terrible, low-effort chapter on neoliberalism. Later on, I remember another chapter with some actual discussion of figures and sensible questions about what we can infer from limited evidence.
I'm not surprised the volume didn't get any wider acclaim. There's some good stuff in there, but there's also a lot of fluff, irrelevant digressions, and ideological bickering. I might be misremembering parts, so don't take these comments too seriously, but overall the book left a bad impression.
Another good avenue is exploring violence in contemporary and recent semi-nomadic people groups, such as the Blackfoot in Alberta, Canada. You will find cycles of peace, but also cases of sustained warfare for trade routes, land, and animal migration routes.
I don't really see how people living under the constraints of the nation they are in are a good avenue for anything related to how stateless tribes lived. That is the main issue with Fry and Söderberg too, they were looking at modern hunter gatherers akin to animals in a zoo (they were literally not allowed to war and they lived isolated in reserves).
Thanks for the post. It's not clear why the Fry and Soderburg 2013 sample is “prehistorically more appropriate” than the sample Pinker uses. The foraging societies in the F&S sample were massively transformed by the time the ethnographies they coded were written (participating in markets, living on reservations, joining the military, etc). Horticulturalists with high mobility, little market integration and state contact, and low inequality may be better analogs on some metrics than the F&S sample See section 2.2 here. https://www.sciencedirect.com/science/article/pii/S1090513824000941
Given that your familiarity with the data showing that the Fry and Soderburg 2013 results are extremely misleading to the point of being largely false I'm not sure why they are used as a counter to the sample Pinker uses. There are other reviews that show less prevalence of war than Pinker argues for. Further, F&S only present a numerator (number killed) without a denominator. This is meaningless for assessing the intensity or importance of war. It's like saying "100 people died from slipping on ice while walking." Is that a lot or a little? Beats the hell out of me because you need a denominator.
While this might be true (and may be likely to be true) “Modal prehistorical societies (nomadic foragers) in general experienced much lower violent lethality rates than claimed or implied by Pinker” I don’t think the data exist to assess this claim. The Meijer survey is the best synthesis of global prehistoric skeletal remains but it's incomplete. The societies we have good ethnographies for that have violence do generally report that a large fraction of deaths are due to violence, but you would be unlikely to know that in a thousand years just looking at the remains. I'd love to be convinced otherwise.
Two papers also arguing for a U-shaped trajectory for violence in human evolution:
Knauft, Bruce M., et al. "Violence and sociality in human evolution [and comments and replies]." Current Anthropology 32.4 (1991): 391-428.
Kelly, Raymond C. "The evolution of lethal intergroup violence." Proceedings of the National Academy of Sciences 102.43 (2005): 15294-15298.
"Steven Pinker documents this long arc in The Better Angels of Our Nature, arguing that violence has declined over long stretches of time and offering several reinforcing explanations. Among them are the rise of states and courts that monopolize force, and the spread of commerce, which makes other people more valuable alive than dead. In other words, governments can raise the cost of conflict, but trade and the institutions that support exchange can also redirect incentives away from predation and toward production. With the benefit of hindsight, the pattern is suggestive if not perfect: societies with stronger rule of law, wider commercial networks, and more open markets tend to be safer and more prosperous than societies organized around tribal politics and coercive redistribution." https://open.substack.com/pub/delphicmirror/p/from-family-to-tribe-to-exchange?r=5d3u3o&utm_campaign=post&utm_medium=web
Great post. I guess as a non-expert, my sense is that the evidence of hunter-gatherer lifestyles is so weak and uncertain, the ideological biases are so strong and almost entirely in one direction (in Fry's cases the desired conclusion is pretty explicit if I remember correctly), and the space for social scientists to select, frame, interpret, etc, results in biased ways is so vast that I don't have much confidence in any conclusions in this area. Another question I have is what is really at stake in disputes about the rate of violence among hunger-gatherers. Even if it's true that the rate of violence is relatively low, it doesn't follow that humans aren't "naturally" violent, competitive, etc. It might just be that with extremely small scale mobile subsistence societies violence doesn't generally pay in terms of expected value. That story would predict that when the incentives change, people can quickly become much more violent, which the history seems to support. Anyway, just some thoughts that occurred to me whilst reading. Thanks!
Thanks! And I feel both of your observations are correct. My level of certainty for direct and even more so indirect inferences about any kinds of numeric rates for pre-historic, nomadic hunter-gatherers is relatively low. Additionally, it does seem fairly clear that whatever the precise rate of violence in the "state of nature," both violence and peace have evolutionary roots or were evolutionarily relevant for human beings.
Very interesting reply to Pinker whose book I've found engaging. If I remember correctly Pinker contextualized violence over time as a percentage of the estimated population. I terms of percent or per capita analysis wouldn't it be clear that by that metric violence has declined markedly overtime particularly over the past thousand years
Thanks for this post, which is both interesting and useful. I look forward to reading your two articles.
I will mention that, in some ways, Pinker can be his own worst enemy. If you read the preface and Intro to the book (which lots of people do), you get the impression that when it comes to warfare, he’s arguing for ultra-steep declines in rates, analogous to homicide and execution; but f you read the chapter on Great Power wars (which apparently few people do), the actual claim is much weaker (trendless fluctuation).
Oh yeah, definitely agree with that!
Great stuff as always. How do anthropologists differentiate between, say, a family feud that turns violent and warfare? Or the modern equivalent of a gang war and other warfare? I think one of the problems here is that, in pre-state societies, by definition human bands were small and I imagine evaluating the archaelogical evidence will make it harder to differentiate run of the mill social/moral violence and warfare.
Thanks! It's quite hard to do that, unfortunately.
For ethnography, you have to infer from the qualitative descriptions and context. When you do that (and there's interpretive ambiguity sometimes, of course), you come up with stuff like like this (from Fry and Söderberg, 2013):
"Of 135 lethal events with unambiguous perpetrator and victim information, 55% consisted of one killer and one victim. In 23% of the lethal events, more than one person participated in killing a single individual, and in 22% of the events, more than one person participated in killing more than one person (Fig. 1). In only one lethal event (0.7%), did a single killer dispatch more than one person (table S4, case 18), and the two victims were children. Tiwi society reflects a different pattern wherein 44% of the lethal events involved one killer and one victim, whereas the corresponding figure for the other 20 societies combined was 64% (supplementary text)."
You can then distinguish, minimally, between ordinary interpersonal events and coalitionary (war-like, intergroup) violence.
For archaeology, it's also hard. Hence the interpretative disputes over Jebel Sahaba and Nataruk. See, for instance: https://www.nature.com/articles/s41598-021-89386-y
"At Jebel Sahaba, the co-occurrence of healed and unhealed lesions strongly supports sporadic and recurrent episodes of interpersonal violence between Nile valley groups at the end of the Late Pleistocene. The projectile nature of at least half of the lesions suggests inter-group attacks, rather than intra-group or domestic conflicts, and the frequency of healed wound confirm that these events were not always lethal and could occur several time during the life of an individual. While the number of parry fractures is higher among female individuals, and the blunt force trauma mostly present on immature individuals, the remaining pattern of lesions on female and immature individuals at Jebel Sahaba is inconsistent with domestic violence."
"A catastrophic single mass burial is highly unlikely and not supported by the archaeological evidence and the demographic analysis. With the exception of a higher percentage of parry fractures in females, there appears to be no patterning in the distribution of trauma or PIMs by rather age or sex. Based on the lesions, the projectile direction also reveals an equal number of posterior and anterior strikes that do not support face-to-face battles. Rather, the involvement of a range of ages and both sexes, with primary (n = 26), double (n = 4) and multiple (n = 4) burials, including some with evidence of disturbance due to the addition of later individuals, indicate small episodes of recurring violent events such as raids or ambushes against this community. This appears to have taken place on a short timescale given the homogeneity of the burial place and practices."
thanks! The quantitative methods of these fields are always interesting to me.
Agreed. :)
Wow, thank you so much for this post!! Ever since I read the work of Zimbardo and Pinker, I've been driving myself mad trying unsuccessfully to disentangle all the complexities and assess the various claims about human nature, our propensity for violence and war and our "natural" state. I haven't been able to work out anything coherent so it was such a pleasure to read your beautifully thorough exploration.
I have one question: have you read Rutger Bregman's "Humanity: A Hopeful History?" by any chance? I don't know what to make of some of his claims and data sources, which I am sceptical of (though I would really, really love to be proven wrong!), but most of the critiques I have read of his work seem too biased to fully trust too.
Thanks. :)
I did read a large chunk of Bregman's book. My impression is similar to yours. There are parts where he clearly oversteps or is one-sided, and I think some of his simplistic framing (the veneer theory) is biased. But generally, I agreed with quite a lot of what I read.
Yes, the veneer theory was one of the framings I found the least convincing too. But I really did enjoy the book overall, so that's great to know. Thanks so much for the response :).
Considering most tribes would have barely, if at all, met other tribes, our violence rates were extremely high.
Have you read The Darker Angels of Our Nature: Refuting the Pinker Theory of History & Violence (2021/2022)? I’d be interested to get your thoughts.
I did read it, though not in whole.
I had high hopes for it, but the introductory chapter (especially the style of it but also some of the substance) really put me off. It doesn't come across as a dispassionate scientific inquiry at all. I felt one early chapter on homicide was okay. Then there was a terrible, low-effort chapter on neoliberalism. Later on, I remember another chapter with some actual discussion of figures and sensible questions about what we can infer from limited evidence.
I'm not surprised the volume didn't get any wider acclaim. There's some good stuff in there, but there's also a lot of fluff, irrelevant digressions, and ideological bickering. I might be misremembering parts, so don't take these comments too seriously, but overall the book left a bad impression.
Another good avenue is exploring violence in contemporary and recent semi-nomadic people groups, such as the Blackfoot in Alberta, Canada. You will find cycles of peace, but also cases of sustained warfare for trade routes, land, and animal migration routes.
I don't really see how people living under the constraints of the nation they are in are a good avenue for anything related to how stateless tribes lived. That is the main issue with Fry and Söderberg too, they were looking at modern hunter gatherers akin to animals in a zoo (they were literally not allowed to war and they lived isolated in reserves).
The fry söderberg study is so useless it's like studying lions in a zoo to see how often they hunt in the wild
The modern hunter gatherers they included to make their case are generally:
Surrounded by police or military outposts
Pushed into marginal, resource-poor lands (deserts/tundras) where fighting isn't worth the calories
Explicitly forbidden from warring by colonial or state laws
Thanks for the post. It's not clear why the Fry and Soderburg 2013 sample is “prehistorically more appropriate” than the sample Pinker uses. The foraging societies in the F&S sample were massively transformed by the time the ethnographies they coded were written (participating in markets, living on reservations, joining the military, etc). Horticulturalists with high mobility, little market integration and state contact, and low inequality may be better analogs on some metrics than the F&S sample See section 2.2 here. https://www.sciencedirect.com/science/article/pii/S1090513824000941
Given that your familiarity with the data showing that the Fry and Soderburg 2013 results are extremely misleading to the point of being largely false I'm not sure why they are used as a counter to the sample Pinker uses. There are other reviews that show less prevalence of war than Pinker argues for. Further, F&S only present a numerator (number killed) without a denominator. This is meaningless for assessing the intensity or importance of war. It's like saying "100 people died from slipping on ice while walking." Is that a lot or a little? Beats the hell out of me because you need a denominator.
While this might be true (and may be likely to be true) “Modal prehistorical societies (nomadic foragers) in general experienced much lower violent lethality rates than claimed or implied by Pinker” I don’t think the data exist to assess this claim. The Meijer survey is the best synthesis of global prehistoric skeletal remains but it's incomplete. The societies we have good ethnographies for that have violence do generally report that a large fraction of deaths are due to violence, but you would be unlikely to know that in a thousand years just looking at the remains. I'd love to be convinced otherwise.
Two papers also arguing for a U-shaped trajectory for violence in human evolution:
Knauft, Bruce M., et al. "Violence and sociality in human evolution [and comments and replies]." Current Anthropology 32.4 (1991): 391-428.
Kelly, Raymond C. "The evolution of lethal intergroup violence." Proceedings of the National Academy of Sciences 102.43 (2005): 15294-15298.
"Steven Pinker documents this long arc in The Better Angels of Our Nature, arguing that violence has declined over long stretches of time and offering several reinforcing explanations. Among them are the rise of states and courts that monopolize force, and the spread of commerce, which makes other people more valuable alive than dead. In other words, governments can raise the cost of conflict, but trade and the institutions that support exchange can also redirect incentives away from predation and toward production. With the benefit of hindsight, the pattern is suggestive if not perfect: societies with stronger rule of law, wider commercial networks, and more open markets tend to be safer and more prosperous than societies organized around tribal politics and coercive redistribution." https://open.substack.com/pub/delphicmirror/p/from-family-to-tribe-to-exchange?r=5d3u3o&utm_campaign=post&utm_medium=web