(This is the second post in a short sequence discussing evidence and arguments presented by Christopher Ryan and Cacilda Jethá's Sex at Dawn, inspired by the spirit of Kaj_Sotala's recent discussion of What Intelligence Tests Miss. It covers Part II: Lust in Paradise and Part III: The Way We Weren't.)

Forager anthropology is a discipline that is easy to abuse. It relies on unreliable first-hand observations of easily misunderstood cultures that are frequently influenced by the presence of modern observers. These cultures are often exterminated or assimilated within decades of their discovery, making it difficult to confirm controversial claims and discoveries. But modern-day foraging societies are the most direct source of evidence we have about our pre-agricultural ancestors; in many ways, they are agriculture's control group, living in conditions substantially similar to the ones under which our species evolved. The standard narrative of human sexual evolution ignores or manipulates the findings of forager anthropology to support its claims, and this is no doubt responsible for much of its confused support.

Steven Pinker is one of the most prominent and well-respected advocates of the standard narrative, both on Less Wrong and elsewhere. Eliezer has referenced him as an authority on evolutionary psychology. One commenter on the first post in this series claimed that Pinker is "the only mainstream academic I'm aware of who visibly demonstrates the full suite of traditional rationalist virtues in essentially all of his writing." Another cited Pinker's claim that 20-60% of hunter-gatherer males were victims of lethal human violence ("murdered") as justification for a Malthusian view of human nature. 

That 20-60% number comes from a claim about war casualties in a 2007 TED talk Pinker gave on "the myth of violence", for which he drew upon several important findings in forager anthropology. (The talk is based on an argument presented in the third chapter of The Blank Slate; there is a text version of the talk available, but it omits the material on forager anthropology that Ryan and Jethá critique.)

At 2:45 in the video Pinker displays a slide which reads

Until 10,000 years ago, humans lived as hunter-gatherers, without permanent settlements or government.

He also points out that modern hunter-gatherers are our best evidence for drawing conclusions about those prehistoric hunter-gatherers; in both these statements he is in accordance with nearly universal historical, anthropological, and archaeological opinion. Pinker's next slide is a chart from The Blank Slate, originally based on the research of Lawrence Keeley. Sort of. It is labeled as "the percentage of male deaths due to warfare," with bars for eight hunter-gatherer societies that range from approximately 15-60%. The problem is that of these eight cultures, zero are migratory hunter-gatherers.

In descending order of bloodiness, the societies mentioned are the Jivaro (who cultivate numerous crops, keep livestock, and live in "matrilocal households"), the Yanomamo (who live in villages and grow bananas), the Mae Enga (who have scattered homesteads and cultivate sweet potatoes), the Dugum Dani (who live in villages, cultivate sweet potatoes, and raise pigs), the Murngin (more commonly known as the Yolngu; the data cited was collected in 1975, after they had been living with "missionaries, guns, and aluminum powerboats" (Ryan and Jethá, 185) for more than three decades), a different tribe of the Yanomamo (who still live in villages and grow bananas), the Huli ("exceptional farmers"), and the Gebusi (who live in longhouses and keep gardens).

While Keeley's research is the basis for this claim, it should be noted that he distinguishes (if somewhat confusingly) between what he calls "sedentary hunter-gatherers" and true "nomadic hunter-gatherers" (War Before Civilization, 31, as cited by Ryan and Jethá). Keeley also points out that "Farmers and sedentary hunter-gatherers had little alternative but to meet force with force or, after injury, to discourage further depredations by taking revenge." The nomads, on the other hand, "had the option of fleeing conflict and raiding parties. At best, the only thing they would lose by such flight was their composure."

Pinker is not so easily excused. It is possible that he failed to recognize this critical research failure for the five years between The Blank Slate's publication and the TED talk in question, and that no one with a copy of his best-selling book and access to the internet noticed the error and pointed it out to him. It is also possible that he was being deliberately deceptive. In either case, while this doesn't warrant discarding every claim that Pinker has ever made about evolutionary psychology, he should probably not be considered a reliable source on the topic. (See Menand, Blackburn, and Malik, for example, for further criticism of Pinker's approach in The Blank Slate.) 

What does forager anthropology have to say about human sexual evolution and the standard narrative, then? Ryan and Jethá offer a wealth of examples in support of their thesis--that in the human evolutionary environment, communal sexual behavior was the dominant paradigm.

Partible Paternity - While we (and the standard narrative's advocates) take it for granted that any given individual can have only a single father, this was not established scientifically until the 19th century. This belief is not universal, however, and Beckerman and Valentine (pdf) have compiled decades of anthropological research on dozens of South American tribes (both foragers and farmers) that believe in partible paternity: that "a fetus is made of accumulated semen" (Ryan and Jethá, 90), and so can have multiple biological fathers.This is not a regional oddity, either--the Lusi of Papua New Guinea had similar beliefs. One of the consequences of this belief is that a woman who wants to give her future child every possible advantage should "solicit 'contributions' from the best hunters, the best storytellers, the funniest, the kindest, the best-looking, the strongest, and so on--in hopes her child will literally absorb the essence of each" (Ryan and Jethá, 91).

It is a key point in the standard narrative that men benefit when they enforce sexual monogamy on their mates and ensure their paternity. But Hill and Hurtado (review here) found that among the Aché, a South American tribe that believes in partible paternity, children with multiple fathers were more likely to survive than those with only one. It should not be hard to see why: keeping all else equal, men in a society in which all children have one father have the same rate of biological paternity as men in a society in which all children have three fathers. Should a particular man in the first society die before his child is grown, his child has no other man to rely upon for resources; should a particular man in the second society die, his child is one of three receiving the resources of the two surviving men. (The second society would then have a strong selection pressure for men who were more likely to provide the sperm originally responsible for fertilization; the next post in this sequence, on comparative anatomy, will cover sperm competition.)

Alloparenting - The standard narrative considers it a given that individuals should have no incentive to raise children which they know are not their own; this is why paternity is important in the first place. While it should be possible to fool a man into thinking a child carries his genetic legacy (and it is important for the standard narrative that this is so), fooling a woman in the same way is rather less plausible. Chimp and gorilla mothers never allow other females in their tribe to hold their young children, probably because females of both species are quite willing to kill infants not their own; this is precisely what the standard narrative tells us we should expect. And yet, in 87% of human forager societies, mothers are willing to allow other women to breastfeed their children. This is not just a lack of rampant infanticide: it's an active expenditure of resources. (See Hrdy's Mothers and Others; review here.)

This is the sort of highly-cooperative situation that might seem like it could only be explained by group selection, and if that were so it would probably be a good idea to discard evolutionary explanations of alloparenting entirely. But positing group selection is unnecessary when there's a much more plausible explanation available: kin selection. W.D. Hamilton identified two different mechanisms by which kin selection might operate, and humans are among the minority of species which fit the criteria of both. (The related Price Equation may offer a more definitive explanation of the process, but I admit that I haven't taken the time to grok the math involved.) Kin selection also explains why alloparenting is more frequently practiced by close relatives (like grandparents, aunts, and uncles) than by distant relatives or unrelated tribe members. 

(If you're still uncomfortable with kin selection's similarity to group selection, take a look at the breeding patterns of naked mole rats and try to explain them any other way.)

The (Un-)Universality of Marriage - Well-respected anthropologists (George Murdock and Desmond Morris, for example) are in the habit of declaring that marriage is found every human society, a finding that provides strong support for the standard narrative; after all, explaining the evolutionary inevitably of human pair-bonding isn't much good if it isn't universal in the first place. Anthropologists are willing to consider all kinds of arrangements to be "marriage", though, creating confusion that is easily amplified by imprecisions of translation.¹

For example:

• The previously-mentioned Aché, who "say that a man and woman sleeping in the same hut are married. But if one of them takes his or her hammock to another hut, they're not married anymore" (Ryan and Jethá, 119).
• "Among the !Kung San (also known as Ju/'hoansi) of Botswana, most girls marry [emphasis theirs] several times before they settle into a long-term relationship." (ibid.)
• Younger Curripaco call a couple married when the woman moves her hammock next to the man's and begins cooking for him. Older Curripaco don't acknowledge the marriage until the couple has had a child and fasted together.
• Modern Sunni Muslim doctrine includes provisions for a form of (often-extramarital) marriage known as Nikah Misyar, which is used specifically to avoid the usual cohabitation requirements.
• The Canela of Brazil don't consider a couple to be married until the bride-to-be has had sex with the fifteen to twenty members of a "festival men's society." If she does well, the men make payments of meat to her future mother-in-law.
• The Inuit sometimes swap wives--permanently--as an alternative to divorce.
• Some Hindu castes in southern India once practiced a form of ceremonial marriage that involved no sexual contact and, usually, no contact between the bride and groom once the wedding concluded.
• The Mosuo have a taboo against acknowledging the existence of a sexual relationship and lack any traditions that even remotely resemble marriage. Naturally, their private and impermanent pairings are known as "walking marriages."

These are not behaviors that the standard narrative should be able to explain; indeed, if it could explain them it wouldn't be paying its rent. When I first noticed this point, I very suddenly realized why I find Ryan and Jethá's thesis so convincing: it isn't trying to explain everything.

The standard narrative is supposed to explain every universal aspect of human sexual behavior, and a great deal more besides. And, its proponents hold, it should be able to explain the behavior of both modern and prehistoric humans with more-or-less equal accuracy. This more than anything else is its failure: it does not acknowledge the mutability of human preference. The current mainstream American standard of female beauty values low body fat², which is a powerful signal of something about genetic fitness. Not long ago, the mainstream Mauritanian standard of female beauty valued obesity² (as some subpopulations still do), which is a powerful signal of something contradictory about genetic fitness. No evo-psych theory should be able to explain both of these desirability criteria in a fashion more direct than "desirability criteria are easily influenced by social pressures."³

Faced with the godshatter of modern human preference, Sex at Dawn passes on trying to provide an explanation. The book's greatest virtue, to my mind, is that it just attempts to discover the patterns of prehistoric sexual behavior, acknowledging that many questions about how humans behave today are better left to other disciplines.

The next post in this sequence will look at what Ryan and Jethá have concluded from the study of comparative sexual anatomy.

(As before, I will be happy to provide whatever additional citations I can to address specific claims made in this post.)

1: To say nothing of polygamous arrangements, which should obviously prohibit any attempt to conflate marriage with monogamy.

2: This statement is true, but that does not imply that I prefer the state of affairs it describes.

3: Except, of course, when they aren't.