Understanding Western Exceptionalism

Transcript of conversation with Joseph Henrich

A few months ago, I reviewed Joseph Henrich’s The WEIRDest People in the World. This week, I was happy to have him on the CSPI Podcast to talk about his work. A lightly edited transcript of the conversation is below.

While I’m a big admirer of Henrich’s books, I remain unconvinced by his answers about whether we have reasons to doubt that inherent group differences have a role to play in societal outcomes. I was particularly surprised when he expressed skepticism about inherent population differences in athletic ability, which I think made his position more internally consistent while at the same time demonstrating why it is wrong. Readers and listeners can judge for themselves.

I was excited to discuss AI alignment, and I hopefully nudged him into perhaps at some point thinking more about and writing on this topic. His work has certainly influenced my thinking on the subject, as we discuss during our conversation.

---

AI Alignment

Richard: Hi everyone, welcome to the podcast. I’m here with Joe Henrich today. Professor Henrich, could you just introduce yourself a little bit? I think you need no introduction to most of our listeners, but for those who do need one, could you just talk a little bit about your background and what you do?

Joe: Of course, yeah. I’m Joe Henrich. I’m a professor in the Department of Human Evolutionary Biology at Harvard. I apply evolutionary theory to understand human behavior and psychology. I’m particularly interested in humans as a cultural species, so focusing on how natural selection has shaped our minds for cultural evolution. I do a variety of different kinds of field research, including historical research, but also ethnographic fieldwork.

Richard: Okay, great. And yeah, you’ve written two fascinating... I think you’ve written three books, right? I know you have at least three. So the two that I’ve read are, The Secret of Our Success and The WEIRDest People in the World. And the first thing I would ask... I read The Secret of Our Success, and I came away with it less impressed with human intelligence at the individual level and more impressed with human intelligence at the collective level. I think that was the intention of the book. And I think it was very eye-opening about how much of our intelligence is not really intelligence in the way we think about a person having a high IQ and inventing something or doing something.

Have you thought about the implications of this for the AI debate? Because people often think in terms of, okay, there’s this individual thing that’s going to have so much brain power. It’s just going to be able to dominate us. It’s going to be able to do whatever it wants. Whatever its goals are, we have to align it. Does your theory about what makes us smart tell us anything about what would potentially make artificial intelligence smart, and what it could possibly do?

Joe: Yeah, I think so. Because the assumption is that if suddenly AI gets as smart as we are or smarter, more computational processing power, it’s suddenly going to be able to do all the stuff we can do. We’re able to do all of that stuff because of the interlinked sets of minds that accumulate ideas and tools and technologies, and actually new ways of processing information. Over time it’s built up often without there being any individual inventor at all. The real worry is when they start linking up the AIs and they start learning from each other.

Richard: Yeah. And do you think that’s plausible? Do they have the mechanism that we had which was, you have evolution, you have genetic selection, and then you have cultural selection, which as you show is very, very powerful. Will they have something like that? Is that something we can assume, or do we have to come up with a theoretical reason?

Joe: No. Right now, to my knowledge, who knows what’s going on. But to my knowledge, they aren’t built like that, but presumably they could be. That would probably be a bad idea, I think.

Richard: Yeah, to build their ability. Can we rule out, maybe not rule out, but say it’s highly unlikely that... You know, the first time I read about the paperclip monster, that it basically just sits there and it figures out how to manipulate us, and then it figures out how to give itself another IQ point, and then it gives itself another IQ point, and then it just gets so smart. It probably is not like that because nothing in our experience is like that. That’s not how we got here, one genius building on top of another. That has interesting implications.

Joe: And we also rely on a lot of serendipity and happenstance. In some sense, what cumulative cultural evolution is doing, is it’s aggregating a whole bunch of lucky coincidences and insights that have occurred dispersed over a broad population over time.

Were Greece and Rome Already WEIRD?

Richard: Yeah, that’s right. Yeah, there needs to be a lot. And I think the selection mechanism, that’s what takes care of it. That’s the key to the whole thing. Yeah, the artificial intelligence story, it’s missing that. I think maybe it could still happen, but you need a good theory for that. And so your other book, The WEIRDest People in the World, one thing people I think have pointed out, some reviewers, is that your argument is basically that the Christian marriage and family plan makes Westerners unique, it cultivates these psychological traits. But people who’ve studied the ancient world, they’ll say, well, we have Athenian democracy, Aristotle’s considered the father of science. What do you say about the argument that Westerners were already WEIRD, this was baked in before Christianity?

Joe: Well, it’s important to distinguish Greece and Rome. Rome has monogamy to some degree, although there’s no sexual constraints on men. But it’s founded based on elite clans. And the theory actually applies pretty well to classical Athens. Classical Athens made a bunch of changes to the kinship system, including bilateral inheritance, and enforced monogamy, where Athenian men could only have one wife. So they cultivated a kind of individualism. Psychologically, they became more similar, and that led to a lot of the changes in the flourishing that we see historically. But I don’t think there’s a very good argument to be made that that cultural heritage then came to the West, other than bits and pieces of it.

A lot of the philosophical tracts and whatnot, the intellectual contributions written by the Greeks, were found after Europeans had already become WEIRD when they flowed in from Islamic societies and whatnot. So that’s too far too late in the story to account for the differences we see. I think the better argument is that Europeans liked those things because they were individualistic and they had a lot of the psychological features that Europeans had come, mostly through independent convergence, to like.

Richard: Yeah. And what about people who say that... You also have Ancient Greece and you have Ancient Rome, but I’ve read a lot of stuff about the Northern Europeans. And so Iceland, for example, they apparently have the longest-running parliament in the history of the world. It was started in 950, at least that’s the claim. I don’t know how much you’ve read about this. And that was actually before they were Christian, that was 100 or 200 years before they were Christianized. There’s an argument that there’s either a cultural or genetic basis to these things that go before Christianity. How plausible do you find that?

Joe: The first thing to realize is that these councils in which members of different subgroups of some larger community, say a tribe, come together to make decisions, that can be found all around the world, including in places like New Guinea, and all throughout the New World, and is super common. Some people have tried to say, oh, that’s democracy. Well, it’s certainly not democracy of the kind that we’re accustomed to. The representatives in these cases are generally representatives of large kin groups, representatives of clans. And so the decision-making mechanism is not individuals making decisions, it’s families or clans negotiating by getting a bunch of clan leaders, generally men, around a fire or whatever, and then making these decisions or adjudicating cases, or things like that. So it does have a representative feel to it, but it’s not based on individuals each having their own separate voice. I think that part is a little bit misleading.

In that sense, Europe looks like the rest of the world, and you don’t see the historical processes that then took place in Europe. You don’t see them appearing elsewhere in the world where there also were these groups of people coming together in what looked like democratic representative groupings. People seem to think if you don’t have autocracy, then you have democracy. Maybe in some broad sense these things could be called democracies, but they’re still built on a logic of kinship, networks, essentialized descent rather than individuals. In these societies, for example, prestige is an important thing. But you can find what anthropologists call big man societies, where people build up through prestige and they build alliances and stuff. That’s super common across human history. That’s not what distinguishes Europe and makes it culturally different.

Richard: Yeah. You could look back and there’s so many ways to understand history, and there’s so many ways to understand institutions. And for you, what you’re looking for when you’re trying to say this is what makes Europeans unique, it’s the individual. From what I understand, you see that at Athens, you don’t really see that as much in Rome. You see the tribe, community. You don’t see that in the Northern European pre-Christian times, and you don’t see that in most of the rest of the world. So you don’t think, for example, if say a historian was analyzing Rome and comparing it to ancient Persia… do you think that they would not say Rome was more individualistic? Would you think that’s a correct interpretation of history? Because it’s all relative, right? 2,000 years ago they’re not going to be like us today. But what would you say to their argument… that perhaps there was just a strand of individualism relative to the rest of the world in Europe that there wasn’t everywhere else?

Joe: Yeah. That’s one of the things I really try to emphasize in the book is to get away from dichotomous thinking. And the theory that’s presented in the book begins with variation in the organization of families. And I try to really emphasize in the book you can find variation just within Han Chinese, in individualism, by looking at the rice agriculture. Because rice agriculture fosters particularly intensive clans and particularly intensive forms of kinship. So you’re going to create variation just within that single ethnic group. You can see something similar in India. You can find this anywhere. And so the point is there’s lots of individual variation. But then you need that question. Why did Europeans go down this path that really accentuated the individualism and the analytic thinking and the impersonal trust, and ended up with quite different institutions which hadn’t been seen before? That’s the puzzle I’m trying to get is, in one sense I already think that... Forget Europe, just analyze the rest of the world, you’ll find lots of variation in those kinds of things.

Explaining East Asian Success

Richard: Yeah. I think you’re right, that there’s something important that needs to be explained, European success. But I thought that it, The WEIRDest People, I think it glosses over quite a bit of East Asian success. You have these psychological differences between Europeans and the rest of the world. I don’t think if you looked at polling data or the kinds of evidence you use – things like polling data, things like ultimatum games and game theoretical, these constructs in the lab – I don’t know if you would find that East Asia really stands out. But then when you look at actual economic growth and social stability, East Asia does stand out. It’s really, the real first world countries are European descent or East Asian descent. You have some rich Gulf monarchies, but they’re a completely different system, they’re just sitting on an oil well. That’s a different thing. How do you think about East Asian success in this context? Is it an anomaly of the theory?

Joe: Well, the first thing that I think that people don’t realize is that because these societies had a lot of top-down rule, and they’d had a long history of agriculture, and they already had some psychological inclinations related to that like temporal discounting and stuff that you get from long periods of state formation, certain forms of agriculture, they’re prepared to take in new institutions and enforce them. In Japan, after the Meiji Restoration, Japan begins adopting institutions and laws from the West wholesale that alter the kinship structure and basically do quickly what it took a lot longer to do in Western Europe as the church expanded. In 1950, China does the same thing. So they try to destroy the clans, they burn genealogies, they begin banning marriage around close relatives. And they’re able to do relatively quickly what it took... Then the one-child policy just ends all cousins.

If you don’t have brothers and sisters, you don’t have cousins. So they’re able to remanufacture the kinship system relatively quickly. And that’s going to lead to all kinds of changes. So not only do you have the advantages of having the state system, they’re able to impose a bunch of the stuff from the West, and they have pre-built institutions. So Europe took centuries to develop all of these things, like universities, which are now common in Japan and China. That’s a western institution. It doesn’t feel like a problem to me.

Richard: Yeah. So your idea is that basically the western package works, and if you have a strong state you can enforce the western package. Is that the idea?

Joe: The machinery of global trade and stuff has already been built by the Europeans. And there’s that, at least for the 20th century, the US is driving the innovation. The US is pouring out novel innovation, some from Western Europe. And then they can be put to work and used in all these other places.

Richard: Yeah. But I guess, is East Asia really the... You have other places of the world that also have long histories of states. Ethiopia, I think has one of the longest histories of statehood. I’m not an expert in Ethiopia, how strong or functional that state’s been. But I know Egypt, you can look at Egypt if you want a really, really long history of statehood. You can look at Iraq, long history of statehood, although obviously with interruptions. Yeah, it’s interesting that it’s all so concentrated in Northeast Asia pretty much.

Joe: Yeah. But one of the interesting things about the cases you brought up, at least Egypt and Iraq, is that you get the spread of Islam through these places. And in the same way that I argue that one brand of Christianity adopted this marriage and family program, which broke everyone down into monogamous nuclear families, Islam, they constrained polygynous marriage, but then they adopted this inheritance rule. And the inheritance rule says that daughters get half of what sons inherit. Well that works fine if you’re a trader like Muhammad. But if your wealth is mostly in land, that means every time you marry a daughter off, part of your land goes away, and you get poorer and poorer and poorer as you marry your daughters off.

What Islamic society started doing was something that’s almost unheard of cross-culturally, except in the Islamic world, is patrilateral parallel cousin marriage. So if you’re a male, you marry your daughter to your brother, to your brother’s sons. And that keeps the wealth within the family and it stops that bleeding off of wealth. Those families get richer and more prosperous, so other groups copy them. But that creates this very endogenous kinship system, which is going to have the opposite psychological effect, than forcing people to build long distance ties. There’s lots of stuff going on in history, you got to take into account the role of it.

Richard: Yeah. I don’t think you mentioned this in the book, but you might have, that often minorities are very wealthy in the Muslim world. So you have Christians from the same ethnic background as the Muslim Arabs that are their neighbors, and they tend to have more economic success. And the Armenians in the Middle East tend to have more economic success. Do you see that as part of, something that makes sense in the context of the theory too?

Joe: Yeah, I think so. I hadn’t focused on that and studied it in detail, but roughly that seems to make sense.

Richard: Yeah, that’s interesting. There’s an alternative theory that basically the levels of development are reflecting IQ differences. So if you go look at East Asian IQ, it’s scoring well for decades now. You talk about a lot of psychological differences in The WEIRDest People in the World. You even have a measure of analytical reasoning, and you have a chart comparing that to rates of cousin marriage. And you find the more cousin marriage the lower analytic reason. Have you looked at or thought about IQ score differences and how they fit into this theory and explain the success of different nations?

Joe: Sure. The problem with the way that many folks think about IQ, is they think that it’s somehow a feature of our genes or something like that, rather than realizing that IQ is a set of specialized, culturally evolved cognitive abilities for navigating a particular environment. IQ, for example, just among Americans, I can tell you the story about other populations as well, just among Americans has gone up dramatically just in the [twentieth] century. It’s impossible for that to have anything to do with genes. But because of a whole series of changes related to nutrition, the nature of television, the nature of schooling, all of this has driven up IQs from what would’ve been a score of 75. Remember, IQs are normalized, so they’re always placed at 100. But what would’ve been 75 or something by modern standards. So you went up from 75 to 100 just by getting everyone to go to school.

And if you look at TV programs since the 1950s, you just have five characters. And now you have ensemble plots where you’re tracking multiple different narratives and different places. It’s training up a certain set of cognitive abilities. And the age at which kids learn their colors has gone from six or seven down to three. And that’s because we have books. “What color is that?” And we differentiate everything by these different colors and shapes and stuff. The system is evolving to train people in a certain set of cognitive abilities. Because of the structure of our society, those cognitive abilities lead to success within those domains and then they get cultivated. And so, IQ is associated with success because of the structure of society. If you are a hunter gatherer in Bolivian Amazon, those skills are virtually worthless. You’ve got to be able to track and have all these other cognitive skills. So you can imagine a hunter gatherer IQ, which would be a compilation of the things that tend to lead to success in the Bolivian Amazon, or something like that.

In my lab we’ve done research, and particularly I had a post-doc named Helen Davis, who has shown how the introduction of schooling destroys some cognitive abilities. So people get dumber when you send them to western schools in some ways, but they also get smarter in the ways that we call IQ. It’s just an interesting way that institutions manipulate our cognitive abilities. So when you see differences around the world in IQ, those can be very much the product of schooling, the importance of certain kinds of cognitive abilities, emphasis on pursuing these kinds of things, all those sorts of things. So it’s a product of... the cause and effect are all screwed up.

Richard: Yeah, I think all that is definitely true. And that’s true, you have the Flynn Effect. But that’s not inconsistent with... And you’re right, what makes for success of somebody in a modern industrial society, it’s not the same as what makes sense in the Australian Outback, trying to survive and track animals and whatnot. That being said, height for example, you could say the same thing. Height has gone up. That doesn’t mean that there aren’t group differences in height, or there aren’t relevant important genotypic differences between individuals and populations. And so the question is, it’s not about whether some groups, actually we might all be equal, or some groups might be better at the hunter gathering. The question is, do these patterns make sense?

And if you look at things like East Asians, I know that they have IQ tests going back decades, and they found impressive results. Before they became wealthy. North Korea actually does very well at the math Olympiads. Which, it’s very interesting considering that North Korea, how poor they are. I don’t know if it’s a math Olympiads, it was some mathematics competition [Note: It was the Math Olympiads]. I don’t want to give wrong information out there. Isn’t there something to the theory that perhaps this is capturing a lot of what makes some groups good at industrialization, or maybe post-industrial societies, and others having had less success in doing so?

Joe: Well, I just haven’t seen any evidence that’s persuasive. In all the cases you’re mentioning, being skilled in those particular cognitive abilities is highly valued in the society. So children are going to be growing up and cultivating those cognitive skills. I think what the big problem is, people don’t understand how plastic our brains and bodies are. If you grow up in Kenya, you might emphasize long distance running, so the top marathoners in the world tend to be Kenyan. That’s not because they get some special genes for long distance running. It’s because the thing you want to do in that society is run really far. So kids from a young age start running really far.

Richard: On the running thing, there’s a book called The Sports Gene, that actually argues that there is a reason that… So he says there’s long distance in East Africa, and then short distance sprinting in West Africa. It’s pretty interesting. I haven’t read the critique, but this applies even to the Western Hemisphere. So the people who came from western Africa, they’re good at short distance running and they’re not good at long distance running. And if you study the NFL, it’s amazing. You have the NFL cornerback position, it’s 100% some years all West African descent. So yeah, the sports is...

Joe: What about the NHL? There’s no Africans in the NHL, is that due to a lack of a hockey gene?

Richard: Well, of course, I think there’s different sports, right? So you have running, which is just you and you’re running. Some things, the equipment and having the income to train, some things are a lot. Hockey is probably least likely to be strongly genetic. There’s actually a book by a guy named Stephens-Davidowitz I interviewed on this podcast who actually calculated the heritability of different sports based on how likely a successful Olympic athlete was to have a child. Or how likely that a successful athlete had a parent, that’s the way he did it, who also succeeded at the sport. And stuff like hockey, stuff that requires a lot of equipment, tends not to be very genetic. Things like basketball, running, you get more genetic influence.

Joe: Remember we’re talking about some kind of group-level heritability as opposed to individual within-population heritability. It’s much harder to maintain population-level differences in genetics because any kind of gene flow is going to interrupt that. And I think the thing about the IQ debate, which again is missed, is that the set of things we call IQ are actually specialized cognitive abilities for navigating the modern world. And they’re not generally good cognitive abilities. There’s all kinds of situations when you need to have spatial navigation or raw pattern recognition, all of which people with high IQ aren’t that good at.

Richard: Is it no correlation between that and IQ, or is it negative? Do you know?

Joe: Yeah, I don’t know offhand. In general, people in cities are horrible at spatial navigation, so cities tend to have higher IQs than everywhere else. It’s likely that they’re not related, or they’re negatively correlated. Because what builds your spatial navigation abilities is having to navigate space from a young age, and needing to find your way across terrain, and know where the nearest city is, and things like that.

Richard: Yeah. But we do see that when different populations grow up in the same society that values all these things, economic success, we do see differences in cognitive performance. So is that an argument for these differences?

Joe: Well, no, I haven’t seen these kinds of adoption studies you would need to confirm that. What I have seen is the opposite. When you do have adoption studies, you don’t see the population-level differences.

Richard: Yeah, I’d like to see that. I’ve seen...

Joe: Korean adoptees, adopted by Canadian families.

Richard: Okay, yeah, I’d have to see that. I saw one from Scandinavia. And in Scandinavia, there were people, I think it was Sweden, and they adopted kids from Korea and then kids from some other countries. And then, there was some effect of the adoption, but the Koreans still scored really high, and the people from other countries that originally didn’t have higher IQs didn’t score as high. It removed some of the gap, but not all of it. I’d like to see the Canadian one. There was the Minnesota transracial adoption study too. Yeah, I’d like to see the Canadian one, because that would be different from those results. Well, we’ll put the links in the notes.

Selection for WEIRDness

Richard: And what about, putting aside group differences and cognitive ability, near the end of The WEIRDest People in the World you give an argument as to why you don’t think the MFP, the marriage and family plan, had a long-term genetic effect on Europeans. And it’s something to do with the cities as a demographic shredder. Instead of me explaining, can you explain that argument of why you don’t think it has had a large genetic effect?

Joe: Right. Well, part of the way, what I think the evidence shows, but the argument is that a lot of the action in the shifting psychology and the emergence of new institutions in Europe occurs in urban centers and towns. And so you have people migrating in from the countryside into places where it helps to be individualistic and analytically oriented and encourages temporal discounting and things like that. Values hard work, things like that. And throughout this whole period, urban areas tend to be urban graveyards. So historians talk about these urban graveyards. One idea would be that people who are interested in that, maybe they have some genes that make them more oriented towards that, move to these places. And then they’re going to tend to be selected out, because it’s much healthier to stay in the countryside and not move rather than move into the city. So if anything, this process would’ve created a selection pressure against those genes because you would’ve been attracted to places where you could put all this stuff to work and then that would’ve worked against you.

Richard: Yeah, that makes sense. And it’s an interesting argument. At the same time you do, in the book, argue that the MFP does reach into rural areas. And so you have these people moving to the cities. At the same time, Europe is not that urbanized here compared to modern standards. It’s not like 50% of people are living in the cities 500 years ago or so. You do have the MFP operating at the village level, and you do have people getting married and mating and all that stuff. And it seems hard to believe that wouldn’t have some effect. My intuition is over one thousand years despite the demographic shredder argument, that also makes sense. How do you feel about that?

Joe: You mean that it would select for those kinds of genes?

Richard: Yeah. Because the demographic shredder argument seems to be, just because the organization…

Joe: Another fact, suppose... One thing about this is that some of the values and some of the psychological inclinations that are WEIRD also lead you to have lower fertility in general. Because if you move apart from other people, you’re independent minded, the less you’re around family seems to encourage fewer children. If you like education, fewer children. It seems to do everything, pushes you in the direction of fewer children.

Richard: Yeah, that’s true. But you would move towards an optimization where if those traits help you succeed, and obviously you’re always selected for more children, you would want people who... You would think you would move towards a selected pressure where you’d get both, you’d get more children and you would get WEIRD individualist behavior, right?

Joe: Well, I think one of the problems with being an individualist is you are cultivating your own attributes and accomplishments. And so if you’re investing too much in siblings and children and stuff, then you’re not investing in cultivating your own unique attributes. It seems like there’s a psychological push to just reduce fertility. Now of course if you have wealth and stuff, that can help you survive shocks and stuff that other people might not survive. It’s not a slam dunk.

Weber, Protestantism, and Islam

Richard: Yeah. How do you think about development of WEIRDness in the context of Max Weber and his work on the Protestant work ethic? I get the impression from your book that you think that Weber puts a lot into the ideas of Protestantism. And I think you do too, but you think that Europe was going in that direction anyway. Do you give a lot to the existence of Martin Luther as an individual, or do you think it would’ve happened anyway, society would’ve gone in that direction?

Joe: Yeah. Max Weber anticipates a couple of my arguments in the sense that he talks about the dissolution of the kinship systems in his book on the European city. And then of course he has his Protestant argument. In some ways my enterprise is to flesh out a lot of that argument and explain how you could ever get a religion that is as individualistic and mentalistic as Protestantism.

And part of that is the origin of why the family shrunk. Weber doesn’t go down the road of pointing it to the Christian Church, a guy named Jack Goodie does that. So it gets you that far. Now your question about whether you would get Protestant religion even without Martin Luther, I think there’s good historical reason to think that would be the case. My argument would suggest that. But then also you have monks like the Cistercians which go all the way back to the year 1000, which are already adopting some Protestant-like stuff. And then you have various failed efforts to start Protestant-like new changes that look Protestant in different places. And I discuss some of those in the book. Even if Martin Luther hadn’t been around, I think you would’ve seen this percolation of new forms of Christianity.

Richard: Okay. Yeah, that makes sense. What do you think about, you close up in the book with a little bit about modern society and how to think about political issues. I think you talk about the American attempts to transform the Middle East, and how that wasn’t... Would it suggest that maybe we can transform other societies, or we can change cultures, but we have to focus on different things? So the war in Iraq was just, build the democratic institutions. And so maybe the idea was, if you want countries to succeed, you should just have a global campaign against cousin marriage. Just propagandize against that and try to do things like that. Do you think that there’s, instead of negative implications in the sense that here’s what not to do, do you think there are positive implications of your theory about how you can potentially improve people’s lives?

Joe: Well, yes. It does tell you what you would need to do. The suggestion is that you need to rewire societies from the ground up. So if you have the kinds of societies that exist in Afghanistan, for example, people’s entire livelihoods, notions of honor, security networks, old age social security type insurance is all built through these intensive kinship structures. It’d be quite a vast enterprise, because you’d have to figure out a way to replace all of that work essentially that these kinship structures do in order to transform that society.

Richard: Yeah. Is immigration a way to do this? There’s two sides of the immigration thing. You could say people’s cultures could come and they could change society, or it could be that society changes them. The western society changes the migrants. And it’s an empirical question, which cause outweighs the other. How do you think about that?

Joe: Well, we have good evidence now from immigration studies that migrants from around the world to the US or Europe or something, they assimilated in three generations or something like that. You can still detect the home country in the second-generation immigrants, but it’s a declining, weakening signal. If you have ethnic enclaves, these things can last longer and whatnot, but there’s another reason to be skeptical about the genetic differences that you mentioned, because immigrants do assimilate so effectively.

Richard: Yeah, I think that’s right. Although, I don’t want to get a digression into the genetics thing again. But if you look at things like crime rates and you compare it to the home countries, I’m just struck by the remarkable ... these consistencies, but I don’t want to go down that road again. What about, you have so much discussion of cousin marriage, but I was struck by how little discussion of the inbreeding problem there was. Isn’t just the most straightforward... You can have these more complicated theories about cousin marriage, but is the more straightforward problem with it that it just causes all kinds of genetic problems for the next generation?

Joe: Well, we do a little bit of analysis on that in a working paper we have that looks at kinship systems and GDP, focusing on satellite luminosity as a proxy for GDP. We can get nighttime imagery of the earth and use the amount of illumination in each pixel as a proxy for the kind of economic prosperity of that pixel. And then we can look at the traditional kinship structures of those areas based on anthropological data. And we made a big effort to link those up. And one of the hypotheses we took on is we said, well what if this is due to the negative cognitive effects of inbreeding? We have estimates for what those are. And the effects of inbreeding are quite small in terms of how much inbreeding you would need to see the kinds of prosperity differences that we find in the light luminosity. Whatever this is, it’s a small fraction of the total differences we see.

The other thing people always forget about is that there’s two sides to the inbreeding coin. One is the negative effects that it has on cognition, which it clearly does have some. But if you have relatively light inbreeding, you also have the benefits of altruism. So people help relatives more, so you’re actually building up a community and a stronger network. So if you think about a fitness maximizing problem, the fitness maximizing solution is not no inbreeding, because of the altruism benefits.

Richard: Yeah, that’s interesting. Yeah. Do you think about that in terms of just distant family, like your cousin is closer to you? Or do you think of that in terms of ethnic groups?

Joe: Yeah, you bring all your relatives closer, right? Because if you’re marrying cousins, then they’re producing children who have relatedness through multiple pathways.

Richard: Yeah. Do you think it makes sense at the broader kinship level? If you have one ethnic population that does a little bit of inbreeding and the other doesn’t, the one that does more inbreeding will be more ethnocentric?

Joe: Well, I don’t think that is likely to be the case. But I do think at the level of the family or clan, extended kin networks, that does work. The thing is kinship falls off geometrically. So by the time you get out to a tribal population, you’ve probably lost those effects at that level.

How Culture Influences Natural Selection

Richard: Okay. Yeah, that brings me to a question about your position in the group selection debate. Could you explain this? Because when I first started reading evolutionary theory it was like group selection is just pseudoscience, it’s nonsense. There’s reciprocity and then there’s kin selection. I’m not exactly sure, I read a little bit of your stuff, I’m not exactly sure where you talked about group selection as a bookkeeping enterprise. You say, well these societies, you could book keep at the individual or the family level, or the broader level. Do you have a fundamental disagreement about the science of evolution here, or is it just like you said, a way to understand data? Does that question make sense?

Joe: Yeah, I think so. I think that where the problems arise is with problems that involve social interaction of various kinds, and especially cultural evolution. But this could also apply to genetic evolution. What this kind of formal mathematical machinery shows is that you’ll often get many different stable solutions. For example, a typical cooperative dilemma, you can have reputational solutions that’ll stabilize cooperation in some group, but it’ll also stabilize non-cooperation in that group, and various levels of cooperation. So once you’re stable, the ingroup selection pressures are keeping that there. So you don’t have this thing, this free-rider problem that people talk about. In the simplistic undergraduate version of this description, there are genes flowing between populations, and when there’s a lot of cooperators in a group, defectors are favored because they can exploit the cooperators.

But in these more complex models, defectors are not favored. And ingroup selection, whether it’s genetic or cultural, wipes out those things. So you have competition amongst these stable equilibria, which can certainly work. And I think it’s very common, and that’s the argument in the case of cultural evolution. And it can go relatively fast, and it can occur if people copy other groups. And so there are these different ways it can occur. Migrants in a genetic model keep their genes for their whole lives and their kids have their genes. But in a cultural model, you can almost overwrite everything within a generation as you grow up. So you don’t have these lingering defector genes around, or lingering defectors. Yeah, I think that’s the biggest confusion is the selection among stable equilibria.

Richard: Yeah, that makes sense. I understand in the context of cultural selection, you have one stable equilibrium here, you have another stable one over here, they compete with each other. Do you have an idea, or is there a disagreement about whether that would create genetically a different kind of population? Because if you started out with any difference between groups, one group might lead to one equilibrium, another group might lead to the other equilibrium. That’ll be genetic and then that group will take over that. Or do you think that’s too small, these differences are probably too small relative to the cultural differences and how fast that changes?

Joe: Yeah. Well, I guess it’s an open question, but I feel like humans are pretty well mixed genetically. But the one place where that might not be the case is when you get these gene-culture interactions. For example, if you have genes that say you’re a cattle-herding population and you’re drinking milk, you could get genes that go along with that lifestyle. Like processing lactose. And then that group could spread by outcompeting other groups, because it’s got this nice package of cattle herding, milk drinking, which allows you to get nutrients from the animals. And then a bunch of things that go along with that. It’s kind of a gene-culture package, and that thing can actually lock up in a way that interacts the genes in a culture in a cool way. But there’s very little work on it.

Richard: Yeah, interesting. Have you read about, I think it was Cochrane and Harpening, about the Amish? And so the idea is you have these cultural traits, and basically they give people a choice. I guess you’d get a year or two years, or whatever, I don’t know what it is. But they let people basically have a chance to leave the community. And some do. So they have five, six kids and they lose one or two a generation. And the idea is they would be becoming more Amish across time. Is that the idea, something like that?

Joe: Yeah, I think so. Yeah, that could work. Or it could be... Yeah. So there’d have to be something about Amish lifestyle which was interacting with certain genes that exaggerated each other, or something.

Richard: Yeah. And that makes me think perhaps in a liberal society where you let people go off and form their own communities and do what they want, you might get more of this. You might have, okay, you can be Amish and you can be a Christian fundamentalist of a different kind, and you could be this. Some groups will have few kids, there’s many demographics who see their fertility plummet. But among those groups that do have some reproduction, you would have heightened... That gets me thinking, do you think that something like this is maybe...

Joe: It would always have to endure. In order to get any genetic evolution it’d have to endure, that’s the tricky part.

Richard: Yeah, exactly. So that would work with the Amish, they’ve been around for hundreds of years. Yeah, you’re right, it would have to... David Reich, in his book, argues that evolution happens at a little bit of a faster rate. Do you have a feeling about Reich’s work?

Joe: I think it does happen faster than some people have argued. Hundreds is pretty quick. It all depends on how much, what the trade is, and all that kind of stuff. So it’s all quantitative in that sense. Certainly, you get some amount of change in hundreds of years, so you just can’t get that much. Can you measure it? Probably. How important is it? Not sure. It’s probably the case that the genes that cause people to be more religious have increased over the 20th century in the US, just because religious people have more kids.

Richard: Yeah. And at the same time, religious observance has gotten down. So at least in the short run it’s having a canceling-out effect. But you think in the long run, the genes would probably...

Joe: Well, and so there’s that paper by Jonathan Beauchamp and others that suggest that education genes are probably under negative selection. But education has increased massively, so culture just keeps trouncing all over genetics.

Richard: Yeah. Although education is a strange trait, because we define this thing called education, right? It’s the same with intelligence, that’s also a social construct, but it seems more real than what standards are for people with a degree.

Joe: I would be interested in your take on this, but I think that’s true of every trait. We always have to decide what to measure.

Richard: So aggression, it strikes me as clear that some people are more, say, quick to commit violence, right?

Joe: You just did a very WEIRD thing in the psychological sense, the thing you nominated was a dispositional trait. But what natural selection is going to work on is aggression in fitness-optimizing situations. So optimally you should be aggressive when you need to be aggressive but not aggressive when it’s going to work against you. So you should be very context-sensitive. Anyway, it’s far more likely to be that aggression as a trait is something that is generally going to be under selection, because selection cares on being aggressive here, not being aggressive there. Right?

Reception in Academia

Richard: Yeah. We could talk more about this. I know your time is limited, so I wanted to get to, how has the reaction been to your book? Because I know you’ve thought a little bit about academic freedom issues, the political culture of universities. Your book I think is probably... It’s not like I think people are going to get super offended by it, but people get offended by a lot these days. And you’re taking for granted that western societies have succeeded in some way. And there’s an assumption that the success, western industrial civilization, is desirable. Maybe that’s not your characterization, but that’s the feeling I get from the book. And that is something that probably shouldn’t be controversial but is. Has there already been any pushback to your work on what you think is maybe political or ideological grounds?

Joe: Yeah. The main pushback, which pops up in a couple of the reviews – although most of them are positive, but it pops up in a couple of the reviews, there has been a little bit of scholarly debate on it – is just that there’s a strong desire amongst people from the postmodernist humanities side that want to put everything as glorifying European, the triumph of European societies, and this very old-timey frame that somehow there’s this ladder of societies and Europeans are at the top. And I think of the book as very much a reply to that. Saying that what led to the rise of Europe and global conquests, and the industrial revolution and stuff like that, had more to do with the fluky details of a particular brand of Christianity which set Europeans down a particular pathway.

Not judging it as morally better or morally worse, throughout the book I point out there’s two sides to every coin. For example, when I discuss analytic thinking, this section is called Missing the Forest Through the Trees, because analytic thinkers tend to focus on the bits and miss the whole. There’s another section called WEIRD People Are Bad Friends, because if you’re going to tell the truth in court, you might do it at the expense of a friend or some other close relative. So there’s all these virtues that you can see it either way. I’m trying to tell this very balanced story, but in any case, my critics want to ignore the fact that it’s very much a reply to people who might tell this old-time self-aggrandizing story. And just try to put it in that category, which is, I think it’s in some sense doing the opposite. I think that’s what the facts support. But anyway, I’ve had to deal with those guys.

Richard: Yeah. It’s almost, the marriage and family plan, it’s almost like the opposite. It’s like the church was selfish and wanted to break down family ties. It’s not like the Christian doctrine, because it was the most loving or most good, or whatever. It was they wanted to break down family ties and this was a way to... And you have this discussion about how they would manipulate people to give them all their money at their death. So in some ways it looks like these people, in early Western Civilization, had some negative traits and were doing some bad things, but it ended up as a by-product having beneficial effects.

Joe: Right.

Richard: Last question, what are you working on now? Is there another big project coming up? Can we look forward to another tome like these last few books in your future?

Joe: Yeah. One of the themes I develop in both of those books is the idea of the collective brain. And we talked about this at the beginning, I’m really trying to push that idea now and really think about all the ways in which this thing plays out. Whether it’s in what makes companies innovative, or explaining the last few hundred years of US innovation, why immigrants are so important to driving innovation, those kinds of things. And including in this innovation and thinking. So not only have we accumulated a body of new technologies and tools, we’ve accumulated more mental tools, and more ways of processing information that make it easier to search the thought space and figure out new ideas.

Richard: Interesting. And what about my idea of applying this to AI? Have you ever thought about doing some research in that area?

Joe: Well, I’m thinking about it. I’ve been trying to figure out where that’s going and what I think about that. Because in some ways I like to think of AI as just another technology. It can be another augmentation to our cognition, and we can get it to do cognitive work that was previously hard for us, like abacuses and calculators, and writing and things like that. So you can just think of it as another one of those. But we’ll see. I’m not exactly sure, I’m still at work on the book.

Richard: Yeah. Well, you can think about it as a technology, but also the alignment problem. I think a lot of people think, even if you could rip this up and you think this is the wrong way to think about it, they think that this is a monster that’s going to get out of our hands. So if you could say, well my theories make this more likely, or my theories make this less likely. Selfishly, that’s what I’d like, that’s the next book I’d like to read. But you’ll work on what you want to work on. Okay, Professor Henrich, it’s been a pleasure. Thank you very much.

Joe: Alright, good being with you.

---