Today, I have the pleasure of chatting with  Joseph Henrich, who is a professor of human  

evolutionary biology at Harvard University  and an author of two of my favorite books,  

The Weirdest People in the World and before  that, The Secret of Our Success. And I was  

just mentioning to you that I remember reading  this many many years ago when I was in college,  

and at the time, I didn’t think I would  get a chance to ask you questions about  

it. But the most proximal reason I wanted to  interview you is, I recently had your colleague,  

David Reich, on and we were discussing certain  things in the record of human history where he  

said, “Eventually, you’re just gonna have to have  Joseph Henrich on and ask him these questions,  

because he’s the one who would know.” So  let me ask you one of the questions which  

I was super intrigued by which he raised,  and we didn’t come up [with] an answer to. 

So one of the things he’s discovered through  his genetic evidence is that 70,000 years ago  

across Eurasia, there’s so many different human  species, from the Denisovans to the Neanderthals  

to the Hobbits, and then apparently, there’s  this one group, which was potentially the size  

of 1 to 10,000 people in the Near East, which  subsequently explodes, and now everybody who’s  

descended from Eurasia descends from this one  group. And so I guess the question is like,  

what happened? What did they figure out? A typical assumption when people think about this,  

if you put it in the Paleolithic, they assume  that it has to do with some kind of genetic  

changes. Now, Reich’s lab, there’s no obviously  big changes in the DNA, so it’s a little bit of  

a puzzle. Neanderthals, for example, had larger  brains, and in primates larger brains usually go  

along with more computational abilities, more  ability to solve problems. So the expanding  

variant out of the Middle East, out of Africa,  might’ve actually been less able at an individual  

level to process information. But if you look  back over the more recent period of human history,  

you can see that it’s a story of  expansions of different populations. 

So for example, in Africa, we have the  Bantu expansion about 5,000 years ago,  

which actually eliminates a whole bunch of  hunter-gatherer populations that previously  

existed in Africa. We have the remnant populations  in parts of the Congo, in the Kalahari,  

in the Hadza, for example, in Tanzania.  If you look at the Austronesian expansion,  

so that’s the peopling of the Pacific, that  was the expansion of one group of people at the  

expense of others, and of course, the Neolithic  expansion into Europe is another example. 

So really human history is a story of these  different expansions. And it could be that  

this expansion across Eurasia, which then led to  interbreeding, so we know it’s the same species.  

Humans interbred with Denisovans and Neanderthals,  as well as probably other species- there’s a ghost  

species in there. This could be just institutional  changes, so if you have institutions, for example,  

that interconnect your population, you can  maintain more sophisticated technology. And  

some paleoanthropologists, for example,  have speculated- with some evidence- that  

the expanding populations had projectile weapons,  so bows and arrows. And humans have periodically  

gained and lost bows and arrows in different parts  of the world. So in Australia, for example, bows  

and arrows are never invented. In the New World,  populations probably didn’t have bows and arrows,  

but then later develop bows and arrows. It’s also really interesting how  

some technologies you would just think of  as extremely… I don’t know what the word is,  

but the new world, like not having the wheel or  something. I guess it kinda makes sense with no  

domesticated animals, but again, it’s like such a… Right. Although the New World, everybody has dogs,  

and you can pull carts with dogs. So I’ve never  really bought that, and of course, you can, you  

can use llamas in the New World to pull  carts as well. People do that today. 

Yeah, so what’s your explanation for  why there’s no wheels in the New World? 

So there were wheels on Mayan carts or  on Mayan toys. My explanation is just the  

collective brain. So almost every single first  invention of something big that we think is  

important for humans was invented in Eurasia. And  Eurasia, as Jared Diamond famously pointed out,  

building on other people’s work, it’s  the largest continent by far. It has the  

biggest population. It’s also oriented along an  east-west axis, which allows ideas and people to  

more easily flow. And there’s a belt, which the  historian Ian Morris calls the Lucky Latitudes,  

which runs from basically southern China all  the way through to the Mediterranean. And ideas  

are just flowing back and forth there, across  the center of Eurasia. But you also ended up  

with more complex state bureaucracies and the  kinds of things that allow you to organize and  

move people around and whatnot. And what’s the explanation for  

why the collective brain leads to state  capacity? Or is it the other way around? 

Well, you can think of institutions that  eventually lead to state capacity as just part  

of the innovative process of the collective brain. So if you have more groups experimenting with more  

different ways of governing  groups of people, gradually,  

you get the accumulation of the pieces that you  can put together into different kinds of states. 

Got it. Okay, so going back to what happened  70,000 years ago, is the basic answer like,  

it could be something like bows, but we  don’t know exactly what it was? Or is it- 

Yeah. We definitely don’t know exactly what it  was. So we know that this population expanded,  

and there does seem to be some tool indications  to suggest more complex technology. Probably,  

technology usually goes along  with social organization. 

So for example, if you look at Australia, which  is a continent of hunter-gatherers, there’s an  

expansion about 6,000 years ago out of northern  Australia which eventually takes up seven-eighths  

of the continent, and they had a new social  organization, including rules about who you marry. 

You had linguistic exogamy, and rituals that  interconnected populations. So rather than having  

local rituals and local myths, many communities  would get together periodically to initiate the  

young men, and this would help bond that whole  group. There’d be an exchange of technology,  

and teaching that goes on at these… they’d  spent a few months in the same place so there  

was a lot of time for transmission. And then in terms of mechanism, so,  

if it’s the case that you have these  sort of convalescing waves of expansion,  

and David Reich’s lab’s evidence suggests  that this expansion was quite violent,  

based on whether the genes are passed down through  the maternal line or the paternal line. When you  

think about the mechanism in which technologies  develop or these social institutions develop,  

how much of it is like, look, there’s all these  different groups that are trying different things,  

and one of them maybe figures something out, and  then they just kind of explode? And how much of it  

is that within each group, they are experiencing  sort of accumulations of learning over time,  

and it’s not necessarily a selective process,  it’s more of a sort of accumulative process? 

Well eventually the groups have to meet,  and they’re gonna compete over territories,  

and a bunch of different things can occur. So  groups can copy each other. So we know that  

from the ethnographic record and historical  cases, sometimes a group will say, “Oh,  

those guys have really good technology  tools.” Maybe they’ll get some migrants,  

or something like that, and then they can adopt  the practices. So that definitely happens. 

But there’s also plenty of evidence of violent  conflict. And the Reich labs evidence and lots  

of other ancient DNA suggests that there  are these dead-end genetic lineages. So,  

I mean, the Neanderthals are a dead end in  a sense, although we interbred with them,  

so they’re in us in some sense, but they don’t  have their own pure lineage or anything like that. 

And if you had to guess, whenever this big  wave happens, is it more of a sort of concrete  

technology? I mean, especially if you consider  the range of expansion that can often happen,  

right? It’s like literally that same group goes  from Siberia to England. It’s hard to imagine it’s  

a single technology lets you explore this wide  range. But it’s also the idea that some cultural  

artifact is enough that it gets transmitted over  tens of thousands of years and that’s what gives  

you the edge. I’m curious how you think about,  what could possibly be this driving engine? 

Well the way I describe it in The Secret is that  it’s a package of things. So another good example  

would be the Inuit expansion out of the north  slope of Alaska. And they expand all the way  

across the Arctic and eventually get to Greenland.  And they have a whole package of social practices  

which helps keep them interconnected, and then  they have bows and arrows, which the group they  

are exterminating as they go along, the Dorset,  doesn’t have. They have dogs and sleds. They have,  

along the coast they have boats, so they’re doing  whaling. So there’s a whole package that puts  

together that allows them to out-compete  and eventually exterminate the Dorset. 

And one of the things that happens, though,  is the Dorset probably had better technology,  

but they expanded, they spread out, their  languages diversified, they lost contact,  

and they began losing technology.  So you wanna see this as a dynamic  

pulse. There were probably expansions and then  collapses, another expansion, another collapse,  

right? It’s not one giant long march to victory. And then this process, which has happened many  

times where the population gets cut off or  something: When you say it’s a sort of cycle,  

is there some reason why, over time,  that knowledge gets fragmented and breaks  

apart and populations disperse? Yeah, I think there’s a cultural  

evolutionary dynamic that’s part of this, because  languages will naturally diversify and then as  

they diversify, there’s less contact. People  sometimes get inclined to marry endogamously,  

and especially if there’s enough people around  locally, why go all the way over there? And those  

people are getting more culturally different,  so they’re seeming a little bit like outsiders… 

So the tough part that humans have always  had is to stay unified because of the  

natural effect of geography and learning  locally is gonna tend to fragment us. 

In the Secret, you described this interesting  startup problem where if you don’t have that much  

accumulated cultural knowledge, developing  the ability to do social learning isn’t as  

valuable. But if you don’t have the ability  to do social learning, you don’t have that  

much accumulated cultural knowledge in your  tribe or group. So how is this problem solved? 

Yeah. So before I get to solving the problem, I  just want to sketch for the listeners that the  

question is, why is this cumulative cultural  evolutionary process that is so important  

for humans relatively uncommon in the natural  world? It seems like just our lineage. I mean,  

there were a bunch of split-off  lineages, but now it’s just us. 

So to understand that is this idea that you just  mentioned, where you’re imagining an increase in  

brain tissue that’s gonna be costly, and I  can put that towards individually figuring  

the problem out for myself, or I can put it into  learning from others. And in a world without very  

much cumulative culture, there’s not gonna be very  much useful information in the minds of everybody  

else so I should use that brain tissue for  individually solving problems. And so it’s hard  

to get this runoff where their brains get bigger  for the purposes of learning from others. And then  

the question is, how do you get past that valley? So the case I make in Secret is that, say, three  

million years ago, two million years ago, there  were several factors that came together. The first  

factor is the rate of change of environments.  So you get this increase in the fluctuation  

of environments, so you’re getting more  environmental changes. And in cultural evolution,  

a lot of theory shows that there’s a certain  rate of change which is favorable to cultural  

evolution. It’s gotta be slow enough so that  the information of your parents and the previous  

generation is useful, but not so slow that  you might as well just encode it in the genes. 

So that’s one. Second thing is, we’re a  ground-dwelling ape, which means we have  

hands like chimpanzees and gorillas, and we can  potentially use tools and whatnot. But unlike  

them, our ancestors may have been savanna-dwelling  apes, which meant we may have lived in large  

groups. So in mammals, mammals live in larger  groups when they have to deal with predators  

and the predator guild in Africa at that point was  quite thick. There were a lot of deadly predators. 

So paleo-anthropologists think that  our ancestors may have lived in  

large [groups of] savanna-dwelling apes. And if  you have a lot of individuals, if the culture is  

sparse, the bigger the group, the more chance  that there’s someone doing something useful in  

the larger group. And so that means it’s easier  to get across the threshold. So those are three  

of the main factors that might have allowed our  lineage, as opposed to all the other lineages  

around, to cross this. We’re already a big-brained  primate, we had hands, we’re living in these large  

groups on the savanna, and the climate changed  during this period, so as to make us, yeah. 

And how big were these groups? Well, I mean, nobody’s really sure,  

but maybe a few hundred individuals. Okay. And before the agricultural revolution,  

was there still this transmission of  information across different groups? 

Yeah. So lots of different evidence suggests  that groups were moving trade goods. So well  

back into the Paleolithic we see trade.  Often we see some genetic transmission. 

Right, the expansions and so forth. Yes. And so, talking about the expansion,  

that’s something where the ancient DNA is  useful because the Neanderthals in Europe,  

the DNA suggests they lived in very small groups.  But the DNA of the expanding groups suggests a  

larger population. So those would’ve been two  different collective brains there as well. 

So in The Secret, you discuss a lot of these lost  European explorers with modern technologies as  

of at least a couple centuries ago encountering  peoples who have for tens of thousands of years  

discovered ways of hunting and processing foods  and so forth that, without which even these people  

with modern technology will starve. And as I was  reading that I was wondering: I’m not sure how  

I understand the process by which, if there’s  a 10-step process to making sure this bean is  

actually nutritious, and without any one of those  steps, you might poison yourself or something. And  

at no point do you understand why this process  works. You don’t have a scientific explanation.  

How do you even learn that in the first place? Right. So one of the things we know, even in young  

children there’s a tendency to preferentially  learn from healthier, more successful individuals.  

So if you’re processing it better- so something  like bitter cassava, which has cyanide in it. If  

you just eat bitter cassava, it won’t taste  great. So if you then rinse it somewhat,  

it’ll taste better and maybe you could eat it, but  you’re going to accumulate cyanide over the long  

run. So it doesn’t kill you right away. But if you  do this whole long process that the populations  

in South America developed, then you’re  totally fine, you never get any accumulation. 

So you can imagine that initially this is gonna  be very strong and people are gonna sort of do  

sensible things… but then it gets a little more  mysterious. And we know this because bitter  

cassava gets transported to Africa, and Africans  immediately begin eating it improperly processed  

and getting goitre and the cyanide processing that  goes along with it. So then that’s gonna be a slow  

evolution where groups that do this are gonna be  more successful and individual families are gonna  

be more successful. So, for example, you might  have a household where they process the cassava  

more seriously than another family, and they’re  gonna have more kids and they’re gonna be like,  

“Oh, that family’s really good.” And then  people copy what they do in all kinds of ways,  

but one of them could be copying recipes. Hmm. I guess it depends on the mechanism  

of selection here, because when you consider the  different ways in which two different individuals  

might be different or two different households or  even two different groups, I guess it makes sense  

why cyanide poisoning is such a deleterious effect  that it is a noticeable or quite a strong signal.  

But I think you discussed some other ideas in The  Secret where it’s like, the sort of spices in a  

region sort of match the antimicrobial properties  or the antifungal properties you need to stay,  

I guess, clean or whatever in that particular  region. But it sounds like a small effect,  

and how could such a small effect actually  create a strong enough signal that when  

you’re deciding who to copy, you notice  that this family is healthier because they  

spice their food in a certain way? Or this  group takes over the area because they’ve… 

Yeah, I mean, it has to be a big enough  effect to matter, but I think if you go  

back to a world where there’s a lot of improperly  processed meat, people don’t have refrigerators,  

and the leading cause of death in children is  diarrhea. People carry high pathogen loads.  

If you can knock that pathogen load down, we  know from modern research if you wanna make  

people healthier and even smarter, right? IQ  goes up if you knock the pathogen load down. 

Got it. So this process of cultural  accumulation you’re talking about:  

you think it is not strong enough to pick up  minor increases in fitness and it actually  

has to be quite a significant thing for it to… I think it operates a lot like natural selection,  

in the sense that natural selection will pick up  tiny things if you give it long enough. Cultural  

evolution will do the same thing if you give it  long enough, because just small differences in who  

you’re paying attention to will affect things, but  it might take 1,000 years as opposed to decades,  

right? But we know culture can spread adaptive  traits super quickly if it’s a really big effect. 

Yeah. So the situation in which our ancestors  found themselves, in some ways it’s sort of like  

epistemic hell, in the sense of you don’t know  why certain things are working, but you do know  

that if you break with tradition, you might just  doom yourself and your family. And that maybe,  

as you discuss, causes these religious beliefs and  taboos and mystical understandings of the world to  

rise, where you just think like, “Look, you’re  gonna burn in hell if you don’t do this 10-step  

process for refining your beans or something.”  In some sense, this requires you to abdicate  

reason because you have to just be like, “I don’t  understand this. this is the way it’s been done,  

so we’re gonna do it.” Basically: how much  reason do you necessarily have to abdicate  

to survive in the ancestral environment? I mean, one of the ways we research this  

is by going out to societies that make bows  and arrows or have food taboos that protect  

them from dangerous marine toxins and ask  them questions about why they do it and  

see if they [understand] and they don’t  understand the underlying causal stuff. 

Right. But then why do they say they do it? Well, a typical answer is, “It’s our custom.”  

or “It’s important to people around here to  do it this way, so we do it this way.” So  

just Hadza bows, for example- I didn’t do this  research, Jacob Harris and Kim Hill and Rob Boyd  

did it- but they asked Hadza about their bows and  about how they work and what the mechanics are,  

so they understood some stuff. So you use the bow  and you get some mechanical understanding. But if  

you asked him, “What if you used a different wood,  different materials,” and they had never tried  

anything else but what they learned, how to make  this bow. So they couldn’t speak to a lot of that  

stuff. And the things about the compression of the  wood is very important- didn’t understand that. 

Okay, so that’s quite interesting, right?  Because you would think that you have to  

transmit cultural knowledge over time, but you  also need to experiment in order to innovate.  

But it sounds like because of the belief in these  customs, you would be less inclined to innovate. 

Right. And the thing is, once something  gets good, doing it differently almost  

always makes it worse. And then there’s also,  some things are just different because people  

do it randomly different, like they make a  mistake. So I think people often underestimate  

the power of error in generating novelty. One question I have is why this process of  

cultural learning… So humans have extended  childhoods, presumably to give us more  

time to accumulate culture, and then we live after  menopause, presumably because older people, our  

grandparents, can teach us about the situation we  find ourself in. And 18 years is such a long time,  

and when I think back to- I mean, obviously I  didn’t grow up in the ancestral environment,  

but when I think back to what I was doing as a  teenager I don’t think I was learning that much. I  

was just less productive for no obvious biological  reason. So I guess I don’t intuitively understand  

why this process of cultural learning takes  decades, and why it can’t happen more rapidly. 

Well, you probably, as an adolescent,  were in school, right? So you were  

probably accumulating some stuff in school-  maybe. And here’s an interesting fact about  

hunters. So anthropologists who studied hunters,  hunter-gatherer societies in different places,  

and the physical peak that at least males have is  in their early 20s, right? That’s when they run  

the fastest. They got the best eyesight. But the  best hunters in the community are 36 to 40, right? 

They’re not as fast as those guys, but they just  know, they know the track, they know the animals,  

they know exactly what to do in these different-  And then hunting skill begins to decline because  

basically the physicality of things begins to  catch up, and so there’s this kind of cycle.  

And at 18 young hunters aren’t even producing  enough food to feed themselves. So it’s not  

until they get into their 20s that they’re  actually in surplus and bringing food home for  

everybody else. And you need to know hundreds, at  least, animals, you need to know animal behavior,  

you gotta be able to spot tracks and spoor,  and we’re just very knowledge-dependent in  

terms of our hunting and gathering. Yeah. Given what you just said and  

also the experience of these European explorers  suggests that you just gotta know a lot of shit  

to make it in any sort of environment on Earth.  But then, there’s all these other animals and  

they seem so dumb and they seem able to get by.  So why did humans have such a hard time of it? 

Yeah, well, so I think we offloaded a lot  of stuff into culture. So in one of my Lost  

European Explorers, there’s a case where they’re  in Australia and they have camels. And the camels  

escape, and now central Australia has lots of  feral camels. So the camels survived the lost  

European explorer challenge, because they  have innate instincts. They can smell water  

a mile away. They can detoxify foods in their  own- they have a complex digestive system that  

detoxifies- we’ve lost all that, and we’re  worse than chimpanzees at detoxifying foods  

because we have all these cultural practices  that do the work. So we’ve externalized  

detoxification and a lot of digestion, actually. I had this independent researcher and internet  

writer called Gwern Branwen on my podcast a  couple months ago, and we were discussing AI.  

And I asked him- so his theory is very much like  the brain became bigger, more intelligence. And  

so then I asked him, “Look, if it’s this simple,  why did it take so long for evolution to discover  

intelligence in the first place?” And he had this  interesting answer, which is that in terms of the  

signal that evolution is giving you, there’s a  very narrow gap between skills that are so useful  

that they should just be distilled as an instinct  into your genome, and then skills that are so  

worthless that are not worth learning in the  first place. And so this narrow gap that’s like,  

you need that generalization ability. It’s not so  primal that, you know, you’re not gonna culturally  

learn hunger, it’s just gonna be in your genome.  I don’t know if that sounds like an interesting  

explanation or helps explain anything. Well in cultural evolutionary models,  

a typical thing would be to let genes compete  with individual learning to compete with cultural  

learning. And it turns out the rate of change  of the environment affects that. So if the  

environment is changing slowly, then you should  put it all in the genes. And if it’s changing  

at moderate speed, then culture is the best  way to go, and if it’s fast, then individual  

learning. And the idea, the individual learning  is favored because if it’s changing too quickly,  

the previous generation doesn’t know anything  worth knowing because the world you’re dealing  

with is just so much different from their  world. And so it’s this intermediate range. 

So myself and others have argued that the  increase in the frequency of change during the  

Plio-Pleistocene transition 2.5 million years ago  is a change that increases the value of cultural  

transmission of learning from others. I’m not sure  what your guest was thinking. I mean as our brains  

are not very good at solving problems, otherwise  the lost European explorers could survive. And  

human brain size has been declining for the last  10,000 years. So we’ve actually been getting  

dumber. Fewer neurons, less computational power. Why has brain size been declining? 

Well, the collective brain argument suggests  that at a certain size, you’ll begin farming  

off specialists. So because there’s a store  of knowledge in the society, and we can all be  

generalists and learn how to do all the different  skills. But at some point it makes efficiency  

sense for us to specialize in different skills. In order for that to be the case though, we have  

to have social agreements of some sort that allow  us to trade or exchange things like that. So,  

but then once we’re specializing, we don’t  necessarily need as large a brain and we, because  

we distribute the overall brain power amongst the  society. So it could be that we’re becoming more  

of a superorganism. And you see the same thing  in ants. When ants get specialized occupational  

casts, their individual brains shrink. Yeah. I think David Reich’s lab had a  

result a month or two ago where they showed that  the selective pressure on at least the European  

samples, which is the samples they studied,  was that there’s been selection for greater  

intelligence over the last 10,000 years. I don’t  know if you saw that or what you make of that? 

Well, they didn’t say anything about intelligence.  They did use the polygenic score for education.  

But there was no education 8,000 years ago. Which is correlated, right? 

Yeah, it’s correlated. But then the question  is, we don’t know what goes into that,  

right? Is that actually computational software?  I mean, people do well in school because they  

stick to it and they can sit in the  same place for long periods of time. 

I guess it would be interesting for them to  sort of deduct the docility hit polygenic  

score from the education, because we don’t  have a polygenic score for intelligence yet,  

right, so which is why you need some proxy. Well there are IQs, so there are polygenic  

scores for IQ. I think they actually did that  one as well, like you said, it’s correlated with  

education. But the things that go into giving  you an IQ, I mean, the correlation between  

brain size and IQ is only about, well, across  populations it’s .24, so it’s not very big. 

The other thing is IQ is massively misunderstood,  so the way to think about IQ, which has been going  

up over the 20th century, by quite a bit, so if  you rescale modern scores to 1900, it’s about 70,  

it’s called the Flynn effect. I think that  those are cognitive abilities for navigating  

the institutional world that we’ve constructed. And a huge mistake would be to assume that  

those are the right cognitive abilities for  the next century. So it’ll be a different  

constellation. We study herders and cognitive  abilities of herders in northern Namibia,  

and if you’re in northern Namibia, you gotta be  able to move through the landscape, so being able  

to just pick a direction and know where you’re  going and not get lost is super important. 

So different suites of cognitive abilities  are favored in different environments. This  

idea that there’s this generalized  thing applicable across all human  

environments just doesn’t… I mean, education  massively increases your IQ, right? If you’re  

uneducated, you have a totally different IQ. And then is the way in which your horsepower  

is directed, is that a thing that you think is  basically set by the time you’re an adult? Or,  

if somebody’s 30 now and the AI thing is gonna  happen next and they have to totally reorient  

away from knowledge work, is that a thing? Well, I think it’s a continuous scale. If you  

look at human brains, they’re developing and  continuing to add new connections and stuff,  

at least into the mid-20s. Now, there could be  even more plasticity after that. Unlike chimps,  

we’re still not totally myelinized for  our whole lives.

So there's definitely room. It’s just there’s less room.

The kind of social learning for which our learning biases are fit, right?

Like where you pay attention to elders because they survived somehow

and they’ve accumulated the knowledge from past generations,

how much do you think that actually basically applies

to the way in which the modern knowledge work economy accumulates knowledge?  

And how much do you think is just an artifact of the  kind of environment our ancestors found themselves in?

Right now we’re in Silicon Valley, or in  San Francisco, and it’s very common here for  

a 20-something to make a new product or service  which, without that much- they don’t have that  

much context of how the world works, but they  can make big, big innovations and big changes.  

But at the same time, people on average maybe are  more productive in their 40s and 50s in terms of  

wages or something. So basically, this sort of  social learning you’re describing, how much does  

it actually describe the world as it exists today? Yeah. And so, super important, so we’ll take the  

patent database and if your same-sex parent- has  to be same sex- patents in a particular domain,  

you’re nine times more likely  to patent in that same domain. 

And how do you separate the genetic  versus cultural effect there? 

So these are very fine domains, so we’re talking  about, like, natural adhesives versus synthetic  

adhesives. So unless you think there’s genes  for that. And same thing with transistors  

and electronics, so very fine domains. And if you  grow up in Silicon Valley, you’re much more likely  

to patent, in general, but you’re probably gonna  patent in computers. But if you grow up in Boston,  

it’s gonna be biotech. So having a father or  mother who did biotech and then growing up in  

Boston, you’re even more likely to patent in  biotech. And this is true even if you move to  

New York. So if you look at people in New York  who are from Silicon Valley and from Boston, the  

Bostoners are more likely to patent in biotech,  and the Silicon Valley kids are more likely to  

patent in computing. So all the same rules apply. Well, I’m not sure because maybe the location,  

you learn a lot from the environment.  But the fact that it’s just kids, and so,  

if you think these old rules of cultural  accumulation and collective brains apply,  

you would think, “Ah, you’re 50 by the time you’re  writing your first useful line of Python code or  

something.” When in fact, a lot of these big  innovations come from people who are much,  

much younger, before they should have accumulated  much of the know-how that this theory implies. 

So the model here is what to focus your  efforts on, so it’s the throwing model.  

So if you grow up in Silicon Valley,  you focus your efforts on learning how  

to code or whatever. And so that means that’s  where you’re likely to make the innovation. 

Right. But in terms of, you know, when there’s  2% growth a year and a lot of the technologies  

that are dominant in the economy today, or a  lot of the industries, didn’t even exist many  

generations ago. Basically, does that suggest  this model of- should we just mistrust our  

instincts of who to give prestige to and what  kinds of people we want to pay attention to,  

whether it’s professors or elders or whatever? Yeah, so that’s definitely something I’ve thought  

about and I wrote about in Secret, which is that,  as the rate of cultural change gets faster- and  

we sort of talked about this with environmental  change- the value of older and older members of  

the previous generation declines because the world  that they grew up in, and that they honed their  

skills to, is quite different from the current  world. So you would expect the degree to which,  

I mean, optimally, you would look less  far back or you would look to relatively  

younger individuals to get your inspiration  from because the world they adapted to is  

closer to the world you’re gonna need to adapt to. Yeah. And speaking of Silicon Valley, let me ask  

you a little bit about AI. So one of the reasons  to suspect some incredibly sharp discontinuity  

from the world as it exists today to a world with  AIs. And by AI, I don’t mean just, like, GPT-4,  

I mean, like, replacement at least for anything  you can do on a computer screen, the AI can do. 

One of the reasons we expect this hard  discontinuity is that they have potentially  

the step function increase in social learning and  the ability to accumulate knowledge that maybe  

humans had, or the magnitude of which maybe  humans had when, between them and non-human  

primates. And in particular, the fact that you can  just copy everything you know. Like you don’t have  

to teach a young person, right? The constraints  of biology mean you can’t just replicate brains. 

You have much more efficient communication.  You don’t have to communicate through words,  

you can just shoot your brain state across, the  population size can grow arbitrarily large. So to  

the extent of, the collective brain is the size  of the population and how interconnected it is,  

do you just expect we wouldn’t even  recognize the kind of world these AIs  

could make as a result of their cultural skills? Yeah. No, I mean, I, I definitely think it’s  

pretty interesting, and holds great potential for  expanding the collective brain. There are little  

things in there which might make one worry.  So if you study the history of innovation,  

you find out that, for example, serendipitous  meetings are super important. There’s a great  

paper on Silicon Valley showing that companies  will cross-reference each other’s patents more  

likely when the people at those companies tend  to frequent the same coffee shops, and they  

track people on their cell phones and stuff to  figure this out. So serendipitous meetings are  

important and improper copying. So a huge number  of innovations are mistakes where somebody copied  

incorrectly and then got something better. I read this interesting theory that maybe  

evolution designed transcription and translation  to have more errors than it could otherwise have,  

just so that you can have this sparse reward  when you’re close to the right sequence. 

Yeah, yeah. I think that’s a very interesting  area of research, and it makes good sense to me. 

I’m not sure of the current state of the  evidence, but there are… Different parts  

of the genome are more or less susceptible  to mutation, which is kind of interesting. 

Yeah. So going back to AI, maybe then  another way to phrase it is, look,  

you’re talking about these serendipitous meetings  where you can learn something another person  

knows. And the great advantage these AIs have  is that they can sort of meet with everybody at  

once. Future versions of AIs, you could really  imagine holding the whole internet in context,  

right? I mean, we’ll be the equivalent of those  people isolated in Tasmania according to the AIs,  

right? Because they just have everything in  context. You can get a PhD in every single field,  

and you can amortize that across all your copies. Yeah, so are you sort of banking on, in the next  

10 years, you’re just gonna be living in  the singularity or something because of your  

belief in the value of cultural knowledge? Yeah, I see there’s various potentials,  

and I’m particularly interested in using AIs to  augment problem-solving in human groups. So you  

can imagine getting the humans together because  the humans have the big advantage of having stuff  

they care about, right? There’s stuff they  want to invent. So the AI is still a tool  

at some point. So I’m interested in that, but  I’m interested in how these things like running  

out of training data is gonna be dealt with,  the value of making mistakes and serendipity,  

if you get rid of all those things or how you’re  gonna reintroduce them, those kinds of things. 

Yeah. I mean, it’s quite funny because until  recently, people were saying the big issue  

with LLMs is that they hallucinate  and make mistakes. And at some level,  

hallucination is no different than creativity. Well, may- maybe there’s a way to harness that,  

right? But maybe we just  didn’t know how to harness it. 

That’s right. What do you make of the idea  that you could have AI firms or AI tribes,  

whatever way to think about it- is if the  effective population size that can communicate  

with each other is such an important contributor  to how much progress a group in history was able  

to make. If you could just literally run billions  of AIs and they have this immediate ability to  

communicate with each other. And again, I’m not  talking about current models, I’m talking about  

future, human equivalent. Basically I guess  I’m just throwing out a bunch of different  

intuition pumps and I’m curious which one you  find most promising or most interesting, or did  

you just find all of them not as convincing? Yeah, no, I don’t have strong opinions about  

any of this, but I do think that one thing  in order to make all that work is, humans  

are constantly getting hit with shocks, right? So  there’s economic shocks, there’s weather shocks,  

there’s conflict with other groups, there’s  pandemics. And the shocks have big effects on how  

things go. They inject new information into the  system. So I’d worry about a system that is too  

homogeneous and doesn’t have enough noise adding  shocks and new challenges and things like that. 

One of the things missing from our conversation is  the kind of creativity that cultural evolution has  

figured out. So to give you an example, at some  point in human history, religions invented big,  

powerful, moralizing gods. And this may have  increased the people’s ability to cooperate.  

So forget about the technological element. Some  of the most important features of innovation  

over human history have been institutional  and things that get people to cooperate. 

So in my work, I argue that one branch of  Christianity resulted in the transformation  

of the families in European societies into  small monogamous nuclear families. I’m  

not sure an AI would have thought of that.  Cultural evolution thought of it because of  

how it affected how the societies operated,  so it sorted this out over historical time. 

Yeah. Can I play with that idea? Yeah. 

Because that’s quite interesting. So again, I  think this might even reinforce the advantages  

of AIs in this sense. So, to the extent that  this sort of random fluke by the church led  

to this modernity and the great divergence  or whatever, and it was a result of, there’s  

a bunch of variation, and then you can select  over some group that’s doing the right thing. 

The advantage you have with AIs is that you  can have much more high fidelity replication  

of culture and you can explore a wider range of  potential cultures. Sorry, and it sounds really  

vague when you talk about AIs in this way. So  what do I mean? One problem companies have today  

is that, suppose a company’s working really well.  Maybe it’s early Apple, or early Tesla, or SpaceX  

or something. And suppose it has to do with the  culture, the people, something, it’s not clear  

how you replicate that culture, not only across  different institutions, right? If I’m running  

a company, I don’t know how to copy SpaceX. Not  only can you not replicate it across institutions,  

you can’t even replicate it across time. So over time, many institutions tend to decay,  

their culture fades as people leave or die or  something. And then imagine if all of SpaceX,  

at least maybe at the time you  thought it was the most effective,  

are just AIs where you know every single byte,  and when you can make a copy of it 1,000 times,  

and you can put it against every single hardware  problem we have in the world. And then you can,  

if you think there’s another team that might  have some different culture which is better,  

you can do the same thing with them. And culture  here includes not only how they think about  

their work, but also even how does the board make  decisions, right? Do they do this Monte Carlo tree  

simulation? And there’s so many different things  you could do here. So because of the wider range  

of possibilities you can explore and the fact  that you can have higher fidelity transmission  

of cultural information, maybe the ability  to do this random evolution also increases.  

But I’m curious to get your take on that. Yeah, as a kind of broad sketch, I think that  

is pretty interesting. I worry about too high a  fidelity replication just because it’s important  

to take the details of historical context  and time into account. So that same thing,  

if you fix it, it might not really work a decade  later when everything else around it has changed,  

right? So it’s kind of a moving target.  But you could fix that just by having  

lots of different variants that are different  versions of that. So yeah, that would be cool. 

Sorry, that gives us a good opportunity to go  into WEIRD. But before we get there, one more  

question about before the agricultural revolution.  You know, obviously many groups around the world  

were different and many of those differences  were probably inspired by the fact that they  

were living in different environments, so they  needed to come up with different technologies  

and adaptations to survive best there. How much  just random differences that had nothing to do  

with their environment do you expect there to  have been, in terms of… I don’t even know what  

the right category would be, but in the modern  world, you might imagine something like, does a  

society have slavery or not? And maybe pre-modern  societies all had slavery, so that’s not the right  

way to think about variation, but that kind of… Yep. So I think there’s lots of reason to think  

there was lots of variation. And one of  the ways that researchers study this is  

they look at ethological variables  and they compare it to phylogenetic  

variables. And here they mean cultural phylogeny. So if your ancestral populations had matrilineal,  

matrilocal social organization, how likely are  you to have it, controlling for ecology? And it  

turns out both of these things matter. People  adapt to their ecology, but you can still see  

the signal of past society. So some degree of  fidelity of transmission of social institutions,  

how we make our baskets, those kinds of things.  There’s lots of examples in the modern world,  

and we see this all over the place. So if you  look at gender inequality in the modern world,  

you find that a history of the plow leads  to greater gender inequality. So males had  

a particular advantage in using the plow  because it requires upper body strength. 

This meant males became the dominant force in  economic production at the household level. And  

that even today in populations where most people  aren’t farmers, this persists. So this is cultural  

persistence that has to do with whether you had  the plow. So this is work done by Alberto Alesina  

and Nathan Nunn. A former postdoc of mine now  at the Harvard Business School, Anke Becker, has  

done a similar thing with pastoralism. So if your  ancestors were pastoralists, and pastoralists, for  

various reasons, have quite strong gender norms  that persists into the modern world and still  

shapes, for example, female entrepreneurship. Interesting. Yeah, and the reason I’m especially  

curious about this is because it informs the  following question. If the Industrial Revolution  

didn’t happen in Europe but started somewhere  else, how different would the world look like  

today? So obviously you discussed the fact that  breaking apart these kin ties was necessary for  

the Industrial Revolution, but to the extent that  was necessary, whichever place had the Industrial  

Revolution first would have had that, right? But,  I mean, separate from the technologies or cultural  

practices which were necessary for the Industrial  Revolution in the first place, still how different  

would the world look? How much variation  was possible given our level of technology? 

Yeah, I have a lot of trouble trying  to answer that because, I mean,  

China or some place in the Middle East might  have been the obvious alternative place for  

the Industrial Revolution to happen. But  I feel like you have to give those places  

a lot of stuff that developed only in Europe. So for example, universities begin spreading in  

the High Middle Ages. So you’d want them to have  universities. You’d want them to have universal  

schooling, which began to spread and wasn’t  present in these places prior to that. So that  

begins to spread in the 16th, 17th century.  And so by the time you add all this stuff,  

it basically starts to look a lot like Europe. And these are all things that are global now,  

right? So the universal schooling that  we find around the world today, you know,  

begins with the Protestant Reformation in Germany  and then later England. University’s models  

are the European university, and globalized. I guess one obviously very salient example of  

variance which might not have been replicated,  but I’m curious if you think it might have been,  

is, it seems like the British Empire was  the first major institution which decided  

that slavery is just morally wrong and we’re  gonna throw our weight around to eliminate it.  

And I don’t know if you think that sort of  naturally follows the development of social  

technologies that the Industrial Revolution  would have brought about, but that seems super  

contingent. But I’m curious if you disagree. Yeah. Well, so my story is that the rise of  

the Industrial Revolution in Europe  has to do with the consolidation of  

Europe’s collective brain. And one of the  things that requires is trust in strangers,  

and at least the beginnings of moral universalism. 

And it’s that moral universalism that eventually  causes the British to say, “We’ve got to stop  

with the slave trade thing.” It’s a moral decision  that they made because it’s no longer consistent  

with the changing moral values over time. Right, right. Okay, so we’ve been dancing  

around the thesis of your book following The  Secret, which is The Weirdest People in the World.  

And before we really jump into it, maybe you can  give me a summary of what the thesis there is. 

Yeah. So the first observation is that there’s  a great deal of global psychological variation  

around the world. So European, American,  Australian populations tend to be highly  

individualistic. They’re inclined towards  analytic thinking over holistic thinking.  

They have a lot of impersonal prosociality, so  trust in strangers. They’re against conformity,  

willingness to cooperate with strangers. So  the question is, how can we explain the global  

variation in these features of psychology? And towards the end of the book,  

I actually connect these features of  the psychology to economic differences,  

including the Industrial Revolution that happened  in Europe, which reshapes the world. And the  

story is that the key event is the spread of  a particular form of Christianity into Europe,  

where the Catholic Church- what becomes the  Catholic Church- systematically dismantles  

the intensive kinship systems in Europe,  leading to small monogamous nuclear families. 

And this transformation leads to the creation  of new institutions. So by the high Middle Ages,  

you get the rise of guilds, which are voluntary  groups of craftsmen and self-help societies,  

because people don’t have their families to  rely on. People begin moving around. There’s  

occupational sorting into different occupations.  You get urbanization rising. Charter towns are  

on the rise. Universities pop up. New kinds  of monasteries pop up. And then Europe begins  

to urbanize, and you get new kinds of law that  are based around the individual. Contract law.  

And then eventually this leads to a lot  more innovation because ideas are flowing  

around Europe and then eventually the Industrial  Revolution. So that’s the argument in a nutshell. 

Can you explain again what exactly  the Church did which led to the  

kin-based existing system breaking down? Yeah. So the Church outlaws polygyny,  

and so that stops elite males from having multiple  wives and concubines and whatnot, and creating  

kind of a giant family through that. It outlaws  cousin marriage going all the way out to sixth  

cousins at some point, and that included spiritual  kin and other kinds of non-genetic relatives as  

well as the cousins. It frees up inheritance, so  it has inheritance by testimony rather than normal  

patrilineal inheritance. And a simple example  here is in most societies, you inherit access  

to land corporately, which means you and your  brothers and stuff all own the land. And it might  

be your uncle is actually in charge of the land,  your father’s brother. And so you can’t sell it,  

you can’t move it around, and you’re also tied  to it. It’s where your ancestors are buried,  

and so there’s this big importance of land. So the  Church allows people to give land to the Church,  

the Church becomes the largest landowner in  Europe, because you can do it by testament. 

So those are just some of the examples of the ways  it transforms the family structure, and eventually  

you get these monogamous nuclear families which  are basically unheard of around the world, at  

least if you look at the anthropological record.  And you can really see this when you can compare,  

individuals can move to European cities as  individuals or nuclear families. In China when  

you move to the city, you maintain contacts  with the folks back home in the village,  

and people flow back and forth and you get these  little enclaves of different clans and stuff in  

the cities, the connections. And this is  really important because your ancestors,  

and there’s rituals that have to be done  back in the home community. But Christianity  

does away with all those ancestor rituals. Okay, great. Let’s jump into it by starting  

before- much before- even the church. I wonder  how much of the things that make Europe weird  

existed even in antiquity. If you read about  how Roman society worked, or how these Greek  

polities worked, already you have an emphasis on  nuclear families. You have these sort of universal  

norms around, “Hey, we believe in republicanism,”  or, “We believe in democracy.” And the structure  

of the system matters more necessarily  than… I guess at least before the empire,  

right? So yeah, was Europe already weird  before the things that its church did? 

Well, there were certainly some interesting things  going on in Greece and Rome. But I don’t think  

there’s good evidence that you had the kinds of  monogamous nuclear families that you would find  

later. I mean, even European law is built around  patrilineage. So if your father’s still alive, you  

are not a full citizen. You’re in the pater house  of these large families. Definitely intensive  

kinship, for sure, patrilineal, patrilocal. Women, of course, don’t have any rights.  

And republicanism, but the formation of Rome is  built around a series of elite families. So it’s  

a clan operation, and they call it republican  because the clans all have some say in what’s  

going on. So people sometimes see representative  government where it’s not really individuals being  

represented, but families. And so this fools  a lot of people into thinking there was a lot  

more individualistic voting and things like that,  whereas there’s actually no voting. So in Greece,  

in Athens, Athens is unique because it does  a bunch of things in the laws of Solon,  

that break down the intensive kinship. So for  example, in Greece, you get monogamy for the  

first time, and Athens is considered unusual.  So males can only marry one Athenian woman,  

which has potentially positive effects among the  competition among Athenian men, but they can have  

as many slave concubines as they want. So when  Christianity spreads, it ends the whole slave  

concubine thing, which was also common in Rome,  and that’s another effect on this whole thing. 

Yeah. But I guess the extent to which these  practices were already formed hundreds or  

thousands of years before the church’s  intervention, maybe this suggests that  

Europe was already on this trajectory and the  church didn’t necessarily move the needle that  

much. Or what do you make of that thesis? Well we know that the… in the book, I talk  

about data where we can look at contemporary  kinship practices, and we can look at the  

number of centuries that that part of Europe  was under the church… so we have a database of  

the diffusion of bishoprics, and there’s a clear  connection between the intensive kinship practices  

and the number of years under the church. Hmm. And then if you think about what the  

church basically did, I think your work is often  used to justify this idea of Chesterton’s Fence,  

where if you don’t necessarily understand a  cultural practice, you should keep it around  

because… the stuff we were talking about, The  Secret, where you don’t understand why the 10  

steps lead to the bean being edible, but you  should trust the sort of wisdom of the ages. 

If you think about what the church did here,  right? Like, isn’t this sort of like a… not  

necessarily a justification, but at least an  example of just breaking down Chesterton’s  

Fence? The church doesn’t really understand  why these kin-based networks that have existed  

for thousands of years might have utility. It  still just gets rid of all of them. And that  

results in this lottery ticket that leads to  The Great Divergence. So is there evidence for  

anti-Chesterton’s fence from the Weirdest People? Sure. Sure. So what you’re talking about is just  

the idea that culture has imbued institutions and  various practices with a logic that we might not  

understand. So it’s not calling for never changing  the institutions. It’s saying make sure you figure  

out the logic and the cost and benefits. So I was an expert witness in Canada for  

the attorney general of British Columbia,  asked the Supreme Court of British Columbia,  

whether the statutes against polygyny were  legal. And so my job was just to inform the  

court that polygyny has this unfortunate effect  of the elite and high-status males tend to get a  

disproportionate share of the wives, and that  creates this pool of low-status unmarried men. 

Now you can decide what you want to do with that.  But you need to understand the kind of underlying  

social dynamics that you’re gonna unleash  if you legalize or decriminalize polygyny. 

Right. But again, going back to the Church, the  Church didn’t understand the positive or negative  

effects they’re gonna have, right? They just like,  “Let’s rip the Band-Aid, let’s see what happens.” 

That’s right. And it actually benefited  the Church because it released people’s  

responsibilities to their families and allowed  them to migrate in and join the Church, and more  

heavily invest in the Church. And these later  prohibitions against priests marrying and stuff  

was all an effort to get greater investment in  the Church, because you weren’t torn between  

helping your son and then investing in the Church. Hmm. And then that end result which benefited the  

Church, was that their conscious intention  in breaking apart these kin-based ties? Or  

was that just the accidental byproduct? Yeah, I mean, in the record, I’ve never  

been able to find much evidence that they were  thinking about the destruction of kinship ties.  

There is a great quote from Saint Augustine  where he talks about the benefits of forcing  

people to marry more distantly. But, you know,  this is something that was debated repeatedly in  

church councils all across Europe. And  it doesn’t seem to pop up very often. 

And the Church doing this, for obviously selfish  reasons, was that in any way related to why the  

Church spread through Europe in the first place?  Or did it end up spreading for separate reasons  

and it won out over the other potential religious  competitors for separate reasons? And basically,  

was this part of the selective mechanism  which allowed the Catholic Church to  

become dominant in Europe in the first place? Yeah, I think so, although it’s hard to know.  

There was a lot of randomness going on in the  diffusion of the Church. But also as the Church  

arrived, these places, over a period of time,  became more successful. So you can see rising  

urbanization in the wake of the Church’s  arrival. That would have meant more trade,  

their citizens’ rights began developing because  they were trying to attract citizens to the towns.  

That would have given the Church more cachet,  ability to move. There’s also the appearance  

of these voluntary associations which are new  kinds of monasteries. And so the monasteries  

begin spreading throughout Europe. But they’re all  linked in a network because, like the Cistercians,  

they’re all connected and they have big meetings.  So they’re actually spreading a lot of knowledge  

around Europe as well. So that’s part  of this kind of collective brain story. 

And if you compare what’s happening in  Europe at this time to the rest of the world,  

so starting with China: in 1500, the population  of England, where a couple centuries later the  

Industrial Revolution starts, is three  million, and then in Ming dynasty China,  

it’s somewhere between 100 million to 160 million.  And if we take your previous perspective of the  

size of the collective brain really matters  a lot, the size of the collective brain in  

China just seems so much bigger. So what is  the best way to understand what went wrong  

here? Why weren’t they able to use their… Right. So when you’re thinking about China,  

the first thing to remember is that for  a lot of the history, we can actually  

see that size difference mattering a lot. So a lot of European invention, or a lot of stuff  

used in Europe, is flowing in; gunpowder, paper,  printing press, stuff, is flowing into Europe. And  

so okay, so then what happens after about 1000  CE? Well, the argument is that the destruction  

of the kinship group opens the floodgates to  people moving around. So a recent analysis that  

we’ve done is, after a bishop arrives at a 1.5 by  1.5 grid cell in Europe, people begin flowing in  

and out of that grid cell more. And what we do to  calculate that is we have a big database of a few  

million famous people. And we have birth and death  locations. So we find out that the Church arrives  

and suddenly people are free to move around. So you have a flow of individuals around,  

and you have rising urbanization. So Europe  passes China around 1200 in the percentage of the  

population that lives in cities. And cities are  where a lot of the action is, cities and towns.  

And you have the diversification of occupations.  So normally clans would specialize in different  

crafts, and you’d learn from your clan brothers  and clan fathers how to do the occupation. In  

Europe you get guilds, and you get masters and  apprenticeships developing where strangers will  

become an apprentice to a master, learn from him,  move somewhere else as a journeyman, learn from  

that master, and then eventually set up a separate  shop. Lots of opportunities for a flow of ideas. 

So it has to do with how the kinship  system transformed the movement of people,  

the rising of urbanization, the nature of  guilds, and then eventually you get universities  

and things like that. So it greatly intensifies  the interconnectedness of the collective brain,  

and the amount of cognitive diversity. What do we know about India during this period?  

Because from what I’ve read, from a perspective  of how much written history we have, it’s kind of  

a black box. But we know that there was trade  between India and other parts of the world,  

and we know it had a big population. Do we know  why India before the Mughal period or before the  

British period wasn’t… yeah. What was the effect  of not having maybe Abrahamic gods or.. with the  

kind of other cultural practices India had? Yeah, definitely not an expert on this one,  

but some of David Reich’s evidence suggests  that the caste system is quite old. Because  

you can actually see it in the genetic system.  And the caste system is not good for innovation,  

because if you happen to be good at another  skill but your caste doesn’t do that skill,  

you can’t switch over. So it’s gonna prevent  the sort of available genetic diversity.  

Complex families, intensive kinship, there’s  reason to think that those things were all  

important. Yeah, so, patchwork of polities. But there was lots of interesting ideas,  

actually, that are developed in India, and they  move into Central Asia, eventually end up in  

the Islamic world, and then get into Europe. So  for example, Arabic numbers are actually Indian  

numbers. Zero was probably developed… I mean,  Indians were huge with numbers. It’s kind of  

interesting. Perhaps related to the religion. Not that this would be a justification for the  

caste system, but is there any  way in which the specialization  

it engendered would be good from the perspective  of this collective brain perspective? 

So specialization is good. And so the interesting  thing about most human societies is specialization  

automatically- or not automatically- it tends  to evolve along some kind of kinship line. So  

in Oceania, there were different clans. In  each clan, there’d be like a canoe-building  

clan and a warrior clan. So that allows  specialization, but it doesn’t allow you  

to harness the genetic diversity, because  it’s not like the canoe-building clan had  

special canoe-building genes. It’s just that  you would pass down this cultural knowledge. 

Now, a better system, but it seems hard to  evolve, is one in which we all opt, select,  

into occupations in which we think we’re good  at. But for that world to exist, you need a  

world of voluntary associations with emphasis  on the individual rather than on the group.  

So this is the world that evolves in Europe,  once you demolish the intensive kinship units,  

because otherwise you’re gonna get castes and  clans and all these things we see elsewhere,  

where we see a division of labor, but it’s  transmitted through a kin network, the knowledge. 

And why wasn’t there another sort of stultifying…  because we began the conversation by talking  

about, there does seem to be this cycle where  over time, because of cultural differences,  

because of kin, whatever else, over time groups  will tend to diverge. Languages, culture, other  

things you mentioned, tend to diverge over time.  Why didn’t the same thing happen in Europe where,  

let’s say, in the third century or fifth century,  the Church starts doing this stuff, and then by a  

thousand years later, if you’re part of this guild  or whatever, now that’s become like the new kin,  

and now there’s actually less mobility again? Well, so there was a lingua franca, Latin,  

and so intellectuals would all write and  communicate in Latin for a long time. So  

even though they were speaking different  dialects of French and German and whatnot,  

they were able to communicate in Latin. And  Christendom basically formed an overarching  

network that helped. And one of the key things  that this world religion does, and other world  

religions do it too, is you have to marry other  Christians, but it dissolves the tribal line. 

So Europe has tons of tribes in the  pre-Christian world, but because of  

how you have to marry other Christians, you had,  you know, Celtics marrying Franks. And in fact,  

the early arrival of Christianity into England  is when a Frankish princess is marrying an Angle  

in Kent. And so they’re marrying back and  forth. And this is gonna dissolve the tribes  

because intermarriage, “what’s the kid?”, you  know? So that’s how you get rid of the tribes,  

but you need a world religion to do that. Right. And so it seems like there were two  

very important features that were relevant here.  One, it seems like obviously the competition  

between different European groups which gave  incentive for monasteries or universities  

or cities to attract the best people and  make all these advancements. But secondly,  

the fact that they were descended from a common  empire, from the Roman Empire. Basically, in a  

world where the Roman Empire never existed, do you  still have this collective brain in Europe emerge? 

So I think the Roman Empire plays a big role,  because, for example, there were Roman roads in  

parts of Europe which allow people to flow. And  you have communities which were part of Rome and  

relatively sophisticated. Now, they go into a  bit of decline, but there’s still the memory of  

the empire. So, you know, Charlemagne wants to  be crowned emperor of the Holy Roman Empire in  

800 even though there’s not very much left of it. But it was still an ideal, and that Carolingian  

Empire actually had real effects because they  worked with the pope in order to enforce a lot  

of the marriage and family programs. And the  Carolingians had their own agenda because they  

were trying to use these marriage and  family practices to weaken some of the  

other aristocratic families. But they had the  religious tools that they could put to work,  

and if you didn’t have those religious  beliefs, then you couldn’t put them to work. 

Sometimes people defend ancient conquerors  based on a similar idea of they spread  

knowledge around the world, and they set up  these lines of communication. People say that  

about Alexander and then causing Hellenistic  expansion and trade and so forth. They say  

that about Genghis Khan and the Silk Road. They  say… I mean, obviously this led to the Black  

Plague, so… so it’s tough where that nets out. They say that about Napoleon and the Napoleonic  

Codes and so forth. Where do you basically  come out on that story of: have the great  

conquerors been good for the collective  brain or bad for the collective brain? 

Well, I’ve never actually focused on that.  I mean, I guess my immediate reaction was  

there must be a better way to transmit  the knowledge than the whole conquest. 

And then I guess there’s also the open question  of how the disruptions that are caused by them…  

How much did they set back the- And population decline, spreading  

disease and stuff, so certainly  it could have been done better. 

Right. And going back to David Reich’s work, If  your theory implies that there should have been  

a decrease in genetic similarity in Europe  as of, after the fifth century or something,  

as a result of the culture, the Church’s  practices, do we see that in the genetic record? 

Yeah. Yeah, so Europeans are quite well  mixed compared to other populations. And  

so we were talking about India. You know,  if you do a principal components analysis  

of this two-dimensional plot of the genetic  similarity you can actually see the class,  

and people are pretty spread out. If you put the  Europeans on the same plot, they’re well mixed. 

Right. But do we see the period of greatest  mixing starting in the fifth century? 

Yeah, good question. I don’t know the answer  to that question. So that would be… That’s a  

nice piece of evidence, so… You know, there’s a  couple of groups working on medieval ancient DNA,  

so hopefully we’ll have more answers on  that question. I mean, there is enough  

now of pre-Roman burials, so Bronze Age type  stuff, showing that early Europeans definitely  

had complex, intensive kin groups, patrilineal,  patrilocal resonance kinds of stuff, polygyny. 

And then stepping back, I’ve read there’s  so many different theories of the Industrial  

Revolution and of modernity. And there’s Robert  Allen, the economic historian who thinks it’s a  

result of higher wages because of the Black Death.  There’s Gregory Clark, who thinks it’s because of  

this positive eugenics in England where the upper  classes were having more babies. I mean, there’s  

like 20, 30, probably hundreds of different  theories of why it happened, where it happened,  

and when it happened. And so each one of these,  when I read them, sounds plausible. I don’t have  

a knockdown argument against any of them. But I’m  not sure how to think about which one is correct,  

and yet, if we’re stacking all of these up  against each other… why you think this story  

is any more accurate than any other story  that people tell? Is it overdetermined? 

Well, I mean, the first thing to think about  is, those other authors- possibly Greg Clark  

is an exception, but they don’t think about  the psychological variation that exists. So  

there’s good reason to think there’s all  these psychological differences, and you  

have to believe they don’t cash out. And if you  think they exist, then you have to explain them,  

and so you need a theory that explains them.  So Bob Allen, he’s a blank on the psychological  

variation. So let me give you some evidence that  I think is pretty good evidence. So we have a  

database of the diffusion of Roman bishoprics  through Europe. And then what we do is we look  

at what happens when a bishopric arrives. And  I mentioned before, you get a flow of people. 

But another fact is, if you look  at the production of famous people,  

these places where the bishopric has arrived start  producing more creative individuals relative to  

non-creatives. So there’s an uptick, and it keeps  going up for centuries, so the relative increase,  

and so that suggests you’re producing more authors  and inventors and writers and whatnot. And then if  

you take each of those, and you correlate it with  modern patents, places that spent more time under  

the church produced more patents between 1980  and 2014 based on European patent data. And this  

holds if you just compare regions within the same  country. So you can see long-term effects of this,  

and you can see immediate effects in the  historical record on the production of creatives. 

So what would you make of a theory that says,  “Look, all of these theories are basically  

correct. All of these stories are basically  correct.” And it’s them stacking up that leads  

to the Industrial Revolution rather than any one  of them being the most important proximal cause.  

And maybe there’s reasons, like, let’s say there’s  10 of these stories and you had to stack them all  

up and China had five of them, but they also  had maybe independently, like three other ones,  

but they also had two stories which go  against them or something. And it’s like,  

no one of them is that proximately important, it’s  their combination, right? So, I guess your theory  

seems compatible with Robert Allen’s that coal  being cheap and wages being high was important. 

Well, I guess one of the things that I think is  a mistake is to focus too narrowly on England,  

because England was benefiting from ideas that  were flowing in from France. There was a lot of  

great science being done in France. And, I don’t  think Gregory Clark’s right. I can explain almost  

everything he explains with cultural evolution,  and he doesn’t even really take that seriously. 

Oh, interesting. Yeah. So patience,  

one of the things he argues is patience. In WEIRD,  I use the exact same data- I actually get the data  

from him- to show that you have this increase in  patience. But we know that people can culturally  

learn patience, and this can all be culturally  transmitted, and in this world, it leads to  

more success. So if you’re the kind of person who  doesn’t waste a lot of his money and we begin to  

value thrift and stuff, which Protestantism does,  then we should expect there to be an increase in  

fewer murders and lower interest rates. What’s the reason to think that… I mean,  

his main explanation, I think is genetic, right?  Is there some reason to think that that is less  

likely to be as important a factor as culture? Well, we know that populations- migrants- into the  

US and Europe shift their psychology over a few  generations of being there, and we know there’s  

been natural experiments done by economists  like Chris Blattman where you actively try  

to teach people, train them essentially, to  discount the future less. And that seems to  

be culturally transmittable. So this suggests  that- I can’t rule out that there’s been any  

genetic evolution, but I can show that culture  can operate on this quickly and powerfully. 

Right. I think one of his key pieces of evidence,  I don’t remember the exact numbers on this or even  

the particular detail, but it was something  like if you look at the inheritance of wealth  

or titles or- I forget what the exact thing is-  the correlation across generations points to the  

kind of pattern you would see with genetic  inheritance rather than cultural spreading. 

Yeah. He doesn’t actually fit any cultural  evolutionary models, and if you include that  

people are learning from parents and learning  from their social milieu, your parents determine  

where you are in the social milieu, so you’re  surrounded by people who are being more patient  

and behaving in certain economic ways, then  that’s gonna have a huge effect. I just told  

you about how if you grow up in Silicon  Valley, you’re gonna patent in computers.  

It’s not because you have genes for computers. Yeah. Given the persistence of these cultural  

effects, how should that change how we  think about whether it makes sense to  

have more immigration from societies that are  already WEIRD or whether that matters or not? 

Right. So we have a paper under review right  now in which we show that from the 1850 roughly  

to 1940, a big driver of US innovation both  the quality and quantity of patents, is the  

cultural diversity of counties. And we actually  use immigration as a kind of way of showing the  

causal effect of this. So what you want is a lot  of cultural variation. Now, if you get people from  

societies that are more distant from whatever  the current US culture is at the time, it’s  

gonna take more time until they’re able to fully  integrate. So if they’re coming from [a] very  

distrusting society, they’re not gonna be able  to latch into the collective brain immediately. 

So you see this in around 1900. There’s data from  people, mostly coming from Southern Italy and  

coming from Germany and Britain, and the Southern  Italians are coming from a society that has a high  

distrust of strangers. So their immediate effect  on the patenting system is low. But the strength  

of these cultural traits seems to go down by an  order of magnitude each generation. So the second  

generation, you can still see where they came  from, but they’re a lot more like the majority  

culture than their parents. And then it goes down  another order of magnitude. So yeah, so you just  

have to do more work essentially to assimilate  people, enculturate them. But the potential value  

of having diverse ways of thinking, different  ideas, can be really valuable, I think. 

What do you make of, so Garett  Jones is an economist who has this,  

he calls this spaghetti theory of assimilation.  I don’t know if you’ve heard of this. The  

idea is basically, look, immigrants do  assimilate, but they also assimilate us,  

and so whatever cultural practices they have will,  in proportion to their size of the population,  

have an effect on wherever they migrate to. And his example is when Italians came to America,  

spaghetti, which is traditionally an  Italian dish, also became an American dish,  

right? So they assimilated American culture  as well. And sub in traits like trust or  

whatever there. Basically, this sort of  reciprocal assimilation has implications  

of whether we want immigration from non-WEIRD  societies. What do you make of that idea? 

I think that the spaghetti example or the pizza-  pizza or spaghetti, both good examples- is a  

great thing, right? Because American food is  a fusion of cuisine from lots of societies,  

and that leads to quite good food. Same thing  you can see [in] music. Things like jazz come,  

have African rhythms in them. And so  rock and roll is along that lineage of- 

But we don’t want, like, do we want the  diversity in low trust and high trust? 

Well, the question is, do social interactional  traits operate the same way as these things?  

And if you have ethnic enclaves, then you’re  gonna get low trust in those ethnic enclaves.  

You get things like the mafia, right? So you don’t want that. But if you had  

an immigration policy that distributed  people, and in a high-trust society,  

you benefit by being higher trust because you can  do things like collaborate and start companies  

and stuff, which you have to be high-trust to  do that. Otherwise, you don’t do it in a very  

effective way. So there is a pressure for  low-trust people to become more high-trust.  

But if you’re in mostly high-trust people,  there’s not an incentive for you to move down.  

And social interactional traits are  different than things like food types. 

In psychology, there’s been this problem of  the replication crisis where a lot of the  

main results are hard to replicate and it’s  not clear how much of this is real science.  

Have you looked into how many of the WEIRD  results are based on, of the differences  

in psychology between different populations and  so forth? How much of that actually replicates? 

Yeah, my sense is it replicates quite well.  So we published our paper in 2010, and in, oh,  

late 20 odds, Armin Falk, who’s an economist,  and a bunch of others, Ahnke Becker, Ben Enke,  

published a paper where they measured economic  preferences in 80,000 people around the world.  

And that just showed big variation in things  like patience, various kinds of reciprocity,  

altruism, so the kinds of things we  would expect. And then other large-scale  

projects have similarly shown lots of variation. The groups we have today, which are not WEIRD,  

when you study them, do you think that the  fact that they have resisted modernity for  

so long suggests that they actually are weird in  their own way? Which is to say that this is not  

representative of the way that Europeans might’ve  been thinking in the second century; rather it’s  

like the Hadza’s particular weirdness leads them  to be averse to modernity in a way that is unique? 

Well, that’s always a concern, and it’s especially  a concern when you’re studying hunter-gatherers  

because a lot of the hunter-gatherers that are  left in the world today are [in] places where  

agriculturalists couldn’t easily move to. So that’s something we’re thinking about  

all the time. The ways around that are to  go to places where agriculturalists didn’t  

go or couldn’t go. So we have a lot of good  ethnographic history on people in Australia,  

and then we have the Arctic populations, which,  you know, big sections of Paleolithic Europe were  

probably kinda like Northern Canada is today. So  at least environmentally it’s not crazy, because,  

you know, Ice Age Europe. And places like the  Aleutian Islands, the West Coast of California,  

we have lots of ethnographic evidence,  and that there was no agriculture there. 

So just putting together all these different  lines of evidence help us develop a picture.  

I wouldn’t wanna bank everything on the Hadza.  Of course, we can’t do experiments with some of  

those groups. But we can see whether the Hadza  look like these other hunter-gatherer groups. 

Oh, speaking of the Ice Age: do you have a  take on why before the Ice Age agriculture  

didn’t develop even though genetically there  probably weren’t that big a difference between,  

like, 20,000 years ago versus 10,000 years ago? Yeah, I don’t have a clear picture of that. It  

is the case that the Holocene was a particularly  long stretch, a particularly long interglacial.  

There were some long interglacials, but the  soonest one before that was 120,000 years ago. 

Other than that, they were getting broken up. I  sometimes wonder if we may someday figure out that  

there was actually a little bit of agriculture.  And it would’ve all been destroyed by the Ice Age,  

right? So it’s not clear that there would be  any heritage of it left. So we would have to  

find some remnants of some domesticated crops. Although I was asking David Reich about this,  

and I forget the reason he said, but he did say  that if there were societies before the Ice Age,  

we would have evidence of them in  the archeological record or their  

genetic record or something like that. Yeah, if they had gotten to any scale. But  

we know from modern societies,  there are groups, for example,  

in the Amazon that have root-based agriculture  that wouldn’t leave any archeological record. 

This is a question I also asked to David  Reich. There’s, you know, with modern  

LMs and just generally with newer techniques for  processing information and understanding evidence,  

there’s a potential for answering questions  that maybe we couldn’t have done before,  

especially from maybe a cultural perspective  where you can actually consider lots of different  

cultural artifacts at once in the context of you  can run cultural simulations or something. Maybe  

those are more fanciful ideas, maybe there’s more  practical ideas. But if there’s one question you  

have which it’s sort of bottlenecked by having the  right data or being able to process the data the  

right way and future LLMs or AIs could help us  there, does something immediately come to mind? 

Well, this gives me a chance to tell you something  we’ve been working on. So we were talking about  

this kinship hypothesis, the idea that kinship  intensity affects psychology. And when I presented  

the ideas that are in The Weirdest People in the  World, I get hard times sometimes from historians  

who will have some kind of very European-specific  story about why this happened that has to do with  

European royalty or coal or something like that. And so my move is to not try to get into the  

weeds and start reading Latin texts, but to go  test it somewhere else. So if it’s true that  

kinship intensity should work like this, then  we should be able to test it somewhere else.  

So we have a large corpus, two different corpus  from China. And so we have late-imperial China,  

something called the Gazetteers, which were  produced across the prefectures, and they had  

a stylized content. So this is the same genre,  thousands of books, and then we have about 7,000  

books going back 2,000 years in Chinese history.  And the techniques that the AI is allowing us to  

do is to get measurements of psychology from  the texts. So we take a questionnaire. This is  

a technique developed by my post-doc, Mohamed  Attari. We take a psychological questionnaire  

that’s been validated, it’s in English, and  we wanna get an equivalent in ancient Chinese. 

So we run it through a semantic similarity  comparison, and we look for quotes from the  

ancient Chinese corpora that match each  sentence in the English corpora. And then  

we rebuild the questionnaire in ancient Chinese.  So that’s our psychological measure for something  

like individualism or collectivism or moral  universalism. And then we take each book or each  

paragraph in each book, and we do a comparison,  cosine similarity comparison, between the two  

sets of text there, and that allows us to stamp  each paragraph with a measure of individualism  

for example. And then we do that for the entire  corpora. And this allows us to track psychological  

change across space and time in China. And then we  can correlate that with kinship intensity, and we  

get the same correlations that we do in Europe. Interesting. How worried are you about the fact  

that, if you think about modernity, was a result  of finding this one cultural variant- at least  

according to your story- which then helps us  develop the better technologies, all these  

new institutions and so forth. How worried  are you about the fact that the modern world  

because of the spread of WEIRD, has just much less  cultural variants and because of this monoculture,  

a potential variant which might be useful  in the next transition just wouldn’t emerge? 

Yeah, I am worried about that, because just  the destruction of languages. So we’re losing  

languages left and right. English has a particular  form that you don’t find in lots of languages,  

so that’s just an example of the kind of cultural  diversity we’re losing. So that’s definitely a  

worry. I do think that we’re getting new variants.  Japan adopts a lot of WEIRD institutions. But it’s  

really creating a new third unique thing. It’s  got a bunch of Weird elements, but Japanese law,  

despite being the same as the US, operates  very differently. So unlike Americans,  

Japanese don’t tend to sue each other and they  tend to use mediation and things like that,  

even though the superficial law is pretty  similar. So I do think there is creation of  

novelty which could be useful. But yeah, in  general, I worry about the loss of novelty. 

I also should say that I think that the sort of  impersonal world of impersonal institutions in  

WEIRD psychology is quite fragile. For example, in  the US in rural areas, you see moral universality  

declining since 2008. This is based on John  Haidt’s YourMorals.org data. And so there’s  

an increasing difference in the morality of  urban areas versus rural areas in the US. 

Hmm. I’m sure you get asked about this a lot, but  fertility seems to be declining around the world,  

and there doesn’t seem to be any existing  cultural variance, other than maybe the Amish,  

who can resist it. Do you have takes from a  cultural anthropologist perspective of what’s  

happening here? And a follow-on question here  is, should we really encourage these cultural  

enclaves like the Amish to the extent that  this anti-fertility meme is so viral that  

they can’t be part of the common culture and  still preserve fertility, you really need  

this closed-off societies in order to preserve  fertility. How seriously do you take that idea? 

Well, I definitely take declining fertility  pretty seriously. And especially if you’re  

a collective brain guy like me, just having  fewer people is gonna be a big problem. And I  

think that there will be spread of ideologies or  religions or something which are pronatalist, and  

those groups will have a big advantage in cultural  evolution because a community that is pronatalist  

tends to produce more pronatalist babies. Christianity spread using that. Mormonism  

spread in the 19th century using that trick.  So I just feel like cultural evolution is gonna  

find the combination and there’s gonna  be some pronatalist groups spreading. 

There are a lot of different societies in  the world today. Is there some explanation  

of why none of them have the existing  variants necessary to keep fertility high? 

Well, from a cultural evolutionary perspective,  this is relatively new, right? So the demographic  

transition is only really late 19th century. Lots  of the world is just getting hit after 1970. And  

we also have things like rising female labor  force participation which is gonna stifle it,  

rising female education. And so once that  stuff maxes out, then you’ll see variation  

among different groups in terms of the  number of babies they produce, right? 

And this thing unfolds in demographic time scales,  so we’re not gonna see it for a few generations,  

but there’ll be some group somewhere that  will be producing more babies than everybody  

else. And the reason why I think religion is  the likely one is because people do things  

because they think God wants to do it. And if  people come to think that God wants them to have  

more babies because it’s a way of worshiping  him or getting to heaven or whatever their  

religious configuration is, then that’ll be  a group that produces more babies. I mean,  

Catholics were defying the demographic transition  for a while. They just seem to have stopped. 

Right. How worried are you about the fact that…  I mean, you discuss in WEIRD that throughout this  

period in European history, European states are  at war most of the time. How worried are you about  

the fact that, because of a decline in war between  great powers, that this build-up in cultural  

mutational load that you were talking about or  maybe where things just get less efficient over  

time and there’s no selective pressure to get  rid of that inefficiency. How worried are you  

about that just making governments or nations  less efficient over time because there’s no  

outer loop loss function that says if you mess  up enough your country might not exist, or your  

group might not exist? Which we were talking  about the waves of Yamnaya or the Anatolian  

hunter-gatherers just conquering everything. If  that doesn’t happen, will we just see culture  

and nations and governments degrade over time? I think the answer to that is yes. The one way  

that countries or whatever the political  institution is could address that would be  

doing this thing that I’ve curiously called  “domesticating the competition”. So sports  

teams constantly renew themselves by competing.  Firms live and die and renew themselves over  

time. Like you said, companies will start  off being super great and then they get too  

big and then they get kind of inefficient  and then eventually they disappear. None  

of them last forever. And that seems to be  true of political units. But in principle,  

you could have at least some renewal processes. So  democracy potentially provides a renewal process,  

although there are things like the institutions  that the government builds that are hard to renew.  

So one simple political idea is I think when  you create a new department, it should have  

the same thing that cells have where at a certain  time, they time out and you gotta make a new one. 

And that’s because just the way human  bureaucracies, institutions work is  

they kind of corrode from the interior, from the  inside, just the way cancer spreads in a cell. So  

you just gotta kill it and make a new one. So we  could institute that. But I guess the idea hasn’t  

quite caught on yet. I mean, there are bits and  pieces of it around, so it’s not like it’s unused. 

Yeah, I guess maybe the thing that’s less  understood or is not appreciated is that  

the reason that over time in history  institutions have improved is because  

of the selective process that at least at  certain levels of selection no longer exists. 

Yeah. I mean, the Roman Empire didn’t  say, “You know, we gotta redo everything.”  

“Why not adopt democracy and we could have a  republic.” Nobody does that. Things fall apart. 

The researcher Stuart Armstrong has this idea  called Chesterton’s metafence. And here’s what  

it states: In our current system, democratic  market economies with large governments,  

the common practice of taking down Chesterton’s  fences is a process which seems well established  

and has a decent track record and  should not be unduly interfered with. 

Can you say any more about what the author has  in mind? I would like an example or something. 

Yes. So I think this is a rebuttal against this  common idea that we shouldn’t be arbitrarily  

changing- like suppose a new technology  comes about and we’re worried about the  

risk it might pose to society. Suppose the young  kids are doing something different culturally and  

we’re worried about like what effect that  might have on the culture and so forth. 

He’s basically saying, “Look, this has been  happening for five centuries and this process  

by which new cultural variants enter the mix  and so forth has worked really well for us,  

even if it’s happening at a rapid clip  and we shouldn’t interfere with it.” 

Well I don’t think I have strong feelings  on this. I’m always focused on trying to  

understand the process of cultural evolution,  and it is true that people often resist cultural  

change that in retrospect we think is good.  But of course when you begin to make cities,  

you lead to all kinds of epidemic diseases. So  if you’d anticipated that, you might have worked  

on some of the public health procedures before  you built the cities. So foresight can be good. 

Right. I guess the sort of bigger question here  is, it makes sense in the kind of environment our  

ancestors faced why they’d have intuitions against  progress, because if you already have technologies  

that are well fit to the environment, any  change is likely to be deleterious. And whether  

we find ourselves in a world that’s different  enough where we can just, very intentionally  

disregard our worries about changes in culture or  technology, or whether it’s similar enough that we  

should actually… I mean, there’s two ways you can  interpret The Secret, right? One is, “Oh, look,  

this is why we should care more about culture”  and so forth. Another is just, “Oh, wow, they’re  

living in such a different world. I understand  why my intuition against progress- where it comes  

from, and now I can totally disregard it.” Right. So I would say that the lesson from  

Secret is not to disregard it, but just to  be cautious because we could be dismantling  

things that are really important for the  structure of society. And you shouldn’t  

just dismiss valuable cultural practices as the  relics of a medieval age, or the relics of a  

pre-enlightened age or something like that. So ritual is a good example because rituals  

seem to have real psychological benefits in  binding people in community and helping develop  

self-regulation. But it’s easy to be an atheist  like me and say, “Ah, rituals are stupid. We gotta  

stop doing them.” Turns out they’re doing a bunch  of stuff and if you don’t wanna lose that stuff,  

you gotta figure out another solution. A question that arises in my head is,  

look, we began this conversation talking about  all these societies in the past, even when they  

figured out something successful, they did not  themselves know why it was successful and so they  

just had to ascribe it to custom. They just say  like, “Oh, this is the way we always made bows”  

or hunted or something. Why think that we have an  answer for why the Industrial Revolution happened?  

Why think it’s any different than the Inuit trying  to explain why they make their bows a certain way? 

Oh, you mean, how or what  we’re up to is different? 

Yeah. Well, I mean, we’re trying to  

apply science, which has been very successful  in all these different areas. So, you know,  

evidence, and we all put out our arguments and we  go through the process and some evidence is better  

than others and that’s just our epistemology. And I’m interested in the cultural evolution  

of epistemologies. So societies have varied  over time in what counts as good evidence  

and what counts as a good argument. So one  of the psychological changes that we see  

emerging in Europe compared to other places  is how important is what the ancients say. 

So in lots of societies, if someone says  something and you can say Aristotle or  

Confucius didn’t believe that, then it’s like, “Oh  gosh, I guess I’m sunk.” Whereas at some point,  

Europeans decided, “Actually who cares what  Aristotle thought. We know a lot more than  

he did.” So this is a big epistemological  change. And then the emphasis on empirical  

evidence in science is not something you find  in earlier traditions within Europe, and it’s  

quite variable across different societies. So you should think about the very standards  

by which we count a good argument and good  evidence is itself culturally evolved. 

Yeah. And how much do you worry about the lack  of cultural variation in science in particular,  

where we have like one big institution, the  NIH, which funds most science. And there’s  

similar processes of consensus and peer review and  whatever, that educated whether you get a grant,  

whether work is accepted, whether it’s considered  worthy scholarship and so forth. Should we think  

that maybe we should have more variation in those  methods because maybe the scientific method isn’t  

complete, there’s epistemic tools could be  improved or it could be different and so forth? 

Yeah, absolutely. So polycentricity. There should  be lots of difference, competition amongst these  

groups, no single funding sources… lots of  different priorities. One thing I point out in  

this book I’m working on is that scientific  papers tend to be more impactful when the  

authors are from more diverse societies. But interestingly, people are biased  

to work with people from their own  society. So what we actually do is  

different than what would maximize innovation. I’m sure you’re gonna address this in the book,  

but I’m already curious now. It seems like the  internet hasn’t been as big a deal as somebody  

who’s thinking from this zoomed out collective  brain picture might have anticipated in the  

’90s. Maybe it’s boosted the rate of innovation  somewhat, although it’s not clear just looking at  

growth rates or productivity growth rates  or something that it has. Why didn’t the  

internet cause the equivalent of another round  of urbanization in our ability to do research  

or economic productivity or so forth? Yeah. And that’s something I’m really interested  

in figuring out. And one thing that I’m still-  I don’t have a full answer to this question,  

but one thing that is clear is that there’s  something special about face-to-face interactions. 

So something like whether even in the 21st  century, whether two cities have direct long  

haul flights between them or direct flights  between them increases the flow of ideas  

between those places. But we know these places  are connected by the internet, whereas that  

effect doesn’t seem as good. And part of this  is that people have to build trust, probably,  

before we start sharing ideas. And the more  different someone is from us, the more you need  

that trust. So some of the research suggests that  this effect is even bigger when places are more  

culturally different. And this is interesting,  because they’re probably more valuable to each  

other when they’re more culturally different. But  that’s then the face time and being in the same  

place is even more valuable. So that’s the kind  of direction I’m going. And also, I still think  

that there should be exactly what you said,  which is that the internet should have caused  

more innovation. I’m just not satisfied that we  fully determined the best way to figure that out. 

You had a very interesting chapter in Weird  where you talk about… a lot of innovation  

through history is more serendipitous than we  imagine, and it’s the result of the collective  

brain being big enough to discover these things.  But if you look at the sort of track record in  

science and research, there do seem to be certain  individuals who make independent discoveries  

across many different fields that are each quite  important, your Newtons and Einsteins and John von  

Neumanns and so forth. And if it was just the  result of the right person is in the right lab  

who sees the right observation at the right time,  this repeated excellence by certain people seems  

less explainable. So I’m curious if the collective  brains theory can explain what’s going on there. 

Right. Sometimes I’m interpreted as saying  that there are no genetic differences among  

individuals, [I] definitely think there are  genetic differences among individuals that  

affect their likelihood. But I think when we  take the individual, we often import into our  

thinking about them the person’s life history. So  for example Einstein, when he was a patent clerk,  

so he wasn’t succeeding as an academic, he and  his friends got together in something called  

the Olympia Academy, they called themselves,  which was just a group of, like, five people,  

who would get together and they would read the  interesting books at the time. And if you look at  

the books that they read, and historians have done  a lot of work on this, all the major ideas that go  

into special relativity were read and processed by  the Olympia Academy before they do it. So the idea  

that people think that time is relative; well,  people were kinda talking about that, and they  

were talking about multiple dimensions. And like  Henri Poincaré, at the same time, comes up with  

the same equation as Einstein, but he doesn’t give  it as radical an interpretation as Einstein does.  

He was thinking of the equations as kind of fudge  factors, trying to make the math work. So there  

was an almost simultaneous invention of special  relativity, and the ideas were all circulating.  

And Einstein happened to be in a place which  allowed him to put all those things together. 

As a patent clerk, he was actually processing  devices to synchronize trains in different parts  

of Europe. And so he had to think about the  amount of time a signal coming from different  

parts of the world. And there were a bunch of  these. And we know the patents that went through,  

and we know that he had to look at these, there  were only two guys. So he happened to be in a  

particular place and time in terms of the ideas  circulating and what he was doing on a day-to-day  

basis that really did give him an edge. Now,  you could then say, “Well, what about general  

relativity?” Well, there were a couple other guys  who were probably gonna get general relativity.  

And Einstein himself was worried that he was  gonna get scooped, because once you get the one,  

it’s just a matter of- I mean, it’s not “just” a  matter of figuring out, the math is really hard.  

But there were a number of other people  who probably would’ve gotten the math. 

Yes. I guess then there’s also still the  question of why, in 1905, Einstein himself  

discovers not only special relativity, but  Brownian motion and the two other things,  

right? But I guess you would just say it’s  the result of he had the right reading group? 

I mean, obviously he’s a special individual.  Although he did spend the whole second half of  

his career trying to show why quantum mechanics  was wrong. So his Brownian motion paper is  

foundational in quantum mechanics. But then he  decides God doesn’t play dice with the universe,  

and he doesn’t like the stochastic nature of  quantum mechanics, which has more or less proved  

true, and it’s part of our modern technology and  stuff like that. And he spends the whole second  

half of his career fighting it. So that’s a  case where his intuitions didn’t serve him  

well. And he turned out to be wrong for decades. And, sorry, I think that’s also a case in which  

your point about having the right ideas  available to you is important, because,  

obviously I don’t know much about this, but from  what I understand, the people who believe in the  

many worlds interpretation believe that Einstein  was on the right track, and he just didn’t  

have this idea of the ever-ending multiverse  available to him, which would’ve explained his… 

That there’s this branching  thing, and just a matter of  

figuring out which branch you’re on. Right. And so there’s no probability. 

Yeah, yeah. Yeah. Okay, so we were just discussing,  

look, Einstein was reading the right things  and making sure he had enough context to come  

up with these big discoveries. You’re somebody  who has connected ideas across many different  

fields. How would you describe the input function  for the way in which you come up with new ideas,  

and how much has it been informed by… I mean,  you were like, you started off as an aerospace  

engineer, and you’ve done anthropology across  so many different societies, on fieldwork and  

so forth. How do you basically think about  the Joseph Henrich production function? 

Yeah, I mean I try to implement the ideas of the  collective brain. So in my lab at Harvard, I have  

post-docs and graduate students with backgrounds  in evolutionary biology, psychology, anthropology,  

economics. I mean, some of my favorite  collaborators are economists. And we’re just  

bringing ideas from very different places, but  across the social sciences and even the biological  

sciences, we’re often trying to explain the same  kind of phenomena, economic decision-making,  

cooperation, things like ethnicity, why does it  exist, stuff like that. So the way we silo science  

is a big problem, and especially in the social  science, I think it’s even a bigger problem. 

So I’ve been a professor of anthropology,  psychology, economics, and evolutionary biology,  

and just even the standards of evidence and  what constitutes good research and how you  

tackle a problem really varies. So the academic  disciplines are like different cultural worlds,  

different tribes, and just pulling the best  from them has kind of been my approach. 

And among the more polymathic scholars, do you  guys have, I don’t know… do you and David and  

a couple of the other ones in this category, is  there some shared group you guys have, or, is it… 

No, it’s a series of different networks. So  David Reich and I, for example, are always  

talking about how we wanna read culture from the  genes. So if humans have had a long history of  

gene-culture co-evolution and we wanna understand  what Paleolithic populations were doing, we  

might be able to see imprints of it in the genes. So in, in my own work, for example, we’ve measured  

cousin marriage across different populations and  shown it correlates with runs of homozygosity  

in the genes. Something like polygeny can be  revealed by looking at the ratio, the variance in  

Y chromosomes to X chromosomes. Fire: we probably  have some special genetic adaptations for dealing  

with the toxins in fire. So if you wanna know  when humans got fire, and you see the gene that  

gives us immunity to toxins in fire, then we  can infer that fire is older than that, right? 

Interesting. And, so obviously the input from  different scientific fields in your work is  

obvious. But from a more philosophical  perspective, people compare your work  

in the same tradition as Burke or Chesterton  or something. Do you personally find that  

your philosophy has been informed by  reading them, or is it just independently  

converging on some of the same themes? Yeah, I mean, any convergence is completely  

independent, because I haven’t really spent much  time at all. I mean, I’ve gone back. After people  

said, “Your work is kinda like Hayek,” I went  back and read some Hayek. I did read The Fatal  

Conceit in graduate school, but aside from that,  I didn’t read very much. And then same thing with  

Adam Smith. I mean, I’ve tackled The Wealth of  Nations, and I eventually read The Theory of  

Moral Sentiments. But it was pretty far into  the process. I had picked up on themes they  

developed before I knew they had developed them. I want to suggest that maybe you should add a  

couple of AI people in your rotation. Ok. 

Just because I think this perspective is  incredibly informative, what I think is  

maybe the most important question of our period,  of what does this transition towards a society  

of AIs look like? And, sorry to bring it back  to AI, but one other consideration here that  

your work has really informed me on, is: I was  sort of maybe hyping up how big a deal I expect  

societies of AIs to be, but something your work  informs is the idea of a single superintelligence  

is not the place in which to expect these  big impacts of John von Neumann times 10. 

But then again, there is a key advantage in the  scope of social learning they can do as a society,  

but, like, the idea of a single superintelligence  being super powerful is maybe less likely as a  

result of your work. And I, anyways, I don’t know  what you think about that interpretation, but… 

Yeah, I think that’s right because the  assumption, the sort of model that people  

seem to carry around in their heads is that  humans have done all these things because of  

our individual brain power when really, it’s been  the power of cultural evolution and a network of  

minds gnawing away at problems over long periods  of time and gradually accumulating not just the  

obvious tools, but also the cognitive  tools for, for addressing these things. 

I buy the idea that individual IQ isn’t the most  relevant factor to understanding why discoveries  

are made. It’s not about having the right genius  or something. But what’s the reason for thinking  

that the average IQ of a society isn’t super  important in how much progress a society’s  

gonna make, not just the population size? Right. So the first thing that we need to  

do though is to zoom back out and think about what  we mean by IQ. And so Michael Muthu Krishnan and  

I have made the case that IQ is just a set  of cognitive abilities that leads to success  

in 20th century contemporary institutions that  have come to dominate the world. So it’s a set,  

it’s a culturally evolved system, and we talked  before about the Flynn effect which illustrates  

that, and the fact that IQ is associated  with all these positive outcomes now, but  

certainly wouldn’t be in pastoralist societies. So that would mean that if you raise the average  

IQ of your society, it might lead them to have  more abstract thoughts and do better science.  

And so that’s, that certainly fits. But the  interesting thing for me is that the world of the  

future is gonna be quite different than the one  now. So the set of cognitive abilities which is  

gonna be favored in the new AI world is not gonna  be the same set that was favored in 1900 when they  

designed the IQ test. So for example, in the IQ  test they ask you to remember digits backwards.  

I’m not sure how useful remembering lists of  digits backwards [is]. It was in a previous world,  

where we had to write everything down and  remember a lot of stuff. But in some sense,  

we’re interfacing with our technology and we’ve  got to figure out what are the set of cognitive  

abilities which is gonna make people best able  to solve problems? And like we talked about,  

it’s even the case that the most creative  people aren’t the highest IQ people. 

I guess one of the things I’m trying  to say is that the minds that might  

lead us into the new world might not be the  ones that have the highest IQ because once  

you’re sort of augmented with AI and all  these kind of technological aids we have,  

the specialized thing that leads someone to  do something creative- probably not gonna be  

the same abilities that did it in 1910. Yeah. I guess my real opinion is that,  

if in the world where that’s true- and I think  that’s probably gonna be true- then it really  

doesn’t matter at all because AIs will be  doing everything. But in the world where,  

let’s say AI plateaus or something, then I would  expect roughly the same kinds of skills that have  

mattered for the last couple of centuries to keep  mattering for a modern, technological society,  

which is analytical thinking and the ability to  understand science and technology and so forth. 

But maybe the AI world you’re imagining  is different than the one I’m imagining,  

but I still think that people are gonna  figure out what problems we need to solve,  

unless we’re just gonna tell the AIs to figure  out what the problems are and then solve them. 

They might not be that good at  that. I don’t know. We’ll see. 

You shared with me this draft  of work about collective brains,  

where you show that ants have developed many of  the technologies that humans have, like farming,  

and livestock, and division of labor and so  forth. And so maybe there’s some amount of  

blind selection, and it doesn’t matter if it’s  natural selection, if it’s cultural selection,  

it’s the size of the group which can go through  this learning process that matters, and how many  

people are available to experience the learnings  or get the selection process acted upon them. 

But I guess the big difference in where maybe  individual IQ comes back into the picture here is,  

I don’t expect there to be any amount of natural  selection which will allow ants to land on the  

moon or make a computer chip or do, you know,  make a nuclear fusion plant or something. And  

so is the kind of broad generalization we  see with humans, and we might see to an  

even greater extent with AI, a product of you’re  really bottlenecked by the IQ of the individual?  

There’s no amount of collective learning that  can get you to the moon if you’re an ant? 

Right. Well, one thing to keep in  mind is that most human societies  

over most of human evolutionary history  didn’t get to the moon. So this is one  

particular group of humans. And  one of the things we talked about  

earlier in this conversation was the  cultural evolution of epistemology. 

So it was the improvement in our what constitutes  evidence, what constitutes a good argument that  

allows us to get to science and accumulate this  kind of knowledge to do these kinds of things.  

So I see that as part of a continuous trajectory.  But it’s just that we have new cognitive tools. 

Right. I guess, but… so there’s a really  interesting point that even the epistemic  

tools which let us get to the moon are themselves  a product of cultural evolution. But again,  

that seems bottlenecked by the fact that if  you’re just dealing with chimpanzees or something,  

there’s no amount of cultural evolution that  will result in the scientific method, that- 

Right. It needs some gene culture co-evolution. Right, right. So you mentioned in the Secret  

the tool use started something like  two million years ago, and fire,  

we started domesticating fire closer to  800,000 to a million years ago. Intuitively,  

it doesn’t seem like tool use is that that much  simpler than fire, but is that a misunderstanding?  

Why was fire so much [more] recent? Well, so what we know from other  

species is that lots of animals use tools,  and particularly chimpanzees use tools. So  

we can assume tools in the common ancestor.  Now, what we see in the paleoanthropological  

record is the increasing use of stone tools. And these are pretty simple stone tools. You  

can see a cutting edge there, but not very much  fancier than that. And then fire is… a lot of  

animals are afraid of fire, and they have to run  from wildfires and stuff like that. So whatever  

your story about humans is, you have to overcome  the fear of the fire in order to tame it. Probably  

humans first found wild fire and somebody  approached the fire instead of running away,  

which is the usual thing to do. And then got some  of it and then put it to use, I guess. So I think  

it’s the innate fear that animals have of fire,  which is we don’t hang around when things go off. 

Interesting. Okay, final question,  what’s the next thing you’re working on? 

Well, the big thing is this book on  collective brains, so I’m gradually  

working through that. One of the ideas that I  really am excited about in this book is that  

we evolved to think in collectives. There’s  this assumption that psychologists have had  

to understand human decision-making and how  our minds solve problems is that we should  

put people by themselves in experiments and  see how they do. But real human societies,  

hunter-gatherers, they actually work in groups.  And when we want to solve a problem, the first  

thing we want to do is check with our friends  or ask other expert members of the community. 

So it’s the idea that we think collectively and  solve problems collectively in a kind of naturally  

distributed brain. And Hugo Mercier and Dan  Sperber, for example, have argued and shown with  

various lines of evidence that many of the sort of  irrational biases we have, mitigate or disappear  

entirely when we solve problems as groups. So  it’s almost like we evolved to have that positive  

interaction and correct each other’s errors. Right. That’s quite interesting that you care  

more about the portfolio of intelligences  than any one person being calibrated. 

Right. Excellent. This is super fun.  

Thanks so much for coming on the podcast. All right. Great to be with you.