# 2025-06-28 - Robin Hanson - David Deutsch, Robin Hanson, and Daniel Martin on Culture [YouTube](https://www.youtube.com/watch?v=7zUQBs-ju6w) Duration: 01:21:25 ## Transcript ### Robin Hanson 00:00:00 - 00:00:17 And now we're recording. Hello, I'm Robin Hanson. This is... Hello, Daniel Martin. Hey, with my collaborators as well, David. David Deutsch. All right. We are here to talk about culture. What about culture should we talk about, Daniel? ### Daniel Martin 00:00:17 - 00:01:34 Well, David, you've written about cultural evolution many times over the years, in The Beginning of Infinity, and then this 1994 article, The Evolution of Culture, that was recently put online. But I thought I'd bring you to discuss cultural adaptiveness, drift and monocultures, since Robin's been writing about that. So Robin, you wrote this year. We need sufficient variety and selection pressures regarding our behaviors relative to the rates of environmental change and internal cultural drift. Without good parameters, culture should drift into maladaptation. Today our parameters still look good, really the types of behaviors that are easy to vary individually, individual level. We have record rates of innovation and tech and business practices, but they look more worrisome, really the cultural norms and game theory equilibrium parameters that are hard to change individually. So you're saying in the last hundred years the adaptiveness of cultures, meaning their survival, has dropped due to, I'll let you state why, I would say, from what I've read, a decoupling of survival of memes from the actual survival of their holders in wealthy countries. And the pressure, and secondly, the pressure towards global monoculture so that features become, more features become hard to vary. And these are exacerbated by status markers, and so on. So is that a good description? ### Robin Hanson 00:01:34 - 00:03:46 Roughly, but I fear you're going over so many points so quickly, and David may well have reservations about many of them, so that we should go methodically one step by step. And so I think the first point on which he might well object, I don't know, is this idea of different levels of culture in terms of the ease of evolution of them. So I take the analog to say DNA evolution with species. There's evolution within a species and evolution of the features that define a species, and that evolution is harder because it's hard to vary a species within the species, the defining features. And so that in biological evolution, we have different niches in the world, and when they have fragmented niches, then they have many small species. And when they are large integrated niches, like an ocean, they have a few big species, and that the fragmented species have more evolution of the features that define a species because there are more of them. But the integrated environments have more evolution within species because they just have larger species, which can then share more innovations more quickly. And that's a distinction I would hope you might embrace with respect to DNA evolution. And then I would suggest there's a similar set of levels with cultural evolution when, if say business practices and tech is something I personally can vary without being punished greatly about it, then we have evolution of those much more easily because it's, you know, we can just try lots of different things. But when a culture, like a community or a language or whatever, shares a bunch of things that are hard to vary within, like status markers, norms, game theory, equilibrium parameters, then that's a higher level of culture. And my focus is on the problem of that higher level. I completely grant that stuff at the lower level that we can individually vary relatively easy. We're doing great on that. But that's my first pause is, what do you think, David, about this distinction between the things that are easy to vary within a culture and the things that are sort of tied to a culture that we need variation of cultures to have evolution of those? ### David Deutsch 00:03:46 - 00:05:50 Oh, okay. I was going to say I can agree with everything, but the last sentence, evolution of cultures, that suggests the equivalent of a species selection. Yes, I think it's just generally speaking a mistake because it assumes that. Cultures. Breed true. And so if they're losing the evolutionary battle, they will go extinct. But the variation in cultures or the. The breeding true of cultures is entirely dependent on what individuals do. So you can have. A culture that. That where individuals decide. Decide. That having a lot of children is bad. And then that culture will go into some kind of decline over generations. But the minority of them who don't adopt that will become dominant and then the culture won't go into decline after all. So and the same holds in biology with species selection. Species selection can only sort of apparently have a hold over different variants of a species to the extent that the. Individuals aren't evolving, but the individuals are evolving. So I don't believe in species selection, but apart from that, levels of levels of. Selection and levels of evolution and levels of culture that all sounds very plausible, though I haven't thought much about it. ### Robin Hanson 00:05:52 - 00:07:38 So I think I was right that this is a potential point of disagreement that we should make sure we get clear before we get on to other things. So if we just stick with the DNA species thing. The usual story is that, say, mating structures are harder to change within a species, because if you change your mating structures, you can't mate with other elements of your species and then you know you can't have descendants. So there's a standard story that mating structures are more stable in a species. And that often what defines a different species is having different mating structures, which then means it can't mate with other species. And so if you think of evolution of the structures that are important to mating, you might see that those are harder for an individual in a species to just deviate from because they're coordinating with other members of the species to share enough of the mating structure to be able to mate. But of course, when speciation happens, it does happen through individuals in a species changing their mating structure and then having a set of nearby related individuals changing enough together so that a speciation can happen, right? So speciation still happens via individuals. And biology apparently has produced higher mutation rates in mating structure things. That is, DNA replication just apparently has designed higher error rates, higher rates of change with respect to mating things exactly to encourage speciation because otherwise you wouldn't have a set of species. ### David Deutsch 00:07:38 - 00:08:24 Okay, that's new to me, that idea. But if we go down to a lower level, I would expect not mating, but just the biochemistry of reproduction. I would expect things like ribosomes and the DNA code to be the slowest varying things, not because there's any mechanism making it harder to vary them, but simply because, as you said, if they vary more than a tiny amount, then it will affect the reproduction of the species. ### Robin Hanson 00:08:24 - 00:09:13 Right. So the claim isn't that it's not possible to mechanically vary these things. The claim is that it's not possible to get rewarded for variations if they are not coordinated with other variations. So you know of game theory, presumably. So I was a social scientist trained in game theory. And the key concept in game theory is that there's an equilibrium and that there are multiple possible equilibria, but in a particular equilibria, no one player has any particular incentive to deviate from the equilibrium. So the key concept is even if we'd all be better off in some other equilibrium to the extent we all see the equilibrium, then we need to behave according to that equilibrium. ### David Deutsch 00:09:13 - 00:09:27 But it's the genes that have to be better off in order for some evolutionary change to happen. So not the host. The host might be harmed by a particular variant. ### Robin Hanson 00:09:27 - 00:11:01 I was going to say I think the analog of, say, the ribosome is the school. So schools, I think, are the slowest changing institutions in society because they have the most direct effect on causing their memes to be replicated to the next generation. So one key element of cultural evolution is status because if people copied cultural things at random, it wouldn't work. We have to preferentially copy things that are better than average for cultural evolution to work. And one of the key heuristics we use about what to copy is social status. So that makes social status markers a key element of culture. And so that if a community has bad social status markers, that can hurt the community, even if other things are good. That's, in my sense, an example of an equilibrium coordination feature that we coordinate around who we respect. And one of those in our society is that we respect people with a lot of education. And that if we respect that too much, we could all suffer, even if each one of us individually doesn't have still if we deviate from getting as much education as we can, we will personally get less respect and have fewer descendants, etc. So that's an example of a culture feature as opposed to an individual feature, a shared thing, a thing a community shares, which is how much respect for education. ### David Deutsch 00:11:01 - 00:11:09 Wait, having more offspring is not the most important thing for a meme. ### Robin Hanson 00:11:09 - 00:11:17 Of course, of course. It's just an example of a consequence. It's not all consequences. ### David Deutsch 00:11:17 - 00:12:35 In Britain, it used to be just a few decades ago that if you had an upper class accent, you had an advantage in life and people would try to copy you where they wouldn't try to copy someone with a lower class accent. And this has pretty much changed by now. They're still like, I don't think the meme war between that and its opposite has reached a conclusion yet. But certainly, you know, with prime ministers, if someone came from Eton, nowadays, they, someone was educated in Eton. Nowadays, that's a disadvantage, whereas only 20, 30 years ago, it was actually an advantage. Right. So that's a status marker of accent. And there's been a change over time in the status that certain kinds of accents have. And that just shows that, but in a particular society, say, at a particular time when, say, 50 years ago, Eton accents were highly respected, then that individuals in that society had an incentive to acquire the respected accent. And if they said, the hell with society, I like my accent, they would suffer even if. ### Robin Hanson 00:12:35 - 00:14:18 Right. That's an aspect of education, which I'm saying is quite easy to vary. It like it's subject to fads. It has varied, but it's hard to vary locally. That would be the key claim. That is, if you and your friends decided you don't care about Eton accents 50 years ago, you and your friends would suffer. Society collectively can change these things, and it does. And that's an important part of the story to think about. But the point is, as an equilibrium, individuals don't have much of an incentive to deviate. Right. They get punished for deviating. Okay, I can. Right. Now I get what you mean. I think I can say that I think that's possible. Yes. Okay, well, so my claim is about this level of things that it's hard. You get punished for varying individually, not that you can't vary individually that you get. So, for example, as I said, we respect education a lot in our society, and that encourages individuals to get as much education as feasible at the most prestigious places they can. Even if that takes a very long part of their lives. And that may hurt us all collectively, but individually, that's still your incentive is to get as much education as you can. And that's a shared feature of our society. As you know, centuries ago, education didn't nearly have as high a status, and so people pursued other sorts of things to gain status. And that had consequences. They pursued military prowess and money and family connections, whereas we pursue education. And so our societies are different as a result of these different features, but the features are hard for individuals to vary much. These are shared features of a community. That's the key idea. ### David Deutsch 00:14:18 - 00:14:31 But didn't the shared features of the community begin as shared features of a tiny sub community, which then other people could be then? ### Robin Hanson 00:14:31 - 00:15:17 So shared cultural features like the value we put on education or say how anti racist we are, are features that change over time, and they change over time in part due to cultural activists who are initially small groups of people who then push for cultural change that over time becomes large and shared. So, but, you know, and there are, it's interesting to ask how do cultural activists get away with defying current existing cultural values and norms in order to change them and gain status for changing them. That's a key feature of the modern era. Well, it wasn't true until a few centuries ago. You know, the modern world is full of cultural activism in ways that the ancient world just didn't have that much of. ### David Deutsch 00:15:17 - 00:15:32 But, yes, I doubt that in most cases, the failed ones fail because they're punished, except in the sense that people don't listen to them or stop listening to them or something. ### Robin Hanson 00:15:32 - 00:15:59 But that, but the being punished as you know being excluded from society, I think is the rare case. I didn't mean punish in any other than the game theoretic sense of not the maximal payoff. So the idea of a game theory equilibrium is that your strategy is determined by the equilibrium and such that if you deviate you won't get as high a payoff. That's the sense in which I would mean you're punished for not following the equilibrium. ### David Deutsch 00:15:59 - 00:16:01 Right. Okay, so I'm with you so far. ### Robin Hanson 00:16:02 - 00:18:40 Okay, so if we can talk about this. So just like with again with species, if you're willing to accept the analog of species, again, large integrated environments habitats like the ocean have a small number of large species. And then they're going to have rapid evolution within the species that is a larger species has more places for innovations to spark more places from the spread. If you have many small species, then they will have slower innovation within each species because they're more limited in terms of the number of places innovations can start and spread to. So that says not only is the evolution of species something that matters, it apparently matters even more than the evolution within species because otherwise, you know, rivers, rainforests, coral reefs, those sorts of very fragmented environments are actually today more often places where the species we have today came from in terms of their origins. So that says it's actually pretty important to have this evolution of species and analogously, you might think it's important to have evolution of cultures. So one example is corporate cultures. Corporations have cultures inside and then different corporations have different cultures. And by analogy, we have the issue of a large corporation can have more innovation within the corporation because there's more different places for innovations to start and spread to the rest of the corporation. But corporations often have shared culture features across the whole corporation that are hard to vary within the company, like, you know, what standards they have for who should be promoted or, you know, what looks like a good worker. And so some industries have a few big firms and other industries have many small firms and we can ask which industries are more innovative. And it looks like the fragmented industries are actually more innovative than the few big firms, which also suggests analogously that for corporate cultures, innovation of the cultures matters at least as much as the innovation within a corporate culture. And so that's more about culture. ### David Deutsch 00:18:41 - 00:19:14 It seems like civilization is more analogous to the company cultures to me than it does to genes because you have minds involved. So the business, right. The market sets the constraints of what businesses can exist, although they're not when you create knowledge that also sets some unknowable constraints. But it's built by minds as well. And so any one problem they can think they're way out of because they're not a blind watchmaker like an evolution. So why is our ship like we need to change some parameters or we're going to hit an iceberg rather than we need to set up our ability to make new knowledge to get over these problems. ### Daniel Martin 00:19:14 - 00:19:16 What do you say? ### Robin Hanson 00:19:16 - 00:20:13 So, the thing I've been belaboring here is just this idea that cultures have shared features. And that if you just have a few huge cultures, they might not evolve those shared features very well. That's the key thing I'm trying to point to the analogy with corporate cultures and with species. There's a difference between having many small things versus a few big ones and that the few big ones will do great internally. Like the species do a great internally and the corporations do better internally when they have they have a bigger scope. But evolution of the things that they share within that scope, that evolution is worse when you have a few of them and better when you have many of those units that that's the key. I'm belaboring it because if you accept that, then I think you'll accept everything else. I'm going to say really easily. But this is the one thing I fear that it's that you might not accept, which is why I'm trying to go through it very slowly. ### Daniel Martin 00:20:13 - 00:20:21 I'm suggesting the set might be small. But what do you think David? As long as people can complain, right? As long as we have some tolerance for people complaining. ### David Deutsch 00:20:21 - 00:21:02 Yeah, there is this thing that memes can change not along heritable lines. Like they can change sideways in a culture. So that's not DNA heritable, but it's cultural heritable. Yes, but it makes a difference. So even gene variant that spreads like wildfire will take many generations before it covers the entire culture. Whereas an idea can cover the entire culture in much less than one generation. ### Robin Hanson 00:21:02 - 00:21:36 Right. Like, you know, gay marriage, for example, was so there's a literature of cultural evolution of, you know, academics who study it. And I've accepted their stance. I don't know if maybe you disagree with it, but the stance is roughly the key difference between DNA evolution and cultural evolution is just the inherited. We have new inheritance paths. That is, you have a wider set of potential parents of culture, but it's still roughly the same process. It's just through a wider set of parents from which you can inherit culture. ### David Deutsch 00:21:36 - 00:23:43 So I don't think that's the biggest difference, although I don't know whether the biggest difference is relevant to what you're going to say. But let me just say that a bigger difference is that a variant meme has to run the gauntlet of two different selection processes alternately. One is the selection of which meme a recipient of memes will adopt. And there are always very many more memes, unlike like in the in the biological case, there are always very many more memes around than any one person can adopt. And then secondly, once it's adopted, there is the question of it's going to be enacted, whether it's going to be enacted. And both things are subject to the choice of the recipient. And although you can have memes that cause the recipient not to make many choices by, you know, anti rational memes and so on, in as far as the mechanism of inheritance goes, there are these two things people can take on board a meme, but not pass it on. And it's a bit of a miracle that any memes are can can pass this test. But as I said, there are, it's quite common for memes to spread like wildfire in much less than a generation. And because they have this property of both being taken on board and enacted. And I mean, another minor thing connected with that is that a meme has to be enacted to be passed on, whereas a gene doesn't have to, you know, I can have a gene for repairing my bones when they're broken, which are never enacted in my expressed as the biologists say in my lifetime. And yet I can still pass them on. Right. Whereas that cannot happen with memes. ### Robin Hanson 00:23:43 - 00:23:54 It's true. I accept those and I don't that much care which is the most important difference. I will not make the claims about that. I'm just I'm happy to accept those are true facts about culture and they are different from DNA evolution. ### Daniel Martin 00:23:54 - 00:24:05 And so we, we agree roughly on these features that are different about cultural evolution. Why do you say enacted David, when you like vote up or down on some internet site, you're causing it to be more visible. ### David Deutsch 00:24:06 - 00:25:03 Yeah, well, when you say a meme, you know, when you when you pronounce a word in a new way, whatever that's an action. Even if even if it's just saying things, it doesn't have the property that it's copied into the recipient. It's up to the recipient, and the recipient may often does mishear what you have said or what you have written. And because creativity is required to receive a meme to enact a meme requires less creativity, but to receive a meme is a purely creative process, and it's fallible. And so one property of successful memes is that they're easy to understand, at least within a certain culture, they're easy to understand. But anyway, I don't know if this has anything to do with your point. ### Robin Hanson 00:25:03 - 00:27:37 So I've been focused on elements of culture that are things that most people will have to have a variation of. You know, and so I'm less focused on these things that are very sort of rarely expressed unusual in a wide space of possibility things. So, for example, if we say what are the main causes of fertility decline in the world today, I would find a half dozen cultural trends. And that each of those cultural trends seems to be the sort of thing that everybody needs to have some version of it, and that are hard. There's costs for deviating from the shared consensus on these items in the larger culture. So one of them is the high value in education. That's discouraging fertility because young women are eager to get as much education as they can to gain as much respect as possible. And that's a limitation of fertility. Another is they say switch from cornerstone to capstone marriage norms. So when you and I were young, the norm would be that you get married young and not fully formed. You don't know where you're going to be in the world and you just find it out together. And now the norm is that you wait until you figure out who you are and find a secure place, then find somebody else who matches that place in the world and then think about having kids. And that also discourages kids, but it's a shared norm in that when the whole world has the capstone norm and you have a different norm, it's going to be hard for you to find a marriage partner harder. You will suffer because others are looking to follow the capstone norm and you're trying to do the cornerstone norm that will get you to disadvantage. Just like you have a disadvantage if you say hello school, I don't care about school. I'm just going to do my thing. Other people won't respect you as much. You will suffer a cost. A third is a rise in say the norm of how much attention parents should be paying to their children. How many hours a day or week should they be constantly attending to their children, being there supporting them, doing things with them? That norm has changed. I presume you remember as a child, parents didn't pay remotely as much attention to you and me as parents are paying to their kids today. But now if a parent today tries to deviate and goes back to the old norm of parenting, they'll be accused of neglecting their children, like letting their child walk to school, for example, is apparently a problem now because you should have walked with them. So I could go on, but these are examples, I would say, of cultural norms that are at the level of a culture, not that you can't deviate from them and defy them, but that you will suffer a cost from doing so. ### David Deutsch 00:27:37 - 00:27:46 Yeah, suffering like I said and you agreed. It can be a misleading term here because... ### Robin Hanson 00:27:46 - 00:27:57 You will not get maximum payoff in your equilibrium, right? The meme won't get maximum payoff. It won't be transmitted to as many people. ### David Deutsch 00:27:57 - 00:28:44 Yes. So, you know, and I think these trends like having fewer children or paying more or less attention to them or women wanting careers are not primarily caused by memes that tell you to do that. Women want to have careers because of completely different memes that say that they ought to be in charge of their own lives and that kind of thing. And whether that causes... For some memes, it's actually part of the attractiveness of the meme that it is stigmatized by the older generation because you can be cool and hip that way. ### Robin Hanson 00:28:44 - 00:28:46 Right, but you need your generation to like it. ### David Deutsch 00:28:46 - 00:28:48 Well... ### Robin Hanson 00:28:48 - 00:28:50 Somebody needs to like it. ### David Deutsch 00:28:50 - 00:28:56 Well, so if you want to have children with a partner, you need one partner. ### Robin Hanson 00:28:56 - 00:28:57 Right. ### David Deutsch 00:28:57 - 00:29:19 And if there is a small group of people that are enacting one thing, and for example, this thing about women controlling their own lives, there have always been women who have controlled their own lives for centuries. Right. It's just that not many other women emulated them. So... ### Robin Hanson 00:29:19 - 00:29:27 Right. Because of cultural context. It wasn't a random thing. We can see why in that culture people weren't rewarded for doing that. ### David Deutsch 00:29:27 - 00:30:38 It may not have been... No, again, you're saying they were rewarded or not, but I think it's better to say people weren't persuaded by that. They weren't persuaded that that's a good way to be. You know, they looked at Mary Wollstonecraft and too few women thought to themselves, I want to be like that because I'll have more control of my life. No, because people will hate me. I mean, people hating Mary was part of the package. That was part of being a revolutionary, a trendsetter, and so on. Some people did try to follow it, but too few. Until, I don't know, until maybe the pill was invented. Again, not a cultural thing, just a mechanical thing in the 1960s. Maybe that's what tipped the balance. So it wasn't a matter of what culture thinks of those people. It was a matter of ideas that they had about what they should do in their lives. ### Robin Hanson 00:30:38 - 00:31:07 So the idea of norms and the idea of status markers is exactly the idea of things where it's not just about you and your head. It's about what other people around you think of what you do. That's the whole point of norms is status markers. These are social coordination mechanisms that create correlations in what people do and how they treat each other. That's the mechanism by which it's harder to individually deviate, right? ### David Deutsch 00:31:07 - 00:31:23 That's a mechanism, but another mechanism is whether some idea like a fad spreads through the population like wildfire, regardless of what the culture, like a couple of decades before, thought of that. ### Robin Hanson 00:31:23 - 00:31:36 Again, I would call that a change in the norms and status markers. That is, if at one point everybody shames you for doing something and later they think that's cool, that's because initially the norms were against it and later the norms are in support of it. ### David Deutsch 00:31:36 - 00:32:07 That's a change in the norms, which definitely happens. But I'm still trying to distinguish that from the fact that at a time norms exist and they have a force. You're saying that people change their individual minds because they're responding to the norms, but I think that often it happens the other way around. People change their minds and the change causes a change in the norms, like with gay marriage. It happened amazingly fast that the norms were against it and now the norms are heavily in favor of it. ### Robin Hanson 00:32:07 - 00:32:24 But it sounds like you're accepting there is this thing called a norm. It's a thing that exists. It's not a figment of our imagination, right? And it has force. It influences behavior. Norms definitely influence it. Okay, so then we could talk about different societies having different norms to their benefit or harm. ### David Deutsch 00:32:24 - 00:32:54 Yes, that's true. I was just resisting that individual behavior is determined or overwhelmingly determined or something by the norms of their society. Clearly when we look at a society's norms and we look at the behavior of the society, we see a substantial correlation, right? The behavior of individuals is quite correlated with the norms of that society. Causation could go either way to account for that correlation. ### Robin Hanson 00:32:54 - 00:33:16 Okay, but there is some shared causation that is producing the correlation between norms and behavior, right? So we can think of a society as having maladaptive or adaptive norms in the sense that these will produce adaptive behaviors or maladaptive behavior. ### David Deutsch 00:33:16 - 00:33:24 Yes, and they can also have maladaptive or adaptive fads which were not caused by the culture. ### Robin Hanson 00:33:24 - 00:33:33 Well, we could think of that as another part of the culture. We could think of fads as part of a culture, but they aren't necessarily the same as the norms of the culture. ### David Deutsch 00:33:33 - 00:33:52 I do want to distinguish between one thing that you say happens, which is that people don't do a thing if other people hate them for it. And the other way around is that people do a thing regardless of whether people hate them for it or even because other people hate them for it. So both those things I think can happen. ### Robin Hanson 00:33:52 - 00:34:05 But surely not at an equal level of magnitude, right? If it means anything to say that a society has a norm, it's that in fact most people in the society are following the norm. If most people are defying the norm, it's not really a norm of the society, is it? ### David Deutsch 00:34:05 - 00:34:34 Well, you can say that most people follow the norm of, I don't know, hair length or something. But it's not at all clear that people choose their hair length according to what other people want them to do. Sometimes it's the exact opposite. In some societies, people don't care what your hair length is, and then you're free to choose your hair length any way you like, according to influence. ### Robin Hanson 00:34:34 - 00:35:08 But then in other societies, say in 1960s America, young men, long hair was very much disapproved. Or say certainly 1950s America, there was a norm against men with long hair. And then you would suffer a substantial cost in terms of your advantages in society if you chose to wear long hair. Now, there might be some particular niches in which you could get away with making that an advantage, but for most people, in most situations, for young men, it would be a disadvantage to have long hair. In 1950s America, for example. ### David Deutsch 00:35:08 - 00:35:18 Yes, and that changed. It wasn't that the norm as a property of the culture changed, and then that changed people. It's the other way around. ### Robin Hanson 00:35:19 - 00:35:54 I'm not trying to make a claim about the fine grain in time causal order of changes here. I'm just trying to say that different cultures can have different norms, and that will be to their benefit or harm. And that the evolution of that requires that there be multiple cultures, because it's hard for individuals in a culture like that to change their culture. So if we think about the analog with species, it takes fragmented habitats with many different species to have a lot of evolution of species. ### David Deutsch 00:35:54 - 00:36:09 Yes, in any one species that's huge in a big habitat, individuals could try to defy the species. I don't know what it's like with bacteria who can swap bits of DNA or RNA. ### Robin Hanson 00:36:09 - 00:36:13 Across species boundaries, yes. So would that change? ### David Deutsch 00:36:13 - 00:36:24 I don't know. So that's the situation with memes. We don't have to wait for inheritance. ### Robin Hanson 00:36:24 - 00:36:49 We do see distinctive bacterial species. Even though bacteria can do this exchange, we definitely see clumps of bacteria that share many features together that interact together on the basis of those shared features. And presumably they are coordinating together with these shared features, and that if any one of them deviated, they would suffer the cost of deviation of a shared coordination of a bacterial species. ### David Deutsch 00:36:49 - 00:37:12 If the gene that they swap with some other species helps those other species, then it might spread like wildfire in the other species, and the original species might be unaffected. Or it might be swamped out or whatever. And that's what it's like with memes. ### Daniel Martin 00:37:12 - 00:37:23 Yes, the group selection arguments. Why can't we just choose Robin? We just choose to have a culture where we allow some variation and we don't have this problem. ### Robin Hanson 00:37:23 - 00:38:51 Well, I think it's hard to just choose your culture per se, whatever features you're trying to do. That's kind of the point of a culture having shared norms, status markers, et cetera, that makes it hard for individuals to just defy their culture. If there's a pressure against them. I think we all agree it would be tricky for issues where we all have to move the force, making us all agree on the thing. Whether or not that's caused by a meme or not. I mean, like in our society, again, education has a high value. Now, you might say you don't actually use the stuff you learn from education. It's mostly a way of showing off that you have capacities. In principle, you could show off those capacities by playing video games. And some people might say, the hell with school, I'm going to play video games, and that's how I'm going to show off my capacities. But that just won't go typically very well for you in the sense that you won't get prestigious job offers and people who want to marry you from the basis of your video game performance. Now, that may change in the future. Some people are in that. But that's the point of saying the whole culture would need to be changing its attitude toward how impressed they are by video game performance, because in order for and that may eventually happen. But at the current moment, if you had a son or a relative who is young and they were considering this video game strategy, you might well have some words of caution for them about how that's going to go for their life. Right. ### Daniel Martin 00:38:51 - 00:38:57 Should they be living the life just for themselves for their own purpose, David? And then you can't make the argument that this wouldn't go well for. ### David Deutsch 00:38:57 - 00:39:42 Well, words of wisdom, if they're not coercive words of wisdom, presumably can't do any harm. But I think to predict to predict that somebody will suffer the consequences in 20 years time of choosing video games rather than quantum physics requires a prophecy that I don't think is possible. The quantum physics itself was a thing that had no job prospects when let's say quantum computers were first invented and now has massive job prospects. Winston Churchill played with his toys and that was instrumental for him. Yeah, that's a good example. Yeah. ### Robin Hanson 00:39:43 - 00:40:20 So, yeah, the concept I want to appeal to is this idea that there are different cultures and in each culture they have some features of the culture like the weight they put on education cornerstone versus capstone marriage how much attention you're supposed to pay the kids. I want to be able to talk about cultures as things that exist with these features. And if I can do that, then I can make the rest of the argument about cultural change adaptation but this is the thing I want to be able to refer to if I can't point to this and talk about it that I'm lost here. Is there a way to talk about cultures as things that have some features. ### David Deutsch 00:40:20 - 00:40:36 Yes, but I think you also want them to be causally autonomous, whereas I think they have massive causal input from other things, such as individual creativity. ### Robin Hanson 00:40:36 - 00:41:48 Of course, but the analog with species is species aren't autonomous I mean species exist in a physical world and they have big features that have been selected for by their world. If they had different features they might be selected out but still species have features. We can look at a species and say, this species has long claws that one has thick fur. We can point to species and talk about their features and therefore we can talk about whether those features are adaptive, the features of a species and similarly for culture we should be able to point to it and say how much weight does it put on education, is that an adaptive feature of the species or not. In the 60s and 70s, we could say one of the features of our culture is that it is reproducing faster than its resources can supply and within one generation that reversed and now people are. But I presume you accept that I can talk about the features of a species that doesn't commit me to the idea that species won't be changing. And the same for culture, I can point to a culture with features and I'm open to the idea that the culture could be changing and even perhaps rapidly but still I need to be able to talk about species and their features or I can't talk about the cultural evolution of cultures. ### David Deutsch 00:41:48 - 00:42:04 So as species can have a feature like being doomed to resource depletion and you can have ideas, starting in one individual that may that can spread throughout the culture within one generation. ### Robin Hanson 00:42:04 - 00:42:27 And so, yes, to talk about yes certainly you know the feature changed. Right. Again, we could talk about these cultures having features that could change but nevertheless, they are a certain feature at the moment and we can talk about even the distribution of features of cultures in a space of possible cultures. Yes. ### Daniel Martin 00:42:27 - 00:42:34 I don't think this would be fully complete unless we discuss the solutions, Robin, because it would be kind of like yeah yeah let's get on to that. ### Robin Hanson 00:42:34 - 00:43:46 Okay, so I mean I just want. I mean, you have this initial, you know, few sentence summary, and at the beginning here and I just want to like rephrase that in these terms we've been talking about because that's just the thing I wanted to be clear about. Imagine a space of possible features of cultures, like, you know, that's one way we think about evolution there's a space of possible features or characteristics of an organism or a, you know, whatever it is we're talking about. So imagine a space of possible cultural features. And then you can think of in the individual species as a at a point in that space. And then over time, these points may well move. And then they may well split and multiply and grow or decline and go away. That would be the net effect of cultural evolution so cultures are at places in cultural space. And we might think of there's adaptive regions in that space, regions where if a culture is there it's more likely to grow and multiply and other less adaptive regions where they're more likely to decline and disappear. ### David Deutsch 00:43:46 - 00:43:49 So the axes aren't clear to me. ### Robin Hanson 00:43:49 - 00:44:16 So the axes are just these different features of, you know, how much weight do you put on education, how much attention should parents play to kids, these sorts of features of cultures that have costs for individual variation within them, that is, or even like, how much we disapprove of sexism or how much we approve of patriotism or disapprove of war or democracy. ### Daniel Martin 00:44:16 - 00:44:29 Is David frozen? Oh yeah, looks like it. Now he's back. Okay. You're back. Was I gone as well? ### David Deutsch 00:44:29 - 00:45:00 You were gone from us and now we're all back together. Okay, I was just saying, I may be interrupting your exposition but I think in this space, I think there's a lot more there than cultures. There are many, many more ideas in people's minds in a culture than just the ideas that constitute the culture. ### Robin Hanson 00:45:00 - 00:45:01 Absolutely. ### David Deutsch 00:45:01 - 00:45:03 Unlike DNA. ### Robin Hanson 00:45:03 - 00:45:14 Right. This, this is not intended to be a full description of everything in culture. It's not a full representation of everything. It's a reduced representation that lets me focus on the thing I want to focus on, which is the shared features of cultures. ### David Deutsch 00:45:14 - 00:45:22 It's being affected continually by the other ideas that aren't in the culture. ### Robin Hanson 00:45:22 - 00:48:26 Yes. And in fact, there are little swirling eddies of mini cultures that either grow or are destroyed themselves. And the shared features of the culture are just the very unusual ones that have got themselves spread to the entire population. I agree. So, as I said, I wanted to focus on this idea that cultures are units and they have some shared features. And I wanted to say there's all these other features that in fact we don't have problems with it. So the problem I'm pointing to is with this particular thing I'm trying to describe. And I'm saying, you know, lots of other stuff exists that we don't have problems with and they're doing great. But here's the thing where there is a problem. So if we think about cultures as units and then they have features that are shared and then that's a limited set of all possible features of behavior, etc. But still we can think about the health of cultural evolution as this process by which, say, the adaptive region moves around in the space. And the points of culture need to track that adaptive region. That would sort of be the first task of natural selection is not to invent new things. It's to like save old things. Not die and break because as things change, they need to be repaired. So we can think of there's a space of possibilities. There's an adaptive region where points in that space do the best. And that place moves. And there's a set of points, some of which are right in the middle of the region and that are that grow and multiply and other points that are out of the region and declining and disappearing. And the point of natural selection is to track an adaptive region. One of the points is not the only point, but this is like the first task natural selection needs to achieve is to correct errors and manage to track changes of the environment. And okay. And so there's four parameters of this process. One parameter is how many points are there. That's the variety question. A second parameter is how strong is the selection pressure stuff in the middle growing stuff far away declining. That's a parameter of the process. These are subcultures or like they are these things that share internally. So whatever unit has to share say for respect for education. If it's about how much do we respect education, it's whatever the cultural unit size is such that we need to have that the same internally and then other cultures can be different. So that might differ for different features. But again, I want to just point out there's four parameters here. There's a variety, which is the number of points. There's selection pressure. How strongly did the stuff inside grow versus outside shrink? Third, there's the rate of change of the region that you're trying to track. If it changes very slowly, that's easier than if it's moving around really rapidly. And finally, there's the rate at which these points just randomly wiggle, you know, independent of this adaptive selection force. ### Robin Hanson 00:48:26 - 00:49:41 There just could be other forces that make them move around. So those are the four key parameters of this process. And I'd say you can imagine some parameter values here where it all works great, right? You've got a million points. The selection is really strong. The region doesn't move very fast at all. The points don't move very fast at all. This thing is just going to track that region. And that's the way the world was three centuries ago. Three centuries ago, the world had hundreds of thousands of little peasant cultures, each of which had a thousand or so people near the edge of survival, lots of wars, pandemics, famines. The world wasn't changing very fast. That is, the world economy doubled roughly every thousand years. So the adaptive region you're trying to track was moving very slowly. And cultures were very conservative. They tried not to change very much, which meant that they weren't doing much random moving. So all four of those parameter values, say three centuries ago, I would say made for a healthy cultural evolution process here where this cloud of points would track the adaptive region. And then the claim is since then, these parameters have moved off in the bad direction. ### David Deutsch 00:49:41 - 00:49:57 Are you sure that 300 years ago is a good place to think of it? Because there were wars of religion, there were wars of politics, there were new ideas like nationalism and maybe 500 years. ### Robin Hanson 00:49:57 - 00:51:01 I'm happy to go 500 if you like, if you think 300 is too recent. But still 500 years ago, again, the world had a population of say half a billion. Each of these cultures was a thousand or 2000 people on average, a little peasant culture. So we divide one into the other and we get, you know, what, 500,000 cultures, right? And the benefit that the memes got is that they survived. So the society didn't change. The, you know, culture stayed adaptive would be the way I would say it is that. And that meant the people in them stayed adaptive because people have to be in an adaptive culture to be adaptive. That's part of what it is to be adaptive. Adaptive in the sense that they can successfully pass on their memes. And they can't if the population dies and goes away. So in order for memes to spread, there have to be people to spread them. So there is a interdependence of the people and the memes. ### David Deutsch 00:51:01 - 00:51:09 Yes, I think it was rare for populations to disappear altogether. ### Robin Hanson 00:51:09 - 00:51:44 What would happen is that they shrank and the survivors were the most committed to keeping their culture alive. If you imagine some local peasant culture that just decided to pick a whole new norm of marriage, say, and that just went badly in the next famine or something, then that culture might just disappear and the people in it might just be spread to the other nearby cultures. And then the people wouldn't all disappear, but the culture would disappear. That new attitude toward marriage would just go away. Yes, because it wasn't adaptive in that context. ### David Deutsch 00:51:44 - 00:52:05 Yeah, yeah, that's what I call a fad. So in your model, Robin, you're admitting that the degree to which entire cultures are good units of selection can vary a bit. So that can vary. Axes can be added and removed to the graph. And then sometimes the evolution can be dominated by axes that you hadn't predicted in the past. You're happy to accept all that? ### Robin Hanson 00:52:05 - 00:52:57 I think so. Yes, that is, we're focused on the features that are shared in culture. So we're definitely just not focused in this imagined diagram. We're not focused on all the other features of behavior that are easy to vary, say, technology, business practices, things like that. And we're here suppressing them from this analysis because they're not the thing we're focused on. Here we're focused on the features shared in culture. So each of these peasant cultures has an attitude toward marriage, toward war, toward property, toward family, death. And those are going to be shared in that culture, that is, you don't eat the dead. Most of these peasant cultures say it's really bad to eat the dead. If somebody eats the dead then they're going to be shamed and disapproved. So they will not eat the dead, and that'll be a feature of the culture. We're focused on features like that. ### David Deutsch 00:52:58 - 00:53:01 And eat the dead won't be copied, I think. ### Robin Hanson 00:53:01 - 00:53:11 Okay, but then that cultural element of the culture is maladaptive. Right. If the culture adopted it then the culture would tend to go away. ### Daniel Martin 00:53:11 - 00:53:23 I think even if David maybe doesn't agree with the whole framing, maybe he'll just agree with your solutions. So, okay. Okay. So, yes. The key point here is to say, in the last few centuries variety is way, way down. ### Robin Hanson 00:53:24 - 00:54:21 Okay, selection pressures are way, way down. The rate of change at which we're trying to track is way up the economy is doubling every 20 years or less. And the rate at which these things vary has gone up because we've come to celebrate cultural activism, people trying to cause cultural change. So now all four of these parameters are in a bad regime, arguably, and therefore this cloud is no longer tracking the adaptive region. And therefore plausibly drifting into maladaptation. And so that would be the argument to expect cultures to be becoming maladaptive, including low fertility as say the best sign. But it would just be the symptom of a deeper cause that this process of adaptation is broken and we're no longer tracking the adaptive region. So that's the hypothesis of a problem based on these parameters. And I agree that's not an absolute, you know, settles the issue sort of thing. It's a plausibility argument only. ### David Deutsch 00:54:23 - 00:54:39 Yeah, I'm trying to get my head around this picture because the points in this big space of ideas, there's a cloud of points which constitute a culture, then there's an adaptive region. ### Robin Hanson 00:54:39 - 00:54:48 So each point is a culture. That's the idea. That is the points are cultures, not that a cloud of points is a culture. Maybe that's confusing. ### David Deutsch 00:54:49 - 00:55:10 The points are cultures. But then, okay, I'm trying to imagine how to include the fact that there are many more ideas around than the ones that constitute cultures. So, and those are ideas. ### Robin Hanson 00:55:10 - 00:55:24 Yes. So you could think of, we're looking at a subspace of a much higher dimensional space. And if you focus on the higher dimensional space, you'll see a lot more detail and your modeling and things. But we're trying to focus on this projection of the high dimensional space into the subspace of the features of cultures. ### David Deutsch 00:55:26 - 00:55:45 Even so, the cultures are made up of people. Right. So the people are changing their values and that changes the entire culture. So the point, the culture, is actually vast numbers of points. ### Robin Hanson 00:55:47 - 00:56:25 I'm not sure. Right. So, I mean, you might be familiar today with sometimes there are graphs of nations where each nation is a point, and it's looking at average income, average fertility, you know, square feet housing per person. When you look at a graph like that, you're obviously suppressing a lot of information, the variation with each in each nation, for example. But still, it often does make sense to think about nations as points in a space of possible nations. If you want a law of motion of that graph to like, either explain why it moved from here to here in the past or predict what it's going to do in the future. ### David Deutsch 00:56:27 - 00:56:31 It might be that you're missing all the determinants of what makes it move. ### Robin Hanson 00:56:31 - 00:56:55 It might be. Right. But still, for example, you can look up, there are some websites where you can show points, other nations, increasing in income over time, fertility changing. You can watch the cloud of points move around. And then you can ask yourself to what extent is the thing as I'm seeing an explanation of these changes versus other things I'm not seeing. But still, you can make sense of that graph and that video showing the points moving around in space. ### David Deutsch 00:56:57 - 00:57:08 Yes. So a point is trying to move towards a region where it gets replicated in the next generation. ### Robin Hanson 00:57:08 - 00:58:07 You might say there's a healthy region for nations and nations in that region will cause more other nations to emulate them and have more children, etc. Like win more wars, a space of possible or you could do it for universities. For example, you can have a space of possible universities, which universities are growing and thriving and attracting more students and funding. Other points are declining and becoming right. That's a thing. You could make up points. This is just a way we abstract for many sorts of units of analysis. You could do it for firms. For example, you could have different for-profit companies, how big they are, how many countries they sell in, what's their product line, what's their debt to equity ratio. You can analyze units. Of course, you know you're suppressing enormous amounts of detail when you aggregate these things to single dimensions here. But that's often still a way to think about larger units and how they change over time. ### David Deutsch 00:58:07 - 00:58:29 I can imagine a big firm which is right in the middle of the advantageous region and along comes a fad that makes most of their employees think that what the company as a whole is doing is evil. Now you couldn't have predicted that without predicting the fad. ### Robin Hanson 00:58:29 - 01:00:09 But you might have this parameter of do the employees think the company is evil? Now you will see that along that dimension, this company moved a lot. There was a huge move of the company from along that dimension of employees think it's good and employees think it's evil. That's not saying we know what the cause of that, but we can see in the space that we can represent the fact that the company is now thought of as evil as a change in that space. The space is still representing that change, even if it's not representing all the causes of it. So that could allow you to explain what's happening to the company in terms of the parameters that you've chosen and think are important. That is, the adaptive region will be in terms of whether the employees think it's evil. Well, you may not have realized that because when you were setting up this thing that didn't occur to you that there could be such a thing as people thinking that their company is evil. Right. If you're going to do an empirical analysis, you'll have to decide what to measure and you may miss things. But here we're just doing a theoretical conception of like, just imagine the space and imagine that there are adaptive regions in the space and that some of the points are in the regions and some are out. So we don't actually need to have listed all the dimensions of the space here to just think about there's a space. And this is the key concept of natural selection, really, right? That there's a space of possibilities and that there's regions in the space that are more adaptive than other regions and that points in the space will grow and multiply and points out of the adaptive regions will decline in a way that I would say this is our fundamental concept of natural selection. Yes. ### Daniel Martin 01:00:09 - 01:00:18 It is a force, but so what is like a general we are doomed problem. How bad is this? ### Robin Hanson 01:00:18 - 01:00:46 So, yeah, let's talk about solutions. So this is an unusually hard problem to fix. That is, if you think we've drifted off from the adaptive region. Like, one way to fix it is to just destroy everything. You know, society collapses to, you know, vast, you know, low tech simplicity and everybody's in little farming villages again. Yeah, that would solve the problem but it's, you know, seems way. ### David Deutsch 01:00:46 - 01:00:56 Our memes don't get replicated. So like random memes or random variants do get replicated and therefore society doesn't work anymore. Is that what you're proposing? ### Robin Hanson 01:00:56 - 01:01:38 Well, I would actually say more concretely, we have these insular fertile subcultures like the Amish and the Haredim who are doubling every 20 years and have been doing so for a century and that's how the Christians took over the Roman Empire by basically doubling every 20 years for three centuries. So my default scenario is that our dominant world culture declines and is replaced by groups like the Amish and Haredim and then they go on to rise again and do similar things. That's that would be my default scenario. It's not human extinction, but it is the loss of what we treasure about our dominant world civilization. So the Amish, Haredim, they're not into free speech, democracy, open inquiry. There's a lot of things that we treasure about our dominant world culture that they don't value. It's the loss of those that I fear. ### David Deutsch 01:01:39 - 01:01:48 Yeah, so that would be bad and it's not just that we like them, it's that they are functionally necessary for societies to continue to improve. ### Robin Hanson 01:01:48 - 01:01:58 Well, I certainly would hope so, but of course if the society that has them is losing and the ones that lack it are winning, then something is wrong. ### David Deutsch 01:01:58 - 01:02:05 There's never any guarantees. We can make mistakes of an arbitrarily large kind and we could do much more. ### Robin Hanson 01:02:05 - 01:02:39 And we may be doing that, but yes, so what I would like is to say, so I would say the analog is you're on a ship heading for an iceberg. You can either turn the whole ship or you can get off on lifeboats. Like we can try to fragment and create new subcultures that just separate from our dominant world culture and then like have new features that grow and that's what kind of the Amish and Haredim are doing, but it's actually really hard. Or we can try to influence our shared world culture and move it in a direction that avoids the iceberg here, like change our shared culture as and that's clearly possible, but also very hard. ### David Deutsch 01:02:40 - 01:02:58 Well, I certainly agree with that and I think the lifeboat thing has a disadvantage, although it might lead to ultimate survival. It has the disadvantage that the, if I can extend the metaphor, that all the possible lifeboats are far less powerful than the ship was originally. ### Robin Hanson 01:02:58 - 01:03:08 Right. When the ship hits the iceberg, it's suddenly going to be a lot less powerful and the lifeboats may be the most powerful boats around. ### David Deutsch 01:03:09 - 01:03:45 Yeah, they might be, but yeah, so I agree that the way I put it, sort of the Popperian way, is that we should be looking for aspects of our culture that are maladaptive. And we should argue that they are and we should argue that people should change their ideas and their behavior. And that's turning the ship. That's appealing to the shared world culture and reasonable people and to try to convince them that their culture has problems. ### Robin Hanson 01:03:45 - 01:04:48 Yeah. That they should change them. But I mean, a key problem there is because we have made cultural activists our most prestigious people, there are huge numbers of people out there already fighting in this argument for what they claim would be better cultures and why we should agree with them and move the culture in their direction. So we're competing in this very loud voice of all these other people pushing culture in other directions. And the question is, can we get them to listen to the adaptation voice? Because, as you know, many people think adaptation is bad. That is, say, social Darwinism is this terrible word to describe terrible people who try to make their cultures adaptive, because, of course, they do that by genocide and war and, you know, terrible things. And that's part of the obstacle here. Like, one potential solution is you just get people to be more social Darwinist as the collective culture. But, like, people are all set up to hate that idea. And you have to overcome that. ### David Deutsch 01:04:48 - 01:05:01 Yeah, yeah. Ideas are powerful, though, just because there are a million voices out there doesn't mean that if you are right, you can't outshout, out-yell them. ### Robin Hanson 01:05:01 - 01:05:45 So I'll just describe two other solutions here. Okay. One is, I've got more solutions. None of them are that attractive in the sense of being that likely, but I have some more. One is, we let capitalism go wild running everything. Capitalism own governments and run them. Capitalism make for-profit orphanages. Just all aspects of culture, capitalist organizations own and run them. And capitalism contains enough strong selection pressures that plausibly, though, that would produce adaptive behavior because capitalism just faces strong selection pressures. And if it's not adaptive in the long run, those things will die. We have to argue for that to be introduced, which is the... ### David Deutsch 01:05:45 - 01:05:54 Yes, but people hate this idea, as you probably know. Yeah. Although Mr. Milei got himself elected somehow. Indeed. ### Robin Hanson 01:05:54 - 01:06:15 But it's really striking that for-profit is the dominant form of organization for most small organizations in the world, and there's almost no large scale governments that use it. There's pretty much none. And that's a really stark... apparently above some level, people say, nope, we do not want for-profit governance here. That's just the end of it. But you can see that's a potential solution. ### David Deutsch 01:06:15 - 01:06:18 Yes. And the fourth one? ### Robin Hanson 01:06:18 - 01:08:27 Well, the fourth one is I happen to be the inventor of an alternative form of governance, which I think has the potential to just be much more effective than other forms of governance. It's called futarchy based on betting markets. Slogan is vote on values, bet on beliefs. And so if a big chunk of the world, a big region, say, adopted this powerful form of governance, and then it's a form of governance where you give it a concrete goal with a metric, like you've given an outcome metric to achieve, and then it's very competent at achieving that metric, then we get that region to pick a metric that happens to be inconsistent with civilization collapse. So it's not... I don't think they would... and it has to be a metric they will treat as sacred, which means that they'd be ashamed to drop it and they'd be proud to sacrifice for it. Because... so for example, an example would be when do a million people live in space? So I think we're at the point where if civilization peaks and declines over the coming centuries, that will put off when a million people live in space for many more centuries. So if you set up a futarchy that says your goal is the earliest possible date when a million people live in space, then the governance mechanism will have to say, OK, well, how do we prevent decline? Because that's getting in the way of our goal. And then because people treated this goal as sacred, when it demanded sacrifices, which it would, and things they didn't like, they would be proud for that instead of bothered by the fact that they were being asked to make sacrifices. And then when some people suggested, well, let's just dump this goal, to hell with it, people would be ashamed of that because, again, it was their sacred goal. So you can see this is a big ask. I have to prove that there is, in fact, a much more competent form of governance and get it adopted. And then that isn't enough. They have to adopt a sacred goal, but that's not enough. It has to be a sacred goal that happens to be inconsistent with civilization decline because then the mechanism will, in fact, do what it takes. So this is a meta solution. It says, I don't know how to solve it. Just create a competent agent and hand them the problem and say, please solve the problem. And if they're competent enough, that works. So you can see it's a big ask as well. ### David Deutsch 01:08:27 - 01:08:53 Yeah, I mean, there are other ways for an idea to be persistent, like a goal like, you know, million people living in space is a possible goal. There are other ways for that to be a persistent goal for as long as it takes to achieve it, other than making it sacred. For example, it could be right. And then it's much easier to argue for it. ### Robin Hanson 01:08:53 - 01:09:43 Well, here we want to go. So the question is, could we just make the goal adaptiveness and say, hey, government, please make our culture adaptive? And unfortunately, I just don't think that has the chance of being sacred to people. Even if it's true in some sense, I don't know if it really would motivate people enough to actually adopt it and persist with it. I would like it to be true, but I have to. So I did some polls of what people thought would be a goal that a lot of people could get around over a long time. And so this million people living in space was pretty much the top of the list. I had 32 other options I could show you. But let's just make society adaptive. Just did not poll very well. So that's why I've looked for a more indirect solution. I agree the direct solution would be the clean one. Just have the government say, please make us adaptive, make us last and grow into the long term. ### David Deutsch 01:09:43 - 01:09:51 Yeah, I think that that goal, as stated, is not nuanced enough. It doesn't have... ### Robin Hanson 01:09:51 - 01:09:59 I'd be happy to propose whatever variation you thought was more nuanced. I'm probably going to support it. I'm not going to object to it. The question is, can you get people excited about it? ### David Deutsch 01:10:03 - 01:10:12 In a way, that is what happened. I'm not sure that the early Christians outbred other people. ### Robin Hanson 01:10:12 - 01:10:15 They doubled every 20 years for three centuries. That's literally how it worked. ### David Deutsch 01:10:15 - 01:10:17 But they got lots of converts. ### Robin Hanson 01:10:17 - 01:10:29 Their main form of convert was babies who were set up for exposure that they just raised as their own babies. That was a very reliable way to get converts. ### David Deutsch 01:10:29 - 01:10:39 I'm not sure that that's this prevailing theory among historians of early Christianity. But maybe I can imagine that being true. ### Robin Hanson 01:10:39 - 01:10:57 They also were willing to stay in cities in pandemics and treat the sick. And then those sick that they saved would also join the Christian church. That was another main mechanism of conversion in the early Christian church. It wasn't just preaching on the street corner and having people believe them. That wasn't the main mechanism of conversion. ### David Deutsch 01:10:57 - 01:11:03 But these are mechanisms of conversion. It seems to me, Robin, you should be happy with a partial solution. ### Robin Hanson 01:11:03 - 01:11:07 Sure, absolutely. Show me. ### David Deutsch 01:11:07 - 01:11:15 If we just make it until we have our civilization then expand enough that it fragments due to the speed of light. ### Robin Hanson 01:11:16 - 01:12:21 Human level AI is actually not a solution here. But it might be a temporary reprieve of the symptoms. In general, when you have a sick patient and they have symptoms that are killing them, it's good to fix the symptoms even if that doesn't fix the underlying problem. We don't rely on human level AI just to solve the problem. But once we have civilizations in space that are separated, then the ship isn't heading towards an iceberg that it will crash. If we do nothing, it will just diverge. So, I think that getting across star systems would in fact produce substantial variety that would be sustainable. But it's just getting to that point. I fear that that's just not in the cards the next 50 years. I'll live with some partial solution that buys us 10 years. Look, I am desperate here. I really mean to say none of the solutions I've come up with seem very attractive or reliable. So I am really wide open to solutions here and happy to endorse anything. Some people think we just need more cults, more religious cults, other cults. And it's true. Cults are a kind of cultural variety. And if we just embraced cults and had more of them, that would help. I don't know if it was going to help enough, but sure. I'm all in favor of more cults. ### David Deutsch 01:12:21 - 01:13:00 Funnily enough, I think our culture, let's say the civilization of the West, the thing that's wrong with it is not that the bulk of the ideas are not adaptive. It's just that there's been an accumulation of fads. And we happen to be, the culture happens to be vulnerable to certain types of fad. Like you said, glorifying dissent, just any kind of dissent, rewards people who just dissent for the sake of it. ### Robin Hanson 01:13:00 - 01:13:02 Exactly. Exactly. ### David Deutsch 01:13:02 - 01:13:10 Like an organism can just have a parasite that occupies only a small fraction of the volume of the organism. It can still kill the organism, right? ### Robin Hanson 01:13:10 - 01:13:12 Yes. Yes. ### David Deutsch 01:13:12 - 01:13:33 So, but that means that thinking of the goal as curing the society as a whole may not be the right way to think about it. It may be the Popperian way of looking for bugs. Finding the flaws and squashing them, absolutely. ### Robin Hanson 01:13:33 - 01:13:34 Yeah. ### David Deutsch 01:13:34 - 01:13:39 Like looking for the vermin in the house and fumigating the house to kill off the vermin. ### Robin Hanson 01:13:39 - 01:13:54 Absolutely. That's if you can get people behind it. But that's the question. Can you get, but again, I'm desperate. I am not squashing any proposed solutions here. I do not at all mean to be dissing any proposed solutions. I mean to say, help, for God's sake, we're in trouble. ### David Deutsch 01:13:54 - 01:14:11 In regard to the economy, you've got Milei and you've got Thatcher and they didn't have to bring anything down. They just needed to make tiny tweaks, actually, in order to set the society on a much healthier direction. ### Robin Hanson 01:14:11 - 01:14:45 Well, I mean, for example, overregulation, I think, is one of the key ways in which society has been going maladaptive for a long time. And there have been these times when overregulation reduces, but they've been small exceptions to the larger long term trend. Even the same for fertility. There was the baby boom in the World War where fertility was going up, but that didn't last. So I am happy to support and encourage any reversal of any of these trends everywhere. But I am worried about whether they're sustainable and can last long enough to really be enough. ### David Deutsch 01:14:45 - 01:14:54 Well, I think worrying, well, actually not worrying in the unpleasant sense, but thinking about these things is highly desirable and necessary. ### Robin Hanson 01:14:54 - 01:15:17 And so I'm trying to get other people worried. If I can find a way to get you worried or express my language in a way that already expresses the worries you've already had, I'm eager to find ways to sort of make common cause with others. There's no way this is going to be solved by me, by myself. Obviously, the only plausible chance we could have is to get more people to think about and worry about the problem. ### David Deutsch 01:15:17 - 01:15:24 It could be that the underpopulation problem is actually soluble by very small changes. ### Robin Hanson 01:15:24 - 01:15:42 I think it is. But the problem is whether people are willing to do that. I think the main obstacle to most of the reasonable fertility fix is, in fact, the cultural changes it would cause. People look ahead and they go, oh, but if you did this, that would say have fewer women going to, you know, grad school or whatever. They go, we don't want that. ### David Deutsch 01:15:42 - 01:15:54 Yeah, there's this idea you could just have one jurisdiction where you try something and then people see that it works. And now they don't mind as much. But people do have a memory that we forgot a lot about communism and what happened there. ### Robin Hanson 01:15:54 - 01:16:25 Right. And the world has much more convergent culture than many people realize. I think COVID was a pretty dramatic example of how the whole world basically did COVID pretty much the same way, even though there's hundreds of countries that could all do things different. They didn't. And if you look at nuclear energy, you know, banking, medical ethics, a large range of regulation areas, basically the whole world does them pretty much the same, even though the world could support enormous variety and experimentation. They don't because people want to be part of the shared world culture and they want to do things the same as the rest of the world does. ### David Deutsch 01:16:25 - 01:16:32 Yeah, I think each of the individual ones you mentioned, we will solve. So it's just if there's too many of them, we can't solve them fast enough. ### Robin Hanson 01:16:32 - 01:16:55 That's the worry. Absolutely. Yeah, there's a time deadline here. I don't know, three centuries. You have three centuries, maybe two, until, say, the Amish and the Haredim just take over. And that's how long you got. And so it's not five years or 10 years. It's a nice long chunk there, but it's a really hard problem. ### David Deutsch 01:16:55 - 01:17:00 Yeah, I think we're all on board supporting creative solutions to this. This was fun. ### Robin Hanson 01:17:00 - 01:17:27 Yeah, I agree. All right. So I guess, I mean, you've had some worries. I've described some worries. Are they the same worry, different worries? Have I added to your worries? How would you frame the worries I've described in the worries you already had as, you know, near relative to each other? ### David Deutsch 01:17:27 - 01:17:54 I think they're not quite the same because, but the difference may not matter to you. I think that this is all about individual ideas. And the changes are all about individuals changing their minds because they've been persuaded by something. And this can turn on a sixpence. ### Robin Hanson 01:17:55 - 01:18:08 I agree. That's how cultural activism works. You know, as cultures change through cultural activism, which works through individuals being persuaded to join different sides of cultural battles. Yes. So it's individuals who decide to join them. ### David Deutsch 01:18:08 - 01:18:46 It's not just like, I think, take Harry Potter. So Harry Potter was created by one person sitting in a cafe and writing stuff. And the memes spread everywhere and caused changes. They caused changes like changing the law in Scotland about trans, you know, things that you wouldn't have expected it to have an effect on. And I think that's how fads change. And we're the victim of some very bad fads. ### Robin Hanson 01:18:46 - 01:18:58 So you want to promote both the individual and the cafe writing that book, but then all the other people who grow, who build on it and share it, that's also part of the process. Without that, it also doesn't happen. ### David Deutsch 01:18:58 - 01:19:17 Well, she didn't have to do much to make it happen. But other people did. Yes, but they did it. They weren't trying to improve society. They were trying to get an interesting book to read. You know, they were just trying to improve their individual lives and they didn't give a thought to the effect on society. ### Robin Hanson 01:19:17 - 01:19:23 So right. Cultures change both through cultural activism and through other processes that aren't mediated by cultural activism. ### David Deutsch 01:19:23 - 01:19:35 Right. You know, people tend to want to credit the activists perhaps more than they deserve because. But still there is a force of cultural activism. It does have influence. It may be overrated, but it's real. ### Robin Hanson 01:19:35 - 01:19:46 And we're happy to take any influence here. So if you can write, if anybody who's listening to this can write a novel that helps deal with this problem that people just want to read because it's fun, that would be great. We very much approve. ### David Deutsch 01:19:46 - 01:20:02 I agree. I agree. I think that's how it always has happened. And that's also how the bad things happen. And maybe that's why I'm not as worried as you are. ### Robin Hanson 01:20:02 - 01:20:31 Well, as you know, bad things have happened in history, right? Like empires have fallen really. The Roman Empire fell, that sort of thing. So there have been like large long-term bad trends, right? And long-term large good trends. The risk for fear here is that we're facing a big long-term bad trend, but we would love for it to be reversed. And if anybody listening has a way, please take the initiative and start. ### Daniel Martin 01:20:31 - 01:20:39 Yeah, are you going to write a book? Make sure the book is also good. That would help rather than just the only thing about it is to make people understand this. ### Robin Hanson 01:20:39 - 01:21:10 Right. And so I'm trying to write a book on this, but part of what's put me off is like just realizing how hard this problem is and thinking I really need to frame it well and think about, you know, how to, you know, framing matters a lot in these sorts of things. What sort of analogs you make, what terms you use. And then so that's why I'm very eager to sort of find other terms like you, David, are using and see how I can change my framing and terminology to integrate better with other uses. ### Daniel Martin 01:21:10 - 01:21:22 Well, thank you both for talking today. I guess we've given others something to listen to. And so I will stop the recording and we will share files soon. ### Robin Hanson 01:21:22 - 01:21:24 Great. See you.