#1329 Pierrick Bourrat: Evolutionary Transitions, Heritability, Adaptation, and Multilevel Selection

00:00 Ricardo Lopes: Hello everyone. Welcome to a new episode of The Dissenter. I'm your host, as always, Ricard Lops, and today I'm joined by Dr. Pier Ric Boat. He's senior lecturer and AR Fellow at Macquarie University in Sydney, Australia. He specializes in philosophy of biology, and today, We're going to talk about topics like evolutionary transitions, individuality from an evolutionary perspective, heritability, adaptation, multi-level selection, the extended evolutionary synthesis, and other related topics. So Dr. Borois, welcome to the show. It's a pleasure to everyone. Well,

00:39 Pierrick Bourrat: it's a pleasure for me too. Thank you for having me.

00:43 Ricardo Lopes: OK, great. So let's start with uh evolutionary transitions. I mean, what does that term mean exactly? What do people mean by that?

00:55 Pierrick Bourrat: So what the terms mean evolutionary transition, I would say it's quite a vague term, and I would say so the topic really started to be of general interest for biologists with the book of Maynard Smith and Saass Marie in 1995, the Major Transition in evolution. And so it's a broad category, and this framework has somehow been somehow criticized because it's kind of unclear what the transition is, and it can be very abstract what a transition is is a form of change in some sense, but Maynard Smith and Sasmari were saying that basically there is on one side a change in the level of individuality. And on the other side we could say a change in the way information is transmitted from during, during the evolutionary, during an evolutionary process. So people have them and different things, and I would say the topic I worked the most on is evolutionary transition in individuality. Which is in evolution, when do, uh, the level at which uh we consider uh individuals in a population, uh somehow shift from one level to the other. Now, I've been working with, uh, with, uh, in particular, a colleague called, um, uh, Andrew Barron, who is also at Macquarie University, who has been talking about evolutionary transition in cognition or cognitive architecture. So this is a more recent topic. And here, so if we try to link transition inability and transition in cognitive architecture, it's not necessarily very easy to do. Um, BUT I'm currently with him. We are working on a way maybe to reconcile and to see what would be a way to reconcile, and the idea is basically that when you have, when you're on one side of a transition, so before the transition, people have called the transition an event. It's more like, it's more like it's a spatial temporal event rather than a singular event typically. Uh, WHAT, what you have before the transition is a way to explain or to predict the evolutionary patterns, the evolutionary trajectory, and when you have a transition, it's not the case anymore. You need to invoke a different model or a different explanation. And I think the transition is a period in between, between these two, these two, these two ways to model evolution or to predict, explain. And so that would be one way to somehow, uh, reconcile the two. We need also to distinguish maybe a transition from something that we might call evolutionary innovation. It's not because something is new that the explanation that we have, uh, needs to change. So for instance, mutation is part of a classical population genetics model, and, uh, we don't need to invoke something new, although, uh, some mutation might be. Uh, MIGHT be innovative in a stronger sense and we might want to call them a transition. So there's a lot of uh play with Word in this, but if we want to distinguish a transition from an innovation, we, we will have to do some work on that,

04:16 Ricardo Lopes: right? But what are the main evolutionary transitions in our evolutionary history? What are the ones you would point to?

04:26 Pierrick Bourrat: OK, so if we take what I've said earlier, so if it's a model relative notion, if you talk to different people, they are going to have different answers as to what is, uh, what is a transition. If we talk about transition in individuality, uh, classically, uh, so you have the, the, the first, uh, the first, um, biological entity that emerge. Simple replicators, but I don't particularly like the view to see everything from the replicator's perspective. Uh, ALTHOUGH I'm originally coming from this perspective. I mean, I've been hugely influenced by Dawkins and, uh, but I think like there are, there are issues to see everything from a replicator perspective, which we can talk maybe. Later, but so you have the origin of life, which is basically probably the 1st, 1st major transition or first transition in individuality. Then you have this classical way to say, well, you have the simple replicators and then maybe chromosomes or ancestors of the structures. It's very hard to know now what basically using modern modern biology to know what were the ancestors. Uh, BUT basically, uh, you also have also the, the transition from prokaryotic to a chaotic, uh, cells, which happened many times, uh, in, in different, uh, taxa. Sometimes you have a double where, uh, so, um, I don't remember when, when algae has 4 membranes which we can, uh, think that there were two events of engulfment, uh, from one to the other, from, from, um, To create a eukaryotic cell, basically, then you have many transitions from unicellularity to multicellular organisms. Then you have the transition from eusocial to superorganisms. You could also maybe think about some multi-species. Some people think multi-species entities such as holoons are a new form of transitions, but that is just for individuality here in the. Of transition in cognition, cognitive architecture. So my colleague has a, has a framework where you have different cognitive archi architecture that emerged during an evolutionary process and you have these different transitions here, yeah, which

06:57 Ricardo Lopes: I'm. Yeah, just to touch on one quick thing that you said there, and we maybe will come back to this when later we talk about things like multilevel selection and the extended evolutionary synthesis. I mean, so, uh, because you mentioned Richard Dawkins and you mentioned that you don't really like to just, uh, focus on the, on replicators. I mean, so you don't come at these topics from a gene-centric perspective, right? I mean, you have a

07:26 Pierrick Bourrat: more

07:28 Ricardo Lopes: expensive view of it, or

07:30 Pierrick Bourrat: yeah, I mean, the problem with, with the notion of, I mean, I come from a replicator perspective, I would say. Uh, SO when, when I was, uh, when I was younger, I read, I, I was into memetics and all cultural evolution. I actually, I come more from cultural and human, human, uh, human evolution and human evolutionary psychology, uh, but so I was interested in this framework, but, um, and it's, it's an interesting framework. It's a good starting point framework, but over the years, basically one of my One of my obsessions has been the notion of scales in evolutionary theory. And in particular, if you think about the replicator and you change the scale at which you look at an object, basically a replication does not become a replication anymore. If you change the scale, if you give more details about the scale, and I think for instance in cultural evolution with the memetics, it's particularly a case where we don't have a good A good physical template and a particular scale at which we can look at the cultural phenomenon. They are described by so many different scales that talking about replication, it's not, doesn't give me any real understanding of the processes here because I can really describe the process always as one which is a replication if I abstract away. Uh, IF I abstract away the, the description, basically, so that, that's my main reason why I don't like the, the replication. I mean, it's not that I don't like, it's, I find it limited, yeah,

09:09 Ricardo Lopes: yeah. So let me just take a quick step back to ask you because we've been talking here about individuality and transitions in individuality, but what is individuality actually in evolutionary biology? What, what does that mean?

09:27 Pierrick Bourrat: So, uh, similar topic, um, similar difficult question to answer because, um, so I would say like probably the simplest way to talk about individual for me, what I mean by it is basically a unit of selection. So an entity or an object more than an entity that natural selection sees, and I use the square code because it's a, it's a metaphor, of course, but basically it's Yeah, a description of, so a biological entity is a biological material, let's say, that we can describe in a way that behaves as a unit of selection. Now, of course, the notion of a unit of selection is itself very complicated. And so I'm going to rely here on, on the framework proposed by Lisa Lloyd, and she has a kind of 3 to 4 ways to understand what one means by a unit of selection, which is an interactor, and an interactor comes from the notion of interactor comes from David Hall, who was criticizing. The notion of vehicle of Richard Doin as being a bit too passive, uh, so he considers that interactor, and many people do actually, um, basically they are the objects that interact with their environment. OK, so in the minimal sense that's an object, it's somehow bonded, not necessarily physically bonded, but causally bonded, and this physically or causally bonded object interacts with its environment and If you have a population of objects interacting differently with their environment, and that leads to different evolutionary success. So here we are in the territory of fitness, which is also a real can of worms. It's very, it can be very complicated, and if we give a spin on scale, it's becomes even more, even more complicated. But basically you have this evolutionary, these objects that are in a population, and they have a different evolutionary success, and that's minimally what a unit of selection. I then you can have objects that have very special modes to be successful because for instance they reproduce in a particular way, so they reproduce very with high fidelity, but I think, and I was more on this side a few years ago, but after having read more about the work of Of Lisa Lloyd and a few other people, I think we can have a more a more expanded view of what a more minimal view of what a unit of section is just as an intertractor, but then you can go and say, well, it might be a reproducer, so an entity that reproduce in a kind of modular way. Uh, OR it could also be, that's what people have meant by unit of, uh, unit of selection, which is basically a unit of adaptation where adaptation is understood as a cumulative product of multiple rounds of natural selection. And she has another, another way to talk about it, which is another notion of unity, which is also a Um, beneficiary, and I must say the beneficiary, I'm not sure, I'm not sure which she says was especially as a way to talk about the Dawkins and view of evolution where ultimately the gene is the is the beneficiary, but basically a gene is always a part of another larger object and so we could say that the beneficiary is It's to me it's unclear what the beneficiary. People might have some view. They might have some view as I believe the beneficiary is this entity, but for me, yeah, it's going to be all of them at the same time.

13:39 Ricardo Lopes: Yeah, no, no, I, I mean, the, the, the conversation surrounding units of selection has become more and more complex over time. I, for example, just not very long ago, I interviewed Dr. Javier Suarez. He wrote. The book with Elizabeth Lloyd units of selection, I think it was published by Cambridge University Press or something, and, uh, I, I'm in the, all the discussions surrounding, uh, replicators, um, um. Uh, uh, WHAT, what were the other terms? I mean, I mean, I'm missing that replicators,

14:15 Pierrick Bourrat: reproduce,

14:20 Ricardo Lopes: yeah, yeah, yeah. All of those, uh, I mean, it's, uh, it's a very complicated kind of

14:26 Pierrick Bourrat: discussion,

14:28 Ricardo Lopes: particularly for people, particularly for people who are not very familiarized with philosophy of biology.

14:35 Pierrick Bourrat: Yeah, I could say that the level of selection units and level of selection is a, it's a, for me it's a very interesting topic because it has all to do with basically levels of explanation. Often you have to do, it has to deal with part-hole relationships, emergence, reductionism, so it's a and so there are all these angles that you need to take into account at the same time, which is why, why it's, I think, kind of complicated, yeah.

15:08 Ricardo Lopes: Mhm. OK, so I would like to ask you about a specific topic that I read about in your work, uh, because you talk about a particle level and collective level inheritance. So, uh, tell us about each of them and how do you look at the relationship between them.

15:28 Pierrick Bourrat: OK, so, uh, so particular, particular and collective level. So basically a collective, so it's an abstract description. An abstract description of a collective could be, for instance, an organism, and the particle would be the cells or the genes. So it's just a placeholder to have a two-level relationship basically. So I got interested in the topic of the relationship between levels of inheritance or heritability. It's basically a mirroring question about levels of interaction or units of selection, basically. Uh, AND typically, uh, when I say unit of section, so selection is also a difficult, uh, it's an ambiguous term because people sometimes mean just the selection process. So you choose basically, nature chooses some entities versus others, but there is another notion which is a response to sections. So I'm using here what quantitative. Uh, Geneticist described, which is selection and response to selection, and basically selection, uh, so it's what I've said, and the response to selection is, um, is basically plus heritability. And heritability. Basically what heritability does, it records what selection doctors what selection choose, it records it and sometimes even amplify those differences. So I was interested in the notion of inheritance between two levels because people have been talking a lot about selection. And kind of less about about inheritance. And so if there is a difficult relationship between selection, it's even more complicated once you talk about about selection. And so I was kind of, I've read quite a lot of work on this, and I come from a biology background where I've done like several years of population genetics. And what I've noticed is that people, when they talk about heritability, for instance, which is a population level statistics, does not correspond to what people intuitively mean in a tiny model where you have heritability, which means more the capacity to transmit with some fidelity in your your your offspring, and so It also heritability does not depend on the number of parents you have. It could be many, many, many parents who would still have, you could still have a high level of heritability. So when people say, well, if you have too many parents, for instance, they cannot mean heritability in the sense of the statistical population. THE statistics basically they mean it in a different sense. So I've investigated a bit the link and one claim that was often made is that basically you can't have a heritability as a collective if you have a heritability at the particular level, you don't have necessarily heritability at the at the higher level, and what I've shown is like it's it's false basically if you have the simple in the case of the simple possible. WHICH is an additive trait. So basically an additive trait is like if you have one gene, for instance, where there are two different, if you have multiple genes, they all have, uh, you can all add their effects. And for a trait like this, or like let's say. Yeah, you have A, B, C, and D, and they all contributes plus 1 to particular phenotypes. Basically, for all these traits, you trivially have heritability at the collective level. And in the case where you have a non-additive trait, so you can't give a very, a very, um, you can't give a complex analysis of, of uh non-additive trait because non-additivities are so many functions that could be non-additive. What you get here is a typically a very unstable notion of heritability. If you're in a population where reproduction occurs in a very, in a well-mixed way. If you don't have population structure, if you have population structure, then you can, uh, you can have uh heritability also at the collective level. Uh, SO it's a kind of very complicated. It's very complicated analysis, but it was just starting from this point that people say, well, you can't have heritability at the collective level, uh, when, when you, uh, start from a heritability of the, of the particular level, I, I think it, it was problematic here, so that's where I started to work on this.

20:10 Ricardo Lopes: OK, but on the topic of heritability, let, let's just clarify a little bit more what that term means, because I mean just to put it in simple terms or to explore a simple example, uh, people who are not very familiar with genetics, particularly, they might hear something like, oh, a trait like height is 70%. Charitable and what they interpret that as being is that oh it's 70% genetic and sometimes they even make a jump and and assume that means that it's 70% determined by genes or genetically determined. I mean, is that really how we should think about heritability?

21:02 Pierrick Bourrat: OK, so heritability, so there are multiple definitions of heritability. So I've given you, like, you can first of all distinguish between this more like vernacular or casual notion of heritability, where Jesus means like, is it inherited in some sense. Is it passed on with some fidelity to the next generation, but as I said, that's not what heritability means when we talk in a scientific journal about heritability. And even here with heritability we have several notions of heritability. The broadest category, the broadest distinction is between narrow sense and broad sense heritability. And so broad science is typically used in the context of development to see whether genes explained. And so basically, genes here include all the effects of genes, including the effects of um interactions between genes. Um uh SO episthesis is included dominance. And so forth and narrow sensoritability is more to predict what would be the, the, the value of the pop the value of the next generation, basically. So it's just the additives. So when, when an organism or organism produced, uh, basically its genes are somehow partially um Uh, become independent of the, the, the, the combination of recombination, but basically, uh, the, the, the combination changes and so we only see, uh, at the next generation, uh, what's called the additive effects. And so that's what neuroscience is, but there is, so that's one, so neuroscience is based on what's called a variance approach. Uh, WHAT I call the approach, but there is another more, uh, general in some sense or more abstract notion of heritability, and so the But basically the variance approach rely on the kind of causal models that you have where you have additive, uh, additive effects, you have dominance, you have episthesis, and so forth. Uh, THE, basically the, the regression approach is we just do a, a correlation between power and. So it's a regression, but it's based on a kind of, uh, a core correlation between the two. Uh, AND in this case you can, you can do between any, any types of objects basically. They're not necessarily, you can do it in the case of cultural evolution, but we use a approach where basically you are just referring to genetics, you can't, uh, you, you can't just Transfers is genetics. Now if by genetics you mean something more abstract than just a gene in the DNA, like something like an evolutionary gene or a replicator or something more abstract, then you could also develop a similar, a similar model. So the regression approach is much more, much more general, but also much more statistical. And correlative and so to go back to your question, so that's just kind of the background. So it's, I would say it's a kind of a mess, to be honest, because people use different, uh, different approach in different, um, in different scientific discipline and if you, if you see this word in this discipline does not necessarily mean the same thing in another discipline. There might be some bridge. Often there are some bridges, but it's, it's, it's not, it's not a 1 to 1 where I can transpose the concept in one discipline to the other. Sometimes you need to do some work to do that often, but when people say like it's 70% genetic, it means like assuming a particular background of variation where the environment can vary. So it's already very limited, and if you're in an experimental setting where the conditions are very stable. Then it doesn't represent the variation that you observe in in nature, but in this case, the ratio basically of the variation on uh genetic variation uh explains 70% uh uh. 70% of the variation that we see in the phenotype, that's what it is basically, uh, so we can't move from this population level description strictly to an individual level description. I mean there are some bridges, of course, uh, people sometimes say that there are no bridge. I don't agree. There are, there is a bridge, but it's a very limited notion, uh, and actually it's a, it's a single number which summarize something. And the population level, so it has not much information about individuals. It has some information, but it's extremely limited, basically, uh, and, uh, yeah, and the jump from the population level description to the individual level discipline cannot be done, uh, cannot be done, uh, straightforwardly.

25:57 Ricardo Lopes: Mhm. And in cases where there's missing heritability, is that heritability necessarily in the DNA?

26:08 Pierrick Bourrat: So the missing heritability problem is precisely a problem where you have different notion of heritability. The kind of basic abstract conceptual tool is the same, but it's not applied in exactly the same in the same. Context and so the missing heritability problem uh occurred when people started to do GWAS, or genome wide association studies which are based on SNP, uh, on uh singular nucleotide polymorphism and so the, so basically a SNP is a kind of mutation, uh, it's a kind of mutation, but, and it could, we could say it corresponds somehow to the notion, yeah, point mutation. Uh, BUT there are so many other notions which abstractly are a kind of mutations that are not recorded by a snip. And so the way heritability is traditionally done is in um is in family studies, so you compare twins or parents of spring and so forth. And so not only uh this kind of mutation are transmitted, but many other things, uh, and some things might be uh uh. Somehow linked to DNA. So for instance, I don't know if you have like a massive change in DNA like an inversion of chromosome or something like that would count to me as somehow genetic too, but not genetic code itself. Uh, AND, and basically the, the missing heritability problem is arise because when we do these family studies. And when we do genome-wide association study and techniques from genome-wide association studies, we have a gap between the two. And this gap is unexplained and it's a missing heritability right and my, I mean, and I'm not the only one to argue that, but my argument is like at least a lot of it could be explained if we were referring to the same biological ship state. And so the missing heritability probably, uh, probably take into account, um. Um, It does not take into account some mutation, some things that we would consider mutation, uh, in, in the. When the, the heritability estimate is made. Well, in the case of um of uh of family studies, basically, what we might have is an overestimation here because there are some correlated factors that are our DNA, the environment, and so forth that, uh, that, uh, even if we try to separate them, they are still connecting. So basically if we work on these two sides, maybe there is a way to, uh, to make them more connected.

28:55 Ricardo Lopes: Mhm. Right. So, um, of course, we're going to open another can of worms here, but, but I have to ask you also about, uh, adaptation because of course people in evolutionary biology, evolutionary psychology, and so on talk about adaptations or even adaptationist approaches to, uh, I mean, to different. OF traits, but what is an adaptation actually? And this, when we talk about that, is there really sort of a consensus on a particular set of criteria that people, uh, biologists or philosophers of biology use to classify a particular phenotype as an adaptation?

29:45 Pierrick Bourrat: Well, um, so, so the adaptation is an interesting concept, and I think, um, adaptation without being clear about what the environment in which an adaptation is, uh, an adaptation to which environment, uh, we are, we are referring to, uh, talking about adaptation in a vacuum like this. It's, it's, I think it's very unhelpful because we need, we need to know what is the environment to which we are referring. So evolutionary psychologists, I think, had a very good, very good move to being clear in terms of adaptation that they are referring to a particular, a particular environment of reference, that's what I would call an environment of reference, which is the evolution, the environment of evolutionary adaptiveness. And I think as a framework it's a very healthy thing to do because at least if you say it's an adaptation, you're going to say according to which environment I'm referring and of course people have started to criticize them because sometimes very unfairly I would say, because, uh, because basically, well, there is not a single, we can't consider that there is a single environment for a single. Uh, FOR, for the species, there were variations and so forth. I completely agree. Uh, BUT at the same time, we need, we need a way to try to see whether there are some, uh, even if it's just statistical or on average, whether we have some average difference, um, whether that works. So I would say the first step to me, if someone talks to me about adaptation, I want to know what against which environment they are referring. And the notion of time scales become important too, because environment over time changes. And so if the environment changes, changes, and we are not clear about the time scale at which we are referring, then, uh, then we, I, I think it, it doesn't make much sense. Other than we can say, well, look, there is a structure here. And we can see that it's advantageous in all possible environments that we can come up with or that we are biologically irrelevant in the life history. Then the notion of adaptation, if it's not really properly contextualized, I think it's not going to be very, very helpful. Now when scientists in a particular field talk about adaptation, typically they don't make the environment explicit, but it's implicit. Uh, BUT then, then we go back again to this transfer of, uh, an expression in one field to another field where the hypo, the, the assumptions are different and so forth, and it can become very, and I think people might even talk very easily past each other because of, because of that, uh, but I would say like an adaptation. So if we leave this to the side, like an adaptation is somehow, so we can talk about it, the process of adaptation, so to become more adapted or better adapted to the environment by natural selection. Uh, SO that's kind of a process which immediately leads to a kind of product. So at the end of a round of selection, you will have individuals that are better adapted, which will produce, who, which will produce more, uh, more offspring, which will be more evolutionarily successful. And then there is this notion like that evolutionary, uh, well, let's say neo-Palean, like, the, the, the tradition of Dawkins, behavioral ecologists, and so forth, um, uh, sociobiologists, um. And, uh, evolutionary psychologists talk about, which are basically, uh, especially in the mind, of device, devices, uh, or that are, uh, structures that are the, the result of multiple cumulation, cumulative events of, uh, of selection, uh, which do not need to be, uh, fixed. It can be a very plas um device that leads to very plastic, uh, plastic change. Uh, AND so an adaptation can also be mismatched. So if the environment is, uh, is currently very different or different in some relevant sense, it can lead to be, uh, an adaptation to be, uh, maladaptive and to lead to, uh, a decrease that would still count as an adaptation, uh, according to this kind of historical notion of adaptation, which is the accumulative product of, uh, of natural selection.

34:23 Ricardo Lopes: Yeah, but, yeah, but I mean, when, when it comes to, to the notion of the environment of evolutionary adaptiveness, I mean, uh, uh, just, just to mention, I'm not using this as an argument from authority nor anything like that, but back in 2019, I interviewed the late Robert Rivers and I actually asked him about that and he, he was very creative. Critical of that notion also because he thought that there wasn't one single environment or even the idea put forward by people like Lida Cosmith is that it's not really one single homogeneous environment, but it's like a statistical composite of the environments we evolved in during the Pleistocene. I mean he also disagreed with that and I, I, I mean I, I, I think that there's also um another set of problems in terms of how evolutionary psychologists make certain assumptions about uh Not just the conditions we evolved in, but also what might have changed for since the advent of the Oloin because, because, because I mean, I think that the, uh, I'm not saying that all of them, but most evolutionary psychologists, um. Have, um, uh, an idea that for the past, I don't know, 10 to 14,000 years, human psychology must have not changed that much and they don't take it seriously enough things like uh gene. Cultural co-evolution and cultural evolutionary theory and the work of people like, for example, Joseph Fenwick and Robert Boyd and Peter Richards and then, and others like that. Uh AND um I, I mean, I, I just think that sometimes, and, and when it comes to the conditions we evolved in, another criticism of mine that, uh, I mean, I've been, I've become. I've been emphasizing that more and more is that uh I mean, very unfortunately I don't think that evolutionary psychologists are very well versed or sufficiently versed on anthropology to really make proper claims about what might have even been the statistical composite of An environment of evolutionary adaptiveness to then make those broad claims about what is what is supposedly universal in human psychology, whether it is particular traits or whether they're referring to information processing mechanisms. I mean, because sometimes it's also not. Completely clear what they're referring to there. So I, I, I, I mean, what do you think about those,

37:24 Pierrick Bourrat: uh, I mean, it's, I'm, I, I'm, I'm very happy to, to hear that. Um, WHAT, what I wanted to say is that something that did good was to, um, I mean, and they who are, who are evolutionary psychologists in 2020. 6 versus 92 when the when the adapted mind came, it's very, it's, it's very different. I mean there is a lineage of of of people, but basically what I think they did right is to be clear about uh that you can't talk about an adaptation without an environment which if you talk casually to evolutionary biologists sometimes they say oh this is an adaptation, this is not. And to me I'm like I don't have enough information to understand what do you mean by that. And so that's, that's something that I think, and the evolution, the environment of evolutionary adapted it, it's clearly a very crude tool. Like there is no question about that. Yes, statistical, I mean, if, if you, if you take a statistical environment, for instance, and that it goes in one direction in this part of the environment and the opposite direction in this other part of the environment, you do the sum and you have no difference. So basically there is no effect. So all this notion of average effect, I mean we do need average. I mean we do statistics, so we do need average, but we need to know average again what are the relevant average we need to do, and sometimes we need to be much more fine grained to disentangle different hypotheses. So I would say like. The general project as a starting point is, in my view, it's a, it's a, it's a, it's a good starting point. And then probably yeah, people, some people are more radical, some people do not understand statistics, some people uh do not unders do not understand evolution or that they don't have the same background. If you do, if you do evolutionary psychology, for instance, there is a likelihood you come from psychology purely. Where you would not have the same training as, as a biologist in evolutionary biology coming to evolutionary psychology. So there are all these, and people will have more or less my view of evolution. People will be more or less adaptationists and so forth. Um, YEAH, so, yeah, no, I agree, I agree that you can't just look at things from purely the classical evolutionary psychology, at least as it was, uh, as it was defined. With this, with this 1892, on the other hand, I think like we, we should be able to, uh, to, um, to talk about adaptation as the, as of human behavior. I don't think I find it very complicated to, to describe, but I'm open to the possibility to, to, to talk about it.

40:10 Ricardo Lopes: Yeah, I, I just think that very honestly, um, I, I, I'm not, this is not a broad criticism. I mean, I know that there are evolutionary psychologists out there that write. Their best to do that, to familiarize themselves, for example, more with evolutionary biology, with genetics, with anthropology, or to collaborate with people that are experts in those fields, but I think that honestly sometimes certain people make very broad claims and and then I go and talk with the evolutionary biologists, with the geneticists, with the anthropologists, and I mean because. They're making claims that should involve evidence from those fields and then I go talk with those experts and they just tell me that that's just not true.

41:02 Pierrick Bourrat: So

41:04 Ricardo Lopes: and then it can be a little bit embarrassing. I mean, uh, just

41:07 Pierrick Bourrat: saying, yeah, yeah, of course, of course, yeah, you know, I mean it's, it's, it's first of all, it's an interdisciplinary, uh, discipline. So yeah, it's, uh, it's gonna be, uh, very, and also like, I mean, I don't know like the bold. Claim, I'm also, what I don't, what sometimes I don't like is we don't apply always the same standard to to evolutionary psychology to other disciplines. So I'm a bit on the side where, well, it has been beaten up for so many years that we don't need to, to be, yeah, of course we need to correct what is incorrect, but there are some positive aspects in it which I, I'm trying to, uh, yeah, to look, to look at. And personally, personally I've always been interested in in evolutionary. Psychology and I believe I have a kind of sophisticated understanding of, of, um, of evolutionary theory and so I don't see why, why we couldn't do it or why it's dangerous or all these things. Uh, I don't believe so. I think we should be able to, to do it. We should, we should strive to do it well, basically, and, uh, and we should criticize it when it has to be criticized, but it shouldn't be like, oh no, we can't, we can't, uh, we can't do this type of, uh, of, um. Of research is dangerous. I don't, I don't think, I think it's dangerous to not, to not do it because someone else will, someone will do it and they will do it very badly. So we need like, we need to be able to, um, to do it, uh, well and to be open to criticism, to be, uh, to be open to, um, yeah, to be, to be able to interact. That's.

42:44 Ricardo Lopes: Yeah, sure. So let's talk a little bit about levels of selection and multi-level selection. So at what levels can evolution occur?

42:59 Pierrick Bourrat: Well, again, uh, that's a complicated. It depends what you mean by, uh, units of selection and so forth, but, but basically, so the idea is that, um, So if we take this kind of interactive view as a kind of primary object that, that can participate, so it's an abstract object at the moment, like in the theory that can participate in a, in a selection process, then any object that is an intertractor can participate in a selection process and so you can have selection at any level. Now, The, the, there are multiple difficulties when I, even when I just say that. The first one is that, can I be an intractor if I'm an inter and can components of me also be in tractors? OK. And so if I ask this question, it's not, it's not clear that this is necessarily the case. I would say that if you refer to the same level, so sorry, if you refer to the same uh time horizon at which you are looking at an evolutionary process, OK, and if you refer somehow to the same environment, I think if you are a full interactor understood as a cohesive entity that interacts directly with its Environment, then there is no space for any other objects in the hierarchy because they are composed, they are composed, they are my components. There is no space for them to occupy this role, basically this role of interactor. OK, so, but now it could be the case that if I change the environment on which I'm looking at or the time scale, time horizon, and so forth, well, I could be an intertractor for this time scale at one particular level and intertractor at another. Over a different time scale at another level. So if I make the broad claim, well, can there be multiple levels of selection? Yes, but can we compete? Can we compare them easily and so forth? I would say no. Mhm. So. Yeah, yeah,

45:19 Ricardo Lopes: and so what would be a multi-level selection and what do you make of it?

45:26 Pierrick Bourrat: So for me, multi-level selection, so the idea of multi-level selection has often been used in the context of the price equation, for instance, if you have heard about the price equation, so it's a, it's a very abstract description of evolutionary change in terms of within. And between level selection. OK, I think the within between doesn't map very neatly. I mean, people have often meant that by natural, by multi-level selection, the within and the between, uh, and they have been talking about it. In particular, in the context of altruism and selfishness, where if you have too many selfish, for instance, in a, in a, in a, in a collective, then, uh then uh the, the altruists can be exploited and that's the rial problem and basically the, the, the lower, the lower of the with the, the collective would, uh, would be uh subverted from within. Um, AND that's in that sense that it has been used and of course like we can show that if you have any population structure, that's not necessarily the case, so you can have the evolution of altruism in that way so you can give a part a price equation partitioning of that, um, and so in, in. If that's what people mean by multiation, I, I'm fine with it. I mean, it's, it's, it's OK, but I don't think it's corresponds to the notion of a hierarchy within and within and between is not the same as top and down, OK, uh, and, and so it's OK to have this description, uh, within between, but I think that's not hierarchical, uh, what I would call hierarchical, uh, evolutionary theory where it's really like, can the, the components. Uh, BE somehow, um, have a different evolutionary interest, and when we start to say it in that way, we can see, not really, because if I, if my component does not have an exact, do not have the same evolutionary interest as me. Then it means that uh there also are gonna in the long term and you have this kind of, it's not very made clear sometimes with evolutionary biology. I think the two notions at least, and I think other notions that we have discussed earlier are going on when we talk about multi-level selection. Is it selection that there is a problem of reproduction? Is it reproduction also at the higher level, uh, and so forth so. It's a multifaceted, uh, multifaceted, uh, faceted. Um, DEBATE which, uh, can be taken from different perspectives and often people are not very clear about which exact exactly perspective they are, they are, they are taking. There's a further, even I could say a further distinction which is that some people have been claiming that, um, basically if you use a kin selection, a lower level description versus a multi-level description, it's all equivalent, OK. Yeah, and yeah, in some sense it's equivalent because I can describe an outcome from different statistical partitioning. So once you try to impose causation onto this, it becomes much more difficult to say that it's, uh, it's the same because the causal story is kind of different. So and if you run dynamical models, uh, Uh, it's, it's even, so you, you don't have necessarily the same dynamics, and so you can have the same outcome with very different dynamics, which means that they would be equivalent in terms of outcome, which in some sense is not so interesting. I mean, it can be interesting sometimes to see whether we can get to the same outcome from a different perspective, but in this more restrained, restrictive. Of causation or or dynamic model it's not uh it's not uh equivalent and most of the debate I would say about equivalence has been about this uh outcome side uh perspective which I mean it's fine but it's not the same it's not in my, in my view it's not the same as I mean it's like a real equivalence between the two.

49:30 Ricardo Lopes: But talking about causation, I mean, can natural selection refer to population level causes or simply to individual level causes?

49:42 Pierrick Bourrat: OK, so it's an, uh, that's the notion, is the notion of, so that that two notion of population that are going on when we can talk about causation in the context of naturalization. There is a natural population. And there is a when you talk about causation typically a very, very, uh, very good, I mean a very interesting framework from the scientific perspective is the interventionist account of causation where you need like a probability distribution of events, uh, and then you intervene on a variable, so you need a variable, OK, with, uh, with a distribution. Of events or state what is viable. And so you have a population here also of events so and so typically to To see whether we have a correlation between two variables, we, uh, change one variable point its value to another. So you can, you can make a difference in one variable, does it lead to a difference in another variable. So intrinsically we could say the notion of, uh, causation. Involve a form of population events. So if, if we mean that by causation, yeah, of course, uh, everything is at the population level because, uh, because we, we need, we need differences which involve, which is more than one individual, uh, uh, um, to talk about causation, but I don't think. We mean although maybe some people that when they're talking about population that this individual organization, but the, the question is whether Are they, um, where, where do, where should we locate causation here in terms of the object? Is it at the population level, and we can't really, uh, population, natural population, so the population of organism or at the organism level. I think that fitness is, is, is an individual level property. Uh, uh, SO we, we could say this individual ethic now to, to say that it has a fitness is very different from saying that we can characterize it very easily, OK, but I think that in the logic of natural selection there is this idea that individuals have fitness. And, and that the difference the difference in fitness in the population which might be due to many different the way they interact with their environment, their traits, and so forth, how they think specific, um, uh, interact with them and so forth. But if you say that, then it seems to me that natural selection is, uh, a core, um, relies on individual level possession. Of course the outcome of evolution is the evolution of the population. So we can have an individual cause that leads to a population, uh, outcome. That's how we characterize that translation. And of course we can also describe this as the population level, but if we give the same, uh, if we first of all, if we give an average of the population level, we are gonna lose some of the information from the lower level and also it doesn't contradict. Like both individual and population. I'm giving you two descriptions, uh, but I think at the core of the logic, it's really, uh, it's really the individual level that, that works the best.

53:04 Ricardo Lopes: Yeah, but is there or could there be conflict between different levels of selection?

53:12 Pierrick Bourrat: OK, so the question is whether there are conflict. There can be conflict between levels of selection. And so I think the answer here again depends on what one means by, uh by conflict, uh, and I think that in the context of, if we, if by conflict, we just mean uh metaphorically, can there be conflict between level of selection? I think, uh, I think it's kind of OK, but I also think that conflict, the notion of conflict can be very misleading if we make it in a more Like in a more real sense rather than just a, a, a, a metaphor. And let me explain why, for instance, we could, we could say, well, is there a conflict between uh uh someone, the, the cancerous cell of someone having a cancer and the whole individual itself, and I think, um, I think the conflict is, is not really there simply because that's argument has been put forward by, by some people. I'm not the only one to say that, but It doesn't make sense if basically We compare fitness over the same time scale. So I'm going back to this notion of time scale, but the fitness of of cancerous cells, unless they can escape, in some cases they do, but in which case we don't have now anymore a constitutive relationship between the cancerous cell and the non-cancerous of the whole organism. OK, uh, but if we have this constitutive relationship, then, uh, if you look at the, after one generation, well, there is no cancer, uh, there is no cancerous, uh, they have a fitness of zero. I can give you an explanation in terms of evolutionary dynamic within a single body, uh, but it would be a mistake. And we could say, well, what if the body, well, what if they can proliferate indefinitely without any imposing any damage to their environment, basically and to the body in which they live, then in that case we could have, we could have potentially a conflict. Uh, WHERE the evolutionary interests are diverging because there is no constraint whatsoever. So this is an argument where we can make it, but it's a very limited argument, and in the real world, it will not work. Uh, BUT now we can talk about conflict in a more metaphoric sense, metaphorical sense, by saying, well, if there was not this constraint, then we expect like a cancer cell to proliferate indefinitely. Uh, AND that would be in conflict with the interest of the, uh, organism, and in that, in that sense, uh, that's fine, but another way to talk about it is to talk about conflict. So that's what I'm arguing for basically. It's when we talk about conflict, it's always at a single level. So, uh, uh, uh, a whole organism. IS basically a complex nexus of cells, but we could, we could talk about the cancerous lineage versus versus the rest of the lineage, and in that case, the conflict kind of makes sense, much better than between levels.

56:29 Ricardo Lopes: Mhm. OK, so just before we get into the topic of uh whether we need an extended evolutionary synthesis and what that would correspond to, uh, I would just like to get very briefly into a little bit of genetics and epigenetics here because it's, uh, epigenetics is something that Proponents of the extended evolutionary synthesis usually would like to include in it. So, um, how should we conceptualize genes? I mean, is epigenetics something that we should think of as separate from genes themselves or not?

57:10 Pierrick Bourrat: Um, SO I, my PhD advisor is Paul Griffiths, was Paul Griffiths, who is a philosopher of biology, and he has worked for many years on the different concepts of the gene, basically. So that's what he's known for, among other things. But so, and I, I took, uh, I, I took lessons from him, of course. And so basically the term of the gene, the notion of gene is very um It is very complicated. Nobody really knows what a gene is. If you ask a scientist, they will tell you something. If you ask another, they will tell you some other things. Typically, and they can switch also between 11 notion to the other without maybe even realizing sometimes. So depending on again, the the the context matters a lot. So in the context of epigenetics, I think like, of course it makes sense to talk epigenetic in the context of um. Of molecular genes, because if there is any other material biological substrate somehow that can be transmitted from one generation to the other, yeah, that's, that's, that's epigenetic, uh, and, and we know that in some cases actually, there is no germsomal separation. So, yeah, anything that is, uh, for any, uh, any species in which there is no germsomal separation. Then potentially, uh, potentially much more than just DNA can be, can be transmitted. Now, when we ask, when we say that, does it mean that, um, uh, yeah, um. If we, if we now move to the evolutionary level, to the evolutionary theory level, a gene, I mean, a theory is, of course, connected to data, but also it's a part of the theory is also highly abstract uh uh apparatus, which we can apply to different, uh, to different, uh, to different objects. And so a gene in that case, it's a bit like the replicator of Dawkins, a very clear, very clear replicator is a gene, and so sometimes we want to, the distinction between the gene and the replicator, it's kind of the same thing, but the point is that no, we need to separate the two. And if we separate the two and we call it an evolutionary gene, this notion which is much more abstract and not necessarily, not necessarily made of DNA, so we could talk about, I mean, the meme was, was an attempt to talk about a form of replicator in the evolutionary gene, if you want to say in the context of cultural evolution, then a lot of claim about epigenetics lose its kind of grip because Because that cannot go against evolutionary theory or even like the way we represent evolutionary theory. Now if by evolutionary theory you mean like this kind of form of abstract description or abstract model, OK, with all the empirical facts connected with one another and so forth, and that's all this you call the theory, and the theory has been, has been predominantly predominantly applied in a tiny. Tiny corner of biology where there is germsomal separation, then you might start to be uh to be wanting to call it epigenetic uh to um to somehow emphasize that it's not just DNA. Right,

1:00:49 Ricardo Lopes: right. OK, so I mean, um, we've talked about epigenetics, we've talked about multilevel selection and other things. I mean, there are people out there, I mean people who follow, of course, the literature on evolutionary biology and particularly philosophy of biology are already well aware of this, but there There are people out there who propose an extended evolutionary synthesis. I mean something that goes beyond the modern evolutionary synthesis from almost a century ago, and they talk about including mechanisms like, uh, epigenetics. They talk about niche construction. They talk about notions from. Developmental biology, they talk about, uh, culture and cultural evolution. They even talk about sometimes, uh, symbolism and symbiotic. So, I mean, do you think that, uh, The the fact that now we know about those mechanisms that play a role in evolution, it warrants an extended evolutionary synthesis, or how should we deal with all that new knowledge that we have about those uh non-genetic ways of inheritance.

1:02:18 Pierrick Bourrat: Yeah, so actually I've collaborated a bit with Iva Jablonka, who is a proponent, and I also gave a review of, so she has a small element on on the extended evolutionary synthesis. So and my, my view on this is that, I mean, I'm happy with the facts. I mean, so long as, yeah, the facts are replicated, that we, we cannot take them as facts, I'm very happy. Similarly with niche construct. I'm happy with, with, with, with niche construction. I mean, I'm a bit, I have the concept as a kind of teleological sense construction. I would prefer the notion of modification, but basically if it's I, I can make sense of it and, and I, and I'm very happy with. Now, the question is whether it goes against, uh, against like traditional notion of uh evolutionary theory, uh, the modern synthesis, and I, so. From coming from a generation where there was no real opposition against those facts, to me, I think like it's somehow. It seems to me that it's a sociological fact where there are these people who were dissatisfied with a particular kind of rigid structure that was very hard to to basically move in any way. And I think like probably some of those people were not really well treated. They were treated like heretics almost. And so, it would explain why, why they really are against, and I think it's really a generational thing where now younger people, younger generation, are not so opposed to, to, uh, I mean, what is the modern synthesis. Yeah, nobody, nobody is a mosynthesis scientist now. We all have integrated those facts and so forth, so I think it's, it's good to remind like what was the historical. Framework in which people were working. Why like those people coming and proposing new things or maybe, oh no, we, we, but I must also say that sometimes there was, there is certainly also uh with the notion of scales, with the notion of context not being very clear and so forth where people probably talk past each other too. So I have um I have a personal story, for instance, where Uh, where I, I, I discussed with a, with a, with a prominent, um, niche construction person where, um, basically, uh, this person was saying, I'm not going to name names, but this person was just, um, was just telling me that, um, well, natural selection is only, uh, is only the, basically only at the causal arrow from, uh, from the environment to the organism or to the object. Well, I think that's not how many evolutionary biologists talk about natural selection. That is the effect of the, of the organism on the environment are also part of, I mean, the extended phenotype would be one example, but they're also part of purely natural selection. But if you have this view of natural selection that is different from, uh, or if you have people with different views on what entails, what is entailed by natural selection, you can easily. Talk past each other and so I was talking with this person. I was like, I don't understand why that, and then I realized that we had simply different definition of natural selection. And if you take this perspective, then you can see, oh yeah, OK, I see what you mean. If that's what you mean by natural selection, that's not what I mean by natural selection, and so forth. But I think a lot of that, uh, a lot of that has has happened with people also having egos and so forth. You could see how it's probably a recipe for disaster when you, when, when, when they try to communicate. So I think there is both this probably a rigid, rigid frame that has not been changed and so forth. And so we have integrated a lot of facts, and probably we haven't sometimes even recognized in the process of integration that's, that's quite different from it was initially. And there is this, this also talking at different levels of abstraction, talking past each other, which must have opened. I personally think that just in terms of term extended evolutionary synthesis, I don't think there is, there is really a synthesis here. It's a collection of facts that do not fit with the with the traditional model, but itself I wouldn't say it's a synthesis. I would say. An extended model maybe or a richer model, uh, but again, it's like, it's just in terms of the words, but synthesis means like there is a body of, uh, yeah, um, and, and, and it will be interesting, but I'm not even sure that all the people who are part of this synthesis agree. On every aspect, for instance, or even like, or even that there is really a core. I'm not sure, and the core cannot be we disagree with the modern synthesis view. It needs to be something more than that, I would say.

1:07:50 Ricardo Lopes: Yeah, but, but I mean, is the idea to go against the modern evolutionary synthesis, or is it, as the name suggests, to extend the evolutionary synthesis? I mean, because of course I know that for example, there are some people who study epigenetics that sometimes at least. ALLUDE to the idea that maybe evolution is not entirely Darwinian, that it can have maybe some Lamarckian aspects to it, even, but, but, but I don't see that at least most proponents of the extended evolutionary synthesis are wanting to go against the modern evolutionary synthesis, but simply to expand it.

1:08:38 Pierrick Bourrat: Yeah, no, no, I think there is definitely the same. I think there are probably some, some hypotheses within the, depending on how you define the modern synthesis, we could say, well, I disagree with this aspect. I don't think anyone in the extended evolutionary synthesis is, it's probably more. A question of emphasize, emphasis. They want to emphasize phenomenons that do not fit well with, uh, with the kind of more rigid frame, and I mean, I'm fine. I'm fine with that, but it's true that, I mean, and also like there is definitely in the scientific, uh, in, in academy, in academia, there's always like sometimes you need to make your, your claim bold and radical so that it has any impact. And certainly, uh, certainly like calling it an extended evolutionary sense, this is probably much better than just say, well, I just want to extend the, if you say that, maybe people will just like, yeah, sure, uh, uh, not realizing that really, you are doing something quite maybe radically different and so forth. So, yeah, uh, it's, um, it's uh. Yeah, but I don't think anyone wants to say, I mean, maybe there are some, one says that natural selection, uh, has nothing to do or that's, uh, genes have nothing to do with, uh, yeah, of course, um, with, with, uh, with evolution, yeah.

1:10:06 Ricardo Lopes: Mhm. OK, great. So, uh, before we go, where can people find your work on the internet?

1:10:13 Pierrick Bourrat: Uh, JUST Pierre Bora.com, uh, and you will have all my, all my, um, all my publications, uh, yeah, so everything, everything is there, basically.

1:10:26 Ricardo Lopes: OK, I will be leaving a link to it in the description below. So thank you so much for taking the time to come on the show. It's been a fascinating conversation.

1:10:35 Pierrick Bourrat: Thank you. Thanks, thanks to you.

1:10:38 Ricardo Lopes: Hi guys, thank you for watching this interview until the end. If you liked it, please share it, leave a like and hit the subscription button. The show is brought to you by Enlights Learning and Development done differently. Check their website at enlights.com and also please consider supporting the show on Patreon or PayPal. I would also like to give a huge thank you to my main patrons and PayPal supporters, Perergo Larsson, Jerry Muller, Frederick Sundo, Bernard Seyaz Olaf, Alex, Adam Cassel, Matthew Whittingbird, Arnaud Wolff, Tim Hollis, Eric Elena, John Connors, Philip Forrest Connolly. Then Dmitri Robert Windegeru Inasi Zu Mark Nevs, Colin Holbrookfield, Governor, Michel Stormir, Samuel Andrea, Francis Forti Agnun, Svergoo, and Hal Herzognon, Michel Jonathan Labrarinth, John Yardston, and Samuel Curric Hines, Mark Smith, John Ware, Tom Hammel, Sardusran, David Sloan Wilson, Yasilla Dezaraujo Romain Roach, Diego Londono Correa. Yannik Punteran Ruzmani, Charlotte Blis Nicole Barbaro, Adam Hunt, Pavlostazevski, Alekbaka, Madison, Gary G. Alman, Semov, Zal Adrian Yei Poltontin, John Barboza, Julian Price, Edward Hall, Edin Bronner, Douglas Fry, Franco Bartolatti, Gabriel P Scortez or Suliliski, Scott Zachary Fish, Tim Duffy, Sony Smith, and Wisman. Daniel Friedman, William Buckner, Paul Georg Jarno, Luke Lovai, Georgios Theophanus, Chris Williamson, Peter Wolozin, David Williams, Di Acosta, Anton Ericsson, Charles Murray, Alex Shaw, Marie Martinez, Coralli Chevalier, Bangalore atheists, Larry D. Lee Junior. Old Eringbon. Esterri, Michael Bailey, then Spurber, Robert Grassy, Zigoren, Jeff McMahon, Jake Zul, Barnabas Raddix, Mark Kempel, Thomas Dovner, Luke Neeson, Chris Story, Kimberly Johnson, Benjamin Galbert, Jessica Nowicki, Linda Brendan, Nicholas Carlson, Ismael Bensleyman. George Ekoriati, Valentine Steinmann, Per Crawley, Kate Van Goler, Alexander Obert, Liam Dunaway, BR, Massoud Ali Mohammadi, Perpendicular, Jannes Hetner, Ursula Guinov, Gregory Hastings, David Pinsov, Sean Nelson, Mike Levin, and Jos Necht. A special thanks to my producers Iar Webb, Jim Frank Lucas Stinnik, Tom Vanneden, Bernardine Curtis Dixon, Benedict Mueller, Thomas Trumbull, Catherine and Patrick Tobin, John Carlo Montenegro, Al Nick Cortiz, and Nick Golden, and to my executive producers, Matthew Lavender, Sergio Quadrian, Bogdan Kanis, and Rosie. Thank you for all.