Making Sense with Sam Harris - #174 — Life & Mind

Episode Date: November 4, 2019

Sam Harris speaks with Richard Dawkins. They discuss the strangeness of the “gene’s-eye view" of the world, the limits of Darwinian thinking when applied to human life, the concept of the extended phenotype, ideologies as meme complexes, whether consciousness might be an epiphenomenon, psychedelics, meditation, and other topics. If the Making Sense podcast logo in your player is BLACK, you can SUBSCRIBE to gain access to all full-length episodes at samharris.org/subscribe.

Welcome to the Making Sense Podcast. This is Sam Harris. Okay, do I have housekeeping today? I don't think so. Today I'm speaking with Richard Dawkins. Richard really needs no introduction on this podcast, but please note that he has a new book out titled Outgrowing God, A Beginner's Guide. In this conversation, we mostly take your questions, and we start by discussing the strangeness of the gene's eye view of the world. We then move on to the limits of Darwinian thinking when applied to human life. We talk about his concept of the extended phenotype and memetics. We look at how ideologies act as meme complexes. We talk about whether consciousness might be an epiphenomenon and therefore might not have been evolved under selective pressure.

And then we talk about psychedelics and meditation. I actually lead Richard in a guided meditation, and the effects of that you can hear for yourself. And I'll have something more to say in my afterward. So now, without further delay, I bring you Richard Dawkins. Be you still, be you still, trembling heart, Remember the wisdom out of the old days. Him who trembles before the flame and the flood and the winds that blow through the starry ways,

let the starry winds and the flame and the flood cover over and hide, for he has no part with the lonely majestical multitude. What poem is that? It's an early one. It's from The Wind Among the Reeds, I think. Well, that was a wonderful reading and and the perfect soundcheck I am here with Richard Dawkins Richard thanks for joining me again on the podcast thank you very much and thank you for coming to the Biltmore Hotel yeah making me go

to your studio this is this is old school I I love it so you know you and I have done a bunch of events together. Yes. I hope we haven't run out of things to talk about. I worry about that kind of thing. Yeah. Yeah. You know, so in the interest of, of not running into that problem, I decided to go out on social media and ask for questions. And this is, you know, this is, that's the perfect algorithm because I can, I know what kinds of questions we've hit in the past. And, uh, this simultaneously gets us what our respective audiences want to hear. And, and I, I have no fear that we're going to cover the same territory in the same way again. I hope one or two of them may have seen my new book. Maybe not, maybe too recently out.

I don't know. So let's just mention the new books or so that we, uh, we've done that. The new book is Outgrowing God. Outgrowing God, yes. And this is for teenagers, right? Yes. It's sort of, quite a lot of complaints have been that it's just like The God Delusion. It actually isn't just like The God Delusion. It's different. And it's sort of designed for teenagers, yes. And we can obviously spend as much time or as little time on these questions as we want and open any doors that they suggest to us. But the first frivolous question is, and this surprises me, this means nothing, but do you

realize that the most prominent atheists are all Aries, right? So you're an Aries, I'm an Aries, Hitch was an Aries, Dennett is an Aries, Matt Dillahunty is an aries i'm an aries hitch was an aries dennett is an aries matt dillahunty is an aries the great uh film director otto preminger was once approached by a starlet on on the set of one of his films she said oh gee mr preminger what sign are you and he said i am a do not disturb sign that's my attitude towards astrology right, I guess Aries don't believe in astrology. So the first question, which I think will set us on a nice path, it won't preempt everything else, but this is somebody who clearly is exasperated with the prospect that we might focus exclusively on atheism or bashing religion. And he says, for goodness sake, get him talking in

detail about the gene's eye view of natural selection, the extended phenotype, the argument surrounding group selection and punctuated equilibrium, the way memetics has rather ironically taken on a life of its own, and so on. Not just God, for for God's sake so this covers a lot of ground and I do want to do those topics justice so let's I think people are so casually aware of the revolution that has been wrought in our thinking based on on our understanding of Darwinism. And it was really crystallized in your book, The Selfish Gene. But it has kind of receded into the background of our thinking.

And it is such a strange view of the mechanics of things and the logic of things. And so maybe let's just spend a little time talking about the nature of replication. I mean, I like to think if it has receded into the background, that's because it's simply accepted which among professional biologists of the sort of field type it has i mean yeah i'm thinking the general public that has kind of lost sight of how strange it is perhaps that the general public it's not well i suppose it is a bit strange and

it sort of is a turning on its head of what used to be the more orthodox view. Darwin saw natural selection at the level of the individual. So he thought of individuals as competing with each other within the species. He was always a within-species competition. And he was, of course, aware that survival is only a means to the end of reproduction. And his other great book, well, one of his other great books, The Descent of Man, is largely about sexual selections. Darwin was thoroughly aware that success at reproduction was also vitally important and any hereditary tendency to be, for example, sexually attractive or good at competing with members of their own sex would also be favored.

But Darwin didn't have gene language. He had no concept of the particulate gene which Mendel introduced. And that particulate view of genetics was actually essential to natural selection because, as was pointed out in Darwin's own time, if genetics was blending as everybody in the 19th century except Mendel thought, if we were all a kind of mixture of our father and our mother, then variation would disappear as the generations went by. Each generation would be more uniform than the previous one, in which case there would be no variation left for natural selection to work on. This was actually advanced as an argument against natural selection. Actually,

of course, it's an argument against manifest facts because we don't get more alike as the generations go by. Mendel solved that problem, but Darwin didn't realize it. I don't think Mendel realized it properly. And it wasn't until the neo-Darwinian synthesis of the 1930s that it was realized that actually a natural selection is all about genes changing their frequency. So some genes become more frequent in the population, others less frequent in the population.

That's what it's all about. So some genes become more frequent in the population, others less frequent in the population. That's what it's all about. I suppose all that I did really was to take that neo-Darwinian view and put it in a more slightly more poetic way and say that that means that the individual is just a vehicle for carrying genes around and passing them on. And it's temporary. I called it a throwaway survival machine. Right. That's the strange idea, what you call it strange.

And that is a bit strange, I suppose, that I call them a robot survival machine. An individual organism is a device for passing on genes. The selfish gene is quite largely about not selfishness. It's often misunderstood because of the title as being about selfishness or even an advocacy of selfishness. It's actually mostly about altruism. The selfish gene explains altruism at the individual level. Right. But if you take a gene's eye view of human life, many strange things happen.

First, you see that there's a logic by which certain genes would have been selected for and the behaviors they would encode would be grandfathered into the human condition. And yet, evolution can't see most of what we care about. The logic of evolution is anything that has allowed these specific replicators to perpetuate themselves has been selected for, right? So we are here to spawn and to ensure that our progeny successfully spawn. And I don't know, I mean, at what age do you think, evolution ceases to care about us? I guess grandparents are still valuable.

Oh, there's no sudden cutoff. It's a gradual process. But the older an animal is, the more likely it is already to have reproduced. Right. And so we're all descended from ancestors, most of whom reproduced when they were relatively young. A few may have been reproduced when they were old. And this, of course, is why we age, because we're descended from young ancestors. And very often, whatever it took to be successful when you were young made you actually more likely to die. And this is especially true, of course, of sexual selection,

where brilliantly colored male birds, say, are more likely to propagate genes for being brilliantly colored, but then dying because brilliant colors attract predators just as much as they attract females. And that's an extreme case, but that's the sort of model for the Darwinian theory of aging. Right, right. But there's still something about the extended family that would have been selected for it.

I mean, you would think grandparents are good for something. Oh, yes, oh, yes. In helping ensure that. That's right. And in those species where grandparents can, well, there may be a kind of changeover point where when you get to a certain age, you can do your genes more good by caring for grandchildren than you can by having more children. That, again, wouldn't be a sudden cutoff point, but that probably is true of humans

and a number of other species, perhaps. probably is true of humans and a number of other species perhaps. But if we're talking about running viable governments and societies into democracies, capitalism, pursuing scientific interests, building technology that doesn't destroy us, these are things that obviously are parasitic on cognitive traits that have been evolved but evolution can't really see these details. No that's right I mean I think that so much of our human life is has gone beyond natural selection. Natural selection put us in the world and the way that we are and our brains and our bodies are designed by natural selection to survive under

wild conditions in Africa. And we've now moved beyond that. And so what we think of as successful in our society has really sort of pretty much left natural selection behind. Not to put too fine a point on it, but this is an observation that several of us have made in various contexts. If you were going to take a rigorously gene's eye view of the human circumstance, certainly as a man, the thing you would want to do most, the thing that you would find most fulfilling in life, the thing to which you would purpose more or less every day is to donate your sperm to a sperm bank so that you could have tens of thousands of children

for whom you have no financial or resource responsibility. The fact that sperm donors are actually paid is thoroughly Andarwinian and is a wonderful example of how far we have actually advanced. It's not that surprising because natural selection cannot build into our brains a kind of cognitive awareness of what our genes, so to speak, would want. All it can do is build in rules of thumb, which would work under natural conditions. And so a desire for sex makes perfect sense because that, for the whole of

history, the whole of, I mean, evolutionary history, has tended to lead to reproduction. But a desire to donate your sperm is something quite different. It's not... Could not foresee that technology. Natural selection can't see that. And there have been a few notorious cases of doctors who have been substituting their own sperm for donors and things like that. But it is, to a naive Darwinian, it is a surprising fact that sperm donors have to be paid. A naive Darwinian would think that they would pay to donate their sperm.

Right, yeah, and pay quite a lot. Yes. So let's just talk about what genes are for a moment so genes are a kind of memory they're they're a kind of encoding of knowledge in the sense that and you know stop me if you think at any point this these analogies break down but i'm hearing echoes of david Deutsch here, where it's knowledge as a kind of solution to a problem. It's a genetically inscribed solution to problems that our ancestors have successfully faced. Exactly. Yeah, exactly. I mean, I have a chapter in Unweaving the Rainbow

called The Genetic Book of the Dead, which sort of takes off from the, is it a Hindu classic? The Tibetan Book of the Dead. The Tibetan Book of the Dead. Sorry, Buddhist, yes. So The Genetic Book of the Dead, I see the genome, well, let's say the genes of a species as a coded document describing ancestral worlds in which ancestors survived. That's sort of true because they are a filtered subset of genes which have helped ancestors to survive. And in principle,

it should be possible in some future date for, when technology has advanced for a knowledgeable geneticist to read the genome of an individual and actually read off a description of the worlds in which the ancestors of that animal lived. To a lesser extent, I think perhaps it's easier, you can read the body of the animal. I mean, I like to think that if you took a whole lot of water-dwelling animals, say mammals, so it would be otters, seals, whales, water shrews, marsupial swimming animals and things, they'd all have webbed feet, say, for example, except whales. And so that's an obvious one. But if you actually made a list of characteristics of water-dwelling mammals and compared it with, say, desert-dwelling mammals,

you'd find a whole lot of things that all the water-dwelling ones have in common, including probably some biochemical ones, some genetic ones. And so that's part of the description. The Genetic Book of the dead describes water or describes desert. Right. And one day, maybe I'll even write a book called the genetic book of the dead, trying to flesh out this idea. Yeah. And of course, it could also look forward prospectively to situations which we, now to take the human case, are not well adapted to figure out.

That's right. I mean, the genetic book of the dead has always got to be a description of the past. And it helps the animal to survive to the extent that the future resembles the past, which on the whole it does. If the world were totally capricious such that you could not predict the future on the basis of the past, then natural selection wouldn't work. But nature doesn't vary capriciously as the years go by. On the whole, tomorrow is pretty similar to yesterday. Actually, there was a specific question that touches on that point that someone asks, why do we need vaccinations or acquired immunity to diseases at all? Why can't the mother pass on her immunity to her offspring?

Wouldn't that be an enormous evolutionary benefit? So we have acquired immunity because on the assumption that the environment does change enough that that's the best algorithm to run. Yes. I suppose the immune system is a kind of short-term, moment-to-moment substitute for natural selection. Natural selection works over generations and equips the animal to deal with circumstances that arise perennially, or at least over a long period. The immune system is all about equipping the animal, adapting the animal to insults that attack it during its own lifetime from moment to

moment. There are always new epidemics, always new viruses cropping up. So that's what the immune system is about. And vaccination is, of course, sort of preparing the immune system. A gaming of that, yeah. But it would seem good to be immune to everything that your mother had encountered. But it would seem good to be immune to everything that your mother had encountered. It would. I suppose we tend to be immune to everything that most of our ancestors have encountered, but just our mother. We don't seem to have a mechanism for passing on the particular.

If the mother's had chickenpox, we don't inherit an immunity to chickenpox. One thing that's framing this part of the conversation for me is I watched your somewhat stifling conversation with Brett Weinstein, who I greatly admire, who I've done many events with, but he had a kind of ax to grind with you around, if not group selection, something he was calling lineage selection. And more broadly speaking, a sense that evolutionary thinking should cover many of the details of human life, like war making, genocide, nationalism, to a degree that you were disinclined to extend it. And also just this notion that religion should certainly be considered an extended phenotype. Memetics generally should be considered an extended phenotype. And I'm just wondering what the... I mean, I can't do a good impersonation of Brett

for this conversation, but I'm wondering just what are your concerns there and what are the limitations in Darwin darwinian thinking when we're talking about high level human social phenomenon and psychological well first of all i i i hugely admire brett weinstein's stand at ridiculous university that he used to be a member of evergreen yeah evergreen yes i mean he's a real hero for standing up against that nonsense the extended phenotype i think is often misused the idea of extending the type is often misused and i think we should remind people what a feeling yes i should a phenotype is that which the genes engineer in a body which not which in a darwinian sense would help the genes engineer in a body,

which in a Darwinian sense would help the genes to survive. So wings are part of the phenotype of genes that help the genes survive and behavior patterns and crests and sharp talons and sharp teeth and things. So we normally think of genes program bodies to develop phenotypes. Phenotypes help bodies to survive and that helps the genes that built them survive. That's the normal way it happens and genes do it by the processes of embryonic development causing the body to develop the necessary phenotypes. The extended phenotype is phenotype which is outside the body in which the gene sits and my classic examples of this are animal artifacts things like birds nests where the nest

especially a complicated nest like that of a weaver bird obviously an adaptation i mean it's just like an organ it's beautifully shaped for a particular purpose, beautifully shaped, for example, long tubular nest to prevent snakes getting in. That is a perfectly good phenotype. Right. But it's not part of the body. Yes, the genes are producing that. The genes are producing that. And they're producing it. They're still doing it via embryology. But the embryology then, as it were, reaches outside the body in the form of behavior, reaches outside the body in the form of behavior, in this case, nest building behavior. But that's only a yet one more step in the embryonic chain of causation. The embryonic chain of causation begins with DNA influencing proteins, and that influences something else, which influences something else. Cell division, neuron production in the brain,

which has the eventual consequence of causing the bird to build a nest of a particular shape. So there's just this chain of causation, starting with DNA protein and going through various complicated steps in embryology. And then the final steps are outside the body. So I call it the extended phenotype. Well, then the idea is generalized to... So I called it the extended phenotype. Well, then the idea is generalized to, for example... I just want to pause here, at the risk of derailing you. I want to pause here to close the door to a certain species of doubt that evolution can explain the diversity of life that we see.

So now I'm just closing the door to the creationists and the intelligent designers for the moment. Because one of the concerns is that when you take any example of phenotype, you take a bat's wing, for instance, evolution could not have produced a bat's wing, de novo, you know, by functional bat's wing. What you need is some incremental path from no wing at all to a bat's wing, and each increment has to survive the logic of evolution. It has to be useful and lead to differential success. So you have to imagine here to explain any speciation and any path by which we have reached the diversity of life that we see, you have to explain how each increment, the first

little bump that became the wing, how that in itself was useful. And many people just throw up their hands there and say, well, there's clearly no way you can do that. So there must be some other explanation. Yes, thank you for reminding people of that. I mean, that is, of course, very important. And of course course evolution has to take whatever is there and modify it so it is it's not like a little bump that appears in this case it's already existing arm yeah in the case of insects probably was a little bump right because that's not using an existing limb but yes you're you're of course right with a bat it's it's literally the hand yeah and as a membrane stretched between the fingers, which is not difficult to engineer embryologically

because in the embryo, there already is a membrane between the fingers, and actually it's carved away. There's a kind of sculpture process whereby the membrane is removed. All that needed to happen in bats is that that sculpting process didn't happen. The membrane stayed.

And of course, the fingers get hugely long. Pterosaurs do it differently. They just have one big finger. Right. And they stretch that between the legs. And birds do it differently again. But in every case, it makes use of what's already there. It modifies what's already there,

rather than starting de novo, which is what a human engineer would do. We start with a clean design on the drawing board. Yeah. But I was saying about the extended phenotype. But actually, before we get there, so what is the argument that some non-functional precursor wing would nevertheless have been useful enough?

Yes, I mean, this is a favorite problem. What's the use of half a wing? There are a large number of animals that don't exactly fly, but slightly increase their, for example, arboreal animals, squirrels, say, who leap from branch to branch. And it's a dangerous process, leaping from branch to branch. And so any slight increase in flight surface, they're not really flight surfaces, but any slight increase in the surface area that's presented to the air will increase the distance that a squirrel can leap.

The tail, the fluffy tail of a squirrel, acts as a sort of rudimentary aerofoil that increases the distance that a squirrel can jump to. Well, now flying squirrels, they're just squirrels, but they have a membrane between the forelimb and the hindlimb, which started out, no doubt, as just a bit of membrane in the armpit. It just slightly increased the, you know, it could just leap one foot further because of that. And then when that was there, then next generation, perhaps next 10 generations, could leap 10 feet further. So you have a steady gradient of improvement. Are there orthogonal gradients that could explain some of these intermediate forms like heat regulation or something? Well, that's been suggested for insects, yes.

It's been suggested that in the insect, they really did start by just bumps growing out of the thorax rather than modifying existing limbs. And it has been suggested that originally these were thermoregulatory or were solar panels. Right. And then when they got out to a certain size for their thermoregulatory function, they then happened to act as aerofoils.

And so they then became wings. And insect wings are moved not by limbs, as I said, but by movements of the thorax. not by limbs, as I said, but by movements of the thorax. So the thorax is, there are muscles in the thorax that contract it in various ways, which cause these flaps to go up and down. Interestingly, some insects flap their wings up and down with a separate neural command from the central nervous system saying up, down, up, down. But other insects have a kind of motor, sort of oscillating motor, where all that the central nervous system says

is switch on or switch off. And the motor itself does a rhythmic up, down, up, down, up, down, up, up, down. And the frequency of the oscillation is determined not by the central nervous system, but just by the harmonic properties of the system. Okay, so back to the extended phenotype. bird's nest, say, there are many cases where parasites manipulate their hosts to increase the chance that they will be propagated to the next stage of the parasitic cycle. So flukes, for example, usually have an intermediate host, which might be a snail, or it might be an ant. And they need to get into their definitive host, which might be a sheep or a cow.

And so in the case of the so-called brain worm in the ant, for example, the worm in the ant burrows into the brain of the ant and changes the behavior of the ant to make the ant more likely to be eaten by a sheep. It crawls up to the top of stems in the heat of the day rather than going down into the ground. So the parasite is a kind of puppet master which is manipulating the ant to get... Well, now, that to me is an extended phenotype because genes in the worm

have their phenotypic effect in ant behavior. Yeah. I think, isn't there some evidence that toxoplasmosis and some other organisms operate in mammals like ourselves and very likely in people in similar ways? Well, to make us more likely to pass it on. Yeah. Yes. Modifying behavior.

I mean, rabies is the classic example. The rabies virus actually makes rabid dog, for example, more likely to bite and froth at the mouth and pass on the virus when it bites. It also makes the animal more likely to roam and wander far and wide rather than stick around at home, which then spreads the virus more widely. So that's extended phenotype of a parasite. And then you can say, well, parasites don't always live inside their hosts. Cuckoos manipulate their hosts. A cuckoo nestling.

Yeah, terrible bird. Yeah, terrible bird. Manipulates the host with beautiful adaptations. I mean, supernormal gape and things like that. This is, again, manipulating the host behavior. The change in the host behavior is an extended phenotype of cuckoo genes. Genes that change host behavior are more likely to survive. Again, it works via cuckoo embryology, but the final stage in that chain of events in cuckoo embryology is to produce behavior which seduces the host, the reed warbler, whatever it is, or the robin, whatever it is.

And so that again is extended phenotype. And then the next, the final stage in my argument would be all bird song, all animal communication, where one animal manipulates another. You can think of as extended phenotypes. So a gene that changes in one animal has an extended phenotypic effect on another animal via a call, a song, a crest, a flash, a conspicuous signal. So my whole vision of animal signaling is a great network of extended phenotypes. Right. Okay. So before we talk about the prospect of something like religion or any other doctrine or institution being an extended phenotype for humankind let's briefly talk about this other species of replicator the meme which is a a term of your coinage which has an importantly different connotation now now that we've all well spent our lives on social media but these are also it's it's not it's actually a decent analogy to yes to the meme i mean i i wanted i wanted to

make the point that that what matters is replication in and genes are consummate replicators and they and they achieve their replication success by manipulating bodies via the processes of embryology but i wanted to make the point that any replicator could do that it doesn't have to be dna and of course on other planets it almost certainly isn't dna if there is life on other planets which there make the point that any replicator could do that. It doesn't have to be DNA. And of course, on other planets, it almost certainly isn't DNA if there is life on other planets, which there probably is. But then I said, well, maybe we don't have to go to other planets because maybe memes, maybe cultural replicators could be the basis of Darwinian selection. There certainly are cultural replicators, no doubt about that. I mean, things spread. Does it matter that they don't randomly vary? The mutation isn't...

I don't think that matters, no. I mean, it incidentally happens to be true that genetic mutation is random. But even that is only random in the sense that it's not guided towards improvement. But mutation is not random in other senses. Mutations are induced by cosmic rays, for example. That's non-random. But mutation is random in the sense that it's not... What do you mean it's non-random if it comes from cosmic ray bombardment? Well, it has a cause. It's predictable that if you subject yourself to... But the specific base pair that's being targeted is is random presumably that's true yes that's true but what's more important is that it's random with respect to improvement

so there's no tendency for for mutation to be as it were anticipating what's necessary for survival it is random in that sense and the great majority of mutations are actually deleterious. Okay, so when we talk about memes, right, so now a meme is almost any cultural product, an idea. That is replicated. That's replicated. So it could be a clothes fashion or something like that or a speech mannerism. Right.

Awesome. Awesome, yes. Which I use with disconcerting frequency. Do you? I never use it. I've given up. I mean, it's such a wonderful word to mean what it really does mean. Now it just means kind of okay. I'm part of the slide into

degradation. Well, language evolves. I am an American. This would be predictable. It's a good case because language does evolve and so we have to accept that. Yeah, and I think there probably I'm an American. This would be predictable. I mean, it's a good case because language does evolve, and so we have to accept that. Yeah, and I think there probably is some randomness and not to say cosmic ray bombardment that accounts for the changes in speech patterns.

But most memes, it seems to me, the changes in them are engineered, at least with some forward thought as to survival. That doesn't matter, actually. It doesn't really matter. I mean, natural selection would still work even if mutation, genetic mutation was engineered. And, of course, it can be.

We are now in a position to do that. Right. That's what genetic engineering is. Which we'll talk about. So the fact that the basis for the change, directed or not, you still have an environment where things are competing and there's differential success. And so the environment is providing a kind of selection mechanism. Exactly, yes. So memes, ideas, ways of doing things, really all of human culture and ideology. This is being continually produced and spread and going in and out of fashion.

And so this is this domain of memetics. And there literally are what are now called memes on the internet, graphics paired with text that spread on social media, that spread various ideas. I don't know. How do you feel about that appropriation? Well, I'm not particularly keen on that appropriation because they are a very specific example of a meme. I would rather think about whether natural selection of a sort

actually guides the spread of memes. And I like the idea of a meme complex or memeplex where something like a religion, like Roman Catholicism, could be regarded as a meme complex. Right. And individual memes might be the idea of life after death or the idea that you have to confess your sins or something like that. The virgin birth. Virgin birth.

And just like gene complexes are sets of genes which flourish in each other's presence. And that, I think, is an extremely important idea in genetic evolution. So there might be something similar in meme complexes. Yeah, so there's a common fate to these various genes and various memes. They're all hitched together. Yeah. That's right. And so I like to think of, say, the gene complex of a carnivore species like leopards, where you have carnivorous teeth, carnivorous eyes, carnivorous brains,

carnivorous limbs, they all go together. And on the other hand, you have antelope, I mean, the herbivore prey, eyes, noses, limbs, etc., which go together. If you suddenly plonked an antelope gene into a leopard gene pool, it probably wouldn't work. It wouldn't cooperate well with the other genes of the leopard gene complex. It would be a very skittish leopard. Yes. Yes, yes, yes. The cowardly leopard. Cowardly leopard. And so a species is a collection of mutually compatible genes,

which go well together, as opposed to another species, which is complex of different genes. Well, I believe you might do the same kind of thing with meme complexes. But the theory hasn't been really sort of worked out. I think it might be. Okay, so we have meme complexes, something like Roman Catholicism, and what was being urged upon you in your conversation with Brett, and I've seen this come up many times before, is that something like Roman Catholicism should be, or religion in general, should be considered part of the extended phenotype of human beings. Yes. I've never liked that. I've never liked...

I think that's taking the idea of the extended phenotype beyond where it should be. And I think it detracts from people's ability to comprehend the idea of the extended phenotype. Extended phenotype is supposed to be a genetic effect which manifests itself outside the body in the same kind of ways. Genetic effect manifests itself outside the body in the same kind of ways genetic effect manifests itself inside the body. And people have some, I don't think Brett does this,

but people have sometimes said to me, isn't a building like the building we're in at the moment, an extended phenotype. And I think that would only be true if, say, there were genes that caused architects to design a different kind of building. And there aren't. I mean, there's no gene that makes an architect more likely to make Gothic arches rather than Romanesque arches. So our mere survival dependence on buildings

is not enough to have it? No, I don't think so, because variation in buildings is not under genetic control. Right. And I doubt very much that variation in religious habits is under genetic control. If it was, then you might make some sort of a case for talking about extending things in time. But it's not like that.

And so I think that it's possible to push an idea too far, and I think that's what's going on here. So what about the prospect that having religions led to differential success of various groups of human beings? Well, that's quite a different idea, and that's worth considering in its own right. And also it's what's worth considering in its own right. And also, it's what's worth considering in its own right, is the idea that individuals having religions might survive better. That's been suggested and might be true.

And this opens the door to what's been called group selection. Yes. And I've never been a fan of group selection. Darwin himself, it wasn't called group selection then. Darwin almost always was talking about individuals surviving better within a species, but Darwin did, again in The Descent of Man, in one passage talk about a kind of group selection where he suggested that groups of humans who had some kind of social cohesion, who behaved

well towards each other, had altruism towards each other, cooperation, would be more likely to survive than groups that didn't. And so that would be a form of group selection, I suppose. In some ways, I prefer to compare that not to group selection, but to species competition. A bit like when the gray squirrel was introduced from America into Britain as a sort of frivolous exercise. We did that to you? Was that a good idea or a bad idea? Terrible idea. And it drove the red squirrel extinct and so i think that's that would be a better analogy for a group like a group that say has a has a warlike aggressive god like yahweh or like

some of the norse gods you could make a case that having an a militaristic god maybe one who rewards martyrs in a martyr's heaven that kind of religion might spread as a kind of group effect as a kind of species effect an ecological competition effect but i call i would call that ecological competition rather than group selection i think we have because it's so let's just create a an example was a let's say that Hitler won the Second World War and we are now living under the thousand-year Reich and everyone who's not a Nazi is now dead so Nazism would have triumphed over all competing political ideologies so that we some level, you can say, well, this is a selective effect, right? There were various competitors for political ways of thinking,

and one has finally dominated and canceled all others. But that doesn't seem to suggest an analogy to the replication model. I don't think it does. No, I don't think it does. I mean, slightly closer would be if, say, within any country, individuals who espouse Nazi beliefs were more likely to survive than individuals who didn't.

So there would be an individual differential survival effect, which probably would also be in the case. That would be a closer analogy to Darwinian selection. And we might do a kind of memetic analysis of that. Nazi memes survive better than anti-Nazi memes, for example. That would be the case of memetics.

Yeah, I think that might actually be the environment we're currently in on social media. Forbear to comment on that. So I guess one final question here. So are there outstanding questions in what is now as though neo-Darwinism, and perhaps you should actually define that term, is basically flawed in a way that should be troubling to biologists

and public intellectuals. Yes, I don't think that. Any flourishing science will change, of course. And Steve Gould was fond of saying that the modern synthesis is effectively dead. And I thought that was a rather irritating attempt at almost self-publicity. Well, he was irritating in many ways, as it turns out.

Okay, yes. But what is your... So first define neo-Darwinism. Well, okay. Neo-Darwinism is the neo-Darwinian synthesis that was a joint effort in the 1930s, really, of, I think, above all, R.A. Fisher, J.B.S. Haldane, Sewell Wright, Ernst Mayr, Theodosius Dobzhansky,

G.G. Simpson and others. And it was seeing Darwinian evolution as changes in gene frequency in populations, that was the population genetic part of it, seeing, well, the paleontological part of it would be seeing major macroevolutionary change as microevolution writ large. So the geneticists were showing how from generation to generation you could get slight changes in gene frequency. And the paleontologists like Simpson was showing that such microevolutionary changes extrapolated over millions of years, tens of hundreds of millions of years, could produce changes from fish to mammal.

So this movement of the 1930s and 40s, we're still in it. It hasn't really changed much. There have been, I suppose, W.D. Hamilton with his analysis of altruism, kin-selected altruism, is one major advance of the 1960s and 70s. But we're still in the neo-Darwinian era. And you don't think there are gaping holes in the theory that should keep people up at night? No, I don't. I mean, there are questions that remain to be answered. One of the big riddles is the evolution of sex, what sex is good for.

And lots of the most distinguished neo-Darwinian theorists have grappled with that problem. The origin of the Darwinian process is still a bit of a mystery. How did the first replicator arise? And was it, it almost certainly wasn't DNA, actually. I mean, the first replicator would have been something else. Would have been RNA? Maybe.

That's a good possibility, and that's one of the more fashionable ideas. But that is still in the realm of theory. It may never become settled because it happened a long time ago and may be impossible to repeat exactly what happened we know the kind of thing it must have been it was the origin of something self-replicating possibly rna so what about epigenetics and the the way in which they this feature of our biology

seems to suggest a almost a quasiLamarckian kind of inheritance. Yes, it's a strange word, epigenetics, because actually, originally, it was just another word for the way we see embryology. I mean, every cell in the mitotically, every mitotically reproducing cell in the body has the same genes. Right.

And yet, only some of them are. All the genes that your brain cells have. That's right, yes. And different genes get turned on. And so the epigenetic environment of a gene in a brain cell is different from that in a liver cell. And so that's epigenetics. The word has been hijacked fashionably recently by people with, as you say,

a kind of Neo-Lamarckian bent to suggest that some of that epigenetic cytoplasmic environment in which some genes are turned on and others are not can get inherited to the next generation. And that does seem to happen in some cases. So examples like the stress experienced by the mother with the infant in utero, the change in hormonal environment there can actually create some durable effect on the expression of genes in the baby. Yes, that does seem to be, of genes in the baby. Yes, that does seem to be. There are a few rare cases like that.

I don't think it's worth the attention that it's been given. I prefer to reserve the word epigenetics for the ordinary process of embryology and say, just occasionally, there may be epigenetic effects which do pass on to the next generation, maybe even to the grandchild generation. But it's not one of these things that goes on forever, like true genetic mutation. So what is the current frontier of evolutionary biology? If you could pick one question. If you'd like to continue listening to this podcast, you'll need to subscribe at samharris.org. You'll get access to all full-length episodes of the Making Sense Podcast and to other subscriber-only content, including bonus episodes and AMAs and the conversations I've been having on the Waking Up app. The Making Sense Podcast is ad-free and relies entirely on listener support.

And you can subscribe now at SamHarris.org.