195 | Richard Dawkins on Flight and Other Evolutionary Achievements

0:00:00.0 Sean Carol: Hello, everyone, and welcome to The Mindscape Podcast. I'm your host, Sean Caroll. Back when I started blogging, so like 2004-2005, the hot topic in the science blogosphere, believe or not, was creationism, and intelligent design, and the battle with natural selection, and Darwinian Evolution. It was a big story. It was not just on blogs, even the New York Times was writing about this. I don't hear about this that much anymore. I'm sure it's still happening, I'm sure that there are still people trying to get intelligent design in high schools or whatever, but you don't hear about it that much. It makes me wonder what the connection is, between what happens in the world and what you hear about, what the hot topics are, what people are actually talking about. It's probably a pretty flimsy connection in some sense. But anyway, that is not the point of this podcast. The point is that, I thought about it because one of the issues that is always brought up by people who are in favor of creationism or intelligent design, is that there are capacities that living creatures have, that don't seem naively evolvable by a series of small incremental steps.

0:01:02.8 SC: They even tried to quantify this idea in the notion of irreducible complexity. Something that was very complex and functional, but if you removed any piece of the operation, it would be completely useless. And the idea was that things like the eye would be hard to imagine, growing gradually. In fact, that's completely nonsense. Eyes are one of the easiest things to grow gradually. Even in a more simplified context, one of the examples was a mouse trap. If you remove any piece from a mouse trap, it doesn't do anything. Of course, someone instantly invented a reducible mouse trap that you could build up piece by piece. I even wrote about it in The Big Picture. So the lesson was that, things that you are skeptical can be built up piece by piece, are actually very buildable. So some good science came out of that understanding how that works. One such example is flight. Either you fly or you don't, in some very naive sense, but even thinking about it a little bit, you know that's not true. There's floating, there's gliding, there's jumping, there are all sorts of halfway houses to flying.

0:02:07.4 SC: Nevertheless, the question of how flight actually arose through biological evolution is fascinating. So today, we have on the podcast, Richard Dawkins, presumably needs no introduction, and he has a new book out, called Flights of Fancy Defying Gravity by Design and Evolution. And it's a pretty broad overview of how flight works in general, not just in biology. So he includes airplanes and space ships and so forth, and then he compares them. And he's talking about like, "Why does Evolution and Biology use certain kinds of flight, but not other kinds of flight?" Which is a fascinating question to contemplate. And I'll confess, we use the podcast, I use the podcast to go beyond the topic of the book, to talk about Biology and Evolution more generally, questions of adaptation versus randomness, if you played the tape of life once again, would you get to the same kind of Ecology, etcetera. The relationship between genes, and of organisms, and species, etcetera, because look, when you have Richard Dawkins on the podcast, you're gonna wanna ask about evolution very, very broadly. So, we have a wide-ranging conversation about all of those things, and even at the end, we do a little tiny bit about Philosophy and Science. So I'm sure you'll wanna tune in for that. So with that, let's go.

[music]

0:03:40.4 SC: Richard Dawkins, welcome to the Mindscape Podcast.

0:03:42.0 Richard Dawkins: Thank you very much.

0:03:43.5 SC: I love the idea that you've written a book about the evolution of flying, because of course, this is one of the classic puzzles for those of us who like to believe in evolution, evolution works incrementally, and flying is either yes or no, I guess, in our naive conception of it. Why did you pick flying as something to write about, for this book.

0:04:02.7 RD: Well, there are things which are yes or no, but flying actually is not one of them. Because flying, you've got everywhere between flying and gliding, just sort of soaring with not proper wings, but just a bit of a membrane.

0:04:17.1 SC: Sure.

0:04:18.0 RD: You think about a leaping animal like a squirrel, which is up in the tree tops, and it's leaping from branch to branch, and however far it can leap, without any additional flight surfaces. There's just a little bit further it could leap with a flight surface. So somewhere up there, there's going to be a branch distance which it could just reach without an extra flight surface and now it can, because it's got the extra flight surface. So, any little membrane and any little extra flap of skin, extra fluff on the tail, anything to increase the flight surface, the surface area of the animal, will get it just an inch further, or two inches further, or three inches further, and then once it's gotten that much further, then there'll be another pair of branches which are just within reach, if it's got a slightly larger membrane. And we see that, in the forests of Southeast Asia, there are a whole range of... And Australia and Africa, but especially South East Asia, a whole range of animals which glide from tree to tree, and they have various kinds of membrane. There is a flying squirrels that have a membrane stretching from their front legs to their hind legs.

0:05:42.3 RD: There are flying lizards which stick their ribs out, and they have skin stretching between the ribs, there are flying frogs which have large fingers, long fingers, and they have extra membrane in between their fingers. There are marsupial gliders, there are gliders that more than one family of rodents just develop doing this gliding trick. Well, it's not a big step to go from there, to actual controlled flight, like a parachute is, to altering the shape of the parachute, and they do that, and you could watch them as they adjust their... They can steer like a parachutist can. It's not that far, to go from that to flapping, and so you could go to... Maybe the evolution of bat flight could come by going from having a gliding membrane to a flapping membrane.

0:06:37.8 SC: And there's a lot of benefits to this, clearly, but you already brought up something I was saving for later. But let's get right into it. Which is the idea that, climbing into trees presumably plays a large role in why flights developed in the first place. Once you're up there in the tree, then there's an evolutionary pressure to get better at gliding and jumping in a way that it's even more or than if you're sitting on the ground, to start.

0:07:02.4 RD: Yes. And so many animals do climb trees. There's squirrels, there're primates. There's no primates that's developed that gliding trick. There are Gibbons that leap...

0:07:12.8 SC: Oh really!

0:07:14.3 RD: At phenomenal distances and swing themselves from their arms, which is almost like flying, but they haven't developed their extra membrane. Oddly enough, there are quite a lot of people who believe that flight in birds, at least, did not come from them being up trees. They think it came from dinosaurs; birds, of course, are dinosaurs, as you know, leaping into the air, maybe trying to catch an insect or trying to catch, trying to pounce on a prey and then pouncing just that little bit further because you have a little bit of membrane to increase your surface area, to catch the air.

0:07:51.9 SC: Actually, this is pointing to something that is an embarrassing gap in my knowledge. I do know that birds are descended from dinosaurs, but are they descended from flying dinosaurs or did flying get lost and then come back?

0:08:04.9 RD: No. No. No. Flying... There are Pterodactyls.

0:08:08.4 SC: Yeah.

0:08:08.9 RD: Which actually technically aren't dinosaurs either.

0:08:11.4 SC: Right.

0:08:11.4 RD: But, they are related to dinosaurs and they independently evolved true flight, true flapping flight. Birds are descended from feathered dinosaurs. Lots of dinosaurs, we now know, had feathers, probably for thermoregulation purposes originally. And birds then... Well, they are dinosaurs, in the sense that they come from within the dinosaurs and there are dinosaurs which are more closely related to birds than they are to other dinosaurs. So that means that birds are really slap-bang within the middle of the dinosaurs. So, they developed feathers for thermoregulation purposes, and then the feathers proved useful for a flight surface after that.

0:09:00.0 SC: A very typical evolutionary trick to repurpose something that was developed for something else.

0:09:03.0 RD: Exactly.

0:09:03.6 SC: But, your mention of the primates is interesting, 'cause I hadn't thought about that. No flying primates as far as we know, if we don't believe in angels, I suppose. But, is that partly because of the size constraints? One of the big things you mentioned right from the start, in the book are the physics constraints. It's easier to fly when you are tiny, and primates tend to be big.

0:09:25.9 RD: Yes, yes, that's right. There are so-called flying lemurs. Which the name suggests that they're primates. That's actually... Unfortunately, they're not primates. But they are related to the primates. So flying lemurs, Colugos, they carry the parachute trick to its extreme. They look like flying squirrels, except that the membrane goes to the tip of the tail, as well as the arms and legs.

0:09:48.6 SC: Yeah.

0:09:48.6 RD: So they're really just one big parachute. If they were really lemurs, then that would be an example of a flying primate, but then strictly speaking, they're not primates, although they're called flying lemurs.

0:10:01.1 SC: I'm gonna put you on the spot here a little bit and ask you to talk about the physics of size. I know that I'm supposed to be the physicist here. But, there are laws relating weight versus surface area and things like that. Tell us how those go into enabling some animals to fly more easily than others?

0:10:19.3 RD: Well, I'm embarrassed talking to a physicist about this, but for the benefit of the audience. As you increase the linear dimension, the size of anything. It doesn't matter whether it's an animal or a block of wood. If you... The weight of... The volume and therefore weight goes up as the cube of the linear dimension. The surface area goes up as the square of the linear dimension. So, that means that the smaller you are, the relatively larger is your surface area compared to your weight. And so, something like a Dandelion seed with a little puff, a little bit of fluff onto it, or a gnat or a mosquito, is so small that its surface area is large, compared to its weight. And so, it hardly need bother to fly, just kind of floating around in the air. There's a tiny little insect in the book, called Tinkerbella, which is named after the fairy in Peter Pan. And that's so small, we got a picture in the book of it flying through the eye of a needle. So that... Its wings just look like one big feather. It's like, it used its wings to kinda rose through the air rather than actually fly in the ordinary sense. So, the smaller you are, the easier it is to fly. If you are the size of a horse, say, then flying becomes essentially impossible. There are... There's one Pterosaur.

0:11:49.9 SC: Yeah.

0:11:50.3 RD: Quetzalcoatlus, which is may maybe as heavy as a horse. And that managed to, presumably glide from high places.

0:12:00.5 SC: I think I'm gonna let you tell us a little bit more about Quetzalcoatlus because it's a pretty remarkable beast, the size of a giraffe. And it seems like a little bit of an outlier in terms of flying animals.

0:12:12.1 RD: Yes. It certainly is. I think nobody knows how exactly how it flew. It must have flown. It's got wings. It probably is maybe jumped off cliffs, who knows what it did. Once up there and you could probably stay up there because you can glide. It's about the size of a light aircraft, about the size of a Cessna, or Piper light aircraft. That's the largest creature that ever flew, as far as we know. There are some pretty big birds as well. There's a relative of the Condor, which is... Well, 'cause it'd be larger than a Condor, the thing looked a bit like an Albatross, again, much, much larger than an Albatross and they probably also glided.

0:13:06.6 SC: It's a good example of the constraints... Not the constraints, but the competing interests in evolution. Flying seems like a good ability to have, all else being equal, but maybe being large is also a good ability to have, and there's a tradeoff there.

0:13:24.3 RD: Exactly, yes. It's a tradeoff and much of the book actually, keep coming back to this theme of economic tradeoffs. And yes, there will be a tradeoff between the advantages of being large and the advantage of flying, but actually flying is not an unmitigated benefit. We know that because quite a lot of animals have actually lost the power of flight.

[chuckle]

0:13:47.8 RD: And there's a chapter on that. I think we call it, "If Flying Is So Great, Why Have So Many Animals Lost It?". The most dramatic example of that is Queen ants and Queen termites, which fly for one purpose only, which is to mate and they take off on a mating flight and an ant queen mates once in her life, only once, during this mating flight, and then she settles down and digs a hole in the ground and builds a nest, founds a nest. Before doing so, having landed on the ground, she bites off her wings or tears off her wings, in different species.

[laughter]

0:14:26.9 RD: So, that's a very dramatic demonstration that flying is not always a good thing. They actually take... Go to the lengths of biting off their wings in order to presumably, function better underground, and of course, worker ants don't have wings, even though both their parents had wings. Then there are lots of birds, which when they get to islands, lose their ability to flight... To fly, things like Dodos, which must have used their wings... A Dodo is a kind of overgrown Pigeon, and it must have used its wings to get to Mauritius in the first place, but once it arrived on Mauritius and there was no predators worth worrying about, and so it lost its ability to fly. And the wings shrank and it could only then waddle around and then became a victim of sailors clubbing them to death.

0:15:25.5 SC: Of course.

0:15:25.9 RD: 'Cause it couldn't even escape. And...

0:15:27.0 SC: Presumably, penguins are a similar story, right?

0:15:30.4 RD: Well, Penguins, yes. That's not quite the same story because in their case, they took up swimming instead. So, they fly underwater. If you watch how Penguin actually swims, it swims using its wings. So, its wings are optimized for underwater flying, which means they're very small. And so it, as it were, put all its eggs in this one basket of swimming, whereas something like a Puffin, which also flies under water, but flies in air as well. So it has to compromise, uneasy compromise between the best shape and size for a wing to be underwater, which would be Penguin-sized, and the best shape and size to be in air, which would be say, Gull-sized. And Puffin is sort of halfway between, so it's not a very good flyer and it's not a very good swimmer, but it does adequately well at both.

0:16:29.3 SC: I guess it's a really good example of the way that evolution works because what I said, what I blurted out was, flying is clearly advantageous, and as you said, not always. I guess what's really true is that flying is fun. [laughter] We all want to do it. It seems like an ability we would want to keep, but evolution doesn't care about that. Evolution just says, "Is this good to our reproductive success?"

[laughter]

0:16:52.3 RD: It doesn't. I bet it is fun. I think if you watch Gulls playing in the wind, it's very tempting to think it's fun, and then you can, as it were, justify that by saying, "Well, let's not call it fun. Let's say they're practicing."

[laughter]

0:17:12.2 RD: That's fair enough, actually. It probably is a skill that benefits from practice. And so, if you do a bit of aerobatics in gusts of wind, when you are not trying to say, escape from predators, then it might serve you in good stead when you are.

0:17:31.8 SC: Well, fair enough. But this does get us right into some deeper questions about evolution. When we see the birds flying around in ways that seem like fun, to what extent can we say, "That's an adaptive behavior that is training them to be better flyers," and to what extent can we say, "They're just having fun and it it's not necessarily improving their reproductive fitness in any way."?

0:17:57.1 RD: I suppose, as an Orthodox Darwinian, I would have to say something like this, that if it were only fun and they were actually wasting time and wasting energy, then a rival bird that conserved its energy and conserved its time would out-reproduce it. And so, I think there's gotta be some kind of... As a loyal Darwinian, I'd have to say, there's gotta be some kind of added benefit to the fun, such as practice. And I think practice is a perfectly respectable function.

0:18:29.0 RD: I think if you watch a young bird learning to fly, they clearly are learning to fly, there's no doubt about it. A bird... A young Eagle on the nest, say, it sort of hops up and down, and flaps its wings and kind of hopping, and doesn't actually take off from the nest, but it's exercising its wings and also probably maturing its flying skills, as it does so.

0:18:54.6 SC: Well, this is good. This is gonna get us into, again, the deep issues here, because it doesn't necessarily... And I'm not an expert here. It doesn't necessarily seem to me, to be the case, that every activity that an animal does needs to have some reproductive success, right? After all, as someone once taught us, it's the genes that are selected for. And the genes could have various effects, some of which are advantageous traits and some of which are just spinoffs, yeah?

0:19:29.5 RD: Yes, that's true. But, of you... In a way, you've sold the past by using a word like fun. And you've almost said, "Well, because I'm daring to use the word fun, I'm kind of conceding that it's not having that additional spin-off effect." If it is, that's fine of course. If, say, the genes are having a what they call a pleiotropic effect. Pleiotropic effect means the gene has more than one effect, perhaps in different parts of the body. It could be like that, but in a way, that's sort of not addressing the issue that we're trying to talk about. We're trying to actually have an argument about whether... And it's an interesting argument. I don't see why one shouldn't argue about whether animals are motivated by the sheer joy of flying. And the way I was trying to give it my best shot by suggesting that the sheer joy of flying could be useful, because of practising for when it's really needed in earnest. So in a way, to use the pleiotropy it's a bit of a cop out. You're kind of saying, "Well, yeah, maybe it's useful for something else." Which is true. It could be. But it's not really getting to the heart of the interesting discussion we were trying to have.

0:20:51.4 SC: Well, let's ask the philosophy of science question then. How... So let's say that we have some purported explanation for why this behaviour is adaptive and helps with selection. How do we know? How do we test that idea, whether it's just from pleiotropy or whether or not there's a specific advantage to this specific behaviour?

0:21:13.9 RD: Well, that's an excellent question. I don't think that's an easy one to answer as in terms of actual evidence. I think what we could say is something like this. We have so much confidence in Darwinian theory, that it would be... We would have to be throwing a lot of useful stuff out if we were to accept that it was purely fun. Let me give another example, which is not to do with flying but which is interesting. Bird song has great aesthetic appeal, or some bird songs have great aesthetic appeal for humans. As Keats said of the nightingale, "My heart aches, and a drowsy numbness pains my sense, as though of hemlock I had drunk." The nightingale's song had a drugging effect on Keats's nervous system. And he would've called it an aesthetic effect, but he also likened it to taking a drug. Well, Keats's nervous system was a vertebrate nervous system, and so is a female bird's, a female nightingale. So, if the nightingale's song could have that effect on the poet, Keats, why wouldn't it have a similar effect on a female nightingale?

0:22:32.0 RD: And, I like this idea because we know that bird song actually does have a measurable physiological effect on the female bird's hormones. This has been shown in doves and in canaries. It actually... Male song in canaries actually causes the ovaries of a female to grow. So, it's as though the male is having a direct hormonal effect, physiological effect on the female. A human physiologist could cause a female's ovaries to grow by injecting hormones, or perhaps he could influence her behaviour by sticking electrodes in her brain. Well, the male bird can't do that. But he can do something equivalent, which is to sing. So there was a man called Hartshorne who actually tried to make the case that bird song was an aesthetic, that birds have an aesthetic appreciation of song, that it actually enjoyed song in the same way as we enjoy music. And he was rather ridiculed for that. But, it's not too far distant what I'm now saying, which is that the manipulative effect of a male bird's song on the female physiology is kind of like an aesthetic experience.

0:24:00.7 RD: Now, if we go now to the question of how the male bird learns to sing, it's been shown in a number of species, that birds, when they're developing, young birds, when they're developing their song, are teaching themselves to sing by trial and error. So, what they're doing is... And this has been shown by experiments. What they're doing is kind of burbling around at random. And every time they hear a phrase that they like... I used that phrase advisedly. They repeat it. So they're learning. They're teaching themselves to sing by repeating those phrases of burbling, of warbling, which appeal to them, turn them on. Now, a male nightingale or a male canary is the same species as a female. He has a similar brain. So, whatever turns him on might turn the female on. Well, that's getting perilously close to talking about it as an aesthetic experience, isn't it? It's saying, the male teaches himself to sing by singing phrases at random. And the ones that he likes, the ones that turn him on aesthetically, are likely to be the same as the ones that would turn a female on aesthetically and sexually. So we can kind of make a Darwinian justification for using the language of aesthetics. And in a similar kind of way, we might do the same thing for the aesthetics of flight, of enjoying it, of having fun flying, of having a fun experience of flying.

0:25:41.7 SC: I mean presumably as physicalists about the fundamental nature of reality, we don't believe that there is something purely human in the idea of aesthetic enjoyment, right?

0:25:52.6 RD: No. Why come up?

0:25:52.8 SC: It has to sort of come up there along the evolutionary ladder. So in some sense, there has to be some related thing in birds or bees, right?

0:26:02.1 RD: Well, maybe not actually has to be, but it's quite plausible that it would, yes.

0:26:08.3 SC: Did you see... This is gonna sound out of left field, but did you see the documentary on the Beatles, Get Back, that came out recently?

0:26:16.6 RD: No.

0:26:16.8 SC: The Beatles, the musical group, not the species of insect. But, there's a famous clip from it, in which Paul McCartney is just sitting, strumming on his bass guitar, and conjuring out of the ether, the song, Get Back. And he's doing exactly what you just talked about with the nightingales. He's just sort of randomly playing a couple of different things, finding the ones he likes and building upon those.

0:26:41.2 RD: I believe there are some notebooks of Beethoven which show development, where Beethoven jotted down phrases. But he used to go out for long walks and he had a notebook, and he would jot down phrases. And you can see some of the melodies of Beethoven which we now know and love, in their embryonic form, developing through the notebook. I've not actually seen the notebook itself, but I believe that's the case.

0:27:11.6 SC: Okay, very good. To get it back to genes a little bit. I wondered for a long time, and now that I have you here, this is a perfect time to ask these questions about the just numerical relationships between the number of different genes we could have and the number of different traits we could have. So, if it would be evolutionarily advantageous to fly or to sing or whatever, how obvious is it, that that pressure on our abilities can correctly shape our genes to enable that kind of thing? I don't know if this is too vague a question, but is there enough...

0:27:49.9 RD: No, that's very interesting.

0:27:51.3 SC: Is there wiggle room in there?

0:27:52.0 RD: I haven't heard it put that way quite like that before. First, as you know, it's a fallacy, to think about traits as being sort of unitary, like a butcher's diagram where there's a...

0:28:05.0 SC: Exactly, right.

0:28:06.9 RD: It's not like that. It's rather that what genes are doing is influencing embryonic processes. So, we think of it as a cooking recipe where you have a recipe which... It consists of a lot of words, and there's no one-to-one mapping between words of the recipe and bits of the dish, that it's a cake or whatever it is. It's rather that the whole recipe maps onto the whole cake.

0:28:37.1 SC: Right.

0:28:38.8 RD: And a little bit of heat here, a little bit of added milk there, whatever it might be, has an effect on the embryology of the cake. And so... But, that's okay. That doesn't take away from the question that you're asking, which is more; is there a limit to the detail that genes are capable of specifying?

0:29:04.0 SC: Yeah.

0:29:04.3 RD: So something like bird song. We've already talked about that, and I said that birds teach themselves to sing by randomly burbling. And you could say, "Well, that's because the genes don't have the informational capacity to specify something as complicated as a song. It has to be done more indirectly than that." On the other hand, well, the shape of a spider web, say. It's not that the genes contain a kind of map of what a spider web should look like. It's more that the genes that can specify a set of rules which the spider obeys, and the shape of the web is an emergent property from the rules being obeyed. A termite mound is a complicated structure built by thousands of termites, and no termite has anywhere, either in its brain or in its genes, a picture of a termite mound. It's rather that each little termite is following little local rules which, when they all obey these little local rules, what happens is that a termite mound emerges. If you've ever seen films of starling... So-called starling murmurations.

0:30:31.0 SC: Sure.

0:30:31.4 RD: Where these starlings fly in these gigantic flocks, tens of thousands of birds wheeling and dancing in the air in the most spectacular rhythm. The whole flock looks like a great big amoeba. It seems to have a will of its own. It seems to have coordination. It's as though there's a master bird which is giving orders. But it's not like that. It's what it is, is each individual bird is following a set of rules, and the emergent property of all those rules is the murmuration of the entire... The flocking of the entire flock. This has been shown beautifully in computer simulations, where what the programmer does is to program a single bird, not a flock, just one bird, with rules. Just half a dozen simple little rules about keeping neighbours at certain angles, whatever it might be. Having programmed one bird, the programmer then, as it were, clones up that one bird and releases thousands of birds in the computer. And then what you see, what emerges, is the behaviour of the whole flock. So, the way genes program what we see as the phenotype is actually done in that kind of recipe-like way.

0:31:56.5 SC: Right.

0:31:57.6 RD: Programming in simple rules, which then... And now, we're not talking about individual starlings behaving in relation to each other, or individual termites behaving in relation to each other, we're talking about cells in the developing embryo. So, the cells in the embryo are obeying simple little rules, local rules. In one sense, there is no blueprint. In a way, the... What molecular biology textbooks say when they say that DNA's a blueprint for a body, that's not true. It's actually quite a bad fallacy. It's not a blueprint. It's a program or a recipe. It's a set of rules which are obeyed at the cellular level and the sub-cellular level. And the consequence of all these little rules being obeyed in all these little cells is that, a body emerges. I don't know whether that helps to clarify the question you were asking, about the number of genes you need in order to make... If you think recipe rather than blueprint, I think that's the key.

0:33:01.5 SC: No. Yeah, so I'm not sure if it does answer the question I asked, but it answers an even better question that I didn't think to ask. Because it's saying that in the space of all complex behaviours, we can imagine complex behaviours that are just the result of some super-genius intelligence, doing complex things. We can also imagine these emergent complex behaviors that result from the concatenation of simple rules and the kinds of behaviours that biology is going to eventually find due to the constraints of genes doing their thing, are these emergent things. It's a lot easier to...

0:33:35.7 RD: Exactly.

0:33:36.3 SC: Find these simple rules that can build up to something complex, than just be complex.

0:33:42.1 RD: Yes. Now, and natural selection only sees the final phenotype.

0:33:46.8 SC: Right.

0:33:47.3 RD: Natural selection sees the behaviour that emerges or the morphology that emerges. Natural selection doesn't see the genes.

0:33:54.8 SC: Right.

0:33:55.5 RD: But it's the genes that get favoured in terms of the numbers of them that get through to the next generation. So the phenotype is the emergent property by which the genes are selected as this kind of proxy for the genes.

0:34:13.2 SC: I know there's a lot going on when we mix up our genome or when our genome gets altered through the generations. Part of it is just sexual reproduction and sharing. There's very tiny details about horizontal gene transfer at some level and, of course, that there are mutations. Is there some feeling for, like in a modern, relatively mature species like birds or human beings, the relative importance of these different factors in changing our genome through the generations? Is it sexual reproduction doing most of the work in finding good combinations, or the mutations?

0:34:53.6 RD: That's a good question. Again, I hadn't thought of it that way. I don't think so. I could be wrong about that, because not all animals have sexual reproduction.

0:35:05.5 SC: Sure.

0:35:07.5 RD: And bacteria have something much more rudimentary than that. Bacteria do a kind of cut and paste of bits of genome in a sort of rather haphazard way. Whereas in eukaryotes, it's become ritualised into the form of meiosis and sexual recombination. But, once you've got proper sex, once you've got meiosis and proper sex, proper recombination, I'm not sure it's possible to easily answer that question. Mutation is the ultimate source of variation, but sexual recombination is the more proximal way in which it appears.

0:35:49.0 SC: Right. I guess what I have in mind is this very vague feeling that I would like to make more quantitative, that the space of all possible genomes is way bigger than we'll ever reach in the history of evolution, right?

0:36:01.3 RD: Yes.

0:36:01.7 SC: Somehow, there's places that we can easily get to in the space of genomes, and other places are just terra incognita. We'll never get there. And I'm wondering, to what extent we understand the difference between where we might get to and where we never will.

0:36:18.8 RD: Yeah. That is fascinating. And I've tried to explore this a little bit, with very simple computer simulations. You could say this something like this. I love the concept of the space of all possible genomes. And the vast majority of that multidimensional space is terra incognita, as you say. So, this is fundamentally why major mutational steps, major leaps in the hyperspace of all possible genomes is almost bound to be to leap to death.

0:37:01.7 SC: Yeah. [chuckle]

0:37:03.0 RD: It's got to be a walk through proximal regions of the hyperspace. And this is why RA Fisher made a nice analogy with the focusing of a microscope. He said, "A well adapted animal is like a microscope which is in focus." Now, if the microscope is not quite in focus, then it requires only a very small tap to the microscope to make it either... To get better or slightly worse. So this is Fisher's argument for why only very small mutations are evolutionarily important. A large mutation is equivalent to bashing the microscope. So it either goes way, way out of focus one way or way way out of focus the other way. Assuming it's already nearly in focus, which it must be in order for the parent generation to have survived. If the mutation is changing something in the child generation, then it's gotta be a small change. Otherwise it's equivalent to moving the tube of the microscope by a whole inch, which is bound to make it worse. So the space of all possible phenotypes has been most vividly because simply explored in snail shells...

[chuckle]

0:38:44.1 SC: Okay.

0:38:46.1 RD: Because the way a... Not just snail shells actually, but shells of other creatures which are not molluscs. It's a tube that unrolls. The snail shell grows, the young snail grows at the margin. And it's a tube that coils as it grows. And a palaeontologist called Raup quantified this by saying there are only three basic variables that determine how this tube unravels itself. There's the rate at which it expands, there's the rate at which it moves out of the plane. If you think about something like a turret shell which goes out of the plane.

0:39:39.3 SC: Sure.

0:39:40.2 RD: An ammonite doesn't go outta the plane, an ammonite stays with within the plane. So that variable is zero, whereas a turritella, one of those things that looks like a top.

0:39:54.1 SC: Oh, yeah. Okay.

0:39:55.1 RD: Has a high one. So that's that one. There's the rate at which the... The rate at which the diameter expands as the tube grows, and then there's a third variable. So just these three variables, what that means is that all possible snail shells will fit in a cube. Just a three dimensional space.

[laughter]

0:40:10.8 SC: A conceptual cube. Yes. Right.

0:40:15.7 RD: Which is beautiful.

0:40:16.7 SC: Yeah.

0:40:17.1 RD: And so Raup actually plotted this and I did a computer version of it as well. And there are great areas of this cube, which are never visited by nature. They are terror incognito as you say.

[chuckle]

0:40:33.9 RD: And there are sort of corners of the cube which are actually populated by real shells. So something like an oyster or a mussel has a huge expansion rate, something like a turritella has a very slow expansion rate. Well, that model, the Raup cube model is something you can generalize conceptually into a hyper cube where it's not just three variables, but dozens of them.

0:41:06.4 SC: Yeah.

0:41:06.8 RD: But it's still the same principle. You can imagine the same principle working if you could do the... You can't visualize it in three dimensions, but you have to think of it in multidimensional space. And evolution, however many dimensions you've got, evolution must be a walk through neighboring areas, you cannot suddenly jump from one part of the hyper space to the other, because that that's almost bound to be totally non-viable.

0:41:35.4 SC: And maybe this is a good way of thinking about convergent evolution, in some sense. I mean, there are very different genomes that can give effectively similar phenotypes. And we see that in the case of flying, right? Lots of different species have developed wings in very different ways.

0:41:52.1 RD: That's right. Well, flights have evolved four times in insects, pterodactyls, birds, and bats. And in different ways in all cases. Therefore, three different principles. But the physics is the same, so the physics of flying is the same whether you're a pterosaur or a bird or a bat. But bats do it by having greatly elongated fingers with membrane stretch between the fingers, pterosaurs do it by having one elongated finger, just the fourth finger, the ring finger. Birds do it by having the whole arm with feathers hanging from it. But once you've dealt with that fundamental difference, the physics is the same. And so you could say it's convergent. There are much more striking examples of convergence. I mean the whole of Australian mammal fauna shows beautiful convergences to the mammal faunas than the rest of the world. And as you know... Yes, so convergence is a wonderful thing and it's great testimony to the power of natural selection starting from different starting points, you end up with the same thing because the functional needs are the same and because the physics is the same. The eyes have evolved many, many a times for the same reason. The physics of forming a usable image using straight light rays is the same whether you are an insect or a prawn or a bird or...

0:43:29.4 RD: Or an octopus... Compound eyes use a totally different kind of physics, but octopus eyes and vertebrate eyes use the same physics... It's the same physics as a camera, which produces an inverted image, whereas the insect compound eye... Just a whole lot of tubes sticking out in different directions. It's a sphere... A hemisphere with lots and lots of tubes. And so depending on upon which tube the lights coming from you can tell what's going on. So in that case, in so far as there is an image, it's not inverted, it's the right way up. But anyway eyes have evolved dozens of times independently. Convergently, in many cases, in the animal kingdom.

0:44:17.6 SC: Isn't there... I forget the name of the species in Australia... But there is something that looks quite like a dog and acts like a dog and plays the role of a dog, but it's closely related to kangaroos.

0:44:27.4 RD: Alas, it's extinct. You're thinking of Thylacinus, the Tasmanian wolf. It hung on in Tasmania, went extinct in Australia long ago. The Tasmanian wolf survived in Tasmania up until the last century. And the last one went extinct... Died in 1936 in Hobart Zoo. It's a great tragedy... It was thought to be a menace to farmers. It was thought to worry sheep, and so they put a bounty on it, and it was hunted to death. And oh, if only, I would love to see it. There's a film, a movie of... An old black and white movie of the last surviving thylacine. And it looks just like a dog, behaves just like a dog. It's got a tail sticking straight out the back, unlike a dog's tail. It's got stripes unlike a dog. And it's got a pouch like a kangaroo. But the behavior is just like a dog. You could imagine patting it and saying, sit up and beg. And it looks just like a dog, because it does the same job as a dog. It was a hunting animal, like a dog. And the skull is almost identical to a dog. There are one or two telltale signs. It's a favorite trick in zoology exams to give a thylacinus skull and they think it's a dog skull.

0:45:55.5 SC: I guess there's different angles you can take on this. I recently talked to Arik Kershenbaum... I don't know if you know him, but he... A zoologist who's thinking about what aliens will be like. And he tries to make...

0:46:08.7 RD: Oh, yeah. Yeah.

0:46:08.8 SC: The case where they won't be that different maybe because of these forces of convergence that we see here on Earth. In contrast, I had my evil twin, the biologist Sean Carroll, on the show, and he's emphasizing the role of contingency and chance, and randomness and unpredictability. Where do you come down on the question of if we ran the tape backward, if we evolved again, how would our ecosystem look like or is it...

0:46:33.3 RD: Yeah, I'm on the side of convergence, so I find... I think that the actual natural selection is so powerful. On the other hand, of course, the physics is not... That's sort of going be the same. If gravity is stronger or weaker, then that will have entirely predictable effects. If gravity is stronger then you will expect that a spider will be built like a rhinoceros. A great big... And conversely, if gravity is weaker then the rhinoceros will be built like a spider, because the... Simple physics, you can work out why that would be... But if the physics was the... If the gravity was the same, and if there is light... There's gotta be light, because it's gotta have a source of energy. But if it's not shrouded in perpetual fog, say.

0:47:29.4 RD: So that light rays can be used to form images, then I would bet my shirt that you will get eyes like... And probably compound eyes and camera eyes, because these are just... Oh, the parabolic reflector is another way of doing it, of course. And there is one animal at least that uses that principle.

0:47:55.8 SC: Really, what is that?

0:47:56.6 RD: Yeah, scallops. Scallops. I think limpets too. I'm not sure about limpets. How you use the reflective principle. But there are not that many ways to form an image, and so assuming that there is light and there's no fog, and if they're not living in perpetual mud or something like that... So that you can actually... Light rays are available... Straight light rays are available, then you will get eyes. Assuming that there's an atmosphere, there must be maybe... Well, yeah... Okay, assuming there's an atmosphere then you'll have ears. Sound waves can... Then, will you get wheels? Well, the wheel perhaps depends upon prior invention of the road, or before you can... Wheels don't much go down a plowed field.

0:48:50.4 SC: Yeah, is the absence of wheels an example of the fact that even though it's pretty straight forward to imagine the usefulness of half a wing, it is not straightforward to imagine the usefulness of half a wheel.

0:49:03.0 RD: It's quite difficult to imagine getting a blood supply and nerves getting past the axle.

0:49:08.4 SC: Well, it could be bone. I don't know.

0:49:10.8 RD: Yes, it's not inconceivable. I think that major difficulty is part of it. And I think also it... As I said before, wheels don't work on rough terrain. So if the planet is hard, solid, flat rock, then I think there's a better chance of something like wheels evolving.

0:49:35.2 SC: Well, it's interesting that you mentioned... Eyes evolved many times, wings did not evolve that many times. You said four different times.

0:49:42.3 RD: Four.

0:49:43.7 RD: Yes, proper powered flight, which goes on indefinitely as opposed to just gliding downhill. That's only evolved four times.

0:49:53.4 SC: And is the insect way of flying, even though they use wings, is it very different? I mean, I can see the relationship in the other ones.

0:50:00.8 RD: Well, yes. The insect wing is not a modified limb, unlike all vertebrate wings, which are modified from a limb, from the arms. One dinosaur with legs too. But with insects, it's an extension of the thorax; it's an outgrowth of the thorax. So it doesn't use up the limbs, and that's a good idea because it means all the legs are available for running. Insect flight muscle is interesting. Most insects have a flight muscle, which is an oscillator. Instead of having a down-stroke and an upstroke that separate like vertebrates do and like some insects like dragonflies do, insects like bees and flies have a flight motor, which is a... It's a kind of high-speed shiver. So it's either on or off. All the nervous system says is switch the machine on...

0:51:04.0 RD: Oh, okay.

0:51:05.6 RD: Or switch it off. And the actual frequency with which the wings beat is mostly determined at the resonant frequency of the wing rather than being under the control of the nervous system. So that's a major difference, but the aerodynamics is probably similar. Some insects are very good at hovering, and not very many birds are. Hummingbirds are, and there's one, Kingfisher, which can hover properly. And they do it by a kind of figure of a sculling motion. So hover flies are brilliant at that, and hummingbirds are too; they too use a kind of a figure of a sculling.

0:51:49.7 SC: Am I gathering from what you're saying that flight only evolved once in the insect kingdom?

0:51:55.3 RD: I can't be sure of that.

0:51:57.6 SC: But as far as we know?

0:51:58.4 RD: I can't be sure of that.

0:52:00.2 SC: Yeah. Okay.

0:52:00.3 RD: No. It may be more than once, and I don't think we know. Possibly, the fact that some insects have this oscillating muscle and other insects have an up, down, up, down, up, down flight mechanism, that might suggest two different independent evolutions of flight.

0:52:21.3 SC: Okay, interesting. Yeah. I guess once you have a thorax... It's weird that the thorax is where it came from, but once there is that capability...

0:52:29.2 RD: Yes. They kind of do it by... In some cases, the muscles actually attach in such a way there's a hinge and it pulls up and down. But in many insects, it's just a deformation of the thorax which does it. The muscles kind of pull on the thorax and the changing shape of the thorax causes the wings to move.

0:52:50.8 SC: And I suppose I should give you an opportunity to talk about some of the quirkier ways that animals have developed flight. There are things that just float and certainly, there are things that glide that you already mentioned.

0:53:03.6 RD: Yes. The book, of course, is also about human flight. And humans also use balloons, use the lighter-than-air principal; balloons and airships; dirigibles. And I don't think I speculate in the book about whether any animal does that, I don't think they do. It's common place to use it in water, and Teleost fish have this beautiful thing called the swim bladder which enables them to change their position of hydrostatic equilibrium; they have a bladder inside. And by altering the amount of gas in the bladder, they can rise or fall in the water the same way a balloonist does, by changing the aerostatic point of equilibrium. But I don't think any animal uses hydrogen, for example, to fly with.

[chuckle]

0:53:58.4 RD: And they could, but it doesn't seem to have evolved.

0:54:01.8 SC: I presume this is just a resource constraint, right? There's not that much access to lighter-than-air gases, whereas there's plenty of access to lighter-than-water things.

0:54:12.1 RD: Well, I guess. Nature can be very ingenious. There is hydrogen in... Well, there's methane, which is lighter than air.

0:54:25.9 SC: Okay.

0:54:27.1 RD: One of the things we have to worry about with global warming is methane produced by cows.

0:54:31.3 SC: Too much, yeah.

0:54:33.0 RD: So animals can make methane, and do, as a by-product. And so, could they make a bladder of some sort to hold it?

0:54:42.1 SC: Yeah.

0:54:42.1 RD: Well, insects and spiders have silk and silk is very good stuff for making balloons with them. Man-made balloons are somewhat made of silk. So you sort of feel that ingredients are not totally absent, but they don't seem to have been put together.

0:55:03.0 SC: You probably know that Carl Sagan speculated about floating gas bags in the atmosphere.

0:55:07.3 RD: I think that's Jupiter.

0:55:07.5 SC: Jupiter, yes. Right.

0:55:08.5 RD: Yes.

0:55:08.8 SC: 'Cause there's no solid ground to land on.

0:55:10.4 RD: Yes, that's right. Yeah.

0:55:11.4 SC: So there might have been easier.

0:55:13.3 RD: Yes. Yes, the atmosphere of Jupiter would be very dense and therefore it will be a little rather like swimming in water. It'd be more like fish having a swim bladder in water on this planet, I suppose.

0:55:29.0 SC: I think maybe the answer to this question is no, but are there examples of animals that fly, in some sense, by making artificial enhancements to their physiology, by like picking up a leaf and using it as a glider?

0:55:44.0 RD: Oh, yes. I think so. Let me think about that. No, I can't put my finger on it, but it wouldn't surprise me if spiders do that. When a spider is about to start building a web, it needs to have one...

0:56:02.1 RD: Line, one string to start it off with. So what it does is to... In some spiders at least, it releases a thread of silk with a little kite on the end. And that little kite floats, and when it happens to hit some, say, a bit of a tree or something, it sticks there. And that gives the spider something it can run along. And then it's a way. Then it could start building its web using that main gye [0:56:32.7] ____ guide rope to start with. So that's kind of flying a kite. Plenty of animals float. Well, plenty of spiders, for example, do a thing called kiting, where they float into the aerial plankton up in that high atmosphere using their own silk as a kite.

0:56:55.1 SC: Am I going...

0:56:55.6 RD: Yes, so that would be a good example, would be kiting in spiders.

0:57:03.2 SC: Am I making too much of a leap to connect this to the whole idea of the extended phenotype? The idea that our genes affect not only our bodies but parts of our environments, and that's gonna...

0:57:13.4 RD: No, I think that would be rather good. I wouldn't mind doing that.

0:57:17.3 SC: Okay. Good.

0:57:18.1 RD: That would be fine.

0:57:19.3 SC: Why don't you explain to the audience what the idea... Your idea of the extended phenotype is? I mean, it's an enrichment of, I think, how we think about how evolution acts on us.

0:57:29.9 RD: Yes. Well, we normally think of genes as acting on phenotypes, which are part of the body in which the gene sits. And so this is commonplace. The wing of a bird is part of their phenotype, and so natural selection works on genes in the bird which improve the aerodynamic efficiency of the wing. That's commonplace phenotype. Something like a bird nest, on the other hand, is not part of the bird's own body. And yet if you look at the shape of a nest, it's clearly a Darwinian adaptation. The shape of it has been shaped by natural selection, which must mean that there are genes for nest shape. In the case of some weaver bird nests, the nests are very elaborate. Ovenbird nests are very elaborate. And so these must have evolved in slow, gradual stages just like the body of the bird evolves. And yet this is not part of the bird's body. So we have genes that have an extended phenotype. The nest is influenced by genes, must be. It's an extended phenotype because it's outside the body of the bird. Artefacts like that are the most obvious example of an extended phenotype.

0:58:48.9 RD: But I then carry the argument further once you've, as it were, bought into that. Then parasites which influence the behaviour of the host in which they sit in order to make that host more likely to be eaten by the next host in the parasite's lifecycle. Something like a fluke, which is a little flatworm which, so to speak, wants to get into a sheep. And it is in the intermediate host, which is a snail. So it needs the snail to be eaten by a sheep. And it makes the... Or there is another one which sits in an ant, and it needs the ant to be eaten by a sheep. So what it does there, it's called a brain worm. It burrows into the brain of the ant and makes a lesion in the brain of the ant, exactly as a physiologist, a neurophysiologist might make a lesion in a brain, which causes the ant to change its behaviour to make it more vulnerable to being eaten by a sheep. It makes it climb to the top of grass stems instead of being good at going underground, which is where an ant would normally want to be, in the heat of the day at least.

1:00:02.9 RD: And so the brain worm, this fluke, this worm inside the ant is changing the behaviour of the ant in order to get itself into the next host of the lifecycle, which is the sheep. Well, the change of the ant's behaviour is extended phenotype of the fluke's genes. It must have evolved by natural selection, which means there must be genes in the fluke for changing the behaviour of the ant. It's extended phenotype not direct phenotype. And then, the next step in the argument would be just to take a parasite which doesn't live inside its host, like a cuckoo where the cuckoo nestling sits in the nest and manipulates a foster parent of another species into feeding it. And again, the change in the behaviour of the foster parent, since it's obviously favoured by natural selection, is extended phenotype of the cuckoo genes.

1:01:08.8 SC: It makes it almost mind boggling in the complexity of trying to analyse this, right, because we can easily see how genes code for proteins. And then you might wanna say, "Well, okay, these proteins help or hurt in some way. They have some functions." But in fact, there's this very complex interplay of the environment it's in and different levels at which we can talk about things.

1:01:31.2 RD: Exactly.

1:01:31.8 SC: This is why it's way easier to be a physicist.

1:01:33.4 RD: It's just another step. It's just one more step. I mean already we start with protein, the gene-inference protein.

1:01:39.0 SC: Yeah.

1:01:39.3 RD: Which changes the behaviour of cells in the developing embryo, like we were talking about earlier, which changes the formation of tissues, which changes the behaviour in the case of the nervous system. And then one more step changes the nest-building behaviour, which changes the shape of the nest. It's just one more step at the end of this long, long chain of causation which starts with protein and goes on through all the different steps in embryology.

1:02:05.7 SC: Well, and then all the way to societies or the way that we technologically alter our environment, right? I mean these, then, are features that come into how our own genomes evolve.

1:02:18.6 RD: I hesitate to go that far.

1:02:21.0 SC: Okay.

1:02:21.7 RD: And the reason is this. Although it's true to say that you could do a genetics of bird nests. It hasn't been done, but not the slightest doubt you could do it because, as I said, these things have evolved by natural selection. That means there must be genes affecting the shape of a bird's nest. But in the case of a human artefact, like say a building. It's not like that. I mean, there's no gene for Romanesque architecture as opposed to Gothic architecture.

1:02:56.9 SC: Right.

1:02:58.8 RD: There might be a gene that makes an architect a good architect, or a bad architect, because he's a good draftsman or something like that. But, I'm almost confident in saying that there'll never be a gene that affects the actual shape of a building.

1:03:14.3 SC: Okay. So...

1:03:14.9 RD: Where there is a gene that affects the shape of a bird's nest.

1:03:18.2 SC: Yeah, no... Okay. I completely get that a 100%, but then, so let me then push back on the idea of putting the gene at the center of everything, for exactly this reason. Once we have knowledge that we can pass down memes, if you would like to call them that, but you know, other whatever forms of information that persists over time and affects our behavior aren't we pushed to consider the co-evolution of those things. Isn't that, isn't that almost an inescapable way of thinking about it?

1:03:55.1 RD: I think so. But by introducing memes, you've... As it were made that work, because they actually could be replicators, they could actually, they could be natural selection, not genetic selection, but memetic selection of memes. And there would then be co-evolution as you say, between genes and memes. And I should think that could happen. I wouldn't rule that out in the way I would rule out extended phenotypic effects on styles of architecture. I think I could imagine memetic selection for styles of architecture and an architect sees a particular way of decorating the tops of pillars or something. "Oh, all that looks neat. I'll copy that."

1:04:44.9 SC: Right.

1:04:46.2 RD: And so that could spread in a kind of quasi genetic way and that I get, that would be nice. I mean, and I bet that happens with all sorts of things like the, well, Desmond Morris looked at the style of painting boats in Malta, but fishing boats have designs written, have painted on them or the figureheads on the front of sailing ships, probably might have been imitated.

1:05:16.2 SC: Sure.

1:05:17.0 RD: That would be memetic selection. And I could buy into that.

1:05:21.0 SC: But I think again, correct me if I'm wrong, but you are skeptical of attempts to analyze selection of the level of groups or something like that. Right? I mean, I'm not at all familiar with all the details. But I know there's a long running debate about gene selection, kin selection, group selection, explanations of altruism, et cetera. Do you think that there's actually... Do you think that debate is healthy and ongoing or we basically know the right answer?

1:05:47.5 RD: Well, you are talking to me about this and I'm a partisan.

1:05:50.8 SC: I am. Yes. Go ahead.

1:05:54.1 RD: I think we do know the right answer but if you talk to somebody else they would disagree.

[laughter]

1:06:00.6 SC: Sure. So what is the right answer?

1:06:01.5 RD: Fundamentally, it's all gene selection.

1:06:03.5 SC: Yeah. Okay.

1:06:05.0 RD: And I make a distinction between replicators and vehicles and an organism is a vehicle. And there is a sense in which you, because the vehicle, because the organism, the animal is the thing that actually does things, it's a thing that actually hunts or escapes or mates, or it's a thing that has sensory organs, it's a thing that has limbs, has hands, feet. So the vehicle is an important unit of evolutionary agency. But it's not what's naturally selected. Because it dies. The only thing that goes on to the next generation and the next and the next and the next is the genes. And so, the unit of selection in that sense is the gene, the unit of selection in the vehicle sense is the individual, it's arguable. And some would argue that the group can be a vehicle in that sense, that would have to mean that a group evolves adaptations for the propagation of the genes that make that adaptation. Individuals clearly do. The shape of a nose, the shape of a wing, the shape of a tail these all affect the individual's ability to pass on genes. Somebody might argue it. I wouldn't, somebody might argue that groups also are vehicles in that sense that there might be a property of a group, which affects that group's ability to pass on genes. I can't see how that would work.

1:07:43.5 SC: Right.

1:07:43.9 RD: It doesn't seem to me to be plausible. And so I don't buy into group selection.

1:07:50.6 SC: Yeah. I think... Again, I have no dog in the fight as it were. But it makes sense to me that the information that is being passed down is contained in the genes in some sense. So there's something special about them. In that sense, if we aspire to a general theory of replicators, is there something new that comes along when you have memes or when you have...

1:08:12.9 SC: Yes.

1:08:13.0 SC: Knowledge?

1:08:13.6 RD: I mean, exactly.

1:08:13.8 SC: That's the other thing. Yeah.

1:08:16.5 RD: I think the value of the meme idea is not so much as a contribution to the study of human culture, although it may be, but rather it's a way of saying it doesn't have to be genes, any replicator would do. And if you cast around and try to think of an alternative replicator, a computer virus would be another example 'cause it's self replicating and a meme is another example. It's self replicating, whether there really is natural selection of memes is an open and interesting question.

1:08:50.9 SC: Right.

1:08:51.6 RD: And I suspect there probably is, but it may not be very evolutionarily significant. Whereas you can look at an animal and say everything about that animal I all about increasing survival of the genes that made it. Could you ever do that for memes? Maybe you could. It's not out of the question. At least they are bonafide replicators.

1:09:14.3 SC: So here's the final topic I wanted to get on the table here. Just to change gears a little bit, but not actually too much. You were very kind recently to say nice things about my book, "The Big Picture" on Twitter. Thank you for that. But you couldn't resist the temptation to take a jab at philosophers along the way while you did it saying that, you...

1:09:35.4 RD: Did I. I guess, I had forgotten that.

[laughter]

1:09:36.3 SC: Something about how I was discussing philosophical things, but in a much more clearer way than actual philosophers would do. But I think, the discussion we've just been having for the last 10 minutes or so, on levels of selection and things like that, this is the kind of arena in which it seems to me, that philosophy has a role to play that, there are scientific questions and also general structural questions about emergence and levels and logic and reasoning and so forth. So I would like to see friendly cooperation between the scientists and the philosophers. Whereas you seem to be a little bit more skeptical that they're gonna be of any help.

1:10:17.9 RD: I like to talk about this because you are one of the few people I know who can do both.

[laughter]

1:10:25.3 RD: And in my naive way, I feel that what philosophers, if they are good ones do is well, is think clearly and help other people to think clearly. But thinking clearly is what I hope we all do anyway.

[laughter]

1:10:41.6 RD: And so, what I don't quite get is why you need to get a degree in philosophy where you study Aristotle and Locke and Kant and, it, before you can think clearly and clearly many people do think clearly. On the other hand, when I attended that seminar in wherever it was, is somewhere in New York state was it that you organized in?

1:11:13.4 SC: In Massachusetts. In Stockbridge, yeah.

1:11:15.3 RD: Yes, that's right. I was totally out of my depths.

[laughter]

1:11:19.7 RD: Because I couldn't follow the jargon. And I am familiar with interacting with many of the scientists there, like, Dan. Well, I think of Dan Dennett as a scientist.

1:11:33.4 SC: There you go.

1:11:34.5 RD: Or Jerry Coyne or... And when I talk to Dan Dennett, he's such a clear thinker and I like to think I am, we can talk together, but without having to have recourse to quoting the philosophers of history. And without the use of philosophical jargon. So I'm fully ready to be convinced that there is more to philosophy than just thinking clearly, and that it is actually necessary to deploy the jargon of philosophy in order to do the job thinking clearly. And you can convince me of that. I just feel completely naive about that matter.

1:12:23.1 SC: Well let me just say something very briefly and hear your response to it. I think that...

1:12:26.2 RD: Yeah.

1:12:29.0 SC: Look, there's plenty of scientists who can't seem to say a sentence without devolving to jargon very quickly. Also, I wouldn't necessarily blame just philosophers for that. But what I think is in the modern academy in universities it's more of a stylistic question of what questions we consider important. I had Dave Reich on the podcast talking about the genetics of ancient humans. And I said what is a species or something like that? And he says, oh, no, I can't talk about that, that's philosophy. And I get it. I get exactly...

1:13:01.8 RD: Oh.

1:13:02.2 SC: In that context.

1:13:03.6 RD: No That's terrible.

1:13:04.3 SC: Well, yeah, so I think that it's less about the substance. What that means that they're doing in philosophy is they're rather than making up a model that makes some prediction for some experiment that we can do in the next five years. They're thinking about foundational questions about what these terms mean, which are unnecessary, what we can get rid of. And I think that style of thinking can... In the right parts, it's not necessary for most of science, but for some kinds of science, that's kind of really useful.

1:13:34.5 RD: I think it's totally important. And I think that's what we all ought to be doing anyway. So we can't talk sensibly unless we do define our terms and find and decide what it is we are talking about, you are of course right that scientists do use jargon all the time, but in a way that's kind of inevitable. If you're a physicist, then you... If you are a quantum physicist, then you are using the jargon of quantum physics, which is incomprehensible to people outside, and that's necessary. You can't help that. Similarly, if you are an immunologist you use jargon. If you are a molecular biologist, you use jargon, you are talking about real things which actually are there in the real world. And they have got names like Cistron and Operon and that kind of thing, you actually do need those. The jargon of philosophy also seems to have a whole dictionary of names. And I don't know what that, what they're for.

[laughter]

1:14:42.6 SC: Fair enough.

1:14:43.6 RD: That's just what I'm saying.

1:14:45.6 SC: I think that's a good place to...

1:14:46.1 RD: I don't... It's probably isn't fair enough. I mean I would genuinely like to be convinced actually.

1:14:50.9 SC: Good. This is a longer conversation. We should have it sometime over... In a pub someplace.

1:14:55.6 RD: Yeah, I think, yeah.

1:14:57.0 SC: But, so let me then just finish up final question, give us your feeling about the state of evolutionary theory, the modern synthesis and so forth. I know there's some people who would argue that with improved understanding of epigenetics and other factors that we have replaced the modern synthesis with something better. Do you think that we... The picture of modern natural selection Neo-Darwinism, the modern synthesis is more or less correct? And it's a matter of firming up the details or are there revolutions to come in that basic understanding?

1:15:33.5 RD: Okay. I think firming up the details is doesn't really do justice to what's necessary because the molecular biology revolution is so profound. And what it means is that the coding of biology is digital, profoundly digital. I mean, we already knew from Mendel that it was kind of digital. I mean, Mendelian genes are digital. They don't blend. But the DNA of Watson and Crick is digital to the umpteenth degree. It's exactly like computer. I mean a chromosome is just a really great computer tape. It's quaternary rather than binary, and that does have found implications. So although it is in a sense firming up the details that covers a multitude of important, interesting things. So I think that the modern synthesis is correct. But I think that people who try to say the modern synthesis is dead they're over reaching themselves, but the modern synthesis, what that had over Darwin was the idea that evolution is changes in gene frequencies in populations.

1:17:03.5 RD: Those were Mendelian genes, but we now know what they are and can express them in digital form. And we can, for example, do taxonomy working out the family tree, the relationships between all living creatures, animals, plants, fungi, bacteria using quantitatively precise counts of the numbers of DNA code letters which they have in common. It's just like with the same kind of precision as if you have alternative versions of the book of Isaiah or something and then you can compare them letter by letter. So this is a huge advance and understanding evolution at the level of DNA, is nothing in the modern synthesis is contradicted.

1:18:01.6 SC: Sure.

1:18:01.8 RD: But the detail is, the detail is huge and fascinating, and Darwin would be thrilled.

1:18:09.7 SC: I think he would be. I think that future generations are gonna have a lot of fun things to add to the whole story. So Richard Dawkins, thanks so much for being on The Mindscape podcast.

1:18:17.8 RD: Thank you very much, indeed.

[background conversation]

[background conversation][/accordion-item][/accordion]