Transcript: Robert Sapolsky on Behavioral Evolution II at Stanford

Stanford professor Robert Sapolsky discusses Behavioral Evolution II in detail at Stanford. In this evolution lecture, he focuses on individual and kin selection, behavioral logic, competitive infanticide, male/female animal hierarchies, sex-ratio fluctuation, intersexual competition, imprinted genes, sperm competition, inbred-founder populations, group and multi-level selection, and punctuated equilibrium. This lecture presentation occurred on April 2, 2010.

Listen to the MP3 Audio here: robert-sapolsky-on-behavioral-evolution-ii-at-stanford


Let’s get going. Various announcements, procedural things. A number of people want more information about grading and what the exams are like, all of that. I think I mentioned one-third of the points will come from the midterm, two-thirds from the final. In terms of the style of the midterm, the midterm is heavily going to be about making sure you got down all the factoids from the first half of the course, that you’ve got the basics of each of our proverbial buckets. The second half of the final is all going to be about integration, thinking across the different categories. So just a sense of that.

Readings. Readings, as they are coming out, the books are not required until the second half of the course. The handout on Monday, I think, said which chapters of the Zebra book you should read. We will shortly get to you which chapters of the Chaos book you should read as well. The readings that are being posted on the CourseWorks, the downloads of various published papers, those are required. I’m clear on those whether this is a paper you should read all of, if this is one you should read the abstract of. Even if you read all of, do not read it in some obsessive, detail-oriented way. The goal is probably to be able to say, in one or two paragraphs, why this paper has something pertinent to say about the topic they fell into. You’re not sitting there having to memorize techniques, middle names of the authors, how many animals, anything like that.

In terms of that, it probably makes sense to read those after the first lecture of whatever block there is. And hopefully, if I get organized, I’ll be able to get you sort of a list of the readings further in advance than one week in advance. Nonetheless, you should probably hold off reading it until after the lecture occurs.

Let’s see. What else? People wanted to get a sense of how long things were going to go. And as we’ll see today, the evolution lecture topic will cover two classes. Molecular genetics, which is what we’ll pick up on Monday, I’m guessing one to one and a half. Behavior genetics following that, one to one and a half, ethology one. Neurobiology, endocrinology, we’ll have one week devoted to intro to the topics. And again, that’s one where this is so important for everybody to be up to speed rather than these being in catch-up sections that week. The whole week, Monday, Wednesday, Friday, will be devoted to that with the TAs teaching it. The following week, three more lectures, more advanced ones. And depending on proximity to the midterm, there may be a half-lecture in there on statistics, or maybe not. So this is going to depend on keeping on schedule. This is a rough approximation. The midterm is going to be a Monday night. You will be responsible for material up to the previous Wednesday, and there will be lots of review stuff. Take a look at the extended notes being posted.

What else? OK. I think that covers most of the procedural stuff. All of this stuff will get posted as well.

So picking up on the other day. What was happening the other day? Number one, the trouncing of Darwin inventing evolution, trouncing of survival of the fittest, probably most importantly, trouncing of behavior for the good of the species, group selection-type arguments. What we saw was, number one, the rationale for the whole thing. There is a vicious, un-fightable logic to why hearts have to be the size they are and kidneys have the filtration rates they have to have in order to solve the challenges of leaving as many copies of your genes in the next generation. And making sense of the evolution of hearts and kidneys and things like that could be the worlds of bioengineers and biomechanics folks with an underlying logic that it’s got to be something that increases the number of copies of genes that you leave.

And the whole rationale for Wednesday’s lecture and today is applying the same sort of logic to behavior. The whole world of just as you can optimize sort of the way one’s neck, how long it is if you’re a giraffe, you can optimize behavioral strategies. And again, also throwing in a caveat, no animal is sitting there, maybe with the exception of some other apes, sitting there consciously strategizing along those lines. One saying, so what would you, as this dandelion, want to do at that point with this ecological challenge, personifying just to make things easier.

What we then barreled into were the three major building blocks for thinking about the evolution of behavior in the framework of contemporary evolutionary thinking. Number one, individual selection — passing on as many copies of your own genes to the next generation as possible by way of your own reproducing. The individual selection — a chicken is an egg’s way to make another egg. I’ve now rehearsed that, so I’ve got that down right. The whole notion of behavior as just being this epiphenomenon in order to do what’s needed to get another copy of the genes into the next generation.

Building block number two. Some of the time, the best way to increase the number of genes you pass on to the next generation is to help your relatives do so following that logic of Mendelian relatedness. And people in the catch-up section, I know, went over issues of, why is it that you share half your genes with a full sibling, a quarter with a half sibling, et cetera? So sometimes the way to maximize is by helping out a relative to do so with, again, constrained by this vicious mathematical logic of, it depends on how related you are to the relative. And thus, you will gladly lay down your life for one identical twin, two full siblings, eight cousins. Off you go.

So the whole notion there of insight into why social animals the galaxy over are so obsessed with kinship and relatedness, the whole world of who counts as an us, who counts as a them in terms of cooperative behaviors playing out along lines of relatedness.

Finally, we saw the third piece, which was reciprocal altruism. You scratch my back, I’ll scratch yours. Many hands make the task less scratchy or whatever. And what you see in those cases, there is a whole world in which you don’t have to be related to have cooperation. And we saw all the domains of that bringing in the formalization — biomechanics person figuring out how strong a leg bone has to be. A game theorist figuring out within the realm of social behavior when you cooperate and when you don’t and what sort of strategies. Game theory, of seeing the prisoner’s dilemma as the building block of that entire field and seeing all the strategies worked out by mathematicians and economists and diplomats and seeing which ones optimize under what circumstance. And then going and look in the real world and seeing there’s all sorts of animals out there that have evolved optimization strategies of when to cooperate and when to defect.

And what we’ll get to as a huge, huge issue by the time we get to the lectures on aggression, cooperation, et cetera is, well, that’s great when you’ve got a cooperative system going. How do they ever start? How do you jump start cooperation in systems? The evolution of cooperation, that’s something we will get to in great detail down the line.

Finally, what we shifted to was now saying, great, we’ve got all these principles in hand here and our big three building blocks and all of that. How would it be applied to making sense of animal behavior out in the real world? And we left with the examples starting someplace there — where did it go, yes, that’s it — of us marching through, know one individual factoid about some species or other species, know that there’s a big difference between the genders in size or there isn’t, know that there’s not high levels of aggression in males, know that females always give birth. Whatever those traits are as we marched through, using these ideas about individual selection, kin selection, reciprocal altruism, you could march through and logically infer what the social behavior of this particular species was going to be like, and you would be right. We saw, for example, in tournament species, tournament species where you have high levels of aggression among males, male-male competition for access to females. As a result, males tend to be a lot bigger than females. They are being selected for muscle mass, secondary sexual characteristics, plumage, big sharp canines for slashing the other guy.

Bringing up this issue of, in a tournament species, who does the male want to mate with? What sort of female are males interested in mating with? The answer being, anyone who will mate with them because there’s like no cost involved. All that’s involved is no parental behavior, none of that, the cost of sperm. And literally, people analyze the relative cost of sperm versus eggs in tournament species. Males are dramatically un-choosy as to who they mate with. In tournament species, males have dramatic variability in reproductive success — 5% accounting for 95% of the matings.

Female choice in a tournament species. What does a female want out of a male? She certainly is not going to get good fatherhood out of the guy. All she wants are good genes because that’s all she could hope for from the guy. A whole world of female selectivity for markers of good genes. And as we’ll see in the sexual behavior lectures, a whole world of males trying to fake out females across the animal world, suggesting they got better genes than they actually do. So we will come to that.

Then tournament species. Do you see female abandonment of kids? Absolutely not, because there’s nobody else who’s going to take care of them. Coupled with that, you see single births among those species, low level. High levels of aggression amongst the males, low levels of parenting. And males tend to live a lot shorter than females.

And what we saw at the other end, the pair-bonded species was a completely different picture. Males are being careful who they’re mating with because the wiring there is you mate, you take care of the kids. High levels of male parenting, females are thus able to abandon them. Females can have multiple births. Males are selected to be as close to females as a male can be. So there’s not a big difference in size. There’s not a big difference in secondary sexual characteristics. There’s not high levels of aggression. There’s not big differences in lifespan.

Who does a female want to mate with? Someone who is going to be a competent father. Thus the whole world of pair-bonded species, where males court with rituals of coming up with food and coming up with things of that sort. So the most striking thing about that was, again, going back to the two skulls I brought in, you could know nothing more about that species other than this is an adult female, this is an adult male. And working through the same logic, you know a huge amount about the private lives of the species and who’s messing around with who in the bushes. And that’s merely by applying these principles.

And likewise saying, here is a picture of a pair in this species and you can’t tell which one is who by gender, that tells you a whole world of predictability at the other end.

So a first example of how much explanatory power you can get out of this. And on the website, I will find all sorts of appropriate pictures of tournament and pair-bonding species for you to look at.

So a first realm of looking at animal behavior and seeing how once we got these principles in hand, individual selection, kin selection, reciprocal altruism, suddenly an awful lot of stuff makes sense. Next domain where that’s the case. And this is one that has had a real challenge to the, “Ooh, isn’t nature benign and animals behave for the good of the species” viewpoint. This one domain probably proves an individual selection framework more than anything else out there.

Back to our inevitable National Geographic specials. And somewhere at the end of whatever the special is, about whatever species, at some point they usually get some narrator with some deep, baritone voice coming at that stage and saying something like, man is the only species that kills for pleasure. Man is the only — and look at how wonderful and benign these rosebushes are. They don’t kill. They don’t have war. Only humans do. And what occurred, by about the mid ’70s or so, was enough field workers from a different species reporting, hey, wait a second. We’re not the only species that kills. And we will see plenty of domains where that makes sense in the lectures to come.

But one particular version really, really demanded some rethinking about animal behavior, which was as follows. What is one of the universals, whether you start with us and go all the way down to slime molds? It’s babies are cute. Everybody likes babies. Babies are adorable. You want to take care of babies. Your eyes dilate as soon as you’re around them. And a longstanding notion that what infants, what babies, what baby features are about are, among other things, are means to reduce aggression. And we will see in the ethology lecture something about that. That was the standard sound bite forever.

And suddenly, you got field biologists coming back from studying various species saying, something’s up there, because in my species I see individuals murdering kids all the time. Whoa, what’s up with that? Infanticide within a species. Suddenly, not only are humans not the only species that kills, we’re like not the only one who goes and kills adorable, little, Disney infants. Suddenly, all sorts of species popping up where there was infanticide. And the huge challenge then became to make sense of this.

This first emerged in the 1970s, studies of langur monkeys in India by someone named Sarah Hrdy reporting this. Soon it was reported among lions, some other species as well. And the first obvious response to this was, wait, this can’t be because I watch all the wildlife specials. There’s something wrong. Oh, there’s some sort of psychopathology going on. This is not a normal population. This is not a normal population because they live close to humans. There’s some disturbance. There’s a toxic waste dump somewhere. They’re not a normal species because you’ve got the wrong color socks on as the observer, whatever it is. And this is not normal. This is pathological behavior.

But long enough time goes by, and it becomes clear in all sorts of species, individuals kill infants. So what is this about? You start to look closely, and there’s patterns. There are patterns to it. The first one being that it tends to be adult males who kill infants. The next being, you look closely, and it’s not random who kills who. It’s males killing infants who are most likely to be the offspring of, you guessed it, individual selection, most likely to be the offspring of other males. Competitive strategies for reducing some other guy’s reproductive success.

So this began to be seen in these species that had infanticide. It was male infanticide of kids who were most likely to be the offspring of other males. But more patterns popped up at that time, which is, well, why don’t you see that in any social species where you have competition? And what wound up being clear after a while is there’s only a certain pattern that you see in species that have competitive infanticide, which is the average interbirth interval among females is longer than the average tenure of a high-ranking male.


What? Here’s what that means. That means you’re some low-ranking guy. You’ve been working out in the gym for years on end. You’re finally in the position to boot out the alpha male and take over the whole group. And damn, every single female in the group has a newborn who they’re going to be nursing for the next two and a half years. So they’re not going to be ovulating, and they’re not going to be ovulating for about three years. And you’re likely to be alpha for only a year and a half or so, where the length of time, on average, that you are going to be able to be reproductively active is shorter than the length of time that females in your species nurse kids and thus are not ovulating.

And suddenly, you have this brutally clear logic that makes perfect sense from everything on Monday, which is, go and kill the kids. Go and kill the kids for two reasons. Number one, by killing the offspring of some other male, you are decreasing that individual’s reproductive success. This competition is leaving as many copies of your genes, et cetera. Number two, by killing an offspring, the female, by stopping nursing, will soon be ovulating. And thus, you see this pattern in langur monkeys, vervet monkeys, patas monkeys, lions, mountain gorillas and such, where it’s always this structure. Competitive infanticide — a male takes over a breeding group and goes about systematically trying to kill the infants.

So we’ve got an enormous violation there of behavior for the good of the species, and baby animals are cute and suppress aggression, all that sort of thing. The clearest demonstration of that is to look at the fact that one of the species where this goes on by this vicious sort of logic of competition, this is one of the species on Earth that is as close to extinction as you can get — mountain gorillas. There’s maybe 600, 700 of them are left. And these are just like stunning animals, and they are disappearing because of habitat degradation, and human presence, and civil war is sweeping through there, and all that sort of stuff. But also they are facing, I think, inevitable extinction in a decade or two because every now and then, a male mountain gorilla will systematically go and kill the infants around in a group by this logic. You can’t ask for a much better demonstration of animals not behaving for the good of the species or the group — this individual selection strategy.

So immediately we start applying some rules to this. Well, when’s a circumstance that a male would come into a group and boot out the previous reproductive male and take over, and where he wouldn’t be all that fast to try to kill all the infants in the group? When would a male take over a group and not be all that thrilled to do this competitive infanticide strategy? Any guesses? [When he’s a brother]. Yeah. Part two, individual selection, kin selection. Kin selection, you don’t see the competitive infanticide if the males replacing are close relatives by exactly that logic.

So then you begin to see some more elegant stuff going on. In some cases, in a bunch of species, instead of males killing infants, what happens is the presence of a new male causes females to miscarry, pregnant females to miscarry. One version of this, and this is seen in wild horses, the new male in there harasses pregnant females to the point of them miscarrying. Same exact logic again. In a whole bunch of rodent species, you see something far more elegant. The new male shows up, and if you were a pregnant female, the smell of a new male causes you to miscarry a litter. And the biology that is completely worked out, going from the olfactory signaling — and we’re going to hear tons about olfactory signaling in weeks to come — going from the olfactory system to release of a stress hormone that disrupts uterine maturation, and you abort. Olfaction-induced abortion.

What’s up with the female with that? Why is that a strategy for her to leave as many copies of her own genes in the next generation? If this guy is around, she’s got a choice at that point, a choice, again, metaphorically. She has a choice at that point. Either she can go through the rest of the metabolic costs of pregnancy, give birth to these kids, and then they get killed. Or at least she can take the lesser of two evils, spontaneously aborts at that point. And soon after, she is ovulating and has a chance again to pass on copies of her own genes. And this is the whole world of — you put a male hamster in with a female who’s just had babies, and he goes and eats the kids. Oh, it’s psychopathological. It’s not psychopathological. Male hamsters are migratory. If there are kids there, they’re not likely to have been mine. That’s the logical thing. Females having been selected for if there is the smell of a new male, miscarry. And the same logic again extending, the smell of a new male, unless he’s a close relative of the previous male. You see the exact same thing playing out there.

Now, all of this, of course, is very sort of male-biased in terms of this. These are females sort of making the best of the bad situation. You’re some female langur monkey, and some new male has come in and is intent on being infanticidal. And you’ve read evolutionary biology also. What do you want to do to maximize copies of your own genes in the next generation? Well, you’ve got this kid here who you’re trying to protect. Would you protect that child to the point of your being killed? Probably not. And you don’t see females doing that.

Under what circumstance would a female be most motivated to defend her child to the point of her being seriously injured? Older females, older females who are less likely to have another child after that. You’re on the scene there, and who else from day one would be most willing to get injured in order to protect the kid? The maternal grandmother, who has no reproductive potential at that point. And what you see in these langur monkeys, the females will defend, not to the point of serious injury. The older the mother is, the more strenuously she defends, and the grandmothers defend even more. Following the same logic here playing out in that way.

But you also have another option, which is to do something much more clever. And this is not in the world after you’ve given birth, but the world of when you are pregnant. Because even in these species that have the competitive infanticide, the new male was also perfectly happy to harass a pregnant female. What have females evolved in a number of primate species? One of the great, all-time ways of manipulating males. So you’re this pregnant langur monkey, and in comes this new male. And you know you’re going to go through like six more weeks of pregnancy, and they’re going to be enormously costly, and the damn guy is going to kill your kids as soon as they’re born. And what can you do and all of that? And you come up with something very clever physiologically. You go into what is called pseudo-estrus, estrus being the term for when female primates are ovulating or sexually receptive, are in heat. And most primate species give external signs of when they’re ovulating. There is engorgement of all sorts of tissues around the vagina and rear end and stuff that humans don’t do. But they have some of the physiological residues, of water retention, for example.

So you’ve got all these external signs of, I am ovulating, except you’re not ovulating because you’re pregnant. You are pseudo-ovulating. You are going through a pseudo-estrus. You generate the picture of it. So what happens at that point with this new guy who’s just shown up, and here’s this like four-month pregnant female who nonetheless looks like she’s going through estrus? Well, these guys sit there then and say, well, well, that’s kind of nice, and they go and mate with her. And like two and a half weeks later, she gives birth. And you know what the guy’s going to do. He’s going to sit there and say, what’s the gestation period in my species? What is it, like five months or something? I’ve been here three weeks, and already she’s given birth. Whoa, what a guy. And they fall for it. They fall for it every time. Females, by having these pseudo-estrus things, the males are not infanticidal afterward when the female gives birth.

So this whole world that initially makes no sense at all — oh, my god, animals killing members of their own species, even babies, making vicious logic when applying issues of individual selection and the qualifications by way of kin selection.

Next example, another one with a similar flavor. And yeah?

[Audience: Wouldn’t evolution favor a male that he didn’t figure it out that it was his children?]

Yep. And I think probably the most technical, Darwinian way of explaining that is the well-known propensity of blood flow to either go to male brains or penises. They just get impulsive at that point. They don’t stop to think and get the gynecology textbook. That there no doubt is an advantage for the tiny subset of male langur monkeys who can actually be like prudent at that point and think through the logic. It works with males. They fall for it. Suggesting that a lot of what’s going on there is not a conscious, cognitive strategy. And this is sort of not a facetious answer, when by the end of next week, we get to looking at, how do animals recognize individuals? With a lot of species, there’s not a cognitive strategy. You’re not able to sit there and figure out, wait a second, this isn’t making any sense. No.


I think what it is, is that there’s not a whole lot of males there who could make sense or who have been selected for making sense of the fact that mating has something to do with babies down at the other end of things. I think there’s not been much selection for being able to cognitively do that. Or to have some sort of rule — if I mate with this individual, any time she produces a baby, I’m going to be unaggressive to the baby, unless it’s within this sort of time span. I think that’s a little too much, too fast, to evolve in sort of a primate system. The males fall for it.

Another version with a lot of the same logic. And this became clear — studies of baboons, Savannah baboons, in the early ’60s, which is you would have some male, who male was having a tension with another adult male. And they’re about to have some sort of clear fight, and one guy is obviously low ranking, and he’s going to get trounced. And here comes the big, high-ranking male, and this guy is terrified. And what does he do? He looks around frantically and grabs a baby and holds it to his chest. 1950s National Geographic special. Why is he doing that? Because infants are comforting, because babies are cute. And everybody becomes less aggressive when there’s a baby around, because oh, my god, who would attack somebody holding an infant? What if the infant’s injured? OK. So that’s the old version.

So applying some contemporary interpretations, what you wind up seeing, along comes the threatening male. The subordinate male grabs an infant and holds it. And what you only begin to see after a while of studying individuals is, it’s not random which infant he grabs. And you know exactly what’s going to come next, a very similar theme to the competitive infanticide. When males are doing “kidnapping” like this, and that is the term to use, they’re not randomly grabbing kids. They’re grabbing kids who are likely to be the offspring of that male. And it is very clear at that point, mess with me, and your kid’s going to get it. And it’s clearly not being played out in any sort of conscious level like that. But what you see is, it’s not random. And it comes with the sort of qualifiers that show you even more how the system is working.

So you’ve got some guy who’s a big, high-ranking male and has obviously fathered all sorts of kids in the troop. And you’re about to trounce this smaller guy who grabs one of the kids who’s more likely to be his than yours. Logical competitive strategy of essentially blackmail. You’ve got a big, high-ranking male who’s dominating everybody else, except he only joined the troop two weeks ago. In other words, there’s not been enough time for him to have kids yet. When he is threatening lower-ranking guys, they’re less likely to kidnap against him than kidnap against the high-ranking male with a long residency.

And what you find even more interesting, which throws you into this world of, well, are these guys consciously thinking about this, or do they just smell that that kid is that guy’s kid, and do they just know? In primates, they’re thinking about it. And the way to realize that is when they screw up and make a mistake. And this one, sort of years ago I was watching some baboons, and there was this low-ranking guy. He was actually middle ranking, and this high-ranking guy was coming at him. And he was a middle-ranking guy who critically was on his way down. He used to be a high-ranking guy. He was aging. In other words, he used to be reproductively active. In other words, there’s some of his kids around in the troop. So he’s sitting there, and here comes this terrifying, high-ranking male who’s coming at him. And our guy gets all agitated, nervous, looks around, and he grabs a kid, and it’s likely to be his kid. He’s grabbed his own kid. Oh, my god, you’re sitting there saying, I’m never going to get a damn thesis out of this. These animals just ruined all the theoretical models. He’s grabbing the wrong kid. So he’s holding the kid there, and you’re sitting there. And you could see the guy’s about five steps away, and he — so he tosses the kid in the air — just as he’s attacked. You will see they act as if they have made a mistake. They are consciously working around something like this. So this whole notion of coercive blackmail and kidnapping, again, it only makes sense once you begin to see the structures of individual selection, kin selection, reciprocal altruism.

Another example, and this one makes perfect sense instantly. You look at the world of female primates in lots of different species, and there’s ranking systems. There is a hierarchy. And what’s the hierarchy built around? You get a rank one below that of your mother. You inherit your rank. Your mom is the alpha female. You’re her first daughter, so you’re number two. And as soon as she has another daughter, your kid sister is number three, until you have a first daughter, who pushes your kid sister down one step. In other words, dominance hierarchies amongst the females are entirely nepotistic.

Whoa. How can you explain that? Part number two, kin selection. Dominance systems in all sorts of social species are built around the nepotism of relatedness. So that makes sense.

Next one, next interesting notion here, which is one of those choices, one of those choices that are not conscious choices. But you’re about to get pregnant, and you have a choice, as whatever species you are, do you want to have a male or a female, or do you want to have a litter of males or females, or do you want have a litter that’s predominantly male or predominantly female? And it comes down to an issue now of two things, which is, how much does it cost to have a female versus a male during pregnancy? And what are the reproductive probabilities of having a male versus a female?

Back to this issue. You are in a big tournament species. And as we saw, there’s high degrees of male variability in reproductive success. 5% of the guys are accounting for 95% of the matings. So you sit there, and what you’ve got is a rule that if you go for a son — like 90% of sons running around there are never going to reproduce. And you hit the jackpot and you have the 5% at the very top, and each of them are going to like father 30 different kids, going for males in a tournament species is a big gamble. It’s a risky move.

Going for a female, however, there is no female primate out there who has 420 kids because she lays eggs like a salmon. Most females there, unless they have some fertility problem, they all have something roughly one to five kids or so over the course of the life of an old-world primate. So female variability is way down. In other words, what’s a conservative strategy to pass on copies of your genes? Have a daughter. What’s a riskier strategy? Have a son. And what that immediately predicts is two things. Number one, you look in dominance hierarchies, and the prediction is that females who are high ranking should show more of a tendency towards having sons than daughters. And females who are low ranking, exactly the opposite. And that’s what you see in a bunch of primate species that have this sort of structure.

Next prediction. You should then predict that when ecological circumstances get tough, when times are tough, you want to go for the offspring that costs less. A female fetus is less calorically demanding than a male fetus. Male fetuses cost more to bring to term than females do. The prediction should be, during times of ecological pressure, the percentage of females being born should increase. And as a measure of the fact that males are more expensive as fetuses, something like 53% of fertilizations in humans are males, about a 53% to 47% ratio. And over the course of pregnancy, the cost, the increased metabolic vulnerability of male fetuses, are such that by the time birth comes around, it’s around 51:49. And it’s not until adolescence that it flips over to the typical pattern of female dominance. You’ve got to have more male fetuses to start off with because they are more expensive, more vulnerable. So this prediction during times of ecological duress, you should get a bias towards more females being born, the 50:50 ratio skewing in that direction. And that’s precisely what you wind up seeing.

And you see all sorts of examples of this. In humans, for example, during periods of famine, food deprivation, the ratio of births skew towards females. What you also see as a measure of that is among humans, a boy, giving birth to a boy, statistically decreases the body weight, is likely to decrease the body weight, of the next offspring. It’s expensive having one of those males. And what you wind up getting, then, is fluctuation as a function of your dominance rank. If you’re high ranking, it’s almost always worth the gamble to go for one of those high-risk, high-payoff boys. If you’re low ranking, go for the far more conservative female. You’ve got this fluctuation around this 50:50 ratio. And this was something worked out by one of the sort of founding figures of modern evolutionary thinking — a guy named Robert Trivers in the 1970s, sex ratio fluctuation as a function of social context. And people have gone and looked, and it’s precisely this.

And you get an interesting bit of conservatism in this, though, which is there’s some circumstance where it makes perfect sense for you to have a gazillion daughters because that’s the time to do it. And at some point, you’re having a gazillion daughters, and everybody else is having a gazillion daughters. And suddenly, males become really valuable because there’s not a whole lot of them around. So the logical thing to do then is to switch over and start having males. And everything else being equal, after a while, with a predominance of males, it’s going to make sense to switch over to females. You have density-dependent selection you will always have oscillating around 50%. Whichever sex is in the smaller number, that one is immediately more preferable. Poor ecological conditions pushes you this way. Dominance ranks push you one way or the other. But in any of those circumstances, you have an oscillation around the mean. Whichever is more common is less desirable. So sex ratio fluctuation.

More stuff. Role of kinship, for example, in vervet monkeys, we talked about that the other day. Play the sound, an infant alarm call, and the mother gets all agitated, and everyone else looks at the mother. They know that she is the mother. Sort of demonstration of awareness of kinship there.

Now, another very interesting social structure to some species, which is we’ve seen all of these tournament species. You have one high-ranking male mating with lots of females and all that sort of stuff and polygamous systems. What about polyandry – circumstances where a female is mating with multiple males? What about circumstances of stable polyandry, where you get, in effect, the inverse of a harem? You get a single breeding female with a number of males? And what is seen with a very, very high predominance is when polyandry occurs, you get a type which is called adelphic polyandry.


And the second I tell what it is, it will make perfect, wonderful sense. OK. Two male lions sharing a pride? That’s like not what you’re supposed to see. It’s one male lion, and Mufasa and his, like, brother gets pushed out of there. And like, that’s not what you see when you study Disney lions. There’s supposed to be only one male is the breeder in the pride. And occasionally you see these prides where instead there were two males. How can they pull that off? They should be doing competitive infanticide, all of that. Who were the two males? Yes, you guessed it somewhere up there before. It’ll be two brothers. When you see cases of two male lions sharing a pride, or flip the other way, when you see a pride being willing to tolerate two males in there instead of one, very, very high likelihood that they are brothers. This technical term — adelphic polyandry. And you wind up seeing one totally wild example of this in humans.

As I went over the other day, looking at our humans, tournament species, they’re pair bonding. We’re somewhere stuck in between. We’re terribly confused. What you see is most cultures traditionally allow polygamy, but most people are not actually polygamous, all of that. And somewhere in there, you got to ask the same question. Hey, is there any polyandry going on with humans? And there is one wild cultural example of this. And this is seen in traditional Tibetan society, and there you get adelphic polyandry. You have the following structure in rural areas. A woman will marry a man. And in the process, she will marry him along with all of his brothers, all his brothers. She marries the entire lot of them. It is adelphic polyandry. And you see that. And like it’s down to the point where here’s this woman with her husband and his younger brother and his younger brother, all the way down to this infant that she’s holding, who’s the youngest brother who’s now one of her husbands. And that’s what you see.

What’s the explanation for this? This is a pattern that you get in agriculturally very impoverished areas where you got a problem. You’ve got five sons or whatever. And with a pattern of land inheritance where you would otherwise split up the land amongst the five sons, that’s going to put each of them below subsistence level. What you need to do is have a way in which they remain as one reproductive unit so you don’t split up the land. You see this adelphic polyandry, a woman marrying this whole bunch of brothers. And it’s in circumstances where it’s trying to keep the small farm plots from being broken up.

But again, the logic there is the same as you see in lions with two brothers sharing a pride. What else? Now, somewhere in there, you begin to get in a realization that not only do you have males competing with males for reproductive success and females with females, but there’s intersexual competition. How could this be? How could a planet that comes up with Valentine’s Day have intersexual competition? What is obvious is the reproductive interests of any given individual may not be exactly the same as the individual she or he mates with. What would be an example of this? Here you have a species in which males are migratory. It’s a tournament species. A male shows up during a mating season, lots of aggression, lots of secondary sexual whatever. And he mates with a female, and she has kids. Hooray. They have parented an offspring, and they’ve shared their genes and increased their reproductive success. Except this being the species, this guy is going to pack up and leave the next day. In other words, he has no investment in the future reproductive health of that female, whereas she sure does. She would be delighted for this offspring to thrive and prosper, but perhaps not at the cost of her future reproduction. What we see here is, between the male and the female, different reproductive strategies. And this opens up a whole bizarre world of intersexual competition.

Here’s one really bizarre manifestation of this. OK. Basic Mendelian genetics. Those of you who got the catch-up this week should have a vague sense of this by now, Hardy-Weinberg ratios. You’ve got one parent has one genotype, just to distinguish the parents and changes the colors there, all of that. What is it that isn’t bothered with from day one when you learn these sorts of things? It doesn’t matter which one is the male and which is the female, which is the father and which is the mother. This is just one of them contributes this profile, one that. What pops out the other side? Basic Mendelian genetics. When making sense of pedigrees, it doesn’t matter which parent is contributing the homozygotic profile, the heterozygotic. It doesn’t matter which parent.

But then, there is a whole world of genetic traits where, in fact, it does matter which parent it comes from, because the same combination of alleles will function differently whether it came from the mother or from the father. And this was a field that sort of emerged in the early 1990s, and these are called imprinted genes, imprinted genes. These are genes where, for our purposes to define them, these are genes which have different manifestations, different phenotypic consequences. These are genes that work differently depending on which parent they came from. Mendel rolling in his grave. This is not how it’s supposed to work. This obscure world of imprinted genes violates this. They work differently depending on which parent you get it from.

Dramatic foreshadowing — when we get to the ethology lecture, the word “imprinting” is going to come up again in a totally different sense. This is a purely molecular term right now.

So you get imprinted genes, imprinted genes. People began to figure this out in the ’90s, that, wait a second, we’ve got a completely different profile in these individuals with a mutation and a particular disease. Here we have these two very different diseases, and they turn out to have the exact same mutation in the same gene. What’s up with that? If you get that gene from this parent, you get this disease. If you get it from that parent, you get that disease. Totally boggling, making no sense at all.

First thing to have to be solved. What’s a mechanism for actually making a gene work differently depending on which parent it comes from? For those who care about these sorts of details, it’s a biochemical process called methylation. You methylate the gene in one parent’s genome and not the other. Don’t worry about the details. The main way is there is a genetic mechanism for making genes work differently depending on which parent it comes from.

So OK, we know how to do it. Why should this happen? Why does this make any sense? And it took a while, and people began to see patterns to this. And finally this was put together by an evolutionary biologist at Harvard named David Haig in the early ’90s, who came up with a unifying theory which explains it all. And it’s very cool.

Here’s what you see. You begin to look at imprinted genes. And for our purposes, the way we can describe them is, this is a gene where if you get it from one parent, it does what it’s supposed to do. If you get it from another parent, it’s silenced. It never works. It’s methylated into silence. These are imprinted genes.

So he began to notice something. When you look at imprinted genes where they are active if they come from the father, they all tend to be genes that promote fetal growth. They’re all genes that increase fetal metabolism. They’re all genes that make for a thicker uterus. They’re all genes driven by hormones coming out of the fetus. They’re all genes that push for greater fetal development.

And then you look at the imprinted genes coming from the female, and they’re all genes that tend to slow down fetal development. What have we got here? We’ve got that intersexual competition played out precisely there. You’ve got this male hamster who’s going to mate and never be seen again because he’s in this category. And what does he want? He could care less what happens to the future reproductive success of this female. He wants this offspring of his to survive. The male imprinted genes push for greater amounts of fetal growth. And the female ones saying, well, that’ll be great if this kid survives, but I also have a future reproductive life to think about. Her imprinted genes tend to counter it.

What would this look like? One of the imprinted genes, first ones identified, which comes active from the male, codes for a protein called insulin-like growth factor. Who cares what insulin-like growth factor does? What’s clear from the name is it’s a growth factor. It’s one of those things that make fetuses grow more. It promotes fetal growth. And then meanwhile, the female hamster has a cognate, an opposing imprinted gene. What’s that one for? That codes the gene for the insulin-like growth factor receptor. And what her version does is make for a less responsive receptor. The male was pushing for more of the insulin-like growth factor. She gets a receptor that’s not as responsive. And you literally have this co-evolutionary arms race there of male imprinted genes evolving to push for more and more fetal growth. Female ones pushing for less and less fetal growth. And you see that coming out most apparently when you see mutations in some of these genes.


For example, another pair of imprinted genes, one from the father, one from the mother, and what you see there is the father’s version promotes invasion, placental invasion, into the uterus. And that’s actually the term that gynecologists use, invasion of the fetus into the uterine wall. It promotes more of that. The female version slows it down.

So what if you have a disease where the male gene relevant to this is mutated and knocked out of action? You solely have the female input, which is decreasing fetal implantation into the uterus. What have you got then? You’ve got a disease where fertilized eggs don’t implant.

On the other side, suppose there’s a mutation in the female part of this pairing. And as a result, all you have is the male input driving for more and more aggressive fetal invasion. What do you have? You wind up with one of the all time bad cancers you don’t want to get, choriocarcinoma, a cancer of the uterus, because it’s growing completely out of control being prompted by the fetus. When you take out each of the voices through a mutation, you see that normally you’re having this tilting, this balancing, of competition between males pushing for more fetal growth at the expense of the future reproductive success of the female, females trying to slow it down. All of the imprinted genes show this.

Amazingly, people now have even identified imprinted genes that don’t work until after birth, and they work in the brain. And what do they do? The male-derived one does things like make for infants that suckle more. Ooh, get more calories out of mom. And the female version tends to blunt some of the suckling reflexes. It’s all built around the strategy of males and females do not necessarily have exactly overlapping reproductive strategies.

So another version of it is one — a male gene which is pushing towards expression of the fetal enzyme fetal lactogen, placental lactogen. And what that does is it makes it easier for the fetus to grab sugar out of the bloodstream from the mother, and the mother tries to counter it. And if she’s not very effective at doing it, or if she’s rather over-effective, you get pregnancy hypoglycemia, pregnancy diabetes. Suddenly, this is mom having a fight with her offspring over how much calories they’re going to get, the offspring being driven by an imprinted gene from the father. Totally cool, totally interesting.

There should be a problem lurking here, though, which is this is a pattern you only see in tournament species. You got a pair-bonding species, and there’s no reason why a male should be saying something like, ooh, I want to like ruin her future reproductive success at the cost of her giving birth. I’m a pair-bonding vole, and I want her to give birth to a child the size of an elephant who’s going to survive, and who cares about her future? They’re pair bonding. They’re in it together for the rest of time. You don’t find imprinted genes in pair-bonding species. You find imprinted genes in tournament species.

And thus, we get back to that same issue the other day looking at tournament versus pair. Where do humans fall into this? And what we’ve already seen is, if humans can come up with choriocarcinomas and things like that, we have imprinted genes. Same punchline again as the other day — in terms of the number of genes we have puts us somewhere in between tournament and pair-bonding species. Again, we are terribly confused.

Let’s take a five-minute break, and we will pick up with more examples.

Let’s get going again. Two good questions just now. One is, where does homosexuality fit into all of this? And where it fits in is about 30 minutes worth of the sexual behavior lecture sometime in mid-May. It is a challenge for some of this thinking. The second good question was, am I capable of speaking louder? I will try. I mumble.

Pushing on. So we’ve just brought in this whole bizarre, unexpected world of trashing the Hallmark cards of intersexual competition. Another example of it, a really interesting one. This was work done by a guy named William Rice at Santa Cruz over some years looking at a number of different fly species. Female flies are polyandrous. They mate with a lot of different males. And you’re a male. You’ve just mated with a female. And what you would like more than anything in order to pass on copies of your own genes is, like, not have some other guy impregnate her, some other male fly guy who’s intimidating you. Not have some other guy do it.

And suddenly, what you see is this interesting world where, because of the mating frequency of the females, she will have sperm from a number of different males inside her at the same time. And suddenly, we get this very strange world, a whole field of research, of sperm competition – sperm competition of regal A teams and gold medals and all of that, of sperm competing with the sperm from other males. And what you see is, in fly species, the sperm of males make toxins that kill the sperm of other males. Whoa. That’s very elegant. That’s very elegant because it, in part, requires you to come up with a toxin that isn’t toxic to yourself. You can solve that. All sorts of molecular tricks for doing that. But this makes wonderful sense. You increase the likelihood of you, the sperm, reaching the goal line and killing the other guys and all of that. Except there is a problem for the females, which is these toxins that the male sperm release aren’t such a hot deal for the female’s health. And what Rice did with these really interesting studies — I won’t go into the details of it, in large part because I still don’t understand what the guy did. But he was able to somehow take populations of male and female flies, and he would hold the females so that they could not evolve in response to whatever was going on with the males — male-male competition for reproduction — while he held constant the female-female competition. The females were not evolving. The males were.

And what he saw over the course of 30 generations was this male-male sperm competition was such that the male sperm were making such powerful toxins that they were shortening the life expectancies of the females. Whoa. That’s not smart. Whoa. That makes perfect sense. House flies are not your pair-bonding swans dying in each other’s arms for life sort of organisms. It’s a classic case of the male only has an investment in the current reproductive bout. And if he can wipe out the competitor’s sperm and, oh, bummer, the female has now got all sorts of rotty necrotic lesions in her vagina, this is not a very reproductively effective female fly down the line. What do I care? I’m off to the next pile of cow dung after that. You have that same structure there, but this is a bummer for the females. The males, left to their own competition, will evolve more and more toxic sperm, which exacts more and more of a price on the future reproductive success of the females.

Now Rice flips things. And now he takes these populations, and he holds the male evolution constant and allows the females to evolve. And over the course of 30 generations, they have evolved a means of detoxifying the sperm. So here we have male-male competition inadvertently being played out in an intersexual realm as well, where, again, this co-evolutionary arms race — totally bizarre. Sperm killing each other, and sperm damaging the female. And that makes no sense if you do for the good of the species. Again, very logical in the context of these models, these models of evolution.

OK, final example. And here’s one that will become very relevant way down the line when we’re are looking at things like the biology of aggression. So you’ve got these social primates. And in lots of social mammals, you’ve got this pattern, which is one of the genders picks up around puberty and moves to another group. It’s to avoid inbreeding and everybody having six fingers and tails, if you’re a species that doesn’t have a tail, and is perfectly logical. And what you need is just some sort of pattern — is that the females or the males of the species which disperse? And no one among the primates has ever found a good rule for which primate species have female exogamy or male exogamy, which gender leaves at puberty. But there’s variability in chimps. It’s the females who leave at puberty. In gorillas, it’s the females who leave at puberty. In baboons, macaques, a few other Old World monkeys, it’s the males who leave at puberty, with a very critical implication.


So you’ve got baboons. Baboons — you’ve got the troop, and you look at the adult males and adult females in there. The adult females grew up in that group. The adult males grew up someplace else and emigrated into here at puberty. In other words, among the adult females in this group, they’re all relatives. Among the males, there’s no relatives.

Meanwhile, next door with the chimps, it’s the females who have left at puberty. You look at the mature animals in the group, and it’s the males who were there with all the relatives who have been there all their life, and the females who have no relatives. So asking a question. Comparing baboons and chimps, which of those species has higher levels of male-male aggression within the group? Baboons, yeah, because they’re all brothers and brothers-at-arms in the chimp group there. And none of the adult males in baboon troops are typically related to each other. Very high levels of intramale aggression intragroup.

So which species, on a certain level, has put us to shame when it comes to the invention of warfare, of fighting between groups? Chimps, because you’ve got bunches of males who cooperate because they are relatives. And what you wind up seeing with chimps — and we will eventually get to this — is you see things that are now termed “border patrols,” where a bunch of males from a group will patrol the edge of their territory. If they encounter a male from the group over there, they will attack. They will kill him, as documented by Jane Goodall. Taken to an extreme, the males of one group will eradicate another group. And if the rule is, I am killing this guy, we are killing this guy, not because we don’t like the look on his face, but because he is a member of that group, this is the United Nations definition of genocide. Chimps have not only come up with organized warfare between groups, they have come up with a chimp version of genocide.

What’s the driving force on that? One of the truly scary things on this planet, which is when all the males living next door are getting along with each other. Because when they do, they suddenly start looking over at this side. Organized males driven by female exogamy in chimps is where the warfare pattern comes from. And as you can guess, when we get to thinking about the biology of aggression with humans, we are suddenly in this world of taking non-relatives who were military recruits and convincing them to pseudo-kinship themselves into being a band of brothers, increasing the sense of kinship there. Chimps do this all on their own.

So now making sense of which primate species have high levels of male-male aggression within group versus between group, between group are ones where it’s the females who leave at puberty. All these guys have been together since they were terrorizing other kids in kindergarten. They’re an organized group against each other. You get intergroup organized conflict in primates that have female exogamy, females leaving at puberty.

So we’ve now worked our way through a bunch of examples here and how to apply individual selection, kin selection, that sort of thing.

A fourth branch has come into the field in the last decade or so. And this is one that has had a lot of controversy, in part because what it is built around is mistaken by a lot of folks as being an outdated concept. But in part because what it is actually implying is very controversial in some realms. And this is this business of group selection somehow sneaking back in the back door there and becoming relevant to the evolution of social behavior.

Group selection. You remember Wednesday. All the wildebeests there, and the elderly wildebeest pushes his way forward and distributes his bank account and then leaps into the river and the crocs and all of that. And that makes no sense at all because animals don’t behave for the good of the species. In individual selection, he got pushed in. Nonetheless, a notion of group selection has come back into thinking, but it occurs only under a couple of specialized circumstances. The version that we just trashed, which was the version Marlin — OK. How many of you guys know who Marlin Perkins was? It’s come to that.

Marlin Perkins. How many of you guys know what — does Mutual of Omaha, does that still exist? Mutual of Omaha’s Wild Kingdom? Is that still a television program? Yes, no? When you were young did that exist? Yeah. OK. Is it still sponsored by Mutual of Omaha? They still do. That’s great, OK, because that one has been going for decades and decades. And they’d somehow always have to segue to Mutual of Omaha in there doing the commercials, building it like, just as rhinos will mate on for hours on end, you want fire insurance for your home or some such thing. But in its original couple of decades, it was hosted by a guy named Marlin Perkins, and Marlin Perkins taught most of America their evolutionary biology. Marlin Perkins is the guy who taught everybody behaving for the good of the species and that sort of thing.

The first version of group selection that got trashed in the ’60s was Marlin Perkins’ group selection. The group selection that’s sneaking back into the field is very different, two different versions. First one, something you will get in various populations. You’ve got a population of something or others, and some biogeographic event occurs, which causes a subset of them to get isolated from everybody else. A land bridge disappears, or somebody drops a lake there, or who knows what. But something isolates a small subset of the population. So these guys go about reproducing by their standards, and these guys go about reproducing by their standards. And what’s going to be a characteristic very soon of the smaller population, they’re going to be more inbred than this population simply because they’re smaller. They’re going to have a higher degree of relatedness among individuals.

So now we throw our second piece of our three building blocks, throw in kin selection. And what that immediately predicts is levels of cooperation will be higher in this group than in this group, because the higher degree of relatedness. That’s great. And that’s great because once we see how these sort of cooperative systems have a larger payoff, you will get a crystallization so that everybody, the second or third cousins, are eventually going to have to be just as cooperative as the siblings are with each other. You are going to fix a trait of cooperation in that population at a high rate. So notice here, in this case, we have a high degree of cooperation driven by kin selection, whereas these guys are going about their usual “savage at each other’s throats” business.

Biogeographic event reverses, and these guys get reincorporated into the main population. And they’re so different by then that they get a different color. And what you’ve got then is, here in the large population, here is a nucleus of animals who are cooperating, and here is the huge, unwashed mass of the ones who are not. And everything we know about reciprocal altruism, all of that, means that these guys are going to start out-competing these guys. And the model that is used is cooperation will have to crystallize outward. Animals will have to join in in these cooperative patterns because these guys will outcompete them.

What do we have? This is called a founder effect. This is a small population that, thanks to being inbred, has fixated some trait that’s advantageous. Where the evolution here moves faster than in here because of the smaller population with the inbreeding, some adaptive trait comes in there. They get re-induced into the main population, and this founder-driven trait quickly spreads throughout the population.

And what you get here is, if the same thing is occurring in the realm of behavior, reciprocal, cooperative behavior, somewhere in here you have a transition from this being a kin selection phenomenon to this being a reciprocal altruism phenomenon. So that stands as one of the models out there for, how do you jump start cooperation in non-relative populations? You use inbred founder populations to drive it there. And just to have a metaphor, all you need to do, think about in some city there’s some occupations, some mercantile something or other, that there’s a gazillion of them, and all sorts of people work in that. And there’s a subgroup of people there who work in this who are all related. And as a result, they do something cooperative and kin selective. They make low-interest loans to each other. And as a result, they’re more successful at business, and they get incorporated into here.

And what you then have is this economic force that everybody else has to join in to this cooperative business of trusting each other, or low-interest loans, or I trust you like a sibling. You don’t have to pay me in full next week. Used to be that only siblings trusted each other like siblings. But a force there for more cooperation, a driving force of inbred kin selection initially. So that’s one way in which people are thinking about group selection.


And here you now have this group out-competing this because of this trait. And people actually use terms like “crystallization.” The trait of cooperation will crystallize outward and fix in the whole population. So that’s one version of it. And this is one, again, we will come back to, this is one of the ways to jumpstart cooperation in a world in which there is no reciprocal altruism. Because you see the problem with that. We all have learned tit for tat is the best one, except forgiving tit for tat might be better, except Pavlov might even be better. And all of those who built around this one requirement, though, which is somebody has to make the first move of being cooperative. That’s got to jumpstart the system. And what we all know is, in a system that doesn’t go along with the rules of can’t we all sing “Kumbaya,” what happens is the first one in the first round who does something cooperative is some schmuck who’s one step behind everybody else for the rest of time. There’s no way to get these cooperative systems to evolve initially. We’ll see there are ways. And this is one of those — have it driven initially by kin selection, and then throw it into the general population. So that’s one version of where group selection has crept back in.

Another version is sort of the more generalized form of it. And it could be illustrated — OK, the following example. You’ve got chickens. You’ve got chickens with an array of traits, and you can see there’s two types of chickens. One chicken is super aggressive and beats on everybody else, and she lays lots of eggs. Another chicken — far more pacific and introspective and less aggressive and has fewer eggs. So now you take any one of the high-aggression females and one of the low-aggression guys and put them together, and who’s going to leave more copies of her genes? The high-aggression female laying lots of eggs.

Now instead, you have a whole group of the high-aggression females and a whole group of the low-aggression females. What’s going to happen in the high-aggression females? They’re all so aggressive that they all injure each other. They all stress each other into lower fertility rates, all of that. And suddenly you have this very important world in which A can dominate B, but where groups of B dominate groups of A. And it’s in that realm that you suddenly get a push towards group selection. When selection on that level — a world of any individual being outcompeted because of a trait. But as a “collective”– and bringing in that word not randomly, as a collective in the collectivist future of low-ranking chickens to throw off their chains — once you have a group of them, the same traits that could be disadvantageous on an individual dyadic level, as a group outcompetes the others. And when you do that, you’ve suddenly got group selection in a classical form, where these animals, because of their traits, they are not behaving for the good of the group. But the very traits that are disadvantageous individually are advantageous as a group.

And it’s another way of thinking about this. If you indiscriminately make low-interest loans to everybody on earth, you are very readily falling into this category. If you are part of a group that makes low-interest loans to each other, suddenly you outcompete the other. A fascinating book. An evolutionary biologist named David Sloan Wilson, who’s been the main person pushing for this idea for decades, has a very broad range of intellectual interests, including religious history. And a number of years ago, he published a book called Darwin’s Cathedral. And he analyzed the birth of clusters of new religions over time as examples of founder effects and group selection-type properties. And he wound up, in great detail, analyzing Calvin and the starts of Calvinism and Calvin and his sidekick Hobbes in Zurich in the 17th century, or whenever it was that Calvinism started, of analyzing how there were all sorts of these little religions popping up there in these little city-states. And what it was about Calvinism — in I think it was Zurich — that took off, where they established some of these inbred cooperative patterns and then beat the pants off of all the other city-states. And soon, all of central Europe was Calvinist. Really interesting book in that regard.

So back has come this whole possibility of group selection. And out of that has come the more broad way of thinking about it now, which is multi-level selection, a very important concept here, which is sometimes selection occurs entirely on the level of a single gene. In other words, have a gene with bad enough of a mutation, and it doesn’t matter how many great other genes you got going for you. Selection will be decided entirely on the basis of one gene.

As we saw in our principles from the other day, much more common selection at the level of the individual, an aggregation of the genotype playing out in terms of phenotype all the different traits there. But what this introduces is some circumstances where selection is at the group level. And it is never at the level of how an individual’s trait plays out individually in a group. It’s how this trait played out individually emerges as a group behavior that you get group selection. And this marks this great peace that’s come between this David Sloan Wilson and the guy who’s probably the most towering figure in this whole field, this guy EO Wilson, Edward O. Wilson at Harvard. David Sloan Wilson and EO Wilson spent years not getting along with each other because Wilson — because EO Wilson is very strongly individual selectionist model guy, and thus he hated group selection. David Sloan Wilson, the other way around. Each of them were endlessly invited to the wrong parties because of the same shared last name and all sorts of social awkwardness there. In the last few years, I don’t know what happened. I don’t know if they arranged marriages between their grandchildren or something. But they arranged some sort of truce. And Wilson-Wilson published a paper last year basically saying, whoa, you know what? Some of the time I’m right, and some of the time he’s right. And isn’t that great, and can’t we all get along, and delineating circumstances where individual selection — individual selection, kin selection, reciprocal altruism is going to dominate circumstances where selection is going to get played out at the group level. They love each other now. They have formed a new language. They share burial customs. And this sort of constitutes a great resolution.

Yeah, depending on the circumstances, the most important thing in evolution might be a single gene, a single organism, or a single group. And the power of a group and aggregate traits that work lousy in the individual is going to be really important down the line when thinking about aggression and stopping aggression and things of that sort.

So all of this is great. This multi-level selection now showing sort of a fancier, more sophisticated way of thinking about individual, kin, reciprocal circumstances, where it works, circumstances where this is the most interesting thing. So that’s great.

So now we begin to look for a first pass of where humans fit into this. We’ve already seen in terms of all of these traits, tournament versus pair bonding, we’re somewhere in between. In terms of imprinted genes, we’re somewhere in between. These are specifics. How do we begin for a first pass to think about humans, human behavior, fitting into these frameworks?

Two problems, or two issues, that are all we’ll touch on at this point. The first one is, there is this temptation, as soon as you think along these lines, to discover, oh, isn’t that interesting? Langur monkeys will have infanticide. Male langur monkeys will kill the kids of some other male. Human males do that sometimes. Oh, langur monkeys might be an interesting evolutionary model for where infanticidal behavior came from in humans.

Oh, isn’t this interesting? Dominance is passed along by males. We do that sometimes. Isn’t it interesting? Kidnapping somehow occurs. And before you know it, you have a world in which people are saying, well, damselflies do something that look like rape. So huh, we’ve got a damselfly model of human behavior in that realm. And two things are happening there. One is taking the traits, a small number of traits, in some other species and inflating the relevance it has for understanding us. The other danger is that somewhere, creeping below the surface, is you see it happens in other species. It’s kind of natural. It’s kind of inevitable. This is a whole world in which people have run off with how this applies to humans in a completely distorted way. We are unique as a species, as is every other species. We are unique. By applying the same evolutionary rules that every other species has, that’s the commonality, not the outcome.

The other challenge is constantly going to be that you get all these great predictive models, and then humans go and screw it up by having quirks and idiosyncrasies and personalities and stuff like that. All I can say in that realm is, back to Wednesday, if making sense of which lions are going to go to the scary speaker in the bushes there revolves, in some part, around personalities, humans, by the time you get to us, blindly assuming these models apply is going to be very shaky.

So just a first pass at how one will have to be cautious for the rest of course in applying all these models. Now, far more critically, is looking at, what are the criticisms that have been brought against this whole framework? This framework of social behavior has evolved to maximize the number of copies of genes driven by kin, driven by reciprocal altruism. Group selection is a rare trait, but when it does, it follows these properties. This is how you explain all the behavior. Know these four or five rules and not only can you look at two skulls of different sizes, but you can explain all behavior and all of human behavior, and its all biology. And eventually, as EO Wilson said in his landmark book of his in 1975, Social Biology, eventually all of the social sciences will be under the wing of evolutionary biology.

No surprise that social scientists were not very pleased with that proclamation at that point. But this was this notion that this was going to explain everything. So what are the criticisms that have been leveled at this style of thinking? And as we’ll see, there are a lot of them. All of the stuff we’ve been thinking about, this basic notion of applying the Darwinian rules as to why giraffe have to have hearts of certain size to behavior, there’s three features that are coming through.


The first one is the notion of heritability of traits. Because what we’ve been discussing throughout the last lecture or so is, there’s heritability of the tendency to kidnap infants at the right strategic time. There’s heritability of the tendency to decide that you like males with big, fancy plumage and long canines. There’s heritability to deciding that your daughter should be highest ranking in the troop. Heritability. An assumption of heritable behavioral traits runs through every single one of these discussions.

The next theme that runs throughout all of them is the notion of adaptiveness. If some trait has emerged out of the evolutionary murk, if some trait has been selected for, selection implies adaptiveness — the notion that everything you see has an adaptive explanation for why it’s emerged evolutionarily.

The third principle, which is not explicitly stated in this field very often but is implicit to everything about it, is these processes of evolutionary change are gradual. At every generation, if by getting into some better behavioral strategy for playing prisoner’s dilemma you leave 1% more copies of your genes than everybody else, come back in 100 zillion generations and that trait will have spread. All of this is played out in generation after generation so that evolutionary change is occurring in gradual, incremental steps.

So what do you do with these three starting points?

The first one. An awful lot of the next two weeks will be looking at how other disciplines look at the issue of, how do you figure out when a behavior is genetic? And one of the first things we’ll learn is how that phrase means absolutely nothing whatsoever. Maybe a better way of stating it is, how do you figure out when a behavioral trait has a genetic influence? And as we’ll see, even the safer way of saying it is, how do you figure out in which environment a certain behavior has a particular genetic influence? How do other disciplines go about saying, oh, it’s genetic. It’s genetic. In this field, among this sort of sociobiological thinking, what you do is an inverse style of proof. You say, OK, if a trait has a genetic component, you would expect it to be heritable along these lines. If it’s more adaptive, it will increase its frequency, all of that. And look, that explains where competitive infanticide has come from. Show me a more explanatory model. Show me a model that is more predictive. And until you can show me one, I’m assuming this trait has some heritability. And we’ll see how, by next week, the more molecular people rip this view to shreds. Tell me, when you’re talking about heritability of some behavior, show me the damn gene. Show me the sequence of DNA. Show me how that produces the behavior. This is where this field ends saying, oh, we can inferentially sort of decide there’s a genetic influence. As we’ll see by next week, this one piece is the main thing attacked by a number of other fields.

Next piece, the adaptiveness — the notion that everything that has evolved has evolved under the scalpel of evolutionary demand for optimization and all of that. And the critics of this view call this the adaptationist fallacy — the notion that everything is there for a good reason. And what they often do in terms of making fun of that is that, oh, you look at people who do this kind of evolutionary biology with this adaptationist emphasis, and it’s one big just-so story.

Why do giraffe have hearts with thick muscles? And here’s the story I make up, and if my story is better than yours, I get tenure and you don’t. Why is that the zebras got stripes? Why is it that males kill infants in this species or whatever? That everything is adaptive. And what you need to do now is come up with a story to explain why it’s adaptive. And if it fits some of those principles of individual and group, et cetera, then you win. You’ve got the best just-so story.

And what is constantly brought as a sort of criticism of that is, show me experimental evidence that it is adaptive. And by the time we get to looking at the field of ethology, you’ll see a whole way in which people do that. Show me that it’s possible to disprove it. OK, tell me a just-so story, where if that were the case, that would prove that this is not what’s going on. You’ve got to have the rules of science. It can’t just be you observe something and come up with the best story and you win.

But what’s most striking in sort of terms of criticizing this is some domains in which there are traits that have not evolved because they are adaptive. They’ve just evolved because they are excess baggage. And this has brought in a whole sort of criticism of the field. Now, some of the time, when you begin to see — it’s kind of hard to see where the adaptiveness is. It’s because it’s a more subtle feature of it. For example, you could look at squid. And someone who is enamored with fish and how cool and aerodynamic, aquadynamic, fish are and all of that, will point out that squid are pretty lousy swimmers compared to all sorts of fish. They’re pretty lousy swimmers compared to fish, but they’re pretty great swimmers for something starting out evolutionarily as a mollusk. This adaptiveness has to be in the context of where it came from and where it got to. OK, so that is a qualifier.

Nonetheless, even factoring that in, of hidden features of adaptation in evolutionary history, there’s lots of traits that are simply not adaptive. And one of the most effective critics of this whole view, Stephen Jay Gould, introduced a term about 30 years ago that summarized this view. Gould was a paleontologist at Harvard, and his thinking will come in a whole lot in next week’s lectures. He teamed up with another guy, a molecular geneticist at Harvard named Richard Lewontin. And in critiquing a lot of this, they came up with a term saying an awful lot of traits have evolved that exist not because they’re adaptive, but because they’re spandrels. So at that point, everybody in the field had to go run to their dictionary because Gould and Lewontin were all being snotty and coming up with fancy, highfalutin culture terms, because it turns out “spandrel” is a term from architecture. So immediately, they already won points off of everyone else for being all cultured and stuff.

But a spandrel, a spandrel is you got some medieval building. Remarkably enough, that’s a medieval building, and — it’s a nice medieval building. And what you get is you have these arches that — there winds up being a space between arches, that there’s space between arches just because you can’t build arches next to each other without there being like this space here. This is solid brick or LEGO blocks or whatever. So you have these arches, and there’s a space in between. And you got this space. It’s not there for any good reason. It’s there because you can’t put two arches next to each other without coming up with this triangular space. And the architectural term for those triangles are “spandrels.”

And spandrels, in architecture, is something that forms between two arches. And as Gould and Lewontin used the term, a spandrel is some trait that occurs simply as the unavoidable outcome of some other traits that are being selected for. You can’t put two arches together without getting some sort of triangular shape in between. And throughout the history of architects making spandrels, these would be artistically built up and all of that, and there would be all sorts of decorative stuff on that. So if you were an adaptationist person, everything has an adaptive purpose. You look at that and, oh, there’s got to be some reason why spandrels have evolved and have these elaborate secondary sexual features on spandrels, little gossamer things flying around and stuff. And there’s got to be an adapt — oh, I know, because people in church will look up at it a whole lot, and their necks will get sore. And when their necks stop being sore, they will decide this is the glory of God, and they will increase their belief. That’s why architects put in spandrels.

No, you can’t build buildings like this without spandrels happening. And just because they were there, why not let’s decorate them. That’s the evolution of spandrels.

And in Gould and Lewontin’s critique, an awful lot of stuff, where the sociobiology types would sit there and say, we’re going to have a just-so story contest now and use our principles and figure out some way in which this is actually adaptive, they’re saying, this actually wasn’t selected for because it’s adaptive. It was an inadvertent byproduct of something else. So here’s an example of that would be given. Humans have chins. Humans have chins. Apparently, all humans do have chins hidden away someplace or other. Humans have chins. And weirdly, we’re like the only primate that does. You look at, like, other apes and stuff, and they have the sort of weak chins that suggest sort of immoral characters and criminality and things like that. We’re the only species that has chins.

And apparently, there’s been some like nutty adaptationist school somewhere back when, trying to make sense of why it is that humans evolved chins. Why it is that human faces come to a point there, and what’s the adaptive advantage? And you can, like, stab rivals — or you can get crumbs from out of the corner on the floor and stuff with your chin. Until somebody figured out that there’s no way that you could have a primate face that has a muzzle foreshortened and a jaw at this angle. You do this, and you do this, and you’re going to get this little spandrel thing sticking out there. And oh, there hasn’t been selection for a chin because you have selection for a hominid face with a shortened muzzle, and this thing pops out there.

And suddenly, all those doctoral theses about the reproductive advantages of chins go down the tubes there. It was merely a spandrel. And Gould and Lewontin’s argument is that there’s lots and lots of spandrels, that this emphasis — everything is adaptive, everything is the outcome of competition — that’s produced that, an awful lot of stuff is excess baggage that merely gets carried along because the evolutionary process isn’t all that efficient. You’re starting with a mollusk if you want to make yourself a squid.

Giving rise to another concept, another term in the field. Andre Lwoff, who was a French — or was it Jacob? They were these two French molecular biology guys, and they both got Nobel Prizes. And one of them said this, and I don’t remember which, but I’m going to say it was Lwoff who did because I like saying Lwoff. Andre Lwoff once said that “Evolution is not an inventor. Evolution is a tinkerer.” It works with preexisting structures. And the notion that you’re going to see this everywhere, you’re gluing stuff together. You’re duct taping this part of the pelvis to this part of the circulatory system, whatever. And you’re going to have spandrels. An awful lot of what those folks go through in these whole song and dances of adaptation are merely spandrels. Evolution is not about optimizing every single trait.

What comes out of that is this notion that an awful lot of this world of competition really isn’t about competition. You stick with these models, the models we’ve been covering for the last two days, and every bit of advantage is going to increase the number of copies of your genes. What’s the only possible outcome? You need to compete. You need to outcompete. It was a critique of this whole view as being one driven very strongly by competition, because all the traits there are driven by having to out-reproduce the next individual, next individual.

Another realm in which this has been critiqued has been by, interestingly, evolutionary biologists from the Soviet Union. And where it was emphasized there — in Western sort of emphasis in evolutionary biology, most of evolutionary selection was about competing with other individuals. Sexual competition, competing for running away from the predator faster than the guy next to you, competing for food, all of that. If you’re sitting there in the Soviet Union and you’re thinking about evolutionary biology, what you’re mostly thinking about is evolving the means of surviving climate demands, surviving extremes of climate, what would be categorized as abiotic rather than biotic, abiotic demands in the environment. And when you look at species where most of their survival is about surviving ecological extremes, environmental extremes, you don’t see a whole lot of competition. That’s a world with far less. And throughout the years, the Soviet evolutionary biologists tended to emphasize abiotic selection, which had far less of an emphasis on competition.

The final piece, though, the gradualism. The gradualism is very intrinsic to this. Because if everything is adaptive, every little bit of advantage is worth fighting for, and competition and killing each other’s infants and killing each other’s sperms and oviducts and all of that. And every little bit is going to make for a little bit of an advantage, and you will have this gradualist change.

And the huge challenge that came to that was from Gould and another scientist in the ’80s, a notion that maybe that’s not what evolution looks like. Maybe evolution is not occurring in these gradualist, incremental steps. Maybe what occurs instead is long periods of nothing happening and then a rapid evolutionary change. In long periods, nothing happening, then rapid evolutionary change, something that they called punctuated equilibrium. And the whole next lecture is going to be about this.

Critical implications — the genetics of how stuff evolves turns out to fit far better the molecular genetics with these models of punctuated, rapid change, and stasis, equilibrium, punctuated change, rather than these gradualist models. And if that’s what it looks like in these steps, most of the time you’re not punching it out for every one-half of 1% reproductive advantage, and it’s all competition. Most of the time, nothing’s happening at all, and there’s periods of severe selection and then long periods in which nothing happens. A step function like that, the role of competition is reduced tremendously in thinking about evolution. So we’re going to spend a ton of time looking at the punctuated equilibrium model. To give away the punch line that I know you are all on the edge of your seats for, I kind of think it’s right in a lot of the circumstances of evolution because it fits much better with the molecular biology.

Finally, in addition to these critiques of show me the genes, everything doesn’t have to be adaptive because there are spandrels, and besides, this is really wimpy science where science counts as making of the best story, and there’s no reason to think that it’s gradual, there’s a mechanism for this as well. In addition to that, there’s always been a final realm of criticism of this style of thinking, which has been a political critique. And in lots of realms of science, you can’t really imagine a political critique as to whether cilia on amoeba flip this way or that way or a critique as to whether, like, plants should have three clover things or four clover things, that there’s not a political factor in that. There’s a huge amount of sociopolitical implications that run through this entire field. Because it addresses issues of like how natural, how evolutionarily ancient, how evolutionarily honed, how adaptive are things like whether a species, a society, has male domination or not, has strict hierarchies or not, has certain degrees of aggression, has certain patterns of sexual coercion. And a whole world arguing that these things have a naturalistic grounding in adaptation versus a whole world that says that’s jibberish. Most of the time, nothing’s happening. And when it does, a lot of it is spandrels. Those have very different political implications.

And from day one, the critics of this entire world have made a point of something that initially seems kind of silly, but may not be silly, which is the entire founding generation of these sociobiologists were white, Southern males. Make of that what you will. EO Wilson from Alabama, Robert Trivers, Irven DeVore, a whole population of these folks, the first generation of these people, were all white males from the American South. Whereas Gould and Lewontin, all of these guys, were Northeastern Marxists.

So we got kind of a major contrast here. And the critique of the sociobiological view is, well, isn’t it interesting that their notion of how evolution works just happens to emphasize the naturalness of a system that rewards them for the inequalities that we have? Isn’t it interesting that their models predict the naturalness of a world in which they’re the ones who benefit most from the notion that this is natural? And from day one, that has been a strong critique. And that’s not like critiquing cilia amoeba theories for political implications. That is a very real one. Because the notions that this is pertinent to making sense of, is rape a human psychopathology, or is rape a competitive strategy, is the fact that children are more likely to be killed by stepfathers than by biological fathers. Does that represent something about the typical socioeconomic pressures in families with step-parents versus the natural biology of gene competition through competitive infanticide? This has a lot of implications.

And from day one, as this field emerged in the late ’70s, there was a huge political agenda that was assumed to be there. Sufficiently so at one lecture, a famous conference in 1977 where EO Wilson was presenting there, a bunch of people rushed the stage and knocked him off the stage and dumped water on him and chanted, whatever it is in German, saying, we will have a society of law and order. This is pretty agitated circumstances for thinking about science. This was a group in Boston called Science for the People, which was a Marxist group at the time there, and saying, all this sociobiological stuff is doing is justifying a world in which it is male-dominated, stratified, and where aggression and competition pays off. And this has been highly controversial from the beginning.

The one counter to it — and I think a lot of that was valid, and I think a lot of it instead defaults into models where this is not making a whole lot of sense. The one thing to be pointed out is, when we see what punctuated equilibrium is about, it’s exactly the sort of world of evolution that produces the sort of world that a Gould or a Lewontin would want it to be, one which minimizes competition in favor of cooperation and a whole pattern of stasis and rapid dialectical change. It’s exactly the world that two Marxist geneticists would say evolution should be about.

So I think that as a first pass, you’re seeing this stuff is dripping in sociopolitical implications. So starting Monday, we will look at this punctuated equilibrium, taking apart this piece of the story.



By Pangambam S

I have been a Transcriber and Editor in the transcription industry for the past 15 years. Now I transcribe and edit at If you have any questions or suggestions, please do let me know. And please do share this post if you liked it and help you in any way.