← LSE: Public lectures and events
LSE: Public lectures and events
Why is it Always 'Us' and 'Them': on the natural history of thinking through groups
01:29:44
Loading summary
Transcript34 lines
- [00:00]
A
Hello and good evening to everyone and thank you for coming to this lecture, which is the first public lecture of the Program in Culture and Cognition, jointly organized by the Institute of Social Psychology and the Department of Anthropology. I am Rita Stutti from the Department of Anthropology and I'm delighted to welcome our speaker tonight, Professor Larry Hirschfeld. Indeed, we couldn't have a better speaker than him to mark what is the first school wide initiative of the Program in Culture and Cognition, as he is one of the pioneers in this area of interdisciplinary research that straddles across anthropology and psychology. Since 2005, Professor Hirschfeld has held a chairman in the Department of Psychology and Anthropology at the New School of Social Research. And before that he was professor in the Departments of Psychology and Anthropology at Michigan, where indeed he started one of the first programs in cultural cognition. He's probably best known for his work on race and for his very influential 1996 book, Race in the Making, in which he makes the groundbreaking move of searching for the origins of racial thinking in the minds of our children. And it is on this, on a topic related to this work that he's going to talk to us tonight in his lecture why Is It Always Us and Them? On the Natural History of Thinking Groups. Please join me in welcoming Larry Hirschhot.
- [01:59]
B
Thank you. I recently went to a talk by Ian Hacking, who probably many of you know as premier analytical philosopher and scholar of Foucault at the NYU Law School. And I was sitting there with a few other people that I knew and I was struck by how well dressed everybody who was coming and sitting in the audience is not typically like my students or colleagues. And then three of these young people who were particularly well dressed got up and started to introduce the speaker, who was not Ian Hacking. So if any of you are looking for a different lecture, I understand. You may leave now. I should probably start by telling you that the title is misleading and that I'm probably not going to tell you why it always assumed them. And I'm only going to speak briefly on the natural history of thinking through groups, but I'm hoping that I'll come close enough that will nonetheless help our understanding of these topics. Whoops. There you go. Sorry. I wanted to thank my collaborators, without whom literally I wouldn't be doing any of this work. Particularly grateful to UDA Frith, who's here in London and who with whom I cooperated in a very junior sense in doing a study that I'll talk about a little later. Basically, what I'm going to do here is very briefly Talk about how people understand other people as having minds. But most of it will be about how people understand people in virtue of their being members to something. And I was originally trained. Well, I was only trained actually, so I don't suppose it was just original as an anthropologist. And it was self evident that when we looked to give accounts of social phenomena and people's behavior, we would look at aggregate level phenomenon in which people were members of something, typically in much of the work, kinship groups. And that that's why they did it, whatever it is they did. But then I discovered that actually a great many people didn't believe that was really the case, Many of whom were psychologists and analytic philosophers. And they. The more I looked at how people would come to represent notions of aggregate phenomena like kinship or ethnicity, say I found that there wasn't really that much work on the topic. And the only work that was really extensive was work on race, which is how I got interested in race as a scientific project. And also turned out that whenever you said something about race and psychology, people would say, well, of course it's that way. The racial environment is so saturated, how could anybody not know that? And so I kept trying to find ways to find people that wouldn't know it in some sense. And I came upon children, really, which is not something anthropologists typically study. In fact, they're considered pests largely in the field. And they're not considered to be actively doing much of anything. But so that. And then I discover, you know, developmental psychology in certain American universities, it's really encapsulated and you're not allowed to learn about other things. And I thought that was a neat thing. And I thought I'd try doing some of it. I didn't fully understand that it was something that people got graduate training in and that it took a while to master some of that stuff. But luckily I was in Michigan, where they were particularly sympathetic and kind and helpful in helping me set up a number of studies which I'm going to talk about here. But again, just briefly. But the way that the talk will go now is that I want to look at the conventional wisdom about things like race and I want to propose an alternative hypothesis. And that is that race, among other social aggregate level phenomenon or imagined phenomenon, could possibly be something that our thinking is governed or our thinking is governed by a program of development that's endogenous to humans and peculiar in many ways to human society and represents a sort of folk theory, not a scientific theory. Because scientific theories of race are hopelessly misguided. Typically, but folk are allowed to have misguided theories. I'm going to give a little review of some of the support for this. And the first part is earlier work I did where we looked at typically developing preschoolers. And then I'm going to talk about a recent study in which we looked at the impaired children's understanding of social phenomena. And then I'm going to end with a few speculations about natural history. I'm not sure if I'm allowed to do this, but if you have questions, feel free to raise them. Obviously, understanding what other people are doing and giving interpretations of their actions is a fundamental challenge that all social animals must meet. Or I suppose non sociologists, asocial animals as well. But for humans it's. And for the human child, it's a particularly important challenge to meet. And there's a good deal of strategy, a good deal of research on the strategies that are employed. And I want to suggest that one of them is fairly well known in a lot of research and the other, there's considerably less research on. Rita has probably done more of this than anyone else that I know of looking at. Rather than how people think about the things because they have beliefs and desires, mental states, it's going to try to understand, explain people's behaviors in terms of the roles that they take or the statuses they have or their membership in some other way in a group that is from which the individual inherits their ideas about how to behave. So that Johnny expects his mother to prefer baking cookies to repairing ovens is not necessarily something that she was sort of born with, as it were, but it's something that in virtue of being born a woman, she was likely to develop. I wish I could put quotes on things I say because I like to distinguish between beliefs that I have and beliefs I have about other people's release. But ultimately, since I don't have a quote, I can. When people object later, I'll be able to say, oh no, that was supposed to be in a quote. So theory of mind module, we know a great deal about this now. It's a developmental program, endogenous developmental program, and that it governs how it is that we develop ways of thinking about people in terms of their mental states. There's a good deal of neurocognitive evidence now that shows that there's dedicated brain circuitry involved with that. And there can be selective impairments. And there's also a good deal of work that shows that what we could describe, I think, as initial states of theory of Mind are available fairly early in the child's life. These are studies that probably many of you were familiar with that shows that infants and I could have. There are studies with much younger infants now showing that they are able to think about people in terms of mental states of one sort or another even before they've had a whole lot of social interactions that would guide them to believe it on their own reasoning about groups. I'm calling it Theory of Society only because I'm trying to steal from the Theory of Mind Popes, an attention grabbing acronym. TOM looks better than toss, but there wasn't much I could do about that. The other one was already sold. So as I said, knowing who a person is is enormously important for understanding what is going on in their mind. And that in many social species that who a person is is a function of their groups through which they belong. And unlike some things, but. Well, I'll leave this elaborated here. But not all groups have the same psychological properties because not all groups are supposed or presumed to have the same properties in the world. And this is true of mental states as well. The idea that you would do something because you have a desire, excuse me, is very different from understanding something because you have a belief and that there are specific developmental courses to each of those two kinds of mental states. Well, there are different kinds of groups to which people can belong. And each of those groups has a different possible way in which they can affect behavior. Art Bhagwad was a newspaper columnist in the United States. I don't know if he ever was syndicated in Great Britain. I don't think his jokes would be very funny here, but it's a little like watching a Woody Allen film in Paris where I was the only person laughing. He had one column about headlines that you'll never see. And one of them, of course, is the he gives here is that white people caused the stock market to crash. Well, of course they did. White people always caused the stock market to crash. Nobody else but white people typically cause the stock market to crash. But you're never going to see this headline because it tells more than we want to tell or that we need to tell. Probably more interesting but less funny account is Sartre discussion of a woman with strident anti Semitism that he encountered. And he asked her why she was so anti Semitic. And she said, well, I had a terrible experience with this Jewish furrier. And he said, well, how come you don't hate furriers? But of course you don't hate furriers in the same way the Inequality of groups is obviously supposed to be pun or the double entente. I'm not sure that it works, but I wanted to guide you to it in case you were wondering why I used that funny phrase, The notion of humankind, which is something I've discussed elsewhere and which other people have discussed, including Ian Hacking, to pray. There are a number of people who have used the term and borrowing it in most respects from the notion of natural kind of philosophy. But here I mean the idea that every society imagines that the world can be sorted into different groups and that each of these groups is a natural fact of the world. They're not constructed or construed. And I'm going to make the claim that all societies make this sort of recognition on the basis of no data. Each group in a humankind is presumed to have a distinct physical appearance and a distinct set of non obvious qualities and potential. Obviously if you keep the notion of race in mind, you'll see immediately how this kind of thing applies and stereotypes and things prejudice. When you think about caste, you'll get very similar kinds of notions. You can think about age in certain societies in sub Saharan Africa where age serves as an important organizing principle to social life and where different ages are thought to have different kinds of obvious and non obvious properties. The important thing here for humankind is that there's these two things. The idea that you have a distinct kind of outer appearance and a distinct kind of inner landscape are thought to be linked, causally linked. And the idea is that they are causally linked because you have a group essence in the same way that biologists and folks, biologists used to, and folks still do believe that plants or animals have distinct essence, although I will argue later that they're not exactly the same kind of essence in their structure. This is sort of as close as I'm going to get in some ways to what I promised in the title. But the important thing here is to recognize these natural groups are not stand alone things. They're not distinct kind of thing that is out there in the world. It's a distinct kind of thing that's out there in the world because there are other kinds of things of the same order that are out there in the world. So that these are always defined in relevant relative terms, even when it's not necessary. So for instance, race, you have people who presumably can as a group and without breaking any laws of nature as it were, reproduce each unto themselves. But they're still seen, races still are seen by people as one in virtue of the Being distinct in virtue of the other. And that's sort of what's behind the notion of white people cause stock market to crash. In many ways caste is different in that although they're endogamous, they need each other in order to exploit the environment. Well, these kinds of phenomena are not something I've discovered. They've been around a long while. And do you need to have some kind of special mental device, special purpose mental device to explain it? Well, no, you don't like conventional wisdom and it's not been seen until recently. I think there are more people now, as Rita's work shows. More people now considering the conventional wisdom is problematic, but this is relatively recent. The idea that's held sway mostly in social psychology where these kinds of phenomena are typically studied, is that group based reasoning is just borrowed from general reasoning we have about anything in the world, general purpose reasoning. There's nothing particularly special about humans as a target of our categorization and our attributions to others in virtue of their category membership. It's just borrowed from general purpose categorization impulses and low level kinds of phenomena like judgments of similarity. And you don't really need more in order to be able to build up the kind of stock of beliefs that humans do about other humans as members of groups. And that differences, the part that I mentioned earlier in terms of inequality between groups can be explained in virtue of something that's not really inherent to the group ness itself, but to some properties, some accidental properties of them. Like race is supposed to be well envisioned by simply opening up your eyes and looking. And whereas you. Some other category, if it were sufficiently marked visually, would be as important and as automatic processing as race. So that it goes back quite a while to work the Siskin Taylor did, where they talked about how race, gender are automatic. You can't open your eyes, you can't go into a room without counting people, seeing people as belonging to a race, a gender or perhaps an age. Now there are a couple reasons that I think that we could be skeptical sort of a priori reasons. The first is that if this kind of social reasoning of this sort about humankind were just a function of general purpose reasoning abilities, the kind that you apply to fixing an automobile or inventing a kinship system or writing ads for cell phones. You know, I haven't been in London in a while. Does anybody do anything but buy cell phones here? I mean, the stores are staggeringly, the place is staggeringly filled with things I can't afford. And cell phones seems to be the Dual category. In any case, if social reasoning, social group reasoning about humankind were a function of domain general reasoning processes, it would be interesting in and of itself simply because it would be like unlike any other important processing that we now believe people do, so that we have dedicated processing for theory of mind and we have other kinds of processing for things like physical objects and their movement, other kind of processing for differences between kinds of flora and fauna and so on. There are many and of course language is one of the most important. And if our understanding or one of our primary ways of understanding other people were simply a function of category effects that can occur in any situation, it would mean that one of the things that's most important to us as humans is one of the least evolved parts of our psyche, which I think would be kind of surprising, not impossible, but it would be kind of surprising. Probably more interesting is how much non human primates use group membership and very, very subtle cross cutting situation specific kinds of group membership to interpret and predict behavior. And it would, here we'd have, and those are almost certainly in some ways special purpose devices, mental devices. So here we'd have something that was less evolved in humans than it is in our primate cousins. Again it could be, but it would be a surprise. Now I put a question mark here because most of these things could be arguably due to something else. What I'm going to suggest is that even if they are explained well by familiarity and liking effects, still the constellation of them seems striking to me. One is the pre verbal infants know a lot about language and they also seem to know something about race and that they show the other race effect that adults show. I'm not sure that that's bad news by the way, because you can, you can distinguish this effect by giving them the barest of diverse environments and it'll go away. This is some recent work and I think it's really striking is how much children are sensitive to different accents and seem to prefer mother's accents to another. Because on one reading an accent is somehow a communication failure in the sense that it's not typical of a community. But of course most accents are typical of some community. So the children distinguish those kinds of phenomena, things that are typical of one community from things that are typical of another is something that I, and particularly language is something I'll come back to later because I think it's important. This study, it's an old one, but I just loved it, is that maybe kids distinguish between adults and children because of height. So they added a control study with Midgets and kids didn't think midgets were children. You know, 12 month old kids didn't just think. And of course kids are very early sensitive to gender and kinship status. I'm not only talking about difference in categories, but actually difference in the status that people hold in virtue of their membership in the category. Well, I'm going to go through. Let me just put all of these up for a second. I'll put the last one up so that Rita has time to get angry. These are products of studies that I could go in in some detail, but I don't think it's necessary here. The inequality of categories, just the idea that some categories have greater inferential potential than others is something that very young kids understand very early, by three years of age, probably a little younger even so that some categories they think are terribly important and other categories are less so. One of the studies we did asked kids about. To compare individuals who varied on one or two properties at the same time. So one of them might be a thin black man and the other might be a heavyset white man. And we asked them to predict who the parent was. Basically doing a couple crossovers of all of this. We found that if you ask them is a body build type, not skinny and obese, but, you know, sort of husky versus wiry, they thought that that was inherited. They thought that even if you switched the child at birth, he would grow up to be husky if his parents were husky and thin, if his parents were thin. And they did the same with race. So on one level you could say, well, they know this about more than one physical difference. When we crossed race with body build so that they had to choose one of those, they overwhelmingly three year olds overwhelmingly chose race. That is, race was the property, the bodily function, the embodied quality which was fundamental to who the person would be and would be the kind of person that they would grow up to be regardless of the environment in which they grew up. The second study was a study done with French kids. And one of the reasons I had done it in France in the early 90s was because unlike the states at that time, when the dominant racial landscape was black versus White, in France there were a number of different racial groups, some who looked more or less similar than others. And black and white wasn't the one that was principally featured in public discourse. It was much more North African American and Europeans. And we read kids either read or showed kids stories in which a person's race, gender, occupation and so on figured, but had no importance to the storyline and the thing that we found is that when they looked at the storybook, they reported no visual information when we asked them. They reported no racial information when we asked them about the person. They did for their gender and they did for their occupation, but nothing for race. On the other hand, when we asked them to tell us about the story and the characters after we read them the story, they did remember information about race. So it suggested to us that what was happening is they were learning about the categories before they were learning about the criteria that determine membership in the the categories. And this fits well with findings that although children, I'll come back to this in a little while. Children are three year old children are really strident racists in many ways. And they, but they're, they're sort of colorblind in terms of their play until about six years of age. And my suggestion is the reason that's true is because they know they don't like blacks, white American kids, but they don't know who among them are blacks. So they just, they don't know who belongs to the category. Constancy is just the straightforward thing I was talking about. If you have different environment in which you grow up, will that environment cause you to be a certain kind of person? Or where you're. Will birth be the thing then gestation be the thing that fixes the kind of person you are? And children understand that that's the case. They believe that race and gender are constant. Now the. Study here that the kids naturalize race is also the idea that if you have a switched at birth, they'll go with the birth parents as the ones determining physical properties at least, and I included Rita here, but she doesn't entirely believe that. But it is for one population at least that her vaso kids made that distinction. And those in many ways were the population that the kids knew less about. So in a sense it was the more automatic judgment. They were relying on a database that was less full than when they were relying on the database for ethnic differences, what we might call ethnic differences and suggest to me possibly that it has a certain priority. Now there's a. Now I want to talk about the possible support for the idea that there's a module for theory of society reasoning. I mean, the effects that I've talked about are not necessarily terribly controversial. They're part of someone's folk beliefs about society in general. They're not bizarre beliefs. They may be a little more precocious than we would have otherwise thought. But what's the chance that this is due to some kind of special purpose endogenous device. Well, one of the things is that they will cross them in ways that don't seem predicted by their experience. So in one study we found that young children thought that in Ann Arbor, Michigan, white kids thought that when we played them Portuguese and English speech that it was black people who spoke Portuguese and white people who spoke English. But in Ann Arbor, Michigan, that's not true. Pretty much all the black people they know spoke English. And many of the white kids, this was in a university daycare, spoke other languages. But they were, I'm arguing, were bringing their understanding of what seemed like physical properties to something that was more interior in a way. And in a study that would have. Sorry, This one I love, I'm sorry I got ahead of myself, but this one of about the 5 year olds predict that members of different categories have distinct way of speaking. She was asking kids to give examples of how different people spoke. And if it was the mommy, how would they talk? And if it was the daddy, how would he talk? And one kid said, oh my daddy, once upon a time, the end. Another kid said, talk like your daddy said, you're fired. All right. There is support for the theory of mind module. I got ahead of myself before. One of them has to do with at least patterns of distinct activation of the of the brain. Now that doesn't necessarily say that this is everything will have to some extent a distinct pattern of activation. That's what our brains are built to do. But these seem to be activations of areas of the brain at least the authors argue that you wouldn't otherwise necessarily expect. And this was the study I met with that I started where she has. She looked at adult Asperger syndrome patients and these were still patients nonetheless who had mentalizing problems. They had some challenges, their theory of mind, but they were as able to use social stereotypes to categorize pictures as were unimpaired adults. Get back to that in a little while back right now. Still those are suggestive. But I'm going to admit that acknowledge that they're not overwhelming data supporting the idea that there's a humankind module that's a distinct cognitive device that may have evolved to meet problems in our ancestral environment. The whole nine yards. And here's what I think might be a good one. Strong evidence would be selective impairment in say, some kind of mental module that we know something about. Say like theory of mind, but not in theory, not in terms of theory of society or humankind module. So if you had an impaired capacity to mentalize and you had an unimpaired capacity to interpret and predict behavior. It would seem to me that it would suggest that there are two modules that can be disassociated and that can be affected by different kinds of insult to the brain. Stronger evidence would include the possibility that in fact not only are they different modules, maybe their course of development is different. Pretty much everything, any sentence that starts with pretty much anything is almost certain, untrue. But pretty much anything that happens interestingly to cognition develops when between the age of 3 and 5 and kids. So if you just find that there are these two things develop between the ages of three and five, you've maybe not made the strongest, maybe it's not the biggest news. But I'm going to argue that if there's a priority of one to the other, it might be indicative of a distinct device of some sort. This is a study that recently came out that I did with Gudefris that I mentioned in the beginning. So we know that autistic children have impairment in their theory of mind capacities, in particular with second order mentalizing. That is the understanding, grasping that someone could believe something that they know isn't true. If kids who have this kind of impairment and this kind of impairment is a, Is a very important one. Kids that tend to fail false belief tasks have significant cognitive impairment and other things as well. If such kids don't have problems thinking about people in terms of their membership in social groups. It may be argued, I think, that these two kinds of reasoning theory, mind and theory of society are governed by independent mechanisms. So to test this, we had four groups of English kids. We had typically developing seven year olds, typically developing three year olds and autistic children who were matched with mental age with the seven year olds the seven year old controls, but who were able to do the false belief task, which is not entirely uncommon. And many of these were obviously older too. And then we had another group who were autistic children who were matched for mental age with the typically developed in 70 years. But they failed at least one of the two false belief tests that we gave them. That is, they were unable to grasp that someone else could believe something that they knew to be false. And the tests for those were the Celian and Smarties test. Anyone who's had done any reading on autism and theory of mind would know these tasks really well. But a simple one is I put this here and close the drawer. You all leave, please don't. I pick it up and I put it in my pocket. And then you come back and Rita says, where's the handkerchief? Wait a minute. I got that wrong, didn't I? Yes, I got it wrong. This is like the waste and test you already got. So it's moved. Ah, Rita didn't leave. Rita was the subject. But Maurice says, rita, where is Jesus? Believe me, it's a good task. I don't want to. Lots of people have done it. Maybe not me, but lots of other people have done it well, and I had nothing to do with testing these kids. But in any case, the bottom line is that the kids will look, the impaired kids will look for where they know the handkerchief is in this case, which isn't one, but where the kid knows it is, rather than the place where the person that they're asking about last saw it. Because they figure if they know it's in the drawer, then the other person knows it's in the drawer. The second task is a reasoning task using stereotypes and do you have these things in the movies and TV where it announces the content may be disturbing. Think about your children. This content may be disturbing about your children. There were two reasoning tasks. One was the preschool racial attitudes measure, and it has 24 racial attitude items and 12 gender items. And it basically asks about common or typical stereotypes and prejudice. I'll get to an example in a minute. And it asks the child to choose between, for the gender items, a male and a female, and for the race items, a black person and a white person. And then there was a second condition which had we took 10 of the items from the other and we instead of just choosing which child the property was true of the. They were given a scenario in which they had to choose between two competing explanations of what the child. So they could have chosen either that a child was doing what they were doing because they wanted to do that, which was a mental state that even the autistic, the children who failed the false belief test understand. Or they could choose to explain what happened in the story in terms. And who was the actor in terms of a common stereotype. So let me give you a couple of examples. These weren't the drawings, by the way. And this is here. These are actually pretty nice ones. Actually, the slide was made up by someone who wanted people to still like me at the end of a talk, I guess. But basically what you have here is, you know, you got a black person and a white person, and you say, this person likes to help people. Which person is it? And on the common stereotype, they'll pick white representation. Some of these got very you know, a child comes across a puppy drowning in a puddle, which child will save the puppy? And they pick the black child surprisingly disturbingly often. I'll give you some data in a minute. Now the conflict task, basically the prem the first item is which one of these wants to play with the dolls? Which one of them is likely to do this behavior? And because it's gender, most kids would choose the woman for playing with dolls and the female character for playing with dolls. Here you see in the conflict condition, Grace, even though she's a girl, she doesn't like to cook for people. And so which child is it who's going to to cook for people? Well, most. Well as you'll see most typically developing kids will overwhelmingly choose the. And when I say overwhelming, you can see they're around 70, 75% will choose for the consulate condition. Excuse me, the mental state, that is whether or not the person wanted to do a light doing that on the first the pram. Just whether or not they knew the stereotypes and endorsed them. They were as you can see, virtually the same. So even from three years on anybody can do this task and from seven years years on any autistic kid can do the task as opposed to mentalizing tasks can't do. This was the most important pattern the three year olds and the false belief failing autistic kids were greater than chance, much greater than chance and differently likely to choose the social role. That is to explain to choose the person that fulfilled the common stereotype even if it went against their mental state, even if they didn't want to do it, this is what they were going to do because they were members of category X. So in a sense they're casting fear for social group membership almost, but not necessarily before. In the case of the three years before they're interested in mental states to some extent if there are questions just so just to sum this up, the kids are equally likely to use social group membership and interpreting and predicting behavior. Their even those with limited capacity are more likely or more likely than normal to choose the social status rather than the mental state. And I didn't mention this, but these were kids. All of the autistic kids were in specialized schools. One of the things they learn in specialized schools is to strategies for acting as if they could mentalize acting as if they could appreciate other people's mental states as the cause of their behavior. They don't teach them how to stereotype. They don't teach them nasty racist prejudice not entirely surprisingly, but they don't get any training in Doing this they seem to be doing on their own. So just where could this come from? My argument is going to be that it's that or has been that it's coming from some kind of device, distinct mental device that favors reasoning about humankind of some sorts as a way of explaining behavior. Well, you know, all right, they're not taught it at the school, but maybe other people taught it to them. And that could be true of the gender stereotypes. They were really obvious. And maybe, you know, I don't think it's true, but maybe in 2005, whenever we collected this data, these data, British women did prefer dolls and cooking to being, you know, firefighters or something. I have doubts, but you know, maybe just by chance they met the wrong people. The other I think is more likely is that there's some kind of mediated learning, that the learning of this kind of material is guided by modular device. And the probably most important aspect of the data in this regard is that the racial items they learned they were judging were not true. They hadn't been running into black kids who were drowning puppies or black kids who steal wallets more than white kids. I mean, these were not. These weren't common stereotypes about their behavior, something that you could learn possibly by looking at them. These were inferences from some kind of other expectation about the differences between blacks and whites. And another reason to expect that they weren't learning them this way is that people don't talk like that among each other terribly often. And certainly with their kids, I mean, they don't sit them down and say, let me tell you about X, Y and Z. In fact, white people in America don't talk about their kids about race at all. One of the striking things is that they just avoid the topic and the kinds of interactions that characterize racial discrimination in many settings. And I would argue probably most settings in advanced capitalist societies are ones that are hidden, that are non obvious. The giving the black candidate an interview that's five minutes less long, the white candidate, the banker that demands more, a better credit rating, maybe just slightly, but enough to change the housing market for the black candidate than the white candidate. Many of these things are sufficiently hidden so that people can actually deny that they're doing them. And there's a whole literature on aversive racism that, that supports this idea and that people's, the attitudes they report are not typically the attitudes that they hold. So that doesn't seem to be a place where they would have learned it. And as I said, there's no training for them to Learn these things. It's not something that they would have been the school would have paid attention to teach them. And because of the. I won't go into it here but it has to do with how many in group and how many out groups judgments they make. It doesn't seem to be something where the kid is going well that kid's like me and this kid isn't. Our study was too small to have actually tested minority kids. But in other studies I and many other people have tested minority kids on these kinds of not the conflict task but the other tasks, the one where the puppy drowning task. And what you find is that the black and white kids judgments are not mirror image of each other. The judgments are not one of in group versus out group. The judgments are one of majority cultural group versus minority cultural group so that the black kids will choose the white the same way that the white kids choose the black doll to the same the same item. So that makes sense.
- [50:58]
C
It is that.
- [51:02]
B
Their judgments are not simply mine versus others. They're I'm somehow figuring out the nature of society and the nature of society is that that group does better than my group. We're going to. I don't think that any of these. These are largely. I think I can take them out there close enough to redundant to be taken out. Modules are precisely because they're modules, because they're distinct programs for learning don't necessarily look like each other, don't have the same scope. But one of the things that seems to be true of many of them is that they play out. They guide development in a way that is often significantly insensitive to the environment. So the environmental differences for instance is from cultural settings don't change the order or age at which thresholds are met. So one study of pygmy children found that their theory of mind reasoning on a or something like the selling end task was the same as kids in the West. But a theory of society module would have to be different because the things that get picked out, like race are things that we invent. Or if you want to take a realist position, realist in quotes, I.e. you believe that races actually exist in the world in the way that folk belief has it. They didn't exist in the world in which our ancestors evolved. People didn't start seeing people that looked very different from each other until they started having long distance overseas travel. So it doesn't make sense to say that we somehow have an evolved device to think about race. We couldn't have an evolved device to Think about it. It would be like having an evolved device to think about spaceships. It violates the order of things. So each, each culture, each society will have its own kind of humankind, own kind of essentialized groups, and that the child has to learn them in order to be a competent actor in their particular environment. So race would presumably not be a terribly useful thing to attend to in South Asia, whereas caste would be and there and age in other parts, in parts of sub Saharan Africa. So that the cultural information seems to be particularly important to the theory of society module. And I'll try to give you an explanation of why. And that's why I think that the cultural adaptation here is to specifically human experience. If there's this theory of society, there's this theory of society module about humankind, it's about a centralized group, groups. It's not going to be something that is represented in some incipient form in non modern human populations. How would kids learn about it then? Well, there are a couple of problems because one, not only might they not hear everything that adults believe, they're not in a position to have witnessed lots of the behaviors that would give them, deliver them the concepts like racial stereotypes. So in order to learn them, presumably you have to do something like the kid has to do something like the following. It has to take infrequently encountered information and weigh it more in terms of building these conceptual categories than frequently encountered variation. Now we know that kids do this in other domains. The dramatic one I think that you can see is in foreign accent children of parents who are non native speakers of whatever language, let's say children of non native speakers in England develop native speaker accents. And they seem to develop them from a very, very early age. They don't need guidance. Somebody doesn't need to come in and say, by the way, don't talk like mom and pop, talk like this. And presumably they're doing that because what they're doing is looking out to see what the community does, what the community believes. And the same thing seems to be true of racial attitudes. Parents and children's racial attitudes do not correlate at all. Parents that try to change, to try to raise their children to not have common stereotypes have no effect either. So even parental interventions make no difference. There's one study that I haven't been able to find the full reference to. So if anybody here knows it, I'd like to pass it on and be grateful. But the study shows that parents who try to teach their daughters counter gender stereotypes have daughters who Learn common stereotypes earlier. And again it. If you were evolving in a certain kind of society, certain kind of cultural environment, it would be good to pay attention to the community rather than pay attention to local and idiosyncratic standards. And that's what the kids seem to do. And one of the interesting things about it is that they do this. The autistic kids do it in learning conditions which seem to be completely the worst. You try to pick out non optimal learning conditions about social behavior live like a significantly impaired autistic child. The ability to scan the environment presumably must be something at a low conceptual level. I don't mean low level in the sense of similarity, but a low level of conceptual attention because it can get it without literally without trying presumably. All right, so this has the last thing I'll hurry through it because this is what I promised in the title. So I have to do this. Could this module be the supposed module, presumed module, claim module be an adaptation to reoccurring evolutionary problems? The long held theory is that we don't like others people from other groups. The us and them part because we're protecting ourselves from predation from this groups. The evidence actually for that is as compelling as it seems on first pass isn't very good. For one thing, a great many of the naturalized groups like gender and age are functions of intra group variation, not intergroup variation. I'll skip a couple of them here. Probably one of the most important things is the in group varian for human in group out group judgments is not the small band. So it's not that your band is protecting you from predation because you don't even consider your band to be the scope of the in group. I'm going to leave that for a second. People I don't think were Hobbesian brutes. And I suppose in this way I part company with some evolutionary psychologists. There's a lot of evidence that this wasn't the case. Although of course with anything like this it's indirect. First contact reports describe people zipping along and mixing in ways that seem completely implausible on the idea of us and thems and tribes and stuff and. And in fact the evidence for warfare which it's been argued there was a lot of in the ancestral environment. In fact there wasn't apparently a lot of it until more recent prehistory. This is the. Our pre cortex is larger. There's a. Dunbar's done the work showing that the size of the pre cortex neocortex is correlated with band size and humans have neocortexes which are bigger than bands which, because we have them that are about easily working with 150 people. Nobody lived in bands with 150 people. Let me see. I'll do this as fast as I can. There is, however, evidence that. That humans did in our ancestral environment, do a lot of moving around, that they didn't seem to be in these distinct groups everywhere. And the going through it too fast. The idea that predation, you needed predation because of scarce resources. Doesn't seem to hold true in terms of how at least relatively modern hunter gatherer populations live. I mean, remember that a lot of, you know, at least for one group of modern humans, they evolved in places that we go for vacations, Riviera. What I would argue is that this thing that creates us as and them is as good as including people, as excluding people. As long as the people that you include are people that are outside your band. That is, you create in groups that are larger than. Than the on the ground, face to face groups that you live in. And there's lots of evidence that this is true. And it would also explain why we can easily know 150 people. Because you could have 150 people spread out over a wide space. And even though you didn't see them often or at all, you could still count them as in groups. So how would you get strangers into your in group? Well, you could do it through training networks, which I suggested was important in the ancestral environment. And that's true. You could create bonds of kinship. I don't suppose anybody talks about exchanging women anymore in anthropology, but I was trained when we still read Levy's Trust. So the different groups, bands exchange. They create bands through marriage. And they also create groups that are imagined. That is, they're bigger than anybody necessarily ever sees. And then they essentialize these people. So what you do is you create a sense of closeness with people who you don't see very often. And so rather than being a way of excluding people and protecting yourself from predation, what you do is you do something that allows you to get salt. When you live far inland, you create in groups that go over time and space. Go over space which allow you to include people that you would other seldom see in your own. Well, I guess I better stop, right? Yeah, stop. Maybe it'll come up in questions.
- [64:41]
A
Well, thank you very much, Larry, for a very good overview of your research and some stimulating thoughts. Especially at the end. I thought the idea of the in group being the out group being created as a. As a way of creating an in group. But in any case, before I ask any question, I want to ask people in the audience and I'm going to take. I think that's my role. I'm going to take questions. But you are asked to speak through the microphone so that everyone can hear you. Here at the front, please. And if you could say who you.
- [65:35]
B
Are.
- [65:38]
A
Before you ask your question.
- [65:40]
B
Thank you.
- [65:41]
D
I'm Andy Wells from Social Psychology. I'm not sure that I understand the general logic of your argument for a modular device because you started off by suggesting that there would be a model module. And then you gave us lots of experimental reasons for thinking that there might be. And then it seemed to me that you contradicted all of that by giving us all sorts of reasons as to why we wouldn't expect an adaptive path to such a module. So I was left a bit puzzled about exactly where the selected pressure for the module came from.
- [66:23]
B
Sure. I'm sorry, that was the. What I was arguing is the selective pressure was not protection against predation. The selective pressure was the ability to exploit environments that would be otherwise difficult to exploit without basically having trade routes and without having ways of bringing together small bands, you know, trans nomads, kinds of things to together at certain times to exploit, to share in exploiting environments like you do in the winter, the animals and so on. Without. What it does is it creates a way of seeing those people as uss, if you will. And that. That would have been the adaptive. That would have provided the pressure for the adaptation rather than the pressure to protect yourself from other groups who were likely to be predators. Didn't work. Oh, that. Oh, sorry. That's a very good point. We get the thems when we start getting settlements and agriculture and the sizes grow up so that the nature of competition for groups to exploit changes. And the kinds of rank and inequalities which characterize small scale societies start to get very big. You get much more concentration of power and fewer hands because of the nature of the environment. And what started out as a great way to foster inclusion becomes a great way to foster exclusion. So it's the bad news. And it wasn't the bad news of our ancestors, it was the bad news of, well, our ancestors, but closer ancestors. Sorry, I should have said that. I did skip through this a little too fast. Obviously.
- [68:28]
A
Here at the front.
- [68:29]
E
Please.
- [68:33]
F
Olivia Harris, Anthropology. I was very intrigued.
- [68:36]
A
Hi.
- [68:37]
B
Hi.
- [68:38]
F
Many years. By the concentration on black kids versus, or rather white kids versus black kids. And Please. That you did towards the end, talk about the reactions of black kids in Your studies. But from an anthropological point of view, there are so many other issues going on in a world white, black opposition that I would be interested to know whether you produce the same results in the French context and whether you were using white versus North African, whether you were always using white as the default. Because that seems to me to bring in so many other issues that it would be much more convincing if you were to do it, say in India.
- [69:18]
G
Where the kind of.
- [69:22]
F
Things, the visual differences were not so obvious.
- [69:26]
B
There have been studies now in other places where it's not white versus black in. I can't remember where she worked, but I think it was in Bali where you looked at groups that. Where you got the political power holders being like the whites and the Americans in the American situation. There are other groups that we've looked at in the United States like Hispanics and we get similar kinds of things. Of course, most races don't look different. They are presumed to look different. You know, read 19th century, early 20th century accounts of Jews in the United States and you don't have to go there. You can go to Shakespeare and you have a description of Shylock as literally a comic figure because he's so ugly. But the difference is it probably wasn't triggered by actual physical differences or if the physical differences involved were probably things like costume more than physical, you know, embodied features. And there's been work with that finds the same patterns with Israeli and Arab kids and stuff like that. But the more important question is why. Why would power be so important here? Why would power and race link up in this way that makes it. While it's not the only system, it certainly is a system that sold well. And my argument is that race is very easy to think it is in virtue of having, in virtue of being part of our built in common sense, the kinds of things that we attribute to race, our expectations that we already come with to the world and that once you have that, you. Rather than race being a psychologically salient, important category. I'm sorry, yes. Rather than being a psychologically salient human category because it's a politically salient category, I would argue it's the other way around. That is because it's used in power so readily. Because it's so easy to think we could do, you know, we could do class. But people don't learn class quite with the same ease that they seem to learn race. I mean, the kinds of properties that get associated with race are properties that allow not only race to be a handy way to distribute power, but a way of distorting the distribution of power. I don't know if that. I may have misunderstood your question. No, I don't think there's anything intrinsic to the argument. It's the way I have done a lot of the studies because it's the way. Way the thing plays out in the United States, largely, as I said, there was one study among French kids. But I don't think there's necessarily anything about the argument that requires that it be black and white. Well, but that's. No, my argument is that race isn't salient because kids see it. If my argument was that, then you're right, it would be a serious problem. My argument is that that fact is, to the extent that it's a fact is a later emerging quality of cognition rather than the thing that drives the cognition. What drives the cognition is the expectation that you need to find the groups out there. And you find them largely by listening to people talk and learning labels. And then you go around and figure out what it is that makes people belong to the category in the United States. It's, you know, it's not the same anymore, but it's largely, you know, significantly black and white. But that's just an accident of our history. I don't think the argument. The argument shouldn't settle on that at all. I don't mean it to. At least I hope it does.
- [73:45]
E
Thank you. My name is Dieter. I'm from Grinnell University. You were talking a bit about language as well, and you talked about how children tend to adopt the accent that is the majority in the community, rather than taking on their parents accent. So I guess the parents is a kind of us, even for the minority in the community as them or you could see, say the opposite. But how would you account for all the syncretic processes that are in language as well? When creole languages are developed, or even in many Western European countries, immigrant children, for example, create a new kind of language. So that's like some sort of in between us and them, which make it less clear.
- [74:30]
B
Language is an enormously important quality of communities, and people use it in part because as humans we know this tremendous amount about it because of the peculiar nature of the way we learn language. And it's used all the time to set up different communities in virtue of the way we speak. I live next door in Ann Arbor, lived next door to a black family, and they had four children. Each of the children spoke a different kind of sub accent that ranged from something that was indistinguishable from the kids. At the university to ones that were sort of hip hop oriented. The language was part of their negotiation of society and the groups that they wanted to be associated with. You write, the US and them has worked out to be an unfortunate aspect of something that I think to repeat was a solution of a problem for how to make more of us. And so when you look at the kinds of any number of kinds of tasks that measure prejudice, you'll find that in group favoritism is much stronger than out group enmity. It's just that out group enmity makes a great way, is a great way to regulate power and authority. The language politics are still about power and authority, but the stakes are a little lower. I don't know if that answers this. I wanted to mention. Well, I'll wait for another question that begs it more.
- [76:17]
A
Here.
- [76:20]
G
Here at the.
- [76:20]
A
Front and then at the back, here at the first row, please.
- [76:26]
G
Senor Javcelovich, Social psychology. I have a question to you about the theory of society module, and perhaps it's related to the question we just had. I want to ask you, if you think in the way you're conceptualizing in that possible module, the us and then dichotomy, or the us and then categorization system would be the most preponderant or indeed the only way through which theory of society would develop in children. And I'm thinking, I'm asking you this because I'm thinking especially about vast area of state studies in developmental and social psychology that have tried to understand the nature of children's engagement with society and what we call the development of societal knowledge in children. So how they come to understand institutions, how they come to understand the functioning of social norms, of rules in a way that go far beyond a system of categorization that would juxtapose a us and man. So I was wondering if your theory of, you know, the whole area of research, of theory of mind, in a way is redolent of the early Piagetian studies on perspective taking and on the, indeed, Herbert Mead's idea of developing of the role of the other. So I wonder if the idea of fear of society now could be perhaps integrated in this tradition of what we call children's development of societal knowledge.
- [78:07]
B
Yeah, but would as many people. Oh, I'm sorry, as many people have come out if I'd added more accurate. I mean, you know, so it's. There are two reasons, right? You're absolutely right. In fact, there are people here who are looking at aspects of what would be theory of society that have to do with, with political, you know, political negotiations and so on. And Gio Hatano did work on how kids learn about banking, for instance. Right? Yeah. And there are a couple of things here that I'll try to mention them as briefly as possible. But you know, when you look at the folk theories, folk theory of psychology, folk theory biology, folk theory of physics, say folk theory of society, sociology, they start looking a lot like late 19th century German universities. And the way they carve up the world, I don't think that's an accident. Not because that's the way the world is, but because it's a great way to describe a faculty that is the modules are almost certainly much, much smaller in the same way we have vision is module, but it has modules for edge detectors and things like that. What's relevant to us who experience the world without modules. That is one of the striking things about modular processing is that we don't notice that it's modular. It seems seamless to us from one kind of information to another, from one source. Source of information to another. You're absolutely right. There are many ways in which society can be carved up in terms of possible. Terms of possible modular competencies. And there are many ways that it can be carved up that have nothing to do with modules too. Occupation, which is really a, a description of habitual activities. Right. I don't think that we evolved with. I don't think we have an evolved device for habitual activities that, for which there's remuneration or something, you know, so there are both things that are a function of historical epochs, obviously. And there are also many ways in which we organize and regulate society that have nothing to do with humankind, that have to do with completely different phenomena in many ways and potentially very different causal principles. So absolutely. I just, I'm trying to be in your face.
- [80:51]
A
Okay. I think I'm going to take one more question there and then I want to ask one myself. And then we go.
- [81:00]
C
Peter Hagerty Social Psychologist at the University of Surrey I'm a social psychologist. So professionally my job is to see biological theories and propose alternatives. To say this is the fundamental attribution error at work. And of course you've courted me into that critique beautifully tonight by showing me some very compelling experimental evidence. And I'm not going to throw social complexity at you because we really have to have an alternative explanation. And your modular one is compelling and seductive. But I don't think that you've considered the full range of ways that we socialize children and I don't think you've considered the full range of ways that we socialize children to think about power, inequalities and degrees to which people are normal. So, for example, when I was listening to your talk tonight, I couldn't help but notice that There were about 20 times where you said people where you really meant white people. And I couldn't help but notice that There were about 20 times when you said race, a racial group, when really you just meant black people. Right. So implicitly listening to you, if I was a child, I would have gotten about 40 cues that normal kinds of people are white people and abnormal kinds of people are black people. Now, I've got some data on my hands at the moment, gathered by one of your collaborators, actually, in which white and black people are talking about race differences. And you see the same difference there just as you do in experiments, just as you do implicit measures. And there are a billion theories that tell you that part of the experience of whiteness is about feeling that you don't have a race at all. So might it not be possible that the thing that is happening, that these three year olds have learned, just as they have learned that there are racial categories at all, is that some kind of racial categories are marked in some kind of racial categories are unmarked and that that in and of itself is efficient. And so there is no module to divide people into us and them. There is no module. But actually this is all being done perhaps in a median way through socialization of language. Because what seems to be missing from the account of human evolution here is the fact that humans are distinctive language users who symbolize stuff.
- [83:01]
B
Yeah, absolutely right. And thank you for pointing out one of the least attractions attractive qualities in my presentation. No, I agree. It has to be in virtue of the kind of information you're talking about that they're learning this kind of from speech. And speech is unbelievably rich and provocative. And the kinds of effects that I got in with typical developing children I think are good evidence, but not, as I said, knockdown evidence. I think that if we can get a double dissociation between impairment and stereotyping, it would be really, really knockdown evidence. And one of the things that I'm planning to do, it's great when you answer questions by telling them something you're going to do. But is test Williams syndrome kids who you think might be good with theory of mind but bad with theory of society or the kind of theory of society I'm talking about, I also would again, but throw back at you the kind of experience that at least some of these kids are getting in, the autistic kids they're in, I can't tell you. It's true and it would be something certainly worth studying, but I would be surprised if they were in the position to listen to me in my talk that way. In terms of an active, in terms of an active attention. I want to listen to him. And I think that what they need is some kind of motivation to do it. And what they need is also some kind of weighing of the importance of the information so that it creates norms of community rather than qualities of individuals or qualities of very small groupings. You know, I'll grant that it's. That it's not a finished project. And for those of you who are graduate students, and I will do anything you want if you do work, because I think there's lots and lots of possible work that could be done, but there are still possible alternative explanations. And I think I have pretty good challenges to them, but they're certainly not an reasonable. And the other thing that I think is just the argument about the a priori argument about modules. Why would it be that we would learn about these kinds of things so differently than we learn about other things when this is really what we think about all the time? Listen to humans, they talk about each other non stop, right? So you're right, they get lots of information. But wouldn't it be odd if we evolved not to care about that in a way, you know, not to be specifically interested in that? So.
- [86:19]
A
Okay, I just want to do some mind reading here and you're very privileged to have in the audience lots of first year anthropology students. And I'm trying to ask a question on their behalf. So you are a first year undergraduate student in anthropology and you are telling us that we might have a toss. And I would like to ask you to spell out for them why as anthropologists we should care whether we do or not.
- [86:55]
B
Well, I can't. I can't possibly answer that. In a sense.
- [86:58]
G
Yeah, one minute.
- [86:58]
B
I have one minute. Okay, good. There's no shortage of good things to know about and to study and for which we have lots of open questions. So I can't answer that in that respect. I think though that for students in anthropology, it's particularly important to recognize that the way our minds are structured is not trivial precondition to the way our societies are structured and that the way we learn to be in the world and come to learn to be in the world, both individually and across generations is in many ways a function of how the child learns about the world rather than the other way around. You know, borrow this from Judith Harris, but, you know, you see a good parent and you see a good kid and you say good parenting makes good kids. After being a parent now for more than two decades, I know perfectly well that a good kid makes a good parent. In the same way that if I play tennis with a good tennis player, I look pretty good, and if I am raising a kid, I look pretty good. If the kid looks pretty good, the adults staggeringly overestimate the impact they have on children and hugely underestimate the impact that children have on adults. So if I were. Let's move a minute ago, if I were recommending why you would care about this as an anthropologist, I would say that if you want to understand why we're doing what we're doing and why society matters, why the structure of society and the trajectories of society matter, why things like regimes of truth are important, then it would be good to learn how it is that we invent regimes of truth. And we do that because in significant measure, because the child finds it easy to learn that, and not recognizing that, I think, allows us to make all sorts of questionable attributions.
- [89:29]
A
Great. Thank you very much. On behalf of the first year students and everyone else in the audience, thank you.