Robin:

We've decided this time to revisit a previous podcast episode, the one on experts and elites It was initially, Agnes was saying that she wasn't persuaded so much by that. That's what she told me a few weeks ago, I guess, and wanted to talk about it more. And we thought we might discuss it together here on the podcast. So we've both gone over the transcript at least again. And I can say that when I go over the transcript, I see that I was making an assumption that you were challenging. It's assumption that's reasonable to challenge and that I should have noticed and tried to discuss that more, i.e. to evaluate that assumption. I think it still is true, but I noticed that it's an assumption that many people have challenged in my life. So for example, years ago, my colleague, Alex Tabarrok, When I brought up the subject of social status, which is a common concept used in social science, he said he just didn't think there was such a thing. It was this made-up word to deal with a bunch of more specific things that made sense to him, but he didn't think there was this general concept. And I just read an interesting paper called Rethinking Prestige Bias by Azita Shalapo, published in 2021 in Synthese. This is also, this paper is challenging cultural evolution primarily through that same move, which I see you as having made in our previous discussion, which is why can't we understand a lot of ways we treat each other as rational ways we react to particulars, i.e. particular things other people might know, particular reasons we might have for copying their behavior. as long as we are reasonable people who use reasons for many things, isn't that the reasonable assumption to make about behavior? And if we can assume that people make copy other people's behavior because of specific indications that they have specific skills, then there becomes less of a place for this general concept of prestige and prestige bias in cultural evolution. And like my colleague Alex was basically intuiting years ago, What's the point of this abstract concept of status when we plausibly have much more concrete indications of other people's powers, skills, resources, contacts, et cetera, that we could use to make social judgments about them? So challenging the idea that there is this abstract concept of status prestige that's worth thinking about at that level of abstraction. I mean, we might use it as a quick summary for these other more detailed things, but the question is, to what extent do humans in particular seek to create this estimate about other people, their overall status or prestige, and then use that primarily to make judgments about them instead of making more particular judgments based on more particular context. So that's the assumption I see. I made that the literature, lots of sociology literature does make, the cultural evolution literature does make, but it's a sensible something you can question and challenge. And again, this paper does that. My colleague did that. And I think many economists that I've known all through my life, their knee jerk reaction to hearing this status thing is, that's kind of an abstract fuzzy thing. Why do we need that? Why can't we think about particulars?

Agnes:

OK, so I have my own thoughts about our conversation, but before that, I just actually want to follow up question on this idea of overall status. It still seems to me that even when you want to use the idea of status, it's restricted. It's not overall, it's restricted to a context. So let's take like you know, I don't know, the manager of a company or something, like, you know, relative to, like, the lower-level employees, that person has status, and that status might be, you know, some kind of sum of a bunch of their powers or whatever, but it's not gonna be relevant, like, how nice their wife is to them or something, or, like, how much status they have in their country club, or, well, there's gonna be, things that, or if they have like a hobby of painting, how good are they at that hobby? So it seems like it's not actually overall, it's in a given context, isn't it?

Robin:

So, I mean, more precisely, we might ask, you know, at what level of abstraction is the concept operating at? So think of law schools. We might say Harvard Law School is the most prestigious law school. And you might say, first of all, that's prestige in law, not in, you know, clowning or something. And secondly, you might say, well, in fact, Harvard Law School might have distinct prestige in corporate law or contract law or constitutional law. And you might question the idea that there is such an overall prestige of Harvard Law School relative to many particular prestiges. So we could ask, you know, there'd be some level of at which the prestige concepts would operate. It might not be all possible prestige, although you might say if you're thinking of inviting somebody to a dinner party or deciding who your daughter could marry, you might invoke a pretty high-level overall prestige concept, but then in other cases you would use a bit more particular. But still, we could ask, well, at what level is Harvard prestigious? Is it at the level of a particular recent area of constitutional law? Is it at the area of constitutional law? Is it at the area of law entirely? Or is Harvard University prestigious in education in general, not just particularly in law?

Agnes:

But even if it were prestigious in education in general, it wouldn't be prestigious in everything. It would be in education.

Robin:

But we might say that a particular person had an overall level of prestige, and part of that was because they went to Harvard. But there is the overall prestige of a particular person, and that might be a coherent concept, i.e., inviting them to a dinner party, having your daughter marry them. Is there a prestige of a person concept that people use and make decisions based on?

Agnes:

Yeah, I mean, I guess I think that seems like an additional claim. The idea that there's an overall prestige, like it seems to me who you want your daughter to marry, there might be lots of things like how much prestige they have in their gym, whether they are running faster than the other people in their gym or slower, that just might be irrelevant to you. And, you know, you might really care about, like, their prestige in their field, or you might not as in terms of who's going to marry your daughter. So it seems to me like the overall calculation, there may be some very rarefied context in which some group of people is inclined to make the overall calculation, but it seems to me on the whole we don't make such a calculation. That is, there's just tons of stuff we're leaving out all the time, and so we're interested in the person's prestige as a dinner party guest, which isn't their total prestige.

Robin:

So, I mean, I think I disagree a bit there, but I'm initially granting that I was too quick to assume that you would agree with me.

Agnes:

Okay.

Robin:

And that it's worth pursuing.

Agnes:

I see. Okay, so I think that's really an additional issue. You could agree that there are elites, because they're going to be elite at something, as I would see it, and that they're different from experts in some way, but still think that there's no such thing as eliteness simpliciter. There's no such thing as expertise simpliciter. There's no person who's just an expert at everything. That's not a kind that you can be.

Robin:

Right, but there might be a kind of eliteness in general. We could go into that discussion.

Agnes:

Okay, so that's one question. Is there eliteness in general? And then, but I wanna actually go, before we go into that, so my takeaway from our conversation, is that my skepticism over the concept of elite is between two possible ways that it might be defined. And I see myself throughout the conversation as trying to get you to tell me which of these two ways you have a thinking about it. And so it's not that I think it's too abstract or fuzzy. It's just that there are two things we might mean by it. And I'm sort of not really satisfied by either one. So one of them is that Um, an elite person is elite because they are, um, good at something where the thing they're good at isn't, can't be defined as getting other people to think they're good at that thing. That is not going to be circular. They're actually going to be good. It's the most straightforward way to get people to think you're good at something is to be good at that thing. So if you're an expert, you have to get other people to think you're good at the thing you're an expert in. The way you do that is. by being good at the thing. And so one way of understanding elites is they have some skills. Like they could be social skills, they could be managerial skills, they could be ways of being and operating in a hierarchy and dealing with difficult social situations that other people would find too hard to manage. And so they're genuinely skilled, they have some skills. And because they have those skills, people believe they have them. That is, they exercise them and then it becomes evident. And, you know, I think we could add, like, they have those skills and maybe part of what it is to have those skills is to be good at getting other people to see that you have them. So that'd be the first way of defining it. There's some actual things that they're good at. And then I want us to be able to say what those things are. The second possibility is the, no, they're just somehow good at convincing other people of something where that thing doesn't need to be the case. They're good at projecting an image or, and, A lot of times you sort of talk that way, not necessarily that the elites are deceiving people, but that their primary skill lies in something like, yeah, convincing other people that they're elites. And I find that unsatisfying, like that can't be a kind of person, as far as I can tell. Okay, so there's like the skill model, there's the just convincing other people that you have the thing, which is going to be circular as a definition of thing, and there's a third possibility that just occurred to me actually, which is like elites are the morally better people. That is, they're superior. They're ethically superior. I think that's an important contender because if you, through most of history, if you asked some random person in the society to say, what makes someone an elite? They would say, well, they have virtues of justice, generosity, courage, moderation. That's what you have to have to be a leader. That is, maybe that will be, they're leaders, and leaders have to be especially good people. And so that's what an elite is. Okay, so these are three possibilities of ways of understanding an elite. And I wanna kinda get clear on what we're talking about.

Robin:

Okay, so first I'm going to talk about a related thing and then go back to this. So I first want to identify the fact that people often coordinate together on shared social constructs. So we might together decide, you know, what is bedtime or the time to be quiet in an apartment. We might decide what's the appropriate speed on a freeway. In a group, we might decide our mood, are we serious or are we playful? There's just many social occasions on which a group of people needs to coordinate on some shared estimate. Now that shared estimate reflects some reality in the world, but still the key thing is it's important that we share the estimate. Right. So, you know, for example, is it about to rain? Maybe a group of people would have to, would want to create a shared estimate of our shared sense of, is it about to rain? Are we going to go out and do things outside? Are we going to stay inside to do something? Right? That shared estimate is about a real thing. Is it going to rain? But there is this other construct of what do we, what is our shared estimate about this? Because we each need to know what that is so that we can react to it. And then we coordinate together through that shared estimate. It's a shared estimate of something, but still the shared estimate is, has a separate reality to it. It's a, it's a thing that matters separately from, is it actually going to rain? Similarly to the speed limit, et cetera, like that. So the idea is that status is a thing like that. It's a thing that we want to agree on, and it's useful to agree on. It may, of course, also be intended to reflect something else, but an important part of its function is that we agree on it. that is in a group of people, not only do we want to agree on like what mood we're in or whether we're going to go outside, but we might want to agree on who among us is higher or lower in some pecking order. And of course, most animals do create pecking orders that they share. So this isn't unique to humans. And part of the function of the pecking order is, and I might reflect who would win with who in a fight, but There can be the perceived ranking in a pecking order that might be different than who would actually win in the fight. And the key idea is the group needs to come to some sort of agreement about that pecking order or the purpose of deciding who takes limited resources, has limited responsibility for certain positions. There's a bunch of ways we use status to ordinate socially.

Agnes:

Right. So can I just ask a question about that? So if we were going to agree about like the weather and come to a group view, we do this sometimes in my household where we're deciding are we going to take the car or are we going to walk in and we have a shared view and we will check our phones. And does the phone say it's going to rain or not would be the basis of a shared view. It would be very weird if we just made something up about whether it's going to rain or not. That's not going to work out very well for us over time. We're going to base our shared estimate on something. And so the question would be, similarly, we're not just going to form some shared estimate on who's better. You'd think we'd base it on who's actually better.

Robin:

If you could see that directly. But in the world when there weren't weather apps, or where the weather was unpredictable or more difficult, you would find it harder to coordinate an opinion about the weather, but you would use some weather markers. some indications, the temperature, humidity, do you hear rain? You know, so was it raining an hour ago? You'd have a set of markers, and then you would probably share some relative weights on those markers. That is, you might agree that if it sounds like it's raining, it's probably raining, or if it rained an hour ago, it's a pretty good chance it will be raining. You would then have some weather markers that would be the indications of weather.

Agnes:

Those would be an indicate, like we would be trying to track the reality. We'd be doing our best. Our best might not be very good at all. We'd be doing our best to track. the reality of whether it's going to rain or not. So similarly, when we're deciding on who should rule, we're doing our best to track some property that would be in some way correlated that you should rule, like the person, for example.

Robin:

Sure. Now, say we're trying to decide whether to attack the neighboring tribe or something, and we wonder whether the signs are auspicious for it. We might look for an eclipse or a strange animal or other things as signs of whether the gods are encouraging us to fight our neighbors or something. In that case, we would also probably have a set of shared markers that we agreed on that would be indications of whether it was an auspicious time to attack the neighbors. But there may not be actually some true relationship between these signs and the actual you know, auspiciousness of attacking, but it still might be functional for the group to have something they agree on, because then at least we all go together, or we don't go together. So there's often advantages in coordinating, even when it's harder to attach to a reality. We might not know if there's reality. We might believe that there really are gods who are giving us signs, or we might not. But even so, it's worth having some shared agreement there. So I just want to point out that you might not be able to tell as easily what is the reality behind it or whether there is one so much when in our cases when it's just valuable to coordinate and agree on things.

Agnes:

Well, I think you would think you could tell. That is, you would think, in fact, that the eclipse is a sign of this thing. And so you can tell by the fact that there's an eclipse. You would always think that. That is, you would never use the eclipse unless you thought it actually is a sign. Right.

Robin:

But if we disagreed about how important an eclipse was, then we might not have a way to settle that dispute in our interactions. But we still might want to know what most people thought, even if we personally disagreed with it.

Agnes:

I think that if, you know, people started to doubt whether eclipses really were correlated with the gods telling you that you should fight, then that would have to be, like, eventually there would have to be some kind of reckoning. Because what the group is trying to do is track a reality of whether or not the gods actually think we should fight. And it's only insofar as at least most people think it really is. That is, people I don't think are mostly interested in what other people think. They're mostly interested in what is the right answer. Do the gods want us to fight or not? They have to come to that answer as a group in order to make a decision as a group.

Robin:

So divination is an ancient, very common tradition where people would roll dice or throw bones or do various things and read the tea leaves, as they say, to see signs for whether they should attack or get married, et cetera. And, you know, a standard story is that people might not be very sure there is much to this thing, but they might see a social benefit from just coordinating around a set of rules about what the divination says.

Agnes:

I guess that doesn't seem plausible to me. Like when I the discussions I've heard of divination in the ancient world, there's never nothing that I've ever read that's compatible with the interpretation that you've given, which is like, well, there's a social benefit to this, even if it's bullshit. People just think that it really is going to tell them. That's how you get the social benefit.

Robin:

But still, if you were to challenge and question that, it might be just hard to penetrate. You might not be able to get very far questioning the widespread belief that the divination was indicating something. That's the thing I was trying to point out here.

Agnes:

their arguments over divination are pretty common. And like, does this omen really mean this? People did challenge that sort of thing. Because it wasn't very clear.

Robin:

I mean, I just made my one point. I don't mean to belabor it. So let's go back to status then. We might say there is some value in ranking people And people do that. That's actually a common thing they do. They'd like to have a shared ranking. And the shared ranking, we can look at what they're used for.

Agnes:

What are we ranking them in? That's my question.

Robin:

Well, we can look at how they're used to try to figure that out. So we're just being social scientists here. We're just seeing that people do rank each other, or animals rank each other in a pecking order, for example. So when animals rank each other in a pecking order, their primary function there could be like, who should I fight for something or who should I not fight? And we might say, you know, top chicken takes the top chicken wife and the rest of us don't fight him because we've decided that, you know, the top gets top. And that's our way of keeping the peace. And then a pecking order helps establish that. And so if there's only five good slots, the top five will get them and people below that won't get them. They're instead of fighting, they're using the pecking order to allocate, you know, scarce resources. So that's what happens in pecking orders of animals. And then, um, you know, we might say, what do we use status for? One thing we use it as to figure out who's our leader, because often we need to make choices together and somebody needs to be the person who initiates and suggests which choices we all make. And then the high end status can be the leader suggesting ordination. High-end status, also like the chicken, might have conflicts over mates or resources, and we might let the high-end status win those conflicts. Then we have, you know, select groups. If we're going to make a committee on something in a department, we might want to decide, well, let's let the most elite people in the department all be on a committee because it's an important decision. We need to make sure, you know, the important voices were heard. in making that decision. So we might, by eliteness, choose who goes on an executive committee. We might choose, on eliteness, who to invite to an elite party. We have a number of different times where we need to rank people. for various purposes, and each purpose might have, you know, a distinct different set of criteria that would be relevant for making that choice. But if the correlation is high enough between the criteria for different kinds of decisions, it might be handy just to have a single overall ranking that's the first thing you go to, and then you might ask yourself, is that good enough or should we just use that? So you can see why there might be a reason to have shared rankings and to know what the shared ranking is, and that could be a way to settle many kinds of disputes that might otherwise cause a lot of conflict and time and anxiety, and we just have this simple way to settle disputes.

Agnes:

In the case of the chickens, what they're doing, the pecking order is marks who would win against whom in a fight. And so it seems to me that that is gonna be relevant for, so that there, that's a property that we're picking out, right? So that's already gonna satisfy me, that is elite, the elite chicken has, is believed to have by the group, right? And very likely because they actually do have an ability to fight other chickens and defeat them. And so I'd say the analog would be like the elite people probably have some abilities. some powers, some genuine powers and skills. Now, what are those powers and skills and abilities? Well, you're probably gonna need to look at the context in which they're elite. So, if you look who's gonna get selected to the top parties, I think it's gonna be people who are cool.

Robin:

What does that mean? Is that much more well-defined than high status?

Agnes:

I don't know that it's well-defined, but it doesn't... Well-described, even characterized. So I think the people who are cool are not necessarily gonna be in charge of it. I mean, the person I know of who is the coolest person I know. Not me. It's not you. But you definitely want him at any party. He is in charge of very few things. I think right now he actually may be captaining a boat somewhere in the Pacific. He has had many, many different jobs. He rides a motorcycle. He's very good looking. He only dates models. But he's not rich or powerful. He just gets models to date him because he's that cool. So that's just, he's got his own way of being the guy you want to have at a party. He's pretty interesting to talk to.

Robin:

Okay, but I could play the same game and reverse it and say, well, let's think of 10 different kinds of parties. Is he really the guy you want at each of 10 different kinds of parties? Or should I have a different kind of cool for each kind of party, right? How many kinds of cool are there?

Agnes:

I think that's right.

Robin:

But you can see, you find this useful concept to call them cool, right? It's not worth the bother to distinguish all the different kinds of cool all the time. Sometimes it's easier just to collapse to a single kind of cool, estimate people on the single axis and be done with it. That you would, you might be perfectly happy recommending for a party without asking me that many details about what my party's like.

Agnes:

Right. Well, I mean, relative to the place where you were going to collapse all the various kinds of status. So the point I was going to go on to make was that I think there are contexts where we put people on the committee because they're the most aggressive. And this committee, it's like the chicken committee, they're gonna have to fight, and we want the people who are going to fight. Those may not be the most esteemed scholars, so they may not be high status in some other sense. And sometimes you want the most conscientious people on the committee. So I think the way to think about it is there's just a variety of human virtues. There's a variety of ways to be good as a human being. And we put people in charge because we think they have the relevant virtues. And I guess you couldn't sum over all the virtues and just talk about the best person. And I guess when it comes to chickens, that's just a matter of whether you can beat up the other chickens. But when it comes to humans, it's actually kind of hard to say what makes a human overall good. But OK, like that, that would be what you were. And so then I wouldn't use the word status. I would just use the word virtue or goodness. That is, some people are better than others. And we have a public estimation of who are the better people. And, you know, much of that... Isn't that status?

Robin:

Well, you could just say that's what status is. You could just identify the two, if you like. But we don't know. What we see is these human behaviors and we're trying to understand what they are. And this is a key feature of social science. People just behave and they don't explain why they do what they do. And we're trying to figure out why they do what they are by watching what they do.

Agnes:

But there's a kind of skepticism that you're having here where you don't, you're not like, well, there's some people who, you know, people treat them as experts, and we don't know whether they really are experts or whether the physicists actually know anything, but, you know, the behaviors are such, we see these respect behaviors in relation to some of the experts, so these are the people that are treated as such. We're gonna be agnostic about whether the experts actually know anything. You're prepared to be like, no, the experts do know stuff, and so then why shouldn't we just be prepared to say the elites are the good people? on the same ground?

Robin:

Well, because we don't know what the concept is. We're trying to figure that out by observing the behavior. Like, it could be moral goodness, but it doesn't have to be. So let's look at it.

Agnes:

And can we go with expertise? Like, or are we just observing behavior there? Like, suppose there's somebody who claims to be an expert shoemaker. And we can observe their behavior and make shoes. And now We can also observe the leader doing leading, right, and doing a good job at it. How are these different from your point of view?

Robin:

So when we see people labeled as experts, and we see them treated as experts, and we see them behave as experts, we see a somewhat consistent pattern of their expertise is targeted to a particular topic area, say shoes. and that they are actually seen to behave well regarding shoes, that is, they make good shoes or pick good shoes, and that they are said to be good at shoes. And then, you know, those all go together and we can say, oh, well, that's what expertise on shoes is. It looks, there's a pretty, it's a simpler story to analyze because there's less complications that make it hard to analyze. I'd say most people who are said to be good at shoes you look at shoes and that's the thing they are focused on and that's the thing they're good at. And it's not so hard. Now, if you talk about somebody being good at culture, well, now an expert in culture, that's harder because it's less clear what culture exactly is. But if you talk about shoes, if you talk about cars, you talk about, you know, programming computers, there are a number of things where it seems pretty clear what it is to be an expert at that. We can look at the kinds of things they're asked to do, the kinds of ways they're evaluated, and there's a consistent story that fits among these various things that are happening that It's about that particular topic. So it's about being able to do things and know things about that particular area. That seems to be a simple story that works for what expertise is. Now you could challenge that and complicate it and maybe the simple story is wrong, but at least we have a nice start.

Agnes:

I just don't understand why you approach these two things in this very deeply asymmetrical way. I just think one can say the same thing about culture, that there are culture critics and they consistently succeed at producing pieces of culture criticism. For example, people read them, they find them insightful, they get accepted to the top journals. They, you know, get applause of various kinds, like they experience success in the avenue of, say, cultural criticism, just as the one guy makes shoes that people want to buy. Now you could say, but are they really good shoes? Maybe all the people who bought those shoes were deceived about whether those shoes were actually of good quality. Yeah, we can have the skepticism about the shoes too, but I'm just not sure why you think you get a grip in the case of the shoes and not in the case of culture.

Robin:

Well, let's just go back to the case of elites. What we observe is that when we look at the kinds of features people are using to identify elites and to call people elites, they're using a much wider range of features than they use for experts. If you have a set of celebrities at a celebrity party or panel or conference, people can be seen as celebrities and elite for a wide range of considerations. They can be especially articulate, especially rich, especially famous, especially handsome, especially energetic, especially accomplished, especially powerful. It seems like a wide range of features of people are considered relevant to their eliteness. And that's a thing to notice about eliteness is that people seem to be elite on the basis of an unusually wide range of considerations relative to, say, the experts. We're experts on shoes. We focus pretty much on the shoes. But for elites, we consider a wide range of features.

Agnes:

So here we have our two disagreements are coming together because it's only because you insist on this idea of, you know, eliteness in general, not with respect to a context that you then get into this problem, which is that we tend to make it in this, we make this really abstract judgment that I don't think we make in this terrible way using all these different features. And I think, no, when we have a cultural critic, what we do is we look at, um, you know, their cultural criticism and that's how we decide whether they're good as such. And

Robin:

That would be an expert in cultural criticism there. Now we're going to talk about eliteness. Are you claiming that eliteness is just a kind of expertise?

Agnes:

No. I mean, I guess you were saying that, well, it's one thing to know about shoes and it's another thing to know about culture. And I was just saying, it seems to me that you can know about culture and we can tell whether or not somebody knows about culture. And I guess we both watched this TV show, The Prophet. I watched it because of you, P-R-O-F-I-T, not the other way, which is what all my family thought it was for a while. And here's a guy who, I mean, I don't know how, he knows about small businesses. He knows how to improve small businesses. Now, that's a kind of expertise where there's a wide range of things that he has to know. It's not just one thing. It's not like making shoes. So there are forms of expertise that are quite general in that way. And then it's like a little bit harder to assess whether somebody actually has the expertise. And maybe the judgment of that is gonna be a little bit more difficult. But then that's maybe how we should think about you know, the eliteness of who do we put on a panel? Who do we put on a university committee? Who do we invite to a party? We invite those people not because they're generically elite, but because they have some virtues, some powers or skills, like being aggressive or being cool or being conscientious. And some of those skills involve, you know, many things put together. And so it's going to be a more approximate judgment. But that same thing is true of the guy who has the expertise of how to help small businesses succeed.

Robin:

So one of the places we can ground these distinctions is in organizational hierarchies. People at the bottom of the hierarchies are focused on relatively narrow set of tasks, and then they're judged regarding their skills and performance with respect to a narrow set of tasks. And then higher levels in the organization, those people have a wider range of knowledge and skills. and they have a wider scope of the issues they're considering. And then we find it harder to say, well, what are they being selected for? What exactly is the criteria that makes a appropriate high-level manager? That's similar to these questions where in some cases you have, you know, relatively localized set of tasks and tests that you can use to evaluate people. And then there are these other contexts where It seems like there's a much broader range of considerations. And, you know, one way to define eliteness might just be appropriateness for situations where you have a much wider range of ways you'll be evaluated or things you might need to do. But we could probably do a little better than that.

Agnes:

This reminds me of your definition of culture where it's like the stuff we don't understand. So maybe eliteness is people who are good at stuff, but we don't we can't really explain why or how. It's the hidden goodness.

Robin:

Although if somebody was very good at jumping jacks, but we didn't really know why, we still might say, well, it's jumping jacks that they're good at. That's what their expertise, even if we didn't understand how it is that they were good at jumping jacks. But here we're focused on the fact that we are not even sure what we would be looking at exactly to evaluate them.

Agnes:

I guess when I just think about some elite people in my life, I have a colleague who's really good, who I would want her on the committee that you described, both because she will fight for stuff and because she's conscientious and she's very well-respected. But I just feel like all of this is actually very obvious about her. If you spend a little bit of time with her, you'll see all these, her virtues just shine forth. You can see them. They're observable. They're not hard to discern. And I think that that's just quite often true, that virtuous people, their virtues shine out of them and are observable and can be perceived just as you can see the person doing jumping jacks.

Robin:

Okay, but we can watch people talking about evaluating people, so we can see that quite often there's people very close to someone, like this colleague of yours, and those people close can see many features about them and many details to make an evaluation on, but then quite often people much farther away are forming evaluations of these people based on descriptions of people closer to them. And these faraway descriptions are much more summarized, lower dimensional, less detail. And at some level, the most distant evaluation you could think of is just a very simple one-dimensional high versus low evaluation of people. That's sort of naturally what you'll get until you get so far away that nobody's even heard of the person, right?

Agnes:

You have to get through the Shoemaker. In junior high school, I did a report on Shoemaker. That's why I use this example where I hung out in the Shoemaker. There was this shoe repair guy. I hung out in their workshop and they let me watch all the things they did. I got to see how they fix shoes and how they repair shoes in this really fine-grained way. I thought it was really cool. Then I wrote a little report about it. But people who read my report, or there might just be further things like Yelp reviews or whatever, and there just might be up or down with the shoemaker. Are they good or are they bad? Right. They might not have the detailed knowledge that I had from observing the shoemaker close up. So that's just going to be true in general of expertise, of eliteness, that there's the close up people who see better and the far off people who get a more approximate and simpler assessment.

Robin:

But those far off evaluations seem to matter more for some sorts of roles than others. And those are what I would call more elite roles.

Agnes:

OK, that's helpful. That's a point I hadn't gotten before. So.

Robin:

So like the person at the top of the company, it's important not just the few people around them think highly of them. It's important that the entire company think highly of them. And so what people think of this person who are far away from them, have only heard about them somewhat, think of them matters a lot for their role. And this is true for many other kinds of elite roles, politicians, even celebrities often. what, you know, it isn't just how well they handle immediate relationships with people around them. That's what is important in their world. Often it's important that people far away in the social networks, away from them, think highly of them. And then those thoughts are going to be more lower dimensional. They can't make as many distinctions. And so we might then say that's kind of what status is. Status is a shared judgment of people from far away. And that there's a social need for that. That is, we often have a need for such evaluations, especially of people in certain roles, high status roles. We especially need a shared evaluation of someone, how it looks, how they look from a distance.

Agnes:

It seems to me, though, one of the things that, I don't know, has marked being a consumer over the past 20 years is the need for up-down assessment of things like restaurants and clothing stores and even doctors, like Yelp reviews, where it does actually seem, in a business, it is actually gonna matter Whether people who are far away get a positive or negative impression because they'll shop there, they have a positive impression. So that for you is eliteness?

Robin:

Yeah. So for example, Mark Andreessen once explained his theory of status in the world of venture capital. which is they're all these people with projects and they see a status hierarchy and they go to the top of the hierarchy first and see if they can get funding there. If they get rejected, they go down the hierarchy. And so the best products, projects are funded by the highest status venture capitalists. And then the fact that they get the best projects gets to reinforce the status and that they consistently have better projects that they fund. And, but there's a perception in the field of what are the high status venture capitalists so that people know to go to the high ones first. And so that if you hear that your funder was high status, that means something to other people. Again, it's this larger world of people needing to coordinate based on shared evaluation. So people far away from Andreessen Horowitz. need to have an opinion of Andreessen Horowitz so that when they hear that somebody is funded by it, they can make a judgment about how much to be impressed by that. And in some sense, that's what status is, is distant shared evaluations that distant people can then use to make judgments about people closer to the person or organization on the basis of hearing that somebody is associated So that's a concrete example. We could say the same for Harvard Law School, and law schools have these rankings, right?

Agnes:

If you work for anything where that is a selective, where the thing in question, the elite thing in question, makes selections, then the eliteness will sort of give a reason. So it'll actually be correct that the people who are funded by the top venture capital firms, that that is a sign that they're actually good.

Robin:

So we could talk about blurbs of books since I know this isn't an issue for you coming up with your new book, which everybody should read when it comes out. We'll talk more about it, I guess, when it gets. But anyway, you know, often what people are doing for blurbers is looking at the name that sort of how famous is the name of the person blurbing the book, right? We all know almost all blurb books are positive. It wouldn't be there if it wasn't positive. So the main information is just how famous is this person. And that's a way sort of an absolute status of people. that's used for deciding who to put on the blurb and whether to pick up the blurb, right? So it's like Andreessen Horowitz, but it's a way in which all these diverse people are all ranked on some overall ranking, and we're using the perception of that ranking to judge the book.

Agnes:

Right, but like, suppose that, you know, everyone agreed that a certain Barack Obama's restaurant was really good. I want to pick something like that where there's just a shared consensus that one burger place is like the best burger place. And is that then also it's more elite? Supposedly it doesn't cost more than the others, but people just think the burgers taste better.

Robin:

Or you might ask, if I name-dropped this burger, how impressed will you be? That's a key function of status, is status by association. That is, people often want to claim status by showing an association with someone else of high status. And what's one of the major contributors to perceived status is association with other people with perceived status. And so if it's a very difficult, hamburger place to get into, and it's very selective, who gets to go into this hamburger place, and I tell you I got to go there, then yes, you might be impressed. And it's not so much about the expense, it's about how hard it was to get in, in general.

Agnes:

So I think that there's going to be a subset of the thing you've defined as status. So you define status as cases where we need to agree as a group from far away about the value of something. So there's a far off assessment that matters. And my claim was that in the age of like internet recommendations, that's just true for more things. It's true for hamburger places and vacation places. And as we are more likely to make selections, like we have access to other people's reviews of books and and food and etc. And so if we if it gets like tons of stars or whatever, then we're more likely to go. But that doesn't necessarily mean it was selective. So selectiveness is so what what the Harvard case and the venture capital case had in common was that the thing in question only accepts some comers, right? And the hamburger place could be one where there are long lines, but you just have to wait on line long enough and then you'll get in.

Robin:

It could also be that it's an insider thing to even know about it.

Agnes:

Right, so that would be another way to add selection. But the point is just the fact that we have to agree about it from far away doesn't get you the idea of selection. That's a separate thing you have to add into your account. And so the Harvard case and the venture capital case are special cases of when we agree from far off.

Robin:

So I would say the generic situation is we need to agree somehow on who's high and low And then we're looking for indicators that we want to share on how to make that evaluation. And so we agree on which indicators count how much. And then we try to agree on indicators that in fact, you know, make sense as something is probably higher all else equal if it has these indicators. So these indicators are a diverse sort, but they include how handsome you are, how energetic you are, how rich you are, how well-educated you are. Also just what contacts and connections you have, what accomplishments you've done and who else endorses you. All of these things seem to count for people's status, and it makes sense that we let them all count if the point is just to agree on something that's robust. I mean, we could have 10 different variables for each person we all agree at from a distance, but that's a lot more work to manage, so we tend to collapse it to just a few, because for most purposes, just a few parameters are sufficient. And we could talk about what are the main, if there's more than one, what are the multiples? When we bother to distinguish people on multiple status dimensions, what are those dimensions? But still, from a distance, you can't be bothered to have very many.

Agnes:

I guess I think the phenomenon you're talking about is just not as big a thing in terms of the actual decisions they can make. So, for instance, booksellers have told me that they don't think blurbs really work. They don't think they affect that much who buys the book. And an extensive Goodreads review that actually explains why the book is good or bad written by nobody could have a big effect. And I think that in terms of like my colleague, you know, in terms of the decisions that are made about her, the ones that are made from a very great distance, even though she's like, you know, an important professor, those are actually less important than the ones that are made a bit closer in. by people who their opinion of her is formed by either close contact with her or, you know, secondary contact. And most of the decisions are actually that sort of determine her, you know, getting promoted or getting chosen for things are grounded not in her height or how good looking she is, but in her virtue, that is, in the fact that she's actually good. So she's put in charge and she's made to lead things because she's good at doing those things. And so, yes, it may be that with celebrities and in certain contexts, we have these very kind of abstract, free-floating judgments from a distance of high and low. But I think that the judgments that we actually use and are operative in our lives track actual abilities that people have. You put the right people in charge.

Robin:

Your income comes from selling a product, which is the label of graduate of the University of Chicago. And that is largely one of these distant metrics that isn't very sensitive to the local details. Most people who pick University of Chicago graduates for jobs don't even bother to notice what major they have, much less what GPA. A degree like that is largely a status marker, which people from a distance use without much attention to the details that produced it. They don't notice what classes they took or what grades they got in particular classes, or even what year they got the degree in. A degree from a university like yours is exactly one of these status markers, and it matters a lot in our world. People who have you know, degrees from prestigious schools, make a lot more money, have a lot more influence our world. There's plenty of data suggesting there's enormous advantages from having these sorts of status markers that are not from, you know, used in connection to their details that produce them. They are abstracted from those details and relied on enormously to great benefit as status markers.

Agnes:

Well, I mean, they might think that the University of Chicago had done a good job selecting people who are gonna have certain traits in the first place. But I graduated from the University of Chicago, and here's a position I have. I'm the director of undergraduate studies, so I run the undergrad program. I don't think I run the undergrad program because I graduated from the University of Chicago or because I got a PhD from Berkeley. I think I run it because I'm good at running it and people can tell that I'm good at running it. And I guess I just think in most institutions, if you look or most small businesses, you look at like who's in charge or who's doing what, there's going to be like a decent amount of correspondence where the people who are good at a certain job Are they going to be the people who are going to do that job? And if it doesn't work that way, then the thing is going to fail because the people are going to fail at what they're supposed to do.

Robin:

But the key thing is that people who don't have sufficiently high status markers are never allowed to try many things. They are never considered or... Maybe right.

Agnes:

So maybe there's like a whole bunch of people who never get a chance. That's consistent with everything I'm saying. But the point is, the people who are in charge are in charge because they're good. They have virtues. And maybe that other people have virtues too, and those virtues are just not that visible.

Robin:

I'm saying- The elite is to be good. Okay. But the point is, when people choose people for roles, they often choose them on the basis of distant status markers, status markers that can be seen from a distance, that are low dimensional, that are crude, and that matters a lot for who is chosen for what.

Agnes:

So I guess that, the statement you just made, it's ambiguous between two things. One, you choose someone for a role exclusively on the basis of their status marker. So we chose our president of our university just on the basis of what schools he went to or how many prizes he won or whatever, with no regard for whether he would do a good job as president of the university, but just on a bunch of status. I find that very implausible. That is, I think we mostly choose people for jobs on the basis of whether or not they would be good at those jobs, and we're reasonably good at making that selection. but it may be that we're choosing from a pool and the pool is pre-selected such that you only get into that pool if you have certain status markers. Similarly, in a way, there's automatically going to be pre-selection. The only people that can go into the pool are people, for instance, who are around. If they are on the other side of the world and they didn't apply for the job or whatever, so the pool is going to be selected in some way, and one way might be by status markers. Within the people who have those status markers, the only ones that are going to get selected for the job are the ones that are good at the job, whether the job is going to be shoemaker or president of the university.

Robin:

In a few months, we're going to be electing a president here in the US, and the two major candidates are Biden and Trump. It's very hard to believe that these are, in fact, the best two people in the nation that we could put together for this role. Their role requires a coordinated expectation that we're going to put the same person in that role. And so perceptions of their acceptability to other people and their status markers matter a lot for roles like that. It's also true for CEOs. So there are many, a part of the feature of elite roles, I would say, is that wider perceptions matter for them. It's not just how they would perform in some private way, it's how people will think of them, whether people are even willing to consider them, how people will react to them, the expectation of what sort of impression they will make on people far away from their immediate detailed behaviors. Those are more important in elite roles. And so status markers of the sort that distant people would see matter more, the perception that they have of what other people will think of them matter more. And that's just the nature of high levels of power and in organizations and in the world, even for a movie. putting an actor on a movie, it matters what people will think of that actor. When they see that person's name on the movie description, will that draw people in? It's the perception of many people of their appropriate eliteness that matters a lot for elite roles.

Agnes:

Yeah. So I agree with that. I mean, I think the president, there's a weird thing, which is, in fact, it's a bit of a desideratum, at least at the moment politically, that actually a whole bunch of people don't perceive them as suitable to all. Like, I think that's a big part of the appeal of Trump is that the people who like him like him because the other side doesn't want him. They don't want somebody that everybody wants, right? So that's a complicated situation that's, you know, maybe that's like advanced version of applying the theory. But I think it's right that the more people you're presiding over, The more it matters, not only are you good at X, but are you going to be believed to be good at X by people who don't have a lot of experience of whether or not you're good at X? That seems right. And or it might at least be like, are you going to at least not be believed to be bad at X by those people? Again, with the exception case of president where it's somehow it can be good that you're perceived to doubt it. And so that seems right, but it still seems to me that there's the core question of whether or not you're good at it. And then there's the question, can you telegraph your goodness very far away, or can you do that at least to a sufficient degree? And it may be that some people who are good at something, like who would be good as a university president, still wouldn't be good enough at telegraphing that they would be good, or they might even telegraph that they would be bad. to people who are very far away to be able to be, you know, put forward as president. I often think with university presidents, actually, it's going to be like there's going to be specific constituencies to whom they have to telegraph it who actually do get to see them close up, like maybe the board of trustees or something. So it's not always going to be an issue of telegraphing it just as far away as possible. Sometimes they're just specific groups to whom you have to be able to communicate how good you

Robin:

But the board of directors, part of their consideration is, will other people not complain about our choice? Will other people at least accept our choice? There's a lot of those sorts of factors that go into these high-level choices.

Agnes:

I once had to give a presentation to the board of trustees, and I recall that the person organizing the event was very, very concerned about how I dress. She didn't want to tell me what to wear, but she also tried to convey to me, please don't wear any of the things that you normally wear. Um, and, uh, she wanted me to look respectable. And I don't think it's that the board, like, I think, you know, within academia or as a teacher, I can dress in weird clothes and nobody does, nobody, like, ceases to respect me for that. And I don't think it's that the people on the board of trustees are worried that other people will respect me if I don't wear a suit or something. I think they won't respect me if I don't wear a suit, because that's their thing. They don't have much of a basis for predicting what's going to get me to be respected inside of a classroom. And I don't think that is what they're... I think they just have their own standards. And, you know, this person wanted to make sure I met those standards.

Robin:

But quite often, we are each evaluated on our evaluations of others. So each member of the board wants to be respected by the rest of the board, and they are each guessing what sorts of things the other people would disapprove of, and then they feel like they should disapprove of the same sort of things.

Agnes:

I agree. So I think within communities there can be like a shared conception of the good. Right. So the board may have its own conception of what makes a good professor and that conception of what makes a good professor may not be the same as what my students have or what my colleagues have. Those two also may not be the same. And so part of what you're doing is you have to be able to project maybe even to a set of different audiences and satisfy their desires. But I don't think that what the board is doing is just trying to predict what the students are going to think. I don't think that's right. I think they just have their own set of standards and they involve things like wearing a suit. I don't think they're under the false impression that my students want me to wear a suit. I think they're just not thinking about that at all. I guess that's what I think. I think people have their own sets of standards within groups. And that's what's doing the work, not just people constantly trying to project and think about what is everybody else thinking about.

Robin:

Surely there are some people in the world whose primary role is driven by impressing other people, say salespeople. Salespeople primarily need to sell.

Agnes:

I don't think they need to impress other people.

Robin:

They need to impress the customer.

Agnes:

No, they don't. They just need to convince them to buy the product. You might convince them by being unimpressive.

Robin:

I've had many salespeople convince me by not being ... But the point is, the salesperson doesn't care fundamentally about the value of the product to the customer. They care just about the customer buying the product.

Agnes:

I mean, it depends on the salesperson, but I think often the most effective ones actually do really believe in their product, because it turns out ...

Robin:

Okay, but their direct incentives are to sell it. So our lawyers' direct incentives are to win cases and or to win clients. There are many people in the world whose direct incentives, even in academics, direct incentives to get a journal to accept their publication, to get the colleagues to give them a job or tenure. There are a lot of people in the world who we can see their immediate material incentives and we can see they're tied to impressions. They aren't tied to the fundamentals behind the impressions.

Agnes:

I think people have someone who that's not true. Someone who their incentives are tied to the fundamental.

Robin:

Well, for example, if you're an investor and you're investing your own money, then if you invest in something whose price goes down, you lose money. And if you invest in something whose price goes up, you gain money. So in that case, you care not just about what other people think about where the price will go, but where it will actually go.

Agnes:

But isn't where it will go a function of whether people will be willing to buy it from you in the future and that's about it? Sure. I don't think we had an example yet of one where you care about the actual thing. If you're a shoemaker or a shoe repair person, the shoe repair people that I observed cared about making shoes that were actually fit people and were stable.

Robin:

because they expected customers to see those facts about the shoes.

Agnes:

Absolutely. But quite often, the easiest and simplest way to impress people in relation to X is just to make X be the case. Actually, that's usually the case.

Robin:

But there are many other cases where people can identify that their main incentive isn't tied to the impression of X, not X itself. And we can distinguish those in the world. We can walk through a detail.

Agnes:

it's not clear to me that you've yet come up with a way to distinguish it. That is, I don't think you've come up with a solution.

Robin:

You just said shoes are, I mean, you asked for an example where people care about the thing, and we said shoes, because they expect it.

Agnes:

But I think that's the same as the academic, and it's the same as all these other cases. That is, they care about, you could say that the shoe repair guy is like a salesman who just cares about the impression that you have that the shoe is good, even if it's not good.

Robin:

Right, and they know that your impression of the shoe will be infected by the shoe itself. That's the key distinction. They want you to like the shoe, but they know that the shoe itself will affect whether you like the shoe. It depends on which features you find out. That is, if I sell something to you only once in a lifetime, for example, then I might not care that you eventually figure out that you don't like it. There are many kinds of salespeople who primarily incentives are to just sell something that people won't necessarily like later because they're rare sales or they're basically fly-by-night sales, so we can distinguish, you know, the strength of the incentives based on the features of the product versus the impression of the product. That's a thing we can distinguish in many different contexts, and there's different relative weights there.

Agnes:

There's like short-term versus long-term features. Those are all features of the product.

Robin:

Some features are so long-term, people never know.

Agnes:

Sure, and that could be true of the shoe, right? Like, so it could be that I put some cheap glue in it or something.

Robin:

Right.

Agnes:

You know, it'll come apart, but you're going to throw the shoe out before then, so.

Robin:

Right, and so that's an important thing to understand about the world as a social scientist to analyze it, is whose incentives are tied to what features of the things they sell to people. Yes.

Agnes:

Right, but it's not clear to me that that doesn't cut across expertise and eliteness fully. That is, lots of experts have incentives, short-term incentives, and lots of elites are going to have long-term ones. So I'm not yet come up with any way in which the elites are

Robin:

I wasn't proposing that as a way to distinguish experts and elites, I didn't think. I was just saying that that's a thing that matters often in the world, is wide perceptions of people often not disciplined by actual later data on what it actually is.

Agnes:

Right, but I guess I thought the whole reason we were talking about that is that you were saying that it's distinctive of elites that there are these far-off assessments that are in effect not based on launch or not tracking much about their actual performance, and that's what I was skeptical about. That is, I'm skeptical that in the case of elites, we're tracking their performance any worse than we are with the experts, where also there's possibility of something that's just good for the short term. I'll let you get the final word, but then after that we should stop because we're over an hour.

Robin:

Well, thanks for discussing this again, and we may yet do a third session of this some future date. Apparently it's got a lot of things to talk about here. Okay. All right. Take care.