Robin:

Hello, Agnes.

Agnes:

Hi, Robin.

Robin:

What should we talk about?

Agnes:

I get to say? OK. I've been reading Foucault, and Foucault, in Discipline and Punish, is interested in a shift, a kind of shift to a disciplinary regime, where in an earlier regime, there were things like torture, and there was sort of the category of the permitted and the forbidden. And in the earlier regime, there was this sense that to be an individual, to stand out from other people, was to be high class or high status. And in the new regime, the new regime is a regime of normalization, where if you're an individual, that means you're a child or you're crazy or you're a criminal or you're a low status person inside of a hierarchy. And so I'm very interested in this normative shift, the shift from the old regime to the new regime. A way that I had of thinking about it is the old regime is an exclusionary regime. That is, there's permitted and there's forbidden. So it's like, it just cuts a bunch of people out. And it's fine with treating the people that it has cut out in really brutal ways, like torture is big. People who are insane are treated in ways that we think of as very inhumane. So that's the old exclusionary regime. And then there's the new regime is a conformist regime that tries to push everybody into the same mold of a certain kind of perfection. So whereas the old regime allowed for the kind of ups growth of greatness, the new regime does not. So anyway, I'm interested in this. transition from the exclusionary regime of greatness to the conformist inclusionary regime?

Robin:

So distinctions that seem like they're pretty similar from me, one would be the distinction between liability and regulation. So in a liability regime, if things go wrong, then you're punished or have to pay. But we don't otherwise restrict what you can do. We might judge what you do as negligent or not, so then we're getting more into the details of what you do. A simple strict liability regime just cares about whether things go right or wrong. The negligent regime cares a bit more about whether we think you are following good procedures. But a regulatory regime is just about the procedures and not about outcomes. So for example, you can have a liability regime with respect to automobile pollution. And I guess they do in some places where they just measure cars on the road and see how much pollution they're putting out and tax that. And in other places, they might say, get your car checked to make sure we have the proper equipment on it. And that's making sure you follow the procedure. And then it's not about the actual pollution, that's an indirect. So that's one analogous concept. And then in education and in professional licensing, one regime might be, you know, say you're an artist, show us your portfolio. We nod, we're impressed. Okay, you're an artist. Or another might be that you have to show that you have mastered a sequence of techniques. And after you've mastered those techniques, we'll call you an artist, regardless of how good your art is. Or similarly, with lawyers or something, we might put you in front of some cases. You can handle the cases. OK, we'll call you a lawyer. That would be an outcome performance regime. Whereas more typically for many kinds of professional licensing, we make sure that you show that you can follow procedures and that you do follow procedures. And many sort of job definitions, a qualified person for the job is somebody who follows the right procedures for doing the job, especially say in government and civil service. Whereas some kinds of jobs like being a novelist, the standard might be, well, did you make a good novel? Do people respect it? If so, you're a good novelist. And I see in my even history of education where I've been, I see different places do this different and maybe a shift over time from, say, early in my life as a physics student, the professor might just give us hard problems that most of us couldn't do and let us see if we could figure them out. And later on in other places, you know, there's this norm you shouldn't assign anything to a student that you haven't showed them how to do recently in class.

Agnes:

Yeah, so I think the the regime of conformism is going to be one that puts a lot of emphasis on the mechanisms by which you smoosh people into the mold that you want them to, and those are going to be like the regulations and the procedures. That's sort of what Foucault means by discipline. So in a In the conformist regime, there's going to be a lot of emphasis on procedures. The thought is, look, anyone can be turned into the normal person if the proper procedures are applied to them. There's also a lot of emphasis on examinations. So he talks a lot about examinations and the role of examinations in You know, kind of developing the student in forming the student and then the other part that's huge is surveillance. So you need a lot of surveillance in the conformist regime. He doesn't call conformist by the way, but. He talks about normalization, that's his word. But you need a lot of surveillance because you constantly need to be watching out for deviations and pushing in the direction of perfection and away from imperfection. So it's like a... There's a process of the regulation by power, that's what Foucault is interested in, that has to operate at the micro level. He talks about teacher, how education worked and how they would have to take some of the better students and give them the job of watching the other students and correcting how they hold the pen or pencil. This is like the birth of the conformist regime. But it's also the world in which everyone can be saved. That is, you can take even the students that are the worst students that can be pushed upwards a little bit to the next group. So that's the sense in which it's not inclusive. So I think that makes sense. That is, that liability is a bit more like the exclusionary regime and regulation is more like the conformist regime.

Robin:

So it occurs to me that reality TV shows are often let us fantasize about the outcome orientation or greatness orientation, you know, that a typical show might have a bunch of artists and then they all have to, I don't know, cook a dish or make some clothes or something. And then one of them gets kicked out every week or something. And then that's definitely not about following procedures. It's about, you know, doing well. And we have a bit of a fantasy about that, right? And in some sense, the things we often celebrate, i.e. art and sports, are things maybe where outcomes matter more. But in the rest of our world, we are very much wanting to move from a regime where we know it when we see it to a regime where we can tell people exactly how to do things. And I do think there is probably some progress when you can not just tell whether something is good, but figure out how to make it good. There's a sense in which you can teach it to more people, and lower the bar for more people being able to do it. And I guess that's good. I guess the worry is just you jump too quickly to grab any sort of procedure that does OK and make everybody follow that procedure. And then you're going to cut off the chance for excellence or greatness because maybe the procedure you define really isn't going to give much opportunity for that.

Agnes:

I don't know how much reality TV you watch, but the only one I've watched is, because my kids like it, The Great British Bake Off. And that's a show where the judges, you know, they look at the products. A lot of what they say in explaining their judgment makes reference to the procedures and whether the right ones were used. And these people were being filmed the whole time, and then they'll shoot back to the mistakes that they made, the bad choices that they made. It's not considered somehow this magical mystery that the thing worked out. It's like the person did it in the right way, and that's why it worked out. So I'm not sure. Now, there may be other kinds of reality TV, like dating shows where there's maybe there's less of possibility. I don't know. Maybe not. It's like, you know, the person just didn't comport themselves in the right way. And that's why they go to the island or whatever.

Robin:

I think there's a difference between seeing a bunch of things, saying one's the best, and if somebody says, why is that best? Being able to point out to some specifics, and something's worse, you can point out some way it's flawed. But that's different than sort of having a whole procedure you could walk them through to produce the best.

Agnes:

Right. But I was saying a third thing, which is that they point to the dough is underbaked or something. And they can go back to, remember when you didn't leave it in the oven long enough? The reason why this failure happened is that you didn't follow some correct procedure. It's not the same thing as saying I'm going to lay out for you the procedure, but there's this sort of conceit of there was a procedural failure. And so at least that's operating in the background, I think, of many of the judgments. But yes, I think you're right that we do have a bit of the faint, like the other regime, the regime of greatness and exclusion is kind of a little bit the regime of our fantasy life now. It's sort of been relegated to romance and fantasy. I think it still appeals to us even though we live more and more in the regime of conformism and inclusion.

Robin:

So there's a standard observation that companies over time accumulate rules and procedures because whenever something goes wrong, people will then say, well, gee, what did they do wrong? And then they will identify some candidate, you know, thing that didn't happen. They say, well, now we need a rule that that should happen. And then you accumulate more and more rules about what, how things need to happen. But then you could have an over accumulation of rules that the rules are too constraining, even though each one was motivated by a particular case where things went wrong. You see, it's much harder maybe to. look at an exemplar thing and say, this thing was great because it didn't follow this rule, so we should get rid of this rule. I mean, you might, you know, more blame on other things. So, and there's this basic, I teach law and economics, and a basic fact is the world is, the law is vastly more complicated than it was, say, two centuries ago. Two centuries ago, basically, you didn't really need a lawyer. And juries could decide most cases, and they didn't need that much advice from the judge, except because the law was considered pretty simple. And then whatever the jury thought the law was, was kind of what the law was. And then over time, we've accumulated this enormous complex law, which then requires expensive lawyers. And many people say, well, that's because the world got more complicated. And I think the world got somewhat more complicated, but not enough to explain the increase in the number the complexity of law. So that seems analogous to rules accumulating in a firm as slowly over time as each thing goes wrong and you add another rule. And so maybe there's a sense in which our whole society is suffering this problem, not just in law, but maybe in norms and schools and elsewhere that we slowly, we see something go wrong or right. And then we think of a rule or a procedure that might explain the variation we see. And then we try to impose it. But the net effect of all that might be to leave very little room for doing anything other than making sure you followed all the rules.

Agnes:

Right. So I want to go back to the thing that you said before, which I think was important, which is that you can see Foucault doesn't explain why this transition happened. He's not a big, he's not into explaining transitions. He's into pointing them out. But you were sort of giving an account that made sense of why you would make such a transition, namely, instead of just being like, oh, look, there's a good result. Oh, look, there's a bad result. You might want to try to bring the good results about. And you do that by doing some examination of how what happened before the good result happened and which things that happened are maybe salient. And maybe you see some things that happened in a couple of different cases. And so you're like, let's try and do that to make the good things happen more. And you might think that's just what progress looks like is systematically trying to bring about the good results. And so it would be shocking and paradoxical or something if systematically trying to bring about the good results leads to you're preventing the good results coming about because you become so burdened with rules that you made up that are maybe only correlations but don't actually cause the good result or something like that.

Robin:

Maybe an example is PowerPoint. as a way of giving talks, right? So there's this sense in which recommending that someone use PowerPoint recommends a structure that will sort of take out the lowest end of the distribution of really bad talks. You can't be terrible, terrible if you do a PowerPoint talk, because you will at least have thought about what you're going to say, and there'll be a thing to look at, and there'll be a structure, and you'll go through. But on the other hand, many people have said, and I kind of believe it, the very best talks are not PowerPoint talks. somebody maybe needs to set aside the PowerPoint and really think about what they want to say and say it well. And that's an example of sort of a structure. And so you might think of at least requiring students to use a PowerPoint talk because the ones who don't just do really badly. But then if you enforce that rule too strictly, then you will not let people explore the higher end of doing really well.

Agnes:

So I just want to say, as a little digression here, I disagree about PowerPoint. That is, I used to not use PowerPoint. And I give really good talks. But when I started using PowerPoint, now I give even better talks. And the key is, you have to completely write the talk, and completely structure it, and be done with it, and then make the PowerPoint. If you make the PowerPoint first, or if you're thinking about the PowerPoint when you're writing the talk, I agree. But basically, why wouldn't being able to have some visual images, in addition to what you're saying, be good? At the very least, it helps engage your audience. And even if your ideas are really good, you have to do something to engage your audience. So that's just PowerPoint. But I agree with you on the general principle, which is that there often seem to be things, strategies that we recommend where they guard against certain mistakes. So they might prevent certain mistakes, but then they have some costs at the high end or something like that. So, like, you know, I feel this is happening a lot with. When I right now, I'm dealing with a set of students who are 1st year students in college, and they're writing college essays for the 1st time, and I have to kind of undo what their high school teachers did to them. which is give them a formula for writing a paper and they are just trying to produce something that fits that formula and it means that they don't say anything interesting. And I'm like, no, your paper has to say something interesting. Like I have to learn something from it. That's the basic thing. And whether it's five paragraphs or however many paragraphs doesn't matter. The thing that matters is that the person who's reading it needs to learn something from it. And so I guess that that seems right. But if we go back to the basic idea, which is that We want the good results to be able to come about more regularly. It still seems to me like learning how they came about would conduce to that. That is, it shouldn't only conduce to having the bad results come about less frequently. I don't see why. Rules couldn't have both of those functions.

Robin:

So I've noticed that I and other economists say only teach some aspects of economics in our classes. There are some important topics like say innovation where they are really important, but we just don't actually know that many things to teach that we could reliably show you. And so we tend to teach the things where we can show you reliably a thing and then you can learn to do it. And only some topics have that sort of feature. So and so you might think, say, imagine teachers were a little bit nervous about whether they can teach or whether the students will learn or whether anybody will believe they know anything. Well, they might want to lean into, you know, finding and showing procedures that they can teach and that people can pick up because that's a proof that they know something.

Agnes:

OK, right, so let's say we had you know, I don't know, we have some artists in our society and like there's 10 artists who make really great paintings. And so then we're like, okay, we want more great paintings. And so we want to look at how these artists produce these great paintings. And then maybe there's one thing we discover that they all did, which is like, I don't know, clean their brushes before they start or something. They all did that. And so then we tell everybody, before you paint, make sure you clean your brushes. And that might even be good advice. But it may be that like- In most cases, at least. Right. It might be good advice in most cases. But the point is that that doesn't actually help you make a great painting. But it might be all that we know. That is, it might be the only piece of you know, regularity we found. And we might find some more regularities, like maybe they paint in the morning. Maybe they started painting when they were very young. And we can find more of these regularities. And as we find them, we're going to develop this system. And then we're going to get very invested in that system. But maybe what we don't yet know is like, what percent of the final result is explained by the total set of regularities that we found? So is that the worry that there's a meta problem there where, you know, like with economics, right? There may be a bunch of stuff that's known and some stuff that's not known. And it would be important to know what percent of the subject matter is explained by the stuff that's known. If it's like tiny, maybe just don't study economics, because you're not really getting anywhere by studying it. But if it's big, then okay, you're getting some substantial understanding by studying it. And I guess economics has to present itself as in the second case, right? They can't be like, we only know a fraction of what explains our subject matter, so you might as well not study this. Like in the way that the world I just described with respect to art, you wouldn't be well off just not studying it. The things I just cited, the keeping the brushes clean and painting in the morning and starting from a young age, If you understand all that, you've understood nothing about producing great paintings, right? You've got a few tiny correlations that don't really matter. The percentage of the final product that you've explained is just really, really small. So maybe that would be it, would be like, what percentage of the final result can you claim to be explaining with your rules?

Robin:

Well, as you know, my colleague, Brian Kaplan, has the book, Case Against Education, because I know he came and did a nine hours with you on that. And I think that's instructive to think there's all these young people who would like to, you know, the world to give them a try. And the world's kind of skeptical because they're pretty young and naive. And so what they want to do is say, well, look, I got this training. And you should, you know, listen to me more and be willing to try me more because I got this training and one story can be, well, what the training actually does it shows that you could make it through the training, even if the training isn't actually very helpful, they would still prefer to take someone who successfully got through the training. And so, you know, you could create this habit of just trying to create training regimes where you pick some set of rules and some set of procedures to show people, because there'll be demand for that, even if those procedures aren't actually very useful, because you'll be able to distinguish the ones who could at least get through the training from those who couldn't.

Agnes:

question I just asked just arises again, namely, what percent of your success at the ultimate thing you're being selected for would be explained by your ability to get through this training regime? If it's really large, right, so if like 95% of what it takes to be a good doctor is being able to make it through four years of college, then okay, it seems like a good system. Maybe we could find a cheaper one, but the question is, but what if it's like only 10% or something? Then it seems like it's, you know, that something very deeply irrational is going on. So the question is, in a way, it doesn't matter that much whether the training teaches you something or you know, is a way to filter out the people who have the property. The question is, what it teaches you or that property, are those things like a substantial part of what it is to succeed or only a small part?

Robin:

I mean, I think they're usually somewhere in the middle. But imagine a community of doctors, and every once in a while, some patient dies. And we wonder, you know, did they do the right thing? And often, of course, they disagree about procedures. But if some of those doctors, like, hadn't had training, we might say, aha, that's why this doctor's patient died, because they didn't have training. So there's pressure to, like, make sure everybody has training as a defense against accusations that you did things wrong. And you could see that happening with lawyers and other people too, just often everybody's very uncertain about if anybody is any good and it's hard to judge. And so they're trying to reassure nervous people or in the case of something going wrong that, you know, they did the best. And one way to do that is to be able to point to procedures and point to trainings and say, our people, you know, did the training and they know the procedures and they pass the test on the procedures. And, you know, what more do you want from us?

Agnes:

Right, but there were times when doctors didn't have any training and nobody did any of this stuff. So it's not somehow written into the nature of things that you have to be employing this defense, the preemptive defense or something like that. And it seems like the preemptive defense only makes sense if the training actually makes a difference. That is, if it's true that the training will tend to produce the good results, If the training doesn't produce good results, then the fact that our doctors went through it doesn't matter.

Robin:

Well, what if we don't know if the training is producing good results? Then we might, out of risk aversion, lean towards saying, okay, I guess you guys better be trained.

Agnes:

I'm starting to just find that risk aversion is like a phrase that people throw in when they don't know how to explain something and they also don't like it, especially if they're economists. I just think, so the world just became more risk averse, that just like an independent change from everything else doesn't make sense to me. it's gotta be that somehow we came to the conclusion that these results that were just springing up on their own were things that we could control or bring about. You could imagine it like the transition from, you know, like forager to farmer, right? So foragers just find food. There's just some food around and then they grab the food and they just go where the food is and they hope that they find it. And if they do, they're like, awesome. Farmers are in the regime of conformity and inclusion. They're making the food come into existence, right, by using procedures. Like they might look at the plants in the forest and be like, wait, we can grow those plants. And they might look at the wild animals and be like, we could cultivate those wild animals, right? So that transition is not like everyone becoming more risk averse. It's an attempt to bring about the good more systematically and like what else is progress but that. So I still feel like, you know, it's not like, oh, well, doctors want to not be blamed. So they make reference to these procedures. The procedures are supposed to teach you medicine so that you can cure people. That's why we have them. And it would be shocking if they didn't do that. Maybe they don't do it, but then it would be shocking if nobody noticed they didn't do it. If the world of the farmers was a world where actually they never grew any crops and they didn't have any domesticated animals, you would be like, wow, why are you still doing all this raking and hoeing and whatever? It seems like you have to think, no, it somehow does work. But then you have to think that all these introduced rules are not counterproductive. They're not stifling excellence. They're promoting excellence. That's their goal.

Robin:

I mean, we certainly have to grant that quite often there is a scenario by which first people don't understand things very well, and they don't do things very well. And then they figure out better how to do things. And then they teach each other how to do things. And then they try to name and describe the procedures and processes by which they can do things better. And then they teach those things. And that's certainly part of the development of civilization and the rise and improvement of things. you could also imagine going too far with that. That is, because maybe it's easier to defend if you just go farther and make up more rules and assure everybody that they're all safe because you have more rules, we could end up with too many rules.

Agnes:

So, I mean, like, there's a thing you said first, and then you said you could imagine going too far with that. And then it was like a bunch of stuff like about people wanting to make excuses. And like the first thing is just the attempt to bring about the good more systematically. And I actually do not see how you can go too far with that. It seems like you should push that dial to the max all the time. You should be bringing about the good as systematically as you possibly can. If there's ways to do more of it by introducing more rules and more regulations and whatever, then yeah, do that. If you're not actually bringing about the good and you're just pretending to do it, yeah, that's a big problem. If you were just doing pretend farming, you wouldn't get any food. But why is it that, I mean, maybe the point is that as you get better at it, you also get more and more inclined to just pretend that you're doing it. Is that what we want to say?

Robin:

So, you know, there are ways at which times people, you know, go out of their way to have an appearance of productivity and expertise when it may not correspond to the substance. You know, once wearing a white lab coat was a sign of being a scientist, then you wanted to wear a white lab coat in your commercial for your drugs or whatever, because people might trust you. So there's, you know, there's the, can be underlying correlations, but then people attempt to take advantage of expectations without fully implementing the actual process that would produce things being better that way. So we started out by pointing to some areas where it looks plausibly like we have gone too far. That was the start of some of the observations.

Agnes:

I disagree with the description, gone too far. That is, I don't think it makes sense to say, you can go too far by trying to bring about the good instead of just hoping that it happens by accident. Something is going wrong, but it's not that that process is going too far. It's like you're trying too hard. That's like the expression. That's another thing people say, you're trying too hard. But like risk aversion. That just doesn't make sense to me conceptually. There's no such thing as trying too hard to bring about the good. If you try harder, you should be bringing about more of the good. There's something else that might happen. Sometimes you might stop trying and instead start pretending, and then that might cause problems. I get that. If they only pretend farming, they're not going to get any crops. But I don't see why or under what circumstances it would be good to stop trying.

Robin:

Well, take the example of companies that add rules every time something goes wrong.

Agnes:

Yeah.

Robin:

Okay, there's a pressure, something goes wrong, you should do something, something needs to be done, and adding a rule is a thing you can do. But at some point, it might be that you have too many rules, and maybe you should start taking some away, or maybe adding a rule is not a good idea, even though something went wrong that it could have been addressed by it. But the people in that process might not have good incentives to back off on adding rules. Similarly, a school that's trying to convince parents and teachers that it's actually going to teach its kids something, will have a pressure to want to be able to show at the end that the students have learned something, so they're going to want to make some, you know, define some procedures and tests that correspond to them so that they can prove that they've learned something, even if it's actually not very useful.

Agnes:

So, I mean, in the case of the rules, it seems like you should add a rule if adding this rule is going to be a net improvement for your company, right? And that may be hard to determine. So if you don't know whether it's going to be a net improvement, you might want to study that or do an experiment or whatever. But like, but I get that if your goal in adding the rule is not to improve the functioning of your company, but only to make someone else believe that you've improved it, that seems like a problem. But there's still a question, like since the dawn of time, human beings have both done stuff and pretended to do stuff, right? So they've both done good things and pretended to do good things. One way to put this question is, why would the role of pretense increase with the attempts to make the world a better place? Maybe that's ultimately the moral of the story, is that the harder we try, the harder we pretend. And at some point, we're just doing more pretending than trying.

Robin:

So I've written on this concept of rot before. I don't know if we've ever discussed it much in our podcast. Yes, we have an episode on rot. Okay, we have an episode on rot. Well, the concept of rot is this concept that you can have a great reset of some sort, but then after the reset, you get an accumulation of things that can be on net bad. The rot is the accumulation of dependencies that sort of make the system overall, you know, burdened by excessive dependencies and rules and things like that. That's part of the concept of Ra. It can be a consequence of, say, a long period of peace and prosperity and stability. And maybe you don't, you shouldn't wish for a great reset again, because the great resets are very destructive. Somebody like you shouldn't maybe wish that your big old company with too many rules is going to die and be replaced by a young company with too few rules. But maybe, you know, in the long run, that's how it needs to happen. But, you know, if in some sense, you're questioning how could rot ever happen? How could we ever be making choices that accumulate into a burden?

Agnes:

I mean, I think that Yeah, maybe I am asking about that, but I guess I think this is a species of rot. I'm not sure it's rot in general. Because there's just, to me, something very odd about the fact that if the only thing that motivates the introduction of the procedures is the desire to more systematically bring about the same goods that were just popping up by accident. We had these accidents and now we're like, I want to create a system for creating painters because we only had a few and we want a lot of them. Seems like that's a good motivation. That's good. We want as much of that as possible. Wherever there's something good that happens by accident, we want to be able to bring it about intentionally. But then somehow bringing the thing about intentionally ends up being self undermining. So maybe, I mean, one thought might be this, that when you start bringing it about intentionally, it, there are sort of expectations of progress. There's like a timeline that comes into existence. There's the expectation that you'll just be getting better and better at it. And then you have to meet that expectation by showing proofs that you are getting better and better at it. And that's why the proof becomes so important. Or like, there's just a question, why is the proof so important in the defense? Why has that become so important in this regime of bringing things about more systematically?

Robin:

There's a static story and there's a dynamic story. I hear you as saying, well, there's this dynamic puzzle and these static stories haven't by themselves explained it. But it is worth at least understanding a static story. by which people don't want to just do things well, they want to appear to do things well, and they want to appear to be knowledgeable, etc.

Agnes:

Sorry, can I interrupt you for a second? Yes. Worry here is not that you haven't explained the change. It's that it's totally implausible to me that in the earlier regime, people weren't just as desirous of wanting to appear to do things well. That's not something that could plausibly have changed over time. So that's not asking you, how did you get from here to here? It's just you're citing a factor that I don't think can change over time. People have always wanted to appear good at what they're doing.

Robin:

So imagine a small company, a new company. It doesn't have very many rules. It might inherit a small rule set from some prior company or something. And it starts with a very small, sparse set of rules and definitions of job rules and things like that. And then as it grows, things happen and go wrong. And at each time something goes wrong, somebody says, well, we need to address that. And they redefine a rule or redefine a role definition or who's an authorization of who's allowed to decide what and who must get approval for what, et cetera. And over time, as the firm gets larger, you get more and more of these rules. So that's within a firm as it gets larger and older, a asymmetry in time. And you might say, well, you know, what was the difference early? Well, early on, you just started out with very few rules, because it was very small. And then as you got larger, you accumulated more of them. And then the idea is that at each moment, when you're thinking about adding a rule, or adding a role definition or something, you know, you might know in the abstract that you could, that might be harmful. But the concern is you will look, if you don't do anything, you will look like you're just unconcerned and incompetent and unattentive to the problem. So in order to seem attentive to the problem that happened and responsive to the problem, then you need to do something. And saying, I've thought about doing something, I've decided that wasn't good, that's not a good idea, just doesn't look very credibly caring or competent.

Agnes:

I still, this is another, like, form of explanation that I don't find that, I don't find it that plausible that something would make you look like you were competent and caring if that thing wasn't actually a way of being competent and caring. People are so gullible that they're fooled by the tricks that they themselves are pulling?

Robin:

No, no, the point is that some people are uncaring and incompetent, and for them, the thing they want to do is nothing. And you're worried about pooling with them if you are as if you are actually competent and caring, you're worried about looking like you're incompetent and caring and but we know what the incompetent uncaring people though they just do nothing, even if doing something would be helpful.

Agnes:

The people who are going to lead you down the path of rot also do a certain bad thing, which is continue to introduce new rules. You don't want to look like them either. You want to look like the thing you want to look like is the person who actually is improving the place. And the way to do that is just to actually improve the place. I don't see why the thing you would do to look good wouldn't just be the same thing as the thing you would do to be good, because the person who are the people who are watching you are just like you.

Robin:

But the idea here is all they can see is whether you do something, and they don't really know whether the thing you did was very good, but they can tell whether you did something, and they're pretty sure that people, you know, you're more likely to be the incompetent, uncaring type if you just do nothing.

Agnes:

But they also know that there's this history of too many rules getting introduced and they're worried that you're going to be that type of person. And so it seems like you'd want to distinguish yourself from that person. And in fact, if you're like the CEO of some new company and like, you know, you're scrappy and you like, it sort of seems like it's. like it's actually probably like they probably have a lot of signs that you care there are probably a lot of other stuff that they can look at to show that you care and they're not just relying on how you deal with this mistake and like uh it just seems like you should also you know be worried about them having that other false view of you and you should be producing you should be doing the thing that um is doesn't allow you to be conflated with any of the bad guys which is to say okay so

Robin:

Going back to basics here we had started out with this distinction between say perfection and greatness.

Agnes:

Yeah.

Robin:

Under a greatness regime we're going to focus on looking at the outcomes and rewarding and appraising people who have great outcomes. Whereas under a correctness regime, we're gonna collect more rules and procedures and make sure everybody follows them and reward and sort people on how well they follow the rules and procedures that we've accumulated so far. And obviously, then we say there's been a tendency of change over the last century, say more toward the rules and procedures. And surely we can grant that a straightforward explanation of that is that in fact, you know, perfection is just more useful in the modern world than it was in the older world. That is, we just have a bigger, more complicated world to manage, and greatness is a nice-to-have thing, but we are willing to, you know, sacrifice it for more reliably making sure things get done not too badly. And the way to do that is to make sure people all learn a bunch of procedures and rules and make sure that they follow them and monitor them on that. So that's, in some sense, our baseline assumption to be considering alternatives to.

Agnes:

I don't agree. So I think the regime of perfection is just the attempt to bring about the greatness more systematically. So I disagree that it is overly focused on avoiding the bad results. I think it's trying to bring about the good results. Um, I think that, um, I think that the effect of trying to bring about the good results might be that. Um, all we ever do is avoid the bad results that might be right. But, um, but the point is, it's like, um. The motivation is just here's a bunch of stuff that just happened by accident. Let's try to make it happen. That's what I take to be the difference between the perfection and greatness. It's just the attempt to systematically bring about stuff that was just. popping up of its own accord. And then the mystery is, how can it be that trying is counterproductive and trying ends you up in this world where what you're doing is just avoiding certain bad results and you never get to the heights that you were before? But maybe somehow.

Robin:

I think you misunderstood me. So just to be clear, I would say the greatness regime just naturally produces high variance. yeah and the um you know perfection regime lower variance and so lower variance you know, is cutting out both tails, and so then you are accepting that you're reducing the high tail because the benefit is reducing the low tail as well, and in some, in many environments, it could be worth reducing variance, even if the price of that is you take away the very, you know, make the very best things happen less often, but the middle of the distribution could just be better because it's just more reliable, so.

Agnes:

But I don't see why effort would have to cut out both ends of the distribution. So imagine we have our hypothetical group transitioning from foraging to farming. And they're like, you know what? You know how we just sometimes find food accidentally? What if we were to try to just systematically bring about the finding of food? Do you think it would just somehow follow that their food is going to be of, like, that they won't have some of the best food that they had, like, that the food is going to be more in the middle?

Robin:

I mean, it's definitely a standard story that, in fact, farming is lower variance food production than foraging, and that's, in fact, why foragers are more collectivist than farmers. Foragers each face a large risk of not bringing home any food each time they go out hunting. And so they're trying to reduce their risk by sharing.

Agnes:

Absolutely. I think you're right that the effort lowers the variance. The question is, does it necessarily cut out the top end? Couldn't it just lower the variance by cutting out the bottom end? If the farmers and the foragers were to meet and each of them were to present their top five foods, do we somehow know a priori that the foragers are going to have the best ones? It's just not obvious to me.

Robin:

I mean, it seems to me, in many of these examples we went through, like the essay writers, you know, by following a formula for writing the essays, they produce this middle-of-the-road kind of essay, and if they're going to, you know, do better than that, then they need to break away from the formula.

Agnes:

Yeah, but why, like, Like, let's say that I cook, right? And sometimes I just throw together the dishes and sometimes I really put effort into thinking about what recipe I'm going to use and the ingredients and whatever, and I'm more systematic. Like, it just doesn't see, I think that if I use a recipe, I'm gonna cut down on variance. But it's not obvious to me that if I use a recipe, I'm guaranteed to cut out the very top things that I could make. I'll never make a really great thing using a recipe. That doesn't seem true. It actually seems like if I wanna make a really great thing, I should use a recipe. And that basically I'm cutting out the variance of the bottom end. So I'm not convinced that we would have to cut out the top.

Robin:

So think about liability versus rules in the roadways. If we just had a liability regime, then if you ever crashed into somebody or caused an accident, we would punish you a lot. But then we wouldn't have many rules about staying in lanes and staying under speed limits and things like that. We just let you drive how you want and we'd hold you responsible when things went wrong. Right. And in that situation, like when the weather was really clear and easy to see, you might drive fast. And then when it's rainy and stormy, you might go slow, but given full discretion. But if we have a speed limit, the speed limit is going to be somewhat crude and it won't reflect, you know, how sunny it was and rainy and how straight the road was and all those things. So rules sort of by their nature are crude and they can't make as many distinctions as you personally might in the situation making a choice.

Agnes:

Hmm.

Robin:

So for example, if there's the five-paragraph essay rule, then they just have to make five paragraphs. And even if they think, no, in this case, six would be better, they just can't vary it. Whereas once you no longer have the five-paragraph essay rule, you'll make the number of paragraphs you think are appropriate to this particular essay. And that gives you more opportunity to do better and worse, depending on your skill at choosing the number of paragraphs.

Agnes:

Okay, so here's one way to think about it, that the rules cut down on the role for the power of judgment. And that might be pretty good if what you're doing is to try to more systematically bring about a kind of result that's already on the table and you can get more and more like knowledge about how to bring about that exact result. But maybe part of what was good about the the exclusionary regime of greatness, is that it allowed accidents to happen. And maybe accidents are an important part of innovation, that is, of new good things showing up that we hadn't anticipated before. Because one of the things that I get into with my students, where they're like, they always want more instruction from me. They're like, but tell us what you're expecting, and tell us what you want, and tell us how to write a paper, and how do I avoid this problem, and how do I do this? And what I tell them is like, I want you to teach me something new that I don't know yet. So I can't tell you how to do that. Because if I told you what to say, then I would already know it. And then you wouldn't be surprising me. I can't teach you how to surprise me. So that I feel that's the fundamental dynamic that we're often in. And they're frustrated because they just want me to help them. But I can't help you surprise me. You have to somehow do that on your own. And so that there's this way in which some part of the advent of good things is being surprised. And the more rules we have in place for systematically bringing about some good results that we're gradually learning how to bring about more and more systematically, the more we're shutting out the possibility of being surprised.

Robin:

So certainly a standard critique of over-regulation in the sort of government regulation regime is the idea that regulation is usually matched to current technologies and practices. And that tends to make it hard to change those practices because the regulations you tend to forbid it. So for example, you know, cars that are planes, you know, are Difficult to make because we have one set of rules for cars and another set of rules for planes and they don't overlap very much. And so you can't make a thing that's both a car and a plane without breaking a bunch of these rules. And similarly, you know, I've been an advocate for prediction markets, financial regulations basically just says no without much envisioning that you might have this other use case for it. So I think that's, you know, basically regulations in general and rules are simple, you know, basically based on a simplification of the world. And if without those rules, you could adapt more specifically to context, which includes accidents and innovation. and that the rule is going to limit again, you know, sometimes in order to do something right, you might need to go over the speed limit. And, you know, with the speed limit, well, that's just not allowed.

Agnes:

So one thought might be, like, the more progress we make, the more the larger the group of people who think we're just about done, right? Or like we're close to the end of the game. And the closer you think you are to the end of the game, the more it makes sense to just create some rules to prevent the outcomes that we already somewhat know how to produce, like get more precise about producing those outcomes. And so it's actually very much related to this point about earlier about you were saying we don't teach innovation in economics. Like, non accidentally right? That's the thing you don't you don't have there's there's no systematic knowledge of how to bring about the accidents. And, you know, or like. If you think, look, we've pretty much almost got economics down. We pretty much, we have like a pretty systematic understanding of it. Then it just makes sense to teach the stuff that we know and just get better and better at producing students who can, you know, recapitulate those things. But the more we're like, well, we're really far from the end of economics, we're like 1% of the way, then the more you want like a fertile field of accidents to spring up so that we can then maybe try to systematically bring about some of those. So there's a tension between the project of trying to systematically bring about some good and The new good showing up where that thing is going to be by accident. There's maybe some kinds of good things that have to pop up accident.

Robin:

That would be so you think you're talking explore exploit, which I know you focused on. So, I highlight that there's a. there's a coordination thing. So imagine as you go through your life, early in your life you're going to explore and try lots of things, and then later in your life you're going to settle down and take advantage of the best thing you found and live your life, you know, with the best person you found, the best city you found, and the best job you found, right, going to the best restaurants you found. And once you've made the switch, then it's more okay to have rules and habits because there's less reason to go exploring because that's done. And in some sense, you need to coordinate this explore versus exploit trade-off across a whole bunch of areas of your life. And the same would be true for an older company. By the time a company is older and stuck in its ways, well, now you as an employee say, there's no point in searching for really different products or ways of doing things here. This company is on its way down and it's pretty stuck. And so I shouldn't even be trying to make things different. So having a lot of rules isn't such a big cost here because we're not really foregoing an opportunity. So there's a sense of which, like in the organization in a life, if you're going to be open to changes and new things finding, you need to sort of have a whole bunch of aspects ready and open together. Because, you know, just trying to open up one aspect isn't really going to work if all the other things are shut down and locked in.

Agnes:

But if we go back to Foucault and what he's noticing basically in the change in Punishment practices and school practices and work surveillance practices in we're talking. the 19th century, right? This is starting early 19th century. So torture basically goes out of style, early 19th century, and then slowly is being replaced by these humane disciplinary procedures and this sort of new surveillance model. And the thought would be that what he's noticing is that, humanity is just becoming less exploratory and more exploitative. And that seems insane because this is actually when a huge transformation is happening. The 19th century, like things have barely started changing. They've just started to really, really change. And so it, like maybe if you said it was happening now, that might have some credibility. But again, once again, it's like me almost saying, well, now time has stopped, so we can be exploiting. But way back in the 19th century, they were just getting started with big changes. Humanity was definitely exploring. So there's something just in terms of the chronology of this I'm confused by the thought that we see all these manifestations of the conformist inclusionary ideology as early as we do, at least according to Foucault.

Robin:

So remember we read from peasants into Frenchmen.

Agnes:

Yeah, that's late 19th century. I mean, that's focusing on like 1880s, 70 to 1910, something like that.

Robin:

Well, it considers the whole century, but it's a trajectory.

Agnes:

I think it had some years in the title. And I think the years were roughly what I just said. I don't think.

Robin:

But I mean, there's a sense in which to enable the modern world of innovation and development, there was a choice that was sensible in some sense to reduce variety within the nation. That is, instead of all these very different peasant cultures who had very different even languages and customs to sort of create this national culture, and that enabled a national scale market and communication and sense of patriotism that enabled wars, et cetera. And you might think of rules and procedures or this concept of perfection as sort of enabling a shared world that they could then innovate more in, right? There's a sense in which, you know, having very flexible little peasant cultures doesn't so much produce that much innovation because they can't just do that much different in a little peasant culture, but in a nation that has a huge scale economy, the economy and, you know, trade across everywhere and all this different division of labor, then you can have more innovation and development, but it does come at a cost of some reduction of certain kinds of variation.

Agnes:

Right. So by the way, I was right. I just looked it up. 1870 to 1914 is in the title. Those are under discussion. So the idea that maybe like the unit of innovation starts to not be the individual. And that's why the individual has to get homogenized and routinized and squooshed into these larger units that are then going to do bigger stuff. then maybe we're just like at our most innovative now as we got like global coordination among the elites and we're turning into this one giant entity that's going to do some amazing thing. And maybe we, like you and me, have just overly romanticized the innovation at the level of the individual. But the way of the future is the innovation at the level of the ever larger group. And that involves huge amounts of conformity at lower scales.

Robin:

So I do think we are making somewhat of a mistake when we sort of buy and regulate by doing it so much through procedures. I think outcomes would be a better approach, but I can see that this energy and momentum of collecting procedures and rules has a point behind it. It's not a random thing. It's sensible, at least in general trust.

Agnes:

OK, we should stop.

Robin:

All right, till we talk again.

Agnes:

Bye.