Agnes:

Hi Robin.

Robin:

Hi Agnes! Long time no see.

Agnes:

So I think we agree on something.

Robin:

Make a note!

Agnes:

This is— I want to talk about a problem that I think we agree that there’s a problem, though I’m not sure that we agree on how to characterize the problem. I think of it as the distant signals problem. And so we might compare somebody who’s trying to impress the people around them, like their teachers of their high school, their parents, their friends, right? And they’re trying to act in ways that those people will count as being, you know, virtuous, noble, intelligent… all kinds of positive words.
And how they would act if instead, the people they were trying to impress were both temporally and spatially very far away from them. So they might be, for instance, administrators in charge of the final decision on like, you know, university president or something in charge of the final decision— tenure decision, to whom, you know, the departmental committee is going to be in some sense beholden.
And, you know, the letter writers for that tenure file are thinking about what the departmental committee can justify to that president. And the person’s graduate school education, right, is in some sense geared towards producing the kind of record that will get them a job, at the kind of school, and then get them tenure…
So this is like the long— at the end of a long signal process. And you know, how do you conduct, say, your high school education, if you’re— in some sense, you have a view to those very— that very faraway audience instead of a very close audience. And I think this— the problem of the way in which your life gets perverted by attending to that faraway audience is maybe the biggest intellectual problem that I directly encounter in my own life.

Robin:

And I think I agree, it’s one of the big problems. I think it would help to contrast it with a case where it’s less of a problem. So if we think about choosing who to marry, we might say, yes, there’s an element that you’d like your spouse to be impressive or attractive to your family, and maybe even to your boss at your job, but there’s less of an effect there. You’re more going to be focused on whether you like your spouse, and if there’s things about your spouse that you like that other people don’t appreciate, it’ll be okay for you to still pick that spouse, as long as the negatives that other people might see aren’t really terrible.
Whereas with something more like a job at a prestigious management consulting firm or law firm or professorship, it matters more that these distant outsiders will approve of the choice. There’s less scope for a department hiring someone who they privately think is great—because they talk with them all the time—who on paper doesn’t look so great, because the department administrator and the president of the university etc. have to pick somebody who looks good on paper. That— I would see, how is the contrast here we’re focused on: The difference in treatment between choosing locally, based on local criteria when you’re free to do that, versus when you’re not so free, because you’re more constrained by these distant observers.

Agnes:

Right. And the question is, when we say— use this phrase, “looks good on paper,” right? So how do we conceptually connect the idea of the “on paper” to the distance of the signal?

Robin:

Right, so paper will show sort of relatively “objective,” quote-unquote, things that can be verified by distant observers—

Agnes:

Okay.

Robin:

—in the same way that, you know, a legal case, you’ll have to present evidence that can be verified. For example, if you just have hearsay from someone, that won’t be allowed, because it can’t be checked.

Agnes:

Right. So there might be two properties of the signal that are sort of important for it to be, like— Or maybe it’s just one. Some signals have to be heavily contextually interpreted.

Robin:

Right.

Agnes:

And like, what you love about your spouse, you might describe like the way they smile or something. But it’s, like, hard, and you need to actually see it, and you need to see it in the context of a conversation, and you sort of— you need to be there. You need to be close. Because being close is going to mean that you have access to all the contextual features with reference to which that signal functions as a signal. And so the on-paper signals are the signals that can travel far because they can be decontextualized.

Robin:

I’d like to give this historical spin of saying, our distant forager ancestors lived in groups of, say, thirty people, and then only met 150 people in their lives. So when they evaluated each other they could take a lot into account. So when somebody was accused of malfeasance or proposed for a role, they just knew a lot about these people.
And you know, they— each signal was taken into account in all this, and it was a more holistic judgment based on just everything they knew about them. But as the arrival of farming, and then later industry, we have these much larger societies, and we don’t all know each other. And then it became more important, what distant observers would think, based on relatively little evidence.

Agnes:

Right. And it’s not just little evidence, but it’s evidence of a very specific kind, right?

Robin:

Right, that can be verified.

Agnes:

Right. That can be— I mean, you know, presumably, even the other kind of evidence can be verified, it’s just it would be a lot of work. And it would— you might only be able to be verified by people who are close by, right?

Robin:

Right. Easily verified. So for example, in a farming community, if you insulted the gods and said certain words, and other— several people heard the words, then they might be able to quote that they heard those words and verify that, yes, several people can testify that they heard those words. And even if it’s just a few seconds out of your life, it can have these greatly disproportionate impacts because all the rest of your life isn’t so verifiable. But they have these quotes from these people who said this is what they heard you say.
And so it became more important to sort of avoid these very short but very clear signals that you were a bad person, and maybe to acquire very particular good signals. Like you killed the lion once, and everybody remembered you killed the lion. And that fame could go among thousands of people in a village area or something, that you were the one who killed the lion. And now that’s a particular, you know, certification that you’ve acquired that will last.

Agnes:

So if we, you know, if the transition from forager to farmers, this transition to more distant signals: Is your thought, like, “Yeah, but…” and we just keep going more and more? Like, are there moments of sudden great shift? Because I feel like I’m seeing a shift. Like, just in comparison to my undergraduate times versus now over the past thirty years. But you know, you’re talking a shift over thousands of years.

Robin:

Right. So I’m tempted to think in terms of the rise and fall of civilizations story, and the story of what happens during a fall era. And so the story— usual story is that, initially, some small band of people who are very motivated and very bonded to each other, you know, dedicated and worked very hard, and they have good norms about doing things well. And it’s a rare thing, because it’s a rare success.
And then it grows. And it grows, it gets lazy and big and complicated. And then people focus less on the overall value of the civilization, they more focus on divisions and they have more veto players. And overall, you know, you accumulate more positions of power, and maybe more kinds of credentials you pay attention to. And that could just be like a generic evolution in peace and prosperity between times of crisis, when you might toss all that out and make sure things…
So for example, like, at the beginning of World War Two, I’m told that they had a military that was, you know, had all these ranks and all these metrics and measuring, and then they started losing a lot of battles. And they said to themselves, This isn’t working. And all of a sudden, like, they had to go back to basics—like, let’s pick people who have won a battle and put them in charge. And just go back to sort of raw competence and simple direct things that like, worked.
But they had built up all these credentials that were being verified from a long distance, but weren’t tracking sort of fundamental abilities. And you might think that’s a kind of thing that might accumulate. Where, you know, we— you know, somebody who went to Harvard, everybody recognizes Harvard is a great thing. Harvard is a great thing, but we don’t— do we know that he’s actually good at anything? Or do we just know that everybody says he’s good because he went to Harvard?
And we might be sort of accumulating these shared metrics from a distance that don’t actually track performance, because performance is noisy. And whether you went to Harvard is not.

Agnes:

Right. So like it seems to me that, you know, what drives this pressure to create, to— you know, for students at a younger and younger age to be producing these signals that can travel far, is that there are some things that are, like, easy to evaluate and quick to evaluate, and some things that are hard.

Robin:

Right. And reliable to evaluate. Like, with a low error.

Agnes:

Right— I mean, it depends on how you understand error, because you—

Robin:

Right, but they need to— low variance.

Agnes:

So if you look at someone’s, you know, SAT score, right, that’s an assessment you can make quickly versus even reading a piece of writing by them which takes longer, right? So it seems to me that the pressure is somehow efficiency. Like if the pressure towards efficiency is what produces…

Robin:

Well it’s not just efficiency, it seems like it’s also that sort of ease of distance. So for example, you know, you and I know that some of our colleagues’ papers in journals are great papers, and some are not so good, but they’re in the same journal at the same time, maybe.

Agnes:

Right.

Robin:

Right? And we like to talk about which were the better papers, but if we do that in a tenure review the administrator will look, and they won’t see that extra difference, right? They’ll just see us pushing our colleague, you know, by claiming they have better than average papers. And so it’s this problem of how do we convince other people on the outside that there’s this metric. And you know, so that’s the difference between reading an essay and saying it’s great versus checking its spelling, or length, or, you know, an SAT score.

Agnes:

Right. So the idea is, like, that you have to explain. It’s not just that you have to evaluate someone else who’s in some sense far away from you. But you have to be ready to explain your evaluation to somebody who’s not going to have very close contact with—

Robin:

Doesn’t necessarily trust you, and your evaluations.

Agnes:

Doesn’t trust you and doesn’t, you know, isn’t able to evaluate— I mean, that’s the important thing, because the administrator can’t read the paper, right?

Robin:

And they might have an incentive to pick people who have sort of lower noise in their estimate versus more reliability. So I’m struck by how— There’s this story that I think is roughly true, that, for example, when I was in high school or early college and I was a physics student, they would just give us a lot of hard problems that most people couldn’t do. And so there would be just noise in the evaluation of each person each week, or even a semester: Did they manage to get the hard problems?
But if you— you know, but that you could learn a lot from trying hard problems. And then there was apparently this move over the last decades to give more smaller problems that are like more sure that you can do. And now we’re in a situation where high school students like have all these assignments, they’re all expected to get 100% on all of them. And if you ever miss any assignment then that’s going to kill you, but you never can excel. Because you’re never given a hard problem that most people can’t get that you get, and you excel by getting it.

Agnes:

So earlier you gave the example of the forager who doesn’t— he’s going to avoid cursing— or the farmers, to avoid cursing at the gods, right? Because that’ll produce the bad signal. But also might try and go kill the lion in order to produce the good signal, right? But now you’re saying that we’re in a situation where we’re producing— we’re avoiding producing bad signals, but we’re not giving people the opportunity to produce good signals. Why not have both?

Robin:

Well right, so you might think there’s two ways to go get glory: You can go try to kill the lion, or you could go to the lion-killing school where they’ll walk you through exercises of lion killing and give you a simulated lion-killing exercise. And then if you just put in the right effort and have the right connections, you will just be pretty sure to get an A and you will get the lion-killing certificate from the school. And you’re choosing to go for the reliable lion-killing certificate or to go out and try to actually kill a lion, which, you know, even if you’re really good, is gonna be risky, and there’s a good chance you won’t succeed.
And so that the, you know, less noisy signals might just dominate. Like for example, you know, you could take someone who went to Harvard, or someone who did a spectacular accomplishment, like wrote a novel that everybody loves, right? But which should I— you try to do? Write this spectacular novel, or go to Harvard, right? But if you’re really good, even if you try to write the novel, you still might fail. Because it’s hard to write good novels even if you’re really good.

Agnes:

It’s also hard to get accepted to Harvard.

Robin:

Right, but if you’ve got the choice to go to Harvard—

Agnes:

Oh, I see. I see.

Robin:

Right. You would just want to go because it’s such a reliable signal. Everybody— So, Harvard, that’s it.

Agnes:

Right. So it’s somehow— there’s somehow this thing where like, look, what we’re trying to do, I guess, is communicate our own worth, right? To other people, right?

Robin:

Right.

Agnes:

And you know, if you’re communicating your worth to the people around you, like, because they can sort of put stuff in context and evaluate you over a long period, maybe you’re going to be— you’re going to end up communicating the worth that you actually have? Right? More? Is that the thought? But whereas if you’re trying to communicate your worth to people who are far away in time and space, you’re less gonna communicate the worth that you actually have, you’re gonna— is that—

Robin:

So this is related to why it’s been so hard to replace college. So you know, many people, including my colleague Bryan Caplan, you know, argue that college is a bit of a waste. It mainly like takes a lot of time, you’re not actually learning that much useful. And it’s sort of sorting people, but why don’t we sort people some other way? You might say, Why don’t you just go work for a firm for four years and do work there, and then show your same abilities there rather than at college. Because it’s not— it’ll cost a similar amount, but you will get wages and actually be productive there and learn more useful skills.
And the problem seems to be that the firm doesn’t give you as standardized an evaluation as the college. Well, there’s all these different firms, and how do you know how good you are compared to all the people who go to all the different firms? Whereas for colleges, there’s a standard ranking of the colleges and there’s a ranking within the school. And so you might rather go to the school and get this predictable ranking than to go to a particular firm and get this story someone says about how good you were, but they don’t know how to compare it to other people.

Agnes:

Right, that’s basically a signal that doesn’t travel as far.

Robin:

Right, because it hasn’t been standardized.

Agnes:

Right.

Robin:

And I think what we really need to do is to get sort of these places you could work instead of college to find a standardized rating system. Just to make it. Even make more standardized tests so that they could compare you. It’s not impossible. In some sense, though, we have all these different teachers from all these different colleges who have all these different students in classes, and they are rating them in somewhat idiosyncratic ways. But somehow we find a way to merge all those things into a GPA that’s comparable.

Agnes:

I guess I don’t see any reason why, like, the working at the firm, why anywhere you create this system isn’t going to produce the bad effects. I mean we should talk actually first a little bit about what the bad effects are: Why is it bad to have a bad communication of your value? And like one thing that I see as bad is that it puts a huge amount of stress on students to conform to a very specific pattern of like, what it would be for them to show their excellence, right? Which doesn’t bring out the distinctive—doesn’t do a very good job of bringing out the distinctive excellences that different people have. It puts mental— huge mental health pressures on them. And…

Robin:

I think it’s not just students. Like if I think of academics, I see many academics, especially ambitious people, smart people from top schools. They say, What should I write my next paper on? And they say, Well, I need a topic that’s sexy and I need a data set and method that’s hard. And if I can find those combinations, I just do it, and I crank it out and I shove it into the publication system.
And they got to accept it because hey, it’s a sexy topic and it’s a cool method. And if you ask, Yeah, but is it important? What does it matter? Does it have legs? Will it like, go somewhere? And they say, Well, that doesn’t matter. That’s not what the system’s rewarding them for. The system can’t reward me for that, because that’s too subjective a judgment. Whereas it’s objective to judge, Is it a sexy topic, and did I use a hard method?

Agnes:

Right. So you could imagine like a school for like kids, right, that had a very, very strict set of rules for how the children have to speak and move and dress and whatever, and used very painful physical punishment on the children to get them to be that way. And to get them to speak a certain way and to move a certain way and to be interested. And if you look at the kids coming out of that system, what you would expect is some of them would be broken by it and destroyed, and have all kinds of mental health issues. But the ones that made it through and succeeded would be really, really similar to one another.

Robin:

Standardized!

Agnes:

They’d be standardized, right? That is, the standardization of the evaluation process also standardizes the people.

Robin:

And there are advantages of that, and we should admit it. So it is a standard trope that’s somewhat true that the modern education system was created in part to create standardized workers for standardized workplaces. Workplaces who took, you know— Well-educated students could roughly know what to expect. They could create jobs slots for those kind of people, have instructions for them, incentives for them. You know, golf clubs for them, whatever it is they would want. And you know, have that standardization.
Sort of large benefits exist for standardization in workplaces. And so we should acknowledge that, but we might want to say maybe it goes too far—what you want is the amount of standardization that is actually useful, but maybe you don’t need a bunch of other standardization.

Agnes:

There’s also a question of whether—and I don’t know the answer to this—but whether workplaces are as similar to one another now as they were then. That is, is there this idea of the workplace?

Robin:

I mean, there still are huge gains from standardization. So we often like give a— celebrate the few workplaces and tasks which have a lot of creativity and variance, because that’s sort of culturally something we like to celebrate. But honestly it’s not actually that important in most jobs that you’d be very creative, say, or even insightful, right? So in fact, like, most of the variation you will see of people not doing the thing they were asked to do is mostly error, not creativity or innovation.
So you can understand why workplaces are actually not so eager for creativity or innovation relative to standardization, but you could still think of this going forward. So like a standard story is, say, the business suit, right? You know, you want standardized workers who have standard ways to talk to each other, times they come in, ways they sit at the desk. But you don’t actually need them to wear standardized clothes.
However, once you’re in a world of trying to show how standard you are, then that continues into many other aspects of your life, including what clothes you wear. And you can just see from the nature of suits or haircuts that there’s incentives to go much farther to standardize on many things, just to show how standard you are. Just to show that you are one of these standard people who fits the mold.

Agnes:

One of the weird things is that I think in a lot of ways we’ve loosened a lot of those forms of standardization and we’re much more, in some sense, anti-conformity. And yet the actual people inside the suits, I wouldn’t be surprised to learn that they are way more similar to one another than they were back when they wore the suits. And so there’s something odd, there’s something incongruous about at least our self-conception in terms of this acceptance of human heterogeneity in the way in which we excel.
And the— a system that takes especially the top people—like, you know, I’m teaching at the University of Chicago, and teaching the top people, and I see them under absolutely like immense pressures of conformity, where small variations in how they dress, and you know, makeup and hair and whatever, are like, are almost like this…

Robin:

And that doesn’t stop in school. So it continues on to the workplace. I think one of the things I didn’t realize about modern workplaces was just how poorly managers are able to evaluate performance in almost all jobs. You might think, Well, once you’re on the job, there’s just whether you do a good job or not. And this is all about like giving you the chance to prove yourself. But it honestly seems like in most jobs in most organizations, it’s hard for them to tell how good a job you’re doing. And there are some metrics—like you could screw things up really badly or something—but otherwise it’s mostly hard to tell.
And so they mostly have to go on, like, What do people say about you? Do other people think you’re doing a good job? And that comes down more to these distant perceptions, including wariness of any weirdness. And you know, that’s more the usual system.
You can even, you know, think of the usual yes-man sort of thing. You know, a lot of organizations, the boss encourages people below them to say yes. The boss says things, everybody says yes, and, you know, you’re not actually getting a lot of information from individual people about what they they think. You’re getting this incentive for conformity because in some sense the boss wants to show they’re in charge and everybody’s unified behind them. And that’s kind of more important than getting diverse input.

Agnes:

Okay, but I mean, you know, presumably, like, you know, maybe there— maybe there’s something important about the boss showing they’re in charge. I’m interested specifically in the idea that the signals are going over the head of the boss. So like, if we think about your— Well, let me actually give an example, okay, from my own life, is that, you know, I’ve found that over time— So students, I think, care in a class. They care about a number of things, but two of the things they care about: One of them is they want to impress me, and they want to do well in the class to get a high grade so as to, you know, impress future people.
And over time I don’t think I’m changing that much. But the, you know, the amount that the impressing me part of the motivation takes up is like shrinking. And the grade is growing. Right? To the point where like, if I assign a paper, you know, the fears over getting anything less than an A and what that would mean for their transcript so dominate their psychological approach to the paper that, like, the papers are just getting less interesting over time.
And in some sense they’re getting more competent but less interesting over time. And they differ less from one another than they used to. And so now what do I— My problem is, in effect, my students are all looking past me, they’re looking over my head, they’re looking towards the future, right? And I want them to pay attention to me.
And so that’s my problem. And I guess, you know, you want to substitute this system where people go to work and to firms, right, but then there’s the standardized system. But isn’t this just gonna be exactly the same? Namely they’ll be doing something at this firm but they’ll be looking over the head of their boss or their manager or whatever, towards the future, to which they’re sending this signal?

Robin:

Right. So I would think of it more like, you know, picking, say, J.K. Rowling as an author, right? So all her life up until the time she wrote her famous novels, you know, she was seen as so-so. Right? She wasn’t at the top of rankings. And then she produced the spectacular thing, these novels that turned out everybody really loved and now she’s a really big deal, right? And the question is, like, when and where is that even possible, to sort of be ranked low and then do something that, you know, everybody says, Okay, well, we underestimated—you are obviously much better than we thought.
So if you think about the students in your class, imagine a student with like a B-plus average who writes one spectacular paper for you. And then you write a letter of recommendation based on that paper and you say, Yeah, I know, a B-plus average, but look at this paper—this is the best I’ve seen in years.
The question is how much will the audience weigh that, right? And that is what might have changed. Once upon a time it might have been a spectacular paper, and a B-plus average was the winning ticket to the next level. And now it might be that they just care about the GPA, and the story about this one great paper just doesn’t cut it.

Agnes:

Maybe. But my experience was not that they thought that I would write them a letter on the basis of a spectacular paper. My experience was they just very directly cared what I thought about them in the way that we all care what our parents think about us: Not because we hope our parents will say nice things about us to other people, right? But because we live inside that little world, and within that little world there’s a ranking and there’s importance, and there’s—

Robin:

Okay.

Agnes:

—you know, and all of that. And I feel my students are living in the bigger world, and I’m just a stepping stone.

Robin:

Right, so my story isn’t necessarily a story about the— what that’s in their head. So it just could have been in the past that because, like, individual professors’ recommendations about you were important, that people were in a world where they tried to impress the professor. They wouldn’t know why necessarily. They would have inherited a world and people had just told them, like, you know, It’s really important to do this, or That would be really great if it happens. And the people might have said, Yeah, okay, your grades aren’t the best but that’s not the most important thing. And then we might have just moved to a world where everybody keeps saying, The grades are the thing: You know, you have to get the top grade. And you could say, Yeah, but I got this other thing. People would say, Yeah, but that doesn’t count so much. And they might just— you know, again, people are just using proximal cues. And they don’t really know why things work. But they know what things kind of work.

Agnes:

And it seems to me though that if we go back to like, what is, you know, what is driving this, I still think I was right to say it’s efficiency. That is, the reason why we want standardization is that it makes assessments quicker. It allows you to assess people from a variety of backgrounds, right? And, you know, it allows you to be like—in a world where you’re hiring someone for a job from your neighborhood, right? Where you know everyone—that’s very different from a world where you’re hiring someone for a job and you’re hiring from around the world, anyone can apply. Right?
And so there’s a kind of— and you might think, Yeah, I want anyone to be able to apply, because I want to be selecting from the talent pool of the whole world rather than just from my block, right? And I saw an ad like this on Twitter, like, this is our new hiring world. We’re going to, you know, wouldn’t you rather be hiring from this larger pool, right?
And I think it’s like, what I’m seeing is that there’s this massive trade-off because like, yes, of course, you’re getting more talent because you’re open to the whole world. But because of the way that you’re assessing the talent, like, the— by the time they get to you those people have been through a training regimen that’s a little bit like the child abuse scenario that I described at that school. And so there’s this quest for sort of like efficiency, and it is leading, I think, to inefficiency.

Robin:

So I’d like to make the analogy to Moneyball.

Agnes:

Okay.

Robin:

So this famous movie based on a book about the Oakland A’s choosing baseball players.

Agnes:

Yeah.

Robin:

And so the key idea is usually the talent scouts would go out and look for the same sort of things. All the talent scouts had the thing they were looking for, so it’s sort of musculature, sort of like chin and attitude toward the game. And there were a bunch of like things that they would mark you off for if you had those things because it made you look weird. And then the Moneyball people, i.e. the Oakland A’s, hired some statisticians to come in and look at people’s record of winning various things. And then to find the outliers, the ones who were actually pretty good but sort of looked bad according to the scouts.
And because they would have this incentive to win the games, that could be a reason why they would pick these weirdos and then excel. But you can imagine a world in which winning the games doesn’t happen. You just pick a set of people, and it’s whether you look good based on whether everybody thinks they look good according to the same norms.
Which is more like academia, say. Right? You know, you hire a professor because you think they sort of have to walk the walk and talk the talk. And what you know is that when other people meet them, they’ll see they walk the walk and talk the talk, and now you’re picking them for the thing everybody else is picking them for, but it’s not necessarily tied to an outcome so much. And that’s the kind of thing I would focus on is whether like there’s a more direct outcome that’s driving, you know, choices, or whether everybody is so comfortable and safe that they can then just pick people on these sort of shared appearances.

Agnes:

So…

Robin:

Think about the military right? Before World War Two, there’s no war. There’s just various people in charge of various things, and some of them have the military bearing, right, and the right connections and like had the right scores in school or something. And then everybody knows that those are the people who look good. And then those people get promoted, and it all works fine when there’s no war. And all of a sudden there’s a war and you start losing. Now you start to go like, Who’s actually winning battles? And you promote them. And now you’re— it’s a different world, right?

Agnes:

Right. I mean, I could imagine that even in, you know the like Moneyball case, right? You might— I can— like I can imagine the following defense of the old way: You know, used to be that—I have no idea if this is true or anything about baseball, and I haven’t seen the movie but I know roughly the, you know, story—it used to be that baseball was like a sport of like grace, and you know—

Robin:

Sure.

Agnes:

—nobility or whatever, I don’t know. I don’t follow baseball. And now all it is is people trying to like maximize the number of times they run around those bases. And that might sound absurd with baseball. But like, if you— it’s not as absurd, if you think of it— if you think about something like, you know, ice skating, figure skating or gymnastics, right? Where if you look at the older like figure skaters and gymnasts, it wasn’t just counting the number of times they flipped, right?
And so as the sport progresses in a certain kind of athleticism it like, at least in some cases, there is less of an emphasis on the less quantifiable but still beautiful aspects of the sport. Like, and so you can look at sports like ice dancing by contrast with ice skating, which has still has a lot of those elements and be like, Wow, there’s this whole thing— figure skating is very different from ice dancing.

Robin:

So you can say there’s this beautiful essence that only the practitioner-elites can really judge. And if we have some sort of competition then that will be lost, because it’ll only be the, you know, this outsiders competition that drives the—

Agnes:

It’s just that there are multiple excellences or virtues instantiated in the activity but some are more easily quantifiable than others. And these systems are going to distort the activity so as to hugely emphasize the more quantifiable ones, which may not—antecedently to doing that process—you may not have thought those were more important.

Robin:

Right, but I want to distinguish two kinds of measures. One is just a measure that’s easy to measure, like Did you go to Harvard?

Agnes:

Yeah.

Robin:

Or are you six feet tall? Or is your skin smooth or something. And then like some sort of outcome/performance, like winning the game, hitting the ball, winning a battle, you know, writing the novel. Things like that.

Agnes:

Right? Like skating beautifully is an outcome.

Robin:

But it might be hard to judge.

Agnes:

Exactly. Or rather I think it’s not so hard to judge. Like, I think when we see the beauty of it we all respond. But it is hard to develop a standardized metric for it—that’s hard. There are many things we can judge to be beautiful, and we all easily judge them to be beautiful, right? But the idea of comparing two different painters or whatever, that’s gonna be hard to have a standardized metric.
And so you’re going to end up taking those things in life, those valuable important outcomes that we all judge as important, and we can see that they’re important but they aren’t easily quantified—you’re going to be pushed away from those towards the ones that happen to be able to be easily quantified, because those are the cases in which we can produce these far-traveling signals.

Robin:

But we can still distinguish these two kind of cases. So one is a case where there’s a set of elite, say, law-school professors who basically just pick each other. And then they might pick each other on the basis of some local standard of what they are impressed by. And that in principle could be just many different things, there’s multiple equilibria there, but they— if they are pretty autonomous, then that— the thing that matters is just, Do you walk the walk and talk the talk according to law school professors?

Agnes:

Right.

Robin:

And then on an alternative world it’s about winning cases in court. And then the people who are the most prestigious are the people who win cases in court. Now, both of these are sort of easy-to-measure things. Except the winning cases in court is a noisier thing in the sense that you can see in any one case whether they won, but like there’s just a lot of randomness in whether you win any one case.
So I’m trying to make the argument that in a world where we put more weight on whether you win in court, that there’ll be sort of more of a tie to some reality. Whereas if you just have law school profs picking each other then that can sort of drift off in all sorts of random directions that may not really represent the elegance of the— the intrinsic elegance is possible, it could just be some insider’s club of them picking each other.

Agnes:

So I guess I wonder though whether like— whether it’s going to be true in every case that tying the evaluation to a partial reality— that is, part of what it is to be good as a lawyer is to win cases, but it matters how you win them and how you select them, right? It might matter with what kind of justice and fairness you proceed, and—

Robin:

Sure.

Agnes:

Right? But so the point is, the winning the cases is the part we can measure, right? And so if we pick on that basis, it’s like picking the gymnast on how many times they can flip. Right? There is a reality. But there’s also a distortion going on. Now we go over to the insider’s club law professor thing, right? And then the question is just, you know, from the fact that they are picking people who are like them there’s certainly a distortion going on, right?
The question is do they as law professors have any kind of like connection to any reality of like, what it is to be good as a law professor? And like, let me say from my own field, philosophy, right? I think philosophers do, when they’re selecting philosophers, I think they do have some connection to a reality of what it is to be a good philosopher. They’re evaluating people on that, like— but they’re evaluating people who are like them. Right? philosophers will pick people whose philosophical talents can show up to people who are like them.
And so really the question is not, Are we tying it to a reality or not? In both cases we’re tying it only to a part of reality. And in both cases, by tying it only to that part of reality, we’re creating a massive distortion. And the distortion compounds upon itself as the people start flipping more and more times and doing less and less of the— right? In both cases. And so it’s just not obvious to me which system is going to produce like, you know, a net more realistic outcome.

Robin:

So I would say we have a set of historical examples that suggest that say, sometimes an empire becomes safe and comfortable and then they reward these various internal dimensions, and then an outside force comes and invades and knocks them out. Or sometimes firms become safe and comfortable and then they get dominated by these internal dynamics of who inside likes who, and then an outside firm comes and knocks them out.
And you know, you might say, If we can think in terms of these objective outcomes, like a nation winning a war or a firm going out of business or a team winning or losing games, we could say that if we tie the incentives and the reward more to those objective, outside measures, then we will have more outside success. Maybe at the expense of some other inside virtues, but this is like fundamentally the choice that humanity and organizations have made for a long time.
And it’s a story that plays out over and over again: That, you know, some new organization like a nation or a firm starts out and grows because initially it’s winning on the outside, it’s getting new customers, it’s winning wars. And then it grows on the basis of that. And then typically there’s some incentive and measuring system that’s tied to those things, like sales and product development and battles. And then it becomes large and comfortable and secure. And then its metrics become more elite, more subtle, more sort of insiders judging other insiders and telling the story that because now we’re safe and comfortable and rich we can pay more attention to the higher things in life.
And then they do. And they, you know, they say they’re judging on the elegance of the, you know, skating rather than the number of flips. And they have this world where they see that and they value that world, but then when there’s outside competition, you know, and they haven’t paid enough attention to the outside metrics, then they lose. And that’s like a fundamental conflict in human history that I expect will continue for a long time.

Agnes:

But you’d think then that if they’re going to lose, then that’s— the system is like self-rectifying?

Robin:

It’s self-correcting over a long timescale. But like at the moment, you know, Google’s very comfortable. It’s very rich. So it can make a long string of bad decisions and still exist. And, say, the United States now is rich and comfortable. So we could have a pretty bad military and pretty bad policies and still last for a pretty long time before we were displaced or destroyed. And so there’s just these long time lags.
The more successful you were initially, the more you built up a reserve, a slack, such that later on if you slack off and do badly, you will still, you know, have time. So you can think of an individual person, right? Early in their career they’re spectacularly successful at sports or something else, they get this big pile of money. And now they can screw up the rest of their life for a long time and spending all this money they have, and you know, the world won’t correct them because they can keep spending.

Agnes:

So one thing that I find jarring is the idea that this thing I’m seeing, this distant-signals problem, is the product of the world having gotten too contented and sort of inward looking, right? And comfortable, and having too much slack and too relaxed, and like that sports player.
And then on the other hand looking at, you know, the students, the undergraduates, the graduate students, the junior faculty, and seeing the insane amounts of pressure and stress that these people are under, where when I compare what I went through, right, the standards that I was subjected to were just much lower. Life was just much easier. I feel like there’s no slack! And everyone is at the very limits of their ability and they’re under so much stress…

Robin:

The Tales of Genji—I’m sure I’m not pronouncing it—is a Japanese novel from like, the 1300s or something. It’s one of the first, I guess, novels we have—it describes this court life in Japan in that period. And it’s a romance, sort of: A man and his many affairs. But it’s both a world that’s just very insulated, like Japan is unified and enemies across the water don’t bother them and so the court life can focus entirely on poetry and art and dress and dance and, you know, beautiful statues.
And there’s a whole world of all that, but each person in that world is struggling very hard to get their way into court and to work their way up. And you know, backstabbing each other, and… right? So you can both have a world that is, you know, prioritizing these high things and ignoring outside forces and competition, but in that world it doesn’t mean individuals are safe and comfortable.

Agnes:

But like, if I just compare— Okay, let’s say the university is a world, right?

Robin:

Right.

Agnes:

Is a kind of world. University of Chicago: So I’ve been in that world now, you know, since the 90s—I mean, not continuously—but if I compare the degree to which the university could devote itself to art and culture and literature in the 90s versus now, I would say less so now. That is, there’s less of a possibility of doing that. Because there’s less luxury now. What I want is that luxuriousness where we devote ourselves to art and literature and we stop caring about productivity or how we’re improving stuff in the outside world. And all the students are looking out and they want to be productive. And so—

Robin:

I think what you want is a secure aristocracy! Which is what tenure kind of produces, right? So you have some early stage of competition or inheritance, and then people are secure. And then they have this culture where they, you know, play with slack and beauty and all their internal conflicts.

Agnes:

I mean, what I want is leisure. I don’t particularly care who— like as long as I’m in the group of people to get the leisure, I’m not saying they have to deserve it for some reason.

Robin:

No, no. Right, right.

Agnes:

Right? But the point is that undergraduates should be in that group, too. The idea of a university, of a college, as I see it, is what Aristotle called spoudaia, serious leisure. They should be at leisure. It’s only when you’re at leisure that you can appreciate great works of literature and philosophy and whatever, and there’s less and less leisure.

Robin:

So if, when you got into— so and some— I’m told like in Japan at the top university, just getting into the top university is enough and you’re set for life. And you don’t have to work very hard at the top university, it’s pretty much party time and then you’ll just get your job, right?

Agnes:

Right.

Robin:

And so if employers just look at that—whether you got into the top university, they don’t care what happened there—then that’s how it’ll play out, right?

Agnes:

Right.

Robin:

But if they’re looking at the people at that top university and comparing them by grades, or other sorts of things, and then there’s these— only a limited number of top slots that all these people are competing for: Well, now they’ll be competing, right? And so you know in some sense it’s about, When in your career can you be secure? Or what kind of positions can you be secure in? And clearly undergraduates are not there yet in our society, although perhaps in Japan they are.

Agnes:

I guess I think that, you know, when I was an undergraduate, if I think back I don’t think I would have said, Oh, I’m set for life. Now I’m secure and so I can devote myself to these things. It’s more like I just didn’t think so much about it, right? I mean imagine if, you know, every day of your child’s life you talk to them about how they might starve to death if they didn’t like, you know, get a job and whatever.
And like what are children, in every decision about what toy to play with, thinking about—How is this going to affect my future? What kind of a childhood would that be? Right? And I feel like, you know, what kind of a college life is it when every single decision of what class to take, what book to read, is made with reference to this distant future. What I want is for them to forget about that distant future and live in those four years. Like, you know, in the space of education.

Robin:

Well like maybe another thing that’s going on here is just, when success is defined more in terms of slots in established organizations as opposed to sort of independent accomplishment. So like think of J.K. Rowling, right? J.K. Rowling didn’t get an elite professorship. Right? She was independent and then wrote a novel, and that’s what carried her to fame, right?
So in a world where you thought, I’m going to, you know, have a long lifetime, and I’m going to keep trying to do spectacular things, and I’m gonna kind of mostly do them on my own, and then eventually one of them might really succeed and carry me to fame and be a big shot—then it’s less about me finding that slot. And having me put in a slot, and having all the right cues that let me be in that slot and more about like what I can accomplish. And so you might say, well, maybe in the past these slots didn’t matter as much.

Agnes:

I mean, I think that there were slots— like if I think about how philosophy worked, right? You know, I don’t know, in like the 1970s or something, getting a job in philosophy. I think it was like you’re just— you didn’t maybe didn’t even write a dissertation, right? But like a professor at the university where you were, you know, a graduate student might call up their friend and be like, Do you have a spot? I have this good student. You know, that’s maybe how you’d get a job. Right? Obviously very in some sense nepotistic. It depends on who your professor knew. It’s like that thing where you’re in the small town, you’re just hiring from a small town, right? There were slots, it’s just that—

Robin:

They weren’t coordinated in some sort of global evaluation system.

Agnes:

Right. And in particular, you didn’t have a set of expectations as to what would be wanted from you as you’re going through your graduate school experience. There was no way for you to optimize your performance for the future. And I think that’s what it is, is that students very much feel like there’s a way for them to optimize their performance for their futures, and that if they don’t do that, you know— so they’re living under that threat.

Robin:

So I’m more of a futurist and I want to like think about the very long run. And I think on the longer timescales you think, the less slack you’ll see. That is, you have slack and some room to play in the short run, but in the longer run you’re more going to be held to account: You and your university and your firm and your nation and your culture, you’re more going to face competition over the longer run.
And so if I look at the long run, I say I want my firm, my culture, my city, my nation, whatever things I’m part of: I want those to be competitive. I don’t want them to feel like they have so much slack that they can just ignore these outside competitive forces and just form some internal culture and standard of art and just pursue it and enjoy it, if that’s going to come at the expense of this long-term competition. Because those things will then fade away and lose out to the alternatives.

Agnes:

Right, so we disagree about that. But I actually think the more interesting thing is, to get back to— I still feel like I haven’t gotten an answer to the question of, Are we in a high-slack or low-slack situation? You tell me that slack is in some sense increasing because more and more the universities are employing this internalistic, whatever, perspective. And I’m saying, But I’m seeing less and less slack because there’s more and more competition between the students.

Robin:

Right, so I want to distinguish slack at the entire culture level or the entire organization level, and slack at the individual level. And those can be opposite.

Agnes:

Okay. So I mean in some sense if you’re gonna tell me that the cultural lack of slack will produce more slack at the individual level, then I’m happy. You know, what I see is that the lack of slack at the individual level—the students feeling like they have to be optimizing their performance for this distant future—that that is producing, I actually think, less efficiency, in the sense that these students could be doing more excellent things that they’re not doing.

Robin:

Right. So think of two worlds of firms. One world of firms is there’s a lot of firms that start small and they grow based on some initial set of people, based on family connections or some other random set of friendships, and some of them just happen to succeed and they take off. And then people in those organizations have a lot of security and they don’t— they have a lot of slack, at least for a long time. And other people who are at the unsuccessful firms, they have to quit and go, you know, find a more successful one. But at least some people have slack here, the ones who happened to pick the successful firms.
Now imagine a world sort of more like management consulting or something, where there’s just objective measures of how good you are as an employee. And all the firms accept that same objective measure of what school you went to or what management consulting firm you worked for. And now there’s a strong competition for these limited number of workers who have this metric.
And now, like, you have to fight hard to get the metric and almost none of the people have slack. Because they’re also easily compared, that they’re all competing against each other. Whereas in the first case, if you happen to be in one of these firms that wins, you just are part of some local culture that’s just very entrenched. And it’s hard for outsiders to come in and compete with you because you just know this firm well, and you’ve been there for a long time and everybody knows you. And now, you know, you at least have slack there if it’s a successful firm.

Agnes:

So in a way the firms, like, protect people.

Robin:

When they’re successful.

Agnes:

Yeah.

Robin:

But, you know, but because people get entrenched in them, and there’s just—each firm has a different culture, a different work style, a different set of products, and you know, then you become more local. So I mean it’s the same way you might imagine, you know, a world of farms spread across the landscape, but each farm is a little different, with a different climate and a little different local…

[Recording software broke, causing conversation end]