Steve Hsu Interview Transcript

There are .. just too many automated errors in this one.

They were all aware of the realities now, which are highly contested. So you have a bunch of these kinds of old, uh, kind of blue blood, uh, academics and humanities folks are having a fun time at their prestigious university. And then as you said, you have the cold war and you have the stem revolution and become, we come out and completely changed their status.

And not everybody has what I would consider a kind of high amplitude exposure to extremes of say capability or hard worker achievement. The official curriculum in California, there was an amendment there to like deny that kids are naturally gifted. You're talking about like anti-vax level kind of like scientific denials. Is that if you're not high M you cannot reason from statistical data.

Hi. Hi, welcome. Welcome. This is the, from the new world podcast. Today on the show, I'm speaking with Steve Hsu. He's a professor of theoretical physics and professor of computational mathematics, science and engineering at Michigan state university. He's also the former vice-president of research. They're the founder of digital security companies, Safeweb acquired by Symantec and robot genius, as well as genomic startups othram and genomic prediction. Finally, he's the author of the intriguing blog information processing, which all of you should read and is linked in the show notes.

The notable topics we're going to discuss today, man, that's a lot, are history and politics of academia, individual differences in intelligence and athleticism, some our experience with high level mathematics statistics as an anti meme, the tension between merit and ideology, math education theory of mind differences between the Midwest and immigrant suburb schools, where Steve and I grew up respectively and informal networks as alternative credentials. We mentioned a few keywords in the show that I want to give a quick note on as we'll talk about.

There are groups of people with exceptional math ability, high M for short, as well as exceptional verbal or writing ability. High V for short as well as ones with both. The first two groups are jokingly referred to as quote unquote, shape rotators and quote unquote wordcels in some tech circles online.

There were multiple failures in my audio setup so part of the audio you're listening to is a backup of a backup. You probably can tell that this isn't as smooth as the first episode, and it isn't as smooth as future episodes either. That being said after the most time I've spent editing any podcast episode, it, it sounds all right. Last, but not least we're not taking any contributions right now.

So the best thing you can do to help us out is by subscribing and letting your friends know about the podcast. You can do this either by social media or the old fashioned way, and there is no better recommendation than yours. Finally, if you have suggestions for future guests, please let me know by leaving a review, wherever you're listening or by emailing contact at, from the new.world without further ado, here's Steve Hsu.

Uh, welcome to the show, Steve Hsu. Uh, the way I really want to start this interview is that, uh, noticed from your work and from your podcasts that we probably hold, uh, very similar positions on questions of merit, natural talent, and self-discipline, that is strongly in favor and, uh, generally believing in all of them. So with that in mind, I do want to ask what kind of experiences in your life shaped, how you see your world and how you came to hold those kinds of preferences, both in your childhood and later as a theoretical physicist.

Um, that is a huge topic. Um, I'm assuming you kind of want a fulsome answer, so, uh, it's okay to go into some kind of detail in my answer to that.

Yeah, it's a, it's a long form. It's a long form podcasts. Keep rolling. I will interrupt briefly if you are, uh, really taking a lot of time, but the odds are that's not going to happen. Great. Yeah. Feel free to direct me how you want, because you know, I'm really, I want to kind of create the show that you want.

So, uh, you know, within, within parameters of my being able to answer the questions, so, um, the issue of natural talent, uh, hard work, uh, success in life or in a certain career, those are topics I think almost everybody is interested in. Not everybody thinks analytically about those things and not everybody has what I would consider a kind of high amplitude exposure to extremes of say capability or hard worker achievement across multiple areas.

So not everybody is really actually qualified to comment on it, even though people tend to have very strong emotional attachments about whatever particular views they take on the subject. So in my own case, um, when it comes to say cognitive ability and intellectual work, I guess a few unique aspects of my upbringing include having a next door neighbor.

When I was growing up, who was in the terminology of the eighties, which is no longer used, was retarded. So he had an intellectual disability. But we grew up together. We lived together for lived next door to each other for many years. So I knew him quite well. So I understand that end of the spectrum, probably better than most people, unless you, you know, you're the father of a downs kid or something, because most people have not interacted over many years with somebody who has a intellectual disability gone to the park and played with them, played in the backyard, had squirt gun fights.

So I think I understand that end of the spectrum somewhat better than the typical person, I was a precocious kid. So, uh, I understand the high end. We can get into that a little bit more if you're interested. When I was, um, I guess I was maybe 10 years old ish. Uh, I took some standardized tests. They were required in Iowa.

So I university of Iowa has a school of education and they produce something called Iowa tests of educational development. And those are not super high ceiling tests. I think the ceiling is only 99th percentile, but there are many, it's a, it's a large battery. So there are many components. And I distinctly remember maybe in third or fourth grade getting my scores back and I'm not sure I was supposed to see my scores, but somehow they showed me my scores or my parents showed me my scores.

And you know, there may have been four or five batteries, many batteries. I mean, some have more to do with math, some having more to do with, uh, verbal or reading. Ability and maybe even some having to do a spatial ability. And I remember getting 90 nines across the board and already at that time, I was thinking I didn't have kind of any kind of sophisticated, no notion of the concept of correlation, but I intuitively understood what correlation was.

And I said to myself, well, how correlated is this? Because if there's a 1% chance of being a 99 in on battery one and similarly for battery two and battery three and battery four, am I really one in 10 to the minus? You know, am I really one in 10 to the 10th in capability or tend to the eighth? That can't be right.

So these things must be somewhat correlated because I can't be that unusual. Although I checked around with my friends and I did, nobody else had all night. So, so anyway, uh, so that got me interested in the field of psychometrics and my father who himself was admitted to Tsinghua effectively. What long story about exactly what Tsinghua university was because this was during world war II, but effectively admitted to Tsinghua.

When he was 16, he was already also a smart guy and S and, and highly accomplished aerospace engineering professor. So, um, you know, he had what then in the pre-internet era was for a precocious kid like me, the magic secret. He had a library card at the university library. And so I would often go with him or he would take me to the university library and just drop me off and he would go to his office and I would check out a ton of books using his faculty library card and the faculty library card lets you check out as many books as you want.

So I would check out pilots and piles of books and already at that age, it was soon after this testing experience that I had, I was able to find books in the Iowa state university library, including the Terman study, which was a, is a multi-volume study by, uh, uh, a psychologist at Stanford who studied gifted children in a longitudinal study.

Um, and, and many books on psychometrics and things like this, uh, technical books on psychometrics. So I was already pretty interested in this stuff because of the weird test results I had obtained and was then able to read up on the subject. So I, I knew a lot about the subject already at a young age and that then made me a careful observer because I noticed that in the Terman study, you can go and get it out of your library.

If you have access to a good library, maybe you can find copies online. You know, the term and study. They, they, they test all the people that are in the study. I think the induction was in high school. And then they follow them through their lives and then they track and they track them. It's a longitudinal study.

So they're tracking all kinds of variables and then they test them again late in life. So for example, they even look at the rank order correlation between where individual termites, these kids were called termites, um, where they were, when they were young, the rank ordering and what the rank ordering was when they were say 65 and how their career choices affected, you know, the differences in rank ordering.

And so for example, one interesting result, which I think people have a hard time accepting is that the termites who had kind of blue collar jobs and they remember this is a very different time in America when Terman was doing his study. So not everybody could go to college or get a PhD, quite a lot of the high IQ kids that Terman was tracking, uh, ended up in, you know, what we would consider now kind of blue collar jobs, but there didn't seem to be a differential effect on their scores late in life, depending on whether they became a college professor and a researcher versus, uh, you know, and so anyway that I was already quite aware of this stuff.

And so as I went through life, I was actually interested in kind of tracking, you know, my own everybody's own life history is their own longitudinal study, right? Although of limited sample size, but I was more interested. I've always been more interested in my life in keeping track of people that I once.

So I'm much more interested in checking out, Hey, this kid that I knew really well in third grade, and he moved away, whatever happened to him, I'm much more likely in the age of Facebook to track the guy down and call them up and say, how are you and what's going on. So, you know, I've sort of artificial or intentionally increased my sample size for the longitudinal study of my own life.

And so I, if you're familiar with general relativity, there's a concept called the fiducial observer, which is, um, a kind of person that's kind of floating following a geodesic and curved space time. And, and you have families of geodesic observers that map out the space time for you. It's a, it's a kind of concept.

It's a, it's a concept that Einstein introduced. So I think of my own lifeline life curve as having various other fiducial observers that, you know, I, I made good calibration measurements on when we were both in high school, but now this old friend of mine is a corporate lawyer in Silicon valley. And I can call them up and compare notes as to how life went and what happened.

So, so although I'm not formally in this area as a researcher, I'm quite interested in it. I know lots of the leading researchers who do this kind of work. So for example, Ben Bo and Lubinski are the two researchers at Vanderbilt who did this SNP, why study in SVP? Why study? So those are studies of mathematically and verbally precocious youth that have been tracked now for over 50 years.

So it's like almost like a secret. Terman study, but more modern. I know them quite well. And actually by coincidence, this is an amazing life coincidence. The reason partially why I was state university library was so well-stocked in books on this subject is because Ben and Lubinski began their career at Iowa state university.

So they were in Ames. I think they may have actually been in Ames, you know, when I was growing up as a kid. And, uh, there was a very strong, um, program at Iowa state for the study of gifted children. So, so there are all these kinds of weird coincidences. So, um, that's a little bit of an introduction to how I got interested in the subject and why I think I know more about it than a lot of people.

Um, the other thing I want to mention is that I was a pretty good athlete. So I was at the state championship level in competitive swimming growing up. And I was a captain. I think they tell me I'm the youngest team captain ever at my high school because I graduated high school when I was pretty young and swimming.

And so I'm familiar with fairly high end athletic performance too. I mean, some of the kids that I competed with became Olympians. And, um, so I actually understand that world better than most people who say their main window into this is sat scores in math competitions. I think I understand a broader aspect of it because.

I know how good certain athletes were when they were young and what they can do in the weight room. And, uh, then what happened to them and how well did they do when college has competitors and how many made the pros? And I just have, I think, a broader sense of all this stuff. Um, in addition to that, having been an entrepreneur in Silicon valley, I saw a completely different set of talents at work.

You might call it kind of generalist talents that are more to be found in venture funds or among tech non-technical tech, founders, business people, Harvard MBAs, that's a whole different panoply of human capability, sort of soft skills ability to, um, read other people's psychology and emotions. That's a whole different set of capabilities are harder to measure, but definitely real.

And, uh, if you're, uh, if you're a careful observer and you've been in a lot of pitch meetings and board meetings and meetings with, uh, billionaires and venture capitalists, you can hedge fund managers. You get, you get a certain view into that world as well. So I hope I haven't gone on too long to answer your question, but that, that at least sets a kind of baseline for my life experiences and level of interest in this topic.

Yeah. I think the point that you made about seeing things near the tail end, I think that's something that's actually incredibly. The way I Intuit this right. Is, is that like most people in a kind of ordinary class, like an ordinary, like high school class, let's say, um, the range of talent between, let's say like the top student and, uh, in the worst is probably actually not that, uh, not that incongruous with a lot of the narratives that we tell about people being, uh, pretty similar and most of the difference being able to be made up by, um, being able to meet up and be made up by just hard work.

Right. But, um, I think actually I have an experience that's like very similar to this as well. Um, I went to the international Olympiad informatics and the USA, uh, computing Olympiad as well. And then in the latter case, there was this really legendary, um, computer scientist, computer, uh, competitive, uh, computer scientists, uh, named, uh, Benjamin Qi.

And, uh, one of the anecdotes I like to tell, so this is the, this is that the, uh, essentially national, uh, championship in, uh, in just, uh, computer science problems, algorithms problems. And, uh, the third day, this it's a multiple day contest. And the third day, what we mostly agreed was the most. And the distribution, uh, looks something like this.

Okay. So, uh, well, more than well, more than half of the participants, uh, solved zero problems, uh, a few solved one problem, the top four or five solved, uh, two, uh, and then, and then, uh, this guy, Ben, he solves all of the problems. And I think like a little bit more than half of the time, uh, after the contest, after the contest, we have a lunch break during the lunch break, he enters another contest, uh, an online one 30 minutes late and he still gets like top 10.

Uh, and this is top 10, not just in the states and not just, uh, for, uh, people who haven't graduated high school, which was for this contest. But this was across everyone across all time across the world. This was on the website code, forces it for people in my audience who have heard of that. So, um, if you've, if you've like, uh, never encountered one of these contests, they're incredibly exhausting.

Everyone, everyone, after the contest, aside from him was just, uh, completely done, completely done basically for the rest of the day. Uh, but he had this kind of, not just, I think, greater knowledge, like you might think in say, uh, in say like a. Uh, history exam or something like that, anything where you're just remembering things, you're just writing them down.

But the ability to just chain together, uh, a very long sequence of insights, uh, in a way that is just kind of incredibly time consuming or incredibly, uh, resource consuming or thinking consuming for, uh, just about anyone else. Even like, even like the person who was second place, who I also know quite well, and you never get these differences revealed in normal classes, right?

If you have an, this competition was say like 20 people, um, if you have just a normal class of, uh, of students, I don't think you get anywhere near even like one, 100th of this kind of gap, even just the gap between like first and second, in this case, you don't get this kind of experience of just being like, um, having the differences so clearly laid out in front of you and trying your very best and still coming up.

So, so far. Yeah, I think I completely agree with that. Um, I would say one place where you do see it is in high school sports. So if you are training really hard and lifting weights and doing stuff in the off season, and maybe your event is the 400 meter dash or something. You might encounter some kid who, you know, in your senior year, you, you get down to 55 seconds or something, and some freshmen comes along and this guy is just a natural talent and he blows you away.

And he's not even training and smoking in the off season. You know? Um, you can just see that, no, you know, maybe I'm doing something wrong and if I train totally differently, I could be as good as Joe, but no, actually Joe is not working very hard and he's just naturally a much, much more gifted than I am.

But when we went to the state meet, Joe could never crack the, make the finals, you know, or he may be made the finals at the state meet, but he couldn't qualify at the national level. So I think, you know, where you have well measured, intense, you know, intense competitions where people prepare for the competitions and there's high stakes.

Um, you can see, you can see that there's a tail there. Another way of saying it, which my wife hates whenever I say this. Cause I, I sometimes say it when we're talking about our own kids, you know, in their various competitions. I always, I sometimes say, honey, there are levels to this thing. And people just don't like that.

A lot of people just can't accept that there are levels to these things. Um, but I agree with you. I, you know, on the subject of math competitions and things like this, I guess I've written scientific papers. And these are two different two or three different people, a Putnam fellow, and an IMO gold medalist.

So I think I've seen, I actually, I probably the most off scale guy that I know well is a guy called Noam Elkies I don't know if you've ever heard that name, but, um, he's the youngest guy a little bit familiar, but I don't think he was the youngest guy ever to receive tenure at Harvard in the math department. And he think he's, uh, well, he was, he would have been a four time Putnam fellow, but he was only three times because he was only college for three years.

And then, uh, and I think he got a perfect score at the IMO his last year. Maybe he missed two points or something. I think he was, yes. He was maybe just edged out by Perlman one year. So anyway, I doubt there are too many people who are more impressive than Nome in terms of just raw power for contest type, magnitude, uh, mathematics.

Um, so I think I've seen the whole range. Uh, you know, I also kind of know people like, you know, ed Witten, who, you know, maybe they're not the greatest at the contest math, but in terms of real, real deep mathematical, uh, research capability, they're really at the top. So I think I've seen it all. Um, and I'm of course I'm intensely interested in the topic.

Yeah. I think people are fairly willing to accept these kinds of natural differences when it comes to athletics. Probably it's because of like mentioned that you actually can just see it, uh, much more often. Yeah. Just in kind of like everyday life living through high school, or even just like looking at the people around you.

Um, but people tend to be less accepting of this when it comes to things like intelligence or when it comes to things, uh, even that are kind of like close to being able to be, uh, or like much more able to be influenced, like, um, like conscientiousness, but people still tend to be like very, um, very just allergic, I think.

Well, I think there are two, I think there are two, I think there are two components to this one is that it's so visible in sport. You know, if you watch someone run the a hundred meter dash, you know, it's just, it's it's this, you can just see it. It's so obvious. Whereas you can't just take your brain out of your skull and weigh it and say like, oh yeah, look, mine is a lot bigger than yours.

Right? So, so there's an indirectness or complexity to the measurement that you're trying to make in order to make a comparison. And secondly, it's, it can be tied up in people's self image. So, so the more it's tied up into people's self image, then the harder it is for them to acknowledge that there can be large, that there can be levels, uh, qualitatively, different levels, especially people who, you know, like the people who are most likely to be writing popular articles for mass consumption are typically people who are somewhat above average in capability.

And they're quite proud of. But then of course they're defensive about people who might be significantly more able than they are. And so then obviously it's not popular for that. It's not among that population of people I eat the popularizer is, uh, it's not, it's not very comfortable for them to want to dwell on this topic.

Yeah. I think this is something that I definitely hope to get on, uh, get onto later in the interview, but I definitely have a lot of theories about the structure of these kinds of, uh, legacy institutions. Right. What type of people they attract, what kind of dynamics they, uh, they, uh, kind of arrange themselves into.

And, uh, that's, that's kind of the focus of my writing right now. Um, but I think before we get to that, there's this, I think very fundamental question of whether it's ignorance or, uh, or refusal, right? So you could just say, like, we were kind of bad at educating people about these things. We should be investing more in kind of, uh, public education.

We should be, um, making these clear presenting the, all of the best studies, uh, and just doing a better job of educating the population on these kinds of questions. But. At the same time. I think that if you just look at the most educated people, right. People who are, I guess, people in the field kind of, well, you're not going to be in the field if you can't kind of accept these, uh, accept these truths.

But I think people who would otherwise see themselves as very well educated or that society would see them as very well educated are still kind of, uh, are still kind of often in denial of these things. Right. So w would you agree with my kind of view of this, or do you think it's more of a, um, more of a question of like, uh, of, uh, of providing the information versus it being like a social problem?

Uh, again, very complicated questions, so they're really different populations of people. So even if you only consider the subset of people who are highly educated, right? Well, some are highly educated in the field of psychometrics. There's not much disagreement in that population about the actual empirical reality, but if you go one step over to other psychologists who are not themselves, psychometricians those people, I think only in perfectly understand the situation and have natural have already like a sort of grown up natural defenses against some of the conclusions.

So, so you have researchers. Consistently trying to, you know, conduct studies, which show the, which, which would show that IQ doesn't actually play a very big role or conscientiousness plays a bigger role or grit or something like this. So even if you go one concentric circle away from the pure psychometricians, you, you immediately start to see active resistance to the concepts.

Um, and then if you go even further, say you're talking to a history PhD or a social scientist or economist, they could, they could be all over the map. They could, they could understand the reality or they could, uh, having read lots of Malcolm Gladwell books, actually be totally confused about the subject.

And, um, when you get into the pure sciences where it's a little bit like suddenly you're talking to some track coaches or some, you know, weightlifting coaches, they suddenly get it because they've been training grad students. They have to make a laborious selection for which grad students to admit to the department, which ones are going to be in their research group, which one they're going to invest time and energy into training, you know, which is the best postdoc to hire, which is the best junior faculty member to hire.

So they they've been in the trenches doing this. And there you find a lot less disagreement about the general reality that there are levels. They may not be familiar with the detailed, new miracle quantitative results in psychometrics, or even how. For example, you would extract the principle principle component of general intelligence and how well does that load on a particular sub battery?

You know, they don't know that, but they can, they can pick it up if you start explaining to them, but they see at least the empirical reality that there are levels and that natural ability plays a big role as does hard work. So the $1 trillion question probably more is, uh, how can we, how can we change this?

Right? What can we do to, uh, make it so that more people know about this? Because I think there are very obviously direct policy consequences, um, most notably in education, but in basically everything right, how you manage an organization. So I think it's just something of crucial importance to get as soon as possible for people to, for people to understand this.

So, so what can we do? Right? Well, her hands tied. It's an interesting question. So let me first answer it in a kind of historical way. Um, now you'll notice that most of the truths that we're trying to uncover are kind of low tech. Like what do I mean by low tech? Like what's involved in these studies, like, what are the very best studies in this area?

They're basically big longitudinal studies. They involve some testing or standardized testing of people and then just kind of recordkeeping and, you know, just keeping track of people. And then, and then some very simple, descriptive statistics, right? It's not none of this. In some sense, difficult compared to the kind of things you were doing in your informatics competitions, this would be considered kind of intellectually trivial activity.

Right. But, um, nevertheless, and so it, it's not surprising that the truth has been known to the people who cared to search for it has been known for some time. Okay. So let me describe to you the meritocracy and the educationals talent selection system that existed still when I was young, but you know, has undergone quite a bit of decay, you know, uh, by the time that you're much younger than me, by the time that you entered the system.

Okay. So in the early days, Harvard and the Ivy league schools were not even themselves, very meritocratic. They, they were mainly, you know, schools for rich kids and alumni legacies and things like this. And there was a period which really largely coincided with the cold war. Okay. The United States was under pressure.

We didn't get a satellite up first. They got the satellite up first, right? That's what the Sputnik moment was. Right. And the U S wasn't a leader in natural sciences and mathematics going into world war II. We only became one because of lots of refugees that we recruited and then subsequent cold war investments in these areas.

So there was a period of time, you know, roughly coinciding with the cold war. And when I was in high school, the cold war was still in full swing. We're all of these systems got perfected and there was a very high ceiling well-run sat test, which kids took. And that test was actually utilized deliberately by Harvard to select Iowa kids in the rough who had none of the Northeastern establishment Polish, but based on their, uh, scores, basically psychometric scores, they might be admitted to Harvard, uh, as you know, uh, exceptionally talented future scholars.

So we had a very high functioning system like this for a long time. And if you went to bell labs or certain places where, you know, maybe even the defense industries where they were really trying to get stuff done, they were all aware of the realities now, which are highly contested. These are all highly contested things now, so I can get canceled for, you know, if I were a university administrator, I could get canceled for having some of these views, right.

Not because it has to do with, even with different, uh, uh, racial groups or even, you know, uh, sexes, but just having this view that, you know, some people are smarter than others and, uh, some subjects are more difficult than others. That's actually quite verboten thinking these days in the academy, or at least you would have to state all these views privately.

You couldn't, you, you shouldn't be broadcasting them. So. Anyway, we went from a well-functioning system, which, you know, it had other problems, like it was more racist and sexist than the current system now. But on the other hand, at least if you weren't, uh, being discriminated against, based on your race and sex, I E or like a white male or something, um, it was a pretty bureaucratic meritocratic system.

It was designed to be so, and it actually functioned. So, and, um, you know, one of the observations made about the Caltech curriculum is that, you know, if you go as an undergrad to Caltech, no matter how smart you are, and I mean, literally you could be a Putnam fellow or an IMO gold medalist. Um, the performance on the materials you have to master in the undergraduate curriculum at Caltech is still normally distributed, even though they're cutting as high as they can on the IQ, uh, you know, variable to get the students in there because these are intrinsically hard topics.

And so, and he, even, even the Putnam fellow guy can't completely sleep in his classes at Caltech. If he, if he doesn't, if he does goof off, he could easily not get an a in the class. So we had a high-functioning meritocratic system and now we're living in the ruins of it. So that's my historical answer.

Now you maybe what you then want to say is how do we get back. Uh, a well-functioning system and then with the problems removed, like sexism and racism and stuff like that, we just start a war with Russia or China, right? Yeah. So it's an interesting, we're entering a very interesting period now where I, I don't know that much about how Russian I expect.

It's probably pretty good, but th the Russian selection of talent, um, uh, yeah, absolutely. Yeah. Like math Olympiad system, their informatics Olympiad system. The problem is that all of their talented people keep immigrating. That's the problem. Yeah. And, but China has a imperfect, but at least deliberately meritocratic, we focus selection system and yes, they also lose a lot of their talent, uh, say to America, but, but less and less over time.

So, um, so one answer to your question might be with natural forces in the absence of ideological actors who want to prevent it from happening, uh, natural, competitive forces might re-institute, uh, some of these Meredith meritocratic systems, but I think right now, if you look at the first river two, it's still in the negative direction, it's still, we're still, we're going to have more ruined before we start recovering.

Yeah. I think one of the models that I use for this is that, um, merit is kind of, or like innovation or like objectivity is kind of an anti-police. When you have a kind of a chaotic state, if you have a kind of a institution where you're not optimizing for anything, right? You have a lot of degrees of freedom.

You have a lot of choices that can be made, uh, where you can basically justify any choice you make with some kind of post hoc explanation. If there is no kind of hard, uh, assessment of whether you're actually doing that well. And then, uh, in, uh, when you have that, you have all sorts of arbitrary things that can take hold all sorts of politics, whether it's old-school racism, uh, or whether it's, uh, kind of adherence to an ideology as is more common now, or whether it's well new school racism.

Right. That's what, that's really what I see a lot of these admissions, uh, these admissions policies as, uh, and really the thing that I've kind of been trying to push, especially. And like, quite frankly, I don't think like right-wingers are pretty receptive to this is that we want to have a kind of imposition of that order.

We want that order of, uh, of objectivity and that selection to be not just like a thing that you add on afterwards, not just a thing that you do after you've kind of removed this, this kind of like, uh, politics, but actually as the cure for, for this politics. So I guess the question. And that I want to ask her that I want to get you to reflect on is you, you've kind of seen the inside of these things, right.

You've kind of seen, and I guess you yourself have like participated in the actual kind of experiment in this forum where I remember you were talking about, uh, your time as a VP of research, that they really wanted someone who was in these kinds of objective fields to be doing all the applications. Um, but do you think that there's a potential of doing like the kind of merit first order first, uh, approach, or do you think we do have to address the politics first before we get to the kind of inherent, uh, inherent order afterwards?

I think you have to look at it in terms of a kind of more sophisticated study of institutions and how they operate and then even more broadly than because it's not the institutional by itself then essentially just political forces now. Yeah. So, you know, first of all, the people who control these things, I E who determines what admissions regime Harvard should use, um, you know, to some extent it's the president of the university, uh, you know, and the president himself is, or herself is hired and fired by the board of trustees.

Um, in a state university that board of trustees might, you know, they might actually be appointed by a governor who's a Democrat or Republican, you know, It becomes rapidly, very complex. And actually this is a surprise to all stem people and to most professors, the success or failure of the institution, even though it's ostensibly as an institution of higher education is almost never judged on how many Nobel prize did you win prizes?

Did you win in the trailing 10 years or how many members of the national academy did you produce or only a few quant oriented stem professors care about that. So when you, when you go to the meeting with the president and you say I've done a bunch of research, sir, and, uh, it turns out, um, Harvard is falling behind Stanford in terms of, uh, output of, you know, you know, these academic prizes or, you know, the president might care a little bit, but you're talking about like 10% of his utility function.

You know, he cares more about raising money, uh, dedicating a new building. Um, you know, if it's a big state school, how well is the football team doing? How well is the basketball team doing? Um, angry, wealthy alumni whose kids can't get into the school because they're not academically strong enough. Um, you know, so it's, I think it's a mistake for professors and, you know, people who are in science and technology to think that the things they care about are actually central to the power centers of these institutions. Now, maybe MIT and Caltech are slightly different. Okay. Because they have a very strong, their, their whole institutional identity is focused on these things.

But beyond those schools, I think there are very few, I will tell you a funny anecdote. I'm just slight digression, but I am a psych psychology researcher named Jonathan Y um, w a I did a study where we looked at per capita production of Nobel prizes, national academy of science, uh, members across the us, and also global universities.

And, um, you know, that if you think about it, that's like the most, you know, that's a, one of the best ways to track, like the quality of, you know, institutions, right. Because you're actually looking at well, what is the probability that a kid that they've admitted to the school and they're going to train for four years later going on in life and doing something really significant for civilization.

Right. It seems pretty, pretty good metric right now. Astonishingly, how many people have tried to compute that metric and published it somewhere? Hmm. Well, there's Jonathan and me. So that's two.

That's it now how many university presidents, when they become aware of this are actually interested and like ask at least we'll ask a few questions about the, uh, oh, that's interesting. What did you compute again? Did you, you normalize to what the, the number of students in the alumni population? Oh, that's clever.

I never thought of that. Um, very few, but I'll tell you a funny story. I sent the information to my Alma mater, Caltech, which is ranked number one among us universities on these kinds of metrics. There was almost no interest, even from the current president of Caltech. Who's a physicist on the other hand without, uh, totally unprompted after one of our results was appeared in nature.

Uh, the director of Ecole Normale Superieure, which is the global number one, and I don't know how much you know about the French system, but that is the most elite university in the world. In some sense, it's small. Okay. It's like half the size. It's even smaller than Caltech. And I'm very, very hard to get into ENS.

Um, and they produced a dozen fields medalist and it doesn't Nobel prize winners. Um, the director of ENS, who's a theoretical physicist. The, the Caltech guy was only an experimental physicist. No. But the, the director of VNS, uh, emailed me and thanked me. And he said, when I next meet with the minister of education, I will be sure to show him these numbers.

So does anybody care, does anybody actually want to accurately measure the things that you and I care about? Nah, not very many people. What was the, what was the interest out of Asia? Just wondering, was there any, I didn't get any feedback, but you know, it just, you know, who knows who read it? I mean, you know who, I mean, who looks at these things, you know, like I'm pretty friendly with, uh, at one point I was being recruited by Hong Kong, university of science and technology.

So I'm pretty friendly with the top leadership there and they appreciate some of these things. They know what they're trying to do. The Shanghai annual world ranking of universities, um, is pretty numbers driven and, and they do even have like, they later added a population normalization for prizes. So, so they would actually adjust like the number of Nobel laureates on the faculty, um, relative to the overall size of the faculty.

So they started getting into some of these more refined metrics and that, that, that we had, uh, we had been interested in. So I think in Asia, they're a little bit smarter about this stuff and over time, you know, it may pay. But overall the it's kind of a deafening silence in terms of, uh, who who's actually mining the shop in terms of, in terms of, uh, making sure that, you know, uh, us use and development of human capital, is it at the high end is actually working well?

Yeah, one of the interesting models actually, I'm working on a, I'm working on a article about this as well, is that, I mean, just looking at the political breakdown, I think the people who care about this problem generally have to, in a ways to address it and they're kind of polar opposites. So you have one end, you have the people who just want to create new institutions that have scenario and on the other, uh, a group that wants to really, uh, use the hand of government in order to, um, in order to change how universities function and the incentives that are there.

And, uh, I think I've struck a middle ground. That is a very kind of generational difference. I'm actually on the, I'm barely a Zoomer, which is, which is just like shocking to, to think about it. But, um, I think generationally, if you grow up after, in, in the kind of like husk of these, like already malfunctioning institutions, you learn that there's kind of like thriving uh, sub circles. Right? So, um, of course, part of this is just from the, uh, informatics community. I know like maybe, uh, like a solid quarter of the MIT, math and computer science people. Uh, but also you have these kinds of networks that spring up where people are just like, oh, I'm going to recommend you to this person.

You guys have similar interests. Uh, you should probably speak and you have the springing up across every university. And I think the percent of people who are serious, uh, which are, which is like not, not a small amount at each of these universities, they can kind of find each other. So I'm actually not sure that either of those polls are correct.

And I think that I'm maybe more optimistic about this, but, uh, I want to see the kind of older perspective. Yeah. Well, but I think you're talking about grassroots. I think maybe what you just said to me is that there is a, I mean, obviously we live in a networked world, right? So you, you were able to reach out and get in touch with me and record this podcast within a week or something.

Right. So, so yes, I definitely believe that. Especially among people, young people who are say, thinking of tech, startups are building projects in the world, they are able, they are able to find other talented people and screened on talent, you know, based on, you know, not, of course not everybody, not everybody who good competes in these competitions.

There's plenty of probably more people who are good, who just never compete in the competitions, but, but still you have a way of finding. You know, talent and, and networking with other talented people. I don't deny that at all. I just, uh, my only point is that the way the institutions are working right now, they're quite suboptimal in terms of, um, being filters for talent.

Right. Um, but the question is, right. Like, do we really have to worry about that at all? Are we just looking at a future where these institutions are just replaced by these kinds of informal networks it's possible? So you could say it depends on what you're trying to accomplish, but, but yeah, I mean, if, if you posit that, you know, the tech economy, especially where it really matters, like in like early stage startups and things, they have good ways.

They, they know the score, they're not fooled. They, they, they have some folk notion of psychometrics. They have some way of filtering for talent or familiarity with this set of libraries or get hub productivity. You know, they have tools that they are using, they are using right now, right. To build, you know, I'm sure open AI has very talented people and they know how to find talented people.

Right. So yeah, they use these networks, they like act I've actively like heard of people who just got recruited by, uh, by something like this. Yeah, exactly. So, and then maybe you would say, okay, even these dumb old professors, like, you know, in the Harvard. Physics department or the Berkeley CS department.

They kind of know what they're doing too. And they're, they're actually doing okay, you know, insofar as they can avoid, you know, what's forced on them by the university for other, you know, ideological reasons. Other than that, they're largely able to screen for talent. I think, I think where it breaks down though, is when you, for example, like the university of California system requires you.

If you're, if you were applying for an assistant professorship at Berkeley, you would be forced to write a diversity statement. And, uh, that diversity statement better conform with the current ideology in the institution, which is in my terminology, kind of woke leftist. And if you write something at variance with that, you might be filtered out.

You might not make the short list, the interview list for that faculty job, even if you were the most qualified person and my friends who are professors in the university of California, not only, you know, vigorously, but only in private, because they're all scared object to that. They also object to the fact that when they go up for faculty promotions, they also have to write a similar statement.

Now, if you, if you're, if you're a social justice warrior in that system to you, this is totally reasonable. You say, look, uh, obviously the problems that we have are due to discrimination, you know, residual racism and discrimination, even among our faculty, uh, maybe it's even subconscious. Um, but by forcing people to write these statements and to spend a lot of time thinking about diversity and how they're going to improve things, we overall get a better educational experience and better long-term output from our system.

So that that's the counterargument. Um, I frankly don't think it's very productive and I think there are a lot of radicalized stem professors who just because of the current power system there, they're not able to say anything about it. So at that level, I think the system is not functioning very well, but the place which probably functioning the best is in Silicon valley startups and things like that.

Yeah. I think I want to take a slight detour now and ask you a kind of question of left field. Uh, have you heard of Amy Chua and her theory of market dominant minorities? Yes. I'm very familiar with Amy Chua. Um, Amy Chua's father was a pretty well known electrical engineering ECS professor at Berkeley.

And one of my older, one of my physics research collaborators is an old friend of his. So I know, I know all about him. He try even from when she was a baby, uh, she she's similar in age to me, but I, but I've heard all these stories about Amy as a kid from these, this other channel. So yes, I know all about Amy Chua.

Yeah. So, so she writes that you have this, uh, this model that's based on, uh, on, uh, non Western countries where you have these, uh, what are, what she calls market dominant minorities. Chinese in, uh, in Indonesia or, um, or, uh, well, historically this is just true. And of course it doesn't justify anything, uh, anything that's been done, but also, um, Jews in America, right.

They're just successful and that's a good thing. Right. And, uh, and there develops a kind of resentment towards them. Right. And, and she writes that this explains Trumpism, uh, because you have these kinds of cultural groups, not necessarily ethnic in the case of the United States, but you have these cultural groups who are like super progressive and live on the coasts.

But I actually think it's in the it's in the opposite direction. Uh, uh, my take on this is that progressivism is actually demagoguery against the self-made in general. And, um, as a kind of correlate Asians in specific. So you have a bunch of these kind of old, uh, kind of blue-blood, uh, academics and humanities folks who are having a fun time at their purchase stages universities.

And then as you said, you have the cold war, you have the stem revolution and become, we come without didn't completely change their status. Now, when people are thinking of academics, they're no longer thinking of, uh, the people who can like, uh, who can like recite Shakespeare or whatever. But they're thinking of, uh, they're probably thinking of like a physics professor, like Einstein, maybe they're thinking of these kind of completely different style of communicators and.

And that what this actually does is that it creates the same kind of resentment that, uh, that she was talking about. And particularly because it's new, it's a group of people who have just kind of just emerged on the scene and they might've been, as you said, people who were selected for psychometrics, people who didn't have that kind of cultural custom.

And now we take most of the grants. We get most of the respect, we make the most money. Uh, that's going to fuel a lot of envy. So I think if you understand it in that way, then you get like this, this inseparable connection between decline in between a real like war on these positive selection mechanisms and the ideology of progressivism itself.

Well, there's okay. There's a lot to unwrap there. Um, let me see, I think the status decline of humanities, and to some extent, social science, although not economics, economics is its own to be its own bubble in terms of, uh, prestige. Yeah, yeah, yeah. But in general, humanities and social science, uh, is kind of in prestige decline relative, you know, two decades ago, there are fewer and fewer public intellectuals that people pay attention to coming from those disciplines.

Uh, more and more people would be coming from the sciences or engineering or something that, that, that are sort of public figures, you know, AI researchers or something. Um, so I think directionally, everything you said is right now, what is the psychological effect on I'll just, just for fun, I'll say the word cell population.

Um, I was actually going to ask you about this later. Yeah. So what do you think we need to explain that to your listeners or not? I was going to put it in the intro. Okay. So in the wordcel population, which I would broaden out to not just be like humanities and social science professors on campus, but also like journos journalists who are they're generating, they're generating most of the bits, which clog our mass information channels right.

On, uh, on, on the internet. Right. So that class of people is yeah. Pretty aggrieved. Right. So being high V low modest M uh, that population of people is kind of, you know, seen their relative prestige decline. And I'm sure they're not happy about it. Um, and not, not just their prestige, but even their income levels.

Um, but let's see. So does that imply an animosity toward psychometrics or testing? I think that is a contributing factor, but I think a much larger contributing factor in my own mind is that if are not high M you cannot reason from statistical data and all statistical all psychometrics or outcome studies, or, you know, determining whether a particular noisy predictor is good or bad and how good or bad is it.

Those things are just off limits now for those people to really deeply understand. So if I say something like I could take a totally, I don't know if this is good for your audience, but I can take a totally different perspective and say at the NFL combine, they measure the 40 yard dash time of every draft pick every, every candidate for the draft and Scouts care a lot about it.

And fans talk about it. You can go online and watch the 40 yard dash. The runs actually at the combine. It's a statistical question. How important is the 40 yard dash time? How good is it as a noisy predictor for how good your running back is going to be in the NFL or wide receiver or defensive back or quarterback?

A wordcel can not address that question because they literally can't look at the data. They don't have any intuition. They don't know what correlation is. They don't, you know, AUC is that they just can't understand the concepts. So if I make a claim, which is psychologically difficult for them, which is that no, this kid who grew up in a laundromat and is an immigrant from China, but he scored really high on this test.

No, he scored really. That kid should be admitted to Stanford over your kid, even though your kid volunteered for 12 clubs while she was in high school, that's a very emotionally difficult claim for the word cell population to accept. And they can't evaluate it because it cannot understand that noisy predictor called the, uh, the, uh, I forgot what it's called now, what the national math test is called.

But, um, the, you know, the, the, uh, is it ASM? E w what is it, uh, anyway, whatever, whatever the test is now that you take, if you're, if you're good at high school math, um, you know, they just can't accept that they, they don't know how to evaluate whether that predictor is something that should be taken so seriously.

Like this is like getting number one in the spelling bee. Right. It doesn't matter why should, that could be admitted in front of my kid. My kid was captain of the, you know, um, build a school library and Ecuador for poor kids club, you know? Um, so they just don't have any basis for comparing these things.

And so it's not, it's understandable that they're hostile to the use of psychometrics that does that. Does that, does that, does that address your question at all? Yeah, I think it completely does. And I think it actually answers the question before, right? Like, why is it so hard to communicate these kinds of truths?

Uh, exactly something that I'm just thinking of. It's neither like ignorance as, and not having the information nor is it just like, nor is it simply denial, but it's actually like inability. Yes. It's literally a hole in their head and there's a blog post. I wrote a long time ago. There was a linguist at Penn who was talking about this tribe in the Amazon called I think they're called the piranha.

I don't know if I'm pronouncing that right, but they don't have words for numbers, like more than three or four. So they have 1, 2, 3, many, many, many, right. And they don't do any kind of quantitative trading with each other where like, I will give you 12 yams for one, uh, sharpen, spear, you know, they don't really do that.

So they don't have certain concepts and you know, it may seem tragic to us, right. That they don't have these concepts. It seems tragic to me that most university presidents or federal judges are totally innumerate. And can't judge statistical arguments. Yeah. That actually, this is one of the questions I wanted to ask is I'm not really familiar on the literature about this.

So actually this kind of takes like a weird segue into postmodernism because the first time I was reading like Derrida, um, who's a postmodernist who talks about the, kind of a way that language shapes, how people perceive concepts and perceive the things in front of them like physical realities even, um, What's this, this is not how things work at all.

It's this is not how, how I think about the world is not how most of my friends about thinking about the world, but like the more I understood the kind of like ethos of these journos or these kinds of like, um, these, these like wordcels, right. I, I realized that that's what he's describing and he, and when it comes to that, like he actually describes them fairly well.

Um, and so I think what's happening is that there's some bit of the population who actually does depend on the existence of these words and these concepts that are predefined for them. Uh, and some other group that's not. And I'm wondering if that's actually confirmed by the research. Well, I mean, so there's sort of two slightly different claims here.

So, so one is that, um, you know, the language and thought are deeply intertwined. And so, you know, you know, I think quite plausibly, if you don't have a specific word for a concept, it may not mean that I can't quickly update you and say, well, you have this word for three. And if you had one more banana, you would have four, let's call it four.

And then they, they have no problem, you know, uh, grasping that and then just updating and using it. Right. But it does suggest that the way their society is structured, they're not using that concept very much if they don't have a specific term for it, right there may or may not be a barrier to them acquiring the company.

But they're definitely not using it very much if, if you don't, uh, if they don't have a word for it. Right. So it's kind of like the distinction between knowledge and common knowledge, right? Knowledge is just something that's true. But, uh, common knowledge is like, you know, that everyone knows this. Right?

Right. So, I mean, I think the implication works this way. If you don't have a word for it, it probably does imply that you're not using the concept very much in your day-to-day life. However, it does not tell you one way or the other, how easy it would be for you to add that concept and use it effectively.

Okay. However, in the case of numeracy, in the case of understanding a normal distribution or hypothesis testing using statistics, or what is the proper normalization of that variable before I compare it to another variable, those things are not easy to learn. And I would guess no more than five, 10, 15% of the population, even if they worked very hard, would have any kind of facility with that statistical hypothesis testing, even to the point where they could just read the newspaper and decide whether, wow, is this a big GDP gain number?

Or is it actually, could this just be a fluctuation? Is there, you know, is that pattern, you know, something I should extrapolate, that's kind of basic reasoning, which, you know, some percentage of the population certainly can do and does all the time. I think that, you know, even in the best educational environment, it's still a very small fraction of the overall population.

That can become good at it, comfortable with it. And certainly the population of ward cells is not able to do it because they've, they they've almost all had decent educations. So, so right. You're not a word cell. You're not a journal university professor word per word cell without having been through a pretty decent education program.

So you were exposed to those concepts, but you just didn't retain them. And you probably can't because you're not wired properly. Where are you though? I don't think these people have to take stats classes. It's a good question. Like admissions. It's a good, yeah. Okay. So that's a good question. So can you get through, you know, can you get admitted to a prestigious university and then major in English or history and never have to confront, uh, these ideas?

Um, maybe, maybe, maybe that maybe a bunch of words cells could pick it up if they wanted to, if they were forced to. And they just, it's a, it's a problem with the education system. Um, but I, I, I suspect they have been exposed to these, you know, because if you, if you read like the economist or the business page, you see terminology, right.

Like, uh, average or, uh, you know, uh, trend line or slope, or, you know, you, you see things and, and I think, and you, you, you must've had some math in high school. Right. And so I think if you have any. Basic aptitude for these things. You, you would, you would understand them, right. By the way, when I, when I say someone's a word, so I don't mean that the job title means their works were sell.

So my, my, my friend, who's a corporate lawyer in Silicon valley. You know, it might tell you if you're a corporate lawyer in Silicon valley, you're automatically word cell by, by category. But no, I mean, he is mathematically literate, you know, even though he majored in political science. So I'm not, I'm defining the category based on the cognitive profile, not based on the job title, but it is, but it is overwhelmingly true that most journalists are in that word.

So cognitive category, some subset are not, but many of them are, uh, most of them I would say. And to me that suggests that even if they really tried, they would not get good at understanding statistical hypothesis testing. Yeah. I think when you lower this a little bit, I think like normal understanding, like a normal distribution, like most people are not exposed to that, but even when it comes to like basically logical fallacies or like availability bias, right.

If you look at like one kind of police shooting and you extrapolate that there there's this kind of like, uh, massive amounts of police shooting, like there's this whole by skeptic magazine, uh, uh, that found like, uh, I don't remember, but some extremely large percentage of the population thought that they were like more than a 1000 in some more than 10,000, uh, African-American men shot every year.

And this is just like an absurd over estimate. The actual number is something like 19 or 20. Yeah, exactly. So, so backing off from something like hypothesis testing, but, and that way, I just want to put a caveat that this is not like a left wing thing. Like you see the same thing on the right with like terrorism or with like immigrant comes and commits a crime rate.

Those are also like, not indicative whatsoever of those, uh, kind of, uh, actual statistical probabilities, but they get tons of coverage and there's a freak out on the right as well. And it's, it's the exact same thing. Sorry. Gone. No, yes, exactly what you said. So, so just having some good number, intuition for magnitudes, let alone actually knowing what a normal function, you know, the normal Gaussian function looks like.

I'm not talking about anything that high falutin, I just, just having senses of numbers and the meanings of averages and things like this. Uh, I still think that's a relatively small part of the population. That's good at it. Yeah. I think actually this ties back to a question that I had and that I wrote down if I was listening to you speak with Richard Hanania, which is that you guys, or like, I think this is mostly him, but I think you kind of agreed that it's kind of easy to predict or to model like someone who is, less intelligent or has like less, um, less like math ability. And I actually really disagree with this. I think that like, without the, without the kind of psychological data, without the studies that have been published, um, I would never, in my entire life have a good understanding of like people who don't have, um, who don't have like at least a bit of math ability.

Sorry, sorry to interrupt you. But I think you're making a very good point. So I know many people who are strong at math who would have a very tough time passing the kind of Turing test where they pretend to be someone who's not good at math. Okay. So, so you're right. They may be there. They might be a little bit on the aspy side and generally their theory of mind for average or below average people is very poor.

Okay. Um, now if that kid were a little more shifted toward normal on the aspy spectrum, so they're a little more normal in terms of personality and affect and stuff like this. And they grew up with ordinary people like, like growing up with a retarded, you know, with a intellectually disabled next door neighbor to the same age, or maybe, you know, their mom was very like non mathematics.

I think for that category of people. I think Richard and I are more in that category. We can easily imagine how real people view the world, but plenty of super smart people don't are not really very good at figuring out how ordinary people view the world. So it's not as simple. There's not a simple answer to that question, I think.

And I'll try to find something to put in the show notes for this as well. Uh, so I'm not sure how strong this claim is. I think I just heard it somewhere, but there are like management theory papers that let's say that this is just statistically true as well. Right. People with more than I think two standard deviations have trouble managing, uh, have, have higher trouble managing on average.

Yes. You know, on the other hand, you can have a guy like bill Clinton, who probably is actually, maybe he's definitely a word sell, but he's, he's very smart in a certain sense, but he has the common touch, right? He, he does have a good sense of how ordinary people think, and he can make himself liked by very ordinary people that are more than two standard deviations below him on some test.

Right. So, but I agree as a statistical observation, it is true that if you, if you randomly select a person to be the leader and the people he's leading are more than two standard deviations lower than that person on some psychometric scale, there could be significant communication problems. Right. So it's a statistical observation, I think.

I don't know because, or I agree with, I agree with this point, uh, I guess I'm just having like, uh, even just the Clinton point, that's just one example. Right? I think that this is actually just like a incredibly widespread problem that this is actually just, I don't know, actually, because now that I think about it now that I think about it, this is also just based on like availability bias.

I kind of need to catch myself here because I know a lot of people who are kind of like high math and are just like, I tell them about these psychology papers and they kind of have exactly the same reaction as me. Right. Which is, um, this doesn't just sound like a mental illness. This sounds like this sounds like science fiction.

Yeah. Is there any good research on like how, how good high math people or high IQ people are at, at like guessing what preferences are? What kind of like expression, uh, lower math people would like generate? Is there any kind of research on that? I don't know of specific research on that. Um, I do want to, this is a little bit tangential to the topic, but it might, it might be interesting when, when people become very good at, especially the kind of algorithmic thinking that, you know, you might, uh, might be very useful in informatics competitions and things.

It's often because they're interested in that they're interested in very rigid rules based processing of information. And those people on average, I think are less good at simulating complicated, fuzzy headed humans, right there. They're a little more spotlight and a little less empathetic. Like, I, I, I like, I think oftentimes, like I know math, high math type people who they prefer analysis or some kind of, you know, more, I don't want to say continuous versus discreet, but some kind of more, um, less rigid kinds of mathematics, um, or mathematical reasoning.

And often they're a little better at thinking about how normal people are. Uh, I don't, I don't think, I guess I'm not expressing this very well, but, but the, the special sub-category that really liked programming and like algorithmic type thinking, I think are the, actually the least likely statistically to have really good insight into like what ordinary people think.

It's very interesting because like I'm on the far and accommodated toric spectrum. So, well, let me give you, let me give you an example, like Einstein, uh, was not particularly an algorithmic type thinker, but he was, he was very high math and also very high verbal, but he had pretty good insight into. And, uh, you know, cause like a lot of times when he's quoted, he's quoted for some very Horry, uh, philosophical type life philosophy kind of statement.

Right. Which is not, not some, not some very pure mathematical type thing, but it's a completely different kind of wisdom. And um, so that kind of person is very different from the kind of guy that I would typically, you know, recruit for a startup who want a coding competition and it's just a 10 X or a hundred X coder.

Right. Um, it's a different profile. It might be high V high M versus a high M moderate V you know, I don't know. But, um, there is a difference. I don't know, actually I think at this point I should probably just resign on this, uh, on this topic because I don't have anything more other than pure speculation, but the questions I have many questions, I think they're good questions.

I think, I think it would be great to see psychological studies of this type. Yeah. Actually the other, I guess I should mention this, even though it is just an anecdote, uh, I'm not sure if this is a thing with like Asian communities specifically, but you have these kinds of stereotypes about like, oh, you have your nerdy kids, you have your jocks, you have your like music prodigies, uh, at, in like high school.

Right. Um, at my high school, those were all the same. Mostly the same people. It wasn't like number one, like, like people who are like best at sports were Allston, like number one, escorts, number one at, um, academics, number one at music at the same time. But for example, I was like, uh, I was, I think unquestionably top math student, top computer science student, but I was also, um, decent, or like, I would say like on definitely like the top end of, uh, of cross-country running, uh, top end, uh, made it into several, like audition only like jazz bands, that kind of thing.

And it was the same pattern with basically everyone at the school. Right. So the same people who are good at I'm good at academics were good at, um, pretty much like everything right. Reasonably good at even if not, if not the best at it. And I think the same might be true with like social, with like social things.

Right. I, I'm not sure if the stereotype of, um, of these kinds of trade-offs is actually true. I don't think there are necessarily trade offs. However, I, and, and this observation, you know, the reason they call it, the G factor is that it is very, it is a general factor that affects lots and lots of different activities you might be involved in also in high school, there's a kind of achievement factor.

Like some kids care about being achievers and some don't, but if you care about being achiever and especially in today's environment where the. Application process is so horrible, then you're likely to be in like, so if you have the achievement oriented, uh, capability, which probably just means that you're hardworking and you care about achieving and then you have high G yeah.

You're probably going to be good at a whole bunch of different things. Um, I might draw the line at real sports. So really hardcore, I'm not dissing cross-country, but I mean, uh, even cross country at the highest level, it's pretty hard wired, but, um, but you show me some good running backs and a hundred meter sprinters, or, you know, 50 meter freestyle sprinters.

And I'll show you a bunch of guys who are on average, not that intellectual. Okay. So, or guys who can bench press three 50, you know, uh, I spent a lot of time with these guys and, uh, it is not the same population as the ones who make the jazz band, at least in my day. Now maybe it's, it's a super peak and this is true that like I like state or like nationals.

That's probably true. Yeah. Maybe I think even if you're a starting running back or point guard at a good high school team, you're a pretty damn good athlete. And most of those guys are not also good at, so I think my high school was like, yeah, I'm not talking about schools where like, they don't care about the football team.

I'm talking about go down to Texas. And our Alabama and see who's the starting running back on the team. And that is not the same kid, usually who's in the math club, but I mean, I think that, I think at like, high-end, it's just like a correlation problem, right. It's just like an or like not, or like a non-correlation unlikely that yeah.

So the general factor of intelligence, it loads, it loads more on the ability to succeed as a pianist or in various clubs than it does on being able to run the ball. Okay. So that is almost no overlap there, except that if you're a little smarter and analytical, you can train a little bit better. Like you can, your off season training is a little smarter and your ability to like, do what the coach tells.

You might be a little bit better, but it's totally overwhelmed by whether you can bench three 50, whether you're over 200 pounds, whether you have muscle, whether you have fast Twitch, you know, uh, explosiveness, those are pretty much totally de correlated from G as far as I can tell.

Yeah. I think I want to do a segue now to these kinds of, I think, uh, free speech or, uh, or, um, politics in academia. Although I already talked about that, uh, before, so, I mean, you've, you've kind of spoken about this, but, but how much is there a problem of self-censorship or consequences for political views and.

Uh, I think it's huge. I think, uh, the technical term now has become preference falsification.

Yeah. I love Timur Kuran. I want, I want to get him on the show. He was the person who coined the term for the audience.

Um, after my, uh, you know, resignation, as vice-president for research, I was involved with an organization called the, I think it's called the AFA academic freedom Alliance and Timur and I were both involved in this.

So we've had many discussions about preference falsification and whether it's real or not. And this organization AFA the core group that created it are called the Princeton 20. So it's about 20 professors at Princeton who were, you know, the core group that got it started. And then there a bunch of other people like myself and Timur and others who were involved too.

And, um, so we've talked a lot about preference falsification, you know, among faculty in the United States. And we all agree. It's a huge problem. And in my own, again, like this could be by a sampling, but among people I know, well, there's pretty much no disagreement. Like, like even like, I know I have, you know, university of Oregon where I was a professor for a long time.

That's a very progressive leftist kind of campus. I mean, you can't get much more, you know, okay. Maybe Santa Cruz or, or Berkeley can rival it for leftiness and eco hippiness. But, but there are there, I don't think there's any school to the left. And, um, so I have plenty of friends who are on the faculty there still, and they're all very left of center, very left of center.

I mean, I was modestly left of center, but they're, they're very left of center and they even, they would agree with me that you have to very carefully control what you say. Otherwise, you're going to get, you know, attacked and nobody wants to get attacked. Everybody. Who's a serious scholar is busy doing their serious scholar thing.

So they don't want to be attacked over some opinion that, you know, uh, they slip out that slips out during class. So our, every professor's very, very scared about this stuff right now. Yeah. And I think this is actually not even like the problem itself, but there's kind of a meta problem that I think is incredibly threatening, which is just the lack of understanding.

I mean, this is actually really close to what we were talking about before, but the lack of understanding of the powerful that these people, these like individuals, these like very high performance people, like it doesn't matter what they say politically, if they're, if they're helping you kind of like, um, uh, work on a new physics breakthrough, or like even more direct, like a medicine breakthrough, right?

You can't just fire these people. You're sacrificing so much every time you do that. But I think there's just this not, there's just no understanding that these kinds of people are, are not replacing. Yeah, so that's a good point. So, so during the cold war, and this is also a different, very cultural, very different time, culturally, like, you know, if you watch madman or something, you can see it.

Yeah. There was a lot more tolerance of the irascible, politically incorrect genius guy. And they would just say, let this guy be, you know, he's, he's doing great stuff, you know, in signal processing or whatever, but you know, he, of course he doesn't have the politically correct views, et cetera, et cetera.

Now it's much more of a problem. Like that guy is going to be a, a big issue. And if he says the wrong thing, if he has like, Tourette's like syndrome and blurts out politically correct things every now and then he's going to get canceled and it's going to be a huge headache for the department head to deal with it.

And the Dean and, uh, you know, the vice president and the president, you know, all, all these things. So it's, it's, it's a very different time and there's little tolerance for that. And, um, you know, you could say if people were more aware of the power law distribution of contributions toward human advancement, that would make them more tolerant of these, um, Excentra geniuses.

And I agree with that, but people a don't understand the power of distribution of human contributions and B even if they did, they, plenty of people don't care. This is the kind of socialistic mentality. Like we don't care about growing the pie. We care about whether we're comfortable in the pie. I think it's actually worse.

Right. So, uh, I'm pretty kind of immersed in reading, uh, sociology, because I actually have, like, I don't I'm, I'll say it, I'll say it. I have a kind of like deep respect, especially for the old thinkers, like especially math favor. I'm always just like talking about facts, vapor. Um, but I think there's a strain in the left end and this isn't just sociology, but of left wing in general, that is just anti doing things.

Well, like, uh, Brian Kaplan has, this, has this like catchphrase, right? The left is anti market. The right is anti leftist. But I actually, I don't think it's just markets like they, I think, um, and, and of course I should give the caveat that this applies to all left wing people. I consider myself like kind of center left, like vaccine relaxed, right in the fight and kind of guy.

Um, but, uh, there is the strain and it's particularly strong in these kinds of academia circles where I think they just don't like it when you have a kind of compass that you're optimizing to. And here's why I think that's the chase. It's going back to the sociology stuff. Uh, there's kind of choose theories of change, two competing theories of change, which is essentially like innovation versus.

Uh, collective decision-making right. And actually the best illustration of this that I've ever seen is in, uh, Lucian's book. Uh, if he has the three body problem short Trelegy in the third book, uh, deaths. And, uh, I'll, I'll just give a spoiler or let's say at the end of the second book. Okay. Um, this guy who's just been like kind of a sign for being like a notable, uh, noted kind of notable person.

He's just been assigned by kind of infinite resources to come up with something to kind of like defeat these aliens. He comes up with a solution just like random guy. And then in the third book, uh, they elect someone who is essentially like the person who has the power over the system that, uh, the old guy invented the old guys kind of like at this point he was like, oh, there's a time skip.

He's old and he's retiring. And they elect a new person. And this person is kind of like chosen for, for like social purposes for being like very appealing. It's a, it's like a vote. And then, um, uh, so she takes over and then, uh, the aliens immediately invaded and, uh, they fail. There's this kind of like huge collapse that happens because of this.

And so this is the kind of like illustrate beautiful illustration, really like, please read the books. I I'm sure you probably have, like, this is referring to the audience, um, where, uh, there's these two kinds of processes. There's social change and then there's technological change. And I think what is truly emphasized in these kinds of left wing, left wing circles is that, uh, is that social change is the fundamental driver of things that happen.

That it's the kind of, um, it's the kind of principle at the heart of everything, right? It's almost like this conspiracy theory. Well, I don't want to go that far. There are things that are actual conspiracy theories and these kinds of academic circles as well, but this kind of like super, all encompassing narrative about social change and, uh, my prediction as to why there's this kind of detesting of the market.

Or I see right now, like in, in just my ordinary life, this detesting of machine learning is that they just don't like, like it when like the other things pointed out. When we actually look at like the long arc of innovation, it's not actually people who are fighting these kinds of like moral fights or people who are rallying these kind of political forces.

In fact, it's these kind of like random, like, like quite frankly, like kind of on the spectrum people, uh, coming up with these, like coming up with these things from out of left field, that's actually creating social change. Well, let me, let me give you my take on this. Um, and you know, it's funny because, well, okay, so you can, you can talk about two more or less Encore.

They're not completely uncorrelated, but two somewhat independent things you might be working on society. What is the internal structure of the society? And that might be maybe primarily redistribution. So making things fair, making the least well-off person much better off, right. And you might say, that's my number one priority.

That that's what we should all be focused on. Okay. Another view is I'm less worried about the internal structure of society or redistribution. I'm just more worried about increasing our capabilities as a civilization. So pushing science and technology forward, growing the economy, you know, productivity, and there's no obvious answer how you should trade those things off against each other, or even whether they need to be traded off against each other.

So those are very strong, uh, empirical questions, right? You have to decide, are they at odds with each other, maybe a richer society can just afford to be more fair. Oh. But no, but maybe the richer societies for some reason are more unstable for the hyper rich to take over and steal even more from the poor.

Right. So there are, there are all kinds of key structural assumptions you have to make about how things work and how things work is going to be different in an era of early industrialization versus a, uh, postmodern post postmodern era with internet and social networks. So, you know, there's a very, very complicated discussion that we can have about all these things.

Um, one of the, just to tell you an anecdote, um, I was good friends with a philosopher named Bob Nozick. Who wrote a book called anarchy, state and utopia. And if you're kind of a libertarian guy, you would be familiar like libertarians love this guy knows it. Um, and I used to talk to him about Rawls, probably you're familiar with Rawls, right?

So the veil of ignorance and the way the veil of scathing a kind of a half rant, a half article about rawls. Okay. So, well, it seems to be largely accepted by good people today. I can put little ironic quotes around good people that the Rawls is veil of ignorance is a really deep and important concept.

And the, so this idea that if you didn't know where you're going to be born in society, what kind of society would you like to have? Right. And so then the conclusion would be, well, since you're going to be randomly born in society, you want to have a very egalitarian society, right? Because you might wait.

No, hold on. This is at the center of my critique of rawls. Is that that's actually not what it says. It says that you want, like, you want the variance to be low, not like the expected value. Sure. Yes, exactly. Right. So, right. So even that arguments, like why, why do you, like, how much are you weighting like low probability outcomes?

Very well anyway. Okay. But let's suppose Rawls Rawls statement is you want the variance to be. So I, I said egalitarian, which I think is kind of like low variance, but that's fair. So you said kind of, like you said, like born at random, I don't actually think that's kind of like what the model is for an born out like worst.

Uh, okay. I, okay. I, my formulation of it was which might not be Rawls's original one. It's been a long time since I looked at Rawls, but, but, and this conversation I had with Nozick, we're talking about the early nineties. Okay. So this is all old stuff, but, but, um, my, my caricature of Rawls is you're going to be randomly placed in society when you're born.

Your soul is inserted randomly into a baby born on that day. And, um, you want low variance or highly egalitarian society. Cause you're, you can't control which baby you're going to end up in. Right. And somehow, maybe you don't really, really don't. You really don't like the worst case outcome. Right? So, um, obviously there's some implication about your utility function here and whether it's symmetric and stuff like this, but anyway, let's just accept all that.

Okay. The point that I made to Nozick, which I was kind of shocked that he found this novel, and then we talked about it for a long time is I said, this is stupid because not only should we randomize over which baby you're born into your, your soul is inserted into, we should randomize over what time in society, the 21st century, the 22nd century, the 23rd century, you're, I'm going to insert yourself.

Into the body, because you can imagine a very egalitarian society that is not making much progress in technology and, uh, productivity. And then if I integrate over future well-being of people in that society, it really sucks. So my version of Rawls would say like, w we should randomize not just over which baby you're going to be born in, but when in the society you're going to be born.

And then that would put some weight on productivity and scientific advancement, et cetera. And I was kind of shocked that Rawls had not considered that. And I was asking Nozick about this. And he was like, well, it just turns out Ross just like, didn't understand technological change. And, um, didn't think about it.

So, but it goes back to that trade-off I was saying like internal optimizations about redistribution of wealth are potentially independent of questions of, you know, uh, what is the growth rate or what is the, uh, the, uh, scientific advancement rate? I mean, in practice they're, they're deeply, deeply dependent, right.

Are deeply, deeply connected then almost always as a trade-off. Is it? Well, I don't know. I mean, I think if you get really advanced and affluent, so you have like, you know, the star Trek food synthesizers built into the wall, maybe you can be very a gala. I mean, that's the star Trek that the Federation universe is very egalitarian.

Yeah, but I'm talking about in practice, right? In practice. We like, we either redistribute where we do, like, we do like overregulation of technology or we don't, well, I guess what I was saying is that as we get Richard, does the Gini coefficient have to go one way or the other, and it's not obvious to me that it does.

So it seems like our we're about as unequal now as we were, when we were much less developed. So I that's a, that's a, that's a economic history question, but I believe there's some data that says like, yeah, the Chico fish doesn't actually change that much over time. I mean, you have certain periods where Robert Barron's grab up lots of the stuff, but overall there doesn't seem to be an obvious trade off between, Hey, we're all getting richer.

So we must be getting more on equal. Maybe the other way you could say, we must allow for some inequality in order to continue to getting to get richer. But the, the idea that, because we got richer suddenly that enables the rich people to grab even more of it. I don't know if that's how well supported that is.

Actually, there's a very interesting question here. I think, which is about is inequality like growing. And one of my, one of my, um, uh, chapters, I guess, one of the things that I've been really thinking and writing about in private. So this is not published is that there is that freedom, variance and inequality are actually all synonyms, in practice at the very least, if not literally. Uh, and what I mean by that is you have these societies where the power law is eating everything. So you have dating markets. The most attractive people are getting much, much more of the data and it's a power law. Uh, Rob Henderson has excellent documentation of this.

You have the same thing, obviously in economics, this is where the power law got started. You have the same thing with attention, social media followings, that power law is actually even steeper than the one in, uh, than the one in economics. So any time you get these kind of, um, massive hyper-connected incredibly free, right?

You can talk to anyone, you can interact with anyone, you can do anything. You get the situation where, um, where the limiter becomes, these kinds of biological realities, or sometimes becomes like randomness, right. Um, and you get these extremely strong power laws. So the question is right, is, is there really like a way to separate kind of personal freedom from inequality?

Well, you know, it's funny that you, you, you, you say you're riding with Biden and stuff, but what you just said is basically like the libertarian. Thesis right. That, that we should, you know, they would say we should allow personal freedom. That's just a, a kind of Axiom for them. That seems like it seems reasonable, right?

We should allow personal freedom as long as you're not harming somebody else. But then a necessary consequence of that is some people are gonna make the wrong choices and end up at the bottom. And some who are going to make great choices and end up as billionaires. And you're going to get this power law distribution.

So I think it's true. What you just said. Uh, maybe we shouldn't allow it. Maybe we should control people. So we don't get such an equal outcomes. Maybe we should ban Tinder tomorrow. I think it's libertarian. If you, if you like intrinsically value freedom as your kind of highest thing, I am not sure I'm I am like fully there.

I think freedom is incredibly important, certainly, but especially like compared to everything else that could possibly happen in the world. I don't think. Yeah, but what I was saying about you is just, I, it was slightly unfair. I was just joking, but, but the, the analysis is correct. What you, I agree with your analysis, that if you allow people to be free and they're highly networked and can exchange stuff, you probably are going to get these power law inequalities.

Right? The reason I just said it sounded familiar like a libertarian mantra is that they, they take as their primary Axiom that we should have freedom. And then they then say, well, and then of course, inequalities and escapable. So quit complaining about. So that that's kind of there. Yeah. And I think this is not an unnatural reaction for you to have, I know you said it as a joke, but I think a lot of people, this is actually especially happening on Twitter.

I just pointed out a thing that's true. Not necessarily endorsing it. Yes. You are value neutral. But the context in which that analysis usually arises as some libertarian guy talking about. So I mean, like, I, I have to be honest here, right? Like I first kind of stumbled upon this by reading Peter Thiel.

I see. Yeah. There you go. Well, he's not a dumb guy, Peter, you might not agree with everything he says, but he's not a dumb guy. Yeah. I think this observation is, I mean, like you can just look at graphs, right. It's just like, obviously true. Yeah. Yeah. No, I mean, I would love to, I mean, I'd love to do a whole podcast on someone who's an expert on like dating markets for younger people. Cause I missed all this stuff and um, you know, it just seems obvious to me that like the, the super good looking guys are gonna get wild booty calls and just they're living in heaven right now. And uh, all the other guys are getting screwed. So, uh, it just seems like that's true.

Although I'm too old to have experienced. Yeah, I definitely recommend talking to Rob Henderson. Yeah. I've been trying to book him for this podcast, but he's been super busy recently, maybe in the, maybe in the future, I've heard him talk about this on other people's podcasts. And even he did actually dealt, like, I think at a T at the time when he was at Yale, he, he did talk a little bit about when he turned on his dating app and then like expanded the air, the geographical circle to include the working-class parts of new Haven, not just the campus and what he found.

So, yeah, he, uh, yeah, he, he would be a good guy to talk to about it. I maybe an OkCupid guy would be good too. Like one of the people from the actual company or people who really, I don't know how much data Rob actually has access to, like, does he have access to any of platform data or not? Hi, I don't know.

Um, so those guys, those guys know a lot. Okay. If you, if they look in, if you're a data scientist at Tinder, because cause they can see, they can probably tell when the, when a hookup happened. Right. And they can, they can do analysis that, you know, obviously academic, social scientists are such a sorry group, but, but stuff, those guys couldn't even dream of.

But, but even like, if you were a good data, like a good data scientist in industry, but you were not at Tinder, they could do analysis that you can only dream up. Right. If you think about it, it's just the data set, right? Like. But this is a data set. We don't have the data, but this is data set. We really care.

We really care about like, I would, I'm kind of interested to know, like what, what are these Brooklyn hipsters doing? Like who's hooking up with whom and, and, uh, if it's all happening on the platform, that it's pretty interesting. Right? It's like the, as in terms of a human laboratory, uh, social science laboratory, it's an unrivaled just, just to riff on that, like, uh, before your time, there was a thing called PUA pick up culture that guys like, basically trying to figure out what the best way to pick women up was the before, before the apps became dominant.

So like you would go into a bar, like when I was young person, you'd have to go into a bar or a party and pick people up. There were, these movements run basically kind of run by like algorithm driven nerds who were like trying to study what are the best ways to pick up women. And, um, there was a TV show that was on briefly where they had wired up an entire bar in west LA.

They would just monitor all the things that were happening. And then afterwards they would have to get a release form from the people to use the video and the audio. But this was real in the fields, sociology slash anthropology that an academic could never dream of doing. And some of these left wing, social yacht, sociologists that I knew at Oregon studied dating.

And I kept pointing them as saying like, you know, you've got to watch the show because this show is doing what you could only dream of, which is like get thousands and thousands of data points and record them on video and audio. And then, you know, anyway, so, um, Um, I'm quite interested in that. Yeah, I actually, this is something that I've been puzzling over about myself, but I want to hear your answer first.

So you have this kind of very mathematical background, theoretical physics. Um, what made you interested in doing this at all? In like looking at social problems, starting a podcast, that kind of stuff well interested interest in social problems like that, that lengthy autobiographical thing I gave you at the beginning of our conversation was sort of pointing out that I'm a little unusual.

Like I got into this interest in psychometrics for, because of those idiosyncratic reasons that I described. Um, because I was, I grew up in a kind of normal Midwestern high school environment. Then, uh, I was on the sports teams. I knew a lot of athletes and stuff just got a pretty broad exposure to everything that was going on in society.

Um, not maybe the way that a typical mathlete would, would experience high school. Like I experienced high school, like all those eighties movies about high schools. That's that was my high school experience. Um, and, uh, so I've always been

can you, give an example, sorry. Part of being a Zoomer needs that they have watched exactly zero of these movies.

Oh, you're kidding. So you've never watched like, um, uh, risky business with Tom cruise or fast times at Ridgemont high or, oh my God. You guys got to watch these movies. You're missing out. Um, I'll tell you a funny story. There was a tradition for senior pranks at my high school, and a lot of the best senior pranks were done by, uh, in our case guys on the swim team, because there were a lot of, we had a very tight group of guys.

They were pretty high to use the modern lingo, high T, high testosterone guys. And, and some of them were pretty clever, like actually academically, not all of them, but some of them were actually academically, so we could plan and carry out pretty unbelievable stuff. So when I was a junior, one of the pranks that they pulled off was they went into the library.

Uh, I think at night they must've had a key, like my brother actually had a set of keys to the whole school. And so did another guy that we knew, um, with don't ask me how they got, well, I guess I can say they got them because, because the, like the gym, my brother was dating a gymnast and the gymnastics coach left them in the locker room and he swiped them or something.

So anyways, so, uh, so, um, they, they had keys, so they went in like one of the, one of the pranks they pulled as they went into the, and the principal every year in our high school would lecture us about how pranks are a tradition at Ames high school, but, uh, destructive, pranks that destroy school property.

That is, that is just forbidden. Right. And, and, you know, so, okay. So we are juniors. Uh, the one, the senior class did the class just ahead of ours was they went in, they took the disassembled, all the chairs and tables in the library and just kept the ball. And the wing nuts and stuff. And so kids came in one school in the morning and it's like, no one could sit anywhere in the library.

There was no furniture, everything was just taken apart. And then the, of course the principal's like, God dammit what the, you know, like that's not what I'm talking about. That was a highly destructive school prank. And we're like, well, we'll return the bolts to you. Like, of course they didn't know it was us.

Right. So we're like communicating them through like notes left in locker, like in the, like in the principal's office. And stuff's like slid under his door. So we're like, no, we didn't destroy anything. We're going to return the bolts. And then, uh, so I was recruited to write a very difficult math formula that they had to solve this math problem so that they could get the locker number, where we had this huge bag of bolts that we returned to them.

And so like they, and we knew they had to go to the calculus teacher to get, find someone on the faculty who could solve the math problem so that they could figure out which locker we had put the bolts in, so they could re assemble all the furniture. So that was our junior year, senior year. Of course we had to top that.

So, so my brother and I, my brother and I were in the same high school class, he's a year older than I am. But, um, so we, our plan was the following. We were gonna steal all the lockers out of an entire wing of the high school. And so during the day we would sneak around and, and the, the lockers were bolted into the concrete wall of the school.

So we had to go in and loosen them and just remove those bolts during the day. So that was operation during the day. And then. We came into the school with skateboards and these big guys basically pulled the lockers out of the, you know, like it would be a, I don't know, a 10 foot section of walkers with everybody's stuff in them pulled off the wall.

Cause we had unbolted them during the day and then, um, and put them on the skateboards and we slid them out into the back behind the school was like a Prairie. So we made like a whole row of lockers behind the school. And um, so kids got to school that morning and their lockers were gone. All their stuff was gone, it was pandemonium.

And uh, and again, they called an assembly. This is the next year. The principal's like, this is not what we're talking about. This is destructive, you know? And we're like, what, what got destroyed? We'll tell you where the lockers are there behind the school. And here are the bolts we took off the wall. So anyway, that's like eighties, uh, I can't, I can't, maybe kids still do stuff like that.

I don't think they do, but we did and got away with it. So anyway, that I don't think I would know either way. My high school was kind of like a magnet school and it was all like, it was all like immigrants said, you know, like, you know, like the 9, 9, 6, or like, I guess I should explain it for the audience.

So the 9, 9, 6 is nine, 9:00 AM to 9:00 PM, six days a week. It's the kind of, uh, it's the kind of acronym or symbol for the kind of Chinese work ethic or like, this is an idea that is popular. That's what I mean. Uh, I went to a high school in Canada called Woolburn collegiate Institute. It's famous for having like one quarter of all of the, uh, international Olympiad and informatics, uh, candidates for Canada.

Wow. So, so you guys actually, did you explicitly, uh, verbalized to each other that we're living in a nine, nine S like our high school, I think is 9, 9, 6. It wasn't every one for sure. But it was like, it was just like culture. I don't think at 9, 9, 6 was as much of a meme. Uh, back then, it was also only like, I think, like 30, 40% Chinese.

Uh, there were also a lot of Asians who I think actually had the same kind of culture, but it was just, wasn't like as much of a, as much of like an, I mean for the 9, 9 16, but that was definitely like how he worked. I think, see my high school, my high school, we had gear heads, we had, you know, weed heads.

We had jocks, we had, you know, a small number of academically focused students. And, you know, it was a, it was a pretty good introduction to the sociology of middle America. You know, um, it's slightly, slightly, uh, shifted in a certain direction because a lot of the kids were children or professors, but, but still overall, we had every segment of society represented.

I knew kids in high school who took steroids, who were on like the football team and the swim team and, you know, weightlifting and stuff, you know, just very broad experience. I think, which made me interested in broader society and stuff like that. Hmm. That's interesting because I look at my path and it's like, it's completely different, but I think listening to it, listening to you speak, I kind of see like a kindred spirit.

Right. Um, the way I came to it is, um, my mother kind of had a very big interest in like Sherlock Holmes when, uh, or like before, and also during my, my childhood. And, uh, she was also incredibly intelligent, uh, and, and very, very strict. And, uh, I was, I was like good at math competitions and stuff like that from a very young age.

But actually I was like very, very bad at trying to paying attention to particulars and making these detailed observations. And now I think like my bio on like sub stack or whatever is like observer of all things, because I've kind of just picked up this habits, but like not picked up kind of how to like, um, taught into me where, um, she kind of instilled this importance of just keeping track of everything, making sure that you don't go on autopilot when you're, when you're speaking with people when you're engaging in even ordinary life, because there's a lot of valuable things you can, you can improve on.

Like you can have constant improve. And I think that those kinds of observations, which I like eventually, like slowly, slowly, um, developed a system and an intuition for that, that was the kind of thing that led me to, uh, really being interested in these social problems as well. I think for me, I I'm in a different total different era.

So the Vietnam war was ending when I was a little kid and the sixties, there was a kind of hangover from the sixties, lots of hippies and radicalism, and then the eighties became much more corporatized. And so I was very conscious of the changing times and the cheat, just changing attitudes in society towards all kinds of things.

So I became pretty intensely interested in monitoring what was happening to say America, in this case, of course, there are much more dramatic changes happening, say in China that I was interested in or other places, but definitely just in American society. Like what, like, if you look on my blog, there's a label called American society, which is just blog posts about what little things I glean about what's happening in American society over time.

So I think I've been interested in from, from the beginning. Um, yeah, I think I clicked on your blog. Like mainly looking for some of the IQ stuff and also some of the, uh, uh, some of the genomic stuff. Uh, that was what I was really going for. But I think those observations, like I know it kind of inspired me to like, do more kind of casual posting because I always.

I always think like the value of people's work, um, or I shouldn't say peoples, but like of my own work really is like very, very narrow. And the, and the more like specific, the more like constantly like useful it is to, to people in a specific audience the better. But I think like the more I've actually like talked to people and talk to, uh, talk to my subscribers and such, uh, people, people kind of like the life stuff.

And it's also not like useless either, right? It's like, yeah, you never know what insights that you've gleaned from your unique life trajectory will be useful to somebody else. And, um, you know, my blog is all over the place. So like there were certain channels that you discovered, but, you know, there's a whole set of channels about like Brazilian jujitsu and ultimate fighting.

And there are whole channels about how to make about walls, like finance theory and wall street and the credit crisis in 2008. Um, one of the things I really like about blogging is that you can just be free and, you know, I, I view myself a little bit, like, you know, in miniature like Tom Wolf, I don't know if you're familiar with Tom Wolf, the journalist.

Yeah. Yeah. The re kind of realist. He was one of the first kind of, uh, I forgot the term, like a realist journalist, but, um, he wrote like a bonfire of the vanities and, um, uh, It's very, very well known for reporting on what was just really what was happening in America over time, uh, over, you know, I would say six seventies, eighties, nineties.

So, uh, is this the future of social science is a like, really, I do think like this, this is kind of like a, I think if you just say this without context and like, I want to provide the context that I'm constantly reading these like sociological texts and these kind of papers that get put out, but I really do think like, it's, it's actually just more productive.

You have like these kinds of podcasts, circuits, right? For people like, uh, Rob Henderson and Joe Henrich and Jonathan height, I think that's actually done so much more for kind of creating a public discussion and creating like a layers upon layers of ideas and these kinds of sociological or psychological topics than academia has done in the past.

And like, w so like, is this the future? Well, let, let me say, there's a huge chunk of academic social science, which is, I think just too academic, like they're focused on a very narrow problem. They're just publishing whatever data they can get. And it's not really conclusive, right. Um, and I really love this kind of stuff where you get someone in a casual environment and you, you talk and then I'll hopefully it's machine transcribed.

So other people can get at it and find it through search, right. Algorithms and stuff. So it's a, it's actually recording the thoughts and observations of people from a certain period of history, right? It's, it's really rich information probably before your time. There was a very beginning time on the internet when people were super open and you could, people would just pour their hearts out on a discussion forum.

There was this time, like maybe it's still true, kind of on Reddit and stuff like this. But, you know, you could say the often, you know, it's a pseudonymous handle, so you can, no one's going to know who you are. But people I think were very, very truthful, just pouring some things out on the internet. This would have been like late nineties, early two thousands.

That era maybe has kind of gone away. But I think as a sociological moment, it was incredible. Like if you go back and read what people were writing in those old forum posts, they were, it was a unique environment. They could be talking to thousands of people, but they were totally synonymous. And, uh, but then there might be a hundred replies, a hundred more comments in the thread.

And so, um, I don't know, it's a unique thing. Um, I do want to point to some rigorous social science, which is not going to emerge from this casual kind of bantering stuff, which I think is super important. Like there's a researcher at Princeton named Marty. I never say his name. Right. It's G I L E N S. And he and his group looked at all the political, this major political decisions done made in the United States.

And they tried to register through polling, whether the outcome was something that the general public liked, whether it was something that the elites liked and they pretty much, they pretty much documented that our democracy is basically kind of a fraud, like, like the notion. Yeah. There's no correlation between what the general public wants and the outcome where there's a strong correlation between what the elites won and the outcome.

And so if you, if you want to compare like our system of preference aggregation versus the Chinese system of preference aggregation, it's not as black and white as you think. And, um, that kind of, that took a huge team of researchers and grad students to just collate the data and then look at the polling numbers and just make sure that they were doing it right.

And that's, to me really good. And I wouldn't have known the answer with high confidence at all, before they finished their project. Right. Because I, I grew up being told that democracy is a real thing in America, right? So to me, that's a kind of a certain scale and a certain level of rigor and a certain maturity being led by senior professors in political science that can kind of only happen in academia.

I kind of want to put a pin in the institution or like the, in the democracy conversation, because I want to come back to that. But like, just in reply to this, like, I don't think the people I'm talking about or like, I see the, I see the, the choices between like the academic system of prestige versus this kind of new, this kind of new, like social thing.

I don't see it as like a, I don't see on the academic end. I don't see data that part of that, because if you, I mean, like, I'm sorry. I think it's my fault. I didn't make this clear to begin with, but like Jonathan height, Joe, Henrich like, these are not like, these are not like purely narrative people. These are people who have put out papers.

Like, uh, they're both there. They are both professors. Um, they have put like a lot of rigorous work into, um, into their theories, uh, and into the actual observations that, that match them. But, uh, the difference I think is just how you, how you promote it, right? So you have these kinds of ideas. And actually the Princeton study got, got pretty far, because I think like if it's, it is really remarkable, but you have all these other academics who are just, uh, getting papers who are being read by like 10 other people in their fields.

Right. And that's how they're, they're, they're imagining their ideas eventually get out into the real world. And then you have these other people who I think are doing still high quality. But they're just promoting it in a different way. So, so that's kind of what I mean by is this the future? I think the thing you just re expressed, I agree wholeheartedly with, I think I just misinterpreted what you were saying earlier.

Yeah. It was my bad, I think like that's actually a very reasonable interpretation of what I said. I just said it wrong. Yeah. I think, you know, like to the extent that there is a, you know, we'll see how this podcast thing or Twitter spaces, all that stuff matures, but to the extent that you have people who have fought, you know, maybe they, there are high G people or whatever, but have thought very deeply and actually engage with the academic literature and have their own life observations, uh, to bring to the table, maybe experiences in a particular environment like Silicon valley or somewhere else.

And, but they make it accessible to everybody. Um, I think there's just tremendous value in that. I, I, I think that, you know, um, you know, you can reach an enormous audience and even like a modest audience, like for one of my podcasts, if, if only a few thousand people listen to it, it's still far more than I would reach, even if I gave a lecture in the largest auditorium on campus.

So something I've been thinking about in relation to. Is a, actually this is not, I don't think I've come up with this idea and nearly close to uniquely. I think many people have come up with us, but the thing I most recently remember reading about it was by, uh, Antonio Garcia Martinez, um, where he writes about like the power of newness, right.

Where you had, like you said, the kind of blogosphere, um, a lot of people just interacting with each other, commenting on blog posts, uh, developing really deep, very authentic, honest conversations. Uh, and he talks about it as kind of a periodic cycle. Right. He talks about it emerging with the printing, press emerging with, um, people traveling to the United States.

Basically whenever you have newness and the now with like Bitcoin, right. Or like with crypto in general, uh, you have these kind of new communities that spring up that naturally attract kind of like idealistic, uh, heartfelt, uh, non-cynical people and also talented people who can actually understand the stuff required for the new, for the new thing.

Um, where is this something that's just kind of like, you want to be able to discover the next thing before it kind of breaks out and that's how you get this kind of, that's how you get those kinds of, um, super productive environment. Um, so I didn't quite understand the point you're making. So when you say the super productive environment, do you mean like, like us talking on this podcast or do you mean.

Yeah, that would be kind of like part of that, right. I'm not sure if podcasts are too saturated now or not, but like basically the idea on the, in the piece, I don't remember what it, what it's called right now, but we'll leave it in the show notes. Um, is that, uh, is that new things attract like talented idealistic people and that just ends up creating good things.

Right. Uh, and so I didn't know how new you consider podcasts now, but certainly the kind of at this point, maybe it's moved on to like crypto or whatever. Um, but essentially that you kind of periodically have these cycles, right. Where, um, a thing or like old institutions become kind of, uh, popularized and taken over by like politics or whatever.

Uh, and then you have these, uh, these new things that spring up and that's where the kind of interesting discussion noises. That's the kind of idea. I certainly agree with that. I mean, I think you, I mean, I don't feel, I feel like I'm kind of on the periphery of this, cause I'm a little too old involved with, you know, and I'm also a star in some census in my day job as stodgy academic.

But, but, um, yes I do. I'm very envious of all the ferment and stuff that goes on and in, in the bay area and Austin and places like this and uh, every now and then, like, I dunno if you're a slate star codex or Astro codex, 10 person. Yeah, Scott has been nice enough to host a few events with me in, for me in the, in, in, you know, in Berkeley or in the valley.

And those have just been phenomenal. The young people in the energy is just unbelievable. It's off scale. Uh, I love it, but, um, I wanna, I w you mentioned podcasts, I just want to, I have to get this anecdote on the record, cause it's a good addict. So, so I'm at a Silicon valley kind of meeting with a bunch of startup founders and VCs and stuff, uh, around two oh 2005.

Maybe this is ancient history for you, but, um, it's a funny story. So I run into this guy called the, I think his name is Evan Williams, who, um, he had sold blogger to Google and was working on a startup called Odeo O D E O, which was producing podcasting tools to make podcasting easier for the masses. And this is 2005.

So if you do a little, if you start counting on your fingers, you realize Evan was a little early on this one because, because podcasts didn't take off for another 10 years or something. So he's working on Odell, but I, at the time thought podcasts had tremendous potential. So I was talking to, I think his name is Evan Williams.

He's a guy from Nebraska. He's one of the Twitter co-founders. So, anyway, so I was talking to Evan about this and, um, I was saying like, oh yeah, I hope this podcasting thing. This is way too early. Of course. But I like, I hope this podcasting thing takes off. I think it will be really amazing if everybody can create content and share it easily, you know, even if it's just audio content and he, I, he, so we start discussing and he's like, um, yeah, there's this other thing we're doing like a, and you, you maybe know the story, but basically within this startup, they, some total side project was experimenting with 140 character text message things.

And that became Twitter. So he was describing that to me. And I'm like, who the hell would ever use that? Like, what's the point of that? Like, I totally didn't get it. And I was like, that's stupid. Like, you guys are wasting time on that. And of course the rest is history. So podcasting got delayed and didn't happen for another 10 years after ODALE.

But Twitter turned into this monster thing, which is the, like the number one place. Now. It seems like where people interact. So, um, anyway, just thought I'd get that anecdote on the record. Yeah. You see this happening all the time, right? Uh, I don't remember whose, I don't remember the author off the top of my head, but there's this guy who wrote a book.

About, uh, the PayPal mafia, right. And the starting studying, uh, founders of PayPal, like teal and David sacks and eventually Ilan as well. And they were all trying to do things that were not PayPal. They were like, there, there are these like, uh, handheld and held electronic things called like Palm pilots.

Oh yeah. You, you can like tap between the Palm pilots and send money. And as a kind of side thing, we'll also let you like transfer through email and stuff like that. Um, and then the email thing took off and they had the wisdom of saying like, okay, th this is seeing success. People want this let's, let's give the people what they want paid and, and, and thus, um, PayPal.

And, uh, the other end Elan's end, uh, x.com. Uh, they, they eventually got the demand, eventually got off the ground. So I think like these, these kinds of spontaneous, so these kinds of like serendipitous events are like probably more common than we think. Yeah, no, absolutely. I mean, a lot of times like, well, there's a, there's a term for this.

There's a Y Combinator term for this, which is fun for the pivot, which is that you, you know, you have a really strong feeling what they're currently doing. Isn't gonna work out, but you like the founders so much that you're going to fund them anyway, because whatever they pivot to has it is an interest going to be an interesting thing.

Yeah. I mean, YC is. It's like a very small amount of money at the beginning. Right. It's like 60 K. It was like, it was some, it was ridiculously, it was like budgeted for like, if they could eat ramen at every meal and, but they had to pay rent. Um, yeah. How much money would they need to survive for three monsters?

It was some ridiculous thing like that. That's how they got the initial, uh, uh bite-size yeah. So I guess maybe the last or the second last, well, the last full, uh, topic that I really want to talk about is the kind of structures or institutions in society. Right? We already talked a little bit about academia, but this is maybe kind of the central theme of this podcast is essentially that you have these kinds of selection mechanisms, these kind of a way of ways of ordering society.

And that this is, um, this is just not working or is at the very least far from optimal. And, uh, I think the best company in off points for this is on the same thing we talked about before, where you have this kind of Eunice or this kind of like almost like natural gating mechanism, right? Like if you don't understand crypto, then you're not going to be doing crypto.

If you're, if you don't understand like how to use the internet and you're not going to be doing, uh, blocks, if you don't understand how to read or write, then you're not going to be doing, uh, you're not going to be making pamphlets at the beginning of the printing press era. Right. So. Uh, I think like one of my, kind of, uh, one of my kind of ideas and this isn't unique to me is that it's actually selection that creates the quality, not the quality that creates the selection.

So essentially you have, like, for example, in like early sociology, right. We think of sociology now, as it's kind of very like, uh, very, um, ideological, corrupted, um, non rigorous field. And, and you can question the earlier people on bigger as well. But if you look at, for example, like max Weber's writing, um, it looks a lot more like a math textbook than like, uh, then like a modern sociology textbook.

He like enumerates these definitions. He makes it very clear. He is like, he spends a page explaining what he means by pure type versus statistical type. And it's like this very kind of to, to like, to like go back to the earlier term, it's like this very kind of like shaped rotator, like pattern, um, where you have to keep a bunch of concepts in your mind.

And I think that what's, I think that what happens at the beginning of these fields is that you have this kind of natural lockout and that kind of disappears. Right. But the question is like, whether we should just be incorporating these kinds of lockouts as much as we can in as many places as we can.

That's a good question. I don't know very much about the early history of sociology. Sociology is a field or Weber, but it does strike me that if you look at analytic philosophy today, so I don't know if you know the distinction between analytic and continental, but analytic is much more rigorous in a way kind of rigorous with definitions and trying to be extremely clear.

It can be, but they could be, they could be approaching something like Rawls or, you know, something very, which is very real world, but they're trying to do it in the way that I think you described Weber wanting to describe society. So, so in that sense, like this, this notion of being very, very rigorous and, you know, at the cost of maybe being too dry or something like this, it's still around.

Um, and so I think if you, if you ask, if you said like, oh, let's look at the subset of analytic philosophers who study like, you know, questions of how we should value certain competing things in society or something. I think their writing would probably be a lot, like the way you described Weber. So it still exists, I think.

But, um, it doesn't predominate because here I'll, I'll just throw something in which is a big monkey wrenches, that one of the ways that departments. So if you think of natural selection or competition among departments, one of the ways that departments, uh, Survive that contribute to their fitness is just how many students they can enroll.

How many students want to take the classes. So there's definitely a push toward making things more accessible and interesting to undergraduates. Um, but maybe you wanted the couple that maybe you, you know, maybe you want to have research institutes, you know, like in France or something that are not coupled to undergraduate teaching at all.

And they're just funded by the government. Um, you know, that's a different way to go, right? We don't have that. We do have that in the United States and the department what's called the department of energy cause we had to make bombs and stuff. But, uh, and much of things that just came out of DARPA is just amazing.

Right? Yeah, exactly. So we unfortunately in the U S well, for better or worse than the U S most kind of pure research is it's overwhelmingly done at universities. Right. And it doesn't have to be that way though. But on the other hand, then you need an extra line item where you say, oh, we're going to spend money on all these researchers and pay their salaries.

And they're not going to have to teach undergrads and generate tuition. So, um, I don't know, this, doesn't answer your question, but it's definitely a, a relevant factor. Yeah. Is this, is this funding system a bit different for like stem fields? Because actually something that I saw a lot was that you had all of these humanities fields constantly kind of advertising their programs.

And math fields where especially math that is that's like my degree, um, was completely different. It was also, it was kind of like, uh, going, it was kind of like, um, they wanted like fewer students. I think that was something that was kind of interesting. Yeah. You're asking a question, which only professors and not even most professors understand this, you'd have to actually get to the level of being an administrator like a Dean or hire to really understand these things or a department chair.

Like what is the political economy of the math department? Can they do well and do well means, you know, get the budget to fill positions when faculty retire or build a nice faculty lounge or fund good graduate fellowships. Can they get that support from the university? If they don't have enough undergraduate majors or do enough undergraduate service teaching for say engineers and biologists, can they just survive on the beauty and elegance of their subject or, or the awards they win?

Like the fields medal. Um, it's a very complicated political economy within the university. And, um, to some extent it's defined by the top leadership. So the more, the top leadership actually care about the fields metal are producing beautiful results in combinatorics the more they're going to cut slack to the English, to the math department.

So yeah, you don't have very many majors, but what you're doing is really important. So we're going to continue supporting you, but I can tell you right now that among American universities, that attitude is very rare. So what you'll find is most, most math departments try to increase the number of credit hours that they, uh, that are accredited to them as a department.

So, um, you know, there's always that kind of, if you want market pressure on all of these academic disciplines, actually, I think I actually just stumbled upon the answer because the place that I went to is very odd. Um, it's a, the university of Waterloo often referred to as like the MIT of the north, but it's actually so much like MIT attracts like some level of high, high stem students.

And like you would think comparatively fewer students than other, um, in other topics. Uh, so the first thing I think I should explain to the audience is that in Canada, you actually apply to like programs. So not only do you apply to like, um, you not only do you apply to a university and have like, maybe you have your preferred major, but you're actually kind of locked in and it's kind of difficult to transfer between, uh, transfer between faculties, especially.

So you apply ahead of time. They know, and they know that unless you're going to inconvenience yourself pretty much, uh, what you're going to, uh, go into. And there's this weird, um, reputation thing where, I mean, like in any country, I think you're going to have a difference between like, okay, here's a school that specializes in a stem, here's a school that specializes in a humanities.

Um, like you have like MIT and Gail, for example. Right. But, um, because Canada is kind of a small population, this sort is almost like a 100% sort in Canada. Um, so you have always like way too many stem majors, uh, applying to Waterloo. So I guess it's just like, not very much a problem for them. Yeah. They may not care.

I mean, they may be a unique kind of school. That's not subject to these kinds of pressures in the U S the budgets are increased less and less coming from the state and more and more coming from individual tuition paid by the families of the kids. And so then the, yeah, go ahead. Yeah. The subsidies in Canada, I believe are much higher as well.

Yeah. So, you know, I mean, all of these things are complicated systems in and of themselves and the outcomes that they produce are sometimes, as I was saying, not even understood by the participants. Like if you grab a random professor. And ask them, well, how has your department a budget established every year?

Like what, how does the Dean get his money from the provost? And, uh, what, what are the factors that determine, you know, most professors don't really understand it. They have a vague model of it. They know it's good to win an academic prize. It's good to have lots of undergraduate majors or credit hours, but they don't really understand the trade-off, how it works.

So, um, and every institution is different. Yeah. So how accurate really are these kinds of like prevailing models of, for example, universities, right. So you have, like, I think the major kind of critical model is something like, oh, there's been a kind of woke, um, bureaucratic dilation of a lot of the processes.

And they're also kind of just like, not really caring about the right problems and they're, um, they're often echo chambers, so it's difficult to question all the assumptions, even in kind of like, I mean, this is another, this is another kind of, uh, book of, uh, or sorry, a can of worms, uh, like, like old assumptions, like string theory.

You don't have to open that can of worms if you don't. But, um, like how, how, how accurate is that model in terms of, there's no simple answer, because if you think, if you think of, I used to say this sometimes, cause my job as VPR. All aspects of research at a big 10 university with 50,000 students. So, so, and, and thousands of professors.

So, so, you know, I, it covers things from gene editing to super computing, to, you know, nanotechnology, to, you know, whatever everything is in there, right. Economics, social sociology. So it's such a broad set of activities that it's hard to characterize. You could make some generalizations, um, you know, uh, is it becoming more bureaucratic?

I would say yes. Uh, some of this is driven by, uh, just regular regulatory issues, like title nine and things like this just require there to be more administrators, tighter scrutiny on use of federal funds. All of these things just force the universities to have more and more bureaucrats. Um, there are also, uh, all kinds of diversity efforts that require, you know, more or less an enforcement arm to make sure departments are doing the right thing.

And that professors have enough training in these areas to do the right thing. Um, all of those trends are definitely there, um, is, you know, uh, do scientific do academic subjects that suck that basically don't produce results. Do they die off? No, not really, not if they could continue fooling undergraduates and to majoring in those subjects.

Um, one, now that I'm not in my position, I can, I can, I can say things like what I'm about to say out loud. I've whereas in pro before I can only say it in private, but, but let's compare productivity of say research in, oh, I don't know, like battery material science and battery technology versus, uh, the ed school.

Okay. So we have a talking the ed school. Well, I mean, I'm not attacking them, but we can just talk about a very simple, met some very simple metrics and then immediately upon phrasing it this way, things become obvious. Right? So compared to semiconductors, in other areas of science batteries are considered a kind of lagging thing, right.

Everyone's always complaining, like, why don't we have better batteries? Right. Cause they're kind of a bottleneck, right? Like every other part of my laptop has improved tremendously in 20 years, but the battery is not improved very much. Right. So, but it's still improved a lot. Okay. So you got maybe like, you know, a 10 X in battery technology over a few decades, right.

And that's considered a laggard, like kind of a laggard kind of, uh, field or maybe internal bus combustion agents. You could say it's 50% more efficient now than, and it took decades to get that because it's very mature technology. Well, let's look at the ed school. Let's suppose you have a goal. Kid comes into my class.

Doesn't speak any French. I want them to get to a certain level of capability in France. In, you know, and can I, how much have I reduced the time required? How much have I improved the efficiency of that in, you know, 50 years of ed school research with billions of dollars spent answer probably zero, maybe negative gain, right?

Uh, I would, I have a kid comes into my class in high school. I would like to teach him algebra one. Do we do it better now than we did it 50 years ago? Maybe it 12% worse. Yeah, well maybe we do it worse. So it could be zero. I want to put a star in that this is normalized to the world. So it's possible. The world just got better.

It's probable the world that just got better. Well, I, I just, I just want to say, like, it still takes one, you know, like there's a, still a one-year high school sequence for algebra one. And the material covered is pretty similar to what it was covered was covered in that class 50 years ago. And so one could argue there's been zero productivity gain despite the billions of dollars and research efforts of all the ed school people.

So my point is that if you try to measure anything quantitatively, you cannot find anything for which ed school activity reached the level of productivity gain of any level, anything in stem, you pick anything instead, like the worst thing is. And it's better 50, you know, simply because if nothing else they could, like they could like effectively use the computing technology they were given over the last 50 years.

Right? Like my heart monitor. Okay. Maybe the basic physics of my heart monitor is not any better than it was before, but at the microchip is better. Um, but I don't even think they've figured out how to properly use computers to actually impact. And they haven't figured out how to properly use RCTs like this.

This is like Sean Martins, baker thing, someone I know. So my point is, if you actually ask for quantitative metrics, you could argue this whole area of endeavor has not, has basically had effectively zero productivity. Yeah, no, the ed schools are so much worse and I'll make it clear for like anyone who's listening.

This is my opinion. Not, not Steve's, but you look at their kind of like recommendations. I think this was a very big headline story recently where the official curriculum in California, there was an amendment there. She liked denied that kids are naturally gifted. We were talking about like anti-vax level kind of like scientific denials, like official like, uh, uh, official specialists that are consulted by these, um, by these boards of education.

But, but even if you leave aside the really controversial stuff, which you could argue could be negative, negative progress, it's hard to find positive progress. Right. And you're talking about, we could, we could do a 50 year look back, you know, that would bring us to 1900. So, you know, like, do we, did we figure out a better way to teach French?

Did we figure out a better way to teach French teachers? Did we figure out a better way to select French teachers? I don't think we made much progress on any of that stuff. So anyway, but then our ed schools in trouble, no, not at all. They have record enrollment and slow. Plenty of kids want to go into, go to ed schools and, you know, cause they have to, to become teachers.

And so, you know, there isn't really like if you ask as a dynamical system is higher ed evolving, like in a useful way. Not, not really. Yeah. So there's actually an even more cynical take. Right. Which is that essentially the ed schools are kind of like, um, you know, like in store where there's like this rainbow bridge that all of these, uh, that all of these, um, forgot what they were called.

Yeah. Like low keys, army is using to come attack like earth. Um, The kind of like super cynical take is that ed schools are the kind of rainbow bridge between these kinds of like super ideological subject and the kind of rigorous subjects. Right? So you get these kinds of like very silly articles and like, uh, journals of like physics education or whatever.

Right. And that are kind of talking about these ideas, applying these like ideological theories to how we should do physics, then they're just like completely, you know, I think if, I think if we could figure out better ways to teach freshmen physics, that is an extremely valuable thing, but I'm skeptical that we've made a lot of progress on it.

So, and furthermore, this is a problem that you identify, which is that it can become just a means of like pushing your ideology either on the students or actually forcing it on the physics department. So that that's, you know, the, the math education and physics education, uh, sub, uh, components of the faculty, and some of these departments are seen as really the ideological Vanguard for pushing a lot of wokeness stuff into that department, into that previously kind of somewhat protected department.

So the problem being, if you have a very activist professor and everybody else is busy doing their main research, it's very hard to, you know, defeat that activist professor. Cause he's, he's, he's, he's really motivated. He really wants to, he's wants to spend all his time on that stuff. So, um, yeah, it's a, it's a real issue.

Yeah. And, uh, this is kind of, I know you've spoken to Richard Hananiah on your show and on his shows as well, but this is the kind of like, um, he had this article called, um, wokeness and sat on statues, right? Where he's comparing these, um, these kinds of loyalty oats or these kinds of programs as the, kind of, uh, as the kind of loyalty tests you had to, um, you had to Saddam, Saddam Hussein nor to Stalin where you have these kind of super vague laws.

And essentially the only way to be on the good side of these laws is to be like more extreme than the other people. Right? So you had the situation where, um, you have the situation where, uh, there was kind of always a sliding window and he describes the same thing happening here, where there's always a sliding window in terms of how far you have to go.

And it doesn't take very long before that has to go into like reality denial. Well, I agree with that. I think, um, you know, w we are on that slippery slope right now, um, from, it's not even like a slippery slope is, and there'll be more things at the bottom. Right. And we are kind of like, we were kind of already in it.

One, one battle. I fought only with limited success when I was VPR was a movement to get rid of the GRE, uh, because it had negative effects on diversity, you know? And, and I, all I asked for was at least a data. Discussion within each department, which was actually conditioned on the real research, because there are real researchers who work on like the predictive validity of GRE and stuff like this.

And a lot of these professors, especially the ones who are activists were not even aware of the literature that, uh, you know, real professionals psychometricians, who had studied the GRE, you know, they had written papers in nature and things like this that were just not even brought up in the discussion.

So as VPR, I was saying to the departments, at least, can you please have D if you're going to have a faculty meeting about getting rid of the GRE at least make sure the actual research on the subject is introduced so the faculty can judge it before they vote. And even that was controversial. I mean, I kind of asked this question before about like the big, the big question is like, what, what can we actually do about this?

Right. What can we actually do to kind of, um, turn the clock in the other direction to make academia, to make academia great. Again, let's say, well, let me, let me tell you about two things that are actually going on. Although the, with, you know, I think different levels of success, there are some billionaire tech type people who want to create new universities.

Now that is an incredibly daunting prospect because universities take decades easily to get established. You're competing against, in very entrenched old competition. And, uh, it would have to be a passion project. Of somebody who's willing to give it sustained effort for many decades and invest billions of dollars in.

Now there are people talking about this and there are some initial efforts, but I think it's, it's thing. It's a tough road on a different scale. I'm seeing, I'm actually involved in some efforts to fund kind of focused research organizations like mini Manhattan projects. So some billionaires are giving away money to fund sort of five year $30 million focused project to try to accomplish X, you know, maybe something in molecular biology or something in, you know, condensed matter physics or whatever.

Um, and that's a different model. Um, that's a kind of thing that could turbocharge, you know, academic research, um, because if you have these little things springing up to try to attack problems that, you know, departments just aren't necessarily set up the right way to do these kinds of things, um, that could revitalize things if there were enough of these things.

Um, so I dunno, I mean, one of the good things about the current income inequality we have is that a lot of the money's in the hands of people who like technology and science. So hopefully that, that will trickle back in, in, in productive ways. Um, the, the, the big thing I think that will happen organically is that the ability to educate oneself, uh, if you really want to, um, overusing the internal.

And then also form some kind of validation, like take a test or something to prove that you really did master X. I think that will just, it's an inescapable that within a decade or two that will exist. The only question is whether companies and society in general will value that in terms of prestige signaling, as much as the training that you get from a more traditional institution.

I think the capability already exists, certainly in some fields like computer science, but, but, um, in other fields it'll arise as well. And the whole question is just what, how society adapts to it. Last question of the show. Uh, what is either one thing you have one thing you see in the world where there is chaos and you would rather there be order or one thing where there is order where you would rather there be chaos.

Oh, wow. Well, the obvious answer for chaos is the Ukraine conflict because, uh, there is at least some tail risk that's going to lead to world war three. So, um, without getting into all the details, you can just imagine ways that if NATO and Russia mix it up the wrong way, you know, it could, it could escalate very dramatically.

So it'd be good to get that settled as soon as possible. And it's just terrible to grind down. You know, if, if we kind of know where the outcome is going to go, like, I don't really see how Russia is going to lose. Then can we get it over with, without grinding down the Ukrainian people and their infrastructure and their economy to zero?

I mean, I just think it's, it's, it's inhuman. I understand it's instrumental, especially for the west to keep this going, but I think it's inhuman. I mean, we ground down the Iraqis and the Afghanis for 20 years and got nothing out of it and did nothing for them. And now we've just forgotten about it, but I fear that we're going to do the same thing to the Ukrainians.

Uh, too much order needs more chaos. That's a good question. Um, wow. I have to think about that. I think there are well I'll, I'll give you a very pedestrian one. The U S has completely the, the, the wired internet access is completely monopolized in the U S so there's basically like people typically have very limited options and we pay a heck of a lot more than people pay in other countries for inferior bandwidth.

So I think that's an, a very egregious example of a monopoly that needs to be broken up and a lot more competition needs to be allowed. Thank you so much for being on the, on the show. And, uh, I know you have to go, so we'll, uh, respect your time and leave it here.

So that was the interview with Steve Hsu. I'm back and with better audio quality. We certainly covered plenty of ground today. And I think longtime listeners and readers are especially exhilarated right now. You heard a lot of what's interesting and underlying some of the thought that I put into my writing and into podcasts like this one, and also a lot of just fun, enjoyable content.

Unfortunately, we didn't get to talk about genomics, which was an area of Steve's personal expertise that I also have a great interest in, but there's always next time. And if you want to listen, when he eventually appears on the show again, or to the plenty of other episodes that we have on offer, then you can subscribe to the show.

And as I set up top, we don't take funding right now. So the best thing you can do to help the show is to share, to talk to your friends, to recommend it as only you can do, you'd be surprised at how many people are willing to listen. You can guide them to something interesting, and you can help the show right here.