The Art of Manliness - #512： Why Generalists Triumph in a Specialized World

00:00:00.000 Brett McKay here and welcome to another edition of the Art of Manliness podcast. Now we often

00:00:11.380 think to become a success in today's modern world, you have to specialize and specialize

00:00:16.280 early. My guest today makes the case that actually the most creative, innovative, and

00:00:20.700 successful people don't specialize, they're generalists. His name is David Epstein and

00:00:24.560 he's the author of the book Range, Why Generalists Triumph in a Specialized World. We begin our

00:00:28.980 conversation discussing two different paths to success as embodied by Tiger Woods and

00:00:33.220 Roger Federer and why we're naturally drawn to the former specialized approach even though

00:00:37.620 the latter's generalized approach is in fact the most common way to success. David then

00:00:41.940 explains why our increasingly complex and abstract world requires not only having a depth but

00:00:46.320 a breadth of knowledge and how our education system hinders us from gaining such. David

00:00:50.400 and I discuss why you shouldn't expect to know exactly what you're going to do for your career

00:00:53.800 when you're young, why you should dabble in lots of different activities when you're first

00:00:57.320 starting out in life, why you should keep doing that even when you're older, and why

00:01:00.580 there's a correlation between having hobbies and winning the Nobel Prize. We also dig into

00:01:04.600 why intrinsic motivation is often mistaken for grit, why you shouldn't be afraid to sometimes

00:01:08.600 quit things, and the importance of finding pursuits that fit you if you want to achieve success.

00:01:13.120 We end our conversation with David's argument that our increasing specialization is not only

00:01:16.600 stifling individual flourishing but also getting in the way of scientific advances that would

00:01:20.560 benefit society. Out of the show is over, check out our show notes at aom.is slash range.

00:01:25.700 David joins me now via ClearCast.io.

00:01:33.440 All right, David Epstein, welcome back to the show.

00:01:38.760 Thanks for having me. Again, only six years after my agent told me not to let it be five years before

00:01:44.220 I have another book out.

00:01:45.280 Right. Well, your last book that we had you on to talk about was The Sports Gene,

00:01:49.360 which discusses, you know, why some people are just great at certain sports, but you got a new

00:01:55.580 book out, kind of a left, you said it's a left term, but I think there's a connection. It's called

00:01:59.780 Range, Why Generalists Triumph in a Specialized World. Did The Sports Gene sort of begin the thinking

00:02:07.140 about this book?

00:02:08.480 It absolutely did. And in fact, it sort of led to this book, even though this book is,

00:02:12.960 Departs from Sports After the Introduction, it really grew out of The Sports Gene in the sense

00:02:17.820 that after The Sports Gene came out, I was invited to the MIT Sloan Sports Analytics Conference to

00:02:23.380 have a debate with Malcolm Gladwell. So it's like on YouTube titled The Sports Gene versus 10,000 hours,

00:02:28.640 even though we actually have significant middle ground. And, you know, because he's very clever,

00:02:32.740 and I didn't want to get embarrassed. I'd never met him before. I guessed that he would argue about

00:02:38.260 the importance of a head start in athletic development in very narrow, technical, so-called

00:02:43.660 deliberate practice. So I went and gathered up, you know, all the studies I could find that tracked

00:02:49.120 the development of future elite athletes. And what I saw was a pattern where, in fact, rather than

00:02:56.200 doing sort of the Tiger Woods, where they specialize very early, they have what scientists call a sampling

00:03:01.900 period early, where they gain a breadth of general skills, try an array of sports, learn about their

00:03:07.900 own abilities and their own interests, and systematically delay specialization until later

00:03:13.280 than their peers who plateau at lower levels. So I brought that to the debate, saying, this is your

00:03:18.680 hypothesis, and here's the data that cannot fit with that hypothesis. And afterward, we became running

00:03:25.540 buddies and sort of would talk about it on our own time. I filed it away in the back of my brain until

00:03:30.320 this point I describe in range where I got involved with the Pat Tillman Foundation that basically helps

00:03:37.260 military veterans career change and gave a little talk about this. And they were so hungry for

00:03:43.560 information about how to bring diverse experiences to bear on whatever they were going to do because

00:03:49.920 they felt like they were behind when, in fact, they actually had all these powerful experiences and

00:03:54.180 skills that their peers didn't. And so I sort of started to think, I should really investigate this

00:04:00.040 far outside the sports world and see if we see this same pattern of an advantage accruing to people who go

00:04:05.940 abroad early and maintain breadth even as everyone around them is rushing to specialize.

00:04:11.440 Okay. So let's talk about how you started off with the sports, the sports stick, right? The sports

00:04:15.280 analogy. Yeah. And you mentioned there's two approaches to how we go about or people have

00:04:21.500 about going, becoming an expert. And the one you said, there's the Tiger Woods method. And then you

00:04:27.060 also say in the book, there's the Roger Federer way. So talk about, let's say, so Tiger Woods and

00:04:32.660 Roger Federer. Can you do like compare and contrast between those two approaches?

00:04:35.900 Yeah, sure. So originally I titled the book proposal, Roger versus Tiger. So Tiger Woods,

00:04:40.840 I think his development story is pretty well known. He was physically precocious. He could like balance

00:04:46.160 on his father's palm at six months old or pictures of this in his father's book. He was, by the time he

00:04:51.160 was two, he was on television golfing. And by the time he was three, his father was already giving him

00:04:58.420 like media training for his, his future. And by the time he was four, he was like hustling adults at

00:05:02.900 golf courses. You know, and by the time he was a teenager, he was famous and went on to become the

00:05:07.140 best golfer in the world. And so that is very early specialization is kind of 10, very 10,000 hours

00:05:13.100 rule centric specialization became the analogy for a huge number of books that write about performance

00:05:20.920 and how to get good at stuff saying, just extrapolate this to whatever it is that you want

00:05:25.460 to do. Meanwhile, Roger Federer, his story is much less well known. He played a huge array of sports as

00:05:32.280 a kid. His mother was actually a tennis coach and refused to coach him because he wouldn't like return

00:05:37.020 balls in a normal way and do sort of structured practice. When he actually got good enough to get bumped

00:05:42.020 up a level, he declined because he just wanted to talk about WWE with his friends after practice.

00:05:47.920 And actually when he first got good enough to get interviewed by a local paper and the reporter asked

00:05:53.140 him what he would buy with his first paycheck, if he ever became a pro, he said a Mercedes and his

00:05:58.140 mother was totally appalled and asked the reporter if she could listen to a recording of the interview

00:06:02.600 and he obliged. And what Roger had actually said was mere CDs in a Swiss German accent. He just wanted

00:06:08.180 more CDs. So unlike Tiger, he didn't have these big dreams of being a tennis pro. He didn't specialize.

00:06:15.760 His mother forced him to continue playing badminton, basketball and soccer long after some of his peers

00:06:20.720 were not only specialized in tennis, but working with sports psychologists and nutritionists.

00:06:25.960 And of course, Rogers rose to the pinnacle of his domain as well. And so my question was which model,

00:06:30.880 the Roger model or the Tiger model is the more common one en route to expertise.

00:06:37.500 So, I mean, it's kind of interesting because you mentioned, you know, ever since Tiger Woods,

00:06:41.020 there's been all these books put out about how his approach is the way, if you want to become an

00:06:44.500 expert in your domain, you got to do the Tiger approach. But I feel like this existed even before

00:06:49.340 Tiger, that we had this idea that if you want to become the very best, you had to get started young.

00:06:53.640 I mean, people use examples of like Mozart, right? He started composing music when he was three or four

00:06:58.860 and said, well, if you want your kid to be great and whatever, you got to get them started young.

00:07:03.260 Why do you think we have, why do you think that makes sense? And why do we think

00:07:06.020 that the Federer approach, we sort of dabble and sort of, you know, act like a deletante is looked

00:07:11.060 down upon. And the Federer approach, by the way, turns out to be the normal one for athletes who go

00:07:16.340 on to succeed. And I think it's, I think it's multifaceted. First of all, it's not that intuitive,

00:07:22.020 right? We are absolutely not programmed to think that there could be anything wrong with a headstart

00:07:26.680 or that there is, if you want to be good in X, that you should do anything other than X in order

00:07:33.200 to become the best at that. It's not intuitive. I also think it's partly a holdover from the pre-knowledge

00:07:40.320 economy era where the structure of most every organization that people were involved with was

00:07:46.120 very up or out, right? It was the tasks that people had to do for work were much more standardized and

00:07:52.100 they could do similar things over and over and over again. And so they were used to being narrow

00:07:57.320 and, and quite specialized. And then I think these things like Tiger and this whole genre of books

00:08:04.020 that picked on Tiger and Mozart essentially, and also some chess players sort of stoked our obsession

00:08:11.080 with precocity, right? And what those books do is they use these prodigies to say this, anybody can do

00:08:17.900 this in any domain, but the reality of it is that they pick very particular domains that people who

00:08:24.840 study skill acquisition know are horrible models of almost everything else people want to learn.

00:08:30.620 So it's like this magic trick where they say, look at this classical music prodigy, look at this golf

00:08:35.320 prodigy, do this in whatever you're interested in. But that's where the sleight of hand happens.

00:08:40.280 Because if you look into the research of how people get good at things, golf is like a uniquely poor

00:08:46.040 model of almost everything else that people want to get good at.

00:08:49.320 Well, so yeah, so it's not to say the Tiger Woods approach does not work. It only works in specific

00:08:54.520 domains. So like golf would be one, I think you mentioned chess is one, music possibly. And I guess

00:09:01.600 what all these have in common is that they're very procedural. Like once you learn the steps,

00:09:05.780 you just keep doing the steps over and over again, you can get better at it.

00:09:08.760 I mean, people who study skill acquisition basically classify golf as like an industrial task

00:09:13.200 where you're not really dealing with much human behavior. You're not dealing with teammates.

00:09:17.980 It's non-dynamic. Essentially, you're trying to do the known movements. Like you know the answer

00:09:24.360 and you're trying to execute it over and over with as little deviation as possible. And that is the

00:09:29.060 epitome of what psychologists call a kind learning environment basically. And as is chess to a less

00:09:36.120 degree, but still very much so, as are certain aspects of classical music, but not music overall.

00:09:42.400 And so it's very telling that basically child prodigies always come in a very small range of

00:09:48.420 domains that psychologists who study skill acquisition classify like more like, more like

00:09:54.320 kind of industrial domains and things that are easy to automate essentially. So it's, it's a very

00:09:59.240 small, it's the minority of domains, but you know, this a half dozen bestsellers, at least that I can

00:10:05.980 just think of off the top of my head, use them to extrapolate to everything else. And that's where

00:10:10.440 these inappropriate conclusions are made because in fact, the tiger model may well work for golf.

00:10:15.720 There's actually like a surprising dearth of research on golf compared to other sports.

00:10:19.440 So I don't know the jury's out, but because of the structure of the domain, I can definitely believe

00:10:24.040 that early specialization is the way to go in golf, but in sports where there are other people

00:10:30.560 involved, where the situations are much more dynamic, where you have to react to things quickly,

00:10:34.000 it is absolutely not the way to go. The, the, you want to think of it more like studying language.

00:10:39.840 So we know people who grew up multilingual are better able to learn a third language, even a

00:10:44.600 made up one, if scientists give it to them in experiment without being told the rules. And that

00:10:49.360 looks the same for people who play multiple sports that involve anticipation of what other people

00:10:53.740 are doing or of flying objects. So if you do a bunch of different things early, you're better

00:10:58.880 equipped to be able to pick up any new skills going forward. And so that's what you really want.

00:11:03.960 That, that general framework that allows you to, to become a master learner, basically.

00:11:08.940 Okay. So procedural skills there, those are a kind learning environment because you can just learn

00:11:13.180 the things, then you can master them. I guess the other domains that aren't like that, where they're

00:11:18.060 complex dynamic, you call these wicked worlds or what learning psychology or skill acquisition

00:11:22.860 psychology is called a wicked world.

00:11:24.300 Yeah. And so the kind, so to define the kind learning environment. So if we, it's basically

00:11:30.620 a task where patterns repeat over and over, the task itself is very constrained by very clear rules.

00:11:38.540 And every time you do something, you get feedback that is both immediate and fully accurate. So we can

00:11:45.680 think of, if you think of something like chess, it's very constrained. People aren't allowed to move at the

00:11:50.660 same time, right? You have to pause before another person does something. There's an enormous database

00:11:55.700 of previous games, patterns repeat over and over. In fact, grandmasters rely on pattern study to do

00:12:00.920 what they do. And the feedback for a move comes quick and all the information is available. And this

00:12:06.620 happens to be what makes it so easy to automate, which is why computers, you know, one of the first

00:12:11.860 things they mastered was chess because it's a kind learning environment. On the other end of the spectrum

00:12:15.100 are most of the things that humans want to learn, where not all the information is available. You're

00:12:21.500 dealing with real-time human behavior and lots of things moving at once. Not all the information,

00:12:27.020 there's information that's hidden from you. You may get feedback, but you might not get it all the

00:12:31.520 time. It may be delayed. It may be partial. It may be inaccurate. So Robin Hogarth, who coined this,

00:12:37.240 a psychologist who coined this kind, wicked learning environment, used as an example, a famous physician

00:12:42.440 who was renowned for being able to diagnose, diagnose typhoid, like before weeks before a

00:12:49.400 patient had any symptoms at all. And he would do that by palpating their tongue or feeling around

00:12:53.280 their tongue with his hands. And over and over, he could predict who was going to get typhoid before

00:12:57.440 they had a single symptom. And one of his colleagues later pointed out that he was a more prolific carrier

00:13:02.340 of typhoid than typhoid Mary, because he was the one giving typhoid to these patients by feeling around

00:13:07.540 their tongues. And so in that case, the feedback, the positive feedback reinforced the exact wrong

00:13:13.980 lesson. So that's like a super wicked, like most of the things we're doing are not quite that wicked,

00:13:19.240 but they're more toward the wicked end of the spectrum. Like tennis is further from the kind

00:13:24.720 end of the spectrum than golf, but sports are still, you know, far, far from the wicked end of

00:13:31.320 the spectrum compared to most of the things that people are trying to do in the world of work.

00:13:34.980 Yeah. Business is a wicked world because there's so many constituent parts and they interact with

00:13:40.340 each other complex ways that you can't predict. Politics would be one, just management, like

00:13:44.900 working in organizations, there's all these different people with different interests and

00:13:48.440 you don't know what their interests are. And so you can't, you can't come up, you can't develop a

00:13:52.980 system to manage that. That's a great point, you know, and so, and I, and this, this shows up.

00:13:57.700 So in that, in more wicked learning environments, what you want is breadth. So there's a,

00:14:02.000 there's a classic finding again, in people who study how we acquire skills that's, that goes

00:14:07.140 like this breadth of training predicts breadth of transfer. So what you want to do in a world

00:14:12.580 where you're not repeating the same thing over and over, the trick is to be able to apply your

00:14:17.440 knowledge and skills to situations you've never seen before. So not like golf, not like chess.

00:14:23.000 And if you want to be able to do that, you want to have really broad training because instead of

00:14:27.160 learning procedures, what you're trying to do is learn these general abstractions that are

00:14:32.180 frameworks that you can apply going forward. So, you know, to, to use some research that I cite

00:14:37.400 in range is training people to respond to, well, training on, on simulations to respond to naval

00:14:44.260 threats, essentially training commanders. And they tested all these different methods of training.

00:14:50.140 And some of the people would practice a certain scenario over and over and over and over and over and

00:14:53.900 over and over again until they got really good over the course of that day. I'm responding to

00:14:57.420 a certain scenario. Other people saw a different scenario every time in training. And at the end

00:15:01.960 of the training period, the people who are always seeing a different scenario were frustrated. They

00:15:06.100 felt like they hadn't learned much because they weren't performing that well. Whereas the people

00:15:08.980 who saw similar scenarios over and over and sort of internalized those procedures got better and

00:15:13.760 better. And then when they bring them back and show them situations, they've never seen before

00:15:18.220 the people who were frustrated in training and had this broad training destroy the people who were

00:15:24.720 doing this, who were sort of mastering specific scenarios over and over. And these are again on,

00:15:29.840 on situations that none of them have ever seen before. And so the breadth of their training predicted

00:15:35.180 how well they could transfer their skills to totally new situations. And that's sort of the theme

00:15:40.180 in the more wicked end of the learning environment. So I was just reading some LinkedIn research you

00:15:45.060 mentioned in business that looked at half a million members and what was the best predictor of someone

00:15:50.660 going on to become an executive. And it was the number of different job functions they had worked

00:15:55.640 across in their domain. Actually, if they'd gone to a top five MBA program was almost as about as

00:16:02.500 influential, but that they couldn't tell if that was because of the school or just because of the

00:16:06.040 student selection or whatever. But in terms of things that were more in people's control, the number of

00:16:10.540 different job functions they had worked across. And I think that's a similar finding, right? It's their

00:16:15.060 breadth of training gives them the ability to manage situations they've never seen before.

00:16:21.020 Well, I'd like to talk about some research you highlight in the book that I thought was really

00:16:23.300 interesting that highlights the need of more breadth in your thinking and your skill acquisition.

00:16:29.680 And you talk about the Flynn effect, right? And this is the idea that over the past few decades,

00:16:34.620 IQ scores have been going up every year. But they're trying to figure out like, why is that

00:16:40.200 happening? So why, why is it that we've been getting better at IQ tests? Like, are we getting

00:16:45.220 smarter or is it just the way that we think is changing that make us do better IQ tests?

00:16:50.980 Yeah, it was, it was over most of the course of the 20th century, a rise of about three points

00:16:56.740 per decade. And that's so much that like our great grandparents would look handicapped compared to us

00:17:03.320 if the tests weren't re-standardized. The tests are always re-standardized. So a hundred is,

00:17:07.400 is the average. And we're not like, we don't have better brains than them, right? So the question

00:17:12.540 for psychologists who saw this pattern is, well, why are people getting more questions right? And not

00:17:18.920 only getting more questions right, but getting more questions right in the places where the test was

00:17:24.840 least supposed to change over time. So the most abstract questions, so that the IQ test that showed the

00:17:31.000 biggest change over time was one called Raven's progressive matrices, where you just get this,

00:17:35.600 this was designed to be what's called a culturally reduced test. So like nothing you learn in life

00:17:39.960 or school should affect your performance. So if Martians landed on earth, this is the test you'd

00:17:44.200 give them and it could show how clever they were because it doesn't involve any learning. And it just

00:17:48.180 involves these abstract patterns and ones missing. And you just have to look at the ones that are there

00:17:52.600 and try to fill in the missing one. And scores rose extremely rapidly on that test where it was,

00:18:00.200 where the least change was expected. And it turns out that that's the case. Even if you look at the

00:18:05.640 more concrete tests, we aren't doing that much better on specific subjects, you know, vocabulary

00:18:09.940 and that stuff. But wherever there are more abstract questions, people are doing much, much better. Even

00:18:16.340 in cases where test scores in some countries have gone backward on specific learning in things like math

00:18:22.940 and vocabulary, they've still improved on these more abstract questions. And the evidence suggests that

00:18:29.940 that's because as we moved from a less, from a world where we were less focused on the concrete in front of us and

00:18:36.640 focused on our experience, where we move to a world where work is much more interconnected, it's much more based on

00:18:43.900 knowledge that you have to transfer. Like we get by by transferring our skills to different situations all the time and to

00:18:50.580 different jobs. And we take that for granted now. But that's not something that people 100 years ago were as

00:18:58.380 capable of. The world didn't demand it of them as much. They could be much more comfortable sort of staying in a

00:19:04.240 very narrow lane of knowledge and repeating known tasks and procedures. But as the world has become more

00:19:10.400 complex, we have adapted to that by becoming better at abstract thinking. Which means it's not that one type of

00:19:16.900 thinking is better than the other per se, but we're much more adapted to an environment where we can

00:19:21.780 laterally transfer our knowledge to totally new tasks very effectively.

00:19:26.280 Well, can you give us an example of like, say, an abstract question you would if you ask, say, you know,

00:19:31.040 someone in 1875, they would have like a hard time, you know, answering it because they would be thinking in

00:19:38.780 concrete terms. Yeah. So there are, for example, in, in range, I write about this through a Russian

00:19:47.340 psychologist who goes, who, when the Soviet Union is undergoing, you know, socialist revolution and they

00:19:54.340 are nationalizing this remote farmland in what today is Uzbekistan. And the, so these people who've

00:20:02.840 been subsistence farmers and been able, had to be very, very good at the things they know, but didn't have to

00:20:07.600 know much else are suddenly being connected to the rest of the world. And some of them are so, so

00:20:13.780 they're having to manage work with other people, not just with themselves. And when you ask them

00:20:17.980 things, first of all, if you ask them to classify things like objects or colors, they basically are

00:20:24.840 unwilling to do that. Whereas some of the people who have had some exposure to modernity, they can

00:20:30.420 classify, like if you give them shapes, they'll always liken it to an object, right? So if it's a

00:20:36.600 circle, then it's a coin. And if it's a dotted circle, then it's a watch. Whereas the people who've

00:20:41.200 been exposed to some modernity can classify all the different types of circles together in a group,

00:20:45.860 even if they don't know the word circle, they'll recognize that there's some abstract commonality.

00:20:51.520 And that sort of thing shows up all the way up to sort of formal logic, where if you ask

00:21:00.560 some of these more pre-modern people, for example, you know, who are not working in the world of

00:21:06.640 connected work, things like they'll say, well, cotton grows well where it's hot and dry. In

00:21:12.680 England, it's cold and wet. Does cotton grow well in England? They'll basically refuse to answer and

00:21:20.860 say things like, well, you'd have to ask someone who's been there. In the case where those questions

00:21:25.980 were asked of people who were experts in cotton growing, if you really push them, they might say

00:21:30.760 like, well, it probably shouldn't grow there if it's cold and wet. So you can kind of push them.

00:21:35.660 But if you use a question that uses the exact same logical structures, but something they're

00:21:40.780 unfamiliar with, they can't answer it. So one of the questions that Luria asked was this logic

00:21:47.140 problem where he says something like, where it's cold and there's snow, all the bears are white.

00:21:53.660 And then he says like, in Novaya Zemlya, this town in the far north, there is always snow. What

00:22:00.220 color are the bears there? And they can't answer it. They'll say like, I don't know, I haven't been

00:22:03.940 there. Or you'd have to talk to someone who's been there. And so they're unable to transfer

00:22:08.580 knowledge to situations they've never seen before, which again, doesn't mean that their thinking is

00:22:13.340 worse per se, but it is not well adapted to the kind of transfer between domains that we're very

00:22:18.620 well equipped to do today. Okay. So most of the, the, the thinking we do in the modern world,

00:22:23.740 if you live in a Western modern world is abstraction, right? Like a lot of our work,

00:22:27.800 just abstract, even, you know, the way you interact with the world, like a video game is an

00:22:31.460 abstraction, right? You understand it's not actually a, you're playing Red Dead Redemption.

00:22:36.120 You're not actually riding a real horse. It is a, you know, pixels of a horse. So it helps us think

00:22:40.900 like that. So despite this more abstract world, how do we school? Like, are we educating kids and

00:22:47.100 young people to do well in this abstract world? Or do we kind of go back to that very like concrete,

00:22:53.540 like you need to learn these, these vocab words and you need to know this information, et cetera.

00:22:58.020 Yes. That's a, that's a great question. And to pick up on, on what you mentioned about Red Dead

00:23:01.560 Redemption, not to go backward, but just for a sec, we are so accustomed to abstraction, so good at it.

00:23:08.360 We don't even, we totally take it for granted, right? Like when you are, whatever, like downloading

00:23:14.440 the latest flash update or whatever it is on your computer and you see some bar that's like filling

00:23:18.760 up a progress bar to a hundred percent, that's a massive series of abstractions. That bar is some

00:23:25.580 sort of representation of time, which itself is some representation of this huge number of underlying

00:23:32.200 instructions that the computer is carrying out, which itself is an abstraction. That's actually just a

00:23:37.240 bunch of zeros and ones that the computer is using, right? And so there's all these layers

00:23:40.820 of abstraction. So computer programmers do really well on cognitive tests of abstraction

00:23:44.440 because they have to come up with stuff like this. And so I think in the sense, in one way,

00:23:49.240 I think we're doing a good job preparing people for this world in the sense that if you think about

00:23:55.460 video games and you think about the way we deal with apps and computers now, we've gotten really

00:23:59.220 used to doing things without instructions and trying to deduce the rules and just start learning

00:24:05.340 something by using it, right? This is what Robin Hogart, the kind, wicked psychologist, he says,

00:24:10.820 nevermind golf and tennis. Most of us in the world are playing Martian tennis where there's something

00:24:15.380 going on, but you don't know the rules. Nobody's told you and they're subject to change without

00:24:19.140 notice and you just have to deduce them and they could change at any time. So you may have to deduce

00:24:22.840 them again. So I think the world at large is doing a decent job, but school-wise, not great. I think

00:24:29.500 if you look at, people always complain about the school system today compared to yesteryear. And if

00:24:35.380 you look at tests of basic skills, without question, students now do way better than our parents did.

00:24:42.280 No question. The problem is the challenge has gotten much more difficult because our economy is so based

00:24:49.540 based on transferable knowledge, the improvements aren't keeping pace with what's really sort of

00:24:55.160 needed in the world, basically. Like if you look at tests that were given a generation ago to sixth

00:25:00.300 graders, it's very procedural, very procedural. And if you look at them now, it requires a lot more

00:25:06.580 abstract thinking. And so I think there's some movement in that direction, but it's not fast enough.

00:25:11.580 And what James Flynn, the discoverer of the Flynn effect describes in range is this problem where

00:25:17.820 teachers and professors tend to lean toward didactic information, filling people with information or with

00:25:25.760 procedures on how to do things because it's easy to conceptualize that. It's easy to teach and you see

00:25:31.820 immediate progress. The problem is it doesn't build these more fundamental general skills that allow you to

00:25:38.860 then better learn anything later on. And so he was sort of, he describes in the book, this study he did

00:25:46.540 looking at how well grades in college at an elite college corresponded to the ability of a student to do well

00:25:54.700 on like a really important abstract thinking test that tests the kind of critical thinking that you need in the

00:25:59.340 world. And the correlation was zero between grades in college and the scores on that test, which suggests

00:26:04.940 that there's a real disconnect. And, and if you look at some of the data he's collected, it suggests

00:26:09.900 that the work world is doing a decent, is, is having a larger input into making people better abstract

00:26:16.780 thinkers than school is. So I think we're missing an opportunity there. We're going to take a quick

00:26:20.620 break for your words from our sponsors. And now back to the show. Well, and we can keep talking about

00:26:26.700 school because you have a chapter about approaches teachers take to teaching students. And one way that

00:26:33.500 when you sit in on a class and you listen to a teacher interact with a student, you see this

00:26:38.240 interaction take place where the teacher is asking questions, but also giving hints and help trying to

00:26:45.500 help the student like feel some success early on so they don't get discouraged. But what's the study?

00:26:51.360 What does the research say on that, that sort of teaching method?

00:26:55.460 Yeah. So, so similar to some of the sports research, what it says and what you're describing is this,

00:27:01.640 this scene from this international math study that, that opens chapter four, where this really

00:27:08.480 charismatic teacher is, you know, teaching math to young students. And what happens over and over

00:27:17.380 is the students sort of learn to play a form of multiple choice where she'll pose a problem to them

00:27:22.180 that might require some abstract thinking or some conceptual thinking, or it might not,

00:27:27.900 but either way, the way that they interact with the teacher will cause her to give so many hints that

00:27:32.440 they can always turn what's called a making connections problem, where you're forced to

00:27:36.400 connect different ideas into a using procedures problem, where they just figure out rules that

00:27:40.640 they can apply over and over. And when you do using procedures practice, which you need some of

00:27:46.740 for sure, but the problem, the good thing about using procedures practice is you see progress right away.

00:27:52.660 The bad thing is it undermines future progress because you're not learning how to connect ideas

00:27:58.760 in a much larger system. And so the countries that do better in math education, like Japan,

00:28:07.340 instead of having practice over and over and over again in to practice procedures on problems,

00:28:13.920 you go into a classroom there and I've been there and seen this and the whole class period might be

00:28:19.780 one problem where there's a huge blackboard that's the size of an entire wall. All the kids have name

00:28:24.380 magnets. The teacher will do one problem that can connect a whole different, a large number of

00:28:29.380 different concepts. At each stage, he asks students for ideas. They come up, write their idea down. They

00:28:33.740 stick their magnet next to it. It may be right or wrong. And then other students try different ideas.

00:28:38.060 And at the end of the class, you have this like captain's log of the intellectual journey of the

00:28:42.440 class as they connected a bunch of different ideas and went through false starts in one problem.

00:28:46.600 And so Japan actually has a word that means this kind of writing on a blackboard that connects ideas

00:28:53.620 and tracks the thinking through the class is called bansho. And Japan does much better using these making

00:29:00.560 connections kinds of questions, whereas the U.S. is much more focused on using procedures. And that's just

00:29:07.700 like golf. Using procedures is fine as long as you're going to face the exact same problems over and over

00:29:11.680 again. But what you really want is not to teach the students how to use procedures, but to teach

00:29:17.080 them how to pick the right strategy for a type of problem. And that's a totally different thing.

00:29:22.460 And this actually relates, I don't want to go on too long on this, but this relates to what I thought

00:29:25.280 was kind of the single most surprising study in the book for me, if you want me to share that one.

00:29:31.280 Yeah, go ahead.

00:29:31.900 So the single most surprising study to me was this one in chapter four done at the U.S. Air Force

00:29:37.680 Academy, because I don't think you could set up this. There would be no other way to set up this

00:29:41.440 study. And what happened was these researchers wanted to look at the impact of teaching on student

00:29:47.260 achievement over time, not just in a single class. And the Air Force has this incredible setup for this

00:29:52.460 experiment because it brings in students every year. They all have to take a certain sequence of math

00:29:58.080 courses and they are randomized to their professor in the first class. And they all have to take the

00:30:04.020 exact same test. It's created the exact same way. And it's basically the whole grade for the course.

00:30:08.660 And then they are re-randomized the next year to the follow-on course, Calculus 2 or whatever it is.

00:30:13.180 And then they are re-randomized again the next year. So you get these multiple steps of randomization

00:30:18.620 and the students' abilities coming in are spread evenly across these classes. You can really see the

00:30:23.960 effect of professors. And what the researchers found was that the teachers who were the best at

00:30:31.140 promoting good test scores in their own class systematically undermined the future performance

00:30:37.940 of their students in future classes. So those teachers would teach more narrowly because they

00:30:44.380 knew what the students had to learn to do well on the test. The students would rate those teachers

00:30:49.520 really well because they could see instant progress. They would do well on the test. And then they would

00:30:54.060 underperform in all their future classes. Whereas the teachers whose students rated lower because they

00:30:58.500 were more difficult, they taught much more broad concepts, connected ideas. Those students often

00:31:04.300 struggled on the Calculus 1 exam and then overperformed in all the subsequent courses, which is really

00:31:11.560 counterintuitive. And we're not programmed to think that we could be making progress before our eyes and that

00:31:16.640 could be undermining our long-term development. But again, it's the same thing you see in the sports

00:31:21.000 literature and the math learning literature and in a whole bunch of other domains that I talk about

00:31:25.800 in range. Well, it's like that Navy study, right? The group that learned the procedure on that, how to

00:31:30.460 handle a certain situation, a certain war game. They did well initially, but when they got put in

00:31:34.660 something different, they just, they got tanked. Exactly. So if you want somebody, if your goal is for the

00:31:41.800 individual to be able to respond well to situations that aren't exactly like something that's come

00:31:47.360 before, what you want to do is, is help force them into an environment where they have to learn to

00:31:53.600 connect ideas and build these, these abstractions that they can fit to new situations as opposed to

00:31:59.060 giving them procedures that they know how to execute as long as they're seeing exactly something

00:32:03.740 they've seen before. And I imagine there's like, kids are really good at finding procedures. So if you

00:32:08.080 have, if you're a parent with a kid in school and you notice that, man, they're doing really well

00:32:12.220 all the time, well, it might be because they just figured out a procedure. Like they have a knack for

00:32:17.320 finding the procedure and they're just following that procedure and it might not benefit them in the

00:32:21.440 long run. Yeah. I mean, one of the things that cognitive psychologist Nate Cornell says in that

00:32:26.300 chapter is that ease, difficulty is not a sign that you aren't learning, but ease is a sign that you

00:32:34.420 aren't learning. Uh, so if something is too easy for you, then you're not really learning. It's when

00:32:39.820 progress is made too fast. That should actually be kind of a warning sign. The problem is we've set

00:32:45.200 up our evaluation systems, you know, and we're just oriented. It's more intuitive just to be oriented

00:32:49.800 toward before your eyes progress. Like, you know, what could be bad about that? What could ever be bad

00:32:54.880 about a headstart? But it, it turns out that whether it's in math or sports, um, the way to develop the

00:33:02.700 best 10 year old is not the same as the way to develop the best 20 year old. Okay. So if we want

00:33:09.560 to excel in this, this world of abstraction, where you're going to be faced with problems you've never

00:33:13.560 encountered before, we want to get a broad range of skills, a broad range of things. We can make

00:33:19.160 connections. Let's talk about some ways we can do that. Um, you give this great example. I never

00:33:23.740 heard of this group. It's, uh, the Vigile de Coro. Did I say that right? It's that it's, this is a tough

00:33:30.220 one. It's technically pronounced Filia. Filia. Filia de Coro. This was like. It's spelled Figli

00:33:35.020 del Coro, but it's pronounced Filia. It's Italian. Yeah. So this was a group of musicians at a,

00:33:40.300 basically a convent, right? In France, like in the 1700s where they became world renowned and famous.

00:33:46.460 And it was kind of counterintuitive how they got to that point. Yeah. So in, in Italy, they were in

00:33:51.240 Venice and they were in, in the 17th and 18th centuries and, and not just musicians, but they,

00:33:58.000 the Filia, which means that just means daughters. Filia del Coro means daughters of the choir

00:34:02.100 technically. And they were actually basically, so Venice had a, had a very vibrant sex industry

00:34:07.740 at the time. And that led to a problem, which was babies, especially baby girls who didn't have

00:34:16.960 fathers would end up in the canals sometimes. And this was recognized as a social ill that needed to

00:34:22.960 be fixed. And these institutions were established. They weren't technically convents. They weren't,

00:34:25.840 they weren't technically religious, but they sort of had quasi monastic rules that they ran by and

00:34:30.520 they were attached to churches. And basically they had like the, the most famous one called,

00:34:38.420 they were called hospitals, but they weren't really hospitals because the Ospedale del Pieta,

00:34:42.400 the house of mercy, basically it had like a, like a luggage check thing. You know, when you go to an

00:34:47.300 airline and if you can fit your carry on in that check thing, then you can bring it on. They had one

00:34:52.040 in the door where if you put the baby and the baby fit in there, you can ring a little bell and

00:34:57.020 they'll come pick it up and they'll raise it. No questions asked if it fits in the carry on luggage

00:35:01.300 thing and that they called the scafetta. And they would try to raise these girls to become

00:35:07.720 self-sufficient. And the, the Ospedale was its own like internal economy and all these things.

00:35:12.960 And at a certain point people started, they would pay the girls as they'd want to make themselves

00:35:16.940 sufficient for learning different skills. So they'd give them a little money reward.

00:35:19.900 And at a certain point people started donating used instruments there. And the girls would realize,

00:35:25.120 Hey, you know, they can, one, it's fun. And also it counts as new skills if they learn a different

00:35:29.980 instrument. So they would start trying to learn a whole bunch of different instruments. Some of

00:35:33.900 these instruments, musicologists no longer even know what they were because they were these like

00:35:37.500 experimental instruments. And by in, in the course of learning this large number of instruments,

00:35:44.960 the people who ran the Ospedale started to notice that they could pick up anything really,

00:35:49.740 really quickly. They had built these sort of models of how to learn music. And so they started having

00:35:54.720 them perform in the adjoining churches. And they were so good that money started pouring into the

00:35:58.960 Ospedale. And so they started having more performances. Antonio Vivaldi, who the composer

00:36:03.940 of the four seasons, which is probably like now the most famous, arguably like the most famous,

00:36:09.800 it's like almost a pop hit and it's 300 years old, basically. Right. There's like a mashup with

00:36:13.620 the song from Disney's Frozen that has a hundred million views on YouTube or something.

00:36:18.000 And they became Vivaldi's muses. He recognized their ability to pick up anything, to pick up new

00:36:25.400 forms of music. And they became his muses and he sort of became their composer and they became

00:36:30.820 the greatest musicians in the world. These orphans of the Venetian sex industry, whose training

00:36:38.420 consisted of attempting to learn as many different instruments as they possibly could, which equipped

00:36:43.940 them with this ability to learn entirely new types of music and new instruments like on a whim.

00:36:50.640 And for a hundred years until Napoleon came in and his troops took over Venice, they were the greatest

00:36:55.840 musicians in the world.

00:36:57.100 Yeah. I mean, one of them, I think you have an example of this one. She got too old to play a

00:37:00.340 particular instrument. So she like just, okay, I got to play this instrument now. And she was able to do

00:37:04.340 it really easy. Like the transition was smooth.

00:37:06.740 Yeah. Well, her teeth, her teeth fell out. So she couldn't play the wind instrument. She'd played

00:37:10.460 anymore. So she just switched to one of the other ones and, and, and kept performing.

00:37:13.740 So these are, these are people who didn't get like that headstart training where they're, you know,

00:37:17.400 the mom and dad was having them sit down and do, you know, scales, you know, from morning till night,

00:37:22.980 they were just messing around with these instruments. And then you also like jazz musicians were the same

00:37:27.240 way as some of the greatest jazz composers. Most of them didn't get any formal training. They just kind of

00:37:32.580 picked up a guitar and then figured out how to play it. And they came up with this really complex

00:37:36.960 new ways of playing music.

00:37:39.460 Yeah. Two, two different jazz people that I interview in the, in a chapter on music, one,

00:37:44.280 two jazz players and instructors told me the same joke one time when I was interviewing them.

00:37:48.480 And it goes like this, you, you, if you're a jazz musician, you know, you ask one of the guys

00:37:54.020 you're about to play with, if they can read music and they're supposed to respond, not enough to hurt my

00:37:59.040 playing. And a lot of them do in fact, learn how to play music, but they learn, they, they,

00:38:06.180 they do what I call learning like a baby. So when you learn language, you don't, you don't learn the

00:38:12.460 grammar first, right? You learn the sound, you get thrown in, you're immersed, you, you struggle,

00:38:18.080 you fail, and then you learn the grammar much later, if at all. And that seems to be the way,

00:38:24.140 so that the sort of 10,000 hour school has focused very narrowly on a certain type of

00:38:28.660 classical playing. So I mentioned in range, the Cambridge handbook of expertise, which is like

00:38:33.620 the Bible of the 10,000 hour rule school has entire chapter on music. And it's all on a very

00:38:39.480 particular type of classical music. And then there's just like one offhanded mention where

00:38:43.960 it's like in jazz and modern popular music, it seems like actually the people are much broader

00:38:47.820 and start later and sample more stuff. And then it just goes back to, so it basically just gives

00:38:51.680 that short shrift most of the types of music that everyone listens to and plays now.

00:38:55.700 And those musicians develop much more along the Roger model than the Tiger model, where they sample

00:39:00.980 different instruments. Their, their practice doesn't explode until they find an instrument that

00:39:05.760 fits them. So it, if you look at this research over time, it's not that the musicians who end up

00:39:11.080 practicing a lot, who are exceptional, generally are just practice-aholics. It's they go through more

00:39:17.460 instruments than their peers. So some of their peers will stick with their first instrument,

00:39:22.600 even if it's not a great fit, they feel like they have a headstart and can't switch. Whereas the

00:39:26.680 students who go on to become exceptional, they sample instruments until they find one that they

00:39:30.520 think is a fit and then their practice volume explodes. So it looks more like, less like they

00:39:36.360 personally are just practice fanatics and more like they're maximizing what, what economists call

00:39:41.100 their match quality, the degree of fit between who they are, their abilities and interests

00:39:44.520 and, and what they do. And that, that seems to be the rule for most of music development with

00:39:49.580 some, some exceptions. Well, let's talk about that idea of fit because people have this idea

00:39:54.820 that, you know, particularly if you're a young person, you're in college, you're, you're, you're

00:39:59.200 20 years old, you got to pick a major. You have to know whether you're going to go to medical school

00:40:03.420 or law school, even if you don't even know, like if you like medicine or law, and then you just have

00:40:09.580 to grind it out until you get there and you might find out, man, I just don't like doing this. Like

00:40:15.120 why do we overlook that idea of fit? Um, like you're finding something that we're actually good

00:40:19.720 at. It's interesting. You mentioned that because Theodore Schultz, a Nobel laureate economist

00:40:24.340 sort of chastised his field for overlooking fit. Like we had studied higher education and saw that,

00:40:32.100 okay, it makes people more productive by giving them some skills. But he said, we've overlooked

00:40:36.320 studying the effect on match quality, which is that because people come out of high school,

00:40:41.780 they know very little about the world. They know relatively little about themselves. It's all,

00:40:45.360 you know, everything they know is constrained by their very, their, their roster of previous

00:40:49.260 experiences. And what about when they get to have a sampling period like the athletes, does that

00:40:54.560 influence how good they become and how well they fit with what they're doing? And the answer is it

00:41:01.820 absolutely does. So in range, I discuss research on an economist who found sort of a natural

00:41:08.340 experiment by looking at school systems where people are made to specialize at different times,

00:41:12.380 you know, some in their teen years, and then some can delay until later. And what he finds is that the

00:41:16.440 people who specialize earlier jump out to an income lead, but by about six years into their work,

00:41:22.040 the later specializers catch and surpass them. And the earlier specializers start quitting their

00:41:27.200 careers in much higher numbers because they didn't have enough time to sample and figure out a good

00:41:32.000 fit in the first place or to optimize their match quality. And so there's of course, nothing wrong

00:41:36.820 with getting a law degree or getting a medical degree. But I think the research suggests it's

00:41:41.840 actually a dangerous thing to decide to do before you really know if that's a good long-term goal for,

00:41:47.040 for yourself. So you need a little, little sampling to maximize your match quality. And all the research

00:41:52.920 in this area suggests that when people change careers and jobs, they're set back a little bit

00:41:57.560 in certain specific skills, but their growth rate becomes much higher because each time they do it,

00:42:02.320 they are responding to match quality information and optimizing their fit with what it is that they're

00:42:07.060 doing. Well, going back to like the sports gene, this made me think of like, you know, why some

00:42:11.780 athletes do well in certain sports is because their body is fit for that sport. Like Michael Phelps,

00:42:17.240 you know, he, of course he works hard, right? But like his body's designed for swimming.

00:42:21.420 But if he decided I want to play whatever sport and just like grind it out,

00:42:25.640 he might've been okay, but he probably wouldn't have been reached the level as he did in swimming.

00:42:30.260 Right. I mean, he obviously has these traits like, you know, willingness to practice and all these

00:42:33.460 things. But again, that usually follows match quality, not preceding it. So there's no evidence

00:42:38.420 that he would have been a fanatical trainer if you were a runner, for example. And that's a lot of

00:42:43.580 research I discussed that suggests that once someone finds a fit, it just looks like they're a

00:42:47.460 fanatical trainer or like that's their personality. But like, if you look at the guy who's the world

00:42:52.600 record holder in the mile, who's seven inches shorter than Phelps, they wear the same length

00:42:56.220 pants because their bodies are conducive to different types of performance. And I think

00:43:00.780 one of the many reasons why we see that Roger pattern in the development of sports is that as

00:43:05.840 we push selection earlier and earlier, people aren't even biologically mature yet. And the more,

00:43:10.800 the earlier you push selection, the more in anything, the more likely you put the wrong

00:43:15.540 person in the wrong thing. And that's also why in sports, we see the so-called relative

00:43:20.160 age effect where unlike junior national teams, you see this incredible concentration of kids

00:43:26.060 who are just born early in whatever the selection year is because they're more biologically mature

00:43:29.820 than their peers and their coaches mistake that for their potential. And then that effect

00:43:34.680 disappears at the elite level. So, which suggests to me that it's a really bad system and causes

00:43:39.500 us to deselect a lot of people who would potentially go on to become elite.

00:43:43.100 Right. And so this dabbling period, it allows you to find what you're good. And you said like,

00:43:46.700 you know, these guys who in sports, they dabble, they eventually find their fit. Like Michael Phipps

00:43:51.500 finally finds swimming. Roger Federer finally finds tennis. But this can also apply, you know,

00:43:56.120 to just like personality or your brain. Like, I mean, I think they've done studies like the human

00:44:00.640 brain isn't fully formed until like 25, 26. So you might be almost a completely different person

00:44:07.200 when you were, when you're 26 than when you were at 18, where you, when you were deciding your,

00:44:12.320 your life trajectory.

00:44:14.100 In fact, you're getting an important concept in range called the end of history illusion.

00:44:18.760 So this is the psychological finding that when we all look backward, we say, oh gosh,

00:44:23.620 we really changed a lot because of all our experiences in the past. But we think that we're,

00:44:27.960 we're mostly done or we won't change that much in the future. And it turns out that we're wrong at

00:44:32.060 every stage of development. So personality, the, the, the time of most rapid personality change is

00:44:38.460 from about 18 through your late twenties. And so if you're picking what you're doing at age 16 or 17

00:44:44.200 or 18, you are absolutely selecting a career for someone that you don't even know yet that the

00:44:49.220 correlation between someone's teen years and middle age for particular personality trait for the

00:44:54.040 statistically inclined is usually like 0.2 or 0.3, which means there is absolutely still

00:45:00.140 signs of the previous you in the later you that are distinguishable, but you are a very different

00:45:05.660 person. That's like a low, low, moderate kind of correlation. So you, we all change more than we

00:45:10.340 expect, which leads to some really funny results, by the way, like when people are, people think

00:45:14.360 their preferences will stay the same. So when they're asked how much they would pay to see their

00:45:20.180 favorite band today in 10 years, 10 years from now, the $129 is the average answer, but asked how much

00:45:26.160 they would pay to see their favorite band from 10 years ago today, the average is only $80.

00:45:30.140 Because we underestimate how much our preferences will change. And the way that I discussed this

00:45:35.320 in range is that because we underestimate personality change, we really need to be ready

00:45:40.080 to adjust a lot going forward. Cause we're facing the challenge of how to behave when we don't know

00:45:48.040 the future us or the future world that we'll be living in. And what's the best approach to take

00:45:52.920 when you're facing that problem. So basically if you are 22 years old, listening to this podcast and

00:45:57.720 you, you're worried, you don't know what you're supposed to be doing with your life. Like that's

00:46:00.740 okay. You're going to be fine.

00:46:03.020 Yeah. I mean, I think there's a great, you know, the, the Y Combinator invest investor,

00:46:07.660 Paul Graham, I excerpt a little bit of a graduation speech that he wrote, but never gave in the book

00:46:12.900 where he's, he basically calls this idea that you should have a long-term plan and know exactly

00:46:17.000 what you're going to do. He says, computer scientists call this premature optimization.

00:46:19.920 Like all the, he says, all the successful people he knows take this approach of short-term planning,

00:46:25.540 looking at the opportunities in front of them, taking them and being ready to adjust.

00:46:28.960 And he was saying that just based on his, his experience with startups and people he knows,

00:46:34.860 but that turns out to be like the seminal finding of this research discussed in range called the

00:46:41.220 dark horse project. So these Harvard researchers who wanted to study how people find fulfilling

00:46:45.060 careers and their surprising finding was that those people are systematically averse to like

00:46:52.400 rigid long-term planning. Their, their main common trait is short-term planning. So what they do is

00:46:57.820 they say, here's who I am right now. Here are the skills I have here, the things I want to learn,

00:47:02.340 the interests I want to explore here, the opportunities in front of me, here's what I'll do.

00:47:05.700 And maybe a year from now I'll change because I will have learned something different and see a

00:47:09.480 better fit. And they just do that. And they zigzag their way to a place where they uniquely can

00:47:14.220 succeed and feel fulfilled. And the reason this study was named at the dark horse project,

00:47:18.800 that wasn't what it was called initially was when they were studying these people who find

00:47:23.220 fulfillment, all of those people would come in, you know, in their initial interviews and say like,

00:47:27.840 well, I know you're trying to tell people how to go about finding a career, but like,

00:47:31.140 don't give them the advice to do what I did because I changed paths a whole bunch of times.

00:47:35.140 And like, you know, I didn't, wasn't one for long-term goal setting. And they found like 90% of the

00:47:39.860 people say that they're like, well, I'm not a good model, but in fact, that is the good model.

00:47:44.020 But they think they're all oddballs or exceptions because it's hard advice to give, right? Like

00:47:49.660 when I was at Sports Illustrated, I would get contacted by students all the time asking,

00:47:54.360 I want to, if I want to be a sports writer, should I major in English or journalism?

00:47:57.660 And I, I, I studied geology and astronomy, so I have no idea, but I was still inclined to give

00:48:03.080 them the advice that was like, you know, start doing your journalism internships right now,

00:48:07.200 even though the source of power for me that allowed me to become the youngest senior writer

00:48:12.040 there was the fact that I had this science background. It was totally average when I was

00:48:16.300 a science grad student, but totally exceptional when I was at a sports magazine, but it's hard

00:48:20.720 advice to give, you know? So we've been kind of dogging deliberate practice, uh, against dogging

00:48:26.820 it. Don't starting, starting it too early. You don't want to study it too early. So the idea is like,

00:48:30.520 you're going to dabble for a couple of years, you know, maybe into your early twenties. And then once

00:48:35.140 you find something, that's when you start seeing these high performers starting the deliberate

00:48:39.140 practice, right? Yeah. I mean, there's nothing deliberate practice is great, right? You want

00:48:42.740 to do deliberate practice, I think. And this is, and Malcolm Gladwell and I were just invited back

00:48:47.540 to the MIT Sloan conference recently. And what he said, and again, this is on YouTube. He says,

00:48:54.060 I've changed my mind. I think the fact that to be great requires a lot of practice. I thought that

00:49:01.900 implied that you needed to focus narrowly and start early. And now I feel differently. And I think that

00:49:07.120 was a, that was, I thought what he was saying too, you know, and my ideas evolved since starting with

00:49:13.220 my first book, but, but even more so now. And I think it's the implication that that means that you

00:49:18.240 should, to be great at X, you should only do X as early as possible is not supported by, by the

00:49:23.400 research. So you still need to do a bunch of practice, but I think even after the dabbling, you

00:49:27.440 should keep career streams open. So there's early in range. I mentioned research on people who go from

00:49:34.540 becoming great performers, essentially, whether that's like an athlete or a musician or a surgeon

00:49:39.440 or whatever, to being people who are great at like running an orchestra or managing a sports team or

00:49:44.960 running a hospital. And one of the features of those people, again, it's just another finding of

00:49:49.860 the breadth of training predicts breadth of transfer is the, the scientists studying this says they keep

00:49:54.540 multiple career streams open. He says they're, they're traveling an eight lane highway instead of a

00:49:59.420 one way street. And so even once they get more focused, they kind of keep other interests around

00:50:05.380 a little bit and eventually they're able, better able to transfer into these, these management

00:50:10.240 positions. Um, so I think even there, this, this breadth continues to be important. And so you

00:50:15.040 shouldn't get totally, totally blinkered. Even once you become, even though at some point or another,

00:50:20.280 all of us specialize to one degree or another, of course. Well, you even highlight research that

00:50:24.760 people who win Nobel prizes, cause they've specialized in one particular area of science or

00:50:29.060 mathematics, they are more likely to have like a hobby, like improvisation or painting or music,

00:50:34.940 uh, than just the average population. Way more likely they're, they're 22 times more likely

00:50:41.880 than other scientists to have a serious hobby that usually deals with like aesthetics, you know,

00:50:48.360 music, magic, writing, art, glass blowing, all these sorts of things, generally tinkering. So they're

00:50:55.380 much more like, so national level scientists who get inducted into national academies.

00:50:59.060 Are much more likely to have serious hobbies than the average scientist and the Nobel laureates

00:51:04.700 are much more likely still. And so one of the researchers, one of the phrases that I loved in

00:51:12.220 range was this researcher who studies scientific creativity called the network of enterprise.

00:51:16.400 They have a network of enterprise where they're doing all these different things that from afar might

00:51:21.340 look like it's diluting their thinking. But in fact, a lot of this stuff ends up informing

00:51:26.680 their ability to find problems that nobody else is looking at. So the, the father of modern

00:51:32.720 neuroscience, the Spanish Nobel laureate, Ramon Santiago y Cajal, I'm sorry, Santiago Ramon y Cajal.

00:51:39.420 He has a quote that I loved. There's something like from afar, it looks as if they are dissipating

00:51:45.540 their energies when in fact they are channeling and strengthening them. And it's striking to read

00:51:52.840 Nobel acceptance speeches in recent years, which I did a lot of and see that almost every year

00:51:59.300 a scientist who's accepting an award says, I, something to the effect of, I wouldn't be able

00:52:05.320 to do my research now because you have to be so narrow in looking for applications. So, and that,

00:52:11.600 that, that strikes me as something that's worrisome.

00:52:14.740 No, we can, we can talk about that in a minute. But like, I like this idea that these, these are

00:52:18.660 individuals who have gone deep in one area, but they continue that breadth. I think there's like

00:52:22.660 that, uh, guy from the design school IDEO calls them like T-shaped people, right? So they, they got

00:52:28.800 the vertical going up and down as the depth. And then they have like the, the, the horizontal part

00:52:33.260 of the T, which is breadth.

00:52:35.200 Totally. And, and there's, and that, that comes up in range in a section about inventors where this woman

00:52:40.400 who rose up to what's called corporate scientist at the company 3M corporate scientist is like

00:52:44.640 the highest title. She talks about how like, she's never, people kept telling her not to change

00:52:50.900 directions. And she's like, basically never worked in anything she was educated for. But what she does

00:52:55.840 is she knows how to sort of draw on her peers in order to assemble the I part of the T essentially.

00:53:03.980 She's very broad. She, she basically spends her time figuring out what everyone else is doing.

00:53:07.480 She's obviously very science literate. She has PhD, just doesn't work in her own area and makes her to

00:53:13.660 know what everyone else around her is doing and uses them to help sort of cobble together the,

00:53:19.160 the trunk of the T, but her contribution is more like the crossbar of the T and, and she and other

00:53:26.040 people like her have been able to, to use that kind of breadth with a grounding in a certain area

00:53:33.000 and also drawing on other specialists to, to be really high impact. And I think that's why you see

00:53:38.500 in some of the research I cite that the highest impact inventors are not the deepest specialists,

00:53:44.500 but the inventors who do have an area of expertise where they have some depth, but then they spread

00:53:50.680 their career over the largest number of different classes of technology as, as defined by the patent

00:53:56.240 office, basically.

00:53:56.940 I want to talk about one more thing with, in return to terms of fit, because I think it's

00:54:00.780 very counterintuitive and it goes against what, you know, you hear growing up as a kid. And it's

00:54:05.380 this idea that you should never quit anything, right? Once you start something, you got to see it

00:54:09.560 through. You have to develop grit, which there's been a lot of talk about thanks to Duckworth and her

00:54:14.140 research in her book, but this idea of finding fit, finding that, you know, dabbling until you find the

00:54:19.140 thing that you're, you're, you're good at and that you're, you're, that's made for you. That requires you,

00:54:24.020 you're gonna have to quit stuff. Yeah. I mean, I think the, one of the, one of the sort of underlying

00:54:31.820 messages of range is when you find fit, it will look like grit. And I think we've made a mistake

00:54:39.640 in the way we think about the study of personality, where we look at what people are now and assume

00:54:45.360 that's who they are, but, but we change not only over time, but with context. So I think an emerging

00:54:50.780 promising area of the study of personality studies, what's called if then signatures,

00:54:54.700 where you might say, if David is at a massive party, he's an introvert, but if David is with

00:55:03.800 his team, small team at work, then he's an extrovert. And it's turning out that personality

00:55:08.040 is much more complicated and that we look different in different scenarios. And the same for grit or

00:55:14.160 conscientiousness is, is it's, it's like the psychological construct of conscientiousness,

00:55:18.720 where if you, if you get fit, it'll look like you have grit, the same with the musicians,

00:55:22.700 right? Where they weren't practiceaholics until they found an instrument that really fit their,

00:55:27.240 their skills and their interests. And then they became practiceaholics. The study of this,

00:55:32.340 the famous studies of grit all involve subjects really short term. It's like people trying to get

00:55:39.760 through six weeks of physical rigors at the U S military Academy, or kids who are already in the

00:55:44.920 final of the national spelling bee and trying to get farther. The studies are always done on a very

00:55:49.600 select population of people who have a very narrow, well-defined, very short-term goal right in front

00:55:55.640 of them. And the problem has been, we've extrapolated that to all of life where it doesn't make as much

00:56:01.440 sense, where a lot of the research shows what you want to do is, you know, constantly be evaluating

00:56:09.400 your opportunities. And Steven Levitt, I quote him saying, well, one of his main talent, the

00:56:15.740 freakonomics economist, as people know him, one of his main talents has been to identify when he

00:56:20.360 should abandon a project or a whole domain of study and move to another one. And so that motivated him

00:56:25.720 to do a study of job quitting. And he found that when people are thinking about quitting, basically

00:56:30.300 they, they should, because they move on to something that's, that's a better fit. And Seth Godin,

00:56:35.780 you know, it's given some of the most popular career advice ever, I think says, not only should

00:56:40.480 you be willing to quit, not, not just because something's hard, right? You don't want to quit

00:56:44.280 just because something's hard, but when you start something, you should basically always have in

00:56:48.240 mind criteria under which you would decide to quit. And so I think while preaching grit is incredibly

00:56:57.140 intuitively appealing, you know, I, I critique the science of grit at a lot more length in range,

00:57:02.960 but I think it, the way that we've extrapolated what it means does not comport with a huge body

00:57:10.380 of research on how people find the areas where they can become the most productive and fulfilled.

00:57:16.160 Right. So you might not become world-class gritting yourself into something you're not a fit for,

00:57:20.260 but once you find you're something that, that, that is a, that you have, that is a fit for you,

00:57:24.560 the grit will could just come naturally, right? Cause you'll just want to do it. And then you can

00:57:27.960 become world-class because you're just sort of riding that wave of sort of internal motivation.

00:57:32.960 That's right. And you can think about, I mean, things that, that were sort of viewed by

00:57:36.820 parents as the total opposite of grit, maybe not so long ago, like playing lots of video games now

00:57:43.460 are careers where they've drawn enough interest from people that may not have that sort of grit in

00:57:51.240 other areas. Or, I mean, I was a college athlete, right? And there were, I think it's demonstrably

00:57:56.380 false that grit is just a stable characteristic in people. So there were people who were, I was a track

00:58:02.240 athlete, 800 meter runner. And there were guys I trained with who were absolute competitive demons

00:58:07.880 would claw your head off in a race who had not a sign of competitiveness, not a competitive bone

00:58:16.660 in their body when it came to the classroom and vice versa. And so I think we all recognize these

00:58:21.320 sorts of things, but we don't really, we don't really have like good shared language for talking

00:58:25.780 about them or we don't, we don't think very deeply about them.

00:58:28.060 All right. So yeah, again, like if you're a 20 something, don't feel bad if you're going to make

00:58:31.660 a change in your manger, right? Or you decide you're going to quit law school because after the

00:58:36.240 first semester, you may, you realize I don't like law. Or even if you're 30 or 40 and you decide to

00:58:40.700 quit your job, do something else, that's okay. It might actually, it probably will turn out better for

00:58:44.960 you.

00:58:45.760 Yeah. It's, it's, it's psychologically unsettling, right? And it's, it's, it's riskier as the

00:58:50.320 dark horse project researchers say, it's riskier to stick with that long-term goal before you really,

00:58:55.760 you know, have sampled enough to formulate a good one than it is to abandon it. Like we always feel

00:59:00.980 like we're in a rush, but I think we, you know, you notice as you get a little older, like you

00:59:04.360 weren't in as much of a rush as you think you are. And it's, it's well worth it to put in a little bit

00:59:10.480 of time investment in figuring out who you are and where you can make the biggest impact.

00:59:15.340 So I like, I do this going forward. I started what I call a book of experiments where there are things

00:59:19.620 like, and in 2018, one of my experiments was I spent some, some time volunteering and I wanted to figure

00:59:24.840 out where could I make the biggest impact and where would I learn the most? So I spread my time over

00:59:29.240 about a half dozen different organizations that year found the two where I think I can actually

00:59:33.180 contribute uniquely and also learn something. And now I focus in on those. And so I'm now based on

00:59:38.800 the research that went into the book, sort of constantly running experiments, setting up

00:59:42.780 experiences in a way where I have a hypothesis, the experience can help me test that reflecting on it,

00:59:47.880 and then just keep zigzagging and triangulating my way to a place where I think I can uniquely

00:59:52.560 succeed and feel good about what I'm doing.

00:59:54.960 So we've been talking about, so this idea of range on an individual level, you want to dabble in lots

01:00:00.280 of things so you can find fit. And that means you're gonna have to quit things. Once you're learning

01:00:04.300 something, you want the learning in the beginning to be hard, not easy, because if it's easy, it means

01:00:08.740 you're probably not learning. But I want to go back to something you said about, you read all these

01:00:12.800 acceptance speeches from Nobel prize winners. I want to talk about how this idea of specialization and sort

01:00:20.720 of downplaying the importance of range is affecting us on a broader scale. And you highlight research

01:00:26.460 or, you know, just ideas from Nobel prize winners that our specialization that we've been doing,

01:00:32.020 our hyper-specialization we've been doing in science and in other areas, it's actually preventing

01:00:35.940 breakthroughs from coming through. And they're actually diminishing over time.

01:00:40.000 Yeah. So you can see that even as funding has gone up, breakthroughs have not, basically have gone

01:00:45.460 backward. And you can also see this in outcomes. We, we sort of care more about even than breakthroughs,

01:00:51.700 right? Like we're the most medically advanced country in the world and life expectancy is going

01:00:55.760 backward. Same thing in the UK. And what these scientists highlight is, and some of them study

01:01:03.100 the science of science, right? Like how can we have good science get done? And what they find is

01:01:09.360 basically that you don't want to force people to be too narrow. You basically, we've created a system

01:01:15.780 where we're so focused on applications that we require people to narrow their research such that they

01:01:22.580 can quickly come up with applications. And that, and that's when they're applying for grants. And that

01:01:29.720 is pretty much exactly contradictory to the history of science and where impactful discoveries come

01:01:38.700 from. They typically come from someone who has some question they're curious about. There, there may

01:01:44.780 not be any real clear application. And just investigating that question leads to these huge

01:01:50.120 breakthroughs because the huge, the biggest breakthroughs tend to come where you don't

01:01:52.820 really know what the right question is. Right. And so you have to allow people to explore

01:01:58.580 pretty broadly. And the problem we have now as, as one of the, one of the characters in the last

01:02:06.160 chapter is this guy, who's, you know, arguably the most influential immunologist in the world.

01:02:11.200 And he's starting a program that's meant to de-specialize the education of future scientists,

01:02:18.100 because he says, what we have now is what he calls a system of parallel trenches where everyone is in

01:02:23.740 their own little trench. And they're not usually standing up and looking at what's going on in the next

01:02:28.800 trench, even though that's where their answer is. And so what he's seen in immunology is everyone's

01:02:34.420 studying one tiny part of a complex system in such isolation that we've failed to understand how

01:02:42.320 these systems work and connect ideas. And you can write a grant that is for the study of some

01:02:48.160 system of the body. And you can't even get it funded because the people reviewing it only know

01:02:51.840 about one little aspect and say, well, you know, we don't really, we don't really know about the rest

01:02:55.440 of that. So what we really, the world is divided up into disciplines, not because that's the way

01:03:00.320 we divide the world into disciplines, not because that's the way the world really is. It's a necessary

01:03:05.620 evil for just categorizing the studies we do. So we're in the place of putting the world back

01:03:10.840 together after we study things in individual disciplines. And so I think what he and other

01:03:15.040 scientists who are paying attention to this want to do is make interdisciplinary research and

01:03:22.320 interdisciplinary thinking systematic because the world is interdisciplinary. And we're, we're going

01:03:26.680 in the wrong direction on that as we force people to be more and more narrow. So they see a smaller

01:03:30.940 and smaller part of complex systems. But I mean, do you think there's hope? Like, is it, is that sort

01:03:36.160 of a growing idea or growing movement within science that we need to get interdisciplinary if we want to

01:03:41.060 make breakthroughs or people really entrenched in their, their little silo? I mean, I think there are

01:03:45.760 people like that scientist's name is Arturo Casadevall who are so prominent in their area that they are

01:03:52.820 bringing some attention to it, but I still think it's going against the grain. Like I went to a panel

01:03:57.120 that he was on about, there's a thing going on in science right now called the replication crisis.

01:04:02.820 And it has to do with a huge number of scientific findings turn out like not to be true, basically.

01:04:08.260 And part of the reason for that, Arturo argues, is because of this certain type of specialization

01:04:13.820 where people are doing science before they've really been taught the broader concepts of how

01:04:18.020 scientific thinking should even work. So they essentially don't know what they're doing and end up with

01:04:21.840 bad results. And by the way, in the book, I confess to the fact that when I was a science grad student,

01:04:25.900 I did the same thing, but I didn't realize it until much later when I was a journalist writing

01:04:30.000 about bad science, just kind of disappointing. And so Arturo's on this panel talking about how we need

01:04:35.600 to despecialize science education. And the editor of the New England Journal of Medicine, you know,

01:04:41.720 arguably the most prominent medical journal in the world says, no, you can't do that. Like there's

01:04:47.560 already, it already takes way too much time to, in education for students to become MDs or PhDs.

01:04:53.760 Like we can't add time with this broader conceptual stuff and having them do things that aren't exactly

01:04:58.700 where they're going to work. And Arturo's response was, yeah, get rid of all that didactic stuff that

01:05:03.640 we teach them that goes in one ear and out the other ear in two weeks anyway, and that they can

01:05:07.220 find on their phone. What we've got, he said, is a bunch of people walking around with all the world's

01:05:12.100 knowledge on their phone and no ability to integrate it. So his feeling was, you can get rid of that

01:05:17.360 stuff. Like our tools, our information finding tools have allowed us not to worry as much about

01:05:22.380 teaching that didactic stuff because we can find it. Meanwhile, we've skipped over teaching people

01:05:26.680 the broader concepts of how to even do science. And it's, it's helped land us in this, this

01:05:31.720 replication crisis where it's turning out that most scientific findings are probably not true

01:05:36.560 because, you know, I was a grad student at Columbia university, which is obviously like a

01:05:40.280 reputable institution and skipped straight to learning the particulars of Arctic plant physiology

01:05:44.840 before I learned how science and statistics actually work for me to draw true conclusions.

01:05:50.640 And so I have published research out there now that I'm quite sure would fall to the replication

01:05:54.840 crisis if someone tried to replicate it. Well, it's probably an example of, of learning a procedure,

01:05:59.700 right? Sort of shallowly, superficially you, you, you do it and then you get the, it looks like

01:06:05.160 science, but not really. Yeah. And the thing is one of now with computers,

01:06:10.280 there's anyone can get a huge data set and run statistical programs on it and you'll come up

01:06:16.200 with some positive results. And the fact is like myself, most of the scientists out there don't

01:06:22.280 really understand what they're doing when they're running those statistical tests because we've never

01:06:26.080 even been taught to think about it basically. So again, only as a journalist was, was I made to sort

01:06:31.340 of reflect on what I'd been doing in the past. And, and so that's what I think Arturo wants to do is he,

01:06:36.480 he wants to teach thinking. And that's what James Flynn of the Flynn effect suggests in the second

01:06:41.980 chapter is we have to teach these varieties of thinking. Otherwise, none of this didactic

01:06:48.120 information and procedures really makes sense anywhere except for these incredibly narrow

01:06:52.560 applications. Well, David, there's a lot more that we could talk about because there's so much more in

01:06:56.180 this book. Where can people go to learn more about the book and your work? Davidepstein.com.

01:07:00.180 There's a description of the book and some, some early reviews up there. And I think it's,

01:07:04.200 and some other work and it should be hopefully available in your favorite bookseller.

01:07:10.060 David Epstein. Thanks so much for your time. It's been a pleasure.

01:07:12.240 Pleasure is mine.

01:07:13.040 My guest here is David Epstein. He's the author of the book Range. It's available on amazon.com

01:07:17.180 and bookstores everywhere. Also check out my previous interview with David. It's episode number

01:07:21.020 127. It's about his book, The Sports Gene. Check out our show notes, aom.is slash range,

01:07:26.180 where you can find links to resources, where you can delve deeper into this topic. Also check out

01:07:29.360 David's website, davidepstein.com, where you can find more information about his work.

01:07:40.260 Well, that wraps up another edition of the AOM podcast. Check out our website,

01:07:44.020 artofmanliness.com, where you can find our podcast archives, over 500 there, as well as thousands

01:07:49.020 of articles written over the year about personal finance, health and fitness, style, how to be a

01:07:53.440 better husband, better father, you name it, we've got it. And if you'd like to check out ad-free

01:07:57.860 episodes of The Art of Manliness, you can do so only on Stitcher Premium. And so you can get a free

01:08:02.380 month of Stitcher Premium when you sign up at stitcherpremium.com and use promo code manliness.

01:08:07.660 After you sign up at stitcherpremium.com, download the app on iOS and Android, and you can start

01:08:12.360 enjoying new episodes of The Art of Manliness ad-free. Again, use that promo code manliness to get a

01:08:17.100 free month of Stitcher Premium. And if you haven't done so already, I'd appreciate it if you take

01:08:20.600 one minute to give us a review on iTunes or Stitcher. Helps out a lot. And if you've done

01:08:24.160 that already, thank you. Please consider sharing the show with a friend or family member who you

01:08:28.000 think would get something out of it. As always, thank you for the continued support. Until next

01:08:31.660 time, this is Brett McKay, reminding you not only to listen to the AOM podcast, but put what you've heard

01:08:35.280 into action.

01:08:47.100 Thank you for listening to the AOM podcast.

The Art of Manliness - May 29, 2019

#512： Why Generalists Triumph in a Specialized World

Episode Stats

Length

Words per Minute

Word Count

Sentence Count

Misogynist Sentences

Hate Speech Sentences

Summary

Transcript