Seth Roberts on Diversity, Education and Innovation

Wonderful post from Seth Roberts. Not sure if I even agree but Seth makes me think about how I can better serve my students:

People can do great things in dozens of ways, but somehow student work is almost never judged by how beautiful, courageous, practical, good-tasting, astonishing, vivid, funny, moving, comfortable, and so on it is. Because that’s not what professors are good at….To fail to grasp that students can excel in dozens of ways is to seriously shortchange them. To value them at much less than they are worth — and, above all, to fail to help them grow and find their place in the world after college.

At Berkeley, I figured this out in a way that a libertarian should appreciate: I gave my students much more choice. For a term project, I said they could do almost anything so long as it was off-campus and didn’t involve library work. What they chose to do revealed a lot. I began to see not just how different they were from me but how different they were from each other. One of my students chose to give a talk to a high-school class. This was astonishing because she has severe stage fright. Every step was hard. But she did it. “I learned that if I really wanted to, I could conquer my fear,” she wrote.

One of my Tsinghua students recently asked me: “Are you a brave man?” (She wanted to give me a gift of stinky tofu.) I said no. She said she thought I was brave for coming to China. Perhaps. I have never done anything as brave as what my student with stage fright did. I have never done something that terrified me — much less chosen to do such a thing. Her homework hadn’t been very good. When I read about her term project — conquering stage fright — I realized how badly I had misjudged her. How badly I had failed to appreciate her strengths. I saw that it wasn’t just her and it wasn’t just me. By imposing just one narrow way to excel, the whole system badly undervalued almost everyone. Almost everyone had strengths the system ignored. And it’s a system almost everyone must go through to reach a position of power!

…The glorification of IQ has had a solipsistic aspect and has ignored what should be obvious, that diversity of talents and skills promotes innovation. Without a diverse talent pool, any society will do a poor job of solving the problems that inevitably arise.

Hat tip to Bryan Caplan who is less of a romantic than Seth, or I.


Yeah, it's not necessary to be taught anything in college. You just have to occasionally do things that make you feel all fuzzy about yourself.

Sounds like $150,000 well-spent, doesn't it?

Most of the value of college is being able to say, "I was accepted to X, and I graduated."

So, depending on the X ... yeah.

Lots of waste in this world.

I'm always surprised when it turns out people actually believe the value of higher education is not just signaling in nature, but signaling no more than your ability to get in and out. I mean... I know economists, both serious and armchair, like to make simplifying assumptions, but this seems like a stretch. It's a position to take, but do you really think it's correct, or are you just being snide cause it's fun for you? It's one thing to say higher ed is all about signaling, it's quite another to say it's all about signaling your ability to write a resume at 17 and perform the bare minimum to get through courses.

What else would it be signalling? Those are the 2 things required to get a prestigious college degree

I think the core problem here is an inevitable one: Some important things are easier to measure than others, and the easy-to-measure things probably get overused for decisions. Standardized tests are relatively cheap and easy to give and grade and use, as are grades and majors and school rankings. Many other things that are just as important in success in a job are much harder to measure, and so they aren't used as much.

I imagine there are parallels to this in all sorts of things, like sports and medicine.

As an example (but I don't know enough to make up a good one): Most trips to the doctor involve checking some easy-to-get things like temperature, blood pressure, and pulse. Things that are harder to check (EKG, peak flow, presence of polyps in your colon) only get checked when there's some other reason for the doctor to look there. You can see screening guidelines as a way of correcting for that--tell everyone over 50 to go get a colonoscopy, tell women over 50 to get a mammogram every year, etc.)

I think this is right. See the post above - if it isn't in the textbook, it's "fuzzy." There's lots about people that can't be measured objectively - sometimes not easily, sometimes not at all.

In a way it's the anti-Moneyball. Moneyball was about using objective data (statistics) when everyone else was relying on subjective analysis (boy that #24 sure looks good in the uniform! He's got hustle!) This approach works well in baseball because you're judged on purely objective outcomes -- hits and, eventually, runs. That doesn't mean it's equally applicable to all human endeavors.

One problem is that the word "diversity" has come to be equated more and more with politically powerful identity traits (race, ethnicity, sex, sexual orientation) and less and less with skills and personalities. Some people manage to simply ignore modern diversity-think, despite all the social and legal incentives. One example was the late Steve Jobs. Walter Isaacson writes about the wildly successful top management team Jobs assembled at Apple after his return: "Even though there was a surface sameness to his top team—all were middle-aged white males—there was a range of styles."

"The glorification of IQ has had a solipsistic aspect and has ignored what should be obvious, that diversity of talents and skills promotes innovation."

On the whole, Seth Roberts' suggestion of downplaying IQ in favor of a diversity of talents and skills will tend to benefit whites at the expense of East Asians.

This has opened my eyes. From now on I will not judge waiters on how efficiently they deliver food or mechanics on how good they are at fixing my car but I will judge them on their ability to be "beautiful, courageous, practical, good-tasting, astonishing, vivid, funny, moving, comfortable."


This is a particularly odd post from Alex, given the amount of time he spends arguing that college education today allows students to gravitate toward far too many subjects that have zero real world value. You can either support the idea that it's wonderful for students in a physics class to do interpretive dances rather than taking exams or you can support the idea that colleges really need to heard students more aggressively toward fields of study that create value both for individual workers and society as a whole.

Maybe it's time to drop the false idea that there are fields of study that create value for workers and society. You can herd as many students as you want into physics - a physics course isn't going to do an ounce of good to the vast VAST majority of students when they hit the job market.

Maybe the base idea isn't false, but the specific idea that physics is the way to go is false.

We hear a lot about STEM, but there's varying usefulness in that grouping. The Science and Math portions are probably less valuable in the workplace than the other two. I think the areas where a smart individual can provide the most value to society are engineering, programming (as distinct from academic computer science and perhaps included in engineering as "software engineering") and entrepreneurial business (this isn't often taught in colleges - MBA's are for middle management, finance tends toward zero-sum shenanigans these days, I'm talking about the value created when you start or grow a new business that actually provides a valuable good or service).

It's unclear to me what important problems society has right now that are best solved by physicists or mathematicians. On the physics front, nuclear fusion would be helpful for our energy situation I guess. Math-wise, I'm stumped.

"You can either support the idea that it’s wonderful for students in a physics class to do interpretive dances"
Is this what you got from the linked piece? Because it's not what I got at all.

It's a false dichotomy. No one is saying you should not learn physics. OTOH in the typical job no one asks you to solve Fermats Last Theorem. You ARE likely to be asked to make a presentation or facilitate a brainstorming session.

BTW I studied music but since have started multiple software companies. I would fail a final exam for first year computer science students. Somehow my customers don't care.

> You ARE likely to be asked to make a presentation or facilitate a brainstorming session.

Or work on a team. Or deal with a lackluster colleague. Or win a debate without burning bridges.

It's not so much that these things are unmeasurable, it's that academics likely have very little experience with them. Working independently mostly reading and writing on a small and abstract problem just isn't very much like the real world.

I remember several labs in which working with others was the toughest part of the assignment, and the professor never said a thing about it.

@Scoop...herding rarely ends well for individuals or society. The idea of the post is not that all people are good at all things, but that every person has a strength. It makes perfect sense to me that a libertarian would value the individual.

Teaching organic chemistry, it has not occurred to me to judge students on how good-tasting their work is. However, I have, even in beginning classes, given almost full credit for "elegant" synthetic schemes that had a fatal flaw the student was in no position to appreciate. I know most of my colleagues act the same. Synthetic chemistry involves strategy as well as tactics and there is a long tradition in the field of admiring bold or novel strategies. I also sometimes offer to students who complain about tests that he or she can write a scientific biography of a notable organic chemist in place of doing the final. So far everyone has wisely declined since such a task would involve learning far more chemistry than one would need to ace my final.

Anyone who's taught or graded math at a reasonably high level has given marks for beauty or elegance or taste.

Either that or it's evidence that they have no soul. :)

This is just silly. He teaches a class at Berkley (one of ten best universities in US) which is already highly selective, and then says that academics don't matter (as much). For that matter Tsinghua is no Smallville evening college either. When he's had a couple of years experience teaching students who don't know which side of shakespeare butter goes onto and can't string two words together without one of them starting with an 'F', then I'll take him seriously.

Oh all the comments on this blog are always so full of strawmen. Of course less capable people exist but why is it always so exaggerated around here? It's either you're someone who pulls off As in your Quantum Mechanics course or you're some kind of driveling idiot who can barely operate in the world - there's never an in between.

Who's talking about driveling idiots? (That's ableist talk by the way, you seem to be in need of some sensitivity training). You can't very well deny that Seth is teaching the best of the best (OK the intelligentest of the intelligent) so his experience is necessarily a bit skewed. In fact in another recent post he bemoans the fact that the world is structured to reward the too easily measurable trait of intelligence, and here he is bemoaning the fact that professors are too professorial (breaking news! they have been since 500 years ago). Makes one wonder why is he still at Berkley or Tsinghua? Plenty of places to choose from if he wants to make a difference.

Seth only interacts w/ the right side of the bell curve (Berkely students), so he's blind ot how limited the left side is, so he underestimates the importance of IQ. Academics often make this error.

"By imposing just one narrow way to excel, the whole system badly undervalued almost everyone." Nonsense. Your algebra teacher, in giving your performance in her class a grade, isn't assessing your overall value; she is assessing how well you have learned algebra. True, if knowing algebra is a very important part of a human being's overall value, there will be a positive correlation between her grade and your overall value. But "the system" does not say how strongly positive this correlation is.

But unfortunately when the student goes for a job they're going to be judged heavily based on what their grade was even if it's largely irrelevant to the job in question. For many jobs the History major who got an A is going to be chosen far more often then the Math major with an B+ or B.

Employers want smart, housebroken employees for many positions. They can't legally discriminate candidates based on any standardized tests, but it is legal for them to discriminate by grades and schools, so they do.

But it's not a good measure since it's far easier to get As in history courses then in math or physics and yet many employers treat easy courses the same as more difficult ones.

Math and physics courses don't come with the tons of writing, and don't function nearly as well in the housebreaking department. Physics, after all, deals with reality and its immutable laws.


At least as a parent you get to encourage the beautiful, courageous, practical, good-tasting, astonishing, vivid, funny, moving, etc.

No, no, no! Diversity is understanding that everyone is the same.

Report to sensitivity training on Monday AM.

It's probably a feature rather than a bug. I'm convinced the entire point of organized education is "are you able to learn something so boring that it induces suicide in others."

This is another thing engineering has over other fields. The insane volume of work provided both a lot of data points as well as served as the sled thingy in the tractor pull that stops the truck short. They basically admit that it's signaling and human capital and by God they do it.

Engineering is precisely where it's not all or mostly about signaling. Learning calculus, mechanics and strength of materials changes how you view the world in fundamental ways, even if you don't have to do contour integrals on paper at work. Real classical education (Latin and Ancient Greek besides a couple of contemporary European languages, classical canon etc., stuff which got gradually thrown out of 'liberal education' after the war) also changed the worldview, although in different ways. Modern education? Adolescents start doing community work in high school because it is a requirement for top schools, not because they want to.

@Anonymous coward, Just because you are "required" to do something doesn't mean you there's no learning. Maybe you learn how quickly finish your boring prof's make-work assignments, so you can get on with your life? There's another skill for the work place.

I don't see anything inherently wrong with the premise, but if the purpose of the course is to learn calculus, then of course the only way to excel is to learn calculus and demonstrate your knowledge by solving problems. If your grade reflects anything else, it is useless.

The way students demonstrate their overall worth to an employer is usually by taking an internship (or, god forbid, a lower level job) and woking their way up.

The same limitations in what the teacher looks for, are the same limitations that those who would hire us also need. This is why, when we contrast our actual abilities in sets of skills portfolios that could be matched, wealth creation could actually be far more exponential at individual to individual levels.

Maybe I'm getting old, but the author's approach seems somewhat sloppy and undisciplined. I think my grandfather would have called it "Half-Assed."
In some areas (arts?) one COULD argue that every unique interpretation is valid; in areas like structural analysis or thermodynamics, you'll run in to trouble.
But even in the arts, mastering the basics is essential for one to become "Good." (i.e. Musical technique through practice. Lots and lots of practice. Please!)

The astrologers got it wrong — it's not the Aquarian Age now but the Age of the Half Ass.

The nicest thing one of my undergraduates ever said to me: "You certainly taught me to spell 'baloney'."

It's beatiful and actually is exactly as it looks on the website photos,I'm very glad on it!

Tenacity > IQ.

What are the chances that when Seth Roberts needs a surgery, he will look for the most beautiful surgeon who is funny and cooks good-tasting food to perform the operation?

Comments for this post are closed