NꙮW WITH MꙮRE MULTIꙮCULAR ꙮ

Hungarian Education III: Mastering The Core Teachings Of The Budapestians

[Previously in series: The Atomic Bomb Considered As A Hungarian High School Science Fair Project, Four Nobel Truths]

I.

Someone summed up my previous post as “Hungarian education isn’t magic”. I would amend that to read “Hungarian education isn’t systemically magic”. As far as I know, there’s only one Hungarian educator with magic powers, and (like all good wizards) his secrets are maddeningly hard to find.

Laszlo Polgar studied intelligence in university, and decided he had discovered the basic principles behind raising any child to be a genius. He wrote a book called Bring Up Genius and recruited an interested woman to marry him so they could test his philosophy by raising children together. He said a bunch of stuff on how ‘natural talent’ was meaningless and so any child could become a prodigy with the right upbringing.

This is normally the point where I’d start making fun of him. Except that when he trained his three daughters in chess, they became the 1st, 2nd, and 6th best female chess players in the world, gaining honors like “youngest grandmaster ever” and “greatest female chess player of all time”. Also they spoke seven languages, including Esperanto.

Their immense success suggests that education can have a major effect even on such traditional genius-requiring domains as chess ability. How can we reconcile that with the rest of our picture of the world, and how obsessed should we be with getting a copy of Laszlo Polgar’s book?

II.

Let’s get this out of the way first: the Polgar sisters were probably genetically really smart. The whole family was Hungarian Jews, a group with a great track record. Their mother and father were both well-educated teachers interested in stuff like developmental psychology. They had every possible biological advantage and I’m sure that helped.

J Levitt proposes an equation to estimate a chess player’s IQ from their chess score. It suggests that chess grandmasters probably have IQs above 160. Plugging the Polgar sisters’ chess scores into his equation, I get IQs in the range of 150, 160, and 170 for the three sisters.

This is biologically impossible. Even if both Polgar parents were 170-IQ themselves, regression to the mean predicts that their children would have IQs around 140 to 150. It’s mathematically possible for there to be an IQ that predicts you would have three children of 150, 160, and 170, but I doubt any living people have it, and even if they did there’s no way they would marry somebody else equally gifted.

[EDIT: Thanks to a few people who pointed out some problems with my math here (1, 2, 3). I still think that having three supergenius-IQ kids when you and your spouse show no signs of being a supergenius yourself (Laszlo Polgar’s daughters could beat him at chess by the time they were 8) is pretty unlikely, but I admit not impossible. I still think arguing about this is unnecessary thanks to the points below.]

On the other hand, I’m not sure Levitt’s right. Chess champion Gary Kasparov actually sat and took an IQ test for the magazine Der Spiegel, and his IQ was 135. That’s not bad – it’s top 1% of the population – but it’s not amazing either.

This is what we should expect given the correlation of about r = 0.24 between IQ and chess ability (see also this analysis, although I disagree with the details). And the contrary claims – like the one that Bobby Fischer’s IQ was in the 180s – are less well-sourced (although Fischer was the son of a Hungarian-Jewish mathematician, so who knows?).

If it were possible to be a chess world champion with an IQ of 135, then maybe it’s possible to be a “mere” grandmaster with IQs in the high 120s and low 130s. And it’s just barely plausible that some sufficiently smart people might have three kids who all have IQs in the high 120s and low 130s.

But this just passes the buck on the mystery. 2% of people have IQs in the high 120s or low 130s, but 2% of people aren’t the top-ranked female chess player in the world. The Polgar sisters’ IQs might have been a permissive factor in allowing them to excel, but it didn’t necessitate it. So what’s going on there?

III.

“Practice” seems like an obvious part of the picture. Malcolm Gladwell uses the Polgars as poster children for his famous ‘10,000 hours of practice makes you an expert at anything’ rule. The Polgars had 50,000 hours of chess practice each by the time they were adults, presumably enough to make them quintuple-experts.

Robert Howard has a paper Does High-Level Performance Depend On Practice Alone? Debunking The Polgar Sisters Case in which he argues against the strong version of Gladwell’s thesis. He points out that there are many chess masters who have practiced much less than the Polgar sisters but are better than they are. He also points out that even though the sisters themselves have all practiced similar amounts, youngest sister Judit is clearly better than the other two in a way that practice alone cannot explain.

I don’t know if the case he’s arguing against – that practice is literally everything and it’s impossible for anything else to factor in – is a straw man or not. But it seems more important to consider a less silly argument – that practice is one of many factors, and that enough of it can make up for a lack of the others. This seems potentially true. This study showing that amount of practice only explains 12% of the variance in skill level at various tasks, and is often summarized as “practice doesn’t matter much”. But it finds practice matters more (25% of the variance) in unchanging games with clear fixed rules, and uses chess as an example.

So suppose that the Polgar sisters are genetically smart, but maybe not as high up there as some other chess masters. We would expect them to need much more practice to achieve a level of proficiency similar to those chess masters, and indeed that seems like what happens.

(all of this is confounded by them being women and almost all the other equally-good chess masters being men. It’s unclear if the Polgars deserve extra points for overcoming whatever factor usually keeps women out of the highest levels of chess.)

But I’m actually still not sure this suffices as an explanation. According to Wikipedia:

Polgár began teaching his eldest daughter, Susan, to play chess when she was four years old. Six months later, Susan toddled into Budapest’s smoke-filled chess club,” which was crowded with elderly men, and proceeded to beat the veteran players.

The study linked above suggests that Susan practiced 48 hours a week. During those six months, she would have accumulated about 1200 hours of practice. Suppose the elderly Budapest chess players practiced only one hour a week, but had been doing so for the last twenty-five years. They would have more practice than Susan – plus the advantage of having older, more developed brains. So why did she beat them so easily?

Maybe there’s a time-decay factor for practice? That is, maybe Susan had been practicing intensively, so she got a lot of chances to link it all together as she was learning, and also it was fresh in her mind when she went to the club to go play? I’m not sure. If some of those veterans had been playing more than one hour a week (and surely the sort of people who frequent Budapest chess clubs do) then her advantage seems too implausible to be due to freshness-of-material alone.

IV.

That leaves two possibilities.

First, Susan could have benefitted from some form of malleability. A lot of people claim there’s a “developmental window” during which children have a unique ability to learn language. If cats see only vertical stripes for the first few weeks of their lives, they never learn to see in horizontal. Maybe if you teach your kid high-level chess at age 4, they’ll be able to recruit systems that adults could never manage, or reorganize the fundamental structure of their brain to conform to chess better, or something like that.

Second, Polgar might actually have some really good educational methods besides just “start early and have a lot of practice”. I assume this is true, but I’m having a lot of trouble finding them. Shockingly, Polgar’s book Bring Up Genius is out of print and totally unavailable anywhere – I guess the book-reading community heard that someone wrote down a way to reliably turn any child into a genius which had a great real-world track record of success, and collectively decided “Nah, better read Fifty Shades Of Grey instead”. I’m not sure at what point I should start positing a conspiracy of suppression, or whether that would be better or worse than the alternative.

The book seems to possibly be available in Hungarian under the title Nevelj zsenit!, but I can’t tell for sure and a lot of the Hungarian sites suggest it’s out of print even in that language. There may have been a recent republication in Esperanto called Eduku geniulon!, but I can’t find that one either. If anybody knows where to find this book and wants to send it to me, I will figure out some way to translate it and review it. I’d also be willing to pay for costs and even pay extra for your time if it helps. Come on, Esperanto-speakers! This is the only chance you’ll ever have to be useful!

V.

One thing I know without reading the book: Polgar says that his method should work to create geniuses in any field, not just chess. He said he chose chess kind of on the whim of his eldest daughter. From Wikipedia:

Polgár and his wife considered various possible subjects in which to drill their children, “including mathematics and foreign languages,” but they settled on chess. “We could do the same thing with any subject, if you start early, spend lots of time and give great love to that one subject,” Klara later explained. “But we chose chess. Chess is very objective and easy to measure.” Susan described chess as having been her own choice: “Yes, he could have put us in any field, but it was I who chose chess as a four-year-old…. I liked the chessmen; they were toys for me.”

It’s disappointing that he decided to stick with chess for his other two daughters. The study linked above suggests that chess is unusually amenable to practice. What would have happened if he’d tried to train his kids in art? In mathematics? In entrepreneurship? I’m not sure, and I’m really tempted to have some kids and find out.

(be right back, going to change my OKCupid profile to include “must be interested in n=1 developmental-psych experiments, have access to a rare book library, and speak either Hungarian or Esperanto”)

I mentioned this plan to a friend, who protested that this was cruel and tantamount to child abuse. After all, how can you force someone to spend their entire childhood indoors, studying mind-numbing chess problems day in and day out, instead of enjoying themselves like normal kids?

First of all, this isn’t how the Polgar children (or adults) describe their experiment. From The Guardian:

Starting with his eldest daughter, Susan, Polgár was careful to treat it as a playful activity, turning it into a fantasy of dramatic wins and losses. Whereas Earl and Kultida Woods had coerced perfection from Tiger, the Polgárs encouraged enjoyment, By the time Susan had turned five, she was excited by playing and spent hundreds of hours practising. She was entered into a local competition and treated it as fun, winning 10-0, causing a sensation.

Meanwhile, her younger sisters were intrigued and László allowed them to feel the pieces, seeing them as toys, with no formal tuition until they were five. Interviewed recently, all three girls described playing the game as something that they loved doing – it never felt like a chore. Instead of messing about playing Monopoly, netball or going to the local swimming pool, chess was just what the Polgár family enjoyed…Polgár understood that coercion was less valuable than small children’s need to enjoy fantasy play. Consequently, his daughters all seem to have grown into satiable, well-balanced people rather than success addicts.

But more important – I responded that the Polgars claim to have spent about 48 hours a week practicing chess. I spent seven hours a day in school, so if my teachers assigned two hours of homework a night then we spent about the same amount of time getting educated. Except what the Polgars got out of it was world-champion-level mastery of their favorite subject in the world, nationwide fame, and (by their own accounts) loving every second of it, and what I got was staring out a window all day as my teacher declared that we were going to make a collage about the meaning of Respect.

The Polgar sisters talk about how they loved their education, had a great childhood, thought their parents were always patient with them and never strict and harsh, and don’t regret anything. How many kids who went to public school can say the same?

An article about Laszlo Polgar mentions that he had to fight the Hungarian authorities to be allowed to home school his children. Imagine being so certain of your own home-schooling techniques that you’re afraid taking your kids to the Fasori Gymnasium is going to stunt their intellectual growth. And imagine being right. And imagine my friend thinking that normal American public school might be better than that. It sort of boggles the imagination.

And I guess I shouldn’t be too harsh, because the public school system tries to do the best it can with an impossible set of constraints. But I’m still suspicious. Who else has the motivation to hide that book?

[EDIT: Thanks to readers, I’ve got an Esperanto copy and a person willing to translate it. I’ll let you know as this develops.]

This entry was posted in Uncategorized and tagged , . Bookmark the permalink.

Leave a Reply

437 Responses to Hungarian Education III: Mastering The Core Teachings Of The Budapestians

  1. TheWackademic says:

    I think that the story of Laszlo Polgar and his daughters seems like an extreme case of survivorship bias. We don’t know what sample size we started out with. For every Laszlo Polgar who succeeded in creating geniuses, there may be 100 or 10,000 who did exactly the same thing and failed, for no reason other than the fact that the “Polgar Method” works 1/10,001 times. There probably aren’t wikipedia pages for all of the kids who played 48 hours a week of chess from age 4 onward but maxed out at a 1900 rating, became CPAs, and lived quiet lives in Evanston, Illinois.

    • Scott Alexander says:

      I don’t know how many people write books about their ability to bring up genius children and court their future wives on the basis of “let’s raise genius children together using my method”. I doubt there are ten thousand of them.

      But I think the stronger counterargument is that he had three kids and his record was 3/3. The most certain way to eliminate survivorship bias is to repeat the experiment. He did. Twice. I’m willing to believe this one.

      • TheWackademic says:

        N=3 would be a persuasive counterargument if they were truly “independent experiments” – say, if he had adopted 3 kids from different countries and turned them all into geniuses. Doing it on 3 siblings is much less impressive – they may be genetically predisposed to chess ability.

        Thinking on this more – there are many stereotypical examples of “successful prodigies raised by ultra-intense parents”: Mozart, the Williams sisters, Terrence Tao, the Polgar sisters, etc. Each example that comes to my mind was the biological child of the parents who pushed them. Are there any examples of adoptees being raised to genius/prodigy-level performance? Otherwise, maybe an alternate explanation would be that “having a strong work ethic and intense desire to succeed is genetic.”

        • Scott Alexander says:

          See Part II – due to regression of the mean, it’s near-impossible that anyone would have three kids capable of succeeding at chess due to biological intelligence alone (although of course high biological intelligence was a necessary permissive factor). That a person obsessed with educating people into geniuses should strike this one-in-a-million chance beggars the imagination.

          I’m not saying these people don’t probably have top 1% IQs. I’m saying that 70 million people worldwide have top 1% IQs, so something else must be going on here.

          • dmorr says:

            FWIW my kids go to a school with a minimum 130 IQ. Both my kids are there, and there are several families with all three kids there, and at least one with 4.

            Also many families with one kid there and siblings elsewhere.

            Obviously there’s some series selection bias going on here, but I don’t think it’s necessarily that rare to have multiple high IQ kids in the same family. My guess is whatever stats you’re using to suggest that it’s that hard are buggy.

          • bean says:

            I agree with dmorr. I went to one of those schools, and from memory it was about 50/50 in terms of siblings getting in.

            Edit:
            Thinking this over more, there were sort of two types of families. There were some who had one kid in the program, and the rest of the kids were smart, but not particularly exceptional. Then there were families where all of the kids were very smart and hung around with those in the program, even if not all of them quite made the cutoff. (Mine was one of the later. My brother and I made it in, my sister barely failed.)

          • TheWackademic says:

            1) “Regression to the mean” is an argument as to why it’s unlikely, not as to why it’s impossible. Eventually, someone wins the lottery.

            2) I’m not saying they succeeded in chess due to biological intelligence alone. I’m saying they aren’t “independent experiments.” They may have had many similarities – adherence to authority, high testosterone levels, long attention span, who knows what – that explained why the same teaching regime worked 3 times in a row.

          • gin-and-whiskey says:

            Like dmorr said, I think your regression-to-the-mean algorithm is buggy. Both of my parents are somewhere in top 1%; all of their kids are also somewhere in top 1%. There are over a dozen grandkids; i have a few of them and although I’ve only formally tested one all of mine are obviously top 1% across the board and at least half of that generation is top 1% if not more. (My mom’s ancestry is a mix of central and eastern european jews, FWIW.)

            Maybe those statistics only hold for top 0.01?

          • Eponymous says:

            One minor point here: regression to the mean is not independent across children.

            Think about it this way: a person’s IQ is the combination of “true genetic potential” plus “noise”. True genetic potential (mostly additive +IQ genes) doesn’t regress to the mean, it just gets passed on to your kids. Noise regresses.

            If someone has above-average IQ, then on average they have a good noise score, and their IQ is above their genetic potential. But this isn’t true for all high IQ people. Thus if a person has a high-IQ kid, you should raise your estimate of their true genetic potential, making it more likely that future kids will also be high-IQ.

            This is why looking at IQ of relatives can be informative.

          • Steve Sailer says:

            The Williams Sisters in tennis were part of a nature/nurture experiment. Before they were born, their dad was was watching white women play tennis on television and told their mom that the two of them could have daughters that could beat the women they were watching.

          • Eponymous says:

            @Steve Sailor

            I don’t think it’s a coincidence that both cases involve women succeeding in a male-dominated activity. Since overall fewer women participate in chess/sports due to relative lack of interest, it’s more likely that such an experiment will succeed. (Lots of boys come to obsess on chess all on their own, so raising your son to be a chess-obsessed doesn’t guarantee him a spot at the top!)

          • Steve Sailer says:

            Most women who achieve a lot in male dominated spheres are not the daughters of single women. In fact, they often follow their father’s profession. In America at least, traditional families seem to be best for raising stereotype-shattering daughters.

          • BlindKungFuMaster says:

            Probably somebody already pointed it out, but regression to the mean is calculated with the information you have available. If you randomly grab a Hungarian couple from the street you’ll calculate a regression to 100, if you happen to know that they are Ahkenazim you’ll expect regression to 115. If come across the average IQ of their parents and siblings, their kids will regress towards that mean, possibly even away from the mean of the general population.

            As a long time chess fanatic I’d say that your Elo is determined by how much you work, how early you start and whether you do the right kind of work i.e. you can play for years without improving, you’ll have to train to improve. Not many veterans do that. After these three factor IQ might play a role, but having seen and read hundreds of interviews with top chess players (there are post game interviews in online live transmissions of tournaments), they don’t strike me as super smart. 125-140 is probably plenty to reach the top.

            Interesting tidbit: When Judit Polar was the top player in the age bracket (I think) <13 in the world, guess who occupied second place? Demis Hassabis!

          • Aapje says:

            Magnus Carlsen interview:

            Carlsen: Of course it is important for a chess player to be able to concentrate well, but being too intelligent can also be a burden. It can get in your way. I am convinced that the reason the Englishman John Nunn never became world champion is that he is too clever for that.

            SPIEGEL: How that?

            Carlsen: At the age of 15, Nunn started studying mathematics in Oxford; he was the youngest student in the last 500 years, and at 23 he did a PhD in algebraic topology. He has so incredibly much in his head. Simply too much. His enormous powers of understanding and his constant thirst for knowledge distracted him from chess.

            SPIEGEL: Things are different in your case?

            Carlsen: Right. I am a totally normal guy. My father is considerably more intelligent than I am.

            SPIEGEL: Aha. How many moves can you calculate ahead?

            Carlsen: That depends on the game situation. Sometimes 15 to 20. But the trick is to correctly assess the position at the end of the calculation.

            SPIEGEL: You became a grandmaster at the age of 13 years, four months and 27 days; and there has never been a younger number one than you before. What is that due to, if not to your intelligence?

            Carlsen: I’m not saying that I am totally stupid. But my success mainly has to do with the fact that I had the opportunity to learn more, more quickly. It has become easier to get hold of information.

            http://en.chessbase.com/post/magnus-carlsen-on-his-che-career

          • Deiseach says:

            Count me as one of the sceptics that Laszlo Polgar found a method for reliably turning averagely intelligent kids into geniuses. As that article about Susan and her sisters I linked told it, the reason the three girls think Judit was the most successful is that she had the hunger for it, the will to win:

            When they meet, they’ll probably discuss another lingering question: What accounts for the sisters’ differing ratings? How could Judit, seven years Susan’s junior, overtake her sister despite far less practice? All three, of course, practiced more than almost anyone else in the world. There are family theories. Though Sofia was often said to have the most talent, in chess and elsewhere — during one 1989 tournament, in Rome, she went on a legendary run, beating a murderers’ row of Soviet grandmasters — she was never driven enough to focus on one thing, Susan says.

            Judit, meanwhile, had a killer instinct.

            “Out of the three of us, I was the most fit to the kind of life required to be on the top,” Judit says. Losses fueled her determination. “I had this drive in me that I wanted to show it was possible and I can do it.”

            And Sofia has two sons who played chess but weren’t super into it, and she didn’t push them – her explanation for the sisters’ success was that their parents (both of them) made it their whole lives’ work to train and steer the girls to success.

            So you take kids of reasonably high intelligence, possibly already inclined towards chess/maths (it wasn’t by accident, after all, that there was a chess set in the family apartment for toddler Susan to find and want to play with; her father not alone played chess but is a chess teacher), spent your time teaching them to play at an adult level from a young age, encourage them, support them, and generally have the entire household revolving around success at chess, you are going to get some kind of a result. From that 2015 Chronicle of Higher Education article:

            It was natural for Sofia and Judit to follow Susan’s lead. (Their parents considered naming Judit Zseni — “genius” — if she had been a boy.) The younger sisters, peering through a small window in the door of the back room where Susan practiced, hated being excluded; chess was their way in. Laszlo believed that physical fitness was vital to intellectual success, so the girls played table tennis several hours a day, on top of their full day of chess and schooling.

            The parents were tireless in their devotion, buying every chess book they could, cutting out pages with past games, gluing them to cards, and storing it all in an old card catalog. They assembled more than 100,000 games; at the time, only the Soviet Union’s restricted chess archive could match it.

            I admit it’s a wonderful achievement. But why didn’t the Hungarian government snap up the Polgar Method and start churning out a stream of young Hungarian geniuses? Okay, so maybe they felt he made monkeys out of them and they weren’t going to encourage him at all. But why hasn’t any other country tried the same? Why has the book fallen into such obscurity that he couldn’t even arrange an English publication?

            Laszlo might finally publish his book in English, too, having at last given up hopes of a six-figure advance. He and his wife have begun wintering in Florida, and they’re planning to meet with Ericsson*, for the first time, this year.

            A fool-proof method, backed up by three genius girl chessplayers, and no publisher in the English-speaking world – not even in the USA – would take a punt on it? Not even the publishers of the tome below?

            Bob Litwin’s Live the Best Story of Your Life, After not making his tennis team in college, Bob Litwin became a history teacher who wanted more out of life. By using a unique brand of storyteller, he became #1 in the world in tennis and has helped thousands of others reach seemingly impossible goals. Bob has sent tips to Djokovic before he won tournaments, he’s coached diamond barons, New York Times editors and some of the most powerful men on Wall Street, and now he shares his secret sauce with us. The 33 coaching sessions included in this book can easily be read in an afternoon by the pool and will increase the odds that every other self-help book on your shelf will actually start working. This is the first book I’ve ever read that made me believe change was actually easy.

            That sounds to me like the proof of the pudding is in the eating, and despite all the claims that “it’s the method, stupid”, people really do think (a) it’s genetics (b) yeah but very few parents are going to live 24/7 devoted to turning Junior into a child prodigy and those that are, will tend to be the ones where ten to fifteen years later we get best-selling memoirs of “I was a child prodigy, yes, but it was as a result of my parents being monsters who warped my life and that’s why I burned-out early and turned to drugs’n’kinky sex as the long list of scandals attached to my name attests”.

            *K. Anders Ericsson, a psychologist at Florida State University, that intense practice is the most dominant variable in success. The Polgárs would seem to suggest: Yes.

            You may have heard of Ericsson. His work was popularized by Malcolm Gladwell in his 2008 best seller, Outliers, which spawned the notion of 10,000 hours of practice, in particular, as a mythical threshold to success.

          • John Schilling says:

            Count me as one of the sceptics that Laszlo Polgar found a method for reliably turning averagely intelligent kids into geniuses. As that article about Susan and her sisters I linked told it, the reason the three girls think Judit was the most successful is that she had the hunger for it, the will to win:

            OK, but if your alleged recipe for producing geniuses produces two ordinary geniuses and one best-in-the-history-of-the-world champion, I’m not sure it’s fair to dismiss that with “Pfah! Sofia peaked at #6 in the world! Only Judit truly excelled!”

          • Bugmaster says:

            @Deiseach:
            To be fair, though, if someone did indeed discover a method to reliably turn genetically smart children into undisputed masters of their field — whatever that field may be — that would be almost as good. I don’t think this process happens organically; I mean, obviously sometimes it does, but I doubt the success rate is anywhere close to 100%.

          • JPNunez says:

            But…raising one, then two, then three kids to be good at chess is not an independent chance. Once you raise the first kid, the second one has an older sibling to share the game with, and the third has other two siblings fans of chess now.

            Hell, this is even simplified by the three siblings being the same gender.

      • BlindKungFuMaster says:

        There is a similar experiment with n>1000,000,000, which makes it pretty convincing I would say.

        I’m currently learning Chinese, which, as countless other westerners can attest, is pretty hard. With the 4-5 hours I can put into this each week, progress is glacial. But even if I had a lot of time for this, I would never approach anything like native speaker ability. In China though, they conducted this crazy child-raising experiment, where every child is completely surrounded by Chinese speaking people all the time, and they have to conduct every conversation in Chinese from the day they start to speak. This may sound cruel to us, but the results speak for themselves (no pun intended). All these kids grew up to be much more competent in Chinese than the vast majority of westerners who try to learn the language.

        I suspect neurologically learning chess and learning a language is very similar. Especially in the case of Chinese where the characters are even chunked visual inputs similar to chess positions. You could think of the grandmaster title as the ability to write a half-decent novel in Chinese. Not everybody can do, but a lot of Chinese could.

        To me the real question is in which areas we can expect massive gains by early starting and total immersion. Music, chess, language, math (is there a difference between technical skills/tricks and deep understanding?), programming? … the further you get away from learning lots of temporal-spatial patterns and the more higher cognitive functions are involved the more doubtful it seems to me.

        • Anonymous says:

          Seems legit.

          BTW, what are you having trouble with regarding Chinese? I’m learning it myself, and haven’t exactly felt that progress is ‘glacial’ despite putting in maybe an hour or two a week into it (split up into a little bit of practice every day).

          • BlindKungFuMaster says:

            Mostly I just forget everything I don’t repeat very regularly: Tones, characters, advanced grammatical constructions. It’s like I just don’t have the right memory slots for these things. When things get too bad I sit down and go through my 1000+ characters a couple of times and then it’s ok for a while, but I feel like I would have to dedicate a lot more time to actually get anywhere.

          • Anonymous says:

            Are you using spaced repetition?

          • BlindKungFuMaster says:

            Mostly I just scramble to keep up with my course and over some holidays I try to learn all the stuff I should have systematically learned earlier. Not ideal, but with a little kid and Phd-thesis to write, tough to change right now.

          • Anonymous says:

            Try some of these: https://ankiweb.net/shared/decks/mandarin https://ankiweb.net/shared/decks/chinese

            It’s a low-effort learning structure improvement, especially if you put that on your mobile device and practice while you ride the train or whatnot.

          • Very Interesting says:

            Unfortunately, if you put in an hour or two per week every week for 20 years your Chinese is still going to be terrible. You might feel like you’re learning a lot because spaced repetition is very powerful (and the anki decks are a great suggestion), but it’s just not going to cut it. And anki decks do nothing for speaking or listening practice, not to mention writing…Even worse, if you’re not practicing with a native speaker your pronunciation will be so poor that you’d have trouble communicating survival phrases in downtown Beijing.

            I don’t mean to be discouraging, but that’s just the truth. Here are some suggestions that will be helpful:

            1) Find a native speaker to practice with. All of the best Chinese speakers I know who learned as adults had boyfriends/girlfriends/whatever other kind of romantic partners who were native Chinese.

            2) Practice all modalities. Reading, Writing, Listening, Speaking. They all reinforce each other

            3) Used spaced repetition for vocabulary. I suggest HSK decks on anki.

            4) Watch TONS of Chinese videos online. Most of it is subtitled (in Chinese), which really helps with both listening AND reading. YouTube actually has a fair amount of stuff. If you’re looking for western media that’s been dubbed and subtitled in Chinese, search for, say, “The Little Mermaid 国语” or whatever else you want with 国语 stuck on there at the end. You’ll be excited when you recognize vocabulary you learned in your HSK decks and feel like you’ll reach mastery in no time. You won’t, but do take pleasure in small victories.

          • liquidpotato says:

            @ BlindKungFuMaster and Anoymous:

            This is the first time I have heard of spaced repetition and the wikipedia entry seems interesting. Can you share a little of your regime? I will see if I can incorporate that methodology into my own learning (for other things)

        • publiusvarinius says:

          > But even if I had a lot of time for this, I would never approach anything like native speaker ability.

          A whole lot of adults learn languages and get close to native speaker ability.

          In fact, it’s rare to find someone who spent 15+ years in a particular language environment without approaching something like native level mastery. Interestingly, all the exceptions I know are American/Canadian/British expats.

          • Anonymous says:

            In fact, it’s rare to find someone who spent 15+ years in a particular language environment without approaching something like native level mastery. Interestingly, all the exceptions I know are American/Canadian/British expats.

            I know plenty of people who, despite many years living and working in a foreign country, still have A1-equivalent mastery of the local tongue. And this despite very obvious benefits of learning the language.

          • BlindKungFuMaster says:

            > A whole lot of adults learn languages and get close to native speaker ability.

            The question is how close. I sometimes hear people who have been in my country for 30-40 years. Their pronunciation is perfect, but at some point they always use some word in a slightly weird way. In chess that would correspond to a slight mistake which puts you on the back foot. Then you’ll be under pressure and more mistakes follow.

            I played my first tournament game when I was 17. I can occasionally draw a grandmaster, but I’ll never have the same deep natural understanding. I don’t have it in the English language and I’ll certainly never gonna have it in Chinese.

          • publiusvarinius says:

            > I know plenty of people who, despite many years living and working in a foreign country, still have A1-equivalent mastery of the local tongue.

            Sure. As I said, I know some people like that as well.

            It does not change the fact that it’s totally possible to get close to native speaker ability as an adult (70%+ of those 15+ years since migration being at close-to-native language proficiency (according to the American Community Survey data).

          • Anonymous says:

            Oh, sorry. I misparsed that.

            Still, it’s not necessarily true that *everyone* can learn. I figure that adult language learning ability is something you can have more or less talent for. These people we both know are probably in a category of people who have greater difficulty with it.

          • Creutzer says:

            I sometimes hear people who have been in my country for 30-40 years. Their pronunciation is perfect, but at some point they always use some word in a slightly weird way.

            Perfect pronunciation is actually quite rare. My impression is that only a small minority of non-native speakers, even those who live in the relevant country, ever acquire an accent indistinguishable from that of a native speaker. (I haven’t checked any studies, my impression comes from hearing foreign university professors in the US.)

            I genuinely wonder to what extent using a word in a slightly weird way isn’t something that native speakers do, too. There are performance errors, and idiolects don’t always align perfectly.

          • Anonymous says:

            @Creutzer

            Agreed. There are natives with C1 proficiency, and second-language learners with C2, and people who are bilingual and have vocabularies indistinguishable from educated natives, but whose accents and idiosyncrasies immediately mark them as strange/foreign. Especially in the case of a language like English, where there’s no real unified standard for what “proper” speech/pronunciation/spelling is supposed to be.

          • Tibor says:

            British especially, in my experience. In fact those Americans who do travel actually seem to be interested in learning local languages.

            The Brits are terrible in this respect. I know a couple of English guys in my home town who have been living there for 15 or so years and whose level of czech is A1 at best. If you’ve lived somewhere for that long, you’re not an expat, you’re an immigrant. But those guys might as well have stayed in England for how much interaction they have with the locals. They all teach English and then they all hang out in the same Irish pub. At the end of the day it’s their life but I don’t see the point in that. Then again I generally find it strange when people seek out their countrymen abroad and band together, but most people seem to tick that way regardless of their nationality.

      • I think three out of three, while impressive, is not as impressive as it at first sounds, because it is probably easier for a child to learn to be a very good chess player if she has a sibling who is learning to be a very good chess player. Most kids don’t have the opportunity to spend a lot of time playing against players at their own level.

        I actually know of another, slightly less extreme, such situation. When I first got to know Frank Meyer’s family, his son John was the under 21 chess champion of the U.S. and his son Gene the under 14 (I think) champion. Both home schooled.

    • Walter says:

      Pretty ballsy to react to someone who called their shot and then backed it up by saying that their experience is just survivorship bias.

      Like, are there any records of these other people trying this and failing? If you don’t need those for your disproof then it would work on anything. (Sure, everybody who I’ve seen called a cab has succeeded, but the process may just be X / much bigger X likely to succeed. There may by billions who have failed and left no mark!)

      • Deiseach says:

        Like, are there any records of these other people trying this and failing?

        And that’s the big question, isn’t it? Are there any others who tried the Polgar Method? This guy is trying to run something allegedly based on the method, but to be frank the whole thing looks slightly dodgy to me. However, it is a chance for parents or prospective parents to sign up and turn Baby into a certified genius!

    • gwern says:

      Amusing example: 2 days ago I was cleaning up some predictions on PredictionBook.com. One of them was about a guy who declared he would do 10,000 hours of practice and join the PGA Tour as prophesied by Gladwell-Ericsson. I had forgotten all about him. I looked him up and… yeah. He totally failed. Never even got close. Gave up a few years ago.

      On a side note, that Polar book really is hard to find. I am doubtful that it was ever published in English: I can’t find any indications on the used book sites that they’ve sold it, the first few media mentions of it do not quote from it or indicate the English-speaking author has ever read it, the only entry in Worldcat gives its location as the Netherlands, there are multiple threads of everyone failing to find it and no one mentioning possessing it, one person says its ISBN was ‘recycled’ (there is no ISBN in the Worldcat entry), and the title in Worldcat and quoted elsewhere is super suspicious: “You may bring op a genius, bring up a happy person!” It’s not just blatantly ungrammatical, it even misspells ‘up’. Maybe it was vanity-published in Hungarian and then a cataloguer mistranslated the title for indexing. I suppose you could get a Serbian, Hungarian, or Esperanto copy and teach yourself the language if you really want to read it…

      (Speaking of rare psychology books – on a more positive note, I did manage to get my hands on Rosenthal’s Experimenter Effects in Behavioral Research and scan it: https://www.dropbox.com/s/nvoounyzv2412gf/1975-rosenthal-experimenterexpectancyeffects-ch3.pdf https://drive.google.com/file/d/0B6xS2SGLSIsPdVJDZmFZTXprbms/view?usp=sharing Unfortunately, almost everything in it is wrong, but still interesting.)

      • rlms says:

        I’ve searched my university’s copyright library and it’s not there, so it presumably wasn’t published in the UK.

      • Steve Sailer says:

        “One of them was about a guy who declared he would do 10,000 hours of practice and join the PGA Tour as prophesied by Gladwell-Ericsson. … He totally failed.”

        On the other hand, he got down to about a 4 handicap which might be at the 95th or 97th percentile (I’m just guessing) of all golfers. He became a much better golfer than I’ve ever been.

        My guess is that to make the pro golf tour you need about 10,000 hours of practice before, say, your early twenties. Back in the 1980s, there were two fine tour pros, Calvin Peete and Larry Nelson, who hadn’t started golf at all until their early 20s. But I haven’t heard of any since then. It seems harder now to start as an adult and make the pro tour because of increased competition.

        Note that there are a lot of superstar team sport athletes who are golf nuts (such as Michael Jordan) who would love to play on the Senior Tour for 50+ golfers. But former 49er QB John Brodie is only ex-jock to win even a single Senior Tour tournament. (We’ll see how retired Dallas QB Tony Romo does in his golf ambitions.)

        So, I think 10,000 hours of practice in golf (especially at an early age) is close to a necessary condition, but it is not a sufficient condition.

        • Steve Sailer says:

          Koreans have come to dominate the Ladies Professional Golf Association but not the Professional Golf Association in this century. That’s probably because intensive nurture goes further in women’s golf because fewer girls than boys are naturally motivated to become fanatical golfers. So a culture like contemporary South Korea that fanatically nurtures young golfers has a bigger impact on women’s golf than men’s golf.

      • sinxoveretothex says:

        As to your point about Worldcat, it may be that Polgar wrote two books, at least based on what I can peace together from this Reddit thread from 5 years ago: https://www.reddit.com/r/todayilearned/comments/sq0dg/til_a_hungarian_psychologist_wrote_a_book_on_how/c4gcoy9/?context=2

    • jhertzlinger says:

      Boris Sidis tried something similar while raising William James Sidis: https://en.wikipedia.org/wiki/William_James_Sidis

  2. Deiseach says:

    Yeah. The trouble with “anyone can raise a genius”, and I have no idea if this applies at all to Laszlo Polgar (and by the bye, what about his wife? I never see her name mentioned, but surely she had some input into the raising of her children beyond “provided the ova and womb to gestate them”), is the lurking suspicion I have that “anyone can raise a genius – if they act like a tyrant, control every moment of the child’s life, drill them like a colour sergeant, make it very clear that parental love, support and approval is indeed conditional – daddy/mommy will only love you if you learn to beat adult chessplayers when you’re not old enough to start regular school yet”.

    I’d love to hear stories of “well, we just decided we’d play it by ear, and little Wunderkind showed evidence of being interested in [six different subjects] but we certainly didn’t want to force them like a potted plant in a greenhouse so while we tried to be supportive we were determined to let Wundy have as ordinary a life as possible”, and I’m sure they happen (Robert Browning appears to be one of these kids who was extremely bright at a young age, and had an indulgent father who gave his kid every opportunity to educate himself but didn’t push him), but a lot of stories seem to be too much in the mould of “forcing my eight year old daughter to play tennis for hours at a stretch and screaming abuse at her if she got tired or hungry or didn’t hit the ball perfectly every time”.

    EDIT: Okay, the post says the Polgar sisters loved chess and were happy, and Susan says she picked it out for herself, but I still do wonder – what would happen the experiment if little Susan wanted to play with the chess toys but Dad said “No, you are going to be the female Hungarian Mozart, so it’s the piano for you!” Would he have bent to a four year old’s will and let her learn chess instead? Maybe?

    • Scott Alexander says:

      Everything I’ve read about the Polgar sisters makes it very clear this didn’t happen. Their parents were loving, they said they had really happy childhoods, and the only fact I can find about Polgar’s educational technique was that he’s against any kind of punishment or anything that seems harsh or makes children love learning less.

      As for the “control every moment of the child’s life” part, read Part V.

      • Deiseach says:

        As for the “control every moment of the child’s life” part, read Part V.

        I admit that this seems to be one of the cases where the parents were not over-bearing and the kids turned out okay in adult life. But it is also, unfortunately, the case that “stage mothers” and “competitive dads” exist and try to force their children into the mould of achieving success for them.

        It would be great if there were some fool-proof and painless method of teaching any child of average intelligence how to be a prodigy at some subject, but I don’t think we’ve got the fool-proof method yet thanks to Lazlo Polgar. The lack of an easily accessible version of his book is surely evidence in that direction; he seems to have wanted to produce an English version and, as you say, with three chess genius daughters that should surely have attracted the interest of the publishers who churn out self-help books such as the interminable Chicken Soup series, yet nobody seems to have been willing to bite. Given that people will happily hand over money for the latest fad, gimmick and guru, the lack of interest may indeed be scepticism about “yeah but does his method really work?” and a strong suspicion that it’s a combination of lucky gene combinations and parental involvement that produced the results, which are not readily replicable by Mr and Mrs Smith in Rustville with their 105 IQ kids Tommy and Janice simply putting the “Ten Easy Steps!” into practice.

        • Muad'Dib says:

          For a chess player who was forced by his parents (his father) to play chess, achieved a lot of success (nearly became world champion and came close a couple more times), gave up chess, then returned of his own will (but not nearly as strong as before): see Gata Kamsky.

      • Steve Sailer says:

        There used to be a little kid who hit thousands of golf balls at the driving range in my neighborhood named Anthony Kim. He became a phenom on the PGA Tour, winning tournaments in his early 20s and being called the Next Tiger Woods.

        Today, at age 31, he is officially retired from golf.

        When other Koreans ask his parents how to raise a star golfer, they reply: “Don’t.”

      • Jiro says:

        Everything I’ve read about the Polgar sisters makes it very clear this didn’t happen.

        But in that case, isn’t it a pretty huge coincidence that the kids all happened to like chess? If the children have preferences for a randomly chosen smart-kid activity, and the parents pick a smart-kid activity arbitrarily, the odds are pretty low that they’re going to pick the same activity for a kid that she likes–let alone for three kids.

        I wonder how much of this is
        1) The kids convincing themselves they liked chess, when they really liked something else more (basically, the opposite of sour grapes, where something they had to do looks better, instead of something they can’t do looking worse), and
        2) The kids only liked chess within the specific environment where parental praise, etc. was greater for chess than other subjects, making them like it more.

        • Chalid says:

          Little kids often want to emulate big kids. So once the eldest daughter got into chess it became much more likely that the younger ones would show an interest.

          • albatross11 says:

            Also, once you have a sibling or two who is really good at chess, you also have a practice partner who will push you. I suspect the Williams sisters benefitted from that.

          • Deiseach says:

            Little kids often want to emulate big kids. So once the eldest daughter got into chess it became much more likely that the younger ones would show an interest.

            That Chronicle of Higher Education article seems to lean that way; Susan was getting coaching (which meant a lot of attention) from Dad, and the younger Sofia and Judit used to watch them and want some of that themselves. And then Judit had Susan as a coach and practice partner, which perhaps spurred her on to achieve more.

            Laszlo Polgar was a chess teacher, so having a chess set lying around the apartment that little Susan wanted to learn to play with was no more a coincidence than a football player’s kids finding a ball and wanting to kick it around.

          • Sivaas says:

            He only became a chess expert/teacher after his daughters took up the game though, right?

    • Simon says:

      I know quite a few kids/families like that, since it is my job to make sure they get the best possible mathematical education; I’m trying my best to be the László Rátz of the San Francisco Bay Area, and I’m learning as much as I can about how to do that as well as possible. For example, these people are good friends of mine: http://www.davisenterprise.com/features/next-generation/gifted-siblings-capture-the-world-around-them/. While the parents are incredibly supportive of their kids’ unusual educational needs and do everything they can to provide the resources that are helpful, they aren’t pushy at all.

    • wintermute92 says:

      Would he have bent to a four year old’s will and let her learn chess instead? Maybe?

      Can you clarify that edit?

      I can’t seem to parse it as anything other than “ok, these kids were happy and not abused, but wouldn’t it be bad if he had abused them instead?” And it feels like there’s a better point than that, which I would like to understand. Possibly something like “does this fall apart if you try to direct it?”

      • Jiro says:

        I would parse it as “this situation is very unlikely to happen except by abuse. The fact that it happened without abuse this one time isn’t going to generalize. Also, there could have been factors that are not abuse, but are a step in the direction towards abuse.”

        • It seems to me that the way it happens without abuse is one of the directions unschooling can go. The kid is very interested in something and the parents, instead of pushing the kid to do lots of other things, supports and encourages that interest.

      • Deiseach says:

        As I said, all too often these cases turn out to be “Dad or Mom wanted me to be a famous athlete/actor/vaudeville star and pushed me into it, and I was too young to know my own mind or stand up to them”.

        So maybe little Susan did want to learn to play chess herself, and I won’t call her a liar if she says that is how it happened. But I do wonder about the happy coincidence that the daughters of a chess-teacher father all independently decided they wanted to play chess more than anything in the world – if we’re discounting genetics and natural inclinations in favour of the allegation by Laszlo Polgar that “any healthy child can become a genius” and take it on face value: any average child, who is physically and developmentally normal, can learn anything at all to a level of excellence that is considered (popularly) “genius”.

        Suppose Mr and Mrs Polgar had decided to test their method by seeing if they could produce a child prodigy violinst or artist, and three year old Susan said “No I want to learn chess instead”. Would he have said “Okay, I’m giving in on this”, which would seem to contradict his theory that you can take a child and teach them anything? Or would he have insisted “Okay, you can learn chess, but music is the major thing here and if chess distracts you from that, you have to give it up”?

        I’m not accusing the man of being abusive! The three Polgar daughters do seem to have turned out to be normal, happy people and that is fantastic. But I do wonder how much of the family mythos Susan is retelling, and what way it really would have gone if she had balked at learning the subject her parents wanted to use as a test case.

        Jiro is correct: what I was getting at is that in this case, it looks like it was family interest in/aptitude for chess, Susan wants to learn to play with these funny toys, father teaches her and uses his method, instead of common “simplify it for her young age” and turns out chess prodigy, then repeats it twice in a row. I do think “three daughters all want to learn chess” undercuts his theory because it does point more towards aptitude for/interest in a particular topic, but it doesn’t seem to be “parents forced them unwilling or not to learn to play chess”.

        But I can imagine cases where the parents would try to use the Polgar Method to turn out a sports star or music prodigy, and never mind if the kid wanted to be a star athlete or baseball player or singer, the whole point of the theory is you can take any kid and turn them into a genius. Ambition and yes, greed, could easily over-ride the part of the theory that says “you can’t force them, it has to be something they’re interested in, and you must do it all with love and encouragement”.

        • Nabil ad Dajjal says:

          Do you remember Scott’s article on fake consensualism? Because the below sounds like that.

          But I can imagine cases where the parents would try to use the Polgar Method to turn out a sports star or music prodigy, and never mind if the kid wanted to be a star athlete or baseball player or singer, the whole point of the theory is you can take any kid and turn them into a genius. Ambition and yes, greed, could easily over-ride the part of the theory that says “you can’t force them, it has to be something they’re interested in, and you must do it all with love and encouragement”.

          If the Polgar Method works and it becomes common knowledge, more parents might decide to take the stage parent / tiger mom route and pressure their kids. So we should be very suspicious of attempts to popularize his methodology so that it doesn’t lead to tragic outcomes.

          Except that if the Polgar Method works and it doesn’t become common knowledge, a lot of children are going to lose out on the chance to truly excel in their future fields. That’s also a tragedy, not just for the children themselves but also for their parents and for humanity as a whole.

          So far you’ve only looked at the potential costs of adopting the Polgar Method but haven’t considered the opportunity cost of not adopting it.

          • Jiro says:

            Chesterton’s Fence would suggest that we should not use such methods until we understand why we haven’t been using such methods all along and can articulate why that reason no longer applies. (And the reason is probably *not* “nobody thought of it” or “people are idiots and are obviously avoiding it because of bias”.)

            Also, it’s easier to break things than to fix things, so I would be inclined to think that bad effects from pressure would be bigger than good effects from possible success.

          • sconn says:

            It seems obvious why we haven’t – parents haven’t had the leisure. Even basic education for one child is a big job. Genius level education, especially for a larger (i.e. pre-contraception) number of kids, isn’t even considered ​by most parents because they’re tapped out just trying to teach them all to dress themselves and read and so on.

          • Deiseach says:

            So far you’ve only looked at the potential costs of adopting the Polgar Method but haven’t considered the opportunity cost of not adopting it.

            Because I am sceptical of the method as claimed. The success of the Polgar sisters is a great story, and we all love stories. This is the “you too can raise a baby genius” story, and I think there’s more to it than simply “pick an average, healthy child and raise them with this Method and you can turn out a guaranteed genius in any field you choose”.

            From the bits and scraps that are around online, it seems like even Laszlo Polgar includes that you have to wait until the child begins to show an interest in something or an aptitude for something before moving into the intensive coaching phase. So it is more complicated than “pick any child, any child at all, pick any field, any field at all, and this Method will reliably and guaranteed turn out a star in it”.

            But we humans like the simple stories that are reduced down to “One Weird Trick”, and we prefer the idea that you, Mr and Mrs Jones, can take your average kid and turn him or her into the next art world superstar (okay, if Jeff Koons can do it, any average six year old can do it) or opera star or top athlete or whatever, regardless of whether they have the interest, talent, capacity or whatever for it. You can’t turn a child who can’t sing* into the next Pavarotti, for instance, no matter what method you use.

            But the simple story sells, so if the Method gets pushed as “you can turn your kid into a star!”, then people will try it. And half-try it, giving up halfway through (because it demands a lot of commitment from both parents). And it also demands that one, at least, of the parents is able to teach a child whatever particular subject necessary, to a very high level. Many people won’t have the time or ability to do this.

            It may well be the One Simple Method to reliably turn out geniuses. But for a mass approach, it can’t rely on “is your father able to coach chess/your mother able to teach ballet?”, it’ll have to bring kids into centres where specialists can give them full, personal attention and submerge them in the particular topic.

            Which is what boarding schools, for one, were supposed to do. There might be a chance for “Polgar Method Schools” to get off the ground, but as a programme for parents to intensively home-school their budding genius – it’s never going to be workable for more than a small percentage of the population, who (a) have the ability to coach, or can learn it themselves, to a high level (b) will put in the kind of “eight hours a day supervising learning” on top of home school teaching, the several hours of table tennis the Polgar girls played (for fitness purposes), and every thing else involved in providing the immersive environment. That’s going to need one parent to work outside the home to earn money to keep the household (probably what Klara Polgar did, as she was a teacher) and again, that’s a level of commitment and investment that produces its own pressures for a payoff (if Junior, after all this time and effort, can’t start showing results in money and success, it’s going to seem like a waste of time).

            I probably am pessimistic about it, but a lot of pedagogical reforms and new techniques have come down the pike and there’s never been One Simple Method discovered yet. I think the Polgar Method relies on a combination of (a) talent in a particular area, which can be the result of lucky genetics (b) intense commitment by both parents over years (c) reliance on the ability of the parents to teach as much as on the ability of the child(ren) to learn.

            *I don’t mean “oh, I can’t sing” in the usual way that’s used, where people can sing a little but don’t feel they’re ‘good enough’, I mean someone who’s the next best thing to tone-deaf or has no sense of pitch or whose voice is thin or who hasn’t got the range.

          • And it also demands that one, at least, of the parents is able to teach a child whatever particular subject necessary, to a very high level.

            Unless I’ve missed something, none of us has actually read Polgar’s book, so we don’t know in any detail what his method is. I don’t think anyone has suggested that he played chess at the level of his daughters.

            More generally, encouraging a kid to follow an interest up to an expert level doesn’t require the person doing the encouraging to have corresponding expert knowledge. Once the kid is well started, you can point him at books, let him audit college classes, or learn things in any of a variety of other ways.

            Beyond that, your main criticism seems to be “if people misunderstand Polgar’s approach, they could do damage.” That’s probably true–just as it’s probably true of your approach to various things, or mine. But it isn’t an argument against following his approach, only against doing a bad job of explaining it.

  3. Doctor Mist says:

    Interesting. But doesn’t

    This is biologically impossible. Even if both Polgar parents were 170-IQ themselves, regression to the mean predicts that their children would have IQs around 140 to 150.

    overstate the case somewhat? Regression to the mean is statistically true, but surely there’s room for a tail. Granted, we have to include the long odds against the family that exhibits the tail also being one whose story resonates so well with a discussion of IQ. But maybe there’s a selection effect at play — if his kids hadn’t been outliers, we would be talking about something else. North Korea, maybe.

    Of course, I could be entirely underestimating what the statistics say about the odds here. If the chance of his daughters being that smart is one in ten billion, then forget what I said. (I might still fall back on anthropic reasoning — we might expect to see it on one inhabited planet out of ten or a hundred or a thousand, but this is the one we saw it on. But since I use the Fermi Paradox to argue that there are not ten or a hundred or a thousand inhabited planets, and the Anthropic Principle to explain why there’s one, it would be hypocritical of me.)

    Edit: Well, I see I was not the first to balk at this point.

    • Scott Alexander says:

      If we believe Kasparov’s IQ 135 estimate (and if we believe the study showing r = 0.25 correlation between IQ and chess ability, why shouldn’t we?) then why should we even have this discussion? We know IQs in the 130s are enough to be a chess champion, and it’s possible for the Polgar sisters to have had them.

      (unless chess talent is heritable in a way separate from IQ, but I’m not sure we need to bring that in here.)

  4. sandorzoo says:

    Polgar’s book is listed for sale here, but I tried ordering it and it never arrived. Maybe they’ll only ship to within Hungary?

    http://www.numero7.com/konyvesbolt/9789630958226/polgar-laszlo/nevelhetsz-zsenit-.html

  5. bean says:

    There are copies of Nevelj zsenit in several libraries in the US. Someone with interlibrary loan access should be able to get one for free, or you could ask about rates at your public library. I have doubts that Bring Up Genius even exists, as Worldcat has one copy listed, at a Dutch library that also has Nevelj zsenit. Worldcat also has no sign of the Esperanto version.

  6. Deiseach says:

    Very interesting article I found while trawling for any mention of Polgar’s book. Looks like nature crept in even under nurture’s rule:

    The official story, as Susan tells it, is that she chose chess. Tottering around their cramped Budapest flat as a 3-year-old, she pulled open a drawer and found a set. “Let’s play!” she said to her mother. Klara didn’t know how. “Wait till Dad comes home,” she said. When Laszlo returned, he was delighted to see Susan’s interest. He was no chess expert, but this could be the venue for his experiment. It was cheap and intellectual; it shouldn’t matter if you were a girl or boy, tall or short. Plus, the game was so dominated by men that a talented woman might achieve great acclaim.

    It’s a narrative that suits the family, which was long hounded by suspicions that the girls had little choice in their upbringing. Sofia, the middle sister, complicates the story on the phone from Israel, where she lives now. Their father, she says, always had chess in mind. He also had Susan study math, but when they needed to pick a specialty, he encouraged chess, Sofia says. The ranking system was one draw; it’d be easier to measure the experiment’s success.

    …By 1988, Judit’s ascension was clear. She was a risk taker, an overpowering attacker; Susan was more defensive, while Sofia, also aggressive, was distracted by a blossoming interest in art and design — she was often caught reading books in the bathroom late at night. That year, Judit won the World Under-12 Championship, the first woman at any level to win an overall world championship. Months later the sisters accounted for three-fourths of Hungary’s team in the Women’s Chess Olympiad; “Polgária,” people called them. Judit dominated, winning 12 of 13 matches. She was on her way to becoming one of the world’s top players, man or woman. The family went home heroes. Passers-by stopped them on the street. It seemed Laszlo had discovered the secret to success.

    “Everyone said, OK, your dad is not a lunatic or just this weirdo who wants to ruin the life of their children,” Sofia says. “Actually, it’s something very special that he’s doing.”

    …Questions will always chase the Polgárs; such is the price of their peculiar type of fame. But for a time, at least, it looked as if they might answer the biggest question of them all. The family has always known that genes undermined their can-do message. Laszlo and Klara were smart and determined. Isn’t it natural their girls were, too? In the 1990s, to test that, Laszlo almost began raising another set of kids; an eccentric Dutch benefactor would cover the costs. But the plan foundered due to Laszlo’s insistence on adoption. He feared raising prodigies, then not benefiting from their success, Susan says. And Klara made it clear: She did not want more kids. No more experiments.

    …Whatever the source of its success, the Polgár experiment will last only one generation. None of the sisters has raised her children in the same fashion. All the kids attend school. For Sofia, it was important for her two boys to learn chess for its life lessons — making decisions under pressure, avoiding paralysis by analysis. But they didn’t have to be champions. Eventually the boys lost interest. She didn’t push them back into it.

    “I also enjoy having a life, you know,” Sofia says. “For my parents, this was everything.”

    • Steve Sailer says:

      At the extreme high end of achievement, nature and nurture tend to both play major roles. For example, the two top American golfers of the previous generation, Tiger Woods and Phil Mickelson, both started swinging a golf club before they were two. There’s video of Tiger putting around with Bob Hope and Jimmy Stewart around age 3. When Tiger won the Masters at age 21 in 1997, the local sportscaster in L.A. mentioned that he first covered Tiger 18 years before.

  7. aeq says:

    This is the article about Kasparovs IQ. His results were 123 (Raven-matrices) and 135 (Eysenck). http://www.spiegel.de/spiegel/print/d-13526693.html

    • Scott Alexander says:

      Thanks!

      • pseudon says:

        According to Kasparov’s recent book Deep Thinking he never took a real IQ test and the questions from the Spiegel journalists partially bored him.

        Also, I got to meet the man today. Funny coincidence.

    • Diadem says:

      I’m kind of skeptical about these scores.

      Chess is clearly a mental activity. Even if whatever mental activity chess requires isn’t 100% correlated with IQ, it would be extremely surprising if there was only a weak correlation. For example some have suggested that memory is more important than pure ‘thinking power’. But surely memory also strongly correlates with IQ?

      And Kasparov is not just any chess player, he’s one of the strongest of all time. For him to have an IQ that’s only barely in the top 2% would be extremely surprising.

      It also doesn’t match up with my experience. I used to play a lot of chess when I was younger, and chess talent and intelligence (as estimated by interacting with people) always seemed pretty closely correlated. All strong youth chess players I know breezed through high-school without any effort. I would guess their IQs to be north of 135.

      People often grossly overestimate how high an IQ of 135 is. In The Netherlands an average VWO (highest tier) high school class will have 2-3 students with an IQ of 135 or higher. People think this is a truly exception IQ, but it’s really only “gets through high school without any effort and above average grades, or with effort and high grades” level.

      • albatross11 says:

        I think we have two different things that are correlated, but not necessarily all that strongly:

        a. IQ is pretty strongly correlated with what we mean when we talk about intelligence, but it’s not identical. If you find our that Feynman had a 120 IQ (which I’ve seen claimed), you don’t think “Wow, Feynman wasn’t all that smart,” you think “Wow, IQ tests don’t capture all of what we mean by intelligence.” Similarly for Gary Kasparov.

        b. Chess playing ability is somewhat correlated with IQ, and probably somewhat more strongly correlated with what we mean when we talk about intelligence. But there is a bunch of other stuff that goes into success in chess–practice, personality traits that help you train effectively, interest, etc.

      • One problem with using high school IQ figures is that, as I understand it, there are two different definitions with different distributions. One definition, mental age over calendar age, only works for children. The other, defined by number of standard deviations from the man, works for adults.

        When I was a councilor at a camp for gifted children, one of the kids had a measured IQ of 201. He was a very bright eleven year old and I can easily believe that he was as smart as the average twenty-two year old. But as I understand the distribution, it is unlikely that he was as far out on it as 201 ought to be because the expected number of such eleven year olds in the world at one time is less than one.

      • nimim.k.m. says:

        All strong youth chess players I know breezed through high-school without any effort. I would guess their IQs to be north of 135.

        Either your high schools are very difficult in comparison to ours, or I think you might overestimating wildly how much IQ really is needed to “breeze through high school”.

        For example I had remarkably easy time because I already knew most of the things in the curriculum (because I was interested in quite much of things ranging from history to physics and read a lot as a kid), but my IQ isn’t anything too much to talk about. I’m also terrible at chess, mostly because despite several attempts I could not get myself to develop any interest in it at all.

        Also, chess is culturally coded as the “smart person hobby”. That’s going to have an effect what sorts of people you’ll meet in a chess club. One probably needs some minimal amount of the thing measured by IQ to be a good chess player, and extra points probably help, but memory and amount of practice and the stuff not measured by an IQ test is going to help at some point far more.

  8. Tibor says:

    The Dutch Royal library in Hague should have a copy if that’s of any help

    • Deiseach says:

      This bloke allegedly has a précis of the principles in the book (he’s trying to flog his own video course series on How To Bring Up A Genius) but he has a Youtube review of the book and the basic principles of Polgar.

  9. topynate says:

    The Esperanto version is scanned and on the web. Anyone who speaks a few words of Esperanto will be able to find it in a few seconds.

    • Toby Bartels says:

      If Esperanto is easy to find and English is unavailable, then you should definitely go for this, Scott. If you must learn Esperanto first, the phone app for [Duolingo](https://www.duolingo.com/) is an easy way to do your 10,000 hours in your spare time, but Esperanto is so easy that you might be able to just sit down and look at it. You know medical terminology, so you know Latin roots, so 84% of the vocabulary should look familiar if you squint; and you can read [the grammatical rules](https://en.wikipedia.org/wiki/Esperanto_grammar) once, in confidence that there are no exceptions.

      • Michael Watts says:

        You know medical terminology, so you know Latin roots

        Medical terminology shouldn’t help that much with Latin roots. Like most sciences, it tends to draw more from Greek roots.

        Compare the actual medical terminology “hysterectomy” hyster-ec-tom-y womb-out-cut-noun to a hypothetical Latin “uterexsection” uter-ex-sect-ion, with the caveat that Latin doesn’t really go for the creation of compound words like that.

    • Winter Shaker says:

      The admittedly non-free language app that I use, which I find a lot more useful than duolingo, has also recently added Esperanto to its list of supported languages.

      Though I’m a little curious why our host emphasised Esperanto out of all the languages that the sisters learned – surely, as a language consciously crafted to be easier than any natural languages, it’s the one that would have taken the least genius to learn?

  10. dustinwyatt342 says:

    I’m a little uncomfortable with the idea that neglecting other subjects in favor of excellence at one subject is a laudable goal with regards to The Pursuit of Happiness and The Good of Society. I mean, I’m not sure that’s _wrong_, but I’m not sure it’s _right_, either.

    • Jules says:

      They seemed pretty happy about it and I would expect Society to benefit from a wider availability of high-performing specialists as well as from more polymaths.

      As a more polymath-inclined type, it does *feel* wrong to me, but for utility I lean towards going against my own instinct on this.

  11. RLM says:

    Here’s the polish version, available on WorldCat. I’ll have my friend request a copy through MIT (unless you can request it easier yourself?). Still looking for the English version, which is proving somewhat harder to find.

    http://mit.worldcat.org/title/nevelj-zsenit-farkas-endre-interjui-zsuzsa-zsofi-es-judit-valogatott-versenyjatszmai/oclc/25202806

    • bean says:

      This has already been noted upthread. I’m pointing this out mostly to ensure that we don’t end up confusing ILLIAD by there suddenly being a 2-year waiting list for copies. James Miller has made a request.

    • No signal says:

      I cannot find anything Polish in that. Just Hungarian line-noise (for a Polish native speaker…)

  12. RLM says:

    I have the esperanto version as a pdf. Please email me directly if you’d like the copy.

  13. Come on, Esperanto-speakers! This is the only chance you’ll ever have to be useful!

    If you want help from Esperanto speakers, maybe you should refrain from sneering at them.

    Admittedly, they’re easy to make fun of.

    Back when I was playing around with constructed languages, usefulness was not a concern: in my youth, I tried to learn Volapük.

    • Tibor says:

      At the end of the day it is nobody’s business to tell other people which hobbies they should have, but I’ve never understood why anybody would want to learn an artificial language when there are so many natural languages to learn with all their lovely idiosyncrasies and quirks. Particularly Esperanto which is, given its simplicity, probably not all that interesting or beautiful. Although somehow it still seems better to me than learning Klingon or something like that, I’m not sure why it does, probably because Esperanto is a “serious” language. Of course, in terms of usefulness learning Welsh or Icelandic is not any more useful than Esperanto, and probably a lot harder. But learning a language that evolved over time somehow feels a lot more real to me than a language which is a product of intelligent design. Maybe it is because I love to look for common etymologies with other languages and historical reasons for why a particular language is the way it is, whereas artificial languages are sort of soulless in this regard.

      • Bugmaster says:

        Well, you could apply this logic to any other hobby, and honestly, I can’t reasonably disagree with you. For example, why would you spend your time on playing collectible card games, when you could spend that time on part-time work (to get more money), exercise (to get stronger), study (to get smarter), etc. ? In terms of utility, hobbies are a losing proposition.

        • Tibor says:

          I am not so much saying “why not do something more useful” but rather “why not do something which is pretty much the same yet more useful” and “what do people find interesting about artificial languages in the first place?” – which is not “I don’t enjoy it, therefore it is objectively boring”, I am genuinely interested in what people find interesting about them (or Esperanto specifically, other artificial languages are even less popular).

          • vV_Vv says:

            Most constructed languages were invented before English became the de facto lingua franca, at a time when multiple languages (English, German, French, Spanish) were competing for that position.

            Constructed languages were supposed to have speculative benefits such as being easier to learn as second languages due to their mixed-origin vocabularies and being more politically neutral by not being associated to any specific ethnic group or nation.

            More speculative benefits included having simpler, more elegant, more “rational” grammars (according to whatever linguistic theory that the language creators subscribed to), not encumbered by the irregularities and idiosyncrasies that natural languages accumulated through millennia of organic evolution, which should have made constructed languages easier to learn, more precise, less prone to ambiguity, etc.

            In practice these supposed benefits mostly failed to pan out, America won WW2 and the Cold War and now everybody happily speaks English while grieving about colonialism.

            I suppose the interest in constructed languages is due to a similar psychological mechanism that causes interest in evenly-spaced rectangular grid cities. It’s something that frequently occurs in engineering: the desire to get rid of complex, idiosyncratic, legacy systems and design everything anew from the ground up, based on more “rational” principles.

            It even occurs with programming languages. Even though all of them are constructed languages, some are undeniably more constructed than others. Why do people spend their time on beautiful but impractical languages like Haskell or Lisp when they could just hack things together with Perl?

          • Bugmaster says:

            @Tibor:

            “why not do something which is pretty much the same yet more useful”

            Isn’t that the same question, though ? Hobbies are totally useless in general, so I think what you’re really asking is, “what makes a person choose one hobby over another ?” Well, I don’t know. I know one person who greatly enjoys collecting small animal figurines. I find them super boring; I enjoy taking nature photos, and that person finds them super boring. It’s just personal taste. One day, we’ll be able to map human brains and figure out exactly how personal taste develops, but for now, it’s fairly arbitrary.

            That said, see another note at the bottom; but first:

            @vV_Vv:
            You may be thinking specifically of Esperanto; but other constructed languages, such as Klingon, Quenya, or Dothraki, have other goals. Usually, the primary goal is to provide some flavor for some fictional race: by presenting their fictional language, the author can demonstrate how their way of thinking differs from ours. Simplicity of grammar is not a consideration, and in fact may be a detriment, since it makes the language feel too artificial.

            In fact, IIRC Tolkien developed Sindarin and Quenya specifically to illustrate the historical split between the Noldor and the Teleri: they used to speak the same language, but then they split up into two separate groups, and their languages diverged.

            Similarly, the Klingon language isn’t supposed to be easy to learn; it’s supposed to sound angry and warlike. It lacks words such as “please” and “thank you” not because someone forgot to add them to the dictionary, but because such concepts are alien to the Klingon way of thinking. You may say “well done”, but that is a statement of fact, not small talk.

            On the other hand, the languages of Westeros and Essos are designed to be somewhat related to each other, harking back to the time when everyone in the world spoke Valyrian (and maybe even before that). History is a very important part of the series (mostly because all the major players keep forgetting it and end up repeating it), and the languages are constructed to reflect that.

            @Tibor, again:
            I mushed these two replies together to show that there’s actually some depth and creative thought behind constructed languages. Constructing a fictional language can be a part of the creative writing process. I personally am not interested in actually doing it, but I can appreciate the amount of effort involved. That said, it’s still all objectively useless, unless you happen to be one of 10 people in the world who gets paid to construct all these languages. A hobby is still a hobby, in the end.

          • Tibor says:

            @Bugmaster: Ok, I guess when you put it that way I can actually find it more easy to understand why one might pick up a “story-based” language (and even more why one would want to construct such a language) rather than something like Esperanto which is essentially a failed experiment.

            I don’t think the question is the same though. Maybe it is because I find so many languages to be interesting that I have to discriminate a lot in which ones I choose to learn. I’d love to speak (in no particular order) Icelandic, Welsh, Cantonese, Japanese, Mandarin, Croatian, Dutch, Rumantsch, Occitan, Persian, Italian, maybe Hungarian on top of the 5 languages that I do speak and probably there are some other languages that could also easily catch my interest (Nahuatl seems cool 🙂 ). But I write “I’d love to speak” rather than “I’d love to learn”, because that is obviously not the case. I might still learn one or two of them (most likely something Indo-European) once my Portuguese and Spanish are at least as good as my German and my German as good as my English, but I definitely won’t learn all (even if I did, maintaining fluency in so many languages would cost an incredible amount of time). When a language is made up and its history and everything around it is hence a lot less rich than that of any “real” language, that puts it well below all of these other languages. Of course, in practical terms, Rumantsch isn’t much more useful than Elfish (or I guess Elvish, when it’s Tolkien?)

          • sconn says:

            In my experience, as a person who has studied Sindarin, Quenya, and Klingon, plus some dead languages (Latin, Greek, and Anglo-Saxon) the main driver is a love of patterns. Constructed languages are often very interesting in their patterns, and you don’t have to worry about very many exceptions. It’s also a bit of a shortcut – you can master all there is of Quenya in a few months, while Chinese will take years and you still won’t sound like a native.

            There’s something kind of cool about knowing something hardly anybody else knows. When I was studying Elvish languages online, on the early days of Internet forums, it was a huge thrill to me as a high school kid to be told by the adults that I was an expert and to get all this positive feedback on my poetry. Doubly so since I knew very few people in the world could even read it. Whereas when you learn a real language, you’ll always be one of the less proficient speakers of it since you’re competing with native speakers.

            It is unsurprising that I grew up to be a Latin teacher. I also know Spanish but there are zillions of Spanish teachers out there, while Latin teachers are in high demand and you don’t actually need to be fluent.

          • Tibor says:

            @sconn: I can understand why one would want to learn a dead historical language, in fact I tried learning Latin by myself as a teenager (my school did not teach it but a few Czech gymnasiums – the proper English term would be a grammar school – do and so there are school textbooks of Latin easily available) but I stopped quite early. I might still pick it up someday, not aiming for fluency but at least some understanding (what bothers me about the way Latin is taught is that it is not taught as any live language, instead the focus from the very start on grammar and things like that which i prefer to learn in smaller doses so that it becomes more or less intuitive. I’d pick up a Duolingo-style Latin course any time).

            But somehow historical languages seem more real to me and therefore more worth it. There is something magical about being able to read actual historical texts written by people who actually spoke the language (although classical Latin with all its rules and complex grammar was probably not really used in everyday speech by almost anyone, I bet even Cicero used a lot simpler language when just chatting with friends, but anyway). Latin is also amazingly interesting in that it provides roots for a lot of modern “sciency” words (and of course most words in Romance language roots and a lot in English as well) and I find searching for common etymologies if various Indo-European languages very interesting (it is also fun when you realize that all so fancy words like the Latin names of various bones literally mean something like “big bone in the back”). Learning vulgar Latin romance languages helps with this already, but if you really want to go back to the root, then you have to learn Latin (and Greek for other sciency words).

            With made-up languages, well, it just feels too artificial to me. I guess that if you really like Tolkien’s work a lot, learning the Elvish language can allow you to immerse in the story even more and if it really takes just a few months to be proficient in it, it is not nearly as big a commitment as learning Latin or even Cantonese.

            I guess there are also two reasons one might like learning languages – the vocabulary and the grammar. Grammar is usually not all that interesting to me – when it is, it has to be something unique about the language like the use of subjunctive in romance languages (or the use of estar/ser which I still haven’t entirely mastered yet), konjunktiv I in German or grammatical aspect in Balto-Slavic languages. The fact that Finno-Ugric languages use a multitude of cases instead of prepositions is not all that interesting since it does not add any new way of expression (cases themselves are fine, but you don’t need 17 of them or how many to get the idea 🙂 ). But I doubt that artificial languages have any such cool unique grammatical features.

            Then you have the vocabulary which I find interesting for how words relate to each other and to words in other related languages (all the way up to large language groups like Indo-European). I guess you can emulate that to some degree in artificial languages (although since they do not belong to any language groups, they will always miss the connection to other languages) and create something interesting if the author is a good linguist (like Tolkien). But from what I heard about made-up languages like Klingon, they seem very one-dimensional (the “no word for thank you” is more cheesy than interesting).

            I guess I can now understand better why one might be interested in picking up a made-up language though, even if it’s not my thing.

          • sindikat says:

            There are plenty of ways to extract utility from Esperanto. Esperanto has a natural selection mechanism, so an average Esperantist tend to be more humanistic/altruistic/progressive and generally more interesting, compared to an average speaker of any other language. Esperantists are eager to speak with other Esperantists, so you’d find that it’s so much easier to find friends online, and you don’t need a specific justification, than if you just know English. Esperanto speakers are more regularly spread around the world. Without Esperanto I wouldn’t meet a Swiss family traveling across Asia on a trailer. Or a Japanese magician. Esperanto is based on Romance languages, so you can trace common etymologies and by being simple and regular, it reveals, rather than obscures, interesting patterns that can motivate a person to study linguistics. Some people claim learning Romance languages becomes easier after learning Esperanto.

            Is Esperanto a failed experiment? Sure, whatevs. So is LessWrong, but it would be stupid not to read the Sequences, because some random pretentious hipster on the Internet said it’s uncool.

            One thing that eludes people is that Esperanto is ridiculously easy to learn. This factor actually changes a lot. Esperanto is not just marginally easier than an average European language. It’s literally an order of magnitude easier. I study English for decade and a half now, and I still struggle with the bloody language, despite using it 24/7. I learned the basics of Esperanto in a month and spoke almost fluently in half a year.

            This means learning Esperanto is not a laborious investment. It’s something that you can devote no more than half an hour per day and still progress much further, than in any other hobby, be it learning a language, learning to code, drawing, gardening etc.

            It’s the natural languages that are useless. They require unreasonable amount of effort, you never achieve true mastery even after years of devoted study, and by the end of the day they are useless, because all the interesting people speak English anyway, and all the interesting info is in English as well.

          • Tibor says:

            @sindikat: Well, the thing is that if you want to understand a culture you have to learn its language. If you want to integrate into a society somewhere you have to eventually speak it perfectly including colloquialisms and dialect words (this may be very important for example in Switzerland – a lot of Germans living in Switzerland complain that the Swiss are not very open towards them but they never bother to learn Swiss German – high German is not the native language there in any canton) and possibly get the accent right* . Natural languages are more than just means of communication.

            From what I gather, Esperanto is supposed to have Romance and partly Germanic vocabulary and Slavic grammar (obviously all of these are probably heavily simplified, especially the grammar – I don’t expect Esperanto to have cases or grammatical genders for example). At least that’s what Wikipedia says. The point is that no matter which country you’re from, at least as long as it’s language belongs to the Indo-european group, you will already find something in Esperanto which is similar to your mother tongue.

            *that’s the hardest part – I can get reasonably close with my accent in at least English and German (and maybe in Spanish if I try really hard) so that I can fool foreigners into thinking I might be a native but I’ve yet to fool a native speaker of any language that I speak…I guess I could get there fairly quickly if I lived in an English speaking country but I actually speak English mostly to non-natives whose accent is even worse than mine…and my German is not quite at the level where trying to master the accent makes sense since I still make mistakes, mostly in word genders – this is the most frustrating part of German since there are little to no rules for which gender a word has.

            EDIT: I tried reading some random Esperanto wikipedia articles and it really seems extremely easy. I could understand about a third of the text without ever having learned it. I recognize some Germanic and Slavic words although most are indeed Romance. I am also surprised it uses accents. That seems like an unnecessary complication. I really find the use of “kaj” very funny, since this seems to be the same word used in a Silesian dialect on the Polish-Czech border and when I see it I imagine the whole sentence in the sound of that dialect which is rather particular 🙂

          • sconn says:

            @Tibor – yeah, for me it’s the grammar. I loved starting Latin with big charts of nouns and verbs. Looks like you prefer the story behind the language more, and reading texts in it.

            Quenya is actually based on Finnish and has cases instead of most prepositions. It’s a really cool feature which I enjoyed learning and using.

          • I guess you can emulate that to some degree in artificial languages (although since they do not belong to any language groups, they will always miss the connection to other languages)

            Surely both Esperanto and Volapük count as Indo-European languages!

          • Tibor says:

            @Larry: Yes, you’re right of course. For some reason I assumed that all artificial languages have completely made up roots.

        • AnonYEmous says:

          it’s a bit more like: why learn to play this obscure card game, when you could learn Magic : The Gathering and then get to play against other people, expanding your potential social circle and enjoyment from playing?

    • vV_Vv says:

      If I understand correctly, Esperanto is the only constructed language that has native speakers and is actually used for communication.

      • Bugmaster says:

        Er, sort of. It’s the only one besides Klingon and Quenya. Although, that depends on what you mean by “native speaker”; if you mean, “someone who speaks only Esperanto”, or “someone for whom Esperanto is the first language”, perhaps you may be right. That said though, a ton of people speak Klingon; and a few do speak Quenya.

        • Machine Interface says:

          The usual assessment for Esperanto is:
          7,000,000 speak some Esperanto
          700,000 speak Esperanto frequently
          70,000 are fluent in Esperanto
          7,000 are native Esperanto speakers (children of bilingual couples whose only common language is Esperanto)

          The main point of interest I can see in learning Esperanto is how dispersed the community is; if you learn Esperanto, you will find people to talk to (through Esperanto networks) in almost every country in the world.

          If you learn Chinese, you will have a lot more people to talk to, but they will almost all be in China.

          • HeelBearCub says:

            I’m guessing that there are more Chinese speakers than Esperanto speakers in most countries?

            But perhaps Esperanto is a signalling device for something else (being in the Esperanto “club”), so it’s not really the raw numbers that matter, but accessibility.

          • Aapje says:

            @Machine Interface

            If you learn Chinese, you will have a lot more people to talk to, but they will almost all be in China.

            I think that you have a better chance to run into someone who speaks Mandarin in pretty much every country than someone who speaks Esperanto. They are also easier to pick out.

            I think that you are mistaking percentages with absolute numbers. It can both be true that Mandarin-speakers are more centralized and also, that more Mandarin-speakers live outside of China, due to sheer quantity. You only need a fraction of the 1.4 billion Chinese people to live abroad for there to be more of an international community of Mandarin-speakers than the community of 7 million Esperanto-speakers.

            EDIT: Ninja’d by HBC 🙂

          • HeelBearCub says:

            EDIT: Ninja’d by HBC

            I’m tricksy that way.

          • Machine Interface says:

            Aapje > this is true, but the sense of my parenthesis “(through esperanto networks)” was that you would not be just hoping to run into esperanto speakers, but deliberatedly track them down in the country you’re interested in.

            Chinese was perhaps not the best example, since there are a lot of chinese communities outside of China, but these people are obviously culturally chinese; whereas, the point of Esperanto is that you will find speakers of it in almost every cultures — not a lot, but they’re there, and they have a network which you gain access to by speaking Esperanto. If on a whim you want to learn about say, Slovenian culture, you can locate Esperanto speakers in Slovenia and talk to them.

            I don’t speak Esperanto myself and I think its advantages and qualities are hugely overstated by the Esperanto community, but I do see the point of having access to a geographically dispersed and culturally diverse community of speakers [of course English accomplishes this too to a large degree].

          • Tibor says:

            @Machine Interface: Can’t you just talk to the Slovenians in English? If you want to fully grasp a culture you have to learn its language anyway. If not, you might as well speak English. Of course, speaking Esperanto is like having a shared hobby and that might make it easier to “break the ice” and start a conversation with someone you don’t know, so you have a community of people around the world who want to talk to you just because you speak Esperanto. Nobody is going to be impressed that you speak English. On the other hand, unless the local culture is extremely closed to foreigners, it is not that hard to find new friends anyway.

        • Creutzer says:

          Quenya can’t possibly have native speakers. The vocabulary is much too small.

          • Bugmaster says:

            AFAIK people have taken Tolkien’s original design and went nuts with it, massively developing both the vocabulary and the grammar — but I could be wrong. I agree that canonical Tolkien-era Quenya is probably too small to have native speakers.

          • albatross11 says:

            SM Stirling’s Change series has a whole survivor community of people who speak Sindarin and ASL growing up, after most technology is caused to stop working by a plot device.

          • Chalid says:

            IIRC Stirling explicitly says that the community had to make up a whole lot of words.

          • Deiseach says:

            There’s a course if you want to learn Quenya! (Great as a resource website, absolutely terrible design; seems like linguistic talent does not necessarily go hand-in-hand with graphic design talent).

          • Quenya can’t possibly have native speakers. The vocabulary is much too small.

            I assume you mean young native speakers.

          • Tibor says:

            David: doesn’t being a native speaker of a language mean that it’s the first language you’ve learned? Or possibly one of the first two if you’re bilingual

          • Tibor:

            I think that’s what it means. I was referring to people for whom either Quenya or Sindarin would have been their first language.

          • Creutzer says:

            Well, that’s what I said, so I’m now very confused what the modifier young was supposed to do.

          • John Schilling says:

            so I’m now very confused what the modifier young was supposed to do

            Well, anyone for whom Quenya or Sindarin was a first language would be at least eight thousand years old by now. Even by Eldar standards, that’s a bit long in the tooth.

          • Tibor says:

            @John: Oh. I guess I am a bit slow/don’t have a sense of humour 🙂 Well, at least I’m not the only one.

  14. The Clever Chameleon says:

    The book has also been translated in Serbian: http://www.worldcat.org/title/svako-dete-je-genije/oclc/440095791 I should be able to get hold of a copy somewhere in October.

  15. topynate says:

    Who else has the motivation to hide that book?

    The suspect would have to be aware of the book, and its potential effect, and yet be opposed to those effects. Perhaps, given its slightly better availability in Esperanto, the suspect would have a soft spot for speakers of the language, whether through personal experience or general internationalist feeling. And of course, the suspect would have to be a very slick operator. A bit of money wouldn’t hurt either, to hunt down all the English copies.

    Prepared to sacrifice improvement in the intellectual power of mankind for the sake of obscuring the genetic constraints on that improvement. Knowledgeable of and favourable towards Esperanto and Esperantism. Speaks at least one of Hungarian, Esperanto and English. It wouldn’t be entirely incredible for the suspect to be from the same community as Laszlo, in fact. Wealthy.

    Probably a bit too overdetermined a profile to actually fit anyone, though.

  16. Freddie deBoer says:

    regression to the mean predicts that their children would have IQs around 140 to 150

    Don’t you think we have too few observations here for regression to the mean to be a serious constraint?

    • Scott Alexander says:

      I’m trying to see if it’s even at all possible given the worst possible assumptions, but some people corrected my math (see above) and I’m not sure this provides strong evidence now.

  17. Ozy Frantz says:

    I would worry (both from an altruistic perspective and from a what-makes-my-kids-happiest perspective) that raising a child in this fashion would involve sacrificing flexibility to an unusually high degree. If the child turned out to be unusually talented in something that wasn’t obvious when they were two, or something changes and the career you picked for them is less fulfilling and has less good working conditions than you thought it would be, or (from an altruistic perspective) you’re wrong about what the highest-impact careers will be two decades from now, all your effort could wind up going to waste.

    • Scott Alexander says:

      I think Polgar says you can wait until up to six before specializing.

      I think this is the same bet as any effective altruist intervention. Sure, ten years from now the cause you chose to invest in building could be totally useless. But you go on percentage, and aggregating across a large number of cases (in the community, not necessarily per individual) you do okay.

      It seems like the worst failure mode (assuming that they get basically unschooling-level results in everything else) is that at age sixteen they decide to do something else, and they’re no worse off than any other sixteen year old who was unschooled.

      • John Schilling says:

        Unschooling usually involves the parents supporting whatever quasi-educational activities the children actually want to partake. If instead the parents say “no, you will study [X] instead”, where X is either what the parents think is the Right Specialty or what the child thought was the Right Specialty at six but no longer cares about, that could plausibly leave the child worse off.

        • wintermute92 says:

          Ok, fair, but there’s no particular sign that this happened?

          I don’t see this as a risk to the kids, at any rate – you just avoid that. I do see it as a risk to the experiment – dropout rates from “the kids lost interest” might be pretty high. Although I bet they’re lower than one would naively guess, just because investment in a topic tends to breed interest in it.

      • Ozy Frantz says:

        ~All studies of homeschooling are really biased towards showing that homeschooling is effective (they are almost always done by homeschooling advocacy groups, which do the usual shenanigans, and some shenanigans which are impressively bad on the order of “using test results sent in by the parents who could presumably not send in test results that make homeschooling look bad”). I think the current evidence on unschooling is a reasonable upper bound but not a reasonable lower bound. So “unschoolers are a year behind” is actually a HUGE ENORMOUS RED SCARY FLAG, because it means they’re biasing it as much as they can and they still can’t get unschoolers to come out ahead.

        I also think that unschooling is really different from whatever education you’d be doing outside their specialty, which limits the generalizability. “Parent follows the children’s interests while providing an enriched environment and gently guiding them to learn things” is a very particular situation and I don’t think it should be treated as the default. Of course it’s hard to say what the default should be given how little we know about their educational methods.

        I tend to donate to things that have results right now and ignore flowthrough effects, which has some flaws but does not require me to predict what 80,000 Hours’s top career pick will be in 2036 to be confident I’m doing good. 🙂

    • BlindKungFuMaster says:

      So the guy who was second in the world after Judith Polgar in the age group under 13. He also was homeschooled to be able to totally concentrate on chess. When he was 13 he decided to do something else. Became a game designer, neuroscientist, entrepreneur, ai-researcher. Demis Hassabis.

      Seems like his early chess training didn’t harm his flexibility much.

  18. Freddie deBoer says:

    “The Polgar sisters talk about how they loved their education, had a great childhood, thought their parents were always patient with them and never strict and harsh, and don’t regret anything. How many kids who went to public school can say the same?”

    Well, me, for one.

    • Joyously says:

      My public school experience was torturous, but my parents were lovely and I ended up fine. I’m not sure whether I “regret” those public school years since, as I said, I’m fine, but I do sometimes wistfully think about how it might have been different if I had been homeschooled by my lovely parents.

    • Scott Alexander says:

      Oh God, you were one of those kids who liked school, weren’t you? That explains a lot.

      • HeelBearCub says:

        Seriously?

        You should bloody well take your own medicine here Scott. Not kind, not necessary, not even true.

        (And I hated almost all of the social aspects of school but loved the classroom time).

        ETA:
        And to make clear what I find objectionable, this is Scott, on his home turf, sourrounded by his own people signalling that someone is a member of one of “those” people in the outgroup.

        • grothor says:

          Any time that I think someone on the internet has said something uncharacteristically crazy/mean/stupid/whatever, I ask myself “Am I certain I understand on what level of irony this person is operating?”

        • onyomi says:

          @HBC

          I don’t think “kind, necessary, true” prohibits good-natured ribbing.

          • HeelBearCub says:

            And why should we take this as good natured ribbing? He has not exactly earned the right to engage in that with Freddie, given his generally poor opinion of Freddie’s positions.

            Even if it’s meant as good natured ribbing, what’s the sub-text? That people who actually liked school are freakish weirdos. And pardon me, Scott does not include himself in that group of people, so he should understand what he is signalling here.

            There is plenty of that from the people who don’t like learning things, we really don’t need it from the people who do. It’s a purity signal.

          • ADifferentAnonymous says:

            Freddie is on the blogroll, and Scott explicitly recommended his blog in the last links post.

            And speaking personally, I dislike everything I read by Freddie but persist in reading it solely because of Scott’s evident respect for him.

            To me, that makes theirs among the friendliest of rivalries, and buys leeway for some ribbing.

            And I kind of liked my (admittedly private) school but share Scott’s skepticism that it’s a remotely reasonable way to spend a childhood.

          • bean says:

            I’ll actually make something of the case for school as teaching social skills. I spent my time through 3rd grade in a pretty normal school, and it didn’t work well. I was a lot smarter than anyone else in my class, and basically just read all the time, occasionally paying attention to the teacher. I had about one friend at a time, although they tended to move away.
            In 4th grade, I started at the regional gifted program, and it was a revelation. There is some value in being forced to deal with a bunch of people you don’t get to select, although for most smart people it doesn’t work well because they aren’t having to deal with people they can actually relate to. It was also a very strong program academically. When I have kids, I’m sure they won’t be in normal schools. If I can find an equivalent program, I’ll probably send them there. If not, there’s always homeschooling, although the majority of homeschoolers I’ve seen have serious social issues. Not all, but enough to convince me that there is some value in school for social skills.

          • AnonYEmous says:

            He has not exactly earned the right to engage in that with Freddie, given his generally poor opinion of Freddie’s positions

            I took it as this and I have no doubt that Freddie does as well. If anything, Alexander is consistently kind to him and refers to him as a good source (both things he deserves, in many cases, but still)

        • Scott Alexander says:

          To explain what I’m saying – Freddie has some posts up which I interpret as saying that schooling doesn’t really improve test-measurable knowledge or ability to get good jobs, but that the education system is good anyway because it helps build community and stuff. I think he wants to keep the school system how it is except maybe purifying it of newfangled neoliberal influences like charter schools and merit pay.

          To me, this sounds a lot like “we discovered incarceration doesn’t actually decrease crime, but the prison system is good anyway because it helps guards have a place to go to connect with each other, so let’s keep it exactly as it is and no harm done.”

          • Bugmaster says:

            I think that social skills, as well as the ability to work with other people toward a common goal, are almost as important as “test-measurable knowledge or ability to get good jobs”. So, even if school doesn’t provide anything but those social skills (I disagree with this proposition, but I’m willing to entertain it for the sake of the argument), it would still be worthwhile, just for the socialization alone.

            I mean, I do agree that if you are so smart that no one around you could even come close to being your intellectual peer, then you don’t need any social skills. You just need an ample supply of food, a quiet room to think in, some lab equipment, and solitude — and then you might discover the next scientific advance that will catapult humanity into a new golden age. But most people aren’t like that; we actually need to interact with each other at some point in order to be productive.

          • onyomi says:

            I am skeptical about our current model of schooling’s ability to impart social skills. It creates a completely artificial sort of social situation which largely isn’t replicated in adulthood.

            In adulthood most people are able to chose whom they socialize with based on shared interests, family ties, etc. Even in a professional context, though you may have to get along with people you don’t like at your company, you generally have people of a wide range of age groups doing a wide variety of jobs, not the same people of the same age group all shoved into a room together to do roughly the same tasks.

            School doesn’t teach you how to make friends as an adult; it arguably handicaps this ability as homeschoolers, like adults, have to seek out friends and activities without having a default social circle forced on them every day.

            One could make a slightly stronger case that school prepares you to work at a company full of people you may or may not like. In my case, it taught me that I could never do a job where the social dynamics were anything like they were in school (primary and secondary).

          • vV_Vv says:

            not the same people of the same age group all shoved into a room together to do roughly the same tasks.

            Factory jobs or even most office jobs are pretty much like this. There may be more age variation, but age differences mostly matter when you are a child, not an adult.

            One could make a slightly stronger case that school prepares you to work at a company full of people you may or may not like.

            This seems important.

            Almost anybody can get along with people they like, but getting along with people you don’t like requires social skills.

          • HeelBearCub says:

            @Scott Alexander:
            You seem to be engaged in “typical minding”. Your experience in school was negative, therefore everyone’s must be, and if it wasn’t they are abnormal.

            It’s also important to remember that school is not the cause of unhappiness that is actually caused by things like adolescence.

            Finally, it seems that Scott wasn’t actually engaged in “good natured ribbing” but his point is actually that anyone who likes school is like someone who likes being in prison.

          • Aapje says:

            @HeelBearCub

            Scott seems to believe that formal schooling is overall a bad/sub-optimal way to educate children, which doesn’t require a belief that it is bad for all children.

            To steelman Scott: school is a pleasant environment for some and a horrible experience for others. Freddie is the person who is typical minding when he only takes the former group into consideration and ignoring the severe psychological harm experienced by a decent subset of students. When it comes to prisons, we know that having a prison is better than having anarchy, but there are alternative forms of punishment that are better than prison. The existence of ex-prisoners who look favorably on how prison changed their life doesn’t change the fact that the statistics show that prison is not a very good way to reform people. Similarly, the existence of ex-students who look favorably on how school changed their life, doesn’t mean that school is better than the alternatives.

          • HeelBearCub says:

            @Aapje:
            That’s not really a steelman, as it inverts the use of “some” and “most”.

            Note Scott’s allegorical statement above about school being a prison that is helpful only to the guards, and only for social reasons.

          • Aapje says:

            @HBC

            The allegory can be a reference to Freddie being an educator and thus defending the current system, because it provides him with a job.

            Arguing that the real beneficiaries of the current system are the teachers, and not the students, doesn’t necessarily imply that school is bad for all students, but merely bad on average.

          • Freddie deBoer says:

            To be frank, I think Scott is a really good example of someone who has spent his whole life believing that he is one of the only smart people alive, and that this colors his perspective on a lot of things. In particular, the point of schooling. If you’re someone like Scott who believes yourself to be in the high echelons of IQ, you are likely to think that IQ is really all that matters for straightforwardly self-interested reasons. You are also likely to disdain other values that could be inculcated in formal schooling. Is that a particularly charitable take on the influences on Scott’s thinking? No. But I’ve encountered it enough times in my life to find it likely. I have argued, many times, that these tests have predictive validity and measure something “real.” The question is, is that real thing an exhaustive catalog of the mental and emotional abilities that people might find valuable? I think the answer’s no. That doesn’t undermine the predictive validity of IQ-surrogates. It just means there’s more to the story.

            For myself, I’ll just say this: not only are quantitative metrics like IQ-surrogate standardized test scores a modern invention, the idea that education could be reduced to such things is a thoroughly modern invention, one that arose in a particular economic and social context. Go back throughout the history of formal schooling and you’ll find that perfectly “hard-nosed,” conservative, capitalist thinkers never would have assumed that the purpose of schooling was as narrow as Scott (and many others) make it out to be. That historical record should be explored and given an adult accounting.

          • The Nybbler says:

            @onyomi

            School works extremely well at imparting the social skills society needs. It teaches each person the skills appropriate for their place in the social hierarchy. The masters learn how to be masters, their toadies learn toadying, the thugs learn thug-appropriate skills, the middle learn how to get along in the presence of the other groups, and the outcasts learn what it takes to survive as an outcast. It also sorts people properly into these categories according to their natural tendencies. This is rough on those who aren’t satisfied with their category, but life is always rough for them.

          • Anonymous says:

            @Nybbler

            I don’t think you’re entirely wrong, but school teaches you social skills in environment of children, not adults. This is subtly different from desirable.

            The role models in school are other children, which is one of the problems with Prussian-model-inspired schools. Instead of looking up to their parents and their parents’ adult friends and acquaintances, schoolchildren are presented with teachers (who are very rarely admirable, and more often of the sort that “can’t do, so they teach”) and other children. Predictably, this means that bullies, assholes, petty criminals, etc have an undue influence on their peers. Especially if they’re a year or two older (which they likely are, due to being halted in a grade).

          • vV_Vv says:

            Instead of looking up to their parents and their parents’ adult friends and acquaintances, schoolchildren are presented with teachers (who are very rarely admirable, and more often of the sort that “can’t do, so they teach”) and other children. Predictably, this means that bullies, assholes, petty criminals, etc have an undue influence on their peers. Especially if they’re a year or two older (which they likely are, due to being halted in a grade).

            Your parents and their adult friends will not always be there to protect you. Bullies, assholes, criminals, etc. will always be there to fuck with you, so you’d better learn it how to deal with them. It’s better to learn it earlier, in a controlled environment under the somewhat distant supervision of the teachers, rather than later, in your adult life, when any error may cause you severe consequences.

          • Bugmaster says:

            @Freddie deBoer:

            I think Scott is a really good example of someone who has spent his whole life believing that he is one of the only smart people alive, and that this colors his perspective on a lot of things.

            That is perhaps a bit harsh, but frankly, I have seen that same failure mode before. I am friends with some people who are much, much smarter than myself; and my totally unscientific perception is that they are simply not used to ever being wrong. Or rather, they are used to being able to easily solve problems that other people find very challenging; so, upon encountering a problem that someone claims to be challenging, their gut reaction is, “ok, the last 99 supposedly challenging problems turned out to be easy, let’s see how quickly I can solve this one”.

            That’s probably the correct response under Bayesian reasoning; however, when that problem #100 turns out to not have an easy solution after all, these smart people tend to have trouble noticing this fact. Physicists are notorious for this, but they’re not alone; other disciplines have the same issues.

          • psmith says:

            As a total dumbass, let me assure you guys that lots of us found school pointless, boring, and miserable as well.

            I also think Freddie’s conflation of “education” with “formal schooling” is pretty telling.

          • John Schilling says:

            You are also likely to disdain other values that could be inculcated in formal schooling.

            Could you be more specific as to what these “other values” that you think Scott undervalues might be? Also, I think the onus is on you to show that they actually are inculcated in formal schooling, not just that they “might be”.

            If the answer is any simple variation on “social skills”, then no. I think it’s very clear from Scott’s writings that he doesn’t undervalue social skills. And after a lifetime of hearing people assert without evidence that schools provide VITAL “social skills”, is there anybody left who gives a damn about the bare unsupported assertion? But if you’ve got something new, let’s hear it.

          • Jiro says:

            Bullies, assholes, criminals, etc. will always be there to fuck with you, so you’d better learn it how to deal with them.

            In the adult world, dealing them means some combination of avoiding them, going to the police or courts, and self-defense using weapons. Schools won’t let you do this.

            Of course, you may just be trying to suggest that we should allow schoolchildren to get restraining orders against bullies, in which case, sure.

          • HeelBearCub says:

            @Aapje:

            The allegory can be a reference to Freddie being an educator and thus defending the current system, because it provides him with a job.

            This is still wildly at odds with your original contention that Scott was talking about a minority of students with poor experiences.

            Scott’s consistent contention has been that the vast majority of students get no real benefit from education. He has stated this over and over, in a variety of ways.

          • Brad says:

            Bullies, assholes, criminals, etc. will always be there to fuck with you, so you’d better learn it how to deal with them.

            In the adult world, dealing them means some combination of avoiding them, going to the police or courts, and self-defense using weapons. Schools won’t let you do this.

            Sure, no one is going to shove you into a locker as an adult. But office politics can certainly involve its share of bullies and assholes. You can’t go to the police for that.

          • Aapje says:

            @HBC

            This is still wildly at odds with your original contention that Scott was talking about a minority of students with poor experiences.

            This is not an accurate reflection of what I said, IMO.

            The general argument merely requires that there are significant downsides to the current model of schooling, while an alternative has equal or better benefits and fewer downsides.

            I suggest that you adopt a more fruitful approach to challenge these beliefs by, for example, questioning whether a better alternative exists. Attacking the premises seems more sensible than attacking the argument itself.

            Scott’s consistent contention has been that the vast majority of students get no real benefit from education. He has stated this over and over, in a variety of ways.

            I was addressing the actual claims that Scott made in this thread and don’t feel obligated to defend all the beliefs that Scott has or may have.

            Although I do want to point out that Scott’s conclusion in his graduation speech seems to be that formal education is very poor value for the costs (money + time), which is a very different claim than “that the vast majority of students get no real benefit from education.”

          • The Nybbler says:

            @Jiro

            Schools teach you that if you’re not the kind of person the authorities want to help, you’re not going to get help from the authorities and if you try, you will be hurt by them instead.

            This is a lesson with vast application to the real world. It teaches you to yield to those favored by authority, lest you get into a conflict which you will not be permitted to win. It teaches you to settle conflicts with other disfavored people outside the view (or at least official view) of authority.

          • Aapje says:

            @Brad

            I was specifically taught in school to not depend on the authorities and to use violence against bullies. In my current working environment, I suspect that this is a very poor strategy.

            In none of the environments where I worked, did I have anywhere near the bullying and asshole behavior that I experience in school. In so far that I experienced the latter, the strategy of explaining my issue with their behavior and asking for them to be more considerate was reasonably successful, while this strategy was actually counterproductive in school.

            My experience as an adult may be partly due to my work environment involving people with moderately high IQ’s, although my experiences were not that much worse in manual labor jobs I did next to school/university.

          • “The question is, is that real thing an exhaustive catalog of the mental and emotional abilities that people might find valuable? ”

            You are attacking a straw man. I don’t think I have encountered anyone who believes that.

          • nimim.k.m. says:

            Most of the local communities that I know of are structured around the local school, including the actual, you know, school-activities that happen in school, but also the all the fairs, after-school sports activities for the kids, space for various community activities, and so on. Parents meet regularly the other parents in the district. It is also the primary way the populace is taught the local version of the secular-national religion (both the explicit rituals like honoring the flag or singing the national anthem and the implicitly in the contents of the curriculum, starting with the version of the local language that is taught and ending with history and geography and such). The modern nationstate is based on the school system.

            (And of course, it’s very convenient daycare for kids whose both parents must go to work. But the emphasis was on the social side of the school.)

            I hear that in US, where people actually go to the church more than once in a year to sing Christmas songs, the local church might have similar function as the social focal point in the neighborhood. But a school has a benefit of being secular, and propagating the values of the local state instead of the particular form of religion around (which is important if you believe in the further existence of the local state in its current form, I suppose).

            It seems quite large Chesterton fence, really.

            edit. To further clarify, there seems to be qualitatively quite large differences between schools and prisons in that sense.

          • Toby Bartels says:

            @nimim.k.m.

            But in the U.S., churches aren’t nearly as neighbourhood-based as public schools are. It’s not just that there are multiple denominations, although that’s a big part of it. But besides that, you don’t drive your kids across town to go to a school outside your neighbourhood unless you have an unusual reason (and even then only if you live in a school district that allows this, and if your only reason is because the other school is rated better, then it will fill up quickly and then no district will allow it). However, you do drive your family across town to go to a church outside your neighbourhood, because that’s the one that agrees with your politics.

            It’s different for Catholics, who are expected to go the local parish church no matter what (but even they don’t need to apply with the bureaucracy for permission if they choose to do differently), and just as well since they go to different schools. So if you live in a heavily Catholic area, then neighbourhood churches are real. And if you live in a rural area, then everybody probably belongs to the same denomination (whatever that is there) and is only within reasonable distance of one church, so neighbourhood churches work there. But among Protestants in cities, you choose the church based on your denomination and your political tribe, not your neighbourhood.

            (Despite writing ‘neighbourhood’ all the time, I am actually an American who’s lived my whole life in the United States.)

          • HeelBearCub says:

            @Aapje:

            ignoring the severe psychological harm experienced by a decent subset of students.

            This is what I was interpreting as saying “a minority of students”.

            Scott has no proposal to reform the education system that I can critique. He handwaves very vaguely at “unschooling”, and that is about it. If you will note in the linked post, he also assumes that unschooling is free, requiring no resources (as he indicates that every student could be able to have the entire cost of their education in their bank account). As I see it, the burden of proof is on him to prove that burning the entire system down (which is what he is asking for) will result in something better.

            If Scott merely wished to state that education can be improved, that outcomes are sub-optimal, I would have no issue with that. But Scott has stated numerous times that education is irrelevant, merely a signal, and that IQ is the only thing that matters (or, at the very least, that he thinks this is the most reasonable assumption given the evidence).

          • Aapje says:

            @HBC

            That school is really bad for a small subset doesn’t mean that it is good for the rest.

            Anyway, I agree that Scott’s position on schooling seems to be mostly based on his personal bad experiences and that he doesn’t really make a consistent case. He has argued for unschooling, but also for vouchers, which are not unschooling.

            He has also argued for ‘more experiments,’ which might be the core of his beliefs: just try a ton of different things and find something better.

          • Brad says:

            @Aapje
            I’m not sure what to tell you. I had my ups and downs in school, but it was never as bad as it apparently was for many of the people here.

            On the flip side, I have been in some pretty toxic work environments.

            I’m not going to say we should blindly extrapolate from my experience, but by the same token we shouldn’t from yours or the others that had terrible experiences. We should all try to form opinions based on data rather than our own idiosyncratic experiences.

          • Aapje says:

            @Brad

            Sure, but when a different commenter claims without hedging that school prepares you to deal with bullies and assholes in later life, I consider a single data point somewhat relevant.

          • Joyously says:

            Standardized test scores aren’t the be-all-end-all, but I have never seen any convincing effort to show that schools in general or public schools in particular improve “social skills” or a “sense of community.” Frankly I think my public school experience stunted my social skills in some ways, particular when it comes to my relationships with men.

          • Joyously says:

            Also: If the major benefit of public schooling is “social skills” then public schools definitely don’t need more money. If you’ve got a box to put the children in and food for lunch that should be sufficient.

          • AnonYEmous says:

            Yeah, this is an important point.

            And let’s address another issue: Public schooling is an issue optimized to teach certain subjects, or at least directs enormous amounts of resources to teaching certain subjects and attempts to get better at doing so.

            But then when challenged that teaching these subjects is kind of meaningless and a waste of everyone’s time and money, they say “Well OK, but the real benefits are social interaction and learning good habits”. OK, so why not create a system optimized to do those things? Why not create a system which encourages and allows for plenty of social interaction, with assistance for anyone who needs it? Many of my most positive potential interactions were fucked over by the fact that the teacher was talking, or had seated me out on an island, or so forth. And as for good habits, those can be learned while also teaching students legitimately interesting or important subjects – not something you can say for high school or much of middle school, at least not for the majority of students.

          • Debug says:

            @Anonymous

            I don’t think you’re entirely wrong, but school teaches you social skills in environment of children, not adults. This is subtly different from desirable.

            I was under the impression that children are primarily socialized by their peers so I don’t think it’s too much of a problem that children are taught social skills in an environment of children. Children need to learn to fit in with their peers and seem to quickly learn to discount

            I do think there is a problem with the average teacher not being particularly great. Locally, we have a teacher over-production problem and perhaps competition between teachers is slightly pushing up teacher quality. Perhaps the education system here could be improved locally by selecting for more intelligent teachers (We have such a large pool and such few positions.. ). The other problem I guess – with teaching in general – is, especially when you have over-production of teachers, that it’s difficult to select amongst the pool of potential candidates who will do the least harm to their students. Currently, selection of teachers based on quality has a large random component.

          • Atlas says:

            @Freddie deBoer

            I don’t normally “white knight”, and I realize that this was just a comment and not an actual essay/article, but I felt that your comment was very uncharitable and poorly reasoned, so:

            To be frank, I think Scott is a really good example of someone who has spent his whole life believing that he is one of the only smart people alive, and that this colors his perspective on a lot of things.

            I don’t think that this is what Scott believes at all. Instead of just deciding what people believe for them without providing any corroborating citations—-a frequent and deeply irritating feature of deBoer’s writing—I will cite some actual evidence that led me to this conclusion. In the post “the Parable of the Talents”, Scott wrote:

            Every so often an overly kind commenter here praises my intelligence and says they feel intellectually inadequate compared to me, that they wish they could be at my level. But at my level, I spend my time feeling intellectually inadequate compared to Scott Aaronson. Scott Aaronson describes feeling “in awe” of Terence Tao and frequently struggling to understand him. Terence Tao – well, I don’t know if he’s religious, but maybe he feels intellectually inadequate compared to God. And God feels intellectually inadequate compared to Johann von Neumann.

            Does describing three successive levels of intelligence one feels completely inadequate beneath sound like the writing of a man who “has spent his whole life believing he is one of the only smart people alive?” I report, you decide.

            As for whether it “colors his perspective”, this is one of those verging on ad hominem arguments I find largely valueless—in a debate on the merits, who cares what personal experiences lead people to the opinions they hold? Surely what actually matters is the quality of the facts and logic they present, which can presumably be refuted if incorrect without recourse to speculation about your opponent’s alleged inferior motives/characteristics?

            If you’re someone like Scott who believes yourself to be in the high echelons of IQ, you are likely to think that IQ is really all that matters for straightforwardly self-interested reasons.

            Maybe this is just an idiosyncrasy on my part, but I feel that it’s bad form to say “John Doe believes/perceives/thinks that x is true”—which suggests that you think this belief is false— without explaining at all what evidence leads you to believe that believing x is the result of faulty perception or incorrect reasoning. In this case, if Scott does indeed believe himself to have a high IQ, this seems eminently reasonable given that he’s a practicing psychiatrist who writes a widely-respected blog that covers many different complex subjects in his free time.

            I also emphatically disagree with the claim that high-IQ people are naturally inclined to selfishly and dishonestly rationalize their success by claiming that “IQ is all that matters”. (I say again that I feel this speculation about personal motives is much less interesting and important than the actual factual question of how much IQ matters.) Charles Murray was censored and attacked at Middlebury because he wrote a book about IQ over 20 years ago by a mob of students who presumably had to do pretty well on IQ-proxy tests to get accepted to Middlebury. Mainstream media publications like the New York Times and Vox that cater to relatively high IQ audiences run lots of articles about how neighborhood quality, school funding, stereotype threat, etc. allegedly explain differences in academic outcomes, while running almost none about the alternative hypothesis that IQ differences do. And just off handedly, I can think of several high IQ people who downplay the importance of IQ, like Noah Smith, Christopher Jencks and Malcolm Gladwell.

            You are also likely to disdain other values that could be inculcated in formal schooling. Is that a particularly charitable take on the influences on Scott’s thinking? No.

            I could take this opportunity to counter by mean spiritedly speculating about how people who deny that IQ matters (the inverse though more common weakman) are just losers who are bitter that they didn’t/don’t do well on IQ/IQ-proxy tests, and they make up dumb rationalizations about how other stuff matters because they’re dumb idiots who want to give themselves false hope. I could do that, but, in addition to this not at all being the correlation I actually observe, I think that would be a counterproductive thing to focus on, even if it was true, because it would move the conversation towards nasty trading of personal accusations rather than objective weighing of relevant facts and logic.

            I have argued, many times, that these tests have predictive validity and measure something “real.” The question is, is that real thing an exhaustive catalog of the mental and emotional abilities that people might find valuable? I think the answer’s no. That doesn’t undermine the predictive validity of IQ-surrogates. It just means there’s more to the story.

            Okay, so, from your statement, you believe that IQ matters, but it’s not the only thing that ever matters at all times. I am pretty sure that this is….like, exactly what Scott Alexander believes.

            For myself, I’ll just say this: not only are quantitative metrics like IQ-surrogate standardized test scores a modern invention, the idea that education could be reduced to such things is a thoroughly modern invention, one that arose in a particular economic and social context. Go back throughout the history of formal schooling and you’ll find that perfectly “hard-nosed,” conservative, capitalist thinkers never would have assumed that the purpose of schooling was as narrow as Scott (and many others) make it out to be. That historical record should be explored and given an adult accounting.

            Judging from “SSC gives a Graduation Speech” and “Against Tulip Subsidies”, it seems to me (perhaps incorrectly) that Scott is well aware that there’s more to a good education than standardized test scores, but is understandably skeptical that these (conveniently hard to quantify) other traits are well developed enough under the current public school model to compensate for its many failings.

          • onyomi says:

            @The Nybbler

            School works extremely well at imparting the social skills society needs. It teaches each person the skills appropriate for their place in the social hierarchy. The masters learn how to be masters, their toadies learn toadying, the thugs learn thug-appropriate skills, the middle learn how to get along in the presence of the other groups, and the outcasts learn what it takes to survive as an outcast.

            But the popular people in high school often have crummy jobs! And the former outcasts may be their boss!

            The main social skill I learned from high school was that I should do my best to avoid the stupid, mean, petty, and/or snobbish people and gravitate toward the smart, nice, interesting people. I do continue to do this in adulthood, but I’m skeptical school helped me learn how. Really, it’s so much easier as an adult to avoid assholes and find nice people, and to the extent it takes skill, it seems like a different set of skills from those I learned in high school.

            I would also note that I am pretty sure I learned counter-productive social “skills” in high school in addition to any useful ones I may have absorbed. In high school I learned to be very suspicious of anyone making friendly overtures to me, because I tended to assume that when the cool people approached me it was to make fun of me. I think this made me more shy and wary as an adult than I might otherwise have been, and in a not very helpful way.

          • onyomi says:

            @Nybbler

            School works extremely well at imparting the social skills society needs. It teaches each person the skills appropriate for their place in the social hierarchy. The masters learn how to be masters, their toadies learn toadying, the thugs learn thug-appropriate skills, the middle learn how to get along in the presence of the other groups, and the outcasts learn what it takes to survive as an outcast.

            But the popular people in high school often have crummy jobs! And the former outcasts may be their boss!

            The main social skill I learned from high school was that I should do my best to avoid the stupid, mean, petty, and/or snobbish people and gravitate toward the smart, nice, interesting people. I do continue to do this in adulthood, but I’m skeptical school helped me learn how. Really, it’s so much easier as an adult to avoid assholes and find nice people, and to the extent it takes skill, it seems like a different set of skills from those I learned in high school.

            I would also note that I am pretty sure I learned counter-productive social “skills” in high school in addition to any useful ones I may have absorbed. In high school I learned to be very suspicious of anyone making friendly overtures to me, because I tended to assume that when the cool people approached me it was to make fun of me. I think this made me more shy and wary as an adult than I might otherwise have been, and in a not very helpful way.

          • The Nybbler says:

            But the popular people in high school often have crummy jobs! And the former outcasts may be their boss!

            This was a popular 80s meme, probably reached its peak in “Back to the Future”. But it ain’t so, it’s just a revenge fantasy. The people running things in the real world were probably popular in school. The smarter variety of outcasts may be well-compensated servants, but they’re servants nonetheless.

          • carvenvisage says:

            @Freddie Deboer

            The one size fits all teach to the test approach *is* pretty ubiquitous in modern public schooling though. Speaking for myself, I don’t object to the underlying idea of public education, I just have less expectation that it can be reformed into something that, among other things isn’t based around a memorisation game, than that people can route around it with things like khan academy and IRL social hubs.

          • onyomi says:

            @Nybbler

            This was a popular 80s meme, probably reached its peak in “Back to the Future”. But it ain’t so, it’s just a revenge fantasy. The people running things in the real world were probably popular in school. The smarter variety of outcasts may be well-compensated servants, but they’re servants nonetheless.

            I certainly wouldn’t claim there’s an inverse correlation between high school popularity and adulthood success, but I also don’t think it positively correlates nearly as tightly as you think.

          • nimim.k.m. says:

            @AnonYEmous

            OK, so why not create a system optimized to do those things? Why not create a system which encourages and allows for plenty of social interaction, with assistance for anyone who needs it?

            This is actually quite good point. It resembles my pre-school / daycare experience (most of time playing and socializing with other children under adult supervision, one lesson per week with curriculum of “let’s color the circles in the book, good work k.m., on the next page, look at that shape, it’s a rectangle, yes, now we color the rectangles, yes you can draw a dog in there if you are already done” (not exact memory but something like that).

          • leoboiko says:

            OK, so why not create a system optimized to do those things? Why not create a system which encourages and allows for plenty of social interaction, with assistance for anyone who needs it?

            I think it’s past time we considered a system with free subjects. The only mandatory classes would be the core skills of reading and basic arithmetic, for young kids. With that done, resources, efforts and time would be spent into coaching kids on how to learn about stuff they’re interested in—how to use libraries, how to search and evaluate things on the Internet, how to practice sports and skills in general…

            There’s still a lot of benefit from having teacher figures, due to basic human psychology—if you just leave a kid to themselves, you risk they spend the whole day trapped on Skinnerian boxes like Facebook or Reddit or MMORPGs (hell, we adults sure could use better preparation on this front!) But I see no reason to force kids to learn a specific set of subjects, when everything shows that they end up not even learning them in the first place.

          • albatross11 says:

            Freddie Deboer:

            I think this is a common problem. There’s the stuff we care about that’s hard to measure, and there’s the stuff we can measure which is kinda like the stuff we care about. So a natural failure is to focus on the measurable stuff even when it leads us to make dumb decisions, because at least we can measure it.

            What we care about is broad and subjective and hard-to-measure: we want our kids to end up as well-rounded, functional, productive adults, and for them to have a good life.

            What we can measure efficiently is how well they did on some standardized tests and whether they graduate high school. For a few long-running studies where we’re willing to spend a lot more resources, we can note whether they go to college, whether they graduate, how much they make, whether they ever end up in prison, their life expectancy, etc.

            All those measurable things are important, but none of them capture exactly the stuff we care about. You could find a set of people who all looked the same in terms of test scores, graduation, income, etc., and some of them would look like people who’d had successful lives, while others would not. As a parent, you can immediately say “this person is successful on paper, but I hope my kids don’t turn out that way” or “this person doesn’t look like a big success on paper, but I think he’s really living his life well!”

            Is there a term for this phenomenon, whereby we all forget about the actual goals we have in favor of the easily-recognized ones? Goodhart’s Law isn’t quite it–that’s about the easily-measured metric naturally being gamed when people get judged for it. (Like state test scores being gamed by deciding who is exempt.)

          • Sivaas says:

            A point for the Skinner boxes of MMOs:

            In college, I was taking discrete math, and was being taught how to get the probability of at least one of a number of independent events occurring.

            I realized I already knew this, because Guild Wars calculated each buff to crit chance as independent, so if you wanted to get your overall chance to crit, you needed to use this technique.

            Other examples:
            – Card games like Hearthstone (Magic, for me, but same concept although different medium) start getting you really interested in draw probabilities.

            – WoW has a tremendous amount of math involved in determining optimal ability usages, to the point where it has a dedicated Monte Carlo simulation program. You pick up a lot of stuff incidentally if you’re trying to play your best. If nothing else, you’ll learn EV calculations pretty quick when you’re trying to evaluate items that give a stat buff with low uptime.

            – A lot of WoW interface stuff is fairly simple coding: one addon in particular, WeakAuras, lets you insert custom Lua code for displaying interface elements. Sooner or later you’ll run into something their UI won’t let you do, and even if you’re starting from copy-pasted forum code, you’re getting exposure to programming logic that may help later on, and possibly tweaking it for your own purposes.

        • enye-word says:

          You have failed at reading tone and you should feel a moderate and ephemeral sense of shame

          • ChetC3 says:

            Since when is failing at reading tone remarkable around here? Given that he isn’t ranting about gulags and superweapons, HBC is doing extraordinarily well by local standards.

        • Glen Raphael says:

          @HeelBearCub:

          (And I hated almost all of the social aspects of school but loved the classroom time)

          Wouldn’t it be fair to round off that experience as not having “liked school”? If you hated some significant part of the experience, the Polgar sisters (by the account given) did better than that.

          (My experience was mostly the reverse of yours – I liked some of the social aspects but hated most of the classroom time.)

          • HeelBearCub says:

            Wouldn’t it be fair to round off that experience as not having “liked school”?

            No. I was really eager to get to college where (I expected) that everyone would be as enamored with education as I was. Of course, I was disabused of that notion.

            And then, by the end of college, I realized that too much of my experience was colored by my own lens, that my inside view of my experience was wildly distorted (by my own inabilities in the social realm).

            But overall, I loved learning in school. Some of the teachers were bad, but nothing like a majority of them. I have fond memories of some truly meaningful educational experiences with some great teachers: Mrs. Weirs, Mrs. Tyson, Mr. Robinson, Mr. Nelson, Mrs “Oh Fish” whose name escapes me, Mrs. Lancaster, Mr. Hicks, Mr. Bud whose last name escapes me, Mrs. Van Damme and a myriad of others in primary and secondary school.

          • Glen Raphael says:

            But overall, I loved learning in school. Some of the teachers were bad, but nothing like a majority of them.

            I loved learning too, but I felt like school was an active hindrance to learning. The problem wasn’t that the teachers themselves were bad so much as that the system of instruction they were following didn’t work for me.

            In classes taught to a textbook, I learned best by simply reading the textbook. The textbook writer/editor was usually much better at making relevant arguments in a clear and concise way at a comfortable-to-the-student rate of progress than was a teacher. If the textbook wasn’t clear, I’d rather go find a better textbook than have a teacher try to re-contextualize it for me in person in real time while talking to an entire class.

            A teacher doing lectures is in kind of a no-win situation: if they cover material fast enough to keep the fastest kids interested the slowest kids get lost. Kids naturally learn at different rates and have different levels of engagement so for any topic you pick, some won’t get it and will want stuff reviewed while others will have read ahead and are bored silly.

            When I was confused about something in the book, I could argue about it with one of the other students and figure it out that way or I could re-read the confusing bit in the book. So the usefulness of the “teacher gives a lecture” part of instruction was largely lost on me. Teachers were useful to have around in classes that teach some sort of physical skill like pottery or music but in classes with a text, reading the text seemed unambiguously better. Since “teacher gives a lecture explaining a topic in the book” constitutes most of the time one is physically on school grounds, that’s kind of a problem.

            (College was much better in that the teachers were smarter and better lecturers who knew their field better and there was more ability to pick especially good teachers teaching especially-interesting subjects.)

          • HeelBearCub says:

            @Glen Raphael:

            The problem wasn’t that the teachers themselves were bad so much as that the system of instruction they were following didn’t work for me.

            And I agree that this is a problem. Not every teacher, nor every instructional method, will work for every child. An ideal school system would use professional educators to help identify how a child learns best and allow them to be taught competently in that manner. I certainly won’t argue that we are close to an ideal education system.

          • onyomi says:

            @HBC

            Studies have actually found little evidence to support the popular notion of varying learning styles. My own experience as a teacher and learner has also largely taught me that there are good teaching and learning methods and bad teaching and learning methods, but not really some methods that are good for some students and some that are good for others, at least in terms of actually imparting information and skills.

            For example, having taught foreign language to many students and having taught myself a few foreign languages successfully, I’ve gradually learned what works and what doesn’t. Some people claim to learn foreign language differently, but I don’t believe them because I’ve never witnessed anyone successfully learn a foreign language differently from the way I know is the way to learn a foreign language (which is not to say I’ve never seen anyone pass a foreign language test using methods useless for actually attaining proficiency–and this may be a key point: the skill of passing tests and actual skills are often widely divergent and non-transferable, depending on how the test is designed).

            Which is not to say that every class is appropriate for every student. Obviously some degree of variance for skill and experience levels is appropriate. But that’s not so much a different teaching style as it is figuring out who’s ready for which materials and methods.

            Then there’s the question of figuring out who should learn what, which is really a different question. Much of what we teach in schools today is probably useless to most students even assuming they could master it to a high degree, which most of them can’t.

            Put another way, I think the idea of figuring out which types of schooling and teaching method are appropriate for which students and attempting to tailor them is a bit of a red herring. There are good teaching methods and bad teaching methods. I think the focus should be on figuring out and reproducing those, rather than on making ten different types of school for ten different types of student (which is not to say I’m against experimentation in schooling; experimentation is the best way to figure out the best practices, which will then tend to catch on).

          • HeelBearCub says:

            @onyomi:
            I think that the 80/20 rule probably holds true in this realm as in many others. And for some majority of kids just about any method of teaching will work as long as it done well. Both of my parents are educators and one of my fathers long running areas of research was in the study and promulgation of effective techniques for teaching Economics. On the other hand, my mother has taught Montessori for 40 years, and while she firmly believes in effective teaching technique, she also believes in understanding the individual child and their needs.

            I will tell you that there is no way on God’s green earth that I would have done good work if you had handed me a textbook and said “learn this” and then come back 6 months later. I needed the formality of the class schedule and lecture hall and regular forced learning checks. But we also have a number of self-professed auto-didacts in this thread who apparently couldn’t learn in that environment. The only class in which I got a D in college was one where the lecturer was useless and his declaration of what would be on a test was “whatever is covered in chapters 1 through 5”.

            When we are dealing with people on the tail of the distribution, techniques that work for the vast majority and are “universal” may not actually be universal.

      • Bugmaster says:

        What does it explain, exactly ? Personally, I liked more than half of my classes in high school; my Physics, Math, and CS teachers were engaging, attentive, and downright inspiring. Of course, some of the other teachers were basically the opposite of that, but that’s life. Like many people, I didn’t particularly enjoy the social aspect of high school, but I think it made me a stronger person anyway.

        So, what’s your claim here ? Anyone who liked high school is an idiot ? I mean, yeah, my IQ isn’t stellar like yours, so you’ve got one data point; but I think you might need to provide more than N=2.

      • Deiseach says:

        I liked school!* Admittedly, from what I’m reading online from various survivors of the American educational experience, being locked in a cage with lab rats between the ages of 2-17 would be a better environment for a child, but the entire world is not the USA, you know 🙂

        *It had books in it. All sorts of books, including old ones. Lying around on shelves, where anyone could get at them. And the teachers didn’t mind you reading them! What was there not to like about this experience?

        • Bugmaster says:

          I’ve personally experienced something like three or four different public education systems; and, in my personal experience, the rat-cage phenomenon has been universal. The quality of education itself can vary greatly (and some of my experiences were quite positive), but the basic homo hominis lupus est remains the same. But maybe the situation really is very much the opposite in private schools ?

        • Skivverus says:

          I went to a school where they let me read, albeit with some grumbling when I did it while they were trying to teach something else.

          Then I went to a public school for a year, and – sure, there were textbooks, and those were interesting enough when there was time, but there was much more “listen to the teacher on [teacher’s specialty topic] for the thirty-seven minutes you’re in this room” and much less “here’s a wall of books and computers-with-educational-games, go nuts”.

          The year after that, I went back to that first school.

        • xXxanonxXx says:

          I survived the American public school system, and grades 1-5 were a lot like that. Particularly my 4th grade teacher had a closet stacked floor to ceiling with books that she’d let me raid any time I cared. Absolutely wonderful woman.

          Starting with grade 6 things started to go to hell though. Increased bullying (not as extreme as plenty of stories I’ve heard, but still). Increased class sizes where I’d end up with bottom performing students and be forced to keep pace. Eventually (and to be fair this was largely because of other factors outside of school) I just didn’t care about anything anymore and started saying I was going to make Columbine look like a day in the park if I had to spend another day there. They kicked me out, and as “punishment” (this still blows my mind) banished me to the town library with a group of other criminal youths. It was heaven. I got along very well with the other miscreants, and was free to do the coursework at my own pace. The library had already been one of my favorite places in town, so it wasn’t as if I was being forced to be anywhere I wouldn’t have been anyway. After lessons were over I’d grab whatever book interested me that day and settle down by one of the antique banker’s lamps in the reading room. All I had to do was pass the standardized tests the Board of Education would bring my way at the end of the term.

          After a year my sentence was up and I returned to school proper, mostly having things sorted at that point. I skipped class regularly (because what were they going to do, kick me out?), involved myself with a small circle of friends and social activities, and basically just cultivated the rumor that I was an unhinged and potentially violent psychopath so everyone else would leave me alone. It’s all funny to me years later, but it put my family through a lot of grief. At the risk of dodging responsibility, I can’t help but think the whole thing could have been avoided if I’d been set loose in the library from the get go. Clearly, this has biased me in favor of homeschooling.

        • Nornagest says:

          Hey, having to fight starving rats for your lunch every day teaches valuable social skills.

        • Joyously says:

          In eighth grade, my teacher threw my book out into the hall because I was reading it under my desk instead of paying attention to him. Pages came out. I’m still mad about it.

        • moscanarius says:

          It had books in it. All sorts of books, including old ones. Lying around on shelves, where anyone could get at them. And the teachers didn’t mind you reading them! What was there not to like about this experience?

          I would have liked that. In all schools I went, the school library was basically off-limits to students. In the younger grades, we needed a (signed) Special Permit to borrow books, and we could only borrow the books that were registered as suitable to our grade. If you were in 4th grade, you could only borrow from the 4th grade shelf, not from those of 5th grade or 3rd grade. Yes, it is makes no sense. The libraries often closed during recess, the librarians hated to have kids around and often scolded and expelled anyone for making any kind of noise.

          High school was a bit better, but I had first to be very friendly to the librarian before I could get the privilege of being allowed to read a book there.

          • nimim.k.m. says:

            In middle school, we didn’t have a librarian, and “the library” was a room slightly larger than a cupboard. Certain teachers had the keys and in theory could let let the bookish kids in to borrow books, but in reality that didn’t happen much. [1] More often, it was just a storage: teachers would get the whatever the books they decided to include in their “recommended reading for your book card” list for that year and bring them up to the classroom.

            High school had a very good (factual, high literature and like) library open for students during breaks, but I don’t think one could borrow the books. At least I don’t know if one could, there was never a librarian present. Thinking about it, I wonder why the library was so good if one could just walk away with a book. I guess doing that kind of thing never occurs to kids who would obsess over Orwell’s Spanish memoirs. (Yes, that’s when by fanboyism started.)

            [1] Bookish kids like me preferred going to the public library right away. Better selection of titles, and you didn’t have to bother anyone to be let in.

        • beleester says:

          AFAIK it depends a lot on the school’s funding – how many good teachers they can afford, how many extracurricular amenities, how much money they can spend on keeping rats out of the cages… etc. And funding is hugely variable because most school districts are, for some bizarre reason, funded mainly by property taxes.

          Hence the reason you hear American parents worry so much about getting into a good school district.

      • James Kabala says:

        I think that many people, perhaps even a silent majority, look back on their school days with an attitude of apathy/stoicism/”Whatever is, is right.” I don’t think that many people outright loved school, but I think fire-of-a-thousand-suns hatred is not nearly as common as commenters here tend to assume.

        • albatross11 says:

          There’s a difference between hellish bullying (which happens sometimes, and makes for horrible stories) and lots of boring waste-of-time classes under indifferent teachers, going over the same material multiple times while the smart kids quietly think “How many f–king times do we need to explain how to write a complete sentence?” My impression is that most smart kids have a lot of that second kind of memory, because if you were pretty bright and were in a class full of normal kids, you were probably bored a lot of the time.

    • rlms says:

      Don’t worry, you’re not alone. My school even had a high proportion of students from unprivileged socio-economic backgrounds!

    • wintermute92 says:

      Wait… are you just saying you thought your parents were patient and never harsh, or following through with the comparison and saying you thought public school was patient and great and never harsh?

      If it’s the latter, where the hell did you go to public school?

      And… is this why you are so much more on board with “public education like we have now, but moreso” than most people? How can we make your experience something actually common?

      • SpaghettiLee says:

        I went to a great public high school; good teachers, good classmates, robust program for the intellectually advanced, minimal bullying and ostracism. Not only do I remember it fondly, I occasionally wish I could do it over again.

        Was I lucky to find myself in that situation? Clearly I was, but: I’d encourage the people here who loathed their schools to do some reflection of their own, on whether their experience was universal enough to stand in for the entire public school system. People like me are not that rare, and we’re not *all* crazy.

    • SpaghettiLee says:

      I don’t know why exactly, but people here who can discuss issues from medicine to politics to threat of annihiliation-by-artificial-intelligence in a measured and humble fashion turn into ideologues when it comes to public schooling: the whole system is a gulag, overrun with thugs and idiots, and anyone who escapes with their mind and sanity intact is a living miracle.

      My theory is that there’s some confluence of personality traits and values that make a person particularly likely to enjoy commenting here that also make them particularly likely to hate public school. It seems more likely than a community full of people who are wildly different from each other otherwise all focusing the same way on one comparatively minor issue, and being dead to rights about it.

      • AnonYEmous says:

        speaking personally, I should have the same opinions as these guys, but I went to schools that were actually set up for, well…me, and people like these guys. So it worked out OK.

        That’s the academic part taken care of. The social part…it’s being nerdy, I guess. I know that my interest in nerdy pursuits decreased more the more I had friends, and so forth.

      • Debug says:

        I suspect there is also a second confluence of personality traits that makes a person particularly likely to enjoy commenting here but also makes them likely to enjoy public school. However, these people – even if they enjoyed school (Such as myself – and I know a few friends who did too) – find themselves looking at the education system and find themselves deeply troubled that they’re not really sure what benefits it provides. In addition, they see the first confluence of personality traits who really hate it and ask themselves “Maybe I was just lucky?”.

        I feel that effectively sorting classrooms based on ability might alleviate some of the issues people have with public school – but maybe the people here who hated school already found themselves in such classrooms. Really, the solution seems to be (as others have said above) to take seriously the fact that the current education system doesn’t work all that well (and stop blaming teachers for such) and try a whole bunch of different systems to see what works. Maybe the best we can do is make school into some sort of opaque daycare. I’m sure we can provide opaque daycares much cheaper then the current education system (And I feel like hiring babysitters per family would be more expensive – but maybe not) – and that might be considered some sort of win.

        • SpaghettiLee says:

          One problem, I think, is that there’s a real possibility that the entire paradigm of ‘go to school, learn skills, get a job’ is about to be blown to bits by automation and the digital economy and so forth, even more than it is already. In that scenario, solutions like finding ways for teachers and students to connect better or not inhibiting the progress of the smartest ones won’t be worth much.

          My gut instinct is that there needs to be more focus on how to think, how to recognize bias and faulty arguments, how to see things from a meta perspective, etc. Facts and data are cheaper than ever, but understanding them and knowing how they work is still rare. Of course I don’t doubt that my own high school teachers would tell me “We’re trying to do this, honestly, but it’s really hard work and hard to quantify and can’t really be taught the way math or history can.”

          • Aapje says:

            Facts and data are cheaper than ever

            Yeah, but if you don’t have a solid base of knowledge, you often won’t know that they exist and more importantly, won’t be able to assess their worth.

      • The Nybbler says:

        It’s not an uncommon feeling. Joss Whedon said about his Buffy the Vampire Slayer: “Sunnydale High School is based on every high school in America because so many kids believe their school is built on a Hellmouth”.

        • HeelBearCub says:

          And yea verily again I will say unto you, do not lay at the feet of high schools that which belongs on the altar of adolescence.

          • The Nybbler says:

            High school has many negative features which cannot be laid on the altar of adolescence.

          • HeelBearCub says:

            @The Nybbler:
            Sure.

            But I think the baseline for kids is more “Lord of the Flies” than something idyllic. The shit comes from us. We make it. It’s important to keep the baseline of “social creatures in adolescence” in mind when evaluating HS.

            As Sartre said, “Hell is other people”.

      • carvenvisage says:

        I don’t know why exactly, but people here who can discuss issues from medicine to politics to threat of annihiliation-by-artificial-intelligence in a measured and humble fashion turn into ideologues when it comes to public schooling: the whole system is a gulag, overrun with thugs and idiots, and anyone who escapes with their mind and sanity intact is a living miracle.

        I really didn’t like school, but in both of the other (local) schools that friends of mine attended, people were literally stabbed. The thugs thing isn’t even an exageration or metaphor in that case.

        (great characterisation, I might use it)

  19. Manx says:

    So, as many have already pointed out, I don’t think you can dismiss the genetics involved as impossible because of ‘regression to the mean.’ Let me point out that it is NOT NORMAL for a 4 year old to sit for 48 hours a week (8 hrs a day x 6 days a week) doing anything! How much time have you spent around small children? If it is true that these girls were actually happy doing this (which is you know, taking this very self-assured author’s claims at face value), then there was already something very unusual about these girls besides their raw intelligence. You’ve pointed out that there are specific mutations – like Gaucher’s trait- that are heritable and highly linked to high IQ. Such a mutation would have a 50% heritability and not regress to the mean. If these girls had either such a trait for intelligence or another such for incredabily long attention span, it would not be so improbable for all three to inherit it. It’s also unclear to me why you would value training chess champions so highly. They focused huge portions of thier lives to a zero-sum game. Did any of these girls go on to win nobel prizes or do anything useful for the world? Ok then. Why do we want to do this to kids again?

    • baconbacon says:

      Let me point out that it is NOT NORMAL for a 4 year old to sit for 48 hours a week (8 hrs a day x 6 days a week) doing anything! How much time have you spent around small children?

      It depends how you categorize things. My 4 year old doesn’t spend 48 hours a week doing anything specific, but he probably hits 20-30 hours a week playing with his dump trucks in the sand box, and during the winter it was probably a similar 20-30 hours with his train sets. If you categorized his play as ‘playing with motor vehicles’ he is >30 hours a week of that most weeks. If ‘practicing’ included all related forms (reading chess books, playing with chess pieces while not playing chess, inventing new variations on chess, teaching their younger siblings how to play chess) then it is plausible for her to hit those numbers without being exceptionally weird.

      • baconbacon says:

        Semi related, I was talking to my wife’s uncle who grew up on a farm, and he was talking about how he spent virtually every waking hour he could remember when not doing chores building things (boats out of milk bottles, pouring his own concrete ramps for trucks and cars ~8 years old, and just perpetually being in the dirt/water).

    • James Miller says:

      Children with autism can develop narrow obsessive focuses and autism probably has a strong genetic basis.

  20. NoahSD says:

    Hi Scott,
    Just FYI, the phrasing “previously in series…” makes it sound like this article used to be part of something else—at least to me.

    • enye-word says:

      Data point: maybe I’ve been ruined by LessWrong, but the phrasing does not make me think that this article used to be part of something else.

  21. Eponymous says:

    On the topic of the psychometrics of top-level chess: it seems that having an excellent memory is highly selected for. There are many tales of memory feats by chess champions, such as Paul Morphy memorizing the entire Louisiana civil code. Magnus Carlsen reportedly memorized all sorts of trivia as a small child (flags of the world; car brands.) Kasparov himself is reputed to have a near-perfect memory (I think he was once quoted as saying, “The day I have to write down a phone number is the day I quit chess.”). Or just the fact that “blindfold simul” is a thing, which boggles my mind.

    Here’s a page that collects stories of memory exploits by great chess players. Regarding Kasparov:

    Garry Kasparov says that he was able to remember all the master games he has played. In 1987-88, the German magazine Der Spiegel went to considerable effort and expense to find out Kasparov’s IQ and test his memory. Under the supervision of an international team of psychologists, Kasparov was given a large battery of tests designed to measure his memory, spatial ability, and abstract reasoning. They measured his IQ as 135 and his memory as one of the very best. Kasparov was asked if he had to re-evaluate the positions on each board every time he had returned to make a move in a simul, or if he remembered the positions all the time. Kasparov replied that he in fact remembered all the positions. He also said he could recall the moves of all the games he had played in the past 6 months.

    Vishy Anand lectured about chess skills and that chess players could easily remember ideas and patterns taken from millions of games. The first skill needed was to develop memory hooks. You learn a few mates and a few tricks. Then you slowly progress, seeing the games of the great players, classic examples that every chess player must know. The great player explains his game, and explains the key moments. A lot of games are accompanied with diagrams of key positions where something interesting happened. Thanks to all these hooks, it is easy to remember these games years later. Anand’s mother says that he has always had a photographic memory.

    http://billwall.phpwebhosting.com/articles/Memory.htm

    • bbeck310 says:

      This fits with an interpretation I’ve had about the way the rating system plays out, based largely on what I’ve read about grandmasters compared to what I know about expert and master level play (disclosure; my USCF rating peaked around 1950). The major difference between experts/national masters (ELO 2000-2400) and IMs/GMs (ELO 2400-2800) seems to be that the IMs/GMs have developed an extensive mental database of positions that are favorable or unfavorable, and what to do with them, so that they can call upon that mental database as a shortcut in their analysis. That is, the IM/GM can analyze a position, figure out that it gets to another position that most chess players would think looks somewhat close, and know that the position is actually a win for one side (“the rest is left as an exercise for the reader.”) The 2000-2400 player still has to mainly rely on heuristics to figure out what to do with unfamiliar positions.

      • Eponymous says:

        I’ve heard a variation of this theory (can’t recall the source offhand, maybe a book by Silman? clearly I’m not GM material!), but I’m not convinced that this is the primary phenomenon that makes memory so essential. I agree with the idea that high level players make use of an extensive mental “database” of chess positions, and which is useful for pattern recognition. But I think that the phenomenon of top players having excellent memories is mainly about calculation.

        Calculation requires forming mental images, manipulating them accurately and quickly, and remembering the results. I imagine that having strong visual working memory capacity (as seen in an extreme form in people with eidetic memory and other savant types) is essential to calculation.

        • bbeck310 says:

          Improved calculation can’t hurt (and it wouldn’t surprise me if you saw the idea in a Silman book, as he’s been my primary guide in adult chess study). But calculation doesn’t explain non-tactical world champion-caliber players very well, or the current crop of preparation-intensive champions–and it doesn’t explain the difference between a 2400 player and a 2800 player very well. I’d find it hard to believe that a world champion like Petrosian or Karpov was that much better at calculation than the average GM.

        • Diadem says:

          Anecdote is not data, but as a young child I was an extremely talented chess player (one of the top players in The Netherlands, which is a reasonably strong chess country), but my memory is nothing special. I am among those 5% of people who have no visual memory at all, and this never seems to have impacted me negatively (although who knows, maybe I’d have been even better with a strong visual memory).

          Memory is important in chess. I won’t dispute that. But I don’t think it’s visual memory. Memorizing chess positions for example is all about pattern recognition. During chess training we used to practice this. Look at a position for 30 seconds, do something else for a few minutes, then reconstruct it from memory. This is easy for an actual chess position, but (near) impossible if you place the chess pieces randomly on a chess board.

          • Eponymous says:

            Huh. Your comment lies at the intersection of several interests of mine, so I hope you won’t mind if I ask you further questions!

            I imagine that as a good player you calculated a lot of variations. My understanding is that most people calculate by forming a mental image of the chessboard and moving pieces around on it. If you have limited ability to form mental imagery, how did you experience calculation? Was it verbal (e4 e5 Nf3 Nc6 Bb5) or did you “feel” moves? Did you have to look at the board when you calculated, or could you do it with your eyes closed? Can you play blindfold chess?

            Also, were you more of a tactical or strategic player? Were you an accurate calculator, especially in positions requiring deep calculations (i.e. long variations, not just total number of variations)? Did you have trouble holding multiple variations in mind to compare them?

            Also, if you are not a visual thinker, are you primarily (exclusively?) a verbal thinker? Do you “think in words” all the time? Also, are you generally intelligent (of course you post on SSC, so mean IQ 140!), and how are your abilities allocated between verbal/math etc?

            What do you mean when you say that your memory is “nothing special”? Do you have good retention of what you read? What you hear? What you see? How are you with names/faces/phone numbers/addresses? Do you navigate by landmarks or forming mental maps?

          • Chalid says:

            I too would be fascinated to hear more about how one can be good at chess without doing any visualization! My experience of playing chess is all about manipulating pictures, though I’m an extremely visual thinker in general.

      • phil says:

        not a chess player

        but, my local library has a copy of https://www.amazon.com/Chess-5334-Problems-Combinations-Games/dp/1579125549/ref=sr_1_1?ie=UTF8&qid=1496237611&sr=8-1&keywords=lazlo+polgar+chess

        I’ve wondered how good a player you would become if you just input each puzzle into a spaced repetition program and went to town

        is the database of shortcuts defined enough that you could brute force them into your head w/ SRS?

    • US says:

      +1 for the memory bit. Memory is a very big deal among the people at the highest level of chess. Strong GMs have excellent memories, not just a few of the people at the top, but pretty much all of them. This is likely even more the case now than it was before the computer age; memorization of lots of long opening lines is necessary for reaching the super-GM level, and it’s frankly obvious when they talk about chess. Svidler banter-blitzing on Chess24 and randomly spotting during a random game that he had that position against some specific opponent in a tournament in Switzerland 15 years ago (or whatever), Caruana banter-blitzing with Eric Hansen and spotting that his low-rated opponent seems to be playing some critical line he saw during (some tournament I can’t recall a while back), Seirawan is providing commentary and during the commentary he reminisces about a similar game he had against (some GM) in 19xx, only he played 9.a3, which is obviously different – this stuff happens all the time. When I was relatively new to chess I had a hard time believing that they could actually remember stuff like that, so I checked out such claims a few times and spotted that they’re usually right. It’s not just opening theory, it applies to all aspects of the game; they remember drawing lines in rook endgames that other people don’t, they understand positions better in part because the remember the strategies you need to apply in order to reach those drawing lines, they likely have a larger mental library of tactical motifs, etc., etc.

      There’s a lot more to reaching the top level than just a good memory, but memory seems to be really important.

      • I’m not sure if memory plays the same role in bridge, but, in my experience, top bridge players can describe a game they played six months before in detail.

        • US says:

          Well, it does sound a bit similar…

          One thing I didn’t mention above is that I’m not really sure what’s going on in the context of chess is purely a selection effect, i.e. ‘only people with a good memory become great chess players’. In my experience it gets easier to remember games and lines as your skill improves; in some sense it might be argued that your ‘chess memory’ improves as you get better at playing the game.

          When people play official rated chess games in a tournament setting, it is required that they write down their moves – this is part of the rules of the game. After a game people will sometimes get together in an informal setting and talk a bit about the game and what happened, if a particular move was quite bad or very good, suggestions for improvements, stuff like that. When I started out playing, I’d never be able to remember a game I’d just played – I always had to look at the moves I’d written down in order to be able to remember precisely what had happened. After I’d played a few tournaments, my need for consulting the written version went down a lot; I could remember most of the moves quite well without having the written version next to me. This seems to be a very common development as people get better; when your chess skills improve you also tend to get better at remembering your games, you get better at visualizing the board without having to set up the pieces, you start finding it natural to discuss particular key moves without providing a great deal of context. I think it sort of links a bit to a common pattern you often see when people learn new things in general; it’s easier to remember stuff if you can relate what you learn to something you already know, to add context. When you start out playing chess it’s all very confusing, lots of rules and concepts and it’s hard to keep track of everything; but then as you learn some decision rules and strategic concepts and gradually develop a framework for understanding the game, it might be that the patterns you learn also enable you to better keep track of the ‘chunks’ that make up the games you play. A way to think about it might be that the beginner tries to remember the individual moves because that’s how you start out, whereas the GM on the other hand think in terms of much larger chunks that are easy to combine in the mind – e.g. the opening setup (e.g.: ‘Ruy Lopez, Breyer variation…’), attacking plan(?), key moves that defined how the game progressed…

          • publiusvarinius says:

            I think this works with every profession. For example, my otherwise absent-minded mathematician friends are often remarkably specific about their own work: “Jones and I used this technique in Theorem 2 of our 1997 paper”.

            And many programmers can relate stories of disastrous bugs, bad coding choices and project management gone spectacularly wrong in excruciating detail (see e.g. the Daily WTF).

          • US says:

            Yep, I agree, it certainly is far from limited to chess. I was mostly trying to remove some of the ‘mystery’ from what to the outsider would perhaps seem really strange feats of memorization etc. It’s perhaps slightly weird if a random hobby player could remember precisely all the details of a game he played many years ago, but someone who spends his entire career and a great part of his life playing/studying/etc. chess? Much less weird. That’s the point; to a strong GM, chess isn’t just a game, it’s like a job. And most people remember quite a bit of work-related stuff because they spend a lot of time working and thinking about work-related stuff. If the work is the type of work that even to some extent puts a big premium on memorizing accurately what you worked with in the past, it’s gets even more understandable that these people are very good at remembering weird details that other people would not really think about (like why the move 14.h4 in XYZ opening is a good idea, and why it worked when I played individual Q in that tournament in Gibraltar).

    • eyeballfrog says:

      “Or just the fact that “blindfold simul” is a thing, which boggles my mind.”

      Blindfold chess in general is something I’m always amazed at. I think I could probably get about 10 moves into the opening before losing track of things. Meanwhile, Judit Polgar once performed the king, bishop, and knight checkmate blindfolded (which requires very careful coordination of the three pieces) and didn’t even use one of the established algorithms.

  22. topynate says:

    If the expected mean is 150, and the sample mean, of the three girls, is 160, and the population standard deviation is 15, then I make the chance of an outcome at least that good to be around 12%.

    That said, 150 implies a population mean of 130. Even for our Hungarian cousins, that’s rather on the high side. As for the parents both having an IQ of 170… well, with a population mean of 130, why not?

  23. ZeitPolizei says:

    So, Laszlo Polgar is still alive. How hard can it be to get in contact with him and ask where his book is?

  24. Eponymous says:

    Incidentally, Judit is at another level from her sisters. She’s simply the best female player ever. She peaked at #8 in the world (2735). Susan peaked at 2577, which is…good for a woman (would make her #3 woman today, with rating inflation; 100 points below the #100 man).

    A story: Susan used to study chess after Judit had gone to bed (Judit is 7 years younger than Susan). When she ran into a really difficult problem that stumped her, she would as a last resort wake Judit up. Judit would look at the problem for a bit, maybe a minute or two at most, and then solve it.

    (Not sure of accuracy of above story; related from my memory, from a story Susan told during live commentary of a chess tournament a few years ago.)

    • Squirrel of Doom says:

      Yeah, I’d say Judit has settled the question if women can be world champions. She was close enough that she’s in that category.

  25. Joyously says:

    One of my friends in grad school was a Chinese girl with a young daughter. She was always worried about her daughter–what charter school to apply for, when she should start pushing her to play piano. She asked me once whether I had been into princesses when I was little, and was relieved when I told her yes (if I could end up getting a PhD in engineering, then princesses hadn’t hurt me and her daughter was just being a normal American girl). She seemed to worry about living up to her husband’s friends’ wives parenting methods.

    After learning that one of my brothers was in med school and the other was a successful programmer, she said, “I should talk to your mom and ask her how she did it.”

    “She didn’t really do anything,” I said. “Your daughter’s parents are both engineering PhDs. She’ll be fine.”

    Though looking back, maybe that’s not true. I also have two sisters who probably could have made it through an engineering program, but chose differently. If my friend thinks success for her daughter is going into science/tech/medicine, maybe she won’t be “fine.”

  26. rlms says:

    I’m not a very strong chess player, but know enough that I’m pretty confident that Levitt’s IQ estimates are wrong. He thinks that someone with an IQ of 100 could reach an ELO of 2000, which I think is a ridiculous overestimate. An ELO that high is pretty impressive, you definitely need to be of above average intelligence to get there. But I think he’s also greatly overestimating the IQ you need to get a moderately higher IQ. I admittedly don’t have much personal experience with strong grandmasters, but I’m very doubtful that you need an IQ of above 160 to become one. The model I’d use is that you need a certain degree of intelligence (somewhere between 110 and 130) to become a reasonable player (ELO of around 1400). But beyond that, success is largely down to practice (amount and quality). Memory and innate chess ability (distinct from intelligence) probably account for more of the remainder than IQ.

    • bbeck310 says:

      Levitt’s equation is silly and implausible, but I don’t think your numbers work all that well either. Take as a key data point the IS 318 program, a chess team out of a predominantly black, lower class middle school in New York City. This is not a population known for generating high IQ kids, and it is a superstar in the US scholastic chess scene. In 2012, its team (including two masters (Elo > 2200) and an expert (Elo > 2000)) of middle schoolers won the US High School National Championship. I’ve met these kids, and they’re smart, but they don’t have the kind of academic achievements suggesting that their unusual chess skills go along with ultra-high IQs. What they do have is a stellar teacher.

      I’ve known plenty of chess players who didn’t impress me as that smart generally but had ratings substantially above 1400, and I’ve known plenty of very smart people who couldn’t really pick up the game. From my experience (I peaked at about USCF 1950 around 8th grade, with my top tournament place being 3rd in the 6th grade section at the 1994 US K-12 Championships), and stopped playing seriously around 10th grade, later as an adult coaching kids for a few years in New York and Chicago), I’d expect that there’s at most a very weak correlation between chess rating and IQ.

      • rlms says:

        I think that supports my numbers. I’m saying that those kids (or other strong players) could potentially have IQs as low as 110 (although I think closer to 120 or maybe higher is more realistic). I don’t think saying strong chess players invariably are in the top 30% of the population by IQ (or even in the top 10%) is unreasonable.

    • publiusvarinius says:

      It’s Elo, not ELO. Got its name from (surprisingly non-Jewish) Hungarian physicist and chess master Arpad Elo.

  27. Mixer says:

    I stumbled upon this, but it demands account creation (and a version of Flash I don’t have):

    http://www.ustream.tv/channel/ydjsnz72p7u

    Looks like a PDF copy, not sure what language. Apologies if mentioned already, but I scanned the comments and didn’t see the link.

  28. baconbacon says:

    I am very interested in this book now. If we are going to set up a fund to buy and translate a copy of the book I will contribute $100.

  29. moridinamael says:

    what I got was staring out a window all day as my teacher declared that we were going to make a collage about the meaning of Respect.

    You had a window?

  30. Z says:

    First, I have to wonder, what kind of woman agrees to be recruited to marry a man so they could test his hypothesis of raising “any child” to be a genius by raising children together?

    Second, I have to wonder, what kind of 4-year old girl likes chess?

    I think the answers to both are: not in the realm of ordinary. I’m reminded of Yao Ming. From his Wikipedia entry:

    In 2005, former Newsweek writer Brook Larmer published a book entitled Operation Yao Ming, in which he said that Yao’s parents were convinced to marry each other so that they would produce a dominant athlete, and that during Yao’s childhood, he was given special treatment to help him become a great basketball player.

    If regression to the mean worked like you propose, it’s “biologically impossible” that Yao Ming is 7 ft 6 in. His dad is 6 ft 7 in and his mom is 6 ft 3 in. There’s more variance in genetics than you assume. For regression to the mean to kick in, one would need a much greater sample size of children from the same parents.

    • baconbacon says:

      This analogy only holds if Yao’s parents were the only couple encouraged to have kids for athletic potential.

      • Z says:

        1) The main point here is his height (very similar if not same heritability as IQ) far outstrips his parents’. Until humans are bred towards intelligence to the same degree as modern racehorses, there’s still room for variance up, not just down, when two exceptionally gifted people have children.

        2) Both of Yao’s parents were professional basketball players. That degree of assortative mating seems rare. From the limited evidence I could find:

        The conjoining of an elite NBA basketball player and a WNBA star might mean future professional basketball progeny if a family happens. But the merging of WNBA and NBA stars in coupledom is rare, despite the similarities in lifestyle—not to mention, height. The recent news that Kevin Durant is engaged to Minnesota Lynx player Monica Wright, got us thinking about other WNBA-NBA relationships.

        Since unions like Monica and Kevin are so rare, we thought we’d provide four other NBA and WNBA relationships, or in the case of one couple, simply a college relationship that started when both played for their famous alma mater. With a dearth of WNBA and NBA couples, we had to include them.

        Going back to Polgar, it’s similarly rare, if not as rare at that level for highly intelligent people to assortatively mate. On top of that, it’s unusual for highly intelligent couples to have children at all, much less three. Polgar’s story seems to me a major outlier. His explanation for his children’s outcomes would be more plausible if he and his wife raised randomly selected adopted children, though I’m glad they didn’t do that.

        • baconbacon says:

          1) The main point here is his height (very similar if not same heritability as IQ) far outstrips his parents’. Until humans are bred towards intelligence to the same degree as modern racehorses, there’s still room for variance up, not just down, when two exceptionally gifted people have children.

          The point Scott was making is that it is very unlikely for two people to have children of that IQ level unless they had EXTREMELY high IQs themselves. Given that the predictions about being able to make them brilliant came before he even had them, and that there aren’t (known) hundreds of thousands of people trying to do this and failing then it lends credence to the idea that his teaching style is the important factor.

        • Tarpitz says:

          Jaden and Jaz Graf-Agassi do not appear to be destined for professional tennis, which to my mind probably speaks well of their parents – both of whom were themselves hothoused by their respective fathers. “Successfully”, I guess, but not terribly happily.

        • Steve Sailer says:

          Track and field stars get married quite often: e.g., Florence Griffith-Joyner was married to gold medalist Al Joyner, whose sister Jackie Joyner-Kersee was married to her coach Bobbie Kersee.

          A track meet tends to be like a movie set: a coed environment with a lot of good looking people with a lot of free time for flirting.

          Golfers, in contrast, almost never get married. The men and women’s pro tours only get together for one tournament per year.

          Tennis tournaments tend to be coed — e.g., the U.S. Open tennis tournament has the men and women play at the same site at the same time, while the U.S. Open golf tournaments are held at different courses at different times for men and women. So, tennis players seem to get married a little more often than golfers, but maybe not as much as track athletes.

        • Deiseach says:

          His explanation for his children’s outcomes would be more plausible if he and his wife raised randomly selected adopted children, though I’m glad they didn’t do that.

          Apparently they were going to try exactly that to prove it was the method, not the parental genes, that was the secret of the Polgar girls’ success but it fizzled out (as you say, and a good thing too):

          Questions will always chase the Polgárs; such is the price of their peculiar type of fame. But for a time, at least, it looked as if they might answer the biggest question of them all. The family has always known that genes undermined their can-do message. Laszlo and Klara were smart and determined. Isn’t it natural their girls were, too? In the 1990s, to test that, Laszlo almost began raising another set of kids; an eccentric Dutch benefactor would cover the costs. But the plan foundered due to Laszlo’s insistence on adoption. He feared raising prodigies, then not benefiting from their success, Susan says. And Klara made it clear: She did not want more kids. No more experiments.

        • Going back to Polgar, it’s similarly rare, if not as rare at that level for highly intelligent people to assortatively mate. On top of that, it’s unusual for highly intelligent couples to have children at all, much less three.

          I don’t have data beyond the couples I have known, but that doesn’t sound plausible. There is no strong reason for someone who is very tall to prefer a very tall spouse. But there is a strong reason why someone who is very smart would like a very smart spouse, ideally one smart in a similar way.

          I can believe that very smart people may have fewer kids than average, but most married couples have kids. Most of the very smart couples I know personally had at least two children.

          Do you have data to support both claims?

        • The Nybbler says:

          @DavidFriedman

          In general short women prefer tall men. Taller women prefer taller men. My evidence is merely anecdotal, but it’s blatantly obvious to short men. And there’s non-anecotal evidence as well.

          On the other hand, it’s often claimed that men prefer women less intelligent than themselves. And there’s some evidence for that as well.

          • Tibor says:

            I’ve never heard the claim that men prefer women less intelligent than themselves and I find it hard to believe. Tallness is a fitness indicator, intelligence is also. People are programmed to find the highest fitness partners the most attractive. They might differ in how much weight they assign each fitness indicator but they are unlikely to assign it a negative value.

          • The Nybbler says:

            @Tibor

            In the US at least, the claim that men aren’t interested in intelligent women is common to the point of cliche. You hear it in feminist laments (usually claiming men are intimidated by intelligent women), you hear about women playing dumb to be more attractive, and I’m pretty sure Dorothy Parker’s “Men seldom make passes / at girls who wear glasses” wasn’t referring to the aesthetics of corrective eyewear.

            So perhaps intelligence isn’t such a fitness indicator. Or perhaps the complaints are nonsense, but I’m not so sure.

          • onyomi says:

            Worries about the supposed bias of men against intelligent women are intense in East Asia. In China there’s a wry joke about female PhDs which implies they’ll never get married.

          • Aapje says:

            This study found that is was very important/a must have to 46% of women (vs 24% of men) that their partner earns just as much or more than them.

            As intelligence is highly correlated with income, it makes sense that men would consider it more likely that a less intelligent women is interested in them and thus would be more willing to spend the effort of wooing her.

            Although the study didn’t look into it, I’ve also heard people claim that dating tends to revolve more around the man proving his ability to entertain her with wit than vice versa. As it is surely easier to be considered witty by a slightly less intelligent than a slightly more intelligent person, if this is true, it would suggest that dating efforts by men aimed at slightly less intelligent women are more effective. And again, we can assume that men will logically prefer women with whom their dating efforts are more likely to be successful, all else being equal.

          • Tibor says:

            @Aapje: That’s mixing what’s easy and what’s desirable. By the same argument you can probably “show” that men prefer women who are less physically attractive. And it might indeed be the case that very beautiful women receive less interest from men than still good looking but not as good looking women, because a lot of men simply will be afraid to approach them. But that doesn’t mean they find them less attractive.

            I also think there is an important distinction to be made between short term “flings” and serious relationships. If men are looking for someone they want to marry and/or have children with, they are going to put a lot more weight on intelligence and personality of their partner than otherwise.

          • Aapje says:

            @Nybbler

            I looked at the Huffington Post story and I think that the most likely explanation is that the men felt that they lost status when the researchers openly told them that their scores worse on the test. This is a far better explanation than that they disfavor intelligent women, because the men weren’t less willing to date the woman if they were told their lower test scores in private. You would expect that they would be less willing to date a smarter woman even if that woman doesn’t know that they are dumber, if they were put off by higher intelligence, rather than the other person perceiving them as being stupid.

            To test this, you’d want to do a similar experiment where men lose to a woman in a game where IQ doesn’t help you win. If they are then also less willing to date, it’s a more general loss in status which makes them reticent to date.

            @Tibor

            Nobody in this thread argued that men find intelligent women less attractive. It was always about preferences, where you can’t easily distinguish between a preference for a more easy conquest vs a preference for a more attractive person.

            PS. Also, the Huffington Post is pretty obnoxiously misandrious.

          • Tibor says:

            @Aapje: I guess you’re right, nobody actually made that claim.

            The thing is that women usually want to look up to the man, quite literally in that they prefer men taller than themselves, all else held equal, and in a more abstract sense in which it is usually used. This is why a story where a girl saves a prince and marries him is not believable. If the prince can’t even save himself he’s clearly an inferiour specimen 😛

            I think that men want to look up to their partners as well (in the abstract way) but they care about it a bit less than women do, particularly in less serious relationships.

            The problem is when one interprets this in the way that men have a superiority complex and hate women who are better than them in something. I’d say that if anything it is more likely an “inferiority complex” in that they feel they are not good enough for some women and might avoid them even though they might have a chance with them.

            I think the same thing is going on with tall women, although I’m less confident here. I’m average height (I guess that 180 cm is still below average by Dutch standards, but the Dutch are the second tallest in the world after the people from the Dinaric Alps) so there are not all that many women taller than me but I don’t find those who are any less attractive. I would assume they would find me less attractive since they’re taller though which would make me a bit reluctant about approaching them. It even works in a relationship. My ex-girlfriend was exactly as tall as me, so when she put on even slightly high heels, which she didn’t do much anyway, she was taller than me. I did not find that all so bad but I preferred it when she wore normal shoes. At the same time, I find it slightly more attractive when a woman is at least 165-170 cm or something, again all things held equal. I don’t know if that is just my personal preference or if it is common for men to find taller women attractive.

          • I see a different implication of “men like women to appreciate their wit.”

            People like to be appreciated, especially for what they value about themselves. If what you value about yourself is being good at working with ideas, you want to interact with someone who has enough of the same ability to actually appreciate what you are doing, not just pretend to. That doesn’t require a woman smarter than you are, but it eliminates women substantially less smart than you are.

          • Tibor says:

            Partially expanding on David’s argument – suppose that men consistently prefer women who are less intelligent than themselves (for whatever reason) and suppose that this is not a quirk of the modern age but something consistent in time. Then we have a problem explaining the lack of a difference in male and female intelligence in humans (there might be a difference in variance but there doesn’t seem to be a difference in mean). Fueling a large brain and making sure that it develops correctly is very costly. If it does not increase your chances of finding a mate with good genes, it is better to focus the energy on something else (looking youthful for longer for example).

            There are a couple of possibilities:
            a) Men preferring women less intelligent than themselves is not a real pattern (if even a consistent minority of men did that it would influence the average intelligence of women)

            b) It is real, but only for some relationships. More specifically it is not real for serious relationships that produce children.

            c) Some men are like this but there is a different minority of men who prefer intelligent women much more than other men which compensates for the decreased incentive to produce comparable intelligence in women. I find this at least anecdotally true – it seems to me based on casual observation that smarter men put more weight on the intelligence of women than less intelligent men do and that they do so in absolute terms, not relative to their own intelligence. Smart women who feel frustrated by the lack of dates might then simply be searching in wrong demographics. A college fraternity party or something like that might not be the best choice (but maybe I am just biased against frat boys 🙂 ).

            d) David’s argument works for women too – you need intelligence to detect intelligence. If you are a woman who wants a very intelligent partner, you have to be intelligent yourself. You don’t have to match someone’s intelligence to recognize his wits but if you want to be able to screen as much of the population as possible you still need to be very smart. I am not sure if this alone is enough to compensate for the supposed lack of preference for intelligence in men.

          • Toby Bartels says:

            Another possibility: genes for intelligence are not sex-linked. Even if they are (and as I understand it, some of them are), they’re more likely to be on the X chromosome (which everybody has) than on the much smaller Y, so it’ll still be hard for evolution to select for sex differences. (There are various ways that this can still happen, but it makes it easier for your other possible confounders to interfere.)

          • Skivverus says:

            @Toby Bartels

            It would not be surprising to me if there were genes contributing to intelligence/intelligence-like mental traits on the X chromosome; the expected result of this would look like “humans have similar average intelligence, but males have a wider variance”, and this in turn is consistent with the relative sex distribution of hobos and nerds (so far as I’m aware, cross-culturally).

            Edit: as John Schilling’s pointed out, if there is a difference in variance, it’s not very big.

          • John Schilling says:

            On the one hand, the Ashkenazi example suggests that the most potent intelligence-linked genes might also be the most dangerous, where having one copy means you get an overclocked brain and having two copies means your head explodes. So it would make sense for a mature evolutionary process (rather than the obviously rushed Ashkenazi hackwork) to put those on the X chromosome, where X-inactivation means you’ll only ever get one copy at work at a time.

            On the other hand, Scott already linked to a study, which looks better than most in this area, that seems to indicate that there isn’t much difference in IQ variance between men and women. Nice theory, meet pesky fact…

          • Tibor says:

            @Tony: I’m not a geneticist but from what I gather that is not an obstacle. You can have genes programmed to do their thing only in the presence of a certain hormone or do one thing in the presence of one hormone and another of stimulated by another.

          • Toby Bartels says:

            @Tinor: Yes, that’s one way to do it, but it makes things more complicated and slower to evolve. I’m not a geneticist either, but my understanding is that sex differences take a lot more environmental pressure to select for. Obviously they do manage to happen; it’s just a complication that should be taken into account.

          • Tibor says:

            John: which makes my case even a bit stronger, you have to fit not just the mean but the entire distribution (assuming that intelligence has normal distribución)

          • Tibor says:

            Oh sorry Toby, I keep misreading people’s names 🙂

          • Toby Bartels says:

            That’s OK, Tibor, I got my petty revenge. 🙂

          • Aapje says:

            We know that historically, a far larger percentage of women than men procreated. So men having higher variance can be beneficial to humanity as a whole if women get to procreate with the genetic lottery winners among men, while the men to whom the genetic lottery was not so generous, simply became a genetic dead end.

            If all uterus’s were needed, it would make more sense for nature to reduce the variance in women, so a very large percentage of them would be capable of producing a healthy baby.

            You’d pretty much automatically get a disbalance between male and female IQs of those who reproduce, as many men with really low IQs would be disqualified, while high IQ men would be very desired, while the retatively fat long tail of highly intelligent men would simply not have the option to all have a highly intelligent woman, as women would not have a far more lean long tail, with women being closer to average… on average.

          • Tibor says:

            @Aapje: Sure, that’s the common argument and it makes a lot of sense – not just in intelligence but pretty much any trait including personality. But if data shows that it is not the case (I don’t know how good that study is but let’s say there are no problems with it), then it is not the case. Maybe it really is not so easy genetically to make intelligence depend on sex as someone suggested here. With personality you do see obvious differences between sexes and they seem to be heavily regulated by the sex hormones and then of course obvious physical differences are obvious. In terms of personality women seem to be less extreme and risk-taking which also makes a lot of sense based on the argument you mentioned. Especially if becoming a “genius” is influenced by behaviour in a nontrivial way then you might see a lot more male Nobel prize winners etc. simply because men are more likely to do things like playing chess 8 hours a day and generally take more high-risk high-return actions.

  31. Plucky says:

    For most fields, I have been under the impression that IQ/intelligence is essentially a threshold matter- you need to have a minimum intelligence level to be able to survive, but once over the threshold the marginal value of intelligence is modest, and usually overwhelmed by other factors. Unless you really do your statistics right, very easy for a threshold effect to look like a high marginal value effect given the amount of noise involved

    • moscanarius says:

      This has been my impression too.

    • Anonymous says:

      I think that it’s actually a double-threshold thing, especially nowadays. If you’re too stupid, you fail at life. If you’re too smart, you fail at reproduction, which is just failing at life shifted one generation.

    • Eponymous says:

      I doubt that this is true. If IQ matters enough that nearly all people in the field are above score X, this implies that additional IQ points are very important in the neighborhood of X. But then why wouldn’t IQ keep mattering above X? That sort of nonlinearity seems a priori unlikely.

      Also, my impression is that the very top people in intellectually demanding fields are simply much smarter on average than the median person in those fields. That certainly seems to be the case in my own field (economics). I’ve interacted with the top people in my field, and it’s readily apparent that their brains work faster than the average person.

      This is supported by some prospective studies of high-IQ individuals, i.e. the SMPY, where students were identified by high scores on the SAT at age 12. Those that scored relatively higher went on the achieve more on average. The same goes for other studies of very high IQ individuals selected as children. Even one of the studies often cited as supporting the “threshold” notion says nothing of the sort when interpreted properly.

      To the degree there is any empirical validity to the threshold idea, I think it’s far more likely that it is due to the extreme rarity of high IQs, and particularly the higher rate at which the frequency of IQs declines as you move up the IQ scale (and also the noise in measuring truly high IQs in the first place). This can create the appearance of a clustering above a certain cutoff.

      To illustrate what I mean, observe that there are 16 people with IQs between 130-145 for every person 145+. By contrast, there are about 6 people with IQs between 115-130 than with IQs 130+. And only 2x as many people 100-115 as 115-130.

      Thus even if every IQ point increases the odds of success by a fixed ratio, given that other traits matter as well you would still find the majority of successful people are clustered around a certain value, which then appears like a threshold.

      (Also, some fields have an externally imposed threshold effect in the form of a certification process, which cuts off the bottom. This is true if it’s e.g. harder to get a medical degree than it is to succeed as a doctor.)

      • moscanarius says:

        It may turn out you are right, Eponymous, but I think we should consider the alternative explanation raised by Pluck, especially given that most fields of human activity are not so abstract as the one you use as example.

        I don’t doubt your impressions are true about the fields of economics (and would bet they will be true for every theory-heavy field), but I urge you to think about other fields that have intelectual requirements, but do not relly so heavily in creative theoretical work. Or fields of work that, though requiring intelectual power, also benefit greatly from other sorts of abilities not correlated with inteligence.

        As an example I tend to think about teaching. Let’s say there is a threshold of 110 IQ points for someone to be smart enough to understand basic math well enough to teach it. There will be a great difference in teaching performance between a 115IQ teacher and a 95IQ that somehow managed to get through a math degree and got the job; but I don’t think there would be so great a difference between the 115IQ guy and his 135IQ colleague in terms of teaching effectiveness. And if we go to greater extremes, I don’t think the 165IQ math genius would be more effective a math teacher in elementary school than the 115IQ modal math teacher. His higher intelligence will certainly matter for advanced mathematics, but is not likely to be so beneficial to the teaching job.

        Or think about a soldier (or policeman, or firefighter): intelligence is required for efficient training and performance, and more IQ points are certainly beneficial… but quite likely, up to a certain point. The 85IQ young man is probably too difficult (maybe impossible) to train and too unreliable on the job, and I am quite sure the 115IQ guy will be better than the 100 IQ one; but would having an IQ of 130 or 145 make the person a sort of super-infantryman or super-firefighter, the way it seems to make a super-good economist? Not likely, I think.

        Or, for the matter, think even of some areas of science which are lean more to the side of practical, manual work: you need to be of above average intelligence to do Zoology field work, but can we assume that a 160 IQ zoologist will necessarily make a better sampling of fish in the Amazon than a 125 IQ one? I am sure the first will make a better story from the data gathered, but will he be better at obtaining the data?

        Or, as people have been discussing here, think of chess. Kasparov is smart but not stellar, yet he is one of the greatest players alive.

        In these examples, we have the situation where an intelligence threshold for efficient work can be seen; but after that threshould has been reached, the incremental effects of having more IQ ponits start to dwindle, and may cease at all in the very highest ranges. After the optimal IQ has been reached (somewhat above the threshold, I guess), an advantage in other abilities becomes more crucial for defining the very best professionals: a great soldier has to be smart, but also has to be physically fit; a great elementary teacher has to have a good IQ, but also has to project authority; and so on.

        Summarizing my point, when you say:

        If IQ matters enough that nearly all people in the field are above score X, this implies that additional IQ points are very important in the neighborhood of X. But then why wouldn’t IQ keep mattering above X? That sort of nonlinearity seems a priori unlikely.

        I think assuming linearity a priori would be just as unlikely for most fields. And this is not to say that your observations are wrong; I think they are quite right for theory-heavy, (non-theory)-light fields like economics or programming. An economist or programmer with stellar IQ will contribute a lot more to the field than a thousand of medium IQ types, and may sometimes singlehandedly revolutionize his area of study; the same does not happen to a teacher or a firefighter with stellar IQ.

        • albatross11 says:

          I vaguely recall (seeing commentaries by experts–I’m not remotely an expert in this area myself!) that you see continuing returns to IQ even for pretty basic tasks.

          • Eponymous says:

            Yup. Standard result, quoted in almost any book on IQ. I looked up a reference in one of mine below.

            I think Gottfredson (pdf) is a common reference here.

        • In the case of the teacher, the high IQ might be a disadvantage–if something is obvious to you it may be hard to see why it isn’t obvious to the student, or even realize that it isn’t obvious to him.

          • Tibor says:

            High IQ as well as high proficiency. It is fashionable to hire native speakers to teach people foreign languages but for complete beginners it might be a bad idea especially if the foreign language has grammatical elements very different from your native tongue. I find it fairly difficult to explain grammatical cases to native speakers of Romance languages or English since they simply come to me naturally. I can come up with some rules but really I have to look it up to state them precisely.

            Similarly, I remember an algebra exam in the last year of my bachelor where the first term was extremely difficult, nobody passed with the highest grade and only two out of about 40 people passed with the second highest (4 grades in total, 4 being failure). The majority failed the exam entirely. Then a week later there was a second term which was very easy – I narrowly failed the first one, then I only reviewed it once more before the second one and went through the last lecture which I missed in the first one (but the subject of that lecture did not end up in that second exam anyway) and passed with the best grade. The professor commented the results with “Why didn’t you learn it the first time already?” Since all of this were basics of general algebra and hence very easy for someone who had been doing that for 30 years, it did not occur to him that there was a significant difference in difficulty between the two exams.

        • Eponymous says:

          Actually, as I recall from Scott’s post on teachers, higher IQ does help teachers. Scott just quotes from this source, so I will too:

          The most robust finding in the research literature is the effect of teacher verbal and cognitive ability on student achievement. Every study that has included a valid measure of teacher verbal or cognitive ability has found that it accounts for more variance in student achievement than any other measured characteristic of teachers (e.g., Greenwald, Hedges, & Lane, 1996; Ferguson & Ladd, 1996; Kain & Singleton, 1996; Ehrenberg & Brewer, 1994).

          My understanding is that IQ scores are good predictors of job performance in many fields…

          Actually, forget my usual hand-wavey “my understanding is…”, I can easily just look up the reference. I searched my copy of Stuart Ritchie’s book on intelligence, and found a table adopted from this paper. Figure 1 shows the correlation between various measures of job performance and measures of cognitive ability. Performance on high complexity jobs correlate with IQ at 0.6, but even low complexity jobs correlate at around 0.4.

          So the data suggest that your intuition is incorrect.

          • That doesn’t tell us where the correlation comes from. It might be that IQ 110 does much better than IQ 100, but 120 does not do better than 110.

          • Eponymous says:

            Of course, a positive correlation does not imply linearity, or even a positive effect for all values of IQ. It does, however, disprove a strong threshold effect, as originally claimed.

            I don’t know why you have an intuition that the result should be highly nonlinear, however. If 10 IQ points are useful at 100, why should they suddenly cease to be useful at 110?

            Perhaps there are diminishing returns after some point, but since high IQs are rare and IQ in a particularly field is restricted by selection this doesn’t seem very relevant.

            Regardless, if you did not know of these results when you formed your opinion on the marginal value of IQ in most jobs, perhaps it would be wise to reconsider your intuitions in light of the evidence.

  32. thedirtyscreech says:

    So I may regret posting this and somehow being bombarded by emails, but a few years ago, I purchased two copies of the Hungarian version of this “Bring up Genius.” I cut one up with the intent to scan it electronically and did actually do a sample scan. The idea was to use Google Translate to see if it was feasible. I just found the sample scan and did the test translation. It’s below. Mostly, you can understand what’s being said, though it’s obviously far from a perfect translation.

    This book is basically a “Expanded Genius!” (1989) edition of my book’s revised and expanded edition.
    From home and abroad that I would like to re-issue my book (s). There was a small change in the title, a formal call
    Instead of now: You can become a genius … title. First I wanted this cinnamon adní for the first edition
    Too, but the driver did not accept it. I did not accidentally stick to the e156 release on the back of the next
    Quote from me: “I do not want to persuade anyone to raise genius. I just wanted to show you how to be genius
    to bring up. I do not call, I do not urge anyone, everyone has to decide for themselves what they want to term, but I give them
    My pedagogical system, and I guide you all along the way I’ve traveled. With the certainty that
    Educate genius, and it is worthwhile because it will be a happy person. ”
    László Polgár

    Foreword for the first run
    Here, generations were taught not to think. Do not doubt, do not argue. Biflázni. Terminate. It is in the upper school
    It could only come from everything. What a pedagogical brutality he was subjected to his physical ignorance in gymnastics
    Math above average child How tortured the young-mindedly interested young brains were uninteresting to them
    Terror of subjects ?! Supernatural education was only suppressed by the distracted genius, the one he had been disciplined. Whoever a
    “Probe,” it could have been a “subordinate.” Here, generations were taught to adapt to non-action, and this
    We are wandering with generations in the desert of quasi-reformism today.
    Our school is not the prosperous school, not the wings, but the school that dangers in the alley, where the palm is a super-
    Who has learned to fit in.
    If a small child here wants to get rid of the mediocre bargaining terror, the authorities are already crying out for the police. Arra a
    Law, it makes compulsory the acquisition of the minimum level of schooling. So that’s the twist
    The one who wants to overcome the average with which to eliminate illiteracy. With a submachine gunman, he arranges the
    Advice to your “disgruntled” dad. For the last 40 years, the children have forgotten the respect of fathers and mothers
    Biblical command, but fathers and mothers also forgot in the chase, in the flat, that the love of the parents’ house
    In their steadfastness, for the joy, happiness, humor, and family teaching of goodness, respect the future of their sons and daughters.
    They forgot to respect the unsatisfactory question of the child, never responding, patiently and with the surplus,
    Sixteen years old only 6k can build into the biological reality of susceptible plasticity.

    Scott, hit me up if this is of enough interest to you to have me scan and google translate the whole book. Or if you actually read Hungarian, just scan it.

    • thedirtyscreech says:

      Turns out I have the full Hungarian one scanned. Maybe I’ll try to translate the whole thing tonight.

    • OldMugwump says:

      It would be much more useful to get a clean scan of the original Hungarian.

      OCR and machine translation are constantly improving – with a clean scan the book can be re-translated in the future.

      And, of course, if there’s enough interest Hungarian-speakers may want to translate it manually.

      • thedirtyscreech says:

        Well, I’m not just going to randomly upload it for anyone due to fears of copyright infringement issues. But if you find some Hungarians willing to translate, I’m all ears.

        • gwern says:

          Don’t be ridiculous. A pseudonymous Internet person is not going to be sued for uploading to the Internet a copy of an obscure out of print Hungarian book written 30 years ago by a 71yo man. This isn’t 1997 and you’re not an American sharing the latest Metallica track.

          • Eponymous says:

            If he is nervous about it, perhaps he could be persuaded to email the scan to someone (like you)?

    • pontifex says:

      I guess it’s pretty awesome that Google Translate can show even that much… but all I can think of, reading that, is “who wrote this… Borat!?” Glorious learnings of superintelligence. Eliminate illiteracy with a submachine gunman.

      • thedirtyscreech says:

        Exactly. It’d be nice to get a native Hungarian to do the translation. I know there’s a free Esperanto version floating around. It’s possible (though seemingly unlikely) that Google could translate that better. I’d imagine that hiring a native Hungarian would be more useful (and cheaper) than someone who taught themselves Esperanto, though. I’m open to providing a PDF or intact hardcopy if we find someone and can pay them.

      • Deiseach says:

        Eliminate illiteracy with a submachine gunman.

        I think that is Laszlo Polgar’s disdain for the conventional educational system coming through clearly, if idiosyncratically 🙂 Education by forced schooling, and if the kids don’t go to school (they prefer to learn at their own pace and the kinds of things they’re interested in), they are called truants and the parents are punished by the force of the state. So they’re trying to overcome what they call ignorance by brute force, which is useless – trying to eliminate illiteracy by means of a gunman 🙂

  33. I’m so glad you found the book (and scans of the Esperanto version really are easy to find — there’s a reddit thread linking to it). But, more importantly,

    Come on, Esperanto-speakers! This is the only chance you’ll ever have to be useful!

    is a mad diss. It’s so great. I saw a comment saying not to drop dope slams on people you’re asking for help. I only mostly agree — honestly, I can’t think of any Esperantist who would mind words crafted with care. 🙂

  34. bodibei says:

    Just finished reading The Little Virtues by Natalia Ginzburg and she has some salient points on this subject.

    A vocation, an ardent and exclusive passion for something in which there is no prospect of money, the consciousness of being able to do something better than others, and being able to love this thing more than anything else-this is the only, the unique way in which a rich child can completely escape being conditioned by money, so that he is free of its claims; so that he feels neither the pride nor the shame of wealth when he is with others. A vocation is man’s one true wealth and salvation.

    The Polgar sisters seem to have this vocation. I would love to have Laszlo’s guidance as I try to help my daughters find and succeed at a vocation.

  35. ryanwc4 says:

    I do believe that there are genetic underpinnings to intelligence.

    But I find it odd that there is so little effort to engage with the null hypothesis, which I might label the status/stress-based replication of the middle class. This theory has interesting support from what I might call micro-population studies of intelligence – that stress has an extraordinary effect on intelligence; that stress is all but eliminated in the comfortable urban middle class; and, as relates to these particular studies, that Jewish Europeans, despite the natural and understandable tendency to portray themselves as Lincolnesque up from their bootstraps types, constituted a significant portion of the European middle class starting centuries ago. And that in particular, that the best histories of the stetls suggest that they held economic dominance over eastern Europe, not despite, but because of their legal status. (It’s an interesting theory that having a government supported monopoly over crafts and trade would constitute a disadvantage. Not sure how that became the default understanding of stetl history.)

    In the 19th century, stetls were hit by something like the same decline that struck American mid-sized towns in the last 50 years, as new transport and communications technologies like rail and telegraph destroyed economic relationships, but also sent large numbers of relatively prosperous Jews from the stetl towns into the cities of eastern and central Europe.

    The argument about Nobel prizes has always been interesting to me, again, for its failure to engage with the null hypothesis, since, if you actually look more closely, a large number of those credited as simply Jewish are the children of mixed marriages. While there might be reasons mixed marriages would produce genetically more intelligent children, such reasons aren’t in keeping with the description of Jewish intelligence that Scott advances. But the ‘replication of the middle class’ theory explains it all quite well. Low stress, greater positive reinforcement, limited amount of the static created by the negative reinforcement that poverty imposes through parental anger.

    Nothing about clusters of physicists in Budapest nor indeed, genetic relationships with intelligence, challenges this theory, since it would posit that lineage-based status would have a significant impact on intelligence.

    And there are enough problems with twin study based on the outplacement of twin children to separate families, along with the extremely small dataset of outplaced twins, that it’s hard to have much confidence in them.

    (In re-reading, I admit that my “the best stetl histories” is a big of a “no true Scotsman” argument.)

    • Steve Sailer says:

      European Jews tended to be quite affluent until their population explosion in the Fiddler on the Roof era left many unable to get the white collar jobs their ancestors had specialized in. This isn’t well understood today because contemporary Jews are encouraged to believe they were always oppressed and impoverished and everybody else is discouraged from knowing anything about Jewish history that might undermine the narrative.

      • bbeck310 says:

        This doesn’t sound right–citation needed?

        Western and Central European Jews tended to be relatively affluent because the Enlightenment came to those countries earliest, and Jews already had the customs and traditions to thrive in a more trade-oriented world. They also benefitted from increased religious tolerance in the Netherlands, Britain, and Germany compared to Spain (US religious freedom owes more to the Dutch influence in New Amsterdam than anything else). Eastern Europe/Russia never accepted enlightenment values, and kept Jews segregated, so Jews in the Russian Empire remained peasants.

        You can see this in US Jewish immigration; Jews who came to the US prior to the 1900-1920 Russian mass wave were largely either fairly wealthy Germans or Spanish exiles (also fairly well-off); they mainly moved uptown in New York or out of the city and were embarassed by their poor, uncultured peasant co-religionists.

        There were no “white-collar” jobs for those Jewish ancestors to have; until the Enlightenment, the vast majority of jobs anywhere were “peasant farmer.”

        • biblicalsausage says:

          Try reading the first chapter or two of Bartal’s _The Jews of Eastern Europe_, or look up the paper “A Natural History of Ashkenazi Intelligence.” Or read Katz, _Tradition and Crisis_. Probably just picking up any book that extensively discusses the history of Ashkenazi Jews should do the trick for you.

          Most Ashkenazim weren’t peasant farmers before the Enlightenment. The vast majority of jobs anywhere were indeed “peasant farmer,” but Ashkenazim were always a minority of up to a few percent of the surrounding population, and there were enough white-collar jobs to go around for them.

        • biblicalsausage says:

          In Poland-Lithuania, for example, the Jews were kept residentially segregated AND they composed a wealthier-than-peasants white-collar class. Both things are simultaneously possible.

        • Steve Sailer says:

          “Western and Central European Jews tended to be relatively affluent because the Enlightenment came to those countries earliest”

          No, Jews did well for themselves in, say, Poland, Lithuania, Ukraine, etc. relative to the average indigenous person. They were invited in to provide Eastern European nobles with a literate/numerate middle class to do rent collecting, tax farming, etc., especially after the Mongols massacred Eastern European cities.

          The Russians asked in Germans to play those roles. The Brits grew their own middle class. The last seems like probably the best strategy if you can pull it off.

          In general, Americans really don’t have an accurate picture in their heads of the social situation of European Jews. Most Americans are familiar only with “Fiddler on the Roof” set around 1900, when Jews were at their relative poorest due to massive population growth. (The Coen Brothers’ enigmatic movie “A Serious Man” is perhaps intended to provide a less self-pitying look at Jews.)

          • Steve Sailer says:

            One interesting question I haven’t seen explored is whether the Jewish population of the Netherlands helped culturally fertilize the indigenous middle class with advanced business techniques and role models for an urban literate culture. A big problem in Eastern Europe was that the big gap in language and culture between the Jewish middle class and the Christian peasantry meant that that the indigenous people didn’t pick up many bourgeois skills from the Jews. In the Netherlands, however, there may have been more cultural transfer. (And there was a subsequent cultural transfer of advanced business techniques from the Netherlands to England, especially after 1688.)

            But that’s speculative. And whether the Sephardics were more open than the Ashkenazi or whether Dutch Protestants were more receptive than Polish Catholics to Jewish bourgeois culture is even more speculative.

          • Aapje says:

            The separation between the Jews and non-Jews was not as strict as the regulations for Jews make appear. No city had a Jewish ghetto. Jews and Christians met as neighbors and increasingly in trade, industry, science and culture. Sometimes some Christians would express anti-semitism, but more often, wonder at and interest in Jewish culture.

            Translated from here.

            Here someone argues that because the Jews weren’t given access to the guilds, they didn’t have a huge influence. However, the author notes that many Jewish traders fled to the harbors and trade cities in Italy, Morocco and the Ottoman Empire and that the Dutch Jews could easily trade with them. It is also noted that 7% of the stock of the Dutch East India Company was in the hands of Jews at the start, rising up to 25% in the 18th century.

            The much more explicit Jewish influence is only visible later, for example, Isaac de Pinto became CEO of the Dutch East India Company 150 years after it was founded. Sephardic Joseph de la Vega wrote the first known book about the stock market based on the Dutch East India Company stock trade, 80 years after it was founded.

            I don’t see how we can determine from the available information how much impact the Jews had on Dutch bourgeois culture. More than nothing and less than a full replacement, clearly.

            It does seem very sensible to me to assume that Dutch Protestant culture at the time was more receptive, for the simple reason that they didn’t try to convert/kill the Jews, in contrast to most Catholic nations of the time.

          • ryanwc4 says:

            There were tax farmers and estate managers, but more often, they were in trade.

            In England, Holland and much of Germany, the key distinction is quite simply the Reformation, not because of any ‘work ethic,’ but simply because Christians were encouraged to learn to read.

        • ryanwc4 says:

          “customs and traditions”

          Or was it just a network effect? I can imagine it would have been tough to penetrate any eastern European cattle market in 1790 if you didn’t speak Yiddish.

          And that network effect was enforced by the ruling monarchs and princes, who established markets by patent and by monopoly rather than just letting them grow wherever they might pop up, under the auspices of whoever decided to market some goods.

          I have no idea what the above poster is referring to in the early 1900s with the term “Spanish exiles.” Exiles from the expulsions of the late 15th century 400-some years before.

    • ryanwc4 says:

      Oy. All those responses on stetl history, which are interesting, and seem to underscore my point, but still no one willing to grapple with the status/stress theory of reduced intelligence as a theory with significant evidence, and one that can not be falsified by population studies.

      It’s not like this is out in left field. I mean, does anyone think the first crane chick always lives and the second crane chick almost always dies because of genetic difference? There are all sorts of examples where status rather than genetics results in reduced fitness. And we know this to be true of intelligence, that low status, and the concomitant stress, has a direct effect on intelligence. For a bunch of smart people to avoid this question seems awfully self-serving – “nah, it can’t be that I’m smart just because my parents were well off enough to have few worries.”

      • The Nybbler says:

        but still no one willing to grapple with the status/stress theory of reduced intelligence as a theory with significant evidence

        What evidence? We know that intelligence is reduced through some gross stressors, no one is disputing that.

        For a bunch of smart people to avoid this question seems awfully self-serving – “nah, it can’t be that I’m smart just because my parents were well off enough to have few worries.”

        My father and all his siblings are pretty smart, and their parents were working-class with a whole bunch of kids. And there’s no end of well-off dumb kids.

      • albatross11 says:

        How might we test this idea? Maybe look to see if cortisol levels as a kid (you can get them from saliva samples) negatively correlate with IQ? Or look for twin studies where one twin went through some family trauma (divorce, death or serious illness in the family) and see whether that affects their adult IQs?

  36. Haukur says:

    I assume that the success of the Polgár sisters comes down to the following:

    a) It’s common enough that all members of an Ashkenazi family have mental abilities well above the average. Many readers of this blog can no doubt testify to that.
    b) Parents consciously trying to raise prodigies does work reasonably often for smart parents with smart kids.
    c) Women tend not to be interested in chess so there was a big opening there. This would all be much less memorable, and perhaps less successful, if it had been three boys. Presumably the positive reinforcement the sisters got from being at the very top of women’s chess had a role to play in maintaining their enthusiasm and confidence. Boys of equivalent initial ability might not have persevered to reach the same heights.

    I doubt there is much left to explain for any secret sauce in that book.

  37. MRockz says:

    I would pay for a digital translation to English of Polgars book. Maybe the costs could be deferred with a crowdfunding campaign? $US20 for a copy? Or get the publishers to reprint as a service to humanity? It’s a pretty well known story, amazing it isn’t available.

    There’s also a documentary from Ruth Diskin films I’ve been waiting to see but they’ve had problems with some of the international digital rights apparently.

    • screwtape says:

      I would pay 20 USD for an English copy. I don’t know what it would take to make that happen, or how many other people would also be willing to chip in to pay a translator.

      Mostly right now I’m wishing I’d picked Esperanto instead of Lojban during my high-school language kick. So close!

  38. dutchie says:

    I’m a (fairly new) Esperantist, and translating the book seems like an interesting challenge with an actually useful outcome. It’s really not hard at all to find a scan of the Esperanto version online and I’ve got a copy. If anyone else is interested in helping to translate, and would like to co-ordinate to avoid duplicated effort, please contact me (I’m dutchie on the #lesswrong and #slatestarcodex IRC channels, the LW Slack and SSC Discord)!

    • dutchie says:

      I have started on this and made a Github repository. Suggestions and feedback are welcome. We’ll see whether this gets DMCA’ed to oblivion. Putting the dodgy scanned PDF in the repo seems like tempting fate rather too much for my taste.

  39. TomA says:

    I think the Polgar story is anecdotal evidence of the power of memetics, e.g. that evolution has predisposed our offspring to be (in essence) programmable during their early developmental years when brain wiring can be proactively modified through repetition and simultaneous activation of multiple brain centers. In other words, perhaps Polgar created several organic versions of Deep Blue.

    • sonnymoonie says:

      Deep Blue was more or less an ordinary chess engine algorithm, with more hardware thrown at it and a team of expert chess engine tuners. A better comparison would be with Alpha Go, which applies some pattern recognition “neural net” techniques from AI research to playing competitive Go.

      The way of thinking that implies evolution has a fairly free choice as to what abilities to predispose an organism to have is highly questionable. Evolution is mainly a process of tending to preserve reproductively relevant abilities that happen to come up, maybe extending some lucky accidents in some species by that process.

      So humans happen to have some learning abilities, that have creative applications such as language and music, far beyond the practical uses for learning abilities that were preserved and developed in earlier animals. (That’s not to be confused with being programmable. Although that’s something humans can emulate to an extent, computers do it better, because humans made them for that.) The feature of learning in humans, that it has windows of being easier or better done at certain ages, for certain skills, seems to be a side effect of how human learning is possible at all, probably not a trait that has a positive value selected by evolution.

      For me, the Polgars are more evidence that chess is a sport that takes training, memory, dedication, and a love of the game. It’s an illusion that chess is highly intellectual rather than merely an abstracted sport that a more intellectual person would have a natural advantage at, all else being equal. That intellectual mystique of chess, and how people were able to invent such a game, is the curious question about the meme of chess.

      • TomA says:

        You are a bit behind the curve here, but I will try to be brief.

        H. sapiens evolved complex language skill about 200,000 years ago (uniquely among all other species), and this enabled our ancient ancestors to quickly pass “wisdom” from generation-to-generation via memetically imprinting sophisticated knowledge and new behaviors in their young during the formative brain development years. This transfer of useful knowledge and skills conferred significant advantage to their survival prospects and hence increased fecundity. This is fundamental Darwinian selection, but in the memetic channel as opposed to the DNA-based genetic channel.

        It’s still a chicken and egg debate about which factor was the precursor, brain encephalization/bicameral partition or the advent of complex language skill.

  40. onyomi says:

    One other confounding factor besides the children being genetically smart (would have proved much more, of course, if the parents had tried out their method on adopted children): they picked the thing to focus on as a result of their children’s own manifested interest. It would have been more impressive, nurture-wise, if the parents had, without being abusive, trained their children to be world-class in something they weren’t naturally that interested in.

    So, the less impressive (but still actually really impressive!), more accurate title might be something like: “if you nurture and support their inborn passions and your children are genetically pretty smart, you can raise children who turn out to be top-notch at the thing they were genetically predisposed to be interested in.”

    This does have implications for parenting, however: specifically, seems like homeschooling, maybe even “unschooling” is probably a better idea than most think. I think one of the fears of this kind of approach is that it will produce “unbalanced” adults–like if you let your kids play Chess at home all day they’ll grow up to be really good at Chess but unable to read or recognize China on a map or know who won the Civil War or something. Oh wait.

    My impression is it turns out that intelligent people just kind of osmose the basic knowledge required to function in their culture and that stupid people are also not going to retain basic information about history, economics, geography, etc. no matter how much you drill them in a classroom setting. Therefore, whether or not your children grow up to have a “balanced” knowledge base is not as much under your control as you think, so you might as well let them follow their passions (that said, we may again be biasing it by assuming unschooling of smart children who naturally gravitate toward some semi-usable pursuit like Chess; it may very well be that most children, left to their own devices, will just play video games all day and not even become competitive video game players).

    • Broseph says:

      Bobby Fischer was the real life example of that unschooling caricature. He became obsessed with chess to the exclusion of all else at the age of seven. He grew up to become arguably one of the greatest chess players of all time. But the guy dropped out of high school, apparently was pretty ignorant of most topics besides chess according to friends, and ultimately died a destitute anti-Semite (despite both his parents being Jews). This is despite getting accepted on a scholarship at a school for the gifted on the basis of his astronomical IQ score (180 supposedly).

      • Eponymous says:

        Bobby also had significant psychiatric issues that nearly torpedoed his successful world championship bid, and destroyed his career thereafter. In the absence of these, he would have been a very financially successful player. This is assuming that removing his psychiatric issues didn’t also remove what made him a great chess player in the first place.

    • Deiseach says:

      Re: your link, I’m sorry to say that my immediate reaction to the opening “Hello, I’m Courtney Plunk” was “I’m sorry to hear that, I hope it gets better” 🙂

  41. P. George Stewart says:

    It’s probably true that we in the West have lost sight of the virtue of sheer repetition as a valuable tool for learning. The famous Chinese term “kung fu” means “skill achieved through repetition”, and applies to any skill, not just martial arts – IOW, there’s a cultural respect for people who put the time in to master something via repetition. Also, wrt Asian martial arts, with the traditional village/clan systems, they start them really young on basic exercises (particularly because the kind of physical conditioning involved in a lot of Asian MA relates to the fascia).

    We’ve tended to think it’s a bit cruel to make kids learn things that way, but the “Tiger Mom” thing showed that from the point of view of Chinese culture, in a tough world it’s actually a sign of parental love and concern, because you’re ensuring your kid’s going to have a marketable skill to a high level. (I suppose whether it’s beneficial or not psychologically has a lot to do with whether it’s done out of pure concern, or mixed with the parent’s own hangups.)

    Anyway, perhaps we should re-inject a bit of that into our modern thinking.

    • Kevin C. says:

      It’s probably true that we in the West have lost sight of the virtue of sheer repetition as a valuable tool for learning.

      I must say that this at least matches my experience as a math tutor. While I’ll admit that at a certain level, “worksheets” can become grinding busy-work, but it seems to me that our schools are erring too far in the opposite direction. All clever and varied pedagogical techniques aside, at least some portion of fluency and proficiency in mathematics comes via sheer repitition. We accept the importance of “practice, practice, practice” in athletics, or in learning a musical instrument. But when it comes to mental skills, we seem to discount the importance of drill and repetition. When it comes to making arithmetic automatic, there’s little that can substitute for just sitting down and adding, subtracting, multiplying, and dividing until it starts “sticking”.

      • Steve Sailer says:

        We accept the importance of “practice, practice, practice” in athletics, or in learning a musical instrument.

        And we’re pretty good at athletics and violin these days.

      • pharmst says:

        I’m completely blanking on an essay I read in the last few months talking about exactly this: IIRC it was written by a woman who came out of High School thinking that she wasn’t great at maths but was a pretty good linguist, but later on she started applying the same tricks she had used to learn languages (repetitive drills to ensure that core skills became completely automatic, lots of “playing around” with ideas to see the same thing in different contexts etc etc) to learning mathematics she ended up at the point that mathematics is now her vocation & professional career.

        If it comes to me, I’ll reply to this comment 🙂

      • gbdub says:

        Does anyone know of some decent texts on teaching math at an early age? I’ve long had a feeling it would be valuable to expose kids to more advanced concepts earlier… I notice a lot of students getting hung up when you start introducing algebra and trig. Ideas of negative and imaginary numbers, variables, and representing curves with equations.

        Fundamentally, these concepts don’t seem much weirder than arithmetic itself (or representing sounds with letters), but for someone who’s spent years working arithmetic, throwing in these new abstractions is a major paradigm shift.

        Would it go better if these abstractions were introduced at the younger ages where the brain is seemingly more amenable to thinking in novel abstractions? If a youngster can learn language more easily than an adult, why not calculus?

        • sconn says:

          I successfully taught basic algebra to some second graders once. They caught on quickly.

          • albatross11 says:

            I’ve introduced my kids to a lot of “advanced” concepts from math, economics, and statistics that most people don’t learn till college. This works fine, as long as understanding the concept doesn’t rest on several years of previously-acquired knowledge.

            For example, a bright ten year old can get regression to the mean, the prisoners’ dilemma and what it looks like iterated, the cut-and-choose protocol for dividing things, infinite series with finite sums and Zeno’s paradox, etc.

            The infinite series thing works well with a cookie. First we’ll eat half the cookie. Then we’ll eat half of what’s left. And half of what’s left. And so on, forever. It’s really easy to see that you’ll never eat more than one cookie (so the series sums to 1), and that you’ll get ever closer to eating the full cookie without ever quite reaching it.

        • US says:

          I’m not aware of any texts, but in terms of the idea of exposing kids to more advanced concepts earlier one resource that has to be mentioned is Khan Academy. Any smart kid who understands English is free to explore more advanced concepts there, at his or her own pace. Adults can track the progress and make suggestions as to which topics might be fun to have a look at. I haven’t used the resource for a while, but it seems close to ideal and if books are written about how to expose kids to advanced topics in mathematics, I would expect at least a few of the people working for Khan Academy to have read them already and tried to implement the ideas when developing their online content.

        • Sanchez says:

          You might be interested in material from Art of Problem Solving.

      • CatCube says:

        Yeah, I remember some years ago being annoyed when working with somebody else who had to punch 6*8 into a calculator.

        There is no substitute for basic memorization in arithmetic. Maybe you can get the kids to memorize 6*8=48 by having them play with blocks many times until they memorize it on their own, but they’re going to have to just remember that fact and be able to recall it automatically. I doubt there’s a faster or more effective way than simple rote memorization of the multiplication tables.

        I actually prefer to memorize basic concepts and then rederive things I need from them at the point of trying to solve the problem, but that still requires memorization.

        • Bugmaster says:

          I think it can be very instructive to demonstrate multiplication from the ground up. At first, we can just do repeated addition, but that gets too boring (and too slow) very quickly. So, we memorize the products of all the single digits, just to save time. Could we memorize more products ? We could, but it won’t have as great of a payoff for the next stage: the standard iterative multiplication algorithm.

          Other algorithms exist, however; for example the “Russian Peasant” method only requires you to multiply and divide by 2. However, it is generally less efficient, which presents a great opportunity to talk about speed/memory tradeoffs.

          From what I’ve seen, American children are taught the standard algorithm as though it were holy gospel, with no history or reasoning behind it, delivered to us humans as a singular revelation that must be blindly obeyed. I understand that this method of teaching is less time consuming, but it still feels like a crime…

        • spork says:

          A fifth-grader showed me a simple connect-four multiplication game recently and went on to beat me at it. Because I’ll be wanting revenge next time I meet him, I went looking for an online version of it so I could practice my strategy. I found this: the AI is too dumb to practice against, but at least it gives you an idea of how the game works. I haven’t seen a better tool for kids to practice the times table. If someone is aware of a version of this game that has a good AI, please reply.

    • Eponymous says:

      I’ve always though that an important aspect of intelligence is how well your brain responds to training, not just how “naturally” good it is in its untrained state.

      Western culture probably places too much emphasis on “natural” raw untrained genius, perhaps as a residue of the Romantic movement. But people whose brains respond very well to training also possess a very useful sort of “natural” ability — namely the ability to learn!

      There’s probably a high correlation between a brain’s untrained power and its ability to learn, but I suspect it’s well short of 1.

      • P. George Stewart says:

        Yeah, very much agree with this, particularly the Romantic thing.

        The hangover from Romanticism is also I think one of the primary causes of a) the idea of the lone genius on the misty mountaintop thinking the world out from his brow (or whatever the saying is) and b) the idea that you have to be tewwibly owiginal in whatever you do – as opposed to the idea of mastering an established cultural groove.

        (Shades here of Jordan Peterson on Openness trait being more associated with the Left, and Conscientiousness more with the Right – we need both, we need people to work with and cherish algorithms that are actually functional, that build and maintain our artificial ecosystem, just as much as we need people to jump outside systems and think outside sundry boxes. There’s been far too much cultural skew towards Openness for too long – as well as Openness’ dark side, destructive critique.)

  42. Ivy says:

    I recently stumbled on a similar book, which is thankfully much easier to find online and in libraries.

    It concerns a different example of world-class achievement as result of a carefully optimized education: that of the utilitarian philosopher John Stuart Mill.

    Mill’s Autobiography starts with an in-depth description of an incredible home-schooling by his father. Mill had read all the Great Books by age 14, and was made to explain their key concepts, summarize arguments, write critical essays and defend them during intensive 1-1 tutoring sessions.

    Mill attributed his success entirely to this homeschooling, and he thought his father’s system could and should be adopted more widely:

    it may be useful that there should be some record of an education which was unusual and remarkable, and which, whatever else it may have done, has proved how much more than is commonly supposed may be taught, and well taught, in those early years which, in the common modes of what is called instruction, are little better than wasted.

    Another piece of anecdotal evidence for “extreme education works” is Renaissance philosopher Michel de Montaigne being artificially confined in a Latin-speaking environment to make Latin his native language.

    • Scott Alexander says:

      I also read Mill’s autobiography; it’s great. If I’m going to look into child prodigies I should probably reread it since I remember he had a lot to say about that, including a pretty good account of a mental breakdown.

      • Steve Sailer says:

        I read J.S. Mill’s autobiography in the 1970s. I took it as a warning about the dangers of Tiger Dadding.

        The most memorable part for me is when poor J.S. finally starts finding relief from his depression in music. But then it occurs to him that with all the great melodies written in the current decade (the 1820s), surely the world will run out of new melodies shortly and he lapses back into depression.

  43. zidien says:

    Ericsson, who is to his chagrin responsible for the ten thousand hours thing, would tell you that it’s deliberate practice that counts. If you don’t meet the conditions for actually learning something, your time is wasted.

    Which, well, no shit I guess, but it doesn’t make things any less complicated.

    Also, c’mon, there’s no way Erdos’ parents were leaving math books around the house by accident, right?

    • Steve Sailer says:

      The guy who took up golf at age 30 with the intention of putting in 10,000 hours of deliberate practice (i.e., no playing rounds of golf for the first couple of years, just working on the practice tee and practice green) and making the pro tour actually did progress pretty rapidly and was quickly shooting in the upper 70s. My impression is that Erricson’s methods worked pretty well for him for awhile.

      Unfortunately, he just hit a ceiling (or floor) at around 75, while pros need to shoot around 65 regularly on a run of the mill golf course.

  44. Glen Raphael says:

    This post reminded me that in Palo Alto I was taught high school physics from an “educator with magic powers”. His name was Art Farmer and the chief evidence of his magic powers was that on some physics test (the California AAPT, I think it was called at the time?) his students utterly dominated the list of top scores, indicating either that he was doing something right or every other school in California was doing something wrong.

    Farmer didn’t become famous, but based on classroom experiments at a couple of schools he published some papers and won some awards and his methods influenced the next round or two of educational reform. Some of his ideas were incorporated into college classes and into the College Board’s recommendations for how high school AP classes should be taught.

    The most relevant paper might be:
    “A new approach to physics teaching” Arthur V. Farmer
    The Physics Teacher 23, 338 (1985)

    I can’t find an un-gated full text but that first page hits the high point, which was that Farmer thought kids might learn physics better if you expose them to ALL of physics first – at least at a vague no-math conceptual level – and only THEN go into greater depth reasoning about examples that can then draw from ALL of physics to show how concepts are connected. Which was not the fashion at the time. Also: something something “multiple representations”.

    • Steve Sailer says:

      My impression is that there are more than a few teachers who are extremely good at what they do, but their theories of how to teach tend to work best when they themselves are teaching, or when one of their students is carrying on the tradition. It’s hard to mass produce great teachers just from having them read about the original great teacher’s pedagogical theory. You kind of had to be there.

      It’s a little bit like studying stand up comedy from reading a textbook about Rodney Dangerfield: “Yank at your necktie a lot.”

  45. b_jonas says:

    You’ve found the Esperanto version, and the comments have already pointed out that an English translation and a Polish translation probably exists. I’d just like to add that there is also a Croatian translation from 1990 under ISBN 86-323-0295-7.

    Note also that the Hungarian has (at least) two different editions, a first edition from probably 1989 titled “Nevelj zsenit!” with ISBN 963-01-9976-9, and revised edition from 2008 titled “Nevelhetsz zsenit” with ISBN 978-963-09-5822-6. I have not read any of these, so I cannot tell if the revised version is better. The Esperanto version 963-86531-0-8 is probably a translation of the first edition, because it apparently predates the first edition and has a translator listed.

  46. enye-word says:

    Here is a link to the Esperanto version, Eduku geniulon!: https://cloud.mail.ru/public/GC63/3QtBTava8
    Here is another link, to that same pdf uploaded to Google Drive: https://drive.google.com/file/d/0B7gGmCVv3euANDVWMklFRGhNYlk/view?usp=sharing
    Here is the magnet link to the same pdf: magnet:?xt=urn:btih:b07670740fe0ab01c22ef909a6f94987087f9500&dn=Eduku%20Geniulon!.pdf

  47. US says:

    “The study linked above suggests that Susan practiced 48 hours a week. During those six months, she would have accumulated about 1200 hours of practice. Suppose the elderly Budapest chess players practiced only one hour a week, but had been doing so for the last twenty-five years. They would have more practice than Susan – plus the advantage of having older, more developed brains. So why did she beat them so easily?”

    a) You have a faulty model of how chess improvement works if you assume that an hour of ‘practice’ necessarily leads to any improvement in chess skill. Already at a not very high level (below-average club player level, probably) you need to put in time regularly just to not lose skill/rating, because if you don’t practice you become ‘rusty’ and e.g. miss opportunities you might not have missed if you’d put in a few hours into chess-related stuff recently. Most players improve for a while, then stagnate at some level. If they leave chess and come back after a while (years), they’ll in my impression usually have to put in some hours to get back to where they were when they left the game; if they start out with the rating they had back then, such people are in my experience (I’m a former tournament organizer) more likely to lose rating in future tournaments than to win rating. Most of that ‘one hour per week’ – assuming they played that much – is best thought of as ‘maintenance’, not ‘skill improvement’.

    b) Relatedly, chess-related activities will vary a lot in terms of how much they are likely to improve your skill level. You might improve a lot more from studying chess (openings, endgames, middle games, tactics, etc.) than from playing random games against much weaker opponents (a classic situation in small chess clubs, especially in the pre-internet era) or from playing 1-minute games online. Even some players who have played a lot of games are quite poor at chess – a few of the players in my local club have played for decades and probably never once reached e.g. 1600. It’s usually more fun to play the game than it is to systematically focus on your weaknesses and to try to improve your chess. It’s also more fun to focus on the things you’re good at than the things you’re not good at, so even people who aim at improvement may focus on the wrong variables.

    c) I already sort of said this, but here’s another way to think about it: The steepness of the learning curve is highly variable in the chess context. Some people (‘talented, smart’) will improve really rapidly. Others …will not. It’s important to keep in mind in that context, and in related contexts, that ‘talent’ and motivation are linked – a big problem with figuring out how important ‘practice’ is in the chess context, compared to ‘talent’, is in my opinion that a lot of the ‘untalented’ stop playing because they get disillusioned by their lack of progress. This is especially true for children. If you try to figure out how important practice is by correlating ‘chess skill’ with ‘hours put in’ but you focus on players who reached a certain threshold/level of skill, you’re neglecting a potentially really important attrition problem.

    d) Her beating veterans is a good story. It’s a less good story that a young girl loses to strong players who’ve played for a long time. If the club had a few decent players, I’d say she probably got clobbered by them, certainly in the beginning, but that part of the story was not told. There’s actually a Magnus Carlsen app that tries to ‘play like him’ at various ages, and any decent player will totally destroy ‘him’ if you’re playing, say, a 7-year old Carlsen. One of the first people I played in my local club was some semi-regular club player in his 40es or 50es; he was a terrible player, around a 1000 rating and an IQ probably south of 85, but he liked the game and the social aspects of the chess club. A 4- or 5-year old beating anybody is a feat (even knowing and understanding the rules at that age is arguably impressive), but I would not make too much of it.

    • bbeck310 says:

      On your 4th point: A good example is current US champion and former child prodigy Hikaru Nakamura. You can see a graph of his USCF rating over time here. While he rapidly improved, he started at the bottom like anyone, and took 4 years to reach national master (2200) (and became the youngest at that time to reach NM).

      More generally, I can go to any area with a decent sized scholastic program and find kids who regularly beat adults. Around a 1200-1400 rating, most kids switch to playing primarily adult tournaments because there’s nothing to gain from the scholastics (unless they’re on a school team). NYChessKids is one of the few tournament organizers I’ve seen who regularly gets strong scholastic sections at non-championship tournaments…which they generally do by covertly giving money prizes for the open section 🙂

    • Kaj Sotala says:

      Her beating veterans is a good story. It’s a less good story that a young girl loses to strong players who’ve played for a long time.

      There’s also the possibility of “adult players see a four-year old girl come into their club and be clearly excited about the game and none of them wants to discourage her, so they all go easy on her”.

  48. harvey-ta says:

    I used to think that Lazlo Polgar’s blank-slatish theory was untrue, until I found out about Black Swedish Grandmaster Pontus Carlsson.

    https://en.wikipedia.org/wiki/Pontus_Carlsson

    Carlsson was adopted, at the age of one, by Ingvar Carlsson – a one time chairman of the Swedish Chess Federation. You might want to persuade Ingvar to write a book if you believe the Polgar achievements are purely down to Ashkenazi smarts.

  49. Machine Interface says:

    “A lot of people claim there’s a “developmental window” during which children have a unique ability to learn language.”

    There’s a simple counter-argument to that claim though: small children spend essentially all their waking hours learning their first language, and literally every human interaction they have during that time is an occasion to practice.

    Adults are lucky if they can dedicate more than 15 minutes a day to learning a second language, and unless they live in bilingual community or in a foreign country, occasions to practice are rare (unless the second language being learnt is English, in which case the “occasions to practice” are called “the Internet”).

    In those conditions it’s not that surprising that children learn language “this quickly” compared to adults (which needs to be nuanced anyway — if I remember correctly, a normally-abled child can well have all the phonology and grammar down by age 3, but still have trouble with metaphors by age 8).

    • publiusvarinius says:

      I think that scientific claim is commonly misinterpreted. I don’t think it’s supposed to apply to any particular language. It’s more like “if you did not learn to speak by age 6, chances are you never will”.

    • sconn says:

      The military trains its translators in an immersion program. After six months, they’re qualified to translate anything. That suggests the important part isn’t the age, but the immersion.

      With one exception: accent. An infant can recognize and distinguish any sound made by any language, but before a year old we stop being able to distinguish sounds we aren’t hearing – for instance, an infant surrounded by only Spanish speakers will lose the ability to tell the difference between B and V. You can regain this ability to some extent, but not entirely, which is why non-native speakers will almost always have an accent.

      • leoboiko says:

        Some misinformation going on. As a linguist who supports the consensus position that Scott is talking about, let me post a summary:

        * Age-related effects—the younger you are, the better you acquire languages—have been shown quite extensively and by decades of replicated experiments.
        * Age-related effects are independent of amount of immersion, length of practice, hours per day, methods, presence of a first language etc. etc. All else held equal, younger=better.
        * There’s nothing magical about phonology. Adults can learn to fix their accents; actors do this all the time. Sure, adults often fail to acquire phonology properly, while kids are guaranteed to acquire it given input; but that’s also the case for morphology, syntax, lexicon, pragmatics and everything else. Syntax is particularly insidious, and certain subtle oddities (like detailed use of determiners and complementizers in English) may never go away for adult foreign learners.
        * Early experiments suggested a ‘critical period’, after which there would be a sharp decline in age-related benefits. More recent experiments have put that into question, and age effects now appear to taper off gradually through adolescence. Age effects are nonetheless real.
        * The most evident age effect are success rates. Adults fail to learn languages all the time; success is highly variable and depends on personality, commitment, resources, time, culture, similarity of target language to known languages, arguably complexity of target language, possibly IQ, etc. Child success is damn near universal. Barring severe mental disability, expose any young kid to any few languages and they’ll pick them up, unconsciously, without explicit teaching, regardless of academic ability, with 100% success, and speak as well as a native. No infant fails to acquire the languages around them, merely through exposure. Adults can learn by immersion too, but success is not nearly guaranteed, and they often need explicit teaching anyway (in phonology, for example, they usually won’t notice a non-native phoneme unless explicitly taught about it, and even after decades of exposure won’t add the phoneme to their repertory; the same go for fossilized constructions, etc. This is in stark contrast to how young children learn.)
        * These age effects have a biological, not social or environmental basis.

        • Anatoly says:

          Thanks, that’s very useful. Follow-up questions:

          1. The meta-analysis you link to seems to find that there’s no age effect in foreign-level acquisition (defined as mostly-classroom, no immersion) vs a considerable age effect in second-level acquisition (defined as involving immersion). Which seems to contradict your “independent of amount of immersion” above. Am I misreading the study, misreading you, or you meant to assert a stronger opinion than the analysis?

          2. Any links to research that backs up the “there’s nothing magical about phonology” part? This seems particularly counter-intuitive to me, though let me qualify that. I understand that morphology, syntax etc. are also usually not acquired perfectly by adult learners, and there’s no arguing that some parts of these are horrifically difficult to impossible to acquire perfectly by adult learners (I think aspect in Slavic languages is notorious here). But anecdotally there seems to be an age cutoff of 7-9 years after which in young learners native-like accent is extremely rare while native-like morphology and syntax are routine. As someone who’s a part of a large immigration wave, I’m very used to encountering people in my community with age of acquisition say 9-15 years old who speak the L2 very fluently and without visible morphology/syntax deficiencies, but still with a noticeable L1-influenced accent.

  50. donniel says:

    Scott, is the title a play on “Mastering the core teachings of the Buddha”, and if so, curious if you’ve read it/how you’ve heard about it?

  51. 6b64 says:

    There is a documentary about them called The Polgar Variant:
    http://www.imdb.com/title/tt4373776/

    • 6b64 says:

      I translated a Hungarian review of the documentary:

      „This movie is about chess, but this is not a chess-movie. An astonishing story of a family whose life revolved around chess, but at the same time brings up universal existential problems: child-rearing, differences between the sexes, the power of love, and the contradiction between a livable existence and constant achievement.”
      – before watching The Polgar Variant I found the words of Yossi Aviram a little exaggerated. But the truth is, even though the movie isn’t perfect, it’s extremely multifaceted. It is difficult even just to list the questions brought up by the story of the Hungarian girls raised to be chess geniuses.
      Yossi Aviram wasn’t a child in Hungary during the Eighties, so unlike me, didn’t grow up together with the story of the Olympic and world champion Polgár girls. He didn’t see them almost constantly in the TV and the papers, didn’t hear the arguments between adults whether it was right or wrong what Polgár did with his family. He [Yossi Aviram] stumbled on the pictures of the girls in an Israeli chess-club, and it occurred to him that they were the only women among the most famous chess champions. He started researching the family, and made the movie with Israeli and Hungarian funding.
      This is the best time for it: every family member, coach, and game partner is still alive and willingly talking. But enough time has passed to have a little perspective on the events, and we can better appreciate the successes and the failures. Even the youngest Polgár girl (we shouldn’t even call them girls anymore) Judit is forty years old, and recently announcend her retirement.
      The director dives with great thoroughness into this exceptionally well documented story. The media followed the Polgár girls since their childhood, so there is a huge amount of interview and film material, and Polgár László himself started taping his family fairly early. This way The Polgar Variant didn’t become the typical talking head documentary, because it can show almost everything it talks about. There are archive recordings of the little children practicing in their small public-housing flat, of games played against grown men, but even that is documented when the middle girl, Zsófi met her eventual husband. No wonder it was difficult to keep all that material in order, and the movie is scattered and digressive at places. On the other hand its is commendable that he didn’t want to pass any of the emerging questions.
      The movie is roughly chronological, it starts off from the Seventies Budapest, when the pedagogue-trained Polgár László figured out before the birth of his children that he was going to raise geniuses. According to his theory every healthy human child can be a genius, it has nothing to do with genes, only the upbringing matters. He chose chess party because it was way more popular at the time then now, and partly because during socialism great importance was attributed to chess successes. The girls didn’t go to kindergarten or school, they studied and practiced at home according to a rigorous schedule, through long hours. Zsuzsa, Zsófi, and Judit now say that it wasn’t a burden on them, but the coaches who worked with them remember a little differently.
      This much we knew. But until now I didn’t hear about what kind of family trauma lay beneath the determination of Polgár László. In the movie he tells it himself that his father lost his first wife and five children in Auschwitz, ninety percent of their family was lost there. He thinks this stood behind his desire to prove himself, this is why he went against every possible authority, and didn’t back down even when he was threatened with having his children taken into institutional custody.
      Even though there are multiple books written about them, I felt all along the movie that this story is so rich that it strains the boundaries of a 70 minute movie. Maybe a television series would be the best format to expand every thread. For example there is the dynamic between the sisters. The biggest burden, as it usually happens, was placed on the oldest girl, Zsuzsa. She pioneered the way for the others and she was the one who suffered the most through the fights between her family and the authorities. The middle girl, Zsófi was said to be the most talented, and she achieved the least (which in her case is still a lot). Simply because she was the one who loved life the most, she was the most interested in the world beyond chess. And finally there is Judit, the best female chess player in history, who had the „killer instinct” that her sisters lacked. They did everything together, and now the live in three different continents.
      Or there is the men’s attitude towards having to play against women, moreover much younger women. For example, Judit’s greatest opponent and idol, Garry Kasparov said that women should occupy themselves with raising children, and the Polgár girls are a circus attraction, not chess players. When Judit beat him, after a quick handshake he literally ran off stage. (Though these days they promote good causes together, and he visited the Polgár-chess-day in 2013.)
      To defy the rules during socialism, raising geniuses versus a (conventionally) complete childhood, outsiderness and fitting in, the relationship of the individual and the authorities, processing success and failure: amazing how many things there are in this movie. This is exactly why I think the inserts are superfluous where the family members living thousand of kilometers apart get emotional listening to Hungarian songs. What happened is more than enough, it has an effect on the viewer without any kitsch.

  52. tjfwainwright says:

    We need to get the​ SuperMemo guy in a room with Scott.

    SuperMemo still being mostly stuck in Poland is enough to refute the “If Laszlo’s book were that great, we would have heard about it” comments for me.

    https://www.supermemo.com/articles/genius.htm#Genius Checklist

  53. Rob K says:

    wrt the starting young aspect, it seems notable to me that Judit Polgar is particularly brilliant in highly tactical situations in speed chess games. Which is to say, she creates positions with the potential for complex move sequences and little time to evaluate them, and outperforms in those.

    That particular strength would support the idea that, by developing chess “fluency”, so to speak, at a young age, she’s marshalled additional brain resources for the task of visualizing and understanding what for most people would be a slower step-by-step calculation. All very strong chess players do that to some extent, but this was her specialty even in the rarefied context of modern super-grandmasters.

  54. Eponymous says:

    As a general comment, estimates of the heredity of IQ rely on the existing variance in genes and environment in the population under consideration. Thus the general result of ~0 effect of shared environment on adult IQ among people brought up in middle class American homes may say more about the variance in environment in this population (as it pertains to IQ development) than about the potential for environment to affect IQ.

    If one buys the Flynn Effect, then the environment of a middle class American home (cum society at large) circa 2017 raises IQ by *30 points* relative to the environment of a middle class American home (cum society) in 1900. And I think we can all agree that America circa 1900 was a relatively intellectual culture compared to an illiterate hunter-gatherer tribe, and that America circa 2017 is not an intellectual utopia. So there’s presumably a lot more room in both directions.

    Thus we have rather strong evidence that a large change in environment can effect an astronomical change in IQ. Isn’t it then possible that a radical change in childhood environment, far outside the norm towards developing chess ability, could raise chess ability by the equivalent of 40 IQ points or so? Particularly if chess ability is more trainable than IQ (which we have evidence is actually quite trainable!)

    (As an aside, if we take the Flynn Effect at anything like face value, and accept the importance of IQ as conveyed by e.g. The Bell Curve, shouldn’t understanding the Flynn Effect be among humanity’s highest priorities?)

    (Yes yes I know, the people who believe in the importance of IQ are skeptical of the Flynn effect, and the contrapositive.)

  55. Bram Cohen says:

    Ranking the Polgar sisters based on their world standing is very misleading. Judit Polgar was at one point in the top ten chess-playing humans on earth when the #2 female was somewhere around #500. In that context her sisters’s achievements look like an unsurprising low level of correlation to a truly singular event. Also chess wasn’t chosen completely randomly, it was something Judit expressed interest in.

    Since then the record for youngest grandmaster has kept going down (I think it at one point was held by Judit). This would seem to imply that whatever methods it is that help teach chess well have been getting more widely disseminated, and that a big part of it is thorough immersion from a young age. It may help to ask chess educators how they’ve gotten better at teaching people chess, since those techniques appear to be entirely reproducible and have gotten better over time.

    • Steve Sailer says:

      If not for Judit’s spectacular achievements, we wouldn’t pay much attention to the Polgar sisters if Judit had only been about as good as her sisters. It’s kind of like the football-playing Gronkowski family: all the brothers are fine athletes, but Patriots tight end Rob Gronkowski is an amazing beast, so we hear about the family as a whole a lot.

      • John Schilling says:

        Agree that we wouldn’t have heard as much about it, but it would I think still be worth hearing about. A guy precommits to “I can teach any random child to e.g. play chess at Grand Master level” and scores three for three, OK, maybe he got lucky or maybe it doesn’t generalize beyond chess or maybe it only works for random Hungarian Jewish children – but if it does generalize, if we can reliably get Grand Master level performance in any chosen field, that’s a pretty big deal even if it doesn’t give best-in-the-history-of-the-world level performance.

      • Tomasyi says:

        If not for Judit’s spectacular achievements, we wouldn’t pay much attention to the Polgar sisters if Judit had only been about as good as her sisters.

        I’m not convinced. The “weakest” of the three sisters, the one who never made it to Grandmaster, was Sofia Polgar. But she still had spectacular achievements – at age just 14, she made one of the famous chess performances of all time in the “Sack of Rome”, beating a string of grandmasters. To have a 14 year old girl beating multiple grandmasters in a tournament caused a massive stir at the time, when many people still believed that women couldn’t play chess at the highest levels.

    • Squirrel of Doom says:

      I think people learn chess from/over the internet now, much like many other special interests, and I think it’s a trend that kids learn more faster than ever because of that.

      It’s also hugely easier to learn now that you can play computers of arbitrary strength 24/7 not. Back in Judit’s day, you had to play other people in person.

  56. ScarecrowBoat716 says:

    Don’t they do something like that in China for dancers and gymnasts? I believe their Olypmians are trained around the clock from an early age. Presumably training them really early when their brains are in hyper-developmental mode gives them a huge head start. Whatever processes help a toddler learn language may also help them learn other complicated subjects in quick fashion.

    • onyomi says:

      Nurture definitely makes a difference if you are okay with borderline or unambiguous child abuse. Otherwise, it seems mostly about not messing them up. Like you can stunt a plant’s growth, or you can provide it with the necessary sunlight, water, etc. it needs to reach its full height, but you can’t make a pine tree into an apple tree. That said, with some wire, you can make a pine tree do some cool things.

      Ironically, it seems like one can derive two contradictory lessons from a strong nature theory: the obvious one is Bryan Caplan’s: your children are basically going to turn out like they’re going to turn out, so just chillax. The other might be the Tiger Mom “tears and tantrums are temporary, glory is eternal!” That is, whether your kids are going to be well-adjusted or neurotic, fucked up adults is also largely out of your control and not much helped by letting them take it easy as children, so you might as well force them to learn some skills which you know will be bankable in adulthood, since one can be confident of the value of bankable skills, if not necessarily of a relaxed childhood.

      • albatross11 says:

        There’s more-or-less a set of “best practices” for raising kids in a normal family environment, and as far as we can tell it works pretty well. Within the range of those best practices, there doesn’t seem to be a lot of stuff that leads to substantial, measurable differences in outcomes for kids–that is, stuff like whether you let the kids watch TV, or whether you send them to public vs private schools, or whether you let them roam the neighborhood vs constantly watch over them.

        We don’t actually know that these best practices are the best ones possible. Perhaps there’s some additional set of practices that give even better results, but either we don’t know about them[0], or most people can’t or won’t manage them for reasons of workload or resources or social acceptability[1].

        [0] For all we know, there’s something as commonplace as reading to your kids in middle-class homes, which would reliably boost kids’ IQs ten points. If not many parents do it, then it won’t show up in the studies of twins or siblings.

        [1] As an example, David Friedman, a regular participant here, was raised by two economists, one a world-famous genius who won a Nobel prize. It wouldn’t be a huge shock if that had a lasting positive impact on him. But since most of us don’t have a spare world-famous genius hanging around to help us raise our kids, that’s a “best practice” that isn’t so easy to use. (Nor do we have useful statistics on that.)

        • Aapje says:

          [Writes down idea for startup: Nobel prize winner baby sitters]

          • Mensa baby sitters would be a lot less expensive. What you want are very smart people who have not been very successful.

            I know someone who published a novel that some think very highly of. I believe she wrote it while she was functioning as the nanny for a Harvard academic family.

          • Aapje says:

            It was a joke, David 🙂

          • tkolbe says:

            I think Mensa baby sitters or the like is a great idea. One downside, potentially, is that the target may be closer to conscientiousness + high IQ which may be more difficult to find among the intelligent but unsuccessful.

          • John Schilling says:

            conscientiousness + high IQ which may be more difficult to find among the intelligent but unsuccessful.

            Are teenagers still allowed to work as babysitters on more than an informal basis? If so, that gives you a potential pool of high-IQ types who don’t yet have more conventional approaches to success available to them.

            And the Polgar experience suggests that you may want someone ~7 years older than the target. OTOH, I don’t think any of the kids I babysat turned out to be rocket scientists. OTOOH, I didn’t have Polgar’s book.

          • When our daughter was very little, she had a baby sitter who was 11 or 12–and turned out to be a very young adult. Many years later I got an email from her.

            She was an air force fighter pilot.

  57. rulerstothesky says:

    “(be right back, going to change my OKCupid profile to include “must be interested in n=1 developmental-psych experiments, have access to a rare book library, and speak either Hungarian or Esperanto”)”

    Wait…you mean your OKCupid profile didn’t already say that?

  58. elialbert says:

    this seems like a good time to mention starchess, invented by the polgar family.
    see: polgarstarchess.com altho the website seems to be down?
    and my version: StarChess Online,
    I tried to reach out to them about it but for some reason they didn’t write back…

  59. littleyid says:

    I don’t know that this trail leads anywhere, but I saw The Polgar Variant a couple of years ago. I can’t find the clip online, but I recall Laszlo explicitly saying that he got his teaching method from the Talmud. I think he then quotes the aphorism “[if someone tells you] ‘I tried, and I succeeded’, believe him”, and says nothing more on the matter.

  60. wjbrown says:

    Is this title of this post a reference to Daniel Ingram?

    • Blue Tribe Dissident says:

      Almost certainly so. I certainly wasn’t expecting an Ingram reference here. In hindsight, Ingram sort of is the Eliezer Yudkowsky of contemporary Buddhism.

  61. harvey-ta says:

    A machine translated (Hungarian to English) of the contents page:
    Foreword (András Mezei) 5
    Another Foreword (János Szabolcsi) 13
    The mysteries of a pedagogical experiment 17
    Instead of introductions: puppets or happy? 17
    Stalemate? Heritage or Cultivation? We move from deadlock 23
    Open: the Citizenship Pedagogy 29
    To school or life? 29
    All children promise 33
    The genius is a treasure or collateral? 41
    If you want, they will be selective … 51
    Esperanto: a common language of Queen and Peasant? 51
    Mid game: 64 black and white 63
    The selected game 69
    The Saxon People 69
    Chess: Step 77
    Chess in Psychology, Psychology in Chess 101
    Chess-right for male prejudice 109
    End game: everyone can win 109
    The Family 119
    Polgárné’s example: the needle is followed by thread 125
    Minority … 131
    The Coronation Juniors: Like Three Happy “Princesses” 137
    You can become a genius – happiness! 145
    Annex 155
    The Talento Foundation 163
    Racing Games – Puzzles 165

    • publiusvarinius says:

      Some corrected/clarified translations:

      Opening: the Polgár pedagogy 29 (the word polgár means a city-dweller, civilian or citizen, which confused Google Translate)
      The genius: treasure or burden?
      Esperanto: a common language of Queens and Pawns? 51
      Check-mate for male prejudice 109
      Mrs. Polgár’s example: the thread follows the needle 125

    • b_jonas says:

      Same question as to thedirtyscreech: is this the first edition or the second (revised) edition?

  62. Baja Roki Thompson says:

    If he wanted to truly demonstrate the reliability of his system, he would’ve raised her girls to be geniuses at parenting like himself.

  63. b_jonas says:

    What is the proof that Polgár László knew in advance that he was trying to raise his children as geniuses, as opposed to telling that story after that his daughters have turned out to be geniuses? Did he pre-register his experiment? Is there some article or interview or classified ad with a definite date?

    • Douglas Knight says:

      Scott mentioned Polgár’s book, which was published in 1989, but appears to have been written much earlier. In 1989 Kárpáti Tamás wrote a book subtitled “Polgár dossier 1974-1989” that presumably contains documents from 1974 (when Susan was 5). What those earliest documents say and how authoritatively they are dated, I do not know. Supposedly László discussed his plans in letters wooing his Ukrainian wife 1965-1967, but I don’t know if they have been published.

  64. markk116 says:

    This oddly ties into a book I’ve been reading about practice and expertise called Peak. I’m not all the way done yet but it seems to me that the father did right is setting them up to engage in deliberate practice. I’d be curious to see if their brains are notably different compared to veteran chess players who didn’t start out at such a young age.

  65. Alsadius says:

    (be right back, going to change my OKCupid profile to include “must be interested in n=1 developmental-psych experiments, have access to a rare book library, and speak either Hungarian or Esperanto”)

    I’m now glad my fiancee isn’t on OKC. There’s no way I could compete with an ad like that. (Weirdly enough, I don’t think I’m even joking)

  66. sconn says:

    So, the magic way to make a child top at a specific skill is to let them pick a skill they’re interested in, and then make them practice it for a zillion hours? That’s unsurprising to me … naturally practice helps you learn things, and interest helps motivate the practice. Without the interest, you’d probably end up with something more like Battle Hymn of the Tiger Mother … successful in the short term, but then one of the kids quits violin and refuses to touch it anymore because she’s so traumatized.

    I suspect this is why autistic kids, even if their IQ supposedly isn’t extremely high, can become top-level experts in one specific skill. With the drive to practice for hours every day, you can get pretty darn good at anything. And naturally the interest is somewhat driven by the tendency of the person’s brain — if you are good at numbers, for instance, you’d probably be drawn to a numbers-related field and find it rewarding enough to keep at it. You can’t just take a random kid and give them a random interest.

    But I don’t know exactly if that counts as “raising a genius.” What they did was raise chess experts, and they got that by giving less time to other subjects. So probably those chess experts aren’t going to suddenly surprise us by getting Nobel Prizes on the side. If you want to raise a child with a lot of general intelligence, the route might be different.

    In Battle Hymn of the Tiger Mother, the author claims all the credit for her daughters’ success — it was her forcing them to study that got them into the Ivy Leagues, right? Her own parents raised her the same way and that’s why she’s successful. But then she also mentions that her husband has been just as successful, despite having been raised by loving but more hands-off parents who let him play after school and didn’t push him to do anything he didn’t want to. As an added bonus, he’s WAY less neurotic than she is and has a clearer idea of what he wants to do with his life. It kind of makes you wonder — what was the point of all that pushing if a child who actually wanted to get to the Ivy Leagues could do it on his own?

    (Yes, Amy Chua’s husband is Jewish.)

    • John Schilling says:

      So, the magic way to make a child top at a specific skill is to let them pick a skill they’re interested in, and then make them practice it for a zillion hours?

      There’s probably more to it than that, but we’ll have to learn Hungarian to be sure. Even so, “let them pick a skill they are interested in” seems weak. It strikes me as highly unlikely that all three Polgar girls independently picked chess as their passion, so even if Laszlo didn’t deliberately steer them in that direction the choice can almost certainly be influenced and the Polgar experience (once we decipher the documentation) may have something useful on that.

      Interestingly, the most dramatic outcome of the experiment came with the youngest daughter. So possibly you’d want to arrange your children to have age-appropriate role models with a passion and talent for the chosen subject in residence during early childhood.

      OTOH, John Stuart Mill was an eldest child, and he pretty clearly followed his father’s chosen path. But the Mill/Bentham method and the Polgar Method may be incompatible.

    • Toby Bartels says:

      Ah, but who got decide how to raise the kids? She’s more successful than him after all!

    • sconn says:

      There are always influences. If the parents hadn’t owned a chess set, the kids wouldn’t have been into chess; and obviously the younger girls just picked up the oldest’s interest. It’s quite reasonable, if you’re trying to raise a prodigy, to guide their interests. You keep them away from stuff you don’t want them to waste time on (like TV and computer games) and introduce them to things you hope they will take an interest in. My kid obsessively memorizes details in stories, so I’m hoping to get him into history since that’s basically the same thing, only more useful than Harry Potter fandom. But maybe he won’t like that, so we’ll try some literature and see if that takes. You get the idea.

  67. Muad'Dib says:

    Here are several related points about IQ scores and Elo ratings:

    1. Sofia Polgar never achieved the GM title; she “only” became an IM and a WGM.
    2. The Levitt formulae are from two decades ago. There has been some concern that there has been Elo rating inflation in the past few decades. Certainly ratings at the very top have risen a lot, but it’s hard to say whether this represents a genuine increase in strength or ratings inflation. Anyway, I wouldn’t trust the formulae from that far back too much.
    3. As a long time (casual) chess follower, my feeling is that most GMs of ratings close to the Polgar sisters do not have IQs of anywhere near 160. Kasparov’s 135 score seems more reasonable. To use Charles Murray’s football analogy, you need to be at least X pounds to be an NFL player, but the top players need not be the ones that are the heaviest.

  68. slitvin says:

    (be right back, going to change my OKCupid profile to include “must be interested in n=1 developmental-psych experiments, have access to a rare book library, and speak either Hungarian or Esperanto”)

    Scott-I’m afraid with such lax criteria you’ll be inundated with potential mates. Might I suggest requiring that the candidate be able to speak BOTH Hungarian and Esperanto? You can also limit the candidates further to native Esperanto speakers only.
    Good luck and please keep us posted.

  69. Polymath says:

    My theory is that chess talent and IQ make an equal contribution to chess success assuming serious effort is made to reach one’s potential. The formula I came up with is
    (Elo-1600)=10*((IQ-100)+(talent-100))
    scoring talent on a Mean 100 SD 15 scale like IQ.

    So someone with 160 of each would be World Champion caliber. Recent champions have had higher chess IQs than standard IQs but the reverse was probably true of Lasker, Steinitz, Euwe, and Botvinnik.

    As for Bobby Fischer: I have it on extremely good authority (my father and my godfather were both close friends of Bobby’s from when he was 9 until he won the world championship) that he was extremely intelligent in the ordinary sense of the word; he was a high-school dropout and never lost his earthy Brooklynese way of speaking, but he was very well-read and very systematic and scientific in his intellectual approach even as a 9-year-old.

  70. diulce says:

    On the other hand, I’m not sure Levitt’s right. Chess champion Gary Kasparov actually sat and took an IQ test for the magazine Der Spiegel, and his IQ was 135. That’s not bad – it’s top 1% of the population – but it’s not amazing either.

    Most IQ tests (and all tests used in a clinical setting) have a real ceiling at around +2 standard-deviations above the mean, or 130 (sd=15). The problems featured in these tests can be solved by most people with an IQ at or above the 98th percentile, which means someone with an IQ around 160 can solve the same set of problems as someone with an IQ around 130. Hence, these tests are not adequate tools to differentiate between IQ levels above (approximately) 130.

    When one scores above 130 on a timed test, the best interpretation should be that his/her IQ is at or above 130. For instance: “Gabriel scored 160 on the WAIS. Hence, his rarity IQ is probably above 130.” Or: “Joanna scored 140 on the WAIS. Hence, her rarity IQ is probably above 130.” An unreasonable inference would be: “Joanna has a lower IQ than Gabriel”. Since the problem sets can’t differentiate IQs above 130, we can only infer that both are above 130. A reasonable inference would be: “Gabriel is faster at solving problems than Joanna”. Often speed of thinking is confused with depth of thinking, leading to the conclusion that faster thinking equals higher intelligence.

    Kasparov’s true rarity IQ is probably much higher than 135, between 170-180 (sd=15). This is corroborated by his theoretical innovations to chess and depth of analysis during live games. Since timed tests can’t differentiate above 130, the alternatives are: hard un-timed tests, which have a true ceiling probably around 160, and analysis of individual intellectual contributions. One who studies chess at a high level can see the brilliance of Kasparov, similar to the contributions of Terence Tao to Mathematics. Also, his non-chess writings point to a much higher verbal IQ than 135 as well.

    This is what we should expect given the correlation of about r = 0.24 between IQ and chess ability (see also this analysis, although I disagree with the details). And the contrary claims – like the one that Bobby Fischer’s IQ was in the 180s – are less well-sourced (although Fischer was the son of a Hungarian-Jewish mathematician, so who knows?).

    The correlation between IQ and chess ability depends greatly on previous training. Given similar high-level training, the correlation should increase sharply. Fischer’s estimate at 180 (sd=15) is actually much more credible than estimating Kasparov’s at 135, although that might not be immediately obvious to someone who didn’t study chess or psychometrics.

    If it were possible to be a chess world champion with an IQ of 135, then maybe it’s possible to be a “mere” grandmaster with IQs in the high 120s and low 130s.

    It’s extremely unlikely. Levitt’s formula seems to give an estimate of the required IQ needed to achieve a certain Elo rating, which of course also requires training. Also it’s easier to be a Grandmaster today and than in the 1960s, because of rating inflation.

    And it’s just barely plausible that some sufficiently smart people might have three kids who all have IQs in the high 120s and low 130s.

    I can think of one other example (Terry Tao and his brothers, all three competed in the IMO, one is a chess IM, one is an obviously absurd genius; low estimate > 140 (sd=15) and > 190 for Terry). We also shouldn’t underestimate the parents’ IQs, in Laszlo’s case he created his own pedagogical system for rearing his children and is an accomplished chess theorist and composer.

    But this just passes the buck on the mystery. 2% of people have IQs in the high 120s or low 130s, but 2% of people aren’t the top-ranked female chess player in the world. The Polgar sisters’ IQs might have been a permissive factor in allowing them to excel, but it didn’t necessitate it.

    This is the case in almost every intellectual pursuit I can think of. IQ rarely necessitates excellence in a specific field, and other factors such as parental guiding, positive rewards in school or pure randomness leads a high-IQ person to excel in chess, or maths, or physics, or etc.

    The Pólgar sisters are probably the result of very high IQs and specific training from a very young age (most Grandmasters learned chess when they were 3-6 years old). They would probably have done well in other areas without specific training anyway, but probably not top-10 in the world. This should echo what most high-IQ people have asked for a long time: identify early which individuals are very smart and give them individual, specialized training in what they enjoy studying.

    • albatross11 says:

      I think we need to make a distinction in our discussion between:

      a. IQ scores, which are the thing we can actually measure with the tests we have.

      b. Intelligence, which is the thing we want to measure with IQ tests.

      It’s clear that IQ scores are correlated with intelligence as we mean it in normal conversation, but also that they’re not identical. Some people will get IQ scores that are a lot higher or lower than their actual intelligence. If someone tells you that Feynman or Kasparov got mediocre IQ scores, you suspect that the test didn’t really do a great job measuring their actual intelligence, because their achievements suggest a huge amount of intellectual ability. Similarly, we can speculate about what kind of intelligence Gauss or Newton or Archimedes must have had, even though nobody had even imagined IQ tests when they were alive. And who knows of modern IQ tests, designed and tested against modern students, would even be useful measures of intelligence for someone whose background and life was a different from ours as, say, Archimedes.

      There’s one other source of confusion here: IQ scores are reported in terms of where you are in the distribution of IQ scores, which are normally distributed with a mean of 100 and a standard deviation of 15. So when we talk about someone with a 146 IQ, we’re talking about someone in the top 0.1% (1/1000) of the distribution. Probably any physicist who accomplishes anything important in his field is well past this level of intelligence.

      But a lot of times, we’re interested in a base level of intelligence, like how smart you have to be to learn calculus or to be able to read and understand a moderately complicated newspaper article. Those are absolute levels of intelligence–you don’t become less able to read a newspaper just because there are more people in your society who can do it. It seems intuitively like there’s a base level of intelligence that’s needed to do a lot of demanding things well, but I don’t actually know that for sure–maybe if we go slowly enough we can get people with 100 IQs through calculus. I’m skeptical, but I don’t really know.

      • diulce says:

        IQ is a relevant metric insofar as it measures a type of cognitive performance that is somewhat consistent across different fields for individuals who have a similar score. To be clear, “different fields” means pattern recognition, spatial, numerical, verbal ability etc., not science, maths etc.

        A general intelligence that could be tested by such a test has been long hypothesized but its first real statistical confirmation was given by Spearman, who called it “the g factor.” G is not a philosophical concept; it is a statistical correlation that Spearman found among school-children, whose grades in different subjects tended to correlate. Nowadays the literature sometimes uses “the g factor” to mean some sort of core intelligence.

        The fact is, whatever the general intelligence is, an IQ test should measure it. Some people like to say that “IQ tests measure the ability to solve IQ test puzzles,” and others are less nonchalant about it but still consider it largely a self-contained arbitrary number, which bears no real meaning outside of IQ testing land.

        When intelligence testing first begun to be studied, it was largely based on ratio IQ, which equals mental age divided by the chronological age. This was mainly because IQ was measured in children to identify slow learners, and intuitively we know that children become smarter as they age. Later on the notion of rarity IQ was introduced, which assumes a Gaussian distribution scale with the mean at 100 and standard-deviation at 15, 16 or 24, depending on the test.

        When I say that Kasparov’s IQ is much above 135, what I mean is that there is a set of mental abilities which IQ tries to measure, and which it measures well within a certain range (70-130), and a test that would be able to measure the rarity of these abilities at higher levels would implicate a score much higher than 135. Similarly we can estimate where on this rarity scale does Einstein, Feynman, or Archimedes stand.

        IQ is not IQ score. A score is an estimate of the rarity of a person’s mental abilities as IQ tests intend to measure. IQ is the rarity itself, which is often approximately equal to the score one receives on a clinical test (WAIS, Stanford-Binet etc.), if the score lies between 70-130 (sd=15). Above this range these tests do not provide a result that accurately reflects the rarity of the abilities that the test is trying to measure.

Leave a Reply