The Polgár sisters success story

Extreme parenting but interpretation is difficult due to genetic confounding

The essence of my pedagogical program is that in my opinion, every healthy child can be raised to be an outstanding person, in my words, a genius. When we began this work with my wife, we read through a large collection of books and studies. We examined the childhoods of many eminent people and noticed that all who became geniuses specialized very early in some field, and we could also document that beside them always stood a father or mother, a tutor or trainer, who were “obsessed” - in the good sense of the word. So on the basis of our research we could rightly conclude that geniuses are not born: one has to raise them. And if it was possible to raise an outstanding person, we definitely needed to try this. So we did, and our attempt brought success. (László Polgár, How to Raise a Genius, 1989)

And so he did! Well, sort of. The story of the 3 Polgár sisters is truly fascinating. Scott Alexander wrote about them in 2017, but for some reason he did not properly add the context to understand this from a behavioral genetics-informed perspective. Since the story is spreading on Twitter (this post has >6 million views), I think it's in order to cover the background context of this story. The basics are these:

• László Polgár, living in communist Hungary, had an idea for how to homeschool children to become "geniuses". He found a wife who liked this idea and they gave it a shot.

• They had 3 daughters, an inauspicious start given that most elite anythings are men.

• Nevertheless, all 3 daughters became chess champions. Respectively, they achieved peak ELOs of 2505 (Sofia), 2577 (Susan), and 2735 (Judit). If you aren't familiar with chess, these are very high ratings, and the last sister peaked at 8 top in the world. Even the least successful was 6th best female player at her peak.

• These results are extraordinary, and are generally taken to show that upbringing can matter a great deal.

Based on this story, one might conclude indeed that geniuses are made, not born. But then again, here's some additional context:

[László] Polgár was born on 11 May 1946 in Gyöngyös, Hungary.[3]: 107  He studied intelligence when he was a university student. He later recalled that "when I looked at the life stories of geniuses" during his student years, "I found the same thing...They all started at a very young age and studied intensively."[4] He prepared for fatherhood before marriage, reported People Magazine in 1987, by studying the biographies of 400 great intellectuals, from Socrates to Einstein. He concluded that if he took the right approach to child-rearing, he could turn "any healthy newborn" into "a genius."[5] In 1992, Polgár told the Washington Post: "A genius is not born but is educated and trained….When a child is born healthy, it is a potential genius."[2]

In 1965, Polgár "conducted an epistolary courtship with a Ukrainian foreign language teacher named Klara." In his letters, he outlined the pedagogical project he had in mind. In reading those biographies, he had "identified a common theme—early and intensive specialization in a particular subject." Confident that "he could turn any healthy child into a prodigy," he "needed a wife willing to jump on board."

What kind of person reads 400 biographies? I would guess the median person reads maybe 20 books in their lifetime, and most of these are novels, not history books. Oh and by the way, László Polgár is Jewish and Klara was a teacher to begin with.

To say that their upbringing and education was rigorous and unusual is an understatement:

– 4 hours of specialist study (for us, chess)
– 1 hour of a foreign language. Esperanto in the first year, English in the second, and another chosen at will in the third. At the stage of beginning, that is, intensive language instruction, it is necessary to increase the study hours to 3 – in place of the specialist study – for 3 months. In summer, study trips to other countries.
– 1 hour of general study (native language, natural science and social studies)
– 1 hour of computing
– 1 hour of moral, psychological, and pedagogical studies (humor lessons as well, with 20 minutes every hour for joke telling)
– 1 hour of gymnastics, freely chosen, which can be accomplished individually outside school. The division of study hours can of course be treated elastically.

9 hours a day, even for young children. Some might consider this child abuse, but László and Klara clearly didn't. I am positively disposed to home schooling and generally think public schooling is a massive waste of time and money, so if some couple out there is thinking of trying some rerun of this, you have my blessing!

In behavioral genetics terms, we can think of the variation in upbringing and schooling experiences that exist. There's quite a lot of variation, but you have to admit the Polgár method is extreme. How extreme? Well, if we think about chess specifically, I would guess most (Western) families teach nothing about chess to their children, or maybe the bare minimum of the rules. Some families play together. I played my father when I was a teenager. Very few families decide on such an extreme and directed effort towards on particular skill. In statistical terms, how many standard deviations is this above the average? I don't know but literally one in a million is probably not enough, and that would correspond to about 4.75 standard deviations. So let's say their upbringing was at least 5 standard deviations (SD) above the mean for chess. (In this case, we will have to ignore that chess upbringing is not really normally distributed, but some kind of non-negative long-tailed distribution, say Poisson.)

But what about the genetics side? Clearly, László Polgár was smart: Jewish, a chess teacher, university educated in the 1960s communist Hungary, an avid nonfiction reader, a successful writer, and "a visionary" who "always thinks big". If we assume he's 135 IQ, we are probably not so far off, but higher is not unreasonable. There seems to be little information about Klara, but if we assume the usual assortative mating of close to r = 0.5, her IQ would be about 115. Given that she was enthusiastic about this grand education plan, we can guess she is probably higher. So what is the average parental IQ, 125 seems reasonable (1.67 SDs above 100), with the allowance that it might be 120 or 130+ (1.33 SD or 2+).

Bringing these facts together, then, we have 3 sisters who were subject to a 5+ SD parenting effort for chess, and they began with likely high general intelligence. Statistically, the sisters will have likely regressed a bit towards the mean, so maybe their average IQ is 120 or 115. Having 3 children with an expected mean of 120 gives rather decent chances (39%) that at least one of them is 130+, and not so small a chance (10%) that 2 of them are 130+. But is 130 even enough?, you should know that top chess players aren't all that bright, not any smarter than many readers of this blog. Gary Kasparov had an IQ around 130. If one does a crude extrapolation based on the number of grandmasters and the known IQ and chess correlation in general population samples (r = .30 to .40), the mean grandmaster IQ should be about 130. They are bright people have who spent a lot of time mastering one particular skill as opposed to doing something else. The thing that distinguishes grandmasters from others is their sheer dedication to mastering one game for years and years, usually from early childhood. Polgár is right about this aspect of extreme talent in some fields. This goes back to Galton's 'triple event':

The particular meaning in which I employ the word ability, does not restrict my argument from a wider application; for, if I succeed in showing—as I undoubtedly shall do—that the concrete triple event, of ability combined with zeal and with capacity for hard labour, is inherited, much more will there be justification for believing that any one of its three elements, whether it be ability, or zeal, or capacity for labour, is similarly a gift of inheritance.

"Zeal" is certainly a word one would describe the Polgár story with.

By the way, after achieving success with his daughters, László went even further in this 1992 interview:

Polgar, who has proven his point that grandmasters can be trained and that they can just as easily be women as men, now wants to break the racial barriers in the virtually all-white chess world. He says that he and his wife, Klara, will adopt a black infant from the Third World and train the child, like his daughters, to become a chess genius."He has ideas in his head that he wants to achieve whether they are right or wrong," says Zoltan Ribli, 40, a grandmaster and former tutor to the Polgar girls.

I can't find any mention of this happening. Quite a number of celebrities have adopted third world children, mainly Africans, and I don't see any success stories about these. Their upbringing is probably not characterized by zeal, but their education surely has anything money can buy.

This is really an n=1 family story, but if we wanted to draw conclusions from it despite this, we might go for these:

• The Polgár sisters are genuinely very impressive, and their father did really set out to prove one could rear children into "geniuses".

• The sheer amount of effort this took from the parents was enormous even by modern tiger mom standards.

• However, it should not be forgotten that the parents were quite bright to begin with. These kinds of extreme parenting stories usually have extreme genetic confounding. This cannot be ignored.

• So: If you are willing to deviate in the extreme in your child-rearing approach, and you found a spouse who also wants to do this, and you both are of reasonably high intelligence, then you can probably raise a child to perform at the elite level in any game that relies upon long-term memory.

• This story doesn't really show that one can rear any child into a chess champion or genius in the ordinary sense (say, Newton). Realistically, almost no one is in position to copy their efforts, most people don't have children that bright to work with, and neither do they possess the zeal or intelligence themselves to structure such an education.