Why Romania excels in international Olympiads

I don't think pillefitz is asking if the author's racist arguments in general are sound. I think he was asking if the arguments in this particular article are sound.

Many racists are actually quite competent outside of when they are trying to argue for racism, and this particular article doesn't seem to be making a racist argument.

It does talk about the possibility of an over performing ethnic group, and suggests Jews as a candidate, but that's not inherently a racist argument. There are in fact ethnic groups in many countries that do over or under perform as a whole, but this is almost always due to cultural factors.

In particular ethnic groups where education and scholarship are a big part of their culture unsurprising tend to over perform in general society unless their culture also is big on minimizing interacting outside their group.

Jewish culture is such a culture and they have long been over represented in intellectual fields unless they are in a country that has high social or legal barriers to keep Jews out of those fields. Heck, this is probably why so many stupid conspiracy theories have them as the people secretly running the world.

It's similar with many Asian cultures nowadays. Just take a stroll through the campus of Caltech or MIT if you want to see that those groups over perform. At both those schools around 48% of undergraduates are Asian. That's way above the 7% of Asians in the general US population. Even if we take into account that Caltech has a disproportionate number of students from California (if you can get into a top STEM school and are from California, are you really want to go deal with Boston weather?) and California has a high Asian population (about 16%) the 48% at Caltech is way higher.

That said, I don't think the arguments in this article are really all that good. He is emphasizing the IMO since 2020 and talking about how 2022-2024 stand out. But when you look farther back instead of starting at 2020 then 2022-2024 are still good years, but not really remarkable years for Romania. Over the period it has been good it has had some major changes in their education system which I think limits what you can infer from recent IMO success. I've got another comment that goes into more detail: https://news.ycombinator.com/item?id=45071972

Absolutely! And that's my point! Broading access to education is a net benefit to society, because fortune prepares an able mind.

> The difference could be connections

In Ilya's case, not really. He was an immigrant twice (first USSR to Israel, then Israel to Canada), and wouldn't have been able to do an MS/PhD at UT without actually being capable - it's not difficult to enter UToronto, but it's difficult to leave with a CS bachelors degree.

I've always felt a broad access system like the Warren-era UCs along with LBJ's "Great Deal" is of better benefit for a country than investing in building isolated ivory towers for education, because we have the ability to better ourselves, and that door should always remain open.

Agreed, there is no way to know the exact degree of differences between the two populations (although any geneticist would reject the notion of primarily unselected genetic drift in the example I described. I specifically outlined strong selective pressures in the environments). But at least we agree that they would not continue to be identical, and that the environmental pressure, over time, would also show up in genetic drift between the populations (there are surprisingly many blank-slaters who would disagree even with this premise).

We have tools in modern science to attempt to differentiate between different underlying components of genetic variation vs environmental variation (identical twin studies, adoption-at-birth studies, general population studies, longitudinal analyses, etc). In lieu of a 1000-year experiment (which we cannot recreate), we can make predictions. If factor X was primarily genetic, we would expect to observe A, B, C. If factor X was primarily environmental, we would expect to observe D, E, F. Then, we can basically incorporate various studies and results and do gradient descent, and observing which (means squared error) is smaller.

None of this is groundbreaking stuff, I know. But I have read arguments from both the primarily environmental camp and the primarily genetics camp, and one of them has to do a lot more mental gymnastics explain the differences between obvious predictions (under this paradigm) and the observed data in the real world.

Doesn't he just say that Romanian Jews overperform?

I don't think there's a need to read the authors words in the worst possible way.

> I understand this as ‘it makes sense for a group with skin color X and hair color Y to be better at school’

It wouldn't make sense for the author to mean it that way, because the "white skin black hair" classification likely includes more non-Jewish than Jewish Romanians.

> followed by ‘it must be Jews’

I mean, it's not that hard to believe. If you look at (Eastern) European history, A LOT of the people of scientific or cultural significance were Jews. Don't pin me on the numbers, but if you check the Wikipedia article of a random Easter European person that left a mark in science, there is an 33-50% chance that person has some sort of Jewish background.

I'm not going to make a scientifict argument about innate intellectual ability - I don't consider myself qualified enough for that.

I can only speak to my own observations. I have taught a lot of people to juggle. Juggling is an interesting discipline because it's pretty distinct from any other activity that most people have participated in in their life. It also has lower risk of "contamination", since people are unlikely to lie if you ask them if they've juggled before. You can also run this experiment with young kids, etc - you will still see very different rates of skill acquisition. I would bet that any other environmental arguments also fall flat - there are no big pro-juggling or anti-juggling cultures out there.

When you get a group of people together and instruct them to juggle, people will pick it up at very different rates. If you have enough of a sample size of observations, you can see extreme variability - some pick it up after 10-15 minutes, others still can't juggle after a year of trying (believe me, I've seen both). I've also observed brothers, both professional jugglers from a young age and both having invested identical amounts of time and passion into the activity (again, have no incentive to lie), progress at very different rates with acquisition of the skill.

From observing hundreds of people learning to juggle, I strongly believe that there is an "innate aptitude for juggling" in every person. Even people without this innate aptitude can learn to juggle and get to a certain level at the skill, beyond which progress will be so slow as to represent a de-facto "ceiling" on their juggling skill.

If you were to make this same argument in intellectual domains (i.e. chess, as an objective example), the argument gets more muddied in that you could make valid arguments that environment has played a role in "predisposing" someone to more rapid skill acquisition. However, I like to think of skill acquisition as more of a y = mx + b equation (not that all skill aquisition is linear). Every person has a certain rate (slope) at which they can aquire a given skill, and a certain intercept (predisposition) towards the acquisition of that skill. This model also works since someone that acquires skill more slowly can expend more time and achieve the same level of competence, but their ultimate ceiling of skill aquisition is lower (since time is finite).

I can make similar arguments about other domains, having spent time with IMO medalists, IOI medalists and other "outliers" in the intellectual domains. The same examples I showed above still apply, but the data is a lot more messy because it's hard in those cases separate the environmental/cultural effects from the genetic effects. That's why I reach for the juggling example as a clearer argument for that.