449 Comments
Comment deleted
Expand full comment

Variants of this story also say that von Neumann had to be brought in to convince him otherwise.

Expand full comment
Comment deleted
Expand full comment

Higher polygenic scores in Ashkenazi Jews.

Expand full comment
Comment deleted
Expand full comment

It's literally a google search away.

Expand full comment
Comment deleted
Expand full comment

And specifically that the Ashkenazi have higher scores than Sephardic & Mizrahi Jews. If it were the Jewish religion requiring literacy, that should apply to all of them (and people should have noticed Jews were smarter than average much earlier).

Expand full comment
Jul 13, 2022·edited Jul 17, 2022

While not quite at the Ashkenazi level, (no one is, except for maybe a few populations like Tamil Brahmins, Parsis, and maybe the Italian Jews), Sephardic and Mizrahi Jews are generally pretty smart, and in countries like Iraq, Iran, Morocco Jews have historically been a huge proportion of the educated and affluent people. Mountain Jews were overrepresented among billionaires in Russia. Within Israel the gap between Ashkenazi and other Jews in EA is only ~.5 SD https://www.journals.uchicago.edu/doi/abs/10.1086/443926 which is smaller than the Ashkenazi-White gap in the US, and the wage gap is also smaller than the Ashkenazi-White gap in the US (there was a twitter thread by @HeTows at https://twitter.com/HeTows/status/1546928804089810945 about this). Indeed non-Ashkenazi have higher SES and EA than the white majority in the US and in France.

The Jewish religion requiring literacy is probably what it is, though not in the sense the environment people imagine. Less intelligent Jews were more likely to leave the community, and literary ability was culturally prized so smarter people had more kids. My guess is some of both effect. The culture drove selection for intelligence. Iraqi Jews are a high IQ population (again, not quite as high as Ashkenazi, but still high), but they are good spatially, unlike the Ashkenazi. There's an obvious reason, Jews in Europe were banned from a lot of professions requiring high spatial IQ, so people with high spatial IQ left. In the Middle East, it was more just the dumber ones couldn't afford the jizya and converted out.

Cultural causes, genetic effects.

Expand full comment

If you are a Jewish man and can't read, you can't get Bar Mitzvahed. If you can't get Bar Mitzvahed, you can't get married. You aren't forced to leave the community, but you aren't going to have a lot of descendants. (Besides, where would you go? The US wasn't taking immigrants in the 1200s.)

I think being the member of a community outside the mainstream tends to bring forth a lot of creative intellectual ability in the appropriate environment. We associate gay men with being leaders in art and design, but it's obviously not genetic, and being good in art and design doesn't make you gay.

Expand full comment

Tangential, but I am inclined to think there's a deeper correlation with gay men *caring more* about art and design; maybe it's purely cultural, hetero men in continental Europe care a lot more about fashion than Anglo hetero men, but I think there's something more fundamental there too.

(my hypothesis, fwiw, is that visual appearance matters more for attracting men than for attracting women, so gay men and hetero women on average care more about fashion and art and design than hetero men and gay

women)

Expand full comment
Mar 2, 2023·edited Mar 2, 2023

"If you are a Jewish man and can't read, you can't get Bar Mitzvahed. If you can't get Bar Mitzvahed, you can't get married."

No. That is not how any of this works. You turn 13 you are Bar Mitzvahed. Period. The party and the reading of the torah portion is not necessary. A Bar Mitzvah has nothing to do with getting married. I don't think that the Jewish marriage laws even require you to be 13 years old.

Expand full comment

That's one of the problems / advantages of Judaism. Different groups have different practices. That's what I was taught at Hebrew school when I was a kid. Maybe they were wrong.

Expand full comment

Where is the lack of data?

Genetics and intelligence data, and Jewish intelligence data are plentiful.

Expand full comment
deletedJul 13, 2022·edited Jul 13, 2022
Comment deleted
Expand full comment

While I agree that finding the genetic mutations would be better evidence, its not the only evidence.

Expand full comment
Comment deleted
Expand full comment

What are your priors?

Expand full comment

First time here I assume?

Trust me, these folk seem very invested in their just-so stories, to the point that I've had arguments that we should ignore the genetic evidence we do have in favour of IQ scores and cod racist typologies when it comes to intelligence.

That said, Scott has raised the idea of Ashkenazi genetic intelligence as an example of rare large-effect genes related to intelligence (all the data we have thus far seems to show that intelligence is massively polygenic and small-effect) that all correlate to really grody genetic diseases (either via partial dominance or tight linkage). I'm not sure what operable approaches you could draw out of such an argument, but it is at least a stab at a testable hypothesis.

Expand full comment

>Trust me, these folk seem very invested in their just-so stories, to the point that I've had arguments that we should ignore the genetic evidence we do have in favour of IQ scores and cod racist typologies when it comes to intelligence.

This sounds terrible, sorry to hear you had to deal with that.

As an aside, the quotation that Unsigned Integer mentions is dismissive from Scott to the point of laziness. What does 'I side with Cochran' mean? That there are no effects from culture & history? In a charitable reading, it's emphasising one side of the story over the other one, but the idea that genetics operate in some independent world where you can give someone with von Neumann's genetic makeup any form of upbringing and they will miraculously become a genius on the basis of genetics is just foolhardy. Any social scientist would look at the genetic explanation on one side and the cultural / historical one on the other and think: 'Oh, so obviously it's both.'

Expand full comment

I will bite that bullet and say that I believe that if you cloned Von Neumann and avoided malnutrition and head injuries the clone would prove unusually talented at whatever field they decided to pursue. I also think they'd test as a "genius" on any standard battery (and clear the threshold by a lot, the threshold is set very low). I don't think they'd be guaranteed to change the world.

I do think that a genuinely good education - lots of private tuition - helps, and luck always plays a huge part in everything, so our clone-Neumann given a very average American upbringing may merely become a very successful professional or small business owner.

Expand full comment

That seems like an isolated demand.

There is no reason to think Jewish intelligence is different from any other intelligence.

If it is intelligence is highly heritable for American twins, Swedish conscripts or Brazilian babies fed on formula milk, it would require a leap to assume the same thing isn't true to some degree for Jews.

Expand full comment

I recall that someone generated a dataset showing decent evidence for selection effects in the form of banking (i.e., lending at interest) being restricted to Jews in Europe and success in banking (as a proxy for higher IQ) being a predictor of higher fertility. I'll see if I can run it down later.

Expand full comment
deletedJul 13, 2022·edited Jul 13, 2022
Comment deleted
Expand full comment
Jul 13, 2022·edited Jul 13, 2022

https://www.metaculus.com/questions/9492/israeli-embryo-selection-for-intelligence/

https://www.metaculus.com/questions/9524/israeli-first-10-on-embryo-selection-for-iq/

Israel is 40% Ashkenazi, but the 40% is likely increasing due to the high TFR of Haredim. Perhaps we'll see Budapest-on-the-Mediterranean in our lifetimes (https://www.metaculus.com/questions/9785/10-embryo-selection-for-iq-when/).

Haredi communities abroad are likely to also adopt embryo selection, Judaism is in favor of IVF (there are some links in those Metaculus questions). Haredi are insular and have high TFR.

Expand full comment

The haredi may be smart but its not necessarily going to be evident if they stay so strictly religious.

https://m.jpost.com/israel-news/israel-under-threat-by-lack-of-basic-education-for-ultra-orthodox-607396/amp

Expand full comment
Jul 13, 2022·edited Jul 13, 2022

Right. I expect more attrition in the future though, at least in Israel. The trend there is in that direction.

Expand full comment

That's right. You needed a flake-ass liberal "woke" community like Budapest or Prague to get that kind of intellectual performance. The right wing reaction was, as one might expect, quite brutal and effective.

Expand full comment

I'd prefer a world where secular intellectuals have children and where conservative religious movements aren't commonplace, but a society with religious groups having children some of whom become apostate and fuel secular society is at least theoretically capable of a a stable equilibrium

Expand full comment
Jul 24, 2022·edited Jul 24, 2022

I agree. I would like 25% income tax per child or something like that, which incentivizes everyone especially educated and affluent people to have more kids, plus state-subsidized embryo selection.

In any case, Israel has a high TFR even among seculars (in the developed world it's really only Israel, France, Czechia that have this), and the Haredi https://www.metaculus.com/questions/7513/-israeli-population-that-is-haredi-in-2050/ https://www.metaculus.com/questions/7571/haredi-share-of-israel-at-peak/ won't take over.

Expand full comment
Jul 13, 2022·edited Jul 13, 2022

There's a common myth among some HBD people (often anti-Semitic ones) that Israeli Ashkenazim are less intelligent due to selective migration. Israel has low PISA but this is due to Israelis leaving large numbers of questions blank and skipping them https://conference.iza.org/conference_files/edu_2021/ofek-shanny_y31696.pdf

- it's a low-trust society. They do much better when bribed, and if you look at Israeli psychometric tests it's much more encouraging.

Also, Israel does not iodize salt https://www.metaculus.com/questions/8360/israel-mandates-salt-iodization-by-2030/ but hopefully will.

I'm curious about other Ashkenazi groups. It seems Budapest Ashkenazi > US Ashkenazi. How do Israeli Ashkenazi rank? Subdivide into Yishuv and Holocaust survivors? How about Haredi? Are they significantly different from the seculars? Yekkes vs Ostjuden? The ones form the Former Soviet Union? I would expect a priori those to be higher IQ because the upper classes were less likely to emigrate. Jews from the FSU also perform better for cultural reasons (in general, Europe >> post-Sputnik US in sciences, especially controlling for demographics and the US having Jews and Asians), but likely it's also genetic reasons, the population that left for the US and Israel in the 70's and circa-1990 is more drawn from the Muscovite upper class than from Anatevka.

It's an interesting question people should study further.

Expand full comment
author

I can't find it in a casual skim of that paper - why do Israelis skip so many questions on PISA?

Expand full comment
Jul 13, 2022·edited Jul 13, 2022

For the fact that they do: look at Page 44 of that paper. They skip and guess a lot of questions, even compared to their score (Jewish Israelis skip/guess more questions than Hispanic Americans and almost as many as Black Americans), and look at Page 14 for the discussion of "low endurance" in Israel. In this paper the scores go up by almost a whole standard deviation on some test after incentives (not financial because those are illegal, just higher grades) are given. It's +1 SD for Israeli Arabs, +.7 SD for Israeli Jews, they cite another paper where it's about +1/3 SD for US students, about +0 for Shanghai students on PISA (this is with financial incentives).

See also Page 43. Israeli Jews and Arabs alike do better on the first part of the test, whereas in the US you see the minorities doing worse in the later parts.

Why? The paper says "low intrinsic motivation". My guess is it's a product of the local culture. Israelis are aggressive, have low social trust, and take shortcuts (I say this a someone who is a big fan of Israel and Israelis). I'll give a couple stories from my Israeli advisor. The first story is that he said he could not give a take-home exam in Israel, because everyone would cheat. The second is that a pollster in Israel asked people, how did you vote in the last election, and the results were *completely* different from that of the last election (lots of people said they vote Labor and actually vote Likud). PISA to some extent measures trust, Romania is also a big under-performer on the PISA and it is low social trust.

This is generally true of both Jewish and Arab Israelis, it's a general fact about the Middle East.

Israelis also don't trust international organizations, and if you give them a test that counts for nothing by some organization, my guess is a lot of them will just check random answers and go outside to play with friends or something.

I can send you some data on the Israeli psychometric tests - they do well, and I'm told it's harder than the SAT, since there is an actual incentive there.

Expand full comment

As an Israeli, who went through the educational system, I have another important point.

The educational system in Israel is dysfunctional, with all the sickness described in "The Case Against Education", but with a local flavor.

Teachers are not incentivized to teach well, and are not rewarded for investing in the pupils. The root cause of this is the teacher union, that insists on seniority based compensation. A starting teacher makes one of the lowest salaries in the Israeli economy, and Israeli kids are probably closer in their behavior to inner-city children rather than well-behaved western Europeans. In short, very hard and unrewarding work. Many, many young teachers leave after the first year. The survivors are basically cruising through at minimum effort. Those that can leave the system do - and thus there is a cooling effect on the proficiency of teachers, especially prominent in math teachers.

Bad math teachers lead to students bad at math, ergo low PISA test results.

Most mathematically inclined students that want to peruse a STEM degree realize mid-high school that they'll need to develop math skill independently, and take external courses or study online.

Expand full comment
Jul 24, 2022·edited Jul 24, 2022

Right. Israeli kids have high rates of truancy in addition to high rates of leaving questions blank on PISA. They don't give a shit. They have very little incentive to do well on any tests or get good grades in high school, since universities are easy to get into. They act like inner-city children and don't take K-12 seriously. I've heard stories of Israeli kids slapping teachers and so on. https://blogs.timesofisrael.com/the-israeli-school-system-is-broken-and-it-needs-to-be-changed-immediately/

Because the country is so small, it's not hard to take STEM courses at a university as a high schooler, they are all within driving distance.

The scientific achievements though are great, as you would expect given the demographics. It's a good case study in how K-12 education really doesn't matter much. ;)

Expand full comment

Low stakes tests like PISA probably differ from country to country on how hard the test-takers try.

Even high stakes tests can have motivation problems when low-scorers give up partway through and bubble in the rest of the way. That happened with the military's 105-page long AFQT in 1980. When they replaced it with an interactive computerized test that gave low scorers easier questions, they had less trouble with low scorers giving up.

Expand full comment

Funny thing is that in Israel it’s illegal to give monetary incentives so they had to give grade incentives. In America and China they have monetary incentives. America got a PISA-style tests of +1/3 SD with $$$, on IQ tests you get like half that by motivating people (makes sense, IQ tests are more engaging than the long PISA word problems), China got basically +0 on PISA from $$$. Israel got +.8 on some PISA-like test (+.7 for Jews and +1.0 for Arabs) with just grade incentives.

Expand full comment

Using such computer adaptive testing, one normally gives subjects items which they will have about 50% chance of getting right because that maximizes the item's informativeness. However, if the goal is both to minimize giving up AND measure the trait, one could have the system give subjects items they have an estimated 80% chance of getting right.

Expand full comment

Not a great idea, that will lead to long tests where people get bored. You need financial incentives.

Expand full comment

The problem with the Haredim is that they consider things like the arts and sciences to be distractions from the primary work of mankind on this earth, to read and study the word of god and carry on traditions. The Haredim were the other great reaction to Jewish emancipation in Europe. They may produce geniuses, but it is unlikely anyone outside the community would recognize them as such.

Expand full comment

If they value intelligence and have lots of kids and at least a few of those kids become secular, then they are going to produce (recognized) geniuses even if the practicing ones are too insular to do things that the secular world cares about.

Expand full comment

That does happen, but it's a pretty rough escape trajectory. Within the culture, education is minimal and knowledge about anything in the outside world - arts, sciences, culture - are excluded. Living a secular life isn't an obvious option. Even English literacy is rare since that is neither the language of the home or the language of religion. It's like being a refuge from North Korea except that North Koreans speak the same Korean as South Koreans which makes it easier.

Expand full comment

So basically, just wait for Israel and diaspora Jewish communities to do embryo selection, which will likely happen in the second half of this century, Judaism is good on bioethics. This, combined with the high TFR of Israeli Jews (both secular and not) and of Orthodox Jews in the diaspora, will give you your wish. :)

Expand full comment

That's interesting. What happens if you have a lot of smart kids in a low trust society with a bad educational system?

Expand full comment

Bond villains.

Expand full comment

Israel happens, lots of people just go to university early and learn stuff from the army.

Expand full comment

I honestly expect embryo selection earlier than that, since the high rate of recessive genetic disorders means that embryo selection for avoiding known diseases is already common in Israel

Expand full comment
Jul 24, 2022·edited Jul 24, 2022

Yeah, Israel has very advanced embryo selection via IVF, which everyone (including the Haredi) is on board with. They have advanced genetic screening, due to the high rate of genetic diseases in at least the Ashkenazi population. Israel doesn't have the "eugenics taboo" of the West, and Judaism is fine with embryo selection to enhance intelligence and health (it's less fine with enhancing physical appearance). Something like 10% of births in Israel already are through IVF and it's going up, because Israelis are still having ~3 kids but are having them later.

Shai Carmi did say that the Israeli health bureaucracy is cautious and is likely to wait for someone else to do some embryo selection for intelligence before subsidizing it on a large scale, though he agrees if it becomes common Israel is likely to be one of the first countries if not the first to do in on a large scale. The first adopters will be in the US or on some offshore havens, and then Israel will do it on a large scale.

He says polygenic embryo selection is currently not approved there but they will likely approve it for disease soon.

Expand full comment

"So, how would we create a new generation of Martians? Can I get an ACX grant to start a new charter city exclusively for smart Ashkenazi Jews? Maybe require all the citizens to be born via IVF for optimal embryo selection?"

Don't forget the oppression and murdering as the kids get older. Ala "The Boys from Brazil."

Expand full comment

Umm... Perhaps a quibble, but I believe Lewis Fry Richardson invented scientific meteorology. Cf. https://en.wikipedia.org/wiki/Lewis_Fry_Richardson#Weather_forecasting

Expand full comment

I purchased the book to read about von Neumann's contributions to meteorology in context, then was disappointed to find the barest passing mention in the intro and none in the body (if the index is accurate).

Not that "scientific meteorology" has a definition, but Vilhelm Bjerknes is generally credited for first placing meteorology into the context of applied physics.

Richardson had a great idea, that if you know the current state of the atmosphere and you know the physics equations that govern its behavior, you ought to be able to predict its future state. He tried a demo; it didn't work.

With computers in the 1950s it might just be possible to calculate a forecast and finish the calculation before the forecasted weather came to pass. This possibility of useful forecasts led von Neumann and dynamical meteorologists like Charney and Platzman to team up to try to do it. It was a group effort, two tricks being to understand why it was essential to start with approximate equations rather than exact ones, and how to compute the numerical solutions with minimal computer resources and without the algorithm going numerically unstable. While von Neumann was irreplaceable for solving the prodigious techical challenges, the meteorological parts were mostly up to the meteorologists.

Expand full comment

Hi, if you're looking for von Neumann's contributions to meteorology in context you would be better off reading a history of the subject. Paul Edward's "A Vast Machine: Computer Models, Climate Data, and the Politics of Global Warming" covers it.

Expand full comment
Jul 13, 2022·edited Jul 13, 2022

Fantastic article. I love John von Neumann. Also, congrats on explicitly coming out as HBD. It's worth mentioning though that most Jews in Budapest survived the war though plenty were killed (see https://encyclopedia.ushmm.org/content/en/article/budapest "more than 100,000" survived compared to ~200,000 at the beginning). It's not just the Holocaust, though of course the Holocaust was awful and had a big effect, but also cities being IQ shredders.

I don't know TFR data for Budapest, but I know BirthGauge has some tweets about Vienna having a sub-1.0 TFR in the early 20th century.

Expand full comment

It would be better to write out your acronyms.

Expand full comment

HBD = human biodiversity

IQ = intelligence quotient

TFR = total fertility rate

Expand full comment

Guess the ban on HBD discussion has ended.

Expand full comment

"cities being IQ shredders"

This is the opposite of conventional wisdom, as I understand it. Could you expand on it a little more please?

Expand full comment

Cities gather high IQ people from the surrounding countryside and towns. But city people gather to work, not primarily to breed. So I wouldn't be surprised if cities are IQ shredders on a genetic level.

Expand full comment

The famous example is Singapore, which purportedly attracts many smart people from elsewhere in South-East Asia and induces them to prioritise their career and not having children (this is obviously a shorthand wavey gesture at the exact mechanics of what depresses fertility rate, which are likely to be complex). Singapore's fertility rate is ~0.78, so if you're skimming the top 10% of people in SEA and subjecting them to that sort of pressure it's a form of truncation selection.

I haven't seen conventional wisdom that swings the other way re: inheritance. Maybe you're thinking of a general Geoffrey West Scale type argument that shows how cities increase productivity, # of entrepreneurs, ..., (and waste etc.)?

Expand full comment

Yes that's exactly the argument I'm thinking of!

My understanding was that cities added value by cramming lots of highly productive people in one place, which increased the odds of some chance meeting producing a viable business, or patent, or book deal, or whatever - because you had many more chance meetings per unit time (and also a second-order effect where it therefore becomes more expensive to live in a city which adds selection pressure on the kind of person who can afford to move there and hence your average quality-per-chance-encounter is likely to be higher).

I assumed that also extended to a chance meeting meaning you fell in love and had lots of kids, but it sounds like you're suggesting cities independently assert a separate effect depressing fertility. It still sounds a bit to me like the effect direction is unclear though - having fewer children (average) with more productive partners (on average) doesn't seem to me to round off to 'IQ shredder'

Expand full comment

It's quite obviously an IQ shredder at the population level:

Scenario 1, a country with no cities. Average IQ 100, TFR 2.0, average IQ of next generation 100.

Scenario 2, a country with a city with half the population. Average IQ of the city: 110. Average IQ of the countryside: 90. TFR of the city: 1.0, TFR of the countryside: 3.0. Assuming 60% heritability and 40% version to the mean, average IQ of the next generation is 106*0.25+94*0.75 = 97.

Numbers exaggerated to prove the point, but if higher IQ people have fewer people when they move to the city, them having those kids with other higher IQ people makes the problem worse not better.*

*Although it does also slightly increase population variance, so the tiny population of very-high-IQ people declines slower than it normally would if the population uniformly reduced its IQ.

Expand full comment
Jul 13, 2022·edited Jul 13, 2022

Might also be worth noting that if there are some wide variety of traits which increase 'intelligence' in various domains, a cosmopolitan city with collisions between the most intelligent members of diverse populations may also substantially increase the *magnitude* of the intellectual elite's advantage. I.e. if Hungarian Jews have trait Y which increases intelligence, and separately Chinese people have trait Z, the fact that the city depresses their fertility does not change the fact that it is the only environment in which a person possessing Y+Z is likely to exist.

Expand full comment

It looks like the human genome doesn't work like that based on the evidence so far (educational attainment is the best proxy for intelligence that we have polygenic scores for since it's easy to collect population-wide educational attainment data, and those scores seem to have components that are present in all populations with varying frequencies, not exclusively confined to certain continental regions), but you're right that that is hypothetically possible.*

*Another suggestion that it's unlikely is that history's catalogue of absolute-top-tier geniuses are not over-represented by people with cosmopolitan-unique ancestry combinations, AFAIK. Von Neumann was close to 100% Ashkenazi Jewish, Tesla was close to 100% Serbian, Newton was close to 100% English, etc

Expand full comment

Your analysis doesn't take into account the increased chance of far-from-mean offspring from the higher IQ baseline and the potential for a ratcheting effect from assortative mating. Perhaps the benefits of cities is that it provides a greater means for differential outcomes due to IQ and assortative mating, enabling an IQ ratcheting effect.

Expand full comment
Jul 14, 2022·edited Jul 14, 2022

The societal benefit from cities is pretty clearly the this-generation increase in productivity and technological invention that you get from concentrating talent.

They are still IQ shredders on a generation-on-generation view; even if the concentration of IQ resulted in an increase in +4 STDEV people in generation 2, that effect would fade away eventually due to regression to the mean, so by generation 5-10 the +4 STDEV (of the original average) population fraction would be lower than it would be otherwise (regression to the mean would be more powerful than the ratcheting effect).

Gut check 1: history's had a lot of million-plus person cities lasting for many generations. Of history's standout geniuses, how many were from lineages that had lived in cities for very long periods of time? Von Neumann's father was from a small city but then his grandfather and great-grandfather were from a small town called Ond; mother's side is 3 generations in Budapest, so mixed evidence. Tesla's family was quite rural. Newton's mother was born in "Market Overton in Rutland" and his father was born in "Colsterworth, South Kesteven District, Lincolnshire, England".

Gut check 2: the demographic that has the clearest evidence for an above-100 selection for IQ are Ashkenazi Jews. Enough Ashkenazi Jews migrated to NYC for us to be able to run a good experiment on whether that population concentrated in a mega-city for generations would result in a dazzling crop of geniuses by generation 5+. Signs point to no; Budapest's generation 1-2 outperformed NYC generation 5 because Budapest got the highest IQ Ashkenazi to start with (based on Scott's analysis).

Gut check 3: Some civilizations urbanized faster than others. The Roman megalopolis left no genetic trace in Italy (according to Razib Khan: https://razib.substack.com/p/they-came-they-saw-they-left-no-trace), and the long existence of Baghdad does not appear to have helped Iraq have more present-day geniuses than surrounding countries.

Expand full comment

This argument, together with the worldwide trend toward urbanization in recent decades, predicts we should see a gradual decline in measured IQ scores. At least we should not see an increase.

But the observed Flynn effect is exactly the opposite. How do you reconcile this data with your reasoning?

Expand full comment

Quite easily, given that IQ is genetics + environment (nutrition, mostly) and nutrition has improved dramatically over that time frame.

Two other key pieces of evidence that reinforce that:

(1) The Flynn effect stopped or even reversed in the rich world decades ago, right about when you'd expect gains from good nutrition to have reached 100% coverage

(2) Polygenic scores for educational attainment in Iceland, at least, show negative selection for the high-education traits over the last few generations. We haven't yet collected that data for more typical countries, but it fits with the expected pattern of declining (average genetic maximum potential for) intelligence.

Expand full comment

Do people ever leave Singapore with the money they've earned to start a family the way people leave San Francisco for, say, Reno? But where do they go?

Expand full comment

I'm not even sure that Singapore is such a big shredder nowadays, the TFR of Singapore is really low but so is the TFR of mainland Chinese and Chinese Malay, who I think are the biggest immigrant populations there.

Expand full comment

Singapore's TFR is 0.9, Mainland China's TFR is 1.7, Malaysia's TFR is 1.9.

Singapore is still quite obviously an IQ shredder.

Expand full comment

Mainland China is 1.3 and it's below Singapore-level in many cities like Shanghai, and Chinese-Malay TFR is also like 1.0.

Expand full comment

> the TFR of Singapore is really low but so is the TFR of mainland Chinese and Chinese Malay

This by no means precludes Singapore still being a big IQ shredder. It can keep skimming smart people from low TFR populations as long as these source populations are much larger than it is.

Expand full comment

Even if they survived, they might not have stayed in Budapest.

Expand full comment

Seeing that bit about Gábor Szegõ being Von Neumann's childhood math tutor reminded me of the fact that Dave Chalmers was apparently a childhood math tutor for Terence Tao.

https://fragments.consc.net/djc/2006/08/fields_medals.html

Expand full comment

Outstanding. I had never heard that before. But did Dave cry with joy for having such a pupil?

Expand full comment

There's also that photo of Erd\H{o}s and Tao in Adelaide. Erdos was visiting George and Esther Szekeres, Hungarian Jews from Budapest who moved to Adelaide.

Expand full comment

Intelligence isn't heritable, so I guess Von Neumann just had a favorable environment growing up.

Expand full comment

Identical-twin studies are are said to demonstrate that intelligence is about 50% heritable.

Expand full comment

Things aren't heritable or not-heritable absent context.

Expand full comment

All human children inherit intelligence from their parents.

Expand full comment

Neither is height or skin color then.

Expand full comment

Interesting replies to what I assume is sarcasm.

Expand full comment
Jul 14, 2022·edited Jul 14, 2022

I'll admit it actually took me a second to realize it was sarcastic — if you look around on e.g. Reddit, you'll find many people un-ironically endorsing this position or a close neighbor.

EDIT: I mean, even the comment directly below is suggesting that there can be no genetic component, because intermarriage or something.

Expand full comment

I mean, I know this is ironic but he was surrounded by a *remarkable* collection of teachers, tutors, and mentors. I suspect you'd get at least solid results out of quite a few basically intelligent people in that scenario.

Expand full comment

Your harping on this subject is tiresome, particularly when this post proves it isn't necessary here.

Expand full comment

It's funny that, when discussing what made for an unusually rich crop of émigré talent from Budapest and thereabouts, Scott spends so little time looking for explanations that have anything to do with the education system in Budapest in the first third of the twentieth century. (It is still good, to judge from my Hungarian friends - and in fact some of its strengths as far as mathematics is concerned have remained constant.)

The fact that a large proportion of people in that group were of Jewish ancestry (or partly of Jewish ancestry, etc.) may in fact be partly a distraction. It's not just that 25% of the population of Budapest fell into that category, but that the percentage was presumably much larger for the middle class. What access to higher education did the working class? And what interest did the nobility have in excelling in studies?

As for genetic explanations - that is hard to either prove or disprove. One thing, though: intermarriage was statistically negligible everywhere before WWI - yet we see it happening quite a lot whenever somebody puts together a list of Great Jews - born before or after WWI, Hungarian or otherwise. It is not just that the lists include people whom Jews would not consider to be Jews if they were not famous, and that comparisons with proportions in the general population are flawed. It is also the case that this is fun to try to make fit with naïve versions of the genetic explanation ("Jews are smart"). Are we dealing with some magical substance that increases in power when diluted?

More sophisticated versions of the same hypothesis may survive this test, but they lack that nice primitive appeal.

Expand full comment

I remember finding out that Olivia Newton-John was Max Born's granddaughter.

Expand full comment

Joan Baez is cousin to physicist John Baez.

The lead singer of the Eels is Hugh Everetts son.

Expand full comment

Joan Baez's Mexican dad was a fine physicist too, the co-inventor of the X-ray reflection microscope:

https://en.wikipedia.org/wiki/Albert_Baez

Expand full comment

I believe this theme of "Jews are smart" has come up before on this blog. Personally I would have liked to have learned more unique insights about the man himself rather than his general background.

Expand full comment

There are some comments about this in there, in the quotes from MacRae's bio. But of course it's all hypothetical. Nobody knows just how or why a genius emerges.

Expand full comment

Agreed.

Expand full comment

Can't believe it! Scott wrote this! Was going to give this my top vote in the contest.

Expand full comment

It's telling isn't it - Scott's reviews usually have a little bit of magic dust sprinkled over them.

Expand full comment

Reminder that the contest essays consistently come on Friday and always have the disclaimer attached to them. (I remember last year making the same mistake when Scott posted his 1001 Nights review)

Expand full comment

This would be a great reminder if I could also consistently remember what day of the week it is :)

And, same on the 1k1

Expand full comment

>It's funny that, when discussing what made for an unusually rich crop of émigré talent from Budapest and thereabouts, Scott spends so little time looking for explanations that have anything to do with the education system in Budapest in the first third of the twentieth century. (It is still good, to judge from my Hungarian friends - and in fact some of its strengths as far as mathematics is concerned have remained constant.)

Scott addressed this question at much more length years ago in this essay-

https://slatestarcodex.com/2017/05/26/the-atomic-bomb-considered-as-hungarian-high-school-science-fair-project/

Expand full comment
Jul 13, 2022·edited Jul 13, 2022

Yes, I read it back then. It is still very cursory. The conclusion is not that "education does nothin'" (he doesn't really learn or tell us much, or nearly anything, about the education system) but simply that it does not produce child prodigies. Fair enough. Of course most child prodigies anywhere go undiscovered, or go nowhere - and very few of them get the chance to learn Classical Greek!

Clearly there are environmental factors at play here outside the _formal_ education system - though part of them can be due to its effect on the parents' generation.

(a) I, for one, would like to know more about what was going in Budapest in the first third of the twentieth century.

(b) As he is forced to conclude in the cited post, it can't reduce to genetics *if seen at the very coarse level of the "ethnic group"* - similar concentrations, and higher ones, have been achieved since then.

Expand full comment

The evidence suggests that in Budapest between 1900 and 1930, in at least some schools and the social environment around those schools (schuls, churches, playgrounds, maybe popular entertainment, street banter, jokes in newspapers, gossip) academic achievement was not disparaged as nerdy/unattractive/low-status, but just the opposite. It would not be surprising if in places where this held true, intelligent kids then tended to develop their intellectual strengths instead of prioritizing things that society incentivises elsewhere. Perhaps it's difficult to achieve such a high density so most such instances are more local: home schooled relatives from a large extended family in the UK, some cousins in a large Indian city who have a famous poet as an uncle and a professor as a role model, a small rural Iowa settlement with a school employing an unusually bookish teacher, and similar stories Scott has written about. At least some of this hypothesis should be testable.

Expand full comment

Right - but testable how?

Expand full comment
Jul 13, 2022·edited Jul 13, 2022

Education does nothin'. I don't think a genetic explanation is at all hard to *evidence*; "proof" is a word often used for "God [or IQ] of the gaps"-style reaching.

Scott has, in fact, looked at this issue a lot; see below link.

Expand full comment

Hybrid vigor is a thing.

Expand full comment

Well, obviously (though we are not dealing with population isolates here).

Note, however, how we are moving away from naïve hypotheses already (of the "forbidden thoughts" variety).

Expand full comment

Hybrid vigor is a thing but not all hybrid pairs have vigor. Jury's still out on humans I think, although we do have suggestive positive evidence from things like <the iconic Hawaiian look from the early 20th century being half Hawaiian half Japanese, or so something I read claimed, maybe James A. Michener's Hawaii?>

Expand full comment

Hybrid vigor reduces the effect of problematic genes, but Ashkenazic Jews were and, to a large extent, still genetically isolated from the European population. If you look at the Martians, most of them had two Jewish parents, so both were from within that isolated group.

My guess is that it was because male Jews had to learn to read the Torah. No Jewish woman was going to marry a man who couldn't read, and a lot of Jewish men would think twice before marrying a woman so pig ignorant that she couldn't read enough to tutor the younger boys.

Expand full comment

The Ashkenazic population actually had enormous contributions from surrounding groups (mostly women) in its founding stage. What you had was little inflow in 1500-1850, say.

Expand full comment

If the educational system played a large role, shouldn't we expect to see lots of non-Jews rising to the highest echelons of math and science? AFAIK, all great Hungarian minds of the time were Jewish.

Expand full comment
Jul 13, 2022·edited Jul 13, 2022

If you define "Jewish" very broadly (including what most Jews would think of as non-Jews: children of converts, people with some Jewish ancestry, etc.)., and choose who gets labelled "great", sure.

At any rate, as said before, the middle class in the capital city was heavily Jewish (under a broad definition of the term), and that's what is relevant; even secondary school wasn't a given at the time for children of the working class, far from it (... and the upper class would have had no incentive to try very hard). You will probably get a very clear majority if you narrow it down to the Bildungsburgerturm, i.e., the relatively comfortable subsector of the middle-class that owed its position to the education system and had a close symbiotic relationship with it. Class habitus runs in families, as do resources, obviously.

Expand full comment
Jul 14, 2022·edited Jul 14, 2022

Are you saying that children of converts are well-represented in the usual examples of disproportionate Jewish achievement, of a magnitude such that positing a genetic component becomes unnecessary?

I would expect that, both in the more general question (representation in Nobels/20th-century prodigies/advanced degrees, etc) and in this particular case, there would be a negligible number of converts and that case can be safely ignored.

Too, I am curious as to how many examples there are of someone with, say, a single Jewish grandfather being counted as Jewish, in this context — I had assumed most "famously brainy" Jews are more Jewish than not, by ancestry; it makes sense to me — in general (and *especially* in the case of Jews) I expect far more endogamy than exogamy — but I'd be interested if you have evidence that's not the case.

(More specific still, I didn't think any of the "Martians" were converts or children of converts -- but I haven't really checked all that hard, either. Or, uh, at all. Still, there aren't very many converts to Judaism at any time or place, AFAIK.)

Expand full comment

Converts *from* Judaism. Who would then not be counted as Jews (if not famous).

Expand full comment

Take Aage Bohr, for instance (and just off the top of my head; he can't be the only Nobel Prize winner in that particular pigeonhole).

Expand full comment

I’ve made this comment before on other posts, but the Jews in Hungary, particularly in Budapest, had a unique situation. Back in 1848, Hungarian Jews were seen as sympathetic to the wave of failed revolutions in that year. So the state decided to punish them by levying an onerous school tax. This was supposed to suck up any resources the Jewish community might put toward another revolution or other political ambitions. Jews were still largely segregated from everyone else, so for 50 years before 1900 Jews were essentially forced to build themselves the best schools money could buy.

Add to this the unwillingness of Hungarian landed gentry to sully their hands with any kind of work. The Jews stepped up to take on the industrialization and modernization of the country. And, for a time, a lot of the wealth they earned went back into schools. A smart Jewish boy born in the latter half of the 19th century had a unique opportunity for world-class schooling plus a worthy outlet for his ambitions. Not to mention the aforementioned feeling that all of this was rather precarious…

Expand full comment

There are two things that make me skeptic about this line of “education technology” explanation.

First, it sounds a lot like an argument for protectionism in education. “Isolate you cormmunitity from outside influence/competition (also, strip it from funding) and you’ll get a great educational system.” Sounds fishy.

Second, this would imply that (i) Hungarian-Jews managed to set up an educational system so extraordinary that it would reliably produce Von Neumann level geniuses. (ii) the knowledge of how to set up that system was completely lost. And (iii), that despite decades of effort and trillions of dollars spent in education world wide, we have not been able to re-discover the recipe for this system.

The war might help you explain (ii), but still (i) and (iii) make me doubt that such an educational system ever existed, or that it could ever exist.

Expand full comment

I think it’s more complex than this. Smart Jewish kids in Budapest didn’t just have access to unusually good educational resources, but the surrounding society made it possible for brilliant people to rack up unusually meaningful and measurable achievements. The Budapest Jews might have been no more or less extraordinary than their circumstances.

In surrounding countries, a the potential for growth was probably spread more evenly among upper-middle-class Christians and lower gentry. But in Hungary the social prohibitions against upper class Christians working in fields like industry and finance were unusually strong. At the same time that Jews were making economic progress due to those opportunities, they were gaining political enfranchisement. I can’t say for sure, but there’s nothing like the fervor of the newly-upwardly mobile to propel the next generation to great things. You can’t engineer the conditions that might have led to a spate of geniuses in early 20th century Budapest, which is why we haven’t.

I forgot to mention my source for the forced spending on Jewish education was Paul Lendvai’s “The Hungarians”:

https://www.amazon.com/Hungarians-Thousand-Years-Victory-Defeat/dp/0691200270/ref=asc_df_0691200270/?tag=hyprod-20&linkCode=df0&hvadid=509234950067&hvpos=&hvnetw=g&hvrand=11823157615448090124&hvpone=&hvptwo=&hvqmt=&hvdev=m&hvdvcmdl=&hvlocint=&hvlocphy=9007216&hvtargid=pla-1258715692963&psc=1

Expand full comment

This is interesting. Your account of the environment that Budapest Jews were faced with (which I agree with), is basically the one that we need to make the genetic explanation work. The smarter Jews, who presumably were already smart on average, faced little competition in banking and industry jobs which allowed them to amass large fortunes and have lots of children, e.g. 10+ children. The smartest of these 10+ children, who are the offspring of smart parents, would go off to profitable careers in business and industry, amass large fortunes, and have 10+ children of their own. In other words, the environment put strong selection pressures for intelligence on Jews. Repeat this process four or five times and you end up with "The Martians."

So I guess our disagreement is not really about what the environment was like, but about the causal mechanism through which said environment produced ultra-smart people. My model of the situation is: {environment} --> {genetic selection for intelligence} --> {Martians}. Whereas yours looks something like: {environment} --> {Martians}.

Assuming this is an accurate representation of the disagreement, I'd go back to the point I made earlier that we know of no environmental intervention that can reliably and sustainably raise intelligence (outside obvious things like nutrition and not hitting people in the head with a bat). The sad truth seems to be (as Fredie deBoer nicely summarizes here: https://freddiedeboer.substack.com/p/education-doesnt-work-20) that people come with a maximum potential intelligence, in the same way that they come with a maximum potential height, and provided that some basic conditions are met (again, nutrition and not hitting with a bat), no environmental interventions that we know of can increase people's intelligence past their maximum potential intelligence.

Given this evidence, I'm inclined to think that the Martians simply had a higher maximum potential intelligence which was the result of genetic selection. The alternative is to assume that Hungarian Jews accidentally landed on an environmental-intervention sweet spot capable of consistently producing world class geniuses, but so sensitive that any departure from it yields people that are only kind of smart.

Expand full comment

Given that one of our key signals that all these guys were unusually brilliant is the fact that they met on the Manhattan Project, I keep wondering if the rarity of combined factors that create people smart enough to invent weapons capable of wiping out humanity is a bug or a feature.

Expand full comment

Thanks.

Do we have much evidence that Budapest Jews outperformed Vienna Jews, other than at the very highest levels of Manhattan Project fields (which could be due to the small sample size)? How about Jews in Prague, the third city of the Austro-Hungarian Empire?

On a per capita basis, how did German Jews perform relative to Austro-Hungarian Jews?

Expand full comment

Scott addresses this issue in an old post about Hungarian-Jews over-achievement (https://slatestarcodex.com/2017/05/26/the-atomic-bomb-considered-as-hungarian-high-school-science-fair-project/), and he briefly mentions it here. He argues that "when we hear “there were X Nobel Prize winning German physicists in the early 1900s”, it sounds only mildly impressive. But when we hear “there were X Nobel Prize winning physicists from Budapest in the early 1900s”, it sounds kind of shocking. But the denominator isn’t the number of Germans vs. Hungarians, it’s the number of German Jews vs. Hungarian Jews, which is about the same."

All this is to say that it is probably not the Hungarian-Jew (HJ) education system that is doing most of the work here, and that there must be some other factor that explains the HJ's "overachievingness" which is also shared by German-Jews. Here, the genetic explanations seems to fit the bill.

Expand full comment

Prague also produced a lot of smart Jews. Look at Hedy Lamarr. She had a theatrical education but taught herself enough about radio and electronics to do some serious inventing. She had a patent on what later became known as TDMA. She wasn't a science major, but she grew up in a freewheeling intellectual environment.

Expand full comment

Jews don't get to consider who is a Jew and who is not a Jew. That's something the antisemites do. It's not like there's a Jewish pope who will excommunicate you in time to save you from a pogrom. It's like this for the Blacks in the US who were classified by racial laws passed by racists. There's no Black pope either.

Expand full comment

Jews actually are very particular about membership (when it comes to people who are not famous). It goes strictly by the maternal line, and attempts to make that rule more flexible have met with enormous resistance. Of course someone can be non-Jewish according to the Jews and still be a target of antisemitism.

(There *is* excommunication (not excommunion), just not by a central authority.)

Expand full comment

Reading this, I was struck by the fact that Albert Einstein is the person who has come to signify genius to the general public. I think he's a very good candidate - there are few historical thinkers who seem so consistently right and insightful about murky and controversial matters (Charles Darwin and Alan Turing are the other two that come to mind). (I think it's also notable that most prominent scientist have philosophical inclinations that would be considered empiricist, while Einstein is one of the very few post-Enlightenment figures that could naturally be thought of as rationalist, in the philosophical sense.)

But Von Neumann seems to have been a different thing altogether. This review doesn't even mention Von Neumann's important work in set theory. (If you've ever learned that the ordinal 0 is the empty set, that 1 is the set {0}, that 2 is the set {0, {0}}, ... that omega is the set {0, {0}, {0,{0}}, ...}, and so on - well, that's Von Neumann's definition, which is much, much easier to work with than Cantor's definition.) He seems to have been incredibly sharp and done important work across many disciplines. But he does seem to have this psychopathic/sociopathic streak, at all levels from his carelessness at driving to his naive game theory about nuclear war. So it's probably for the best that Einstein is the public face of scientific genius.

(It doesn't sound like Von Neumann was ever as casually cruel as Isaac Newton though.)

Expand full comment

Which millions did Isaac Newton propose to murder? The worst I know about him was how he secretly wrote the Royal Society's report supposedly settling the Leibniz-Newton priority dispute [sic].

Von Neumann was a clearly first-rate mathematician, but it's not like he was Hilbert, or for that matter Ramanujan. In ergodic theory, as in computing, he benefitted from being an early entrant in the field.

Expand full comment

Good point about millions. I was thinking of Newton's personal cruelty to people he knew, like Hooke, and to various criminals and rivals when he was in charge of the mint. Not actually as impactful as a nuclear strike on the Soviet Union, though revealing a different sort of character flaw.

It's true that within mathematics, Von Neumann was no Hilbert. But he's clearly much more influential (especially broadly) than Ramanujan. And while it's possible that Hilbert was also as important as Von Neumann in physics, Von Neumann also has direct work in economics and computer science that is incredibly influential there. There aren't many 20th century figures with as much directly influential work in as many different academic fields as Von Neumann. (Nor even as many sub-fields of math as Hilbert - though Von Neumann is close.)

Expand full comment

I was bringing up Ramanujan as an example of "out of the blue" originality and depth, not in terms of influence; Hardy was quite right about both. Ramanujan has certainly been more influential than von Neumann in number theory, though :).

Again, von Neumann was an insightful early entrant in many fields. Of course von Neumann's Ergodic theorem is a basic result, but there's no point in pretending that it is difficult. Of course von Neumann was a founding figure of game theory, but, again, no mathematical difficulty at that stage. Obviously it is important to think about Hilbert spaces abstractly, but my (possibly deeply ignorant) impression is that von Neumann's contribution there was mainly that - axiomatization - and then of course he could use the basic theory, cleaned up by him, extremely well.

I'm sure some of his work can be classified as genuinely hard problem-solving - it is just that that part of his output seems to have a way of avoiding my path. Of course that may in part be a consequence of my ignorance, and again, difficulty is not the only measure of achievement in mathematics.

Expand full comment

I think it's useful to compare Von Neumann's work to that of someone like Kolmogorov or Tarski. It's all the stuff that is important and foundational to a field, and seems in some sense straightforward (though mainly with the benefit of retrospect). It's not as original and deep and creative as the work of people like Gödel or Turing, but it's still extremely foundational, and there's just so *much* of it.

Expand full comment
Jul 13, 2022·edited Jul 13, 2022

I think it may also be easy, with hindsight, to underestimate the level of intellect and originality necessary to *produce* an insight. Most of the mathematics I've learned has seemed obvious — once I've learned it.

I'm super smart, I acknowledge, but I have some small doubt that *even I* could have produced all of the knowledge in a modern university's mathematics department...

*Or could I have?*

No. No, probably not. Right? Unless...

Expand full comment

Then you haven't learned enough mathematics!

But yes, contributions that are "obvious in hindsight" (axiomatizing Hilbert spaces, etc.) are also valuable.

Expand full comment

That's probably fair, though I'd suspect there's a bit more depth and less breadth in Tarski or Kolmogorov.

Expand full comment

“Most mathematicians know one method. For example, Norbert Wiener had mastered Fourier transforms. Some mathematicians have mastered two methods and might really impress someone who knows only one of them. John von Neumann had mastered three methods: 1) A facility for the symbolic manipulation of linear operators, 2) An intuitive feeling for the logical structure of any new mathematical theory; and 3) An intuitive feeling for the combinatorial superstructure of new theories.” - Ulam

I suppose in some sense, that does explain why others said his work wasn't as deep/penetrating as e.g. Einsteins. Most of his tricks were making crisp what he already knew, not in finding something new. Still, that doesn't mean he didn't have a deep understanding of the subject.

Makes me wonder if he would have been a good philosopher. Looking briefly at his opinions on QM interpretations (in a SEP article), they seem much more sensible than e.g. Wigner or Heisenberg. But that's maybe 1 bit of info.

Expand full comment

Strange that things so apparently easy weren't developed until he came along, and that he did this so many times.

Expand full comment

If Von Neumann had a drive to look for new fields where fundamental insights were possible, that's interesting, too.

Expand full comment

Right.

While those who solve hard problems deserve fame, there is much to be said as well for those who discover relatively easy problems that people hadn't noticed yet were even problems.

For example, Francis Galton probably wasn't a Gauss-level genius, but he was extremely productive into old age in a variety of fields, such as statistics, that had, for some reason, tended to attract less talent than had, say, mathematical physics. For example, Galton came up with the correlation coefficient in his mid-60s and the concept of "the wisdom of crowds" (the average of a lot of guesses is more likely to be correct than a single guess) at age 85 in 1907, two years after Einstein's theory of special relativity. Einstein had a whole lot more brainpower than Galton in the 1900s, but Galton was still coming up with new ideas because he was attuned to asking and then trying to answer fairly easy questions that, for whatever reason, nobody else had taken on before.

Somewhat like Galton, Von Neumann tended to take on fairly new fields, harder ones than Galton of course, and come up with solid first steps rather than ultimate insights: e.g., the Von Neumann computing architecture wasn't a genius breakthrough, but it was a lucid description of a workable system that enabled the American computer industry to get up and running fast.

Similarly, almost simultaneously, Von Neumann launched game theory in the mid-1940s by publishing a book on this novel topic that, remarkably, became a New York Times bestseller during WWII. Soon, a whole bunch of smart folks were working on game theory, including applying it to nuclear war strategy to come up with less disastrous ideas than Von Neumann had personally come up with.

Expand full comment

If you do not think von Neumann's work had particular depth I would suggest you read through his collected works. He had plenty of it, and while in pure mathematics he might not have reached the level of insight of someone like Grothendieck, he still had enough to put him in the top tier of mathematicians without doubt, especially when you consider he only spent about 15 years of his life on pure mathematics.

I can't say I understand why people think von Neumann wanted to kill millions either, his viewpoint was that before the USSR developed their own nuclear weapons the US should use theirs to strike the USSR military and disable it. He went to math conferences in Moscow and I am sure he had plenty of friends there, so I have never understood why some people want to make him out to be some genocidal mad man who wanted to kill people.

Expand full comment

Soviet propaganda, I suppose -- obviously, only an evil madman would want to attack the worker's paradise under the leadership of wise and benevolent comrade Stalin.

Expand full comment

Preemptively nuking the soviets strikes me as entirely rational for the time. Almost impossible to fairly judge this without being blinded by hindsight.

Expand full comment
Jul 14, 2022·edited Jul 14, 2022

This was my thought. If I say on Tuesday "There's a 70% chance of rain on Wednesday", but then it doesn't rain, you wouldn't necessarily say "Well, guess you were wrong, idiot".

Similarly, if Von Neumann says "There's a 70% the soviets nuke us if we don't nuke them first", then looking at all the close calls we experienced, I'm not sure I'd say that was an unreasonable estimate. Maybe we just got lucky?

Expand full comment

I mean Jesus was woke to the long term consequences on human flourishing over time of choosing to play zero, and negative, sum minimax games 2000 years ago

Expand full comment

And look how that turned out for him.

Expand full comment

Newton did have people tortured when he was in charge of the mint.

Expand full comment

I think he holds the record for most forgers executed as Master of the Royal Mint, actually, thanks to his innovations in the design of coins

Expand full comment

For what it's worth, the people who are most impressed with von Neumann's accomplishments seem to be rationalists rather than people in his field(s). The physicists I know -- and I've asked directly about von Neumann -- generally feel that Einstein deserves his reputation as the "best" physicist, and the mathematicians I know usually point to one of the greats like Euler or Gauss.

I think this is true of Ramanujan as well. Really appealing to people who like a good Ubermensch story, less so to those only judging their body of work. (Not to diminish either of them, of course -- even getting to the point where most experts have an opinion on your body of work is an achievement most people will never earn.)

Expand full comment

I would like to think that number theorists' opinion of Ramanujan is fairly objective :). It largely agrees with Hardy's: (a) amazingly talented man, (b) the pity - who knows what could have been, (c) he influenced the field nevertheless (partly through Hardy and Littlewood - he had a hand in the development of the circle method, for instance). Of course some of what seemed mysterious to Hardy really came from Ramanujan's having taught himself, and partly rediscovered, the theory of modular forms, to a point well beyond where Hardy had got. That adds to (b): Ramanujan should really have met Hecke as well.

Expand full comment

I think in any particular field he touched, Von Neumann is seen as a very significant early figure, but not the biggest figure. I think this is a really interesting category of person. (The only other one that comes to mind is Dana Scott, who proved some very important results in many different subfields of mathematical logic, despite not being as influential in any one of them as superstars within that particular subfield. I just checked his wikipedia page and saw that he has a similar status in certain areas of theoretical computer science as in mathematical logic: https://en.wikipedia.org/wiki/Dana_Scott .)

Expand full comment

I'll second this. Professional physicist here, and I would rate Von Neumann as important, but certainly not among the top ten most important physicists in the 20th century. Maybe top fifty. My impression (on which I am happy to be corrected) is that the same is true in pretty much all the fields he touched*. Of course, being `top fifty in the century' in half a dozen distinct fields is nothing to sneeze at! But it's very different to Einstein, who was brought up upthread, who was indisputably the #1 most important physicist of the 20th century, possibly ever.

* Possibly in CS he would be top ten? Although I would guess still clearly behind Turing and Shannon?

Expand full comment

Von Neumann's real genius was in putting the computer into the public domain rather than letting Eckert & Mauchly patent it.

Expand full comment

The book actually quotes Mandelbrot on this (from in chapter 8):

'Von Neumann, when I was there at Princeton, was under extreme pressure from mathematicians, who were despising him for no longer being a mathematician; by the physicists, who were despising him for never having been a real physicist; and by everybody for having brought to Princeton this collection of low-class individuals called "programmers".'

(The source given is this interview: https://youtu.be/U9kw6Reml6s?t=185 )

This may also contribute to the subsequent view of his accomplishments.

Expand full comment

In one of Dyson's books (Turing's Cathedral?) Freeman Dyson is quoted on this too saying the faculty at the Institute for Advanced Studies was full of snobs who disliked having lower class people like engineers work there. Similarly some of von Neumann's friends at the end of this MAA documentary on him talk about him spreading his talents "too thin" https://www.youtube.com/watch?v=vQp70uqsBV4

Expand full comment

Some of the people who gave quotes about how impressive von Neumann was were physicists, great ones in fact, Bethe and Wigner for example, both of whom won Nobel Prizes. Of course they were talking about intelligence in general, not as physicists, von Neumann wasn't really a physicist in the sense he mathematized things, I believe Wolfgang Pauli had the quote where he told von Neumann he would be a great physicist if physics was about proofs. Nevertheless most professionals won't really give any strict ranking of who was the greatest and so on because it's hard to give an objective account of different physicists from different times. How would you compare Newton and Einstein for example?

Expand full comment

Well...Einstein was the most important physicist of the 20th century, possibly ever. Is there a field in which the same is true of Von Neumann? My guess would be CS is the most likely candidate, although even there presumably Turing and Shannon would rank higher?

That said, quantity has a quality all of its own, and being a major player in half a dozen fields is pretty remarkable! But if it is indeed true that there is no one field where Von Neumann is clearly THE key figure, then I would say that having Einstein as the `face of genius' instead of Von Neumann seems appropriate.

Expand full comment

Interesting, I'd specifically prefer a major contributor to several different fields over the single biggest contributor to one for the "face of genius". It feels fairer!

Expand full comment
Jul 13, 2022·edited Jul 13, 2022

"Einstein was the most important physicist of the 20th century"

I think such statements are very much up to taste. Stripped of publicity, I could easily argue for Max Planck to be the most important physicist of the 20th century. Or Niels Bohr? Pauli? Dirac? Feynman?

For computer science, one can easily argue that the von Neumann architecture is the basis of every computer and thus he is the most important computer scientist. Or one can argue as easily for someone else.

I would also say that's it is a bit arbitrary what we call a "field". Von Neumann is THE key figure in game theory. Is that a field?

Expand full comment

No, sorry, this is not debatable. Between his contributions to relativity (special and general), quantum mechanics (photoelectric effect, EPR), and statistical physics, Einstein is clearly #1. There is no plausible case whereby Planck or Bohr or Pauli or Dirac is as important as Einstein. As for Feynman, stripped of publicity it is doubtful if he is even top ten (in the 20th century).

I did say CS was probably the best case for Von Neumann to be the key figure, but even there I think Turing and Shannon are more influential? I would appreciate a perspective from a computer scientist though.

Game theory feels too narrow to be a field (more of a subfield), and even there, isn't John Nash more important than Von Neumann?

Expand full comment

I am CS researcher. But as you may have guessed from my previous post, I have no strong opinion on whether von Neumann was more or less influential than Turing or Shannon. I think the question is too ill-defined.

Expand full comment

I’ll take your word for it.

My instinctive response was colored by a sense that Von Neumann is vastly overhyped by internet rationalists, but my assessment of him is mainly based on his contributions to physics and it could be that just wasn’t his strongest suit.

Expand full comment

Well, in computer science he *is* considered one of the founding fathers, that is not just internet hype. The von Neumann architecture is of course the one big thing that got his name. Every computer that has been built in the last 70 years uses the von Neumann architecture. But there are other things. Mergesort, one of the really fundamental algorithms, is by von Neumann.

If you force people to decide for a single name, then I would assume that the most frequent name would be Turing. But I do think that some would choose von Neumann, or Shannon. It also depends a lot on the background. People with electrical engineering background in CS might choose Shannon more often. And people with other backgrounds have other heroes.

Expand full comment
Jul 13, 2022·edited Jul 13, 2022

I think you are badly underestimating Feynman, probably because you are overcompensating for his charisma and popularity. Feynman deserves the main credit for perturbative QFT, the foundation of particle physics. The path integral, in particular, is something almost no one else would have come up with. He also had many other important contributions: the beginnings of quantum gravity in terms of spin-2, the idea that became Fadeev-Poppov ghosts, the "beads on a stick" idea that eventually led to LiGO.

I'd say he certainly rates in the top five of the century, whereas Pauli and Planck do not.

Expand full comment

No I don’t think I am underrating Feynman. I am aware of the contributions that you mention but I don’t think they amount to top five in the 20th century, or even top ten, unless you consider ‘physics’ synonymous with ‘theoretical high energy physics’, which I don’t. I agree Pauli and Planck don’t make the top ten either.

It also seems fair to note that Dirac (who does make my top 5) wrote down the path integral before Feynman, and Schwinger and Dyson were doing perturbative QFT before Feynman also.

Expand full comment
Jul 13, 2022·edited Jul 13, 2022

Are your top 5 / top 10 basically from the Los Alamos crowd? Or are you thinking of people from elsewhere as well?

If I try to produce a list of top 20th century physicists, I end up with (in no particular order) Feynman, Einstein, Dirac, Bohr, Heisenberg, Fermi, Bethe, Schwinger, Oppenheimer, Dyson, Millikan(?), Planck, Pauli, and I think that's about it, but I'm not used to thinking about physics in terms of its "superheroes", and what I know about many of these is largely just what I remember from reading Gleick's bio of Feynman. By contrast, I don't follow who wins Nobel Prizes, and I'm probably prone to forget entire important subfields (I'm not counting string theory anyway, however), and I'm arguably going to lump some people into physics that you might not (e.g. astronomy, cosmology).

Expand full comment

Von Neumann led the team that built the first general purpose stored program electronic computer. He was a key figure in CS. He created a something to study like like Watt did with the expansion driven steam engine. A lot of other people played critical roles in advancing the science, but VN & Watt created something that could drive those studies.

Expand full comment
founding

I'm not sure our world wouldn't be better had the U.S. nuked the Soviets when von Neumann proposed doing so.

Expand full comment

No, but I'm not sure it _would_ be better either.

And faced with such utilitarian uncertainty I think it's reasonable to err on the side of not nuking anybody.

Expand full comment
founding

I'm not sure of that either – the Soviets were fucking terrible!

Destroying them might have prevented a LOT of misery and suffering (on net).

Expand full comment

Most of the Soviet damage to Russia was already done by the 1950s. The US by a first strike would have done even more damage to the Russian people than the Soviets. On top of that, I think the extent of guilt and shame on the US would have left the US broken as well. No one wins a nuclear war.

Expand full comment
founding
Jul 13, 2022·edited Jul 13, 2022

I'm thinking of The Iron Curtain generally and, e.g.:

- The Warsaw Pact

- Stalin by himself

- Their support of the Chinese communists

- The Cold War

- Cuba

- The invasion of Czechoslovakia

- The invasion of Afghanistan

I think it's reasonable to also include a LOT, if not literally all, of the various (nasty) U.S. anti-communism 'efforts' around the world, e.g. Vietnam and Central America, in the list of terrible, awful tragedies that potentially could have been mostly averted had we nuked the Soviets when we had the chance.

Had we nuked the Soviets, North Korea would either never have happened, or we'd have been able to nuke the Chinese communists (or have credibly threatened to do so).

There is a WHOLE LOT of misery that seems, to me, to be directly attributable to the totalitarian ideology we DIDN'T destroy directly when we had a chance.

And, arguably, the U.S. did a _fantastic_ fucking job 'fixing' both Germany and Japan after WW2. If we'd nuked the Soviets first, before they developed nukes themselves, and then installed/setup something like we did in the other former Evil Empires, not only Russia itself, but much of the rest of the world, very well might be MUCH better off.

I think von Neumann very well might have expected a LOT of what eventually happened. He had a lot of up-close-and-personal experience of exactly how _bad_ communism and totalitarianism is in practice.

Expand full comment

" Their support of the Chinese communists"

Morbid question: Is that about 50% of the death toll?

I've read estimated that Mao's great leap corpseward _by_ _itself_ killed between 15 million and 45 million people.

Expand full comment

From 2001 to 2021, the US lead a war against the Taliban in Afghanistan using everything in the book except nukes. Air strikes. Sat intelligence. Gitmo. Occupation by ground forces for two decades. Drone strikes against gatherings of suspected terrorists and civilian bystanders. Infrastructure projects. Democracy. Training of local police and military forces.

We all know the outcome.

How confident are you that Communism could have been defeated just by nuking the USSR?

Expand full comment

>"No one wins a nuclear war."

If you define it as an exchange of nuclear fire, agreed. If you include a war where only one side has & uses nukes, which a preemptive strike on the USSR would've been, then the US clearly won the only nuclear war to have occurred.

Expand full comment

It is probably not a great idea to get a taste for it though.

Expand full comment
Jul 19, 2022·edited Jul 19, 2022

Did he literally mean first strike? Or an American invasion backed by the threat of nuclear strikes if significant resistance was encountered?

We got through the Cold War, and might even get through the next 20 years without nuclear war. But if someone considers nuclear war to be one of the worst things ever, the way to prevent it is to have a hegemon enforcing a unilateral NNPT on the rest of the world.

If life exists on other planets, I'm sure this is the way it's played out on lots of them. It's kind of surprising given what we know of human history that the US had a city-destroying weapon and then . . . just let other people get it, too.

Expand full comment

At the end of WW2, the Soviet Union had superior conventional forces in Europe, and while America might have won a long war it would have been far too costly. Perhaps he meant an ultimatum first, but he definitely meant to launch nukes before engaging in conventional warfare.

Expand full comment

But if the point was to avoid being nuked and they never nuked us in reality, it seems unnecessary.

Expand full comment
founding

NOT 'stopping' the Soviets doesn't seem unnecessary. They caused maybe the most total amount of human misery ever.

Destroying the _other_ Evil Empires sure seems to have worked out great IMO.

Expand full comment

>Destroying the _other_ Evil Empires sure seems to have worked out great IMO.

The ones which are demilitarized and occupied to this day, sure. Would America have been willing to Marshall Plan the nuked Soviets? Somehow I'm very much doubtful about that, and in the absence of it, I'd say that the total amount of misery in the world would only have been increased. But of course, almost nobody in the Definitely Non-Evil Empire would've given a shit, judging by how they steadily memory-hole unimportant de-evilified places like Iran or Libya.

Expand full comment

They haven't nuked us YET. Putin probably has more nukes capable of reaching America than Khrushchev ever did.

It might be wise to bear that in mind.

Expand full comment

Well, reading these comments one might get the impression that you are all but begging Putin to prove you (and von Neumann, retroactively) right. Be careful of what you wish for, I guess.

Expand full comment

History isn't over yet.

Expand full comment

In 1951, when the Soviets succeeded in their first nuclear test, the US had (per https://journals.sagepub.com/doi/pdf/10.2968/062004017 ) 640 nukes. In 1957, when Sputnik 1 was launched, the US had ten times as many, and the USSR had 10% of that.

It is very hard to cause a regime change by air power alone. The Soviet leadership did likely not feel very protective of their civilian population and would probably not have stepped down to save Moscow. Morale bombings do not work. (see https://acoup.blog/2020/07/17/fireside-friday-july-17th-2020/ ) Even if you could magically kill every USSR soldier without touching any civilians, occupying the USSR would still be a rather large undertaking.

Dropping a few 100 nukes on population centers with the aim to slow Soviet industrial development to delay their nuke production would likely have delayed it a bit, but would IMHO not have stopped it permanently. It would also have cast the US as the global villain, normalized the use of preemptive nuclear strikes and pushed about every country in the world to acquire nukes and band together against the US.

https://en.wikipedia.org/wiki/Arms_race#/media/File:US_and_USSR_nuclear_stockpiles.svg

If one looks at the graphs, fighting WWIII early is obviously more advantageous for the US than fighting it later, but the actual outcome of not having to fight it at all is still better.

Expand full comment

Well, yes, but a lot of the argument is "how likely was the historical outcome of avoiding WW3 altogether"? The world came to the brink multiple times during the Cold War - is our timeline one that should have been predicted or one that got exceptionally lucky?

If you, in 1945, thought WW3 had a 50% chance of breaking out in the next few decades and knew that the earlier it was fought the better, would that have justified pre-emptive strike? what is the odds of war were 90%?

Expand full comment

I don’t think he was wrong about nuclear war. In our timeline we’ve come to the brink of it at least three times (petrov, Cuban missile crisis, Able Archer), probably more that we don’t know about. And when he was proposing the first strike second strike capabilities weren’t even known of yet! We were damn lucky to not get a nuclear war in our timeline, from his position a first strike is pretty reasonable. Probably even higher expected utility, and I say that with the knowledge nuclear war didn’t happen!

Instituting a policy of building nukes gets a nuke was the right move in the late forties, if I got time traveled back there I’d tell them as much!

Expand full comment

“ Instituting a policy of building nukes gets a nuke was the right move in the late forties”

Sounds plausible. Though I wonder about the capacities for verifying whether some country was instituting a policy of building nukes. Now that we’ve got all sorts of international atomic agencies, and spy satellites, and so on, verification is easy. But even now you sometimes get a North Korea that manages to sneak a nuke.

Expand full comment

I don't know if it would have been possible for the United States to stop North Korea from building its nukes without using military force. (China might have been in a better position to do so...)

Expand full comment

If you have the nukes and credibly threaten to use them, you can inspect the places you want by force.

It's a totally different world, worse in several ways. But it doesn't have the risk of nuclear war. I don't know how you weigh those things.

Expand full comment

The book’s second chapter gets into von Neumann’s set theory contributions quite deeply. You might enjoy. His interests here would prove central to his work on computers, game theory etc.

Expand full comment

The book’s second chapter gets into von Neumann’s set theory contributions quite deeply. You might enjoy. His interests here would prove central to his work on computers, game theory etc. It’s a theme I was really interested in. How did his earlier work shape his more practical work later on… I’ve spent some time on that.

Expand full comment

> his naive game theory about nuclear war

That's easy for you to say now, because you know the Game Theory that von Neumann invented later.

Also, a first strike on the Soviet Union would *not* have killed millions. The whole point is that only the US had nuclear weapons. According to this [0] the casualties from Hiroshima and Nagasaki were about 200,000 people. I don't know which exact strategy you would use for preventing the Soviet Union from getting nuclear weapons but I think you could do it with about that number of casualties?

When you then consider the amount of misery and death that Soviet communism brought to the world, it's not entirely clear to me that von Neumann was wrong.

[0]: https://www.atomicarchive.com/resources/documents/med/med_chp10.html

Expand full comment

A first strike totally would have killed millions, either directly or in the war it started. Even if you try for a precision strike on their nuclear infrastructure (probably impossible given the limitations of US intelligence in the 40s and 50s), they're just going to shrug and send their armies to roll over Western Europe. Any attack against the Soviets is going to mean all-out war, both conventional and nuclear. And that in turn means the first strike needs to hit the broader Soviet industrial base as hard as it can.

von Neumann was wrong on this, but only because it turned out to be possible to win WWIII without directly fighting the Soviets. If that hadn't been possible, he would absolutely have been right, simply because of how much worse a later war would have been.

Expand full comment

Since the Soviets only had the capability to roll up to Berlin >because< the Allies gave them enormous amounts of stuff (both civilian and military), the idea that they would have been capable of going to the Atlantic in the late 40s without that help is... not realistic.

They probably became somewhat more capable after rebuilding, i.e. the 50s, but I'm still doubtful of a European war actually being successful for them.

Since they had a lot of help from spies (yes, the Rosenbergs were correctly convicted) to produce their own bombs, I suspect that von Neumann urging a strike >before< they could get their own production running was likely the optimum solution.

The other problem would have been that there really was no appetite within the other Allies to continue the war, but this time against an opponent that had just been allied with them until yesterday...

Expand full comment

> Since the Soviets only had the capability to roll up to Berlin >because< the Allies gave them enormous amounts of stuff (both civilian and military), the idea that they would have been capable of going to the Atlantic in the late 40s without that help is... not realistic.

The idea that you can nuke a country with one of the largest militaries and industrial bases in the world, fail to annihilate that military and industrial base (since the nuclear capabilities of the US in the 1946-49 period were not capable of this), and not expect a whole fuck of a lot of your people to die in the ensuing war, is way, way more unrealistic than anything else.

Expand full comment

I completely grant that the Soviets greatly benefited from American logistical help in WWII, and that the Red Army was never quite as formidable as it looked on paper. But that still leaves it with all the stuff that we did give them, and a large and experienced Army. Against a US military presence that has been cut to the bone, and European powers that are bankrupt and focused on rebuilding. It's hard to remember just how bad the military situation was in the late 40s, but there's every chance that they could have gotten a lot further than you think.

And in any case, my point was more that you couldn't ignore the conventional aspects and just destroy the Soviet nuclear program.

Expand full comment

Von Neumann was aware the Soviets had spies in the Manhattan Project before that became public knowledge.

Expand full comment

But did VN know who the Soviet spies were?

Von Neumann and Soviet spy Klaus Fuchs applied together for a patent on a type of H bomb in April 1946:

http://blog.nuclearsecrecy.com/2013/08/23/the-spy-the-human-computer-and-the-h-bomb/#:~:text=One%20of%20the%20most%20enigmatic,and%20dates%20from%20April%201946.

Expand full comment

Julius Rosenberg did spy, and Ethel was at least aware of it; that does not mean that the case against them was procedurally or materially correct (it wasn't). But that's a side issue that won't get resolved here.

Expand full comment
Jul 13, 2022·edited Jul 13, 2022

People in this thread seem to be forgetting that the entire case of the Nuremberg Trials was based mainly on the proposition that aggressive warfare was illegal. Aggressive nuclear warfare is mass murder on top of it. Surely it is a good thing if there wasn't enough appetite for it.

Expand full comment

Well, it wouldn't be "aggressive warfare" since it would be a response to Soviet aggression and had a defensive purpose. Soviets invaded Finland and Poland pre-war and occupied/controlled many other countries after. The same casus belli existed against the Soviets as there was with Germany.

Expand full comment

Using that argument, you can justify aggressive warfare against just about every country - the U.S. most certainly included. It's a good thing international law is not determined by the comments section. Neither is foreign policy; one may sometimes think it could not get worse, but, well, apparently it could.

Expand full comment

"his naive game theory about nuclear war"

Given how Communism blighted the economic development of eastern & central Europe for 45 years (Hungary was on a level with Austria, diverged during Communism, is now converging strongly) and also slowed China's growth, *if* a nuclear war could have toppled the Soviets in the early 1950's (a very big if), it could easily have been a net-positive for human flourishing compared to our timeline, and definitely would have been a net-positive compared to the timelines where there were massive nuclear exchanges in the 1970's/1980's and/or where Communism won/was still around in Russia*. Truman+Eisenhower took a very large gamble that could well have been negative in expected value terms, based on the timeline that we've seen, but it came out positively for us.

*Even in our timeline, we're not done with Communism yet, as the Uyghurs and Tibetans will tell you.

Expand full comment
founding

There's SO many things that probably would have never happened or gone much better had we nuked the Soviets when we could have, e.g. basically all of the U.S.'s own (generally terrible) anti-communism 'efforts'.

Expand full comment

I don't think Von Neumann ordinals are a good way to think about what ordinals are. Fundamentally, ordinals are order types of well-ordered sets. Von Neumann ordinals are a convenient representation for when you need to be able to have them as actual objects in your theory, and are nice for how they formalize the "an ordinal is the set of all smaller ordinals" way of thinking, but honestly, I think for most purposes it's perfectly fine for ordinals (and cardinals!) to be metatheoretical. Fundamentally an ordinal is an order type, not any particular representation of that order type.

Expand full comment

A podcast interview of the author, Ananyo Bhattacharya by Razib Khan:

https://unsupervisedlearning.libsyn.com/ananyo-bhattacharya-the-life-of-john-von-neumann

Expand full comment