Saturday, 16 March 2013

East Asia's Farewell to Alms


“‘How could any man in our village claim that his family had been poor for three generations? If a man is poor, then his son can’t afford to marry; and if his son can’t marry, there can’t be a third generation” China’s poor were continually removed from the gene pool, their places taken by downwardly mobile individuals (Woodblock of farmer, Wang Liangjian, 1939, source)


Mean IQ is unusually high in East Asia, averaging around 106 among Han Chinese, Koreans, and Japanese (Rushton & Jensen, 2005). It falls off as one moves outward from the core area of East Asia, being lower even among the nearby and closely related natives of Mongolia (Lynn, 2007). This ‘IQ plateau’ must therefore have a relatively recent origin, certainly after the advent of agriculture and probably after the rise of State-pacified societies—where most people expected to succeed through work and trade, and not loot and plunder.

Is this ‘IQ plateau’ due to cultural and family values that are specific to East Asian societies? Not likely. Higher mean IQ is observed in East Asian individuals who were adopted at an early age into white American or European families (Clark & Hanisee, 1982; Frydman & Lynn, 1989; Winick, Meyer, & Harris, 1975). It also correlates with better performance at more fundamental mental tasks, like reaction times (Rushton & Jensen, 2005). It seems to be genetically determined.

What caused this genetic evolution? In a recent article, Ron Unz attributes it to the social dynamics of an agrarian society where both land and women were scarce. With limited prospects for starting a family, the lowest strata of society were continually dying out and being replaced by downwardly mobile individuals from the highest strata: 

 […] only the wealthier families of a Chinese village could afford the costs associated with obtaining wives for their sons, with female infanticide and other factors regularly ensuring up to a 15 percent shortfall in the number of available women. Thus, the poorest village strata usually failed to reproduce at all, while poverty and malnourishment also tended to lower fertility and raise infant mortality as one moved downward along the economic gradient. At the same time, the wealthiest villagers sometimes could afford multiple wives or concubines and regularly produced much larger numbers of surviving offspring. Each generation, the poorest disappeared, the less affluent failed to replenish their numbers, and all those lower rungs on the economic ladder were filled by the downwardly mobile children of the fecund wealthy. (Unz, 2013)

In this Hobbesian world, reproductive success went to those with the most business acumen:

The members of a successful family could maintain their economic position over time only if in each generation large amounts of additional wealth were extracted from their land and their neighbors through high intelligence, sharp business sense, hard work, and great diligence. The penalty for major business miscalculations or lack of sufficient effort was either personal or reproductive extinction. (Unz, 2013)

Another factor may have been the imperial examination: “in China the proud family traditions would boast generations of top-scoring test-takers, along with the important government positions that they had received as a result.” But Unz later backs off from this possible cause, noting that only one percent of the population attained the top rank of chin-shih or the lesser rank of chu-jen—too small a percentage to have much evolutionary impact. True, but those two ranks were only the top of a much larger population pyramid. Success at a lower level, such as at the district or provincial levels, still brought some benefits and prestige, and the beneficiaries were a much larger pool of people (Frost, 2011).

Parallels to Clark’s model

All of this sounds much like the model that Gregory Clark put forward to describe the demographic, behavioral and, perhaps, genetic evolution of the English people. According to this model, the English middle class expanded slowly but steadily from the 12th century onward, thereby gradually raising the population mean for predispositions to non-violence, pleasure deferment, and other future-oriented behavior. Although this social class was initially very small in medieval England, its descendants grew in number and gradually replaced the lower classes through downward mobility. By the 1800s, its lineages accounted for most of the English population (Clark, 2007).

Did Ron Unz steal his idea from Gregory Clark? A casual reader might think so. Buried in the footnotes, however, is a mention of a similar paper that a younger Ron Unz had written back in 1983 while a student at Harvard. But at that time few people were thinking along the same lines. In the history of ideas, Ron’s experience is depressingly similar to that of Patrick Matthew, the Scottish scholar who developed a theory of evolution by natural selection a quarter century before the publication of Darwin’s On the Origin of Species.

Let’s hope this article will give Ron a second hearing in the court of academic opinion. Let’s also hope his article will inspire further research, particularly by Chinese geneticists, historians, and social scientists.
 
References

Clark, E.A., and J. Hanisee. (1982). Intellectual and adaptive performance of Asian children in adoptive American settings. Developmental Psychology, 18, 595–599.

Clark, G. (2007). A Farewell to Alms. A Brief Economic History of the World, Princeton University Press, Princeton and Oxford

Frost, P. (2011). East Asian intelligence, Evo and Proud, February 18
http://evoandproud.blogspot.ca/2011/02/east-asian-intelligence.html

Frydman, M., and R. Lynn. (1989). The intelligence of Korean children adopted in Belgium. Personality and Individual Differences, 10, 1323–1326.

Lynn, R. (2007). IQ of Mongolians, Mankind Quarterly, 47, 91-97.

Rushton, J.P. and A.R. Jensen. (2005). Thirty years of research on race differences in cognitive ability, Psychology, Public Policy, and Law, 11, 235-294.

Unz, R. (2013). How Social Darwinism made modern China, The American Conservative, March 11
http://www.theamericanconservative.com/articles/how-social-darwinism-made-modern-china-248/

Unz, R. (1983). Preliminary notes on the possible sociobiological implications of the rural Chinese political economy, unpublished paper.
http://www.ronunz.org/wp-content/uploads/2012/05/ChineseIntelligence.pdf

Winick, M., K.K. Meyer, and R.C. Harris. (1975, December 19). Malnutrition and environmental enrichment by early adoption. Science, 190, 1173–1175.

52 comments:

Anonymous said...

Cochran has a new posting which relates to this and your previous posting

Sean said...

Yes, in olden days testicles were no use to poor Chinese men, they were volunteering to be castrated so they could become eunuchs-servants of the rich.

However I have my doubts about what the cause of Chinese IQ was, because though the Chinese have a lot of high IQs it seems they do not produce the expected number of super high IQs found among Jews.

For instance Ron Unz says Chinese Americans (migrants)* have "15 percent of the highest-achieving students, a performance ratio more than four times better than that of American Jews". Yes, but the top 1% of the total top attaining American high-school graduates, are they four times more likely to be Chinese than Jewish?

Moreover the Chinese do not seem able to produce self made billionaires at a rate that would suggest they have been selected for success in business. I suppose language and the docile Chinese personality may hold them back, but I would expect them to do much better, if Unz's explanation was correct. Not having been to Yale doesn't seem to hold Israelis back. But maybe Chinese are suited to dealing with other Chinese, and Jews do best when they are dealing with non Jews.

*(Weyl: "Of the 800 or so white Rhodesian children who scored 130 or better, no less than twelve were shown to have Iqs in the 180+ range. This again is a multiplier of the normal IQ distribution".)

William Charles Welles. was Scottish too. It is perhaps relevant that Scotland (and Germany) has a claim to be the first country to have universal literacy.

All in all I think the evidence suggests that Chinese have been mainly selected for passing exams that produced a narrow type of intelligence. (Here is Derbyshire being scathing about it 'ANOTHER influence on the way I think about this is my own studies of Chinese history and culture. Candidates for the Imperial examinations in old China had to engage in the same kind of years-long concentrated study of huge masses of accumulated written material that Talmudic scholars have to master. At the end of their studies, for the Imperial examinations, the Chinese scholars had to write an "eight-legged essay" — that is, one conforming to certain traditional patterns of style and presentation. You can find translations of prize-winning "eight-legged essays" in books about Chinese culture. I have one here. It is gibberish. It is content-free. However, if you passed the exam, you got a lifetime job as a Mandarin, a guaranteed income, and a choice of breeding partners.')

I think Jews have been reproductively rewarded, and hence selected, for a type of intelligence that aids business acumen and success in much else (check out what a Talmudic scholar pointed out to Derbyshire's who scoffing at MacDonald for saying Talmudic study was responsible for Jewish IQ, same link). Jewish exam results don't do them justice in my opinion.

M said...

"Ron Unz says Chinese Americans (migrants)* have "15 percent of the highest-achieving students, a performance ratio more than four times better than that of American Jews"

Unz has been cited by Andrew Gelman as having a few problems with distinguishing foreign selected Asians from native born Asians.

On a similar note, there's often a sentiment on the internet that you don't need any kind of super-duper immigrant selection to explain the (greater than White native average) Asian success levels in Western countries.

And for the most part, that's right. They would be more successful (for certain kinds of success at least) even if randomly selected.

Still, I think an interesting factor here is, a little experiment I carried out.

If you look at the distribution of educational qualifications among foreign born Asians, which is available on the US government census page - http://www.census.gov/population/foreign/ - it does turn out that they have quite a bit more eductional diversity (high end and low end) than White natives - http://www.census.gov/hhes/socdemo/education/ - and this is true for all the subgroups lumped as Asian (Chinese, Indian, South East Asian).

If you use the GSS to estimate the IQ of different educational groups - using the variables Wordsum and educ - then it turns out that if you assume each educational level has the same IQ as the White population of the US at the same educational level*, then foreign born Asians should have an IQ standard deviation 1.3-1.2 times as large as the native born White population, with a similar mean IQ.

Assuming there is any correlation between parent and child IQ, this should also carry over into the next generation.

Funnily enough, the Asian American SD on the SAT is also heightened - http://media.collegeboard.com/digitalServices/pdf/research/TotalGroup-2012.pdf to around 1.25 or 1.29 the White SD, on all tests from Reading to Maths.

Naturally, a higher SD converts to more representation at the high end (and no one cares about the low end).

Of course, East Asians do better than Europeans on IQ tests (and derived similar tests) BUT it seems pretty likely to me that the extreme overrepresentation of American Chinese (and to a lesser extent other Asian Americans) in the higher echelons of academia and business is actually also likely to be a factor of a disproportionately large elite migration stream.

*note though - foreign born Asian educational groups seem likely to be slightly above Whites of the equivalent educational level - or at least, this would be the case amongst foreign born Chinese, if not Filipinos.

Also notice I said White Americans!

Black Americans on the GSS routinely have a Wordsum derived IQ around 0.7 SD Wordsum below Whites of an equivalent educational qualification - slightly less at the low end and slightly more at the high end (i.e. IQ gaps are starkest when comparing Whites and Blacks of the same educational attainment level).

The gap at any educational level, based on Wordsum and the GSS, is barely less (and often more) than it is as a whole. Which is remarkable... The US truly does have a parallel education system (either at the same schools or different ones!).

PoW said...

“ However I have my doubts about what the cause of Chinese IQ was, because though the Chinese have a lot of high IQs it seems they do not produce the expected number of super high IQs found among Jews..

...

Moreover the Chinese do not seem able to produce self made billionaires at a rate that would suggest they have been selected for success in business..." ---- Sean said.


IQ is a recent phenomenon – about 100 years-old? Prior to that, the world didn’t have relatively unified measurement of it. China’s last 100 years was the lowest point in her history, arguablely the entire 5,000 year-old history. People were constantly fighting wars…. At the eve of the founding of the PRC, about 90% of Han Chinese adults were functional illiterates – that was only decades ago.

Jews, or strictly speaking Ashkenazi Jews since average IQ of Jews is relatively very low, however, have been in an entirely different situation: the bulk or the absolute majority of Ashkenazis have been raised and educated inside or around the 3 or 4 greatest cities of the world ever seen in the last 200 years, namely Berlin, Paris, London and NYC – being the native language speakers (and the same European race at least by appearances mostly), full enjoying the complete, the absolute, and the best cultural, economical and technological advances of the human kind of the last 200 years brought by the 1st and 2nd industrial revolutions at its very doorstep within very individualistic cultures.

What a fair comparison you have there!

The same comparison goes to that “legendarily objective”Nobel Science Prizes, or “US high school,toppers” to name a few…

The fair comparison, IQ wise, or achievement-wise, is to randomly take some of Chinese and Jews, drop them in the middle of a third country (where neither is a native speaker), ideally non-White and non-east Asian country, then you see who will come out on top academically and technologically years later.

And that Billionaire example…Majority billionaires, just like majority CEOs, I suspect, as you may also aware by first hand experiences, are not high IQers at all. We all know that IQ and Success (measured by money, how convenient! ) are generally positively, yet not highly, correlated, that’s it. Talking about IQ measured by billionaire/capita, some plant-growing tribes in Colombia and Afghanistan probably can claim the smartest race(sub-race) in the world, Ashkenazi who?

As a side note, one important mistake/misnomer I’ve noticed in the recent posts of Frost is the insinuated claim that “Chinese peasants have lower IQ (than ones from cities)”. Not quite, simply because

i. the Chinese are comparatively very homogeneous in the Han Chinese heartlands (except some places in the deep South and the far West), due to countless and constant huge migrations nationwide with all directions, as recent as WWII and theh Chinese Civil War that followed, and

ii. urbanisation of PRC is a recent phenomenon that till today has been largely constrained by “Hukou” system.

These result, contrary to conventional wisdom, that the average Han Chinese peasants today are about as smart as the average Han Chinese in the city, after normalising several small yet obvious environmental factors such as healthcare, education spending/capita, nutrition level and opportunity cost, etc. e.g. Almost ANY Chinese you encounter in the West, no matter how urban he/she is, has a partial extended family live in rural China as “peasants”, and probably the majority of their own and extended family were peasants just 3 generations ago. Decades of scores of China’s university entrance exams, “GaoKao”, prove so. And China nationwide test scores in PISA 2009 also prove so - many major provinces, arguablely one of the typical preventatives, across the whole China, took the test and the scores were very close according to the leaked info, considering that the bulk of the test was conducted with heavier weights on rural population in less developed provinces during sampling process on purpose.

Anonymous said...

But maybe Chinese are suited to dealing with other Chinese, and Jews do best when they are dealing with non Jews.

Racially and ethnically mixed environments are evolutionarily unstable. In a mixed environment, most races/ethnies will be eliminated and or mixed out of existence. The races/ethnies of the world only evolved and exist in the first place due to isolation. Which race(s)/ethny(s)/combination(s) end up persisting or dominating in a mixed environment probably depends on various traits relative to the environment - the various other races/ethnies being a major feature of the "environment". Political and social dominance would appear to be important. In European environments where the major components of the environments were Europeans, Jews were able to obtain and hold certain professional niches. Whereas in parts of the Middle East they had less success. "The 10,000 Year Explosion" notes that in parts of the Middle East, Jews were relegated to certain low status occupations such as cleaning toilets.

Anonymous said...

Which race(s)/ethny(s)/combination(s) end up persisting or dominating in a mixed environment probably depends on various traits relative to the environment - the various other races/ethnies being a major feature of the "environment".

Another major feature of this environment would be whether or not it is a "civilized" environment in which male self-defense and territoriality are suppressed by a centralized power such as a state. If they are not suppressed, then martial and war-making ability and more technically sophisticated combat ability become more important. If they are suppressed, then greater political and social dominance and influence over centralized power become more important.

AK said...

It should be noted, however, than Ron Unz himself has significantly retreated from his theory:

Actually, prior to my getting sidetracked on my Race/IQ article, I was spending a great deal of time researching my old analysis, rereading all the source material I’d originally used—most of which I hadn’t looked at in decades—and reviewing a great deal of the newer Asian demographic research which had appeared since the late 1970s. Unfortunately, by the end of that investigation I found my theory much less persuasive than I had at the beginning.

[continues...]

Ron Unz said...

AK: It should be noted, however, than Ron Unz himself has significantly retreated from his theory:

Yes, I highlight those problems in n.34 of my new article, right near the end: http://www.theamericanconservative.com/articles/darwinism-china-endnotes/#endnote-34. But I've actually slightly swung back again on the issue.

First, relative to their massive current poverty, China's provincial PISA scores are extremely impressive compared to those of Japan, while the numbers for 15M Shanghai are far higher.

Also, consider the relative proportion of American NMS semifinalists. Ethnic Japanese constitute over 1% of California's population and are less than 1% of NMS semifinalists, while ethnic Chinese are 3.8% of the population but 35% of NMS semifinalists, indicating more than a 10x rate of nominal relative academic performance. Certainly much of this is accounted for by the differing age-distribution curves (fewer Japanese are of H.S. age), high white intermarriage, and third/fourth generation acculturation/lack of effort, plus the important impact of highly-selective recent Chinese immigration. But the raw performance difference is so enormous---well above 10x---that I can easily imagine a substantial residual difference remaining once these factors were removed. So I consider it perfectly possible that Chinese are significantly higher-performing than Japanese.

Comparing national IQ scores from Lynn are tricky, because they may often be from unrepresentative urban areas. Clearly, Japanese are a smart, high-ability people, but I could easily see the Chinese being significantly above them. And interestingly enough, the early twentieth century sources claimed that Chinese were always far more successful in business than Japanese, even in Japanese-controlled colonies.

As for Koreans, the evidence seems to put their academic performance intermediate between that of Japanese and Chinese, again with huge uncertainty based on all these distorting factors. Meanwhile, from what little I've read, determining the nature of traditional Korean rural life seems very difficult due because of the conflicting ideologically-charged claims of Japanese and Korean-nationalist scholars, disputing the nature of the society conquered and colonized by Japan.

None of this eliminates my major n. 34 caveat, but I just don't regard the Chinese/Japanese comparison as fully dispositive as I did seven or eight months ago.

Anonymous said...

In this Hobbesian world, reproductive success went to those with the most business acumen....Another factor may have been the imperial examination

Or perhaps it went to those with the best farming ability. There is a book popular in organic farming circles called "Farmers of Forty Centures" which was written a century ago by an American agronomist named F.H. King who toured China, Korea, and Japan to investigate traditional East Asian agriculture:

http://www.amazon.com/Farmers-Forty-Centuries-Organic-Farming/dp/0486436098

Apparently King writes in the book that there were very efficient and sustainable pre-industrial farming methods in East Asia. Presumably those who developed and maintained the more efficient methods were able to yield more food and thus have more kids. And if they were sustainable methods that preserved soil fertility, then maybe the potential agricultural yields were preserved for those descendents who had the farming ability to maximize yields and have more kids.

This would be consistent with the traditional social hierarchy in East Asia which prized and honored farmers and agriculture and looked down upon commerce and merchants:

https://en.wikipedia.org/wiki/Four_occupations_%28East_Asia%29

Scholar-officials and farmers had the highest social status. Artisans and craftsmen had the next highest status. Merchants and traders had the lowest social status. Social status is a major determinant of reproductive success. Farmers would have been the majority of the population, and the fact that they had relatively high status would have meant that they would have done well in terms of reproductive success even if they weren't as financially wealthy as merchants were.

This would be consistent with relatively lower verbal ability, since commercial activity emphasizes verbal ability. Commercial activity primarily consists of dealing with other people, and dealing with other people is primarily done through words, so verbal ability is critical. If commercial ability entailed the greatest reproductive success, one would expect relatively higher verbal ability.

And this is consistent with the similar high levels found in East Asia, since King notes in his book that he found the efficient and sustainable farming methods present in China, Korea, and Japan, and since the cultural norms that rewarded farming with relatively high social status pervaded throughout East Asia.

Anonymous said...

Here is a controversial study of Mongolian IQ by the some Japanese and Mongolians.

http://ir.u-gakugei.ac.jp/bitstream/2309/108073/1/18804306_62_01.pdf

It shows rural six year old Mongolians have average 120 IQ, while urban Mongolian children have average 105 IQ.

Anonymous said...

Apparently King writes in the book that there were very efficient and sustainable pre-industrial farming methods in East Asia. Presumably those who developed and maintained the more efficient methods were able to yield more food and thus have more kids. And if they were sustainable methods that preserved soil fertility, then maybe the potential agricultural yields were preserved for those descendents who had the farming ability to maximize yields and have more kids.

...

This would be consistent with relatively lower verbal ability, since commercial activity emphasizes verbal ability. Commercial activity primarily consists of dealing with other people, and dealing with other people is primarily done through words, so verbal ability is critical. If commercial ability entailed the greatest reproductive success, one would expect relatively higher verbal ability.


Few caveats (I don't find this especially plausible, but nonetheless)

- being a farmer in ancient China doesn't mean they're actually in the field growing stuff - successful farmers employ labor and produce for markets. that's not like being the despised merchents, but the difference is that the farmer is a producer and is not travelling around and seeking new goods, and he is selling a commodity it is hard to scam people over, not that he isn't engaged with markets.

- commercial activity, successful commercial activity, is a question of money and money is a question of mathematics. but perhaps not so much general spatial skills (except those seem to be rather inextricable from math ability for functional reasons whereby quantitative skills involve using back end spatial models).

- the Ashkenazi Jews were to a large extent predatory lenders to commoners and tax farmers for the king more frequently than they were and kind of merchants. That's the kind of niche people should consider as shaping them.

Another caution here would be that I recall lawyer/economist/general bright fellow Nick Szabo's speculations that the apparently inefficiently animal intensive agriculture of the West was really good at producing manure and very good at transporting farm goods on the hoof (cattle and draft animals have broad shoulders).

So an apparently less efficient agricultural method (in terms of area - grass fed cattle being less area efficient than cereals) was much better for market distribution of food in real terms and for maintaining social productivity.

I think Szabo also postulated that this gave markets a much earlier and wider reach in particularly Northwest Europe (at least after adjustment for technological and natural advantages), where dairying and accumulative, stationery farming with pastoralist advantages was most common.

http://unenumerated.blogspot.co.uk/
2011/06/agricultural-consequences-of-black.html

http://unenumerated.blogspot.co.uk/2011/06/trotting-ahead-of-malthus.html

http://unenumerated.blogspot.co.uk/2011/05/lactase-persistence-and-quasi.html

Anonymous said...

"I think Jews have been reproductively rewarded, and hence selected, for a type of intelligence that aids business acumen"

Jews in Europe had a monopoly of banking and money-lending which - it seems to me - would have created a sealed ecosystem for a very specific and enclosed "farewell to alms" effect with selection for success based on those traits most useful for banking and money-lending e.g. mathematical ability.

I think that's a much simpler explanation than the Talmudic study theory (although it may be part of it as well).

(Although i certainly think literacy / education as an *enabler* of selection is part of the story - as pointed at by the disproportionate influence of literacy valuing religious denominations in the industrial revolution.)

Sean said...

If Chinese have been selected for excelling in narrow exam contexts, their test scores may slightly overestimate their potential.

Hideki Yukawa: Nobel prize for physics 1949 (not poor, but he studied only traditional subjects like calligraphy until he was 14). Has any Chinese matched Hideki's contribution yet? I don't think the Chinese exceed the Japanese in practical achievement.

Anyway, if China’s poor were being continually removed from the gene pool, and if Chinese official exams conferred reproductive success on the most intelligent; the real question should be why their IQ is not far higher than it actually is. Why isn't at least as high as the Ashkenazim's?

I think the reason is that successful Chinese men were not constrained in their choice of breeding partners like Jews were; the Mandarins were free to choose the most beautiful women.

Anonymous said...

If farmers were the majority, and presumably as a traditional agricultural society they were an overwhelming majority, and cultural norms provided them with relatively high social status, it's hard to imagine minorities of officials or merchants significantly outbreeding them.

M said...

Unz said : Consider the relative proportion of American NMS semifinalists. Ethnic Japanese constitute over 1% of California's population and are less than 1% of NMS semifinalists, while ethnic Chinese are 3.8% of the population but 35% of NMS semifinalists, indicating more than a 10x rate of nominal relative academic performance

Ron, I had a look at these numbers, and thought you might be interested in the results.

OK, so 1,664,479 SAT takers in 2012. 16,000 NMS semifinalists chosen from amongst these proportionately for states.

In Cali in 2009, 207,301 test takers. Translates to 0.124 * 16000 NMS. So that should be 1992 NMS semifinalists, in total.

In Cali in 2009, there were around 44,026 Asian Americans taking the SAT relative to 71,501 White Americans. Interestingly, that's more like a population size 61% of the White college bound population, as opposed to the estimate of 33% that is typically made (I guess in error).

http://www.domesatreview.com/content/sat-test-demographics-income-and-ethnicity

The Asians had a SAT mean of 1083, with roughly an SD of 1.2x the White SD, while the Whites had a SAT mean of 1092.

Sailer states that the California NSM semifinalist cutoff score should convert to SAT as follows here - http://isteve.blogspot.co.uk/2010/02/national-merit-semifinalist-qualifying.html - (218/3)*20 = 1453. That seems standard across the web.

We based on the SD and means above, we'd estimate that the Asian Californian probability of reaching that score is 2.81x the White (based on SAT, as a normal distribution), almost purely through a larger standard deviation!

Based on the relative population size, we should see 2711 Asians relative to 2540 Whites, assuming a typical normal distribution.

Assuming other ethnicities don't matter (kind of harsh, I admit), we should see an Asian of NMS semifinalists, in California. That's 51% Asian.

Of course, this is more conjecture and, I have to plead ignorance of how NMS works a little here, but I would think that if you had a 2.63 excess of students, you would increase the cutoff (or hey, random sampling a go go).

Readjusting to a required score of 1550 gives around the right number of Californian semifinalists, and an Asian percentage of 59% (which is still typical)

Of the Asians in California, ethnic Chinese are around http://en.wikipedia.org/wiki/Demographics_of_California#Asian_Americans 25% (including self identified Chinese and Taiwanese).

So we should expect 12.75-14.75% of NMS semifinalists in California to be Chinese (or 293 semifinalists).

A trump to that by a factor of 2.5-3.0 is a little surprising!

This might be slightly resolved if we say that the Chinese are slightly disproportionately likely, of the Asians in California, to have parents with high educational attainment.

As I said in my earlier post, all subgroups of Asian Americans have diverse educational attainment - and actually foreign born Chinese do not have especially accelerated educational attainment - but California may be disproportionately a magnet for smart Chinese...

Or Ron's numbers for the Chinese might be slightly off, even though his numbers for the Asians overall are completely orthodox and expected given their large standard deviation of the SAT and similar overall SAT scores to Whites (which largely arises through Asian Americans being a combination of randomly selected Asians and elite selected Asians).

maopai said...

"Has any Chinese matched Hideki's contribution yet? I don't think the Chinese exceed the Japanese in practical achievement."

Han people gave the Japanese civilization. 倭国 (Japan) was only able to rise because China was ruined by nomadic savages in 1644.

Ron Unz said...

M: So we should expect 12.75-14.75% of NMS semifinalists in California to be Chinese (or 293 semifinalists).

A trump to that by a factor of 2.5-3.0 is a little surprising!

This might be slightly resolved if we say that the Chinese are slightly disproportionately likely, of the Asians in California, to have parents with high educational attainment.


Actually, the general Chinese over-representation among Asians in CA and nationwide didn't greatly surprise me, since I expected them to be far stronger academically than Filipinos, Vietnamese, and various other South East Asian groups. But what really shocked me was their 10+x outperformance compared with Japanese. The Japanese TFR has been very low for a couple of generations, and I'm not sure there's any solid estimate of the number of their students competing in the NMS, so that certainly accounts for part of the gap, but there's no way it could account for most of it.

Regarding the relative performance of elite vs. average Chinese, Weyl actually came up with some interesting estimates in one of his books, based on the distinction between "old Chinese" and "new Chinese" surnames, corresponding to early immigrants vs. post-1950 ones, the former being pretty average and the latter usually highly-selective. Generally, at the higher end (e.g. NMS), the performance ratio was a factor of 2-3x, which seems pretty plausible.

PoW said...

Sean,

It’s a bit hard, for me at least, to conceive what is “ being selected for excelling in narrow exam contexts” like, for if there had been such a thing, the Chinese verbal scores would have been off the charts given that the traditional Imperial tests didn’t include any maths but mainly essay-writing, the dull type.

Furthermore, I believe # of Novel Prizes on Science won is a widely abused proxy, even though it’s quite common in www, for intelligence of the underlying population they represent, to name a few reasons:

1. the prizes are usually delayed for decades after the actual achievements they try to award
2. it is highly subjective, subject to even political pressure sometimes.
3. it is highly biased in pre-selection process: to be on the list for pre-selection, one mostly needs to be from renowned, well-networked and well-funded labs(mostly Western), under renowned and well-networked tutors/institutions (akin to legacy admission of Harvard) to say the least that basically eliminate most if not all non-developed countries.
Hence it pro-western, pro-developed (i.e. positively correlated with a country’s GDP and GDP/cap).

This results countless weird things. E.g. Western Germans probably have IQ of 102 and East Germans 85? whne South Koreans win multiple prizes one day while North Koreans none, so South Koreans IQ 105, North Koreans 60? Further, if one day Nigeria wins one, for PC’s sake you know, so Nigerians are smarter than North Koreans, eh?...

As for your comparison btw Chinese Physicists and Japanese ones, I presume, you wanted to mean racial-wise not nationality-wise, there’re more than a couple Nobel-winning ethnic-Chinese physicists actually in the last several decades originally from HK and Taiwan but representing the US.

Thing is, you, like most western observers I’m afraid, tend to look at this Chinese vs Japanese thing via a current simple snapshot, or several snapshots starting from the late 19th century, and conveniently ignoring the entire history prior to it. By the same logic, Greeks or Portugues must be much smarter, on average, than East Germans and Poles under Soviet’s rule? Why not?

Traditionally there were NO Chinese vs Japanese in the eyes of East Asians, for it would have sounded as laughable as Germany vs Norway. Japanese didn’t appear on Chinese radar as any sort of “competitor”, ( for that matter not even on the radar of Korea which was China’s vassal state at a time), until mid 19th century after Japan started to industrialise under Western influence.

Ironically, today’s Japan , technologically the strongest of the East Asia 3 (China, Korea, Japan), was actually the weakest for most of the history. Fortunately and logically, Japan was the first to be conquered by the West when American merchant ships arrived at its shores in the 19th century. Hence Japan enjoyed almost un-interrupted 100+ years head-start on industrialisation vís-a-vís China and Korea… how modern Han Chinese look at Japanese is like what Americans would think if they were somehow technologically surpassed and militarily humiliated, twice, by Canadians or Icelanders due to a series of national misfortunes, while both Chinese and Koreans today look at Japanese with the similar sentiments as how Anglo-Saxons think of Vikings after the invasions, if I have to give an analogy.

Now, the point that the poor being continually removed from the gene pool is Unz’s ( and some others’) theoretical hypnotises. Firstly it’s not necessarily true; Secondly even if it is, it doesn’t automatically overwrite other major factors and forces, as there’re bound to have many, for (e,g, culture) or against(e.g. corruption) the hypnotises in the real world operating at the same time. So from this angle it’s hard to tell who is smarter…

Actually, 1 billion strong Han Chinese average vs several million Ashkenazis, a sub branch of Jews who are in turn also arguablely a sub branch of Europeans given such a long history of inter-breeding amongst them, is almost the textbook definition of what’s apple-orange in my view.

M said...

Unz:Actually, the general Chinese over-representation among Asians in CA and nationwide didn't greatly surprise me, since I expected them to be far stronger academically than Filipinos, Vietnamese, and various other South East Asian groups. But what really shocked me was their 10+x outperformance compared with Japanese. The Japanese TFR has been very low for a couple of generations, and I'm not sure there's any solid estimate of the number of their students competing in the NMS, so that certainly accounts for part of the gap, but there's no way it could account for most of it.

Regarding the relative performance of elite vs. average Chinese, Weyl actually came up with some interesting estimates in one of his books, based on the distinction between "old Chinese" and "new Chinese" surnames, corresponding to early immigrants vs. post-1950 ones, the former being pretty average and the latter usually highly-selective. Generally, at the higher end (e.g. NMS), the performance ratio was a factor of 2-3x, which seems pretty plausible.


Hmmm, based on a definition of South East Asian of Asian West of Bangladesh and South of Canton, based on demographics, and assuming no major age differences, we would expect a total of 23.9% of Californian NMS semifinalists to be South East Asian, of which would expect, of the total:
13.54% to be Filipino, 5.94% to be Vietnamese, 0.94% to be Cambodian, 0.84% to be Hmong, 0.64% to be Laotian, 0.62% to be Thai, 0.36% to be Indonesian, 0.17% to be Burmese, 0.05% to be Malaysian.

Meanwhile, of the other Asians, we should expect, with no differences in age structure, 5.42% Indians, 4.6% Koreans, 4% Japanese, 0.5% Pakistani, 0.1% Sri Lankan, 0.1% Bangladeshi, 0.05% Nepalese, 0.05% Mongolian, and 0.025% Okinawan.

Within foreign born educational attainment structure, the Southeast Asians are not too dissimilar to White Americans, in contrast to the greater educational variance amongst East Asian foreign born and South Asian foreign born, but I do wonder if a lot of their high educational attainment persons are really Chinese - e.g. it is estimated that half of the Vietnamese are Chinese by ancestry, and perhaps those might be more likely to be the ones with university educated parents - which would help explain some more of the Chinese advantage relative to other East Asians.

That and the age structure (and how it relates to educational attainment of migrants) seem like the remaining questions. They could work together.

I do really think a more elite Chinese migration sounds more likely than a kind of Social Darwinism in China but not Japan theory (and a Social Darwinism in both theory seems less likely than either of those).

I mean, we have Taiwanese (Chinese neither urbanised, immigrant, nor Commie). They're pretty good, but... "Thousands of years of Social Darwinism, and I got out of it was this lousy PISA advantage over the Japs that's equivalent to advantage of the the Dutch over the Germans...."!

M said...

PoW: Ironically, today’s Japan , technologically the strongest of the East Asia 3 (China, Korea, Japan), was actually the weakest for most of the history.

The conventional story I hear is that, after about the Ming or early Qing dynasty, Japan started having the best developed proto-industrial systems (on a variety of metrics, such as branded goods, well developed markets, pasteurisation of products).

The normal attribution for this is that the water transport was naturally a much more efficient form of transport in Japan, and that the country was mostly run by practical military men, increasingly of a conflict avoidant caste (after repeated bloody civil war), who didn't restrain commerce in favor of promoting an idealist agriculturalist way of life in the way that the pointy headed statist Mandarins of China and Korea were prone to.

Europeans merchants supposedly most tended to be most impressed with Japan for this (while the pointy headed Enlightenment Absolutists tended to like China better).

There might be no truth in this, but this is the way I've tended to hear it.

Sean said...

Post 1950 immigrants from China being an particularly elite group sounds plausible, and it makes trying to peg the Japanese relative to the Chinese through test scores in the US redundant.

Intermarriage could explain quite a lot; the more successful Japanese men may tend to marry out and have lower IQ offspring, and we don't know how many educated Japanese men get excluded from reproduction altogether by the US mating market (as happens to black women).

I have read that Japanese used to be mocked in China as undersized ('dwarf pirates', 'shrimp barbarians'). Hardly evidence that poor Chinese found it easier to feed their families than poor Japanese or that the social dynamics were similar.

Moreover - and this is a very telling point I think - Japan has repeatedly been able to produce technological products that were the best in the world. (EG in WW2 their destroyers and torpedos. Today their silicon wafers are the gold standard. Wouldn't Japan be more backward if Ron's idea is the main explanation for Chinese IQ?

If Ron's disinterested crusade for merit based policy succeeds, it will feed back and re-enforce itself. Consider, ever increasing waves of brainy Chinese settling in the US; ever increasing restriction on non-Chinese enrolment at elite universities.

PoW said...

“Moreover - and this is a very telling point I think - Japan has repeatedly been able to produce technological products that were the best in the world. (EG in WW2 their destroyers and torpedos. Today their silicon wafers are the gold standard. Wouldn't Japan be more backward if Ron's idea is the main explanation for Chinese IQ?”

Yet again, Sean, I’m not picking on you but it seems to me that this very telling point is still based on a snapshot. By the very same logic another very telling point must be that North Korean IQ=70 I wonder. What are silicon wafer? Probably North Koreans have never even heard about it, let alone making it. lol.

Now seriously, no one argues here that Japanese are dumb. It’s not about absolute but all about relative – i.e. at which starting line with what achievements. Actually, silicon wafers are still not the best of Made-in-Japan (for that matter ASML is not far behind either), material science is. On military, Japan’s anti-submarine related technologies are arguablely the best in the world. On many advanced material technologies Japan today is way ahead of the rest of the world, yes, way ahead. Mitsubishi, Sumitomo alike behind all these have accumulated industrial knowhow traced way back to mid-19 century!

But this is logical. I would have expected nothing short of it from Japan, given its 100+ years accumulated industrial knowhow. Since Japan’s average IQ is about at least the same as any Western colonial power, with about the same headstart in industrialisation, without any interruption of any crazy Japanese civil war or Communism Maoism in between, with free and complete access to int’l capital, int’l technology exchange, natural resources, and overseas export markets particularly after WWI and WWII, will anyone seriously entertain the idea that it won’t produce some technological world champions at least in line with its industrialisation peers at a time such as Germany or England?

But that is not the point.

The point is:

Japan: materials science, semicon wafers, etc, based on 100+ years industrialisation.

S Korea: Samsang (the gold standard in handset at least?), LG, etc, based on about 50 years dedication industrialisation

China: Huawei (the gold standard in the incoming 4G telecom equipments), ZTE, etc. based on less than 20 years industrialisation.

North Korea: Kim Jung-Un (the gold standard in Maoist Communism), Mig-21s and J-7s (donated by the Soviets and China, but maintained by N Korea), etc, based on almost 0 year industrialisation.

Also snapshots? yes, but at least they are extended snapshots across 100 years with their own perspectives.

blastomere said...

Anonymous (March 17 4:35:00) said...

So an apparently less efficient agricultural method (in terms of area - grass fed cattle being less area efficient than cereals) was much better for market distribution of food

Some related points Szabo makes:

[a]Per-acre productivity is very different from labor productivity. East Asia tended to have much higher per-acre productivity (in no small part due to ecological advantages alongside their far lower population of large meat and draft animals), whereas Western Europe had higher labor productivity, which meant a greater percentage of labor available for non-agricultural purposes (war, industry, etc.)

[b] Western Europe made far greater use of draft animals than East Asia. In East Asia, lower-class humans transported the vast majority of food and fuel to market, transported the upper classes around cities, etc., jobs for which Western Europeans almost entirely used draft animals.

[c] As suggested, this is related to genetic differences between Western Europeans and East Asians -- the most blatant, but hardly the only, one being lactose digestion in adults. I would also expect IQ studies would show a higher variation in East Asian societies, due to their greater stratification, to the point where the lowest classes were used in for the same jobs Western Europe used draft animals. A corollary is that Western Europeans were more likely to feed their draft animals and let the lowest classes starve. Which, if any, effect led to a greater Darwinian selection for higher average IQ is IMHO more a matter of speculation.

blastomere said...

Chinese have been mainly selected for passing exams that produced a narrow type of intelligence.

Highly implausible. The vast majority of Chinese over the vast majority of its history, like the vast majority of other agricultural populations, lived, bred, and died on their farms. Only a small minority were literate much less in a position to take the famous exams.

blastomere said...

Comparing Japanese-Americans and Chinese-Americans in California is subject to huge cultural variables.

For example, quite recently in California history (WWII) the Japanese were highl persecuted -- in a fit of war hysteria the whites in charge of California back them basically stole their land and shipped them off to concentration camps. So the few Japanese left living in California are those too stupid, or too poor, to have fled.

By contrast, China has never been a major enemy in wartime, and overseas Chinese were strong allies both in WWII and the Cold War. California, already disproportionately a magnet for non-white immigration, has long had large and dense communities of Chinese-Americans, making it an even stronger magnet for (very disproportionately elite) Chinese immigration.

UncleTomRuckusInGoodWhiteWorld said...

What about Japan...the social system in Japan historically was not like that of China or Korea, it was much more like Medieval Europe actually.

Did Japan just benefit from gene flow from from China (largely through Korea)?

Anonymous said...

I think the focus on the Chinese exam system is missing the wood for the trees. That specific system is just a symptom of a specialized society and a specialized society - unless restricted by caste - always ought to lead to an increase in average IQ imo simply through creating increased competition for those specialized occupations.

If you have a society where 5% of the population perform specialist functions with an above-average cognitive need and a second society where it is only 2% shouldn't you expect the first society to gain a few IQ points over time?

I think the general pattern is:

1)latitude provides a base average IQ in non-specialized societies

2) civilization leads to specialization which provides a boost to the baseline IQ through competition for those specialized occupations proportional to the level of specialization. It's not just the number of mandarins or even local officials. It's the number of innkeepers, artisans, spice merchants etc all boosting the average by a few points.

If so you'd get

1) mid-latitudes
- medium latitude
- early civilization

2) China
- high altitude
- early civilization

3) northern europe
- high altitude
- late civilization
- (partially compensated by height gain from animal based agriculture?)

Caste-based specialization e.g. India generally or Jews in Europe would also increase the average IQ but it would be skewed within the different sub-populations.

Sean said...

China has economies of scale; with associative mating their colleges and high tech facilities are now breeding grounds for a teeming super intelligent knowledge caste.

That caste would swarm into America and its elite universities to take advantage of meritocratic (exam result based) admissions to Harvard ect. An accelerating process of displacement of 'white' students would follow.

Someone needs to remember how under represented the US ethnic majority already is within the 'white' students at elite universities, and think through the implications.

Anonymous said...

[a]Per-acre productivity is very different from labor productivity. East Asia tended to have much higher per-acre productivity (in no small part due to ecological advantages alongside their far lower population of large meat and draft animals), whereas Western Europe had higher labor productivity, which meant a greater percentage of labor available for non-agricultural purposes (war, industry, etc.)

Yes, I believe traditional East Asian agriculture was very labor intensive but very high in per-acre productivity. This is noted in F.H. King's "Farmers of Forty Centuries" and in other sources. Also I believe there were many small plots that individual farmers owned or leased and would farm. Presumably farming ability was a major factor behind reproductive success.

Incidentally, the Chinese character for "man" is a picture of a field and a plow:

http://www.theworldofchinese.com/2012/06/on-the-character-%E7%94%B7/

Highly implausible. The vast majority of Chinese over the vast majority of its history, like the vast majority of other agricultural populations, lived, bred, and died on their farms. Only a small minority were literate much less in a position to take the famous exams.

Yes, also note that elites and urban populations live off of agricultural surplus. And that urban areas are population sinks. Any reproductive advantage enjoyed by urban elites would have had to have been balanced by rural fertility. Otherwise, the urban elites would have ended up eating the seed corn. This is probably why farming held relatively high social status in traditional Chinese society.

It's only really over the last century that rural fertility has been decimated via the spread of communism and capitalism and political centralization which favor elites and urban populations over rural populations, and via media technologies which allow elite and urban status advantage over rural populations to be leveraged significantly.

Ron Unz said...

One additional datapoint. I checked Weyl's 1989 book on American achievement, and in the late 1980s Chinese were over-represented as NMS semifinalists relative to Japanese by about 500%.

That was prior to the huge recent inflow of highly-selected Chinese immigrants, so it tends to support the idea of a significant Chinese/Japanese academic performance gap. Koreans also outperformed Japanese by a comparable margin.

Anonymous said...

That was prior to the huge recent inflow of highly-selected Chinese immigrants, so it tends to support the idea of a significant Chinese/Japanese academic performance gap. Koreans also outperformed Japanese by a comparable margin.

I believe Japanese immigration into the US predominantly occurred during the early 20th century. Mainly poor farm laborers to work on plantations in Hawaii and California. During that time Japan was industrializing and its economy was expanding. It annexed Korea and had aims of expanding into Manchuria and China. So there were increasing opportunities for people and less incentive to emigrate except for the very poor and unskilled.

China by contrast was in social and political turmoil during that time and a wider variety of people including the skilled and able would have had incentive to emigrate.

I'm not sure you can discount selection effects.

Anonymous said...

"Presumably farming ability was a major factor behind reproductive success."

This doesn't necessarily follow if marriages were arranged by families at a young age according to clan-based status rules.

If you follow hbdchick's posts on clan-based marriage structures one of the striking things is how these structures seem almost designed to prevent selection i.e. keep land within the clan no matter what.

Anonymous said...

Peter and Ron:

I made this comment over at Unz's original article, but was interested to see what your response would be here, since it didn't get one there.

Ron:

Hmm…. If heritability of IQ is not that high, and you can systematically identify the heritable components of IQ, then selecting on them, even though it is not meritocratic, will increase the mean IQ faster than selecting directly on the trait.

A person from a smart family who is not themselves smart is likely to have smart genes, but to have got an unlucky shake of the environmental (random development) stick.

If you’re interested in raising mean smartness, then you might do better to help a person have kids who is dumb, but likely from random developmental factors, than some meritocrat who is smart, but from environmental factors.

There might be an ideal level of meritocracy for Darwinian Selection – if someone is an outlier for their family background, its likely to be for environmental (read: random) factors. Such a person isn’t going to pass their specialness on to their kids.

A dumb Darwin-Wedgwood might be a better prospect in terms of passing on the genes for intelligence than a smart meritocrat!

That doesn’t mean the smart meritocrat doesn’t *deserve* more wealth than the dumb Darwin-Wedgwood!


So, it would seem the optimal system would combine some belief in "breeding" (that is, look at a person's family to try and generate a prior over whether they are an outlier due to random environmental factors they won't pass on).

Some system's may be *too* individually meritocratic to maximize selection for intellect....!

Anonymous said...

This doesn't necessarily follow if marriages were arranged by families at a young age according to clan-based status rules.

Well it would follow unless they were somehow able to survive and reproduce without eating food.

Anonymous said...

If you follow hbdchick's posts on clan-based marriage structures one of the striking things is how these structures seem almost designed to prevent selection i.e. keep land within the clan no matter what.

If your ability to produce food is used to feed your own family and clan members, then you are promoting selection for the ability to produce food since they share more of your genes than others.

You would be impeding selection for food production if your ability to produce food is primarily used to feed strangers in a large nation-state or empire or people abroad in another country or on another continent.

Peter Fros_ said...

Given the many comments and my lack of time, I can only offer a few observations:

PoW,
"The fair comparison, IQ wise, or achievement-wise, is to randomly take some of Chinese and Jews, drop them in the middle of a third country (where neither is a native speaker), ideally non-White and non-east Asian country, then you see who will come out on top academically and technologically years later."

- That experiment has been done in a number of Caribbean and Latin American countries, e.g., Brazil. Both groups do well.


"The 10,000 Year Explosion" notes that in parts of the Middle East, Jews were relegated to certain low status occupations such as cleaning toilets"

- The oldest census records from Poland and Algeria show little difference between Ashkenazim and Sephardim in terms of occupational distribution. Craftsmen predominated in both cases. The big difference was the population boom that caused Ashkenazi Jews to increase twenty-fold between 1650 and 1900. That boom was concentrated among craftsmen who worked in family workshops and who increased their workforce mainly by marrying earlier and encouraging their own children to marry earlier. There was thus a feedback loop between business success and reproductive success.
See my post at:

http://evoandproud.blogspot.ca/2007/11/natural-selection-in-proto-industrial.html

Ron,

I'll write about that point in my next post. I believe the Clark model is more applicable to Korea and Japan than might seem at first sight.

Anon,

The distinction between "farmer" and "merchant" was not clearcut. Farmers often engaged in commerical activity, particularly as wholesalers.

Anon,

We only have limited IQ data for Mongols and other northeastern Asian peoples. If we look at Russian PISA scores, the Buriat Mongols tend to perform below the Russian average. See:
http://akarlin.com/2012/08/18/analysis-of-russias-pisa-2009-results/

Sean,

As I understand it, the Chinese are also class-conscious when choosing marital partners. I think the main factor is that both the imperial exam and Clarkian population replacement were selecting for certain mental abilities that tended to downplay innovativeness and risk-taking.

M,

"Asians" are a very heterogeneous group, especially in terms of intellectual performance. I would crunch the data only by national origin.

Blastomere,

"Only a small minority were literate much less in a position to take the famous exams."

And a very large proportion of those people never had any children. The majority aren't necessarily the ones who drive the evolutionary process.

Anon,

Heritability of intelligence is quite high. In North American society it's around 60%. In China, heritability is probably higher among people of the same social class, since that social environment would be more homogeneous than the North American one.

Anonymous said...

The distinction between "farmer" and "merchant" was not clearcut. Farmers often engaged in commerical activity, particularly as wholesalers.

Yes, farmers did sell their produce. But there is a clear distinction between farmers who sell their produce and merchants who specialize entirely in commerce. They're quite different activities and should select for different things.

Sean said...

It seems to me that Ron is substantially correct. However his position on intelligence is a moving target. He makes a very good case that there is a something special about China, and that among immigrants to the US there is a Chinese/Japanese academic performance gap, for the reasons he outlined all those years ago. Yet, he does not accept that low IQ in Ireland (RC celibacy, ethnicity?) was hereditary, and he insists descendants of the Irish immigrants to the US seem to have greatly increased IQ relative to those descended from English immigrants.

Peter: "most people of Irish origin are only part-Irish (including myself). Yes, one can find people who are sure of their Irish ancestry and who can produce genealogical records to prove that their ancestry is entirely Irish. But such people would tend to be better educated... a biased sample"
18 percent of marriages by Chinese- and Japanese-Americans were to American Jews and "it is Asian heritage that loses in such marriages. Jewish ethnic identity and Judaic religious practice characterized most of the 31 intermarried couples they studied in depth, even though only five Asian-American spouses converted. The Jewish attachment seemed to deepen for those couples who had children."


Selection for IQ in the Japanese upper classes may have been of a 'Mating Mind' type:- " EDUCATION for the upper class women was a priority. Poetry recitation and improvisation were important too, with poetry the expected form of communication. Women were expected to know 1000 classic Japanese verses by heart. A woman's elegance was determined by her ability to quote the perfect verse for any occasion. Competitions were held between courtiers to see who could write the perfect verse about the weather, the most notable subject matter. A well-turned pun was the height of elegance." Going by Edo period art, there was reduced interest in secondary sexual characteristics; in aristocratic circles at least. See Here.

(About Asian Jewish interracial marriage . Supposedly all due to 'shared commitment to “tight-knit families, hard work, and educational advancement,”'. But why is it that in so many of such couples - we know about - it's the male who is Jewish, even when he is the obscure one - like Tiger Mom's husband - if not because of the relative masculinity of men from different ethnic groups? IN Japan a quarter of unmarried men and women between the ages of 30 and 34 say they are virgins.)

M said...

Peter: "Asians" are a very heterogeneous group, especially in terms of intellectual performance. I would crunch the data only by national origin.

Thing is, I've tried this approach (disregarding migrant background and assuming each is a representative sample of their nation) with software*, and taking representative population proportions of the various Asian subgroups, assuming IQs with an SD of around 15 and which are about a SD apart between sterotypically high performing and low performing Asians (respectively high East and low West and South) does not work to generate the kind of SD the Asian Americans show on the SAT and other tests. To make this work (and generate the +3 IQ point bonus needed for Asian Americans) it would need both

- East Asian Americans to performing like Ashkenazi Jews (not gonna happen without a selected subsample)

- and more crucially West and Southeast Asians (and probably South Asians) to perform equal or worse than African Americans and Mexican Americans

I regard this as as utterly improbable (data like the PISA and direct examination of unselected Southeast Asian scores in Western countries offer absolutely no support for this), and it is completely incongruous with their levels of educational qualifications (particularly for a South Asian American low performing scenario - even if you aren't interested in the Census data, even the Filipinos have around a 50% Bachelor's degree or higher qualification rate, compared with a White American average of around 30%) and the fact that there is no observable Affirmative Action phenomena for low performing Asian American subgroups. And Taiwanese in Taiwan and Japanese in Japan certainly do not perform with +1SD to European Americans.

Taking the actual educational qualifications of foreign born Asians Americans and USA based Asian Americans (which are to a large extent one and the same) and estimating educational performance on the basis that they are close to White American performance at each level (perhaps the lower performing groups are slightly lower and the high performing ones slightly higher), that does generate the large SDs seen for Asian Americans in the SAT, and the extreme overperformance at the high end, which the modest improvements at the mean would not predict.

No need to go over it further, suffice to leave it that I find it very improbable that the (verifiably) broader spectrum of educational backgrounds among Asians in America doesn't have a very strong role to play in their extreme overperformance and wide range of outcomes.

* The method here is generating normal distributions using Excel's norm.dist function for each Asian subgroup (each with an SD of 15), pooling their results in weighted proportions (as a function of their relative size as a % of all Asian Americans) and then running a mean and SD calculation on the pooled results.

As compared against running a mean and SD calculation against the Asian and European American populations where they are assumed to have IQs in line with their distribution of educational qualifications.

Anonymous said...

Heritability of intelligence is quite high. In North American society it's around 60%. In China, heritability is probably higher among people of the same social class, since that social environment would be more homogeneous than the North American one.

40% environmental (which as far as we can tell, isn't really about class or a similar social environment but is simply "random as far as anyone knows"*) seems like plenty of grist for this kind of effect, except applied to the business of mating **

I am hiring for a job, and there are two applicants that scored 140 on the IQ test that I gave them. I have found that the correlation between IQ and job performance in my shop is 0.70, that is to say the test is less than perfect. If I know nothing about either applicant save the test score then I can toss a coin. But imagine, instead, that I learn that one of the high scorers is an American Black guy, the other one is an Ashkenazi Jew. I can use this additional information to make a better choice, like this:

The average IQ of the US Black population is 85, the average IQ of Ashkenazi Jews is 110. My best prediction is that in terms of job performance there will be regression to the mean, but knowing the group membership of my applicants means I know that they will regress to different means.

My best guess now about the future job performance of the Black guy is 80 + 0.7 * (140-85) or about 119. My best guess about the future job performance of the Ashkenazi guy is 110 + 0.7 * (140-110) or 131. Bottom line: use the information about ethnicity and hire the Jew.


* http://westhunt.wordpress.com/2011/12/28/rischs-conjecture/#comment-398 - Harpending

** http://westhunt.wordpress.com/2011/12/28/rischs-conjecture/#comment-391 - Harpending again

Anonymous said...

"Well it would follow unless they were somehow able to survive and reproduce without eating food."

"If your ability to produce food is used to feed your own family and clan members, then you are promoting selection for the ability to produce food since they share more of your genes than others."

In the WEIRD environment where a nuclear yeoman family owns and runs their own farm there is direct individual selection on individual ability: for good or ill as bigger winners also means bigger losers. I think that was probably an important factor in Europe.

However in a clan-based system someone who turns out later to be a particularly good farmer may have been already married off at fourteen to a specific 1st cousin decided at birth on the basis of a traditional marriage pattern and their skill at farming goes to benefit the clan as a whole because it's the clan's land not the individual's. So it seems to me there is much less selection in a clan-based system - by design - as it's a safety net as well.

(I'm not saying there isn't any selection just that i think there's a danger of projecting the highly selective WEIRD yeoman farmer model on the rest of the world.)

And anyway there were farmers everywhere. If there was something different about farmers in China it would have to have been unique.

.
Personally i think it's population density and it's various side effects.

China had lots of big river valleys leading to high population density along those valleys and the consequent ability to support lots of specialists with the consequent individual competition for those specialist spots.

If you have a river valley that can support 10,000 people, 10% specialists that's 1000 specialists. If you have a river valley that can support 100,000 people, 10% specialists that's 10,000 specialists. More competiton.

If so then the other candidates for a similar high density specialization effect would be India, the fertile crescent and Egypt.

India also had the big river valleys but the caste system sealed off the benefits of competition into specific castes.

The fertile crescent had the river valleys and a head start but were stymied by the Arab conquest?

Egypt ditto.

Reasonably plausible imo.

.
I think you could probably show IQ distribution in China following the big river valleys if you mapped IQ data on a sub-provincial scale small enough to follow the big river valleys

http://www.atimes.com/atimes/China/images/china-rivers.gif

I think you can see hints of this on the provincial scale map here

http://1.bp.blogspot.com/-BCzzIHsWAHc/UB0qu7TWRMI/AAAAAAAAAWI/k81dGOT_gLY/s1600/China's+iq-by-province2.jpg

where it seems the relatively lower IQ hole in Fujian could be caused by relatively higher IQ along the Pearl and Yangtze river valleys.

(For this reason whatever the average Chinese IQ in America is i don't believe it will be the same in China as a whole. I think it will be high along the three river valleys to Beijing, Shanghai and Hong Kong but lower in the gaps. Still high but not as high except along the valleys.)

.
So i think the IQ story will be one of competition but different kinds of competition in different places over different time scales with the Chinese version of the story being very broad-based over the whole population (at least along the big valleys) and over a *very* long time. The Jewish story being much more hot-housed during the period of banking monopoly in between the fall of Rome and the expulsions from western europe and the Euro story being mostly rooted in ancient high latitude survival until relatively recently in civilizational terms followed by more extreme competition over a shorter period of time.

Anonymous said...

"He makes a very good case that there is a something special about China, and that among immigrants to the US there is a Chinese/Japanese academic performance gap, for the reasons he outlined all those years ago. Yet, he does not accept that low IQ in Ireland (RC celibacy, ethnicity?) was hereditary, and he insists descendants of the Irish immigrants to the US seem to have greatly increased IQ relative to those descended from English immigrants."

One way of squaring that circle would be if a population's average IQ was a combination of positive selection effects and genetic load effects.

Then if the Chinese had already shed as much genetic load as was possible without genetic engineering they'd already be at their peak when they arrived.

If on the other hand the Irish had had mostly the same positive selection effects as their neighbors but had more genetic load for some reason they'd be off their peak - reaching it if they subsequently lost that load in the US.

Again personally i think it will turn out to be a function of population density in combination with marriage form.

Anonymous said...

However in a clan-based system someone who turns out later to be a particularly good farmer may have been already married off at fourteen to a specific 1st cousin decided at birth on the basis of a traditional marriage pattern and their skill at farming goes to benefit the clan as a whole because it's the clan's land not the individual's. So it seems to me there is much less selection in a clan-based system - by design - as it's a safety net as well.

I don't see why being effectively taxed by one's fellow clan members rather than by strangers in the form of feudal lords or the Church would lead to much less selection.

Anonymous said...

And anyway there were farmers everywhere. If there was something different about farmers in China it would have to have been unique.

There were farmers everywhere but farming has varied in different ways. There have been societies where farming was done by slaves. In other societies it's been done by independent or relatively independent farmers. It's varied in terms of intensity, productivity, sustainability, etc. I believe unsustainable agriculture contributed to the desertification of the Middle East.

Anonymous said...

This book on China discusses traditional Chinese agriculture:

http://books.google.com/books?id=oPbbaJOpWLwC&lpg=PP1&pg=PA168#v=onepage&q&f=false

"The way the Chinese farmed in the past was so different from European or U.S. farming today that an effort of imagination is needed to understand it. Today it remains very unlike most of the farming in the West.

Chinese traditional farming was close to gardening. It produced high yields on tiny farms by an immense input of skilled labor. This skill was not based on scientific knowledge, but on a detailed practical understanding of local conditions: soils, weather, and types of plant. It was cautious. Too much was at risk for peasants to be very adventurous. But it was not conservative. There was constant experimentation and selection of new strains of plants and animals. Successful innovators were swiftly copied.

During the medieval period, when Inner China was being opened up to agriculture, large landowners who managed at least part of their farming directly had an important function. They had the resources to act as pioneers. Since the 16th or 17th century small farms were generally more productive than large estates. The individual peasant and his family, whether they owned their land or rented it, brought a personal commitment to this garden-type agriculture that hired or servile laborers on large farms could not match. Under pre-modern conditions this commitment more than outweighed the economies of scale available to operators with a managed labor force. With few exceptions, Chinese landlords found it most profitable to lease out their land in parcels that rarely exceeded 3.5 to 5 acres (1.5-2 ha)."

PoW said...

March 19, 2013 at 1:35:00 AM EST
Anonymous said...

& March 20, 2013 at 3:56:00 PM EST
Anonymous said...


I think you are the same person. I was trying to respond to your march 19 comment yesterday and was busy… only wrote the following half a point:

Agreed.

And I tend to add one 2 more factors:

1, [B] population, both quantity and density [/B], which intensifies the competition.

(Note: “Western Europe made far greater use of draft animals than East Asia. In East Asia, lower-class humans transported the vast majority of food and fuel to market, transported the upper classes around cities, etc., jobs for which Western Europeans almost entirely used draft animals ” ---- this was primarily also due to different comparative advantage, namely population size, each had and has at a time. When you have something in abundant, it become cheap regardless its intrinsic value.)

Latitudes provide baseline IQ  earlier intensive agriculture more food earlier and larger citiesmore population and higher densityearlier specialisation, wider specialisation  more competition more wealth more intensive farmingmore population… etc…a self-propelling cycle.

2. [B] Relative social stability [/B] compared with the rest due to earlier “unification” by the first Emperor of China. Comparatively speaking, unification = more stability ( a point also helps more population – see above) = more risk averse ( a political point become cultural point which aversely affect innovations)

Both these 2 points were assisted by China’s relatively caste-free and slave-free society, larger areas of mild latitude terrains at the doorstop for expansion, further stimulated and stabilised by top-down and bottom-up Confucius meritocratic system which comparative peers such as Roman Empire lacked. Of course, imperial exams and ‘FenJia” etc. were important elements, but comparatively they, like some others, were trees, not the wood.

--------
--------

More points today:

i. overall I think that because people tend to underestimate the cold weather in the very beginning which played the decisive part in IQ -- yes, everyone know that , but they STILL underestimate it in practice somehow -- , and they overestimate the society/social factors. (Only tiny group of very ethnically aware and tightly organised tribes such as Ashkenazi Jews who were lucky enough living among some of the greatest technological powerhouses – from the old day Roman Empire, to 18th
and 19th century England, France, Germany etc could develop some IQ points in such a short span of time by narrowing themselves into 1 or 2 specific sectors.)

ii. Compared with Ancient China, Egypt and South Asia (not the post-British Raj modern concept of today’s unified “India”, but countless very small-sized kingdoms) had tropical latitude terrains for expansion, but lack of #1 most important baseline IQ to start with. Their tropical weather, unlike the clear 4-season one, is not necessarily ideal for agriculture either. Neither of them seemed to have enjoyed “unification” and “population” on the scale and span of China. On top of that, they had slavery, deep caste divide, far less innovations which lead to far inferior technologies that led to far more open to be attacked militarily, etc. Therefore, they lagged far behind.

iii. The only place where all the major conditions were available to compete with China was Europe. But Alexander de Great was not great enough like the First Emperor of China... Europe therefore lacked both population (size and density) and stability unfortunately. The flip side of the coin, fortunately, it had more flexibility, less rules, more open, less risk adverse, plus pro-longed inter-state competitions  all the major elements for occasional blips of periods of big innovations such as Industrial Revolutions.

Ron Unz said...

In the WEIRD environment where a nuclear yeoman family owns and runs their own farm there is direct individual selection on individual ability: for good or ill as bigger winners also means bigger losers. I think that was probably an important factor in Europe.

Well, I don't claim to be the world's greatest expert on Old China, but I have read dozens of books and major articles about their traditional rural society, and I just don't recall seeing much mention of clans as an overriding social factor. Land was generally owned and worked by individual, nuclear families, having been divided up equally among all sons, sometimes before the father's death and sometimes after.

Now brothers sometimes helped brothers, cousins sometimes helped cousins, and clansmen sometimes helped clansman. But very often they feuded, fought, oppressed, or cheated each other. I'd guess that the same was true in most European peasant societies. "Clan solidarity" may have been a Chinese ideal, but "Christian brotherhood" was similarly a European ideal.

One important fact to bear in mind is the village was the basic Chinese demographic unit, and in a good fraction of all villages the vast majority of the inhabitants were from a single extended family, or perhaps just two or three. If almost everyone in your community was a Wang, what exactly would be the benefits of "Wang clannishness"?

Also, I think the custom was to almost always take wives from other villages, and that marrying someone of the same surname, even if totally unrelated, was considered improper/illegal/incestuous. So if nearly all wives were from other villages and other clans, that might tend to put a damper on the growth of strong "clannishness."

Again, I'm really not a true China expert, so perhaps someone who is can correct me about any of these points.

blastomere said...

"Only a small minority were literate much less in a position to take the famous exams."

and a very large proportion of those people never had any children.

What was this "very large proportion?" Evidence please, that the reproductive rate of the 95%+ of the population that engaged in predominantly manual labor was substantially less than the less than 0.1% who passed the exams, and that this resulted in a substantial selection effect for the population as a whole?

If anything, the selection effects were likely much more severe on the lower end, among the most poverty-stricken where people frequently died of malnourishment and disease. Sexual selection being the dominant form of selection is a largely post-Mslthusian phenomenon among humans.

blastomere said...

PoW:

Latitudes provide baseline IQ

I tend to agree with you about the importance of differences in baseline genetic endowments among various hunter-gatherers at the dawn of their respective agricultural revolutions. These are a crucial part of our evolutionary history and have become too neglected of late. We didn't start our different agricultural revolutions at the same starting line. We started from a wide variety of very different genetic and cultural positions. (Many didn't start at all until after Columbus).

One interesting related piece of evidence to consider is that, given well-bred crops appropriate to the climate, agriculture nearer the tropics is far more productive per acre than at higher latitudes. This is a big gaping hole in Jared Diamond's selective use of ecological evidence. For example look where ethanol can actually be efficiently produced today, without subsidy. Brazil, with a climate-appropriate crop, sugar cane. The growing seasons are too short at North America latitudes, and the crops adapted too our climate are not efficient enough converters of solar energy, to cost-effectively produce ethanol.

Similar goes for southern and central China, BTW. You can have a far more intensive agriculture and denser population there than in northern Europe simply because of the much longer growing season, the greater productivity of crops like rice adapted to warmer climates, and the greater nutrient availability in the large flat river valleys. Differences in IQ or social institutions aren't necessary to explain the dramatic differences in per-acre productivity. (Whether the causation goes the other way around is debatable too).

But peoples nearer the equator tended to lag or fail at all to develop fixed agriculture that incentivizes the breeding of climate-appropriate crops. Quite possibly for reasons related to the pre-existing genetic endowments in these different hunter-gatherer ecosystems, and with some correlation to general latitude and climate, that you discuss.

That said, the impact of agriculture, where it did occur relatively early, on human genetics is often clear and fascinating. However dwelling on abstractions like IQ tends to obscure some of the most important, clear, and dramatic phenomenon in this area, such as the development of adult lactose digestion in humans, related genetic changes in the corresponding cattle, and the resulting ecological and cultural ramifications that Szabo has studied and described.

Anonymous said...

"I don't see why being effectively taxed by one's fellow clan members rather than by strangers in the form of feudal lords or the Church would lead to much less selection."

A yeoman farmer owns or at least controls their capital (land).

A clan farmer is working clan land.

Imagine the same piece of land. In one case it's split into four yeoman farms with a nuclear family each and in the other case it's farmed collectively by the four families as a clan.

In the first case if one of the yeoman farmers is a good farmer and one is a bad farmer then the good farmer can take the profits he makes (after tax) and gradually buy the land of the bad farmer.

In the clan-based case the skill of the good farmer benefits the clan as a whole and balances out the lack of skill of the bad farmer.

The system that developed in parts of northern europe in the middle ages was more Darwinian imo.

Regardless of the rights or wrongs of this specifically i think it's important to keep in mind that the standard form of farm ownership in the WEIRD countries was different to elsewhere.

Anonymous said...

"The individual peasant and his family, whether they owned their land or rented it, brought a personal commitment to this garden-type agriculture that hired or servile laborers on large farms could not match. Under pre-modern conditions this commitment more than outweighed the economies of scale available to operators with a managed labor force. With few exceptions, Chinese landlords found it most profitable to lease out their land in parcels that rarely exceeded 3.5 to 5 acres (1.5-2 ha)."

Of course if China *did* have yeoman farming as described above then it would have had the same selective effect :)

Sean said...

CLARK 2012: "Elite surnames of the Imperial Era are still slightly overrepresented among modern elites in Communist China. " . [...] "Success in the exams did not only bring glory to the chosen candidates their common ancestor, but also protection for the property rights of the entire kinship. As they obtained more land and wealth, they had more resource to invest in education. In all, the relative status of kinships was more stable over generations than that of individual families (Hymes, 1986). For "outsider" kinships, it may take one or two generations to become rich and literate, and get into the lower gentry class (Shengyuan
, 生员/秀才, the lowest degree, 0.4-0.7% of male population in the late Qing). But it took several generations for a commoner family to produce Juren and Jinshi (5% and 1% of shengyuan, respectively). That is why it always took hundreds of years for an elite kinship to regress to mediocrity. [...] It thus remains to be determined whether it was kin networks, or just the nature of statistical processes, that produces the strong persistence of surname elites in China".