The Evolution Of Chinese IQ

In the discussion at the previous post, in which I took exception to Ron Unz’s theory of the East Asian Exception, he alerted me to so additional work on the matter he’d done as a Harvard freshman on Chinese IQ. You can read his summary of Social Darwinism and Rural China as well as Steve Sailer’s commentary on it.

Ron Unz’s Theory of Social Darwinism in Rural China

According to Ron Unz, Chinese peasants lived close to their Malthusian limits for millennia on end. That is correct. Furthermore, Chinese rural life was “remarkably sophisticated in its financial and business arrangements”, far more so even than in England. I do not have the comparative knowledge to offer informed commentary on this, though I would stop to note that such a system may not have been so much a generator of “selective pressure for those able to prosper” under complexity as a reflection of already high IQ’s. After all on most social, economic, and technological metrics China was far ahead of Europe until the 18th century or so (though there were important exceptions). Furthermore, “virtually all Chinese were on an equal legal footing”, with far fewer of the feudalistic or caste distinctions that proliferate in India and pre-Enlightenment Europe. This is also correct.

This environment included a number of mechanisms that promoted a highly eugenic development path for the Chinese population. Ron Unz says that only the relative affluent could afford their wives for their children. This is not quite correct, or should I say permanently correct, as this issue only heavily manifested itself during times of Malthusian stress, when families opted to kill baby daughters resulting in skewed sex ratios. Otherwise, we should note that Europeans within the Hajnal Line married late and that the poor sometimes didn’t marry at all, so this particular eugenic effect was if anything stronger in Europe.

However the biggest, and most specific to China, eugenic mechanism is argued to be the Chinese custom of fenjia 分家, lit. “family division.” So if, say, a wealthy Chinese family produced four surviving sons, each of them would inherit only a fourth of the family land. The brothers would be back to square one and would have to hustle for money again. A couple of the brothers might be successful and build up wealth again; another would fall into poverty, and the last one would fail to even find a wife and have children. The effect was that every generation, “a good fraction of the poor disappeared from the gene-pool.” As reproductive survivors would tend to be more intelligent and far-sighted, or so the argument goes, this selected for such traits within the Chinese population.

The system of meritocratic imperial exams, which enhanced the reproductive prospects of the very brightest who could pass them, was a further eugenic mechanism but one whose overall impact was “pretty small” compared with “the push from the bottom.”

Finally, Ron Unz compares his theory to Gregory Clark’s book Farewell to Alms, which argues for a eugenic mechanism in England in which the wealthy enjoyed greater reproductive success and, over the centuries, “civilized” the proles via genetic drift through downwards social mobility. As such, the traits of the aristocracy became inculcated in the English masses with all its attendant benefits, e.g. plummeting homicide rates. (This civilization doesn’t seem to have lasted very long however if yob culture and football hooligans are anything to go by). 🙂 He notes that these eugenic mechanisms operated in China for far longer than they did in England.

He also compares the selection pressures facing the Chinese with those that produced the famed intelligence of the Ashkenazi Jew. Unlike the latter, the Chinese didn’t only have to be bright and business-savvy; as a peasant, he also had to maximize “physical endurance, robustness, diligence, discipline, energy-consumption.” As such, selection had a less one-sided skew in favor of intelligence.

 My Critique

This is a nice and elegant theory. It has no obvious contradictions. He is planning to publish his analysis in a formal manner pretty soon. However, before he does so I hope that he will address some of the following counter-arguments and discrepancies.

Re-The (relatively) complex legal environment selected for intelligence. HOWEVER, the Chinese – as do East Asians in general – only perform significantly (hugely) better than whites on visuo-spatial intelligence. That is good for hunting mammoths in the prehistoric tundra and some aspects of mathematics, but not anywhere near as good for navigating complex legal codes in which verbal intelligence is key. However, Chinese verbal intelligence if anything lags the indigenous peoples of most developed European nations. According to 2009 PISA results, Chinese verbal (reading) IQ was 98, which was inferior to Germany’s 102, the US’ 101 (including Blacks and Hispanics), and Poland’s 100; and equal to that of Spain, Italy, Portugal, and Greece.

Here, ironically, Unz faces an additional dilemma: Either he has to reject his theory of the East Asian Exception (i.e. that the Flynn Effect barely applies to them), or he has to rethink his theory of Social Darwinism in rural China.

Re-The eugenic influence of fenjia. The model he sketches out is plausible enough on the surface. That said he has to account for several possible discrepancies.

Korea appears to have a max. potential IQ of about 107, while Japan is slightly lower. Did they have systems of land inheritance that also favored the development of IQ? I do not know. I hope Unz will investigate this matter. A potential problem, however, is that IF they did NOT have their own equivalents of fenjia, then it would be invalidated as a feasible explanation of why East Asian (including Chinese) IQ’s are so high.

Re-Comparison with George Clark’s theory. I don’t think this is a useful crutch to Ron Unz’s China theory at all. So supposedly England had this intensive genetic drift from the top to the bottom. However, today, UK natives (on PISA) score 101; in other IQ tests, the UK’s average is typically set to 100. These numbers are typically lower than those of the  Germanic countries like Germany, the Netherlands, etc. – and equal to the IQ’s of the Nordics, the Western Slavs like the Poles and Czechs, (Celtic) Ireland, and (Celtic-Germanic) France.

Really my critiques boil down to a few main issues.

(1) We need more comparative data on IQ, land inheritance systems in the past, etc. I strongly suspect that for all but a few exceptions (e.g. Ashkenazi Jews) the traits developed in prehistoric times still predominate above all others. After all, pre-agrarian prehistory accounts for 90%+ of homo sapiens sapiens’ existence; and selection pressures back then were FAR stronger because of small population sizes. Noncompetitive tribes got wiped out by hostile tribes or the vagaries of climate with chilling frequency. In medieval times, noncompetitive genes were far likelier to linger on to some degree, firstly because welfare systems – crude and rudimentary as they were back then (e.g. poorhouses; alms, zakat, etc; grain reserves; etc) – were still a league ahead of what can possible exist in a tribal hunter-gatherer society; secondly, because violent as the ancient and medieval periods were, they were vastly more peaceful (and populations were bigger) than was the case in the prehistoric era.

(2) To what extent was fenjia unique to China? Was is common to the East Asian region, or not? If not, why then doesn’t Chinese IQ greatly exceed Korea’s? Did it exist in Vietnam? If it did, why then is Vietnamese IQ substantially lower than China’s? Etc. Also, explain why these mechanisms didn’t result in a particularly high verbal IQ; after all, to understand legal matters, that is really what we need, no?

  • AG

    Great debate! You raise very important question. The east asian countries that use hanzi汉字, chopsticks, and practice confucianism, zen buddism are China (Han), Japan, Korea, Vietnam. At the same time, these countres also have higher national IQ exept vietnam. Other eas asians including Tibetan do not have impressive national IQ. Like rec1man mentioned, 5% of the Indian population is of Oriental race and they dont have a high IQ..

    Does Han Chinese cultues over thousands years influence on Korea, Japan make the difference in national IQ? Or high IQ people tend to embrace Han culture? This is chicken-egg issue. Fenjia might be part of han culture influence. But I do not know for sure.


    • charly

      Korea and Vietnam don’t use hanzi

      • Ricola

        Actually they do, hundreds of years ago. This is why historic texts in these two countries are all written in hanzi. Vietnam even invented a new script based on the hanzi called the Nom script, which is now obsoleted.

    • anon

      .’ Like rec1man mentioned, 5% of the Indian population is of Oriental race and they dont have a high IQ..’

      Rec1man is no authority. His arguments are full of holes and best to be ignored. He claims an IQ ~ 120 for Indian Brahmins which is laughable at best and around 35-45 points inflated. The oriental Indians (NE Indians ) have the highest literacy ~100% and better standards of living despite the hilly (and agriculturally unproductive) environment they live in. I do think they have IQ in range of Maynmar/SE Asians ~90 which is ten points above rest of India.

      Disclaimer: I am not a North East Indian.

  • Thanks for the generous comments, and you raise excellent points.

    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.

    First, I’d imagined that the extreme malthusian conditions of rural China so vividly described in the source accounts had been present for most of the last 2000 years. But that’s not at all clear. The last couple of centuries had been especially difficult times for the Chinese, and it might be incorrect to extrapolate that back to the more distant past. Indeed, there’s even a scholarly dispute about whether serfdom—a totally different system of land-tenure—may have remained widespread as recently as the Ming Dynasty. So the time horizon of my selection model may be much shorter than I realized.

    Even more seriously, I discovered that Japan’s land tenure system had always been strict primogeniture, nothing remotely like China’s fenjia. So the my model wouldn’t apply to Japan, and the Japanese seem about as smart as the Chinese. (I haven’t yet investigated the rural population pressure in traditional Korea, or its inheritance system.)

    Despite these very serious problems, I’m still inclined to try publishing the piece, partly because it was recently cited in the academic literature, and the existing version is such an embarrassment. But I’ll obviously emphasize the major flaws, and let people decide for themselves. After all, the severe selective pressure in rural China wasn’t solely focused on intelligence but probably upon a whole basket of traits, and the Chinese and Japanese differ in all sorts of other ways.

    And I agree that Clark’s evolutionary model for England suffers from similar problems, namely that he’s produced an interesting theory explaining why the English are smarter and longer-time oriented than all the other Europeans. Except they aren’t.

  • “…and that the poor sometimes didn’t marry at all…”

    I don’t remember anymore where I read this, it could have been the novel “Iceland’s Bell” about 17th century Iceland, it could have been somewhere else, but I remember reading about a social convention – not a law, but a universal understanding – that a marriage could only occur if a man was able to offer a woman the position of mistress (feminine of master) of her own household. Two servants living under their masters’ roof were not marriage material.

    “As such, the traits of the aristocracy became inculcated in the English masses with all its attendant benefits, e.g.plummeting homicide rates…”

    Crime might have been bred out of most Euro populations through the frequent use of hanging in past centuries. I think this is what Nietzsche believed.

    ” I strongly suspect that for all but a few exceptions (e.g. Ashkenazi Jews) the traits developed in prehistoric times still predominate above all others. After all, pre-agrarian prehistory accounts for 90%+ of homo sapiens sapiens’ existence; and selection pressures back then were FAR stronger because of small population sizes.”

    In their book “The 10,000 Year Explosion” Cochran and Harpending argued that the speed of evolution was increased by agriculture through two mechanisms:

    1) The switch to agriculture led to a population explosion. The more people per generation, the more mutations per generation, the more rolls of the dice per year. A certain percentage of all mutations are beneficial. Any mutations that are very beneficial will be quickly picked up by the whole population regardless of its size.

    2) New challenges. A radically new diet, radically new diseases, radically new social dynamics. More things to adapt to, hence the speed of change picks up.

    What made the Chinese smart? This is unfounded speculation of course, but isn’t it true that every group that engaged in agriculture in the northern and middle latitudes is smart? Remember, the Han came from northern China, the middle and lower Yellow River valley, which is at the latitude of central Spain. They expanded from there to the south in historical times, pushing out and partly assimilating SE Asian type peoples.

    The Mongoloid descendants of steppe nomads have unremarkable mean IQs. Caucasoid steppe nomads (the Scythians, etc.) disappeared before IQ tests were developed, so we can’t say anything about them besides the fact hat they didn’t leave behind any complex cultural artifacts. It would surprise me if Caucasoid Arctic nomads (the Saami) scored well. We know that the descendants of Mongoloid Arctic nomads score around 90.

    But all the descendants of northern and middle-latitude farmers score well, regardless of whether they’re Mongoloid or Caucasoid. It’s as if northern-farmer past trumped race on this issue. Southern-farmer Mongoloids (SE Asians, Mesoamericans) score in the upper 80s, lower 90s. Southern-farmer Caucasoids (Fertile Crescent, Indian subcontinent) score mostly in the 80s.

    Maybe there was something in that northern-farmer experience that added to groups’ IQ regardless of whether they were Caucasoid or Mongoloid.

    Counterpoints to the above idea:

    Upper caste Indians are really smart, and they’re probably not descended from any northern farmers at all. The Euro part of their heritage would have come from northern steppe nomads who have since disappeared.

    The Ashkenazim, my own group, are said to be roughly 60% Middle Eastern (southern farmer) and 40% European. I don’t think that the question of where exactly in Europe that 40% comes from has been settled.

    To what extent are southern Han descended from the natives of the Yellow River basin? The Chinese language, both spoken and written comes from there. The Chinese state comes from there. I’m sure that a large part of even southern Han genetics comes from there, but how large exactly is that part?

    To summarize, all descendants of northern and middle-latitude farmers are smart, regardless of whether they’re Mongoloid or Caucasoid. None of the descendants of northern nomads are smart, regardless of whether they’re Mongoloid or Caucasoid. Few descendants of southern farmers are smart. Most southern farmers, regardless of whether they’re Mongoloid or Caucasoid, are not smart.

    When Europeans first came to north America, agriculture was in the process of spreading though it from the south. It was still new and patchy there. If the discovery didn’t happen and Amerindians were left to develop agriculture further on their own in north America for a few millenia, another China could have been born. Or not.

    • anonyme

      Atzmon et al. in 2010 gave a range between 30% and 60% European admixture for Ashkenazim, Iberian-descended Sephardim and Syrian Jews. Even within one of those groups there could be variation as there is some evidence for population genetic substructure. In the comments at this link from Razib at Discover GNXP, an estimate of 43-48% is given: .

      • Car Guy

        I think it’s because of this European admixture that Ashkenazi Jews have their renowned IQs. I suspected that the European elements triggered some sort of freakish mutation(s) that greatly benefited their cognitive abilities. Centuries of possible selective pressures just solidified these abilities in the population.

    • charly

      North Chinese look exactly like Southern Chinese so it doesn’t fuck your theory up that Shanghai is the brainy part of China. Also people speaking the language of the capitol is unheard of.

  • Jim

    East Asians aren’t only Chinese, Japanese and Koreans. Mongolians and Manchurians should be included. Tibetans also. That makes six people. It’s obvious that the latter three have ignored. There’s no serious data on them except ‘IQ and Global Inequality’ which puts Mongolian IQ at 101. The Mongolians are an interesting people who deserve more study in terms of IQ selection. They are descendants of fearsome steppe nomads and have long been stereotyped as dumb, yet they ruled both Russia and China and would have easily overrun Western Europe and Japan were it not for natural disasters. One would think they never thought much, yet their shabby tents housed philosophical Buddhist works of a complexity quite advanced. It’s amazing how much educational progress Mongolia made under the Russians in the 20th century, their literacy went from 5% to 98% in just a few decades. It’s true that Mongolia is very poor because of the post-Soviet economic collapse, but I’m sure the people’s IQ has remained resilient. If they only let go of their drinking habit learned from the Russians I’m sure their society will improve very rapidly, because they’re a special breed among the East Asians.

    • AK

      Hard drinking leads to lots of homicides and a low life expectancy, but there is little indication that it has any substantial effect on development. In Russia, for instance, it is the proles who drink the hardest, while the top 25% are sober; the latter also coincide with the cognitive fraction that is most relevant to development.

      BTW, the US was called “The Alcoholic Republic” in the early to mid-19th century. Didn’t do them much harm.

      I am sure that the Mongols have a high IQ for the reasons you stated, although it can still rise some thanks to Flynn. They also perform very well in martial arts for population of their size, probably because of some combination of centuries of selection for such traits as well as a diet that it is basically GOMAD/caveman. They also have the world’s fastest growing economy right now.

      • Jim

        I guess the Mongolians are moving away from Russia and integrating more into their ancestral East Asia. I once went to Ulan Bator and I couldn’t tell if I was in South Korea or Mongolia. So many Korean restaurants and young people dressed Korean style. Their diets are becoming more balanced and First World. I agree with you that we can expect to see some major Flynn in Mongolia, especially with the rapid urbanization and economic development there.

        I think the Cyrillic alphabet really suits their language well. Ultra-nationalist calls to revert to the old script would only decrease literacy.

        As for martial arts I agree. Mongolians pretty much rule Japanese sumo. They already got two yokozunas I think.

        • charly

          South Korea has a very large Mongol community and South Korea is the cultural heart of East Asia so it is not surprising to find Korean influences in Mongolia. Reverse is also true.

      • charly

        Alcohol misuse, leads definitely to problems with cognitive factors. And claims that only the bottom 75% engage in them are not based on reality
        Martial arts are popular with the underclass. It is not a good sign to be good in martial arts.
        Their economy is growing so fat because of exploitation of mineral reserves. Which means the average Mongolian will be fucked and should do the wise thing and move to Korea.

      • Richard Sharpe

        Have you thought about differences in alcohol metabolism in different populations and the effects of the widespread introduction of novel drugs/intoxicants? (The effects seem similar to, although less severe than the introduction of new pathogens.) Certainly, it took the Chinese a couple of centuries to recover from the widespread introduction of opium.

        However, on alcohol, I will note: 何以解憂?唯有杜康。–曹操

    • “There’s no serious data on them except ‘IQ and Global Inequality’ which puts Mongolian IQ at 101.”

      I didn’t know that. I have Lynn’s “Race Differences in Intelligence” here, published in 2006. On p. 240 he says:

      “There is a further anomaly in the intelligence of the peoples of Northeast Asia concerning the IQs of the Mongols of Mongolia and the closely related Samoyeds of Northern Siberia. There are no studies of the intelligence of these peoples but their low level of cultural development and technology suggests that it is not so high as that of the East Asians of China, Japan and Korea.”

      I don’t own a copy of “IQ and Global Inequality.” If you own one, can you quote the place that cites Mongolian IQ at 101?

  • Richard Sharpe

    However, Chinese verbal intelligence if anything lags the indigenous peoples of most developed European nations.

    This seems surprising to me.

    Personally, I know chinese people who are verbally skilled and I have observed that my relatives include those who appreciate word-play and verbal skill.

    In addition, they have a long literary tradition and a veneration of those who combine literary skills with martial skills.

    Could it be that IQ tests do not adequately measure the Verbal Intelligence of the Chinese?

    • Jing

      The problem with Chinese verbal intelligence is that it is measured by a written language test of Chinese ability. Written Chinese is the most difficult and non-intuitive commonly used language in the world.

      That Chinese polities have managed to reach over 90% literacy levels is in and of itself amazing. Other people’s burdened with a cumbersome Chinese language would struggle to reach half that. Can you imagine how many African Americans could actually develop the functional ability of being able to read and write 3000 or so Chinese characters with any degree of fluency?

      • Car Guy

        But don’t the Chinese in the West exhibit the same IQ pattern as their mainland counterparts?

        • Richard Sharpe

          It is not clear to me that we have data that breaks out recent immigrants from those who have been here for one or more generations.

          I think we could expect the verbal scores of recent immigrants to be not as good as their math scores.

        • charly

          You could look at adoption kids, but they always score way above average

      • SP

        “Can you imagine how many African Americans could actually develop the functional ability of being able to read and write 3000 or so Chinese characters with any degree of fluency?”

        I would imagaine, if with life-time dedication and under the gunpoint, 13? 12 perhap? None of them is named Obama, surely.

        AK: This is a moderation note. Please avoid overt racism here.

      • Richard Sharpe

        “The problem with Chinese verbal intelligence is that it is measured by a written language test of Chinese ability. Written Chinese is the most difficult and non-intuitive commonly used language in the world.”

        1. English is far from intuitive on a number of measures. It is is mostly of Germanic origin with a leavening of French/Latin words seasoned with words from other languages. It has a confusing spelling system, and it differs, grammatically from other European languages and from personal experience I know that Chinese speakers have enormous problems with prepositions and tenses (a Chinese person I know once said “my father has already died for seven days” which was clearly literally translated from 我爸爸死了七天。) Of course, I continue to have problems with when to use 了 and other things.

        2. We can expect that the students in China who were tested are functional in written Chinese when tested.

        Is it perhaps that the concepts used in the verbal portion of the IQ test do not map well onto the Chinese world-view and thus cause confusion in the minds of the test takers?

      • Richard Sharpe

        The only way to resolve this would be to compare English Language Verbal IQ tests with Chinese Language Verbal IQ tests. Of course, I could probably not adequately deal with the Chinese test, but I am sure they could be compared.

        Does anyone have links to any that are claimed to be equivalent?

        • HX (Huax)

          Chinese students utterly destroy natives on tests proxies for verbal skill from the SAT Critical reading to British GSCEs to whatever Australia has, an in California exceed Japanese/Korean/Indian students on scholastic tests despite being sourced from a perpetual underclass in Chinese society.

          My theory as I’ve posted before is that social organization raises IQs, but Tibetans and Mongols are no slouches if you account for nutrition.

          What needs to be noted is that East Asian and Northeast Asian IQ structures are very different. The gap between NEA and EA is something akin to gaps found within Caucasoids i.e Iranians/Arabs/Northern-Southern Europeans.

          Japanese people have significantly faster reaction times than “South Chinese” who beat them on overall intelligence (esp. verbal IQs) and “complex” reaction times – having said that, “South Chinese” is problematic because once again the intellectual and IQ profiles of Cantonese, Fujianese and Wu speakers are all wildly different from each other.

          I would break high-IQ East Asians into roughly three distinct macro groups: North Chinese and other Sino-Tibetans, Northeast Asians and Altaic Speakers, and South Chinese (which would be further broken down)

          I’d put it as something like this:

          North Chinese/Sino-Tibetans – High overall IQ (108-110ish), intermediate reaction times
          Northeast Asians – Insanely high VS, good verbal, very basic reaction times
          South Chinese – Taken as a whole, moderate IQ (105-108ish), but with greater extremes (see: Shanghai, Taiwan/Singapore’s TIMSS 95%ile)

          The mechanisms underlying this are fairly intuitive I’d say. A more in-depth explanation is outside of the scope of this one post because I don’t have time to elaborate.

          Oh and hi Mr. Karlin

          • Richard Sharpe

            “Chinese students utterly destroy natives on tests proxies for verbal skill from the SAT Critical reading to British GSCEs to whatever Australia has, an in California exceed Japanese/Korean/Indian students on scholastic tests despite being sourced from a perpetual underclass in Chinese society.”

            Do you have some data or are you only blowing smoke?

            Are you referring to FOBs or even those who have been in the US, England or Australia for a few years, or are you referring to Chinese who were born in those countries?

            • HX (Huax)

              Typically those born there, usually descendants of Guangdong laborers.

              Google California STAR, GSCE A-levels Chinese pupils, etc

          • HX (Huax)

            Very high basic reaction times*

            In short North Chinese benefit from a very harsh selection path and have demonstrated their “survival skills” as a race by enduring not only the Eurasian ice age but a Tibetan one as well, and then 10,000 or so years of complex society thronged by ultra-aggressive and successful nomad confederations. The cycle of civilizational Darwinism is much advanced in Northern China, and as the following 100 years will prove the North Chinese culture and ethno-state will prove to be exceptional in the broad historical sense. As a result of what I’d say is a more optimal cognitive mix, North China has traditionally attracted elites of the elites like Persian/Tocharian/Korean/Xianbei etc royalty which has probably marginally benefited them.

            The South Chinese have seen millennia of complex society as well, as well as migration from the North. The quality of South Chinese elites speaks volumes about the quality of N. Chinese genetic and cultural input – it’s usually Northern or mixed phenotypes that dominate South China culturally and financially. But I say that without dismissing the cognitive qualities of S.Chinese aborigines, which are unexplored (by me, at least). Southern China has also seen significant migration from merchant classes among other nations which seems to continue to this day.

            Northeast Asians are more easy to understand, vs and fast reaction times represent a cognitive endowment that is easily and intuitively adaptable for warfare and hunting. AFAIK it has translated well to economic growth in Korea and Japan. One caveat however is that Koreans and Japanese are not strictly Northeast Asian, which is why Mongolia is a much more interesting case-study in my opinion.

            North/East Asians aside, on India I’d say there is nothing contradictory about exceptional performance South Indian high-caste performance. Southern India has been civilized and organized for thousands of years, and they have a lot of genetic material to work with.

        • Car Guy

          Great, Haux is here. As if one Han supremacist wasn’t enough…

          • HX (Huax)

            I’ve always been here, but I posted under a different name (HX). Stop pulling the race card.

  • SP


    The reason you gave seems likely.

    One thing particularly strikes my attention is poems. Grew up in Europe and went to some Chinese language schools I was forced (also by my patents) to recite many famous Tang and Song Poems. The problem is that they mostly can’t be translated into main European languages including English. Many words are just non-translatable(no such words!) unless one prepares to loss much of the meanings or completely screw up the rhythms. When they do get translated, however, most of them hilariously sound like being written by a 12-year-old “Lady Gaga”. That`s perhaps why the Chinese writers will never get Western-based Nobel Literature Prize by merit IMO, those politically motivated or originally written in European languages aside.

    More shockingly, the reverse by and large also holds true! Many famous poems say in German or English even some pieces of Nobel Prize category sound also like the work of a 12-year-old in Chinese after translated, to the extend that sometimes I wonder if these had been presented to Imperial China Poetry Exam, they would have been disqualified straight away at county level for being vulgar and got these Nobel Category poets´ ar*** slammed by the emperor himself.

    Thing is, unlike the logic of 1+1=2 that everyone can pretty much agree upon, sometimes the West (Europeans) and the East(East Asia) think/reason/speak quite differently than most people thus perhaps some `culture-neutral` verbal IQ tests realise.

  • Richard Sharpe

    If you believe that the mechanisms operating here are elimination of genes for low IQ from the gene pool and the greater reproductive success of those with (certain types) of high IQ due to the civil service exam, then assuming a model where hundreds of genes are involved in IQ, many with two or more alleles, it would seem likely then that it is eliminating alleles for low IQ from the pool. In addition, it would seem that even if they are re-introduced via random mutation or introgression, they will be eliminated again quickly. Further, it seems that IQ reducing mutations are much more likely than IQ increasing mutations, although IQ increasing introgression would likey be preserved.

    Perhaps I don’t understand, but this would seem to imply that variance would be reduced, compared, say, with Caucasian populations, since the upper limit does not move as much as the lower limit.

    Are there any credible numbers on the variance of the Chinese populations tested?

  • SP

    What most commenters from the West build their varies hypotheses upon are the following implicit assumptions that:

    1, All Han Chinese are about the same (oke, even sort of north-south divide, but only about 2 or 3 pts). – just like all Europeans have about the same IQ and a universal variance, right?

    2. Chinese, Japanese and Koreans are about the same. – just like continental Germans, British islanders and nord Finns are all about the same stock, yes?

    3. Japanese IQ, and increasingly S Korean IQ, represent post-Flynn ceiling for all Orientals, which is of course the natural ceiling for all Han Chinese after China is fully developed.

    4. Point (3) is further “confirmed” by varies test results of HK, USA, UK, Taiwan and Singapore.

    All above 4 are patently false, unfortunately.

    [ need to mention here: HK=Cantonese Han Chinese, USA West coast and UK = old timers are Cantonese mainly, Taiwan = Fujianese Han + aboriginals + a bit of everything, Singapore = descendents of mainly Fujianese+Cantonese landless peasants who went out of China striving for better livinghood throughout ages + minority but fastly increasing lower IQ Malays & Indians due to breeding + some recent top IQ economic migrants from China/HK.]

    Guess Jing has to find out what are the Gaokao stats of the last decade on diff han regions. Through all the conversations I’ve had with Chinese students the consensus seems to be that Fujianese and Cantonese are very good, but definitely not the toppers in Gaokao.

    Only from Gaokao stats, particularly heavy g-loaded Gaokao Science, one could gauge where are all sorts of Han Chinese, alongwith the Japanese and Koreans, stand in the rank, in combination with PISA and TIMSS that we’ve already known.

    First be clear on where they each are in the real world and get a grip on basic underlying assumptions, then build models/hypotheses to explain it, not the other way around increasingly a la “Multi-factor Random Bingo Modelling Rec1man Style” ( lol), agree?

    • HX (Huax)

      Link to 2007 Gao Kao scores. Shandong and Henan are on top but I don’t know how representative those scores are. At first glance it sounds about right to me, though, people from that region are pretty smart.

      For sciences Shandong, Shanxi, Henan, Hebei, Heilongjiang, Jilin are all northern Chinese provinces that outscore Zhejiang – more or less all descendants of the same group of people.

      • AK

        Thanks, Huax. I’ll have a new post on this.

      • maopai

        The first table in your link shows a ranking of the minimum score required for admission, not the average score of each region.

        • HX (Huax)


          • Sindilation

            To be honest, this table is completely irrelevant, because different provinces even have different full marks…….Shanghai is at the bottom…

  • Jim


    “I don’t own a copy of “IQ and Global Inequality.” If you own one, can you quote the place that cites Mongolian IQ at 101?”

    “I have Lynn’s “Race Differences in Intelligence” here, published in 2006. On p. 240 he says:
    There is a further anomaly in the intelligence of the peoples of Northeast Asia concerning the IQs of the Mongols of Mongolia and the closely related Samoyeds of Northern Siberia. There are no studies of the intelligence of these peoples but their low level of cultural development and technology suggests that it is not so high as that of the East Asians of China, Japan and Korea.”

    It seems a bit strange that the Mongols can be compared to the reindeer-herding Samoyeds of the Arctic. I’m not saying they are stupid (in fact the Samoyeds may be quite intelligent) but the Mongols should clearly be on another level. First the Samoyeds are only 45,000 and have been reindeer-herding in the far-northern Arctic for the past 5000 years. Mongols are 10 million and starting from the Donghu 3500 years ago have been constantly engaged in complex herding of sheep, goats, horses, cattle, camels and yaks (around 50 million animals at any time). The Monguor are a Mongol-speaking people numbering 250,000 who practice only agriculture and follow Confucianism mixed with Daoism and Buddhism. The Mongol lands have always (for the past 3500 years) been connected to China by an intimate umbilical cord of direct political, economic and social relations which places Mongols firmly within East Asia. They have a history of statehood stretching back 3000 years. In many respects they outdo the Tibetans if you look at the larger picture.

    The Samoyeds have been living in one stagnant social system (reindeer-herding) for the past 5000 years. The Chinese have been living in one closed social system (strictly sedentary agricultural) for the past 6000 years. The Mongols on the other have been living in a much more diverse and challenging social system incorporating steppe nomadic empires, Chinese settled agriculture, South Siberian hunting lifestyle, Tibetan religious complexity and Manchurian pastoral-agricultural hybrid lifestyle. And to this one must add the recent intensive Russian cultural influence.

    So Mongols should be studied within the sphere of East Asian IQ studies. One should remember that the Koreans and Japanese also came from Mongolia and the Lake Baikal region.

    • “They have a history of statehood stretching back 3000 years.”

      I’m pretty sure that they didn’t have real statehood at the time of their conquests of the 13th century. A real state is bigger than any man or any family. A real state cannot be split among the children of a king. It is not any man’s or any family’s property. It has large bureaucratic institutions that tend to endure in a stable fashion through many changes of leadership. It can go on functioning without anyone at the top. The Roman and Byzantine Empires were real states. In the Dark Ages and the early Middle Ages Europe didn’t really have states. They re-emerged there in the late Middle Ages. Ivan III is sometimes credited with creating the first real Russian state in the late 15th century. The word Russia existed for many centuries before that, but there were no states separate from the property of princely families.

      China has had real states, sometimes one at a time, sometimes several at a time, for a very long time. I’m pretty sure that at the time of their peak notoriety (the 13th century) the Mongols didn’t have a real state. As I said before, a real state cannot be treated as a family’s inheritance, to be split among the children, and that’s how Genghis’s heirs treated their conquests.

      • OK, correction: Novgorod was a real state going back to the Middle Ages. It’s more correct to say that Ivan III created the state from which the modern Russian state descends.

      • AK

        To have a state as you define it, you need an agrarian society, and the technological conditions simply didn’t exist for the Mongols to create one in the South Siberian environment they inhabited at the time.

        But I certainly think the Mongols are a high IQ people like the East Asians. They created by far the most advanced nomadic steppe civilization in history. Their military tactics were unprecedented for their time; they exploited the military resources of civilization (e.g. Chinese siege engines), unlike most barbarians; military leadership was based on strongly meritocratic principles, which many countries hadn’t gotten round to implementing even by the 19th century.

        Come to think of it, given their very small population and the climatic features of their spawning grounds, their achievements were nothing short of amazing.

        • Among steppe nomads the entire adult male population fought on horseback. In Russia and the West only the aristocracy was trained to fight on horseback. Most of the population in both the West and China had to spend their time tilling land instead of daily improving their riding and raiding skills. Because of this, before firearms were perfected steppe nomads punched above their weight in war, were more successful in war per capita than settled peoples. The Mongols did more than their predecessors, but there had been precedents: the Huns, Hungarians, Polovtsy, Skythians, Turks, etc.

          I don’t think any of these peoples contributed much to civilization. Modern Turks, Hungarians, Iranians, etc. are mostly descended from the people whom steppe nomads conquered, not from the nomads themselves.

          Anecdotally Kazakhs have lagged behind Russians in Kazakhstan. I don’t know if this has been quantified anywhere in any way. And aren’t the Kyrgyz descended from nomads too? I don’t think independent Kyrgyzstan has been a success story. Almost every East Asian country has boomed in recent decades except for N. Korea (ideology) and Mongolia. It makes one think.

          • AK

            At this point I have to make the always pertinent reminder that not everything in development is reducible to HBD. 🙂

          • HX (Huax)

            Mongolia liberalized fairly late and they are in the process of stabilizing now – in Taiwan, SK and China’s formative economic years GDP growth was likewise all over the map. Their’s is growing at 15%+ now, and will probably stay high due to relatively large mineral wealth per capita.

            Bhutan is another (mostly) EA nation though it’s nominally 25% South Asian (illegals)

            • Jim

              “I’m pretty sure that they didn’t have real statehood at the time of their conquests of the 13th century. A real state is bigger than any man or any family. A real state cannot be split among the children of a king. It is not any man’s or any family’s property. It has large bureaucratic institutions that tend to endure in a stable fashion through many changes of leadership. It can go on functioning without anyone at the top.”

              They did have a state, in a sense, because they were part of or closely connected to a bigger truly state-like entity, the Jin Dynasty (1115-1234). The Borjigin Mongols of Khan were the Menggu Shiwei and they had long been part of the civilized Liao Dynasty (907-1125) and later had intimate relations with the equally civilized Jin Dynasty. Khan was actually given the title “chautquri” (battle ruler) by the Jin Emperor Wanyan Madage. So the Mongols always had an imperial mindset.

              One day a Mongol could be alone hunting deer near Lake Baikal or chasing wild asses in the remotest part of the Gobi Desert (like a hunter-gatherer, developmentally a stage below pastoralism), a few weeks later he could be seated in the presence of the Chinese emperor (in many cases an ethnic Mongol himself as was the case with the Khitans and Tuoba) in Beijing or Chang’an (where half-Mongol Tang emperors ruled) discussing military affairs, trade relations and matters of imperial administration. This pattern can be seen all through history.

              For example (from Wikipedia) ‘the Mongolic-speaking Xianbei originally formed a part of the Donghu confederation, but existed even before that time, as evidenced by a mention in the Guoyu (“晉語八” section) which states that during the reign of King Cheng of Zhou (reigned 1042-1021 BC) the Xianbei came to participate at a meeting of Zhou subject-lords at Qiyang (岐阳) (now Qishan County) but were only allowed to perform the fire ceremony under the supervision of Chu (楚), since they were not vassals by covenant (诸侯).’

              I may not know the fine details but I know that there was a succession of rather advanced empires on the Mongolian steppes starting with the Donghu (c. 1200B.C.-209B.C), Xiongnu (209B.C.-93A.D) and continuing with the Xianbei (93-234), Murong (235-670), Rouran (330-555), Turk (552-744), Uyghur (745-840), Liao (907-1125) and all these before the Mongols of Khan. These states were big and complex enough and had close enough relations with China so I would call them states. They all had a physical territory, a subject population, formalized foreign relations, an organized state bureaucracy, so I think most of the basic criteria are met.

              I’m not sure if the same 3000 year old tradition of statehood applies in Central Asia, in the Stans. It’s far from China. It’s true though that Iran exerted a lot of influence culturally and genetically from the Achaemenids onward. There (in Central Asia) IQ strangely decreases as we move from Mongoloid to Southern Caucasoid: Kazakhstan (94), Kyrgyzstan (90), other three Stans (87), Iran (84).

              • Casian

                According to DODECAD, Kyrgyz have a higher Mongoloid/East Asian component than Kazakhs (70-80% vs 60-70%).

              • Casian

                I think Central Asian underdevelopment (Kazakh/Kirgiz) has alot to do with socio-cultural/geographic vs HBD/genetic factors.

            • charly

              Having large mineral wealth and becoming successful are to different thing. I would even argue that there is a negative correlation. There is a reason why Norway scores so low.

    • Arachanski

      I think Korean/Japanese/Han Chinese had some IQ increasing mutation after they came to Manchuria. If we look at Y-haplogroup history and distribution, that must have happened around 7-8 thousand years ago. That would explain the current IQ distribution in the region completely (Japanese have slightly lower IQ since they have 1/3 Ainu admixture). I also don’t think south Chinese have this mutation (may be as Han admixture but not originally).

      • ObaMahdi

        Hans are not the original aborigines of south china. I suspect the aborigines were australoids similar to philipinos or khmers and south china was quite sparsely populated those days. As the chinese empire expanded or when it was under threats of futher north nomads, hans migrated south. Most of the australoid aborigines moved to places such as present day cambodia, some stay behind and became minority tribes and are mostly assimulated. Even N vietnamese are 70% han by ancestry. One good example is 6th century chinese poems written by northern poets rhyme better in the guangzhou(a southern) dialect

  • Kiwiguy

    Peter Frost commented on Ron’s unpublished paper. In respect to the differences across Japan, Korea & China, he notes:

    “Was this factor strong enough to raise the mean level of intelligence? One objection is that the Chinese civil service exam was only partially adopted by Korea and Japan. Yet mean IQ is similar in all three societies.

    This objection ignores the broader emphasis on education in all East Asian societies. China, Korea, and Japan have long been “exam cultures,” even if we exclude the civil service exam. This exam grew out of values that were embedded in Confucianism and present throughout East Asia:

    Clearly, the higher mean IQs of East Asians cannot be solely or even mainly attributed to the Confucian exam culture. The main cause was the establishment of a State society, its monopoly on the use of violence, and its creation of an orderly, rules-based society. Reproductive success depended on being able to play by the rules.

    The rules, however, were formalized in the teachings of Confucius. One’s knowledge of these teachings became a proxy for one’s ability to succeed in East Asian society. More generally, it became a proxy for intellectual performance, all the more so because one had to memorize Chinese characters (a minimum of 10,000 for functional fluency) and understand an archaic form of the language. Thus, Confucian exam culture might explain some of the differences between European and East Asian intellectual performance.

    But why did this exam culture develop in East Asia and not in Europe? Greco-Roman society similarly valued study of classical literature and proficiency in archaic Greek and Latin (as opposed to the contemporary Koine Greek and Vulgar Latin). With the advent of Christianity, however, classical “pagan” literature became viewed with suspicion. Emphasis shifted toward study of the Bible, and such study usually involved entry into celibate religious orders. Insofar as academic success was linked to heritable predispositions, the overall impact of natural selection would have been negative.”

    • charly

      Weird, but i always taught that Latin was still required until the early 19 century to function intellectually and that Europe has always loved lawyers. But who am i to claim that this sounds to much like finding proof to a predetermined conclusion.

      ps Isn’t the bible written in another language and big?

    • Seeki Yan

      SeekiYan said: According to the statistics of Chinese characters used in newspapers, 90% of the daily used Chinese characters are around 600 only. 97% of the daily used characters are up to 2000 characters. That means if you know about 600 characters you can read and thoroughly understand the newspaper written in Chinese already.
      My own experience in studying English: After studying English for more then ten years and sit for the SAT, in the analogy session, they give a pair of words and then required me to choose the answers from four other pairs of words in the questions. I don’t know any one of it out the ten vocabulary, not in only one question, but in most of the questions. The English speaking people always proud of their rich in vocabulary, but in fact this is a nightmare for a foreigner of different culture to learn their language. After forty years of using English, I still cannot grasp the simple (as native English speaking people say) grammar in English.

    • But Japanese did not take these exams. I believe Koreans did, but not Japanese. So why the high IQ there when they had not received significant immigration from Mainland China for 2,000 years and it was never as much as they got from Korea or their own aboriginal input.

  • Charles – Br0s

    in my opinion, mongols and others barbarians of Siberia are mixed with eskimos
    if you see, the culture of siberians barbarians and mongols are similares of apaches indigenous, there some eskimos tribes in north and northeast of Asia, are almost impossible theys do not meet

    I heard that the IQ of mongolia is a estimated based in mean of your neighbors(Rússia and China)

    • Your knowledge is worse than 5 year old child’s.

  • I shall make some nationalist comments here.

    Celtic Europe (The Atlantic coast from Morocco to Iceland) divided land equally between all the children. This included the English, a mixture of Germans and Celts, until 1066 when the Normans introduced primogeniture.

    The modern rise in standard of living in England began in the Eastern counties after the Black Death when age for men at marriage rose to 28 or so from 23. Women married around 22 years old rather than 20. (This is from memory – it was in the New Scientist and on some now ancient BBC history program). Basically, after a period of overpopulation when expectation of young males was low so women were indifferent, some young males had access to land and were more valuable for marriage than others. We can expect diligence, impulsivity and intelligence to give advantages in a struggle for land. In those days and for much longer, the labouring poor reproduced less successfully than the yeoman class (renters of a farm).

    But intelligence as the key to human achievement is hokum in more ways than I am going to list here. As the Cookie Test demonstrated, other factors are at least as important in terms of achievement. Managed impulsivity is one (do smokers have lower IQ’s – no, they are less able to control impulses). Motivation is another. Only a few highly intelligent, diligent and motivated people are required to make the big advances.

    Other thoughts.
    European Neolithic and Bronze ages – very little warfare in the archaeological record. So, not much coercion, selection depends on personal choices.
    European Iron Age to Black Death – armed warrior aristocracy makes the social decisions. So, aristocratic genes dispersed into population in preference to slave/serf class males who died unmarried.
    Basic English has a German vocabulary with very few Celtic words but in recent years it has been observed that almost all Grammatical variations from German involving verbs are derived from early Welsh. So English doesn’t just have a huge vocabulary. It has multiple sources of grammar. Very tough for someone who speaks a more logically consistent language.