- The American Conservative - http://www.theamericanconservative.com -

China: Debating the “Clark-Unz Model”

As an individual who often regrets his decades-old defection from the academic community, I was remarkably pleased to see anthropologist Peter Frost very generously discuss my recent China article under the rubric “the Clark-Unz Model.” [1]  The senior researcher identified is obviously economist Gregory Clark, whose influential 2007 book A Farewell to Alms [2] had suggested a very similar evolutionary analysis for the forces shaping the British people over most of the last thousand years.

The nearly 100 total comments on that column and Frost’s previous one [3] have most sharply focused on what certainly seems to me to be by far the weakest aspect of my theory, namely that it would predict a substantial performance gap between Chinese and Japanese, given that the traditional rural society of the latter was totally different in nature (although Frost himself argues that there may have been more similarities than I acknowledge).  Obviously, if those two major East Asian peoples are very similar in their abilities, my analysis is probably wrong.

Certainly the conventional wisdom has always placed Chinese and Japanese in the same ability category, and if someone had raised that issue with me a year ago, I would have been very skeptical of any large difference.  But while I was performing the research for my Meritocracy article [4] I encountered some striking data.

California contains almost one-third of America’s total Asian population, and its Chinese outnumber its Japanese by about 3.5 to 1.  But among the high-ability NMS semifinalist students in recent years, there have roughly 750 Chinese names each year as opposed to a mere 15 or so Japanese ones.  Obviously, much of this difference may be explained by factors of cultural assimilation, differences in the age-distribution curves, and the impact of selective recent Chinese immigration.  However, a 50-to-1 difference in the number of top academic students is large enough to catch one’s eye and make one wonder whether there might possibly also exist the sort of intrinsic factors produced by many centuries of disparate selective pressure.  I’d also noticed that although a truly remarkable fraction of all the winners of America’s various national academic competitions had been Chinese, the number of Japanese names was so small that I never even bothered to separately record them.

Obviously, my China hypothesis was meant to be preliminary and provocative rather than definitive, and further study of the relative performance of various East Asian peoples should eventually help to resolve this intriguing question.

Interestingly enough, Clark’s own original version of the model may be subject to much the same critique as my own.  He had argued that many centuries of substantial gains in British future time-orientation and ability had been a large factor in explaining why the Industrial Revolution began in their country, rather than anywhere else in Europe or Asia. Among other things, this would imply that the British were substantially superior to all other European peoples in their intrinsic ability and social efficiency.  But just a few generations after Britain’s economic rise, the peoples of Germany, France, and Northern Italy were exhibiting equal or even better industrial performance, raising doubts about this hypothesis, as does the very middling range of subsequent British academic performance on the PISA and other international tests.

Otherwise, another science blogger responded [5] to my discussion of whether the term “eugenics” may be properly applied to a non-intentional, natural process, and the Chinese language version of my article was posted on several Chinese-language discussion forums, where it drew quite a number of comments, generally either supportive or at least respectful.

 

On a different matter, the published work of my sharpest critic, cancer researcher and ardent feminist Prof. Janet Mertz, was discussed in a The New York Times article [6] on educational policy.  Apparently, Stuyvesant, Bronx Science, and other elite public schools in NYC utilize an objective entrance exam, heavy on math and science, to select their students, and the overwhelming number of students admitted are male, which has attracted controversy.

Prof. Janet Hyde, one of Mertz’s co-authors, denounced these results on the grounds that the Hyde/Mertz peer-reviewed findings had conclusively proven [7] that males and females had equal math ability, and if boys scored much higher on the New York tests, then the tests must be biased and should be replaced, perhaps by the sort of “holistic” application process that was worked so well in selecting students for the Ivy League universities.

Comments Disabled (Open | Close)

Comments Disabled To "China: Debating the “Clark-Unz Model”"

#1 Comment By Gaeranee On March 30, 2013 @ 2:26 am

Mr. Unz, you should check the NMS for the Chinese vs. Japanese for the 1970s. Everyone knows that the Japanese have become wealthy since then so they’ve become underachievers – much like what you’re claiming about the recent collapse in the Jewish achievement. Also, Japanese immigration since then has decreased; thus, weeding out the ambitious immigrants who score well on tests. A better analysis would be to compare Chinese vs. Korean population in the 1990s. Although that may be difficult given that surnames Lee, Chung, Chang, etc overlap.

As for your PISA score analysis, Shanghai doesn’t count. It woud be the equivalent of testing kids from the Silicon Valley and not the entire country that encompasses rural areas. Same goes for Hong Kong and Singapore (city states). Rather, Korea has consistently scored at the top among OECD. But more importantly, according to one study, if you control for poverty levels, the US would be #1.

#2 Comment By kumar On March 30, 2013 @ 6:17 am

“Rather, Korea has consistently scored at the top among OECD.”

Korea itself is tiny. Singapore = 6 million people, HK = 7.5 million people.

Korea = 49 million people, half of them concentrated in Seoul_Incheon Urban Area.

Best is to compare Korea with Guangdong or Fujian, not 1.36 Billion Chinese.

#3 Comment By David J. White On March 30, 2013 @ 4:08 pm

He had argued that many centuries of substantial gains in British future time-orientation and ability had been a large factor in explaining why the Industrial Revolution began in their country, rather than anywhere else in Europe or Asia. Among other things, this would imply that the British were substantially superior to all other European peoples in their intrinsic ability and social efficiency.

I admit that I have not checked into this myself, but one hypothesis I have encountered to explain, in part, why the Industrial Revolution began in Britain rather than in Germany or other countries in Europe, at least, is that in Britain, coal deposits and iron ore deposits are located relatively close to one another, which facilitates making coal-fired furnaces to smelt iron ore. In Germany and elsewhere, coal deposits and iron-ore deposits are evidently further away from one another, which creates transportation problems. As I said, I haven’t checked into this myself, but it’s an interesting point, if true.

#4 Comment By spite On March 31, 2013 @ 7:54 am

My half cents worth regarding the industrial revolution was the relative peace that England enjoyed compared to Germany (the region since it was not yet a nation state), Germany was a battleground for wars from the 1700s to 1815.

#5 Comment By TomB On March 31, 2013 @ 8:42 am

Well in the first place I’m mightily confused. When I glanced at Amazon’s description of —A Farewell To Alms—it says that author Clark’s thesis wasn’t that the Brits were genetically pre-adapted to Industrial Revolution success at all, but instead *culturally* so.

I.e., exactly what I argued in comments responding to Unz’s past threads addressing his provocative idea that genetics/natural selection might be responsible for the success he sees modern Chinese having now.

Anyway while I think I’ve exhausted my objection to Unz’s proposition based on the idea that 1000 or 2000 years is just simply too short a period of time to expect to see natural selection work the genetic changes Unz posits there’s yet another problem that I think afflicts his thought here.

Or indeed *most* thinking that purports to accurately perceive some alleged cognitive/psychological/emotional/behavioral attribute to some certain environmental-natural selection pressure.

To me that is there’s a post-hoc ergo proper-hoc nature about this kind of thing that can seem a bit too pat and that makes me nervous.

And even before this some nervousness would seem greatly in order. After all what is done other than first, some alleged *different* cognitive/psychological/emotional/behavioral trait is identified in some subpopulation, and right there of course might lie a weakness: Because you are talking about a subpopulation by definition almost you are talking about people who now inhabit a different environment from others. (Like the Chinese in modern China.) But then you go and compare them to others and think you perceive where they differ—in industry, ability or etc.—but how do you know that if the general population wasn’t suddenly put into modern China they wouldn’t respond in *exactly* the same way the Chinese are responding?

I know of course that the idea is … comparing modern Chinese to others living in other environments that *seem* to compare: Those not enjoying much previous development and *seem* to have the same present opportunity to do what the Chinese do but don’t. But little environmental differences may make big behavioral differences, no? Let’s just take the subtle and purely intellectual (i.e., probably not genetic) issue of trust: Maybe the difference between the Chinese and others is that the Chinese simply trust their government now when it says they are free to enjoy the fruits of their labor, and others, sizing up their situation, lack that trust. After all, all it takes is that simple, subtle bit of thinking that says you should not so trust and what results? The idea that there’s no way you should you bust your ass to get ahead because there’s no real way you are going to be allowed to keep it.

And then you get to the real post hoc ergo propter hoc aspect because let’s assume you *are* focusing on some attribute (“X”) in some subpopulation that truly doesn’t exist or exists in some markedly lesser degree in all others or even in some others. In other words, your first conclusion is at least valid.

You then go looking into the background of your subpopulation and of course only naturally start looking through its history for its environmental factors that would, to you, seem to have favored X genetically through natural selection.

But, it seems to me, the dice are thereby loaded because I’d bet that for every plausible historical environmental factor your subpopulation has endured that you identify as contributing to X I would bet you could look at the historical environmental factors *every* population has endured that just as plausibly, if not moreso, would seem to put pressure on developing X.

Or to put it another way you are loading the dice by assuming a sort of direct, identifiable relationship between some environmental factor and the production of X, and my sense is that nature doesn’t work in such a linear, blatant way.

Who knows, that is, even with horrendously studied I.Q. where it seems to be known that significant differences clearly *do* exist between sub-groups, what environmental factors favor or less favor same? It’s just horrifically subtle, it seems to me.

And one of the major phenomenon that seems to accompany genetic change is the coincidental benefit. Nature/environment puts pressure on a population to change genetically in some certain, discrete way: E.g., whatever pressure it was that put mankind on its feet with bipedalism.

My goodness however, just think of *all* the additional benefits bipedalism bestowed on us no doubt far and beyond just giving us the advantage initially selected for. No matter *what* you think that selected advantage was for, there’s bound to be *tons* of additional, coincidental advantages it conferred. Advantages that one could *never* guess the cause of, because in a sense they were “uncaused.” Wonderful happy accidents.

And yet what do the Unzians seem to assume? That you can look back from any characteristic/advantage and with at least reasonable reliability find some nice linear cause of same.

To me then at least the way that nature works is such that it can even leave you without any assumptions upon which to even *start* from. For instance, what’s to say that bipedalism *itself* isn’t just a coincident attribute?

If then we can’t even know this, or the cause of bipedalism if that indeed was the prime characteristic selected for, it makes me suspect that we can ever get beyond the mere speculative when it comes to far more subtle things. Even when you indubitably find some distinct, genetic X characteristic, how the hell can you ever ferret out its true “cause” if indeed it can even be said to have one in the manner we are speaking of?

Not to say all this isn’t worth pondering though and deeply interesting and enjoyable though, even if it turns out to be what science writer John Horgan calls “ironic” science which is science (of a sort) that just simply can never be proven or disproven.

Indeed, that may be its very appeal.

#6 Comment By namae nanka On April 1, 2013 @ 8:33 am

a facile explanation?
[8]

reg Mertz and Kane, since reading affects maths, but maths doesn’t affect reading, no surprise:

[9]

“if boys scored much higher on the New York tests, then the tests must be biased and should be replaced,”

Some liberal arts schools got the gender imbalance too after using the standardized tests. I wonder how high the SAT-V difference would be if boys would read more and play less video games. Perhaps putting the latter in the curriculum would make them as boring as school lit. is today.

As for PISA, consider the Finland difference on TIMSS:

solmu.math.helsinki.fi/2005/erik/PisaEng.html

They score near east asians on PISA, yet got trounced on TIMSS by almost 100 points.
And what’s with the fascination with 8th grade maths; from what little I remember of my schooling, 9th grade was the start of real maths instead of the tiddlywinks before. And when the some of the erstwhile mediocre males started outperforming the best girls.
Amusingly, the man behind PISA himself was a slacker in school with bad grades before his father got him in a special school. His elementary school teacher didn’t think he was good enough to be enrolled in a gymnasium.

#7 Comment By namae nanka On April 1, 2013 @ 10:29 am

“school teacher didn’t think he was good enough to be enrolled in a gymnasium”

perhaps it should be “in Gymnasium”

As for La Griffe’s gender gap article, it used data from Project Talent and assumed gaussian distribution. But it looks skewed here:

[10]

and perhaps underestimated the gender ratio at the right tail.

#8 Comment By Dahlia On April 12, 2013 @ 3:41 pm

This is rather late, but I was wondering…

Since I tend to believe that there is a strong relationship between population density and intelligence, wouldn’t the story of *how* peoples dealt with constraints (or the “Malthusian Trap” as others like to say) be more a story of how the shaping and evolving of their *personalites*?
Especially since only small differences exist between China and the smartest European countries, not to mention between China and the other Asian countries?

#9 Comment By Dahlia On April 12, 2013 @ 4:04 pm

crowding, not population density, is more apt. Crowding comes into play with the Cochran-Harpending theory of Jewish intelligence while pop. density does not.

#10 Comment By Nathan Merrill On April 12, 2013 @ 11:39 pm

One major problem with your analysis is that it assumes that Chinese and Japanese last names are equally likely. I know a fair number of people of Chinese descent in the US with Chinese last names, but I know no one of Japanese descent in the US with a Japanese last name.

This will throw off any such analysis done by names if your assumptions about the two groups are not identical, or even if they’re not if the groups are sufficiently small. Given the extremely small sample sizes presented here this is something of a problem – especially given that per scores in their own countries this is simply not accurate.