« Go read Calcutta Libertarian (I'm old school, yo, keep it Anglicized :) | Gene Expression Front Page | Mr. Rogers passes »
February 27, 2003

China, odds & ends

Ikram Saeed asks:

I would agree that China is less diverse than India, but it is still extremely diverse. And south China is the most distinct region of all, with several unintelligible "chinese" languages.

Are Chinese emigrants disproportionately from the south? US and Canadian Chinese populations arise largely from southern China. The only country I know of which has Chinese residents originating from many parts of China is Taiwan.

But this is blog is best when it is quantitive. What is the genetic similarity between northern, southern, and inland residents of China? Compare with India?

On north Chinese vs. south Chinese, Ikram is of course correct that the "dialects" of the south are languages by another name (while the "languages" of Scandanavia are dialects by another name). But, for the lay person, Ikram forgets to mention that all Chinese (Han, Zhongguo-Ren, Huaren, whatever) share a common written language, ergo, their ruling elite share a literary culture & canon. Perhaps an analogy might be made with medieval Latin in Christendom, the difference being that in this case 75% of Christendom shares the same colloquial language to boot.

As for the genetics, this is a complex question (speaking of complex, some bloggers & readers might be interested in this overview of Hakka genetics). Joanna Mountain, working with Cavalli-Sforza, has done work that indicates that south Chinese share an earlier common ancestry with southeast Asians than north Chinese. Historically this is plausible, the Vietnamese ancestral homeland is located just south of the Pearl River Delta while the Thai people still retain a prescence in southern China as the "Dai." Just as Indo-Aryan languages like Marathi have been characterized as "Indo-European with a Dravidian accent," so one could depict the south Chinese dialects in a similar manner. Of course, I have recently come to be skeptical of connecting shared ancestry marked by neutral genes to the genetic heritage of phenotypically significant traits (in other words, south and north Chinese might share different mother & father lines, but their genes that influence phenotype might have been selected so that a common phenotype has developed-this is obviously important from the perspective of human biodiversity).

The idea that the migrants that formed the Chinese Diaspora were self-selected is persuasive on the surface (to me at least). The Taiwanese, Singaporeans & Hong Kong Chinese share about the same per capita income. But, there is a difference: while Singapore & Hong Kong have incomes of 24 K and 25 K, Taiwan is at 17 K (not an economist, so I don't know if "purchasing power parity" takes into account that the two cities almost certainly have sky high cost of living). Could this be because the inhabitants of the two cities are more self-selected than the "Taiwanese" [1]? I would like to know the per capita incomes of the Chinese of Malaysia, Thailand and Indonesia (though the last is a bit confusing, as the Chinese are bifurcated into the "Totok" and "Perakanan" communites, the latter being somewhat like the Baba Chinese of Malaya).

Update: From the message board:

Here's a paper that suggests that there is no difference between Northern and Southern Chinese

Posted by js at February 27, 2003 03:40 PM

[1] Hong Kong is a mostly Cantonese city with a prominent clique of Shanghai capitalists. Singapore is more mixed, but I believe (someone can correct me) that most are descendents of migrants from Fukien. The "Taiwanese" are ~90% Fujianese in origin, migrating over a long period of time from Manchu conquest down to the period of Japanese rule, the remaining 10% are mainlanders, predominantly from north China. The indigenous Formosan Taiwanese are irrelevant to this discussion, and numerically negligible (I'm not a linguistic specialist, apologize for the misappellation Malayo-Polynesian).

Posted by razib at 09:59 AM

The languages of S. China are mutually unintelligible, which makes them more distinct than the Scandinavian languages. The North which is more uniform was settle earlier than the South; the greater southern diversity (my conjecture) must be due either to the dissected geography or else to interference from original languages.

As I remember, the late KC Chang said that North Chinese are physically closer to Koreans and Japanese than to South Chinese.

Singapore and Hong Kong have been international trade cities under British law for over a century; Taiwan was a backwater until 1890, and then a Japanese colony until 1945. The Japanese provided rational administration, infrastructure, and education,but real development only began after 1960 (people said that you would still see oxcarts there in the early 60's.)

In 1984 my students told me that mainlanders were simple and sincere (a Chinese ideal, but not usually really striven for; this was not an expression of Communist sympathies), the Taiwanese more businesslike and selfish, the Hongkongers (?) still more, and the Singaporeans (?) were the most ruthless businessmen in the world. New Yorkers told me that HK matched New York in that respect, so maybe my students were right.

Posted by: zizka at February 27, 2003 12:34 PM

PS. Chinese recognize a process of "becoming Chinese" which means learning to act like a Chinese. Someone who learns to practice Chinese ways will be welcomed into Chinese society according to their education and wealth, which means that an educated, wealthy Westerner will be more accepted in Chinese society than an uneducated or a poor Chinese. (The real story was more complicated than this, but one woman said that "My father wants me to marry a Chinese, but a PhD or MD counts as Chinese").

So anyway, the South became Chinese at least as much by assimilation as by settlement. "Under the Ancestors' Shadow" by Hsu is a description of a Sinified bilingual people. Hsu was an anthropologist and had hoped to study a tribal people, but when he got there he found that they had become almost completely Sinified. So he basically did an anthropological study of Chinese culture. (In "Americans and Chinese" he did an anthropological study of American culture).

Posted by: zizka at February 27, 2003 12:46 PM

historically one could become chinese. the tang emperors (7th through 10th century) were of partial non-han (turkic) origins, but are not looked upon as a barbarian dynasty like the yuan (mongols) or to a lesser extent ching (manchus). also, there are people in south china of muslim origins that are now "han," they often have the surname "ma" (muhammed) and do not give their ancestors pork as offerings. now, could this happen today though? far less likely, the hui ("chinese muslims"-speakers of a han language and physically similar to the chinese around them) are now an ethnic group recognized by the chinese government with certain privileges & benefits. even if one is non-religious, one is still hui. additionally, in modern china, it is not uncommon for han women to marry hui and convert to islam, but the reverse does not happen (from what i've read).

the idea that one can become chinese is i think more problematic in the modern day practice. it is similar to the fact that beurs can not become "french" even though nearly half of the french are descendents of european immigrants themselves (italian & polish generally), while in germany the turks can not become german, though a sizable portion of the natives of berlin can almost certainly trace their ancestor to french hugenots. modern day ethnic definitions have become curiously static as literacy & national self-awareness has spread....

Posted by: razib at February 27, 2003 12:57 PM

Here's a paper that suggests that there is no difference between Northern and Southern Chinese

Posted by: js at February 27, 2003 03:40 PM

One has to ask - if genetics matters and Southern Chinese are different from Northern Chinese then why do the predominantly Southern Chinese diaspora exihibit the same IQ/performance patterns as Japanese and Koreans - even ignoring the self-selection aspect (i.e. take Taiwanese IQ)

Posted by: Jason Soon at February 27, 2003 05:02 PM


i've asked. the first thing i have to say is this-though human biodiversity can predict/intuit a lot from big differences between populations that are geographically and historically (ergo genetically) divergent, for example, zulus, welsh and japanese, it is less usefull and more tenditious for peoples that are more closely associated, for example, the thai, khmer and north chinese. second, as i've noted, there might be a difference in the genetic lineages using mtDNA & y chromosome markers and the phenotypes that might be selected for.

i know from doing some research in my undergraduate years trying to figure out the possible female irish contribution to icelandic genetics that answering historical questions with molecular biology when the groups in question are very close (irish and norwegians in this case) is difficult because the noise level in the data is so high. for instance, i'll be convinced about the presence or lack of irish mtDNA lineages after a dozen studies that point the same way unequivocally. on the other hand, the presence of european y chromosome genes in places like accra, ghana, is a little easier for me to swallow, because the historical process (european sailors forming liasons with local women) is easy to connect to the alien y chromosome lineages.

of course, the simplest answer right now to my mind is that the genetic differences are minimal enough that chinese culture is the prime determinant of economic performance. but remember that the initial question was sparked by comparisons between the chinese and indians, who are genetically different enough to note (and usually cluster on different branches of cladistic trees).

Posted by: razib at February 27, 2003 05:38 PM

Just to throw out a different variable: beyond racial and "cultural" differences, legal, political and institutional differences are of major importance. Chinese are "culturally" very similiar from the mainland to Taiwan to Hong Kong to Singapore and even to the US for the first generation or two, but there are significant differences too which I think can best be accounted for by different institutional environments.

Of course, if "culture" is regarded as "everything but nature and genes" what I said is meaningless. But if you distinguish culture from law/institutions (as you have to if you want to talk about seriously multicultural societies such as India or Singapore), then a different form of analysis becomes possible.

The legal fact that China has had a cash economy and alienable private property in land for well over a thousand years is a major consideration in accounting for their success in the diaspora. This is a "modern" institutional trait.

Posted by: zizka at February 27, 2003 06:17 PM

"The indigenous Malayo-Polynesian Taiwanese" are in fact technically not "Malayo-Polynesian". More precise terms would be "Austronesian" or, better yet, "Formosan". The Formosan languages constitute first-order branches (Tsouic, Atayalic, Paiwanic) of the Austronesian language family. MP is another fourth branch. To say that Formosans are MP is like saying that sisters are their brother's daughters. Although MP was used in the past to refer to Austronesian as a whole (since most AN languages are MP), this usage is now obsolete among specialist linguists.

Posted by: Amritas at February 27, 2003 09:05 PM

The higher GDP of Singapore and Hong Kong compared to Taiwan is because the first two were both former British colonies :)

Posted by: Dan at February 28, 2003 05:56 AM

Singapore and Hong Kong were british colonies, but also major crossroads of international trade and finance. So they had British law, Chinese culture, and the experience of being part of world trade. Being a British colony was less help to nations which were dedicated to providing raw materials.

Japan, of course, the biggest economic success story outside W. European culture and institutions, was never a colony. So Hong Kong and Singapore succeeded for different reasons than Japan, or partly so.

During their rise S, Korea and Taiwan were highly favored by the US and were highly protectionist. (Taiwan had high tariffs -- up to %200 -- and many major state monopolies -- grain, energy, alcohol, tobacco --when I was there in 1983-4). The WTO does not offer the third world these developmental options.

Posted by: zizka at February 28, 2003 02:05 PM

WTO does not offer these options? What the hell does that mean? I mean Pakistan was a strong US ally too. This 'WTO thing' is a stupid left wing excuse if you ask me.

Posted by: Subrato at February 28, 2003 02:45 PM

Lots of the Taiwanese economy was government-run for a long time after the war whilst that was never the case in HK and much less so in Singapore. And it takes a long time for an economy to recover from government meddling -- which is why Britain is still poorer than the USA and Taiwan is poorer than HK.

Posted by: John Ray at March 1, 2003 06:47 AM

On GDP Differences: Singapore and Hong Kong are city states -- cities have very high GDPs. Look at Hamburg's GDP, or New York, or Paris. Compare with the surrounding state / region's GDP. Taiwan is a real (though not real) country, with rural and urban areas. I would bet that Taipei's GDP compares well with HK or Sing.

On WTO: The WTO tarrif reductions and quaot eliminations have made it much more difficult to have a wall against exports, but that's not the only reason the Asian 'tiger's' had high GDP growth in the cold-war. Conventional Wisdom is that these countres engaged in a export dirven strategy, as opposed to the import-substitution strategies of South Asia (incl. Pakistan).

But I would not blame South Asian leaders too much. Post WWII, import-substitution was cutting-edge development strategy. It was the Asian-tiger that were out of sync with then prevailing trends. Pakistan's first and second five-year plans were done by an American advisor, and India mimicked Soviet plans.

Unrelated Anecdote: In the early nineties, a good way to tell is a foreign (or immigrant student was from HK, Taiwan or the Mainland would be to look at their haircuts. HK people had very stylish (for small-town Canada) haircuts, and whole all-black and expensive clothing. Mainlanders had atrocious five-dollar 'bowl' cuts. (Taiwanese fell in the middle, better hair cuts, not so cutting-edge clothing).

The mainland must be richer now, or hair-cut-technology has advanced, because it's harder to diffrentiate along those lines.

Posted by: Ikram Saeed at March 3, 2003 07:13 AM