Wednesday, October 19, 2011

Ideograms, alphabets and civilizations

As someone who never mastered them, I've always felt oppressed by Chinese characters. Perhaps my kids will succeed where I failed :-)

Bertrand Russell reflects on ideograms, alphabets and civilizations in The Problem of China. (Note, technically Chinese is neither wholly ideographic or logographic.)

As everyone knows, the Chinese do not have letters, as we do, but symbols for whole words. This has, of course, many inconveniences: it means that, in learning to write, there are an immense number of different signs to be learnt, not only 26 as with us; that there is no such thing as alphabetical order, so that dictionaries, files, catalogues, etc., are difficult to arrange and linotype is impossible; that foreign words, such as proper names and scientific terms, cannot be written down by sound, as in European languages, but have to be represented by some elaborate device.[15] For these reasons, there is a movement for phonetic writing among the more advanced Chinese reformers; and I think the success of this movement is essential if China is to take her place among the bustling hustling nations which consider that they have a monopoly of all excellence. Even if there were no other argument for the change, the difficulty of elementary education, where reading and writing take so long to learn, would be alone sufficient to decide any believer in democracy. For practical purposes, therefore, the movement for phonetic writing deserves support. [Somehow, modern Chinese have achieved nearly 100% literacy despite the difficulty of the writing system.]

There are, however, many considerations, less obvious to a European, which can be adduced in favour of the ideographic system, to which something of the solid stability of the Chinese civilization is probably traceable. To us, it seems obvious that a written word must represent a sound, whereas to the Chinese it represents an idea. We have adopted the Chinese system ourselves as regards numerals; "1922," for example, can be read in English, French, or any other language, with quite different sounds, but with the same meaning. Similarly what is written in Chinese characters can be read throughout China, in spite of the difference of dialects which are mutually unintelligible when spoken. Even a Japanese, without knowing a word of spoken Chinese, can read out Chinese script in Japanese, just as he could read a row of numerals written by an Englishman. And the Chinese can still read their classics, although the spoken language must have changed as much as French has changed from Latin.

The advantage of writing over speech is its greater permanence, which enables it to be a means of communication between different places and different times. But since the spoken language changes from place to place and from time to time, the characteristic advantage of writing is more fully attained by a script which does not aim at representing spoken sounds than by one which does. [Italics mine.]

Speaking historically, there is nothing peculiar in the Chinese method of writing, which represents a stage through which all writing probably passed. Writing everywhere seems to have begun as pictures, not as a symbolic representation of sounds. I understand that in Egyptian hieroglyphics the course of development from ideograms to phonetic writing can be studied. What is peculiar in China is the preservation of the ideographic system throughout thousands of years of advanced civilization—a preservation probably due, at least in part, to the fact that the spoken language is monosyllabic, uninflected and full of homonyms.

As to the way in which the Chinese system of writing has affected the mentality of those who employ it, I find some suggestive reflections in an article published in the Chinese Students' Monthly (Baltimore), for February 1922, by Mr. Chi Li, in an article on "Some Anthropological Problems of China." He says (p. 327):—

Language has been traditionally treated by European scientists as a collection of sounds instead of an expression of something inner and deeper than the vocal apparatus as it should be. The accumulative effect of language-symbols upon one's mental formulation is still an unexploited field. Dividing the world culture of the living races on this basis, one perceives a fundamental difference of its types between the alphabetical users and the hieroglyphic users, each of which has its own virtues and vices. Now, with all respects to alphabetical civilization, it must be frankly stated that it has a grave and inherent defect in its lack of solidity. The most civilized portion under the alphabetical culture is also inhabited by the most fickled people. The history of the Western land repeats the same story over and over again. Thus up and down with the Greeks; up and down with Rome; up and down with the Arabs. The ancient Semitic and Hametic peoples are essentially alphabetic users, and their civilizations show the same lack of solidity as the Greeks and the Romans. Certainly this phenomenon can be partially explained by the extra-fluidity of the alphabetical language which cannot be depended upon as a suitable organ to conserve any solid idea. Intellectual contents of these people may be likened to waterfalls and cataracts, rather than seas and oceans. No other people is richer in ideas than they; but no people would give up their valuable ideas as quickly as they do....

The Chinese language is by all means the counterpart of the alphabetic stock. It lacks most of the virtues that are found in the alphabetic language; but as an embodiment of simple and final truth, it is invulnerable to storm and stress. It has already protected the Chinese civilization for more than forty centuries. It is solid, square, and beautiful, exactly as the spirit of it represents. Whether it is the spirit that has produced this language or whether this language has in turn accentuated the spirit remains to be determined.

Without committing ourselves wholly to the theory here set forth, which is impregnated with Chinese patriotism, we must nevertheless admit that the Westerner is unaccustomed to the idea of "alphabetical civilization" as merely one kind, to which he happens to belong. I am not competent to judge as to the importance of the ideographic script in producing the distinctive characteristics of Chinese civilization, but I have no doubt that this importance is very great, and is more or less of the kind indicated in the above quotation.


41 comments:

RKU1 said...

There's actually another interesting aspect to this ideogram question, about which I've sometimes speculated...

If you discard your preconceived notions, and objectively examine the national IQ data collected by Lynn, a striking "anti-IQist" conclusion is easily apparent.  Restricting ourselves just to European populations, there are a very large number of cases in which genetically indistinguishable peoples exhibit dramatically different IQ scores, with the sample sizes being far too large for these discrepencies to be explained by random error.  Notable examples are East Germans/West Germans, Ireland Irish/American Irish, Italy Italians/American Italians, Israeli Ashk Jews/American Ashk Jews, and numerous other cases.  The gap is usually in the range of 10-15 points, which is obviously very substantial.

Sometimes these national discrepancies are temporal, with the data showing an apparent rise of 10-15 points within a single generation or so.  A similar example is the case of American-born Mex-Ams, whose GSS-implied IQ seems to have risen about 10 points in the last thirty-odd years.

The common factor in nearly all of these cases is that the low-scoring groups tend to be mostly rural or impoverished in various ways, while their high-scoring counterparts are urban and fully embedded in modern western civilization.  I think the most plausible conclusion is that scoring well on IQ tests tends to require a degree of "cultural priming", which is supplied by urban life but often absent for folk growing up in rural or backward conditions.  This is probably somehow related to the Flynn Effect.


Now here's the interesting part.  Lynn's data seems to show no such "impoverished IQ depression" among East Asian peoples, who seemed to score at 100 or above even when their countries were poor and backward compared to most of the low-scoring European ones.

One possibility is that the EA scores actually *were* artificially depressed, and affluence-adjusted Chinese and Japanese have astonishing mean IQs in the 115-120 range.  But this seems extremely implausible, and I've never seen any evidence for it.

A second possibility is that the genetic structure of EA intelligence is somewhat different from that of Europeans, and requires less "cultural priming" to achieve peak performance, being more robust in structure.  This is an interesting idea which someone should explore.

But a third possibility is that EA culture contains various near-universal elements which provide the necessary "priming" which rural or impoverished Europeans tend to lack.  And I think a plausible candidate for one such element might be the ideogramatic structure of writing, which even impoverished EAs tend to encounter in school.

Presumably, there might be some datasets which would help us to distinguish between these second and third possibilities.

LaurentMelchiorTellier said...

I'm aware of your argument, which I've also thought carefully about for a while. I've found somewhat of a confounding dataset, in the cohort of "South Koreans in Japan".

Although SK's and J's have similar IQs in their respective countries, where they enjoy high status, SKinJ's seem to score 10-15 points below their SK and J counterparts. This would appear to be very similar to (as you mention) Irish in Ireland vs. Irish in the US. Obviously, SKinJ's are generally proficient in kanji / Asian logograms, and are a very large and integrated minority in Japan (1% +). They're by no means an illiterate slave class.
http://en.wikipedia.org/wiki/Koreans_in_Japan

SKinJ's are nonetheless low on the Japanese totem pole - subject to stereotypes of crudeness, dirtiness, and often set to perform unskilled labor. This attitude stems from their importation as forced labor during the period of Japanese imperialism. This IQ gap is similar to Irish, etc.


*
There is a fourth hypothesis that being part of a low-status minority contributes to depressing IQ. 
http://www.huppi.com/kangaroo/L-inferiorIQ.htm

esmith said...

The problem with modern English is that it's, in essence, the worst of both worlds. It lacks the brevity and the preciseness of Chinese ideographic writing. Yet, at the same time, it lacks reliable letter-to-sound correlations present in a truly phonetic language. Which is probably the reason why it is quite unusual to find 5 year old children with adequate grasp of written English, whereas, in Russia, being able to read is often considered a prerequisite among 5 year olds applying to the elementary school.

Sam H said...

Properly, it should be called the Lynn-Fynn effect.

Max B said...

In Russia its common to be able to read at age 6 to 7 . NOT five. English is great language and relatively big discrepancy (if compared to German,French or Russian) between what is written and how it pronounced is the result of lack of language reforms. Russian written language went trough several reforms at the beginning of the century to make it more in line with the development of oral language (and also standardize it across whole country). In any case it is far from being worst. Among the advantages of English language is the ease of learning , rich vocabulary  and laconicity  .

English is much better lingua franca than Chinese could ever be.

Michael Bacon said...

Having lived abroad, in Japan and in Europe, I think the best aspect of English is its malleability.  Basically, there is no one standing at the door trying to keep out new words.  On the contrary, new words spring up all the time -- there's no way to stop them.  I recall a story that a friend told me.  He was a Chinese language teacher in the US.  He spent some time studying and teaching in Taiwan. Once during a class someone said that Chinese didn't have a "word" for a certain concept.  He said that was no problem -- just combine this symbol with that symbol and you had something close enough that worked.  He was politely told that what he proposed just couldn't be done in Chinese.  He asked why and the reply was, well it just can't.  In English (especially in the US) within a couple of months the word would be in some in a popular song and shortly after that in an on-line dictionary.

esmith said...

Even 19th century Russian was phonetic. Reforms mostly did away with silent letters and duplicate symbols.

English and French are victims of their own literacy. English spelling has been mostly crystallized since at least 16th century, even though pronunciation has been drifting - because there were too many literate people and too many printed books around. In Russia, both were all but nonexistent before the late 18th century.

It's not a big deal once you learn the language, but it makes it hard to learn to read. It's a handicap.

RKU1 said...

That's an interesting point about the depressed IQs of South Koreans living in Japan, which I hadn't known.

Otherwise, it's certainly true that most of the non-depressed IQ data from China and other East Asian countries was urban rather than rural.  But the same was true for much of the highly depressed IQ data from East Germany and may have been true for other Eastern European countries as well.  And at the dates in question, I think the per capita GDPs hugely favored the European samples, and were presumably matched by access to modern culture and technology such as television.

Yan Shen said...

[Somehow, modern Chinese have achieved nearly 100% literacy despite the difficulty of the writing system.]

Perhaps the high spatial ability of East Asians greatly assists in their mastering an ideographic writing system?

Yan Shen said...

"The common factor in nearly all of these cases is that the
low-scoring groups tend to be mostly rural or impoverished in various
ways, while their high-scoring counterparts are urban and fully embedded
in modern western civilization.  I think the most plausible conclusion
is that scoring well on IQ tests tends to require a degree of "cultural
priming", which is supplied by urban life but often absent for folk
growing up in rural or backward conditions."

Or maybe people with higher IQs tend to end up in wealthier urban settings and people with lower IQs tend to end up in poorer rural settings. Do people have low IQs because they're poor, or are they poor because they have low IQs?

"Restricting ourselves just to European populations, there are a very
large number of cases in which genetically indistinguishable peoples
exhibit dramatically different IQ scores"

Even within the same country, there are can rather significant genetic differences amongst people in various geographic regions, right? I've read that the IQ of whites in Minnesota is supposedly at least 5 points above the white American mean.

"A second possibility is that the genetic structure of EA intelligence is
somewhat different from that of Europeans, and requires less "cultural
priming" to achieve peak performance, being more robust in structure."

That would mean that Europeans would be better at gaming a test by grinding it out through extensive prep. Maybe you should tell that Steve Sailer fellow about this theory of yours...

RKU1 said...

"Or maybe people with higher IQs tend to end up in wealthier urban settings and people with lower IQs tend to end up in poorer rural settings."

I think that seems very implausible given the raw data.  The groups involved seem genetically indistinguisable, and although one might hypothesize that e.g. the Irish who immigrated to America were smarter than the ones who stayed behind, a genetic gap of around 15 points between the two groups seems implausible.  Furthermore, when we consider the impact of mean-reversion over several generations, the original immigrant-gap must have been closer to 30 points, which seems utterly absurd.  Exactly the same argument applies to all the other same-ethnic IQ comparisons.

Also, consider the temporal comparisons.  Is is really plausible that the "genetic" IQ of Ireland Irish rose over 10 points in just a single generation?  Or that the innate IQ of American-born Mex-Ams followed roughly the same trajectory over a similar period?

The advantage of restricting ourselves to these same-ethnic comparisons is that we mostly remove genetic factors.  Hence, any large variations which remain must be "cultural" or "environmental" in some sense.  And since none of the European countries involved were suffering the sort of serious malnourishment endemic in much of the Third World, nutritional-type factors probably aren't significant.

Yan Shen said...

"The advantage of restricting ourselves to these same-ethnic comparisons is that we mostly remove genetic factors."

Even within a given ethnic group, there's a huge spread in the distribution of IQ/genes. Perhaps poor white Americans score low on IQ tests relative to wealthy white Americans for the same reason why poor Australian Aborigines score low on IQ tests relative to wealthy Europeans.

esmith said...

Where exactly do you get the 15 point gap between Irish and Irish-Americans? And in what direction is it supposed to run? Lynn gives the Irish IQ of 92. In the United States, people of Irish ancestry are among the poorest and presumably least intelligent of all ancestries of European origin, so the value of 95 or so could be reasonable to expect. And there's no such thing as "pure" Irish-Americans - they've been living in this country for over 100 years, so many self-identified Irish-Americans have contributions from other ancestries. That would result in drift towards the mean.

It is harder to explain the Italian vs Italian-American and Spanish vs Spanish-American pattern. In both cases Lynn gives IQ's of 102 and 98 for "mother countries", but their respective immigrants are even poorer on average than Irish-Americans, and their mother countries are among the poorest developed countries in the world. I'd expect that either Lynn is wrong or there is some kind of environmental factor in testing procedures that artificially increases scores of Italy and Spain.

Many other cases you describe can probably be explained by selective migration (villages to cities, poor countries to wealthier countries).

RKU1 said...

Actually, the data I've seen is that Irish-Americans are among the most economically successful white ethnic groups, so I doubt their IQs are below the white American average.  Don't forget that the Irish are heavily concentrated in the cities of the Northeast, which have among the highest per capita income, and are least likely to be found in the South or in rural areas, which generally have the lowest.  Where in the world did you see data suggesting the Irish are among the poorest white Americans?  Maybe in 1875, but that was well over a century ago.

Regarding Lynn's IQ data on the Ireland Irish, I suspect you've allowed him to pull the wool over your eyes.  If you actually look inside his Wealth of Nations book, you'll discover that his Ireland data consists of two studies.  One of these is based on a sample-size of 3500, among the largest he mentions, and shows an IQ of 87.  The other is based on a sample-size of 75, among the smallest he mentions, and finds an IQ of 98.  The figure of 92 is based on a straight (unweighted) average of those two studies, which is totally absurd, and the correct figure to use is obviously the IQ=87, which puts them (roughly) 15 points below Irish-Americans.  I've always wondered if Lynn performed that very strange "averaging" because as a WASP professor in Ulster he was concerned that if he proclaimed the Irish people as mostly feeble-minded, he feared the IRA might shoot him.

esmith said...

For people of Irish ancestry, U.S. Census data (2010 ACS) shows median household income of $56,363. For Swedish, it's $59,224, for Polish, $60,927, for Scottish, $61,439, for Austrian, $65,967. Not a big difference, but statistically significant.

RKU1 said...

Well, according to Wikipedia, median white household income in 2005 was 49K, so the Irish are almost 20% above the white average.  Austrian-Americans are utterly trivial in number, I suspect a good fraction of the "Poles" are Jewish, and both the Swedish and Scottish ethnicities are fairly small, being not much over one-tenth the size of the Irish.  I've never used the ACS data-tool, but my suspicion would be that the Irish income figure is higher than that of the other "large" self-identified white American ethnic groups, such as the German, English, Italian, and "plain American."

Anyway, I never claimed that the Irish were the *most* economically successful white ethnic group---I'd strongly suspect it's the Jews---just that they were *among* the most successful, which is probably correct.

Zhengzheng Zhou said...

2500 years ago written Chinese probably was very easy to learn, as the majority of characters contain a pronunciation clue that's accurate as well a semantic part that also meaningful. Over the centuries both the pronunciation and the meaning have changed greatly. I think without reforms English will be in similar shape in 2000 years.

esmith said...

In terms of educational attainment, the Irish are also behind Germans (33.2%), Italians (34.6%), Norwegians (37.1%), English (37.3%), and Russians (57.1%). The only major white ancestries less educated than the Irish are the French and the Dutch. In these two cases, it's not immediately obvious to me why that would be so, but that's what the numbers say. People of French ancestry are concentrated in Louisiana and they could have been nontrivial intermixing with Native Americans and African Americans.

Matthew Carnegie said...

It could be. But we haven't ever tried educating any other population who natively spoke Chinese in hanzi so we don't know - it might just be less tough than we'd guess. I believe China has some ethnic Russians so maybe we someone could try see how they'd do?

Also I'd guess literacy may be being defined in China as "Can read some level of texts in hanzi that the government decides is literacy (probably not without some justification), even if this is less than their speech and vocal comprehension capacity", rather than the more Western or global typical definition of "Anything you can speak you can read and write". And the former may be somewhat simpler.

...

I think that the persistence of logograms long term in China is due to multiple factors

- The monosyllabic morphemic structure, with relatively few syllables and many homophonous morphemes compared to a Western language, with a very convenient setup for using the rebus principle (which is extensive in written Chinese, although it may play less of a pneumonic role compared to one in which it is simply a generative process for characters learned by rote). This also works better when the writing system is native rather than introduced.
- There are too many syllables in Chinese for a simple syllabary (compare to Japanese, the language which is probably one of the most natural fits for a syllabary amongst languages spoken by many people).
- The alphabetic principle is probably much less obvious for them than it was for the inventors of the first alphabets - the Egyptians - and the other Semitic peoples, whose languages decompose meaning into triconsonantal roots and vowel patterns which are then combined through synthesis to make words. This isn't a particularly clever or useful as a linguistic device, but it would seem to me to make an alphabetic system relatively obvious.

I think once we sort all this out, we might find a case, still, for visuo-spatial skill assisting in the persistance of hanzi in China, but I think the above are still somewhat more persuasive to me.

LaurentMelchiorTellier said...

You two are getting severely sidetracked. 

The question is not about how Irish-Americans compare to other white Americans. The question is about how American Irish compare to Irish living in Ireland.

LaurentMelchiorTellier said...

A second dataset would also appear to confound your theory that hieroglyphs stimulate rural IQ: North Korea.

NK uses no glyph system, theirs is a phonetic alphabet (Hangul). And I think it's fair to characterize most of their population as being peasantry, and intellectually less stimulated than fx. a Japanese cohort. Yet their scores appear identical to SK, and for that matter to Japan. Come to think of it, I'm afraid that every nation in the region with a phonetic alphabet (Vietnam, Thailand and Tibet) seems to be in disagreement with your idea, if you include them as EA's or quasi-EA's - certainly, Korea must be.

RKU1 said...

I think one problem in considering Chinese urban/rural IQ differences is that it's quite plausible there may have been highly selective migration to urban centers over the last century, so it's possible much of the difference might actually be genetic.  Until the last decade or two, I think the PRC strictly enforced its resident-permit system.

But if North Korea solely uses a phonetic alphabet and the (very impoverished) North Korean peasantry has IQ scores pretty similar to the vastly more affluent SK and Japanese populations, that would seem to be be very strong evidence against my hieroglyph-impact speculation.  Same for the other heavily-rural EA populations if their IQs are not substantially depressed despite low socio-economic levels and use of phonetic alphabets.

On the other hand, all of this might tend to strengthen my far more intriguing hypothesis that innate EA intelligence requires less "cultural priming" than the European variety, since so many of the rural/low socio-economic European populations seem to suffer from such IQ depression, and show rapid rises in tested IQ once their countries urbanize and modernize.

A speculative argument in favor of this hypothesis is that for the last millennium or two, the Chinese and perhaps other EA populations tended to live on the Malthusian edge and also with peasant intelligence being a crucial survival trait.  Therefore, there probably would have been strong selective pressure for manifested intelligence to become more "robust" and maintain itself despite severely deprived conditions.  Meanwhile, European peasantry tended both to be less limited by starvation and also less dependent for their success upon intelligence, hence would have been less likely to evolve such mental "robustness."

LaurentMelchiorTellier said...

Indeed, the H2 is at least not readily counter-indicated.

Not to be bulldogging you or anything ;-), but I'm not sure where you've arrived at the conclusion that EA IQ has been stabile over the course of their economic development? The only EA IQ trends I've read (sorry to constantly be going on about China, it's most of what I read) show that Chinese IQ has been steadily increasing during its development, similar to the "usual" pattern of emerging nations. That increase has been continuing in recent years - big parts of China are still undergoing a momentous rise from poverty, and they've been experiencing a big IQ surge, bringing them much closer to Japan/SK/etc.

RKU1 said...

Well, consider the (remarkably high) Chinese IQ results presented by Lynn from the mid-1980s, and compare them with the (very depressed) IQ results he gathered from roughly the same era in Ireland, East Germany, and other parts of Eastern Europe.  I'd think that the per capita GDP of the former would have been vastly lower at that point, even if adjusted for PPP.  Similarly, I'd think most other aspects of urban modernity---widespread contact with TV, automobiles, industrial equipment, electronic equipment---which might contribute to expressed IQ would also have been much lower in China at that point.  I've never investigated any of this, but that would certainly be my impression.

And just consider your own datapoint about the high IQs of (desperately impoverished/often malnourished) North Korean peasants, who have no access to TV or most other modernities.  They must be suffering massive poverty/cultural-deprivation compared to the Irish or East Germans of the 1970s, yet seemingly score about 15 points or more higher in IQ.  On the other hand, if they were well-fed and fully-stimulated, I really doubt their mean IQ would then jump to 120 or so.

I think the notion of varying degree of IQ "robustness" against malnutrition or cultural deprivation seems a pretty plausible hypothesis, based both on empirical evidence and theoretical arguments.

LaurentMelchiorTellier said...

That would be a good case - could you please direct me to the IQ results from the 80s? I can only find Lynns modern results from Wealth'tions.
http://en.wikipedia.org/wiki/IQ_and_the_Wealth_of_Nations

Yes, NK's IQ is slowly rising, despite no access to Starcraft they're at almost 100% parity with SK. I've never heard of a historical period where they seriously diverged, even in periods of famine.

esmith said...

Like I said, seeing a 15 point gap depends crucially on believing that Irish-Americans are the smartest of all Americans. If they are below average, then a 15 point gap turns into a 5 point gap.

esmith said...

If peasant intelligence were a crucial survival trait, we would see uniformly high intelligence in every culture with long history of agriculture and Malthusian pressures. Including many sub-Saharan African and Middle Eastern countries.

Instead, we see high intelligence correlated with recent history of literacy and urbanization. It does not matter much if your remote ancestors were subsisting on agriculture for 500 years or for 5,000. What matters is for how long being able to read was a survival trait.

Korea and coastal China were urbanized early on and even 50 years of mismanagement by North Korean authorities can't reverse multiple centuries of selection.

esmith said...

I'd take all data coming out of China with a grain of salt. There is a risk that the researcher fudged the numbers directly, and there is an even greater risk that numbers are misleading because of the way test subjects were selected. This reminds me of a recent story when China came out on top in a high school academic achievement study, and somehow most news sources failed to notice that the study was comparing elite students from Shanghai against randomly selected students from all walks of life in other countries.

And even that study could get numbers from rural Chinese high school students, those numbers would STILL be misleading because of double digit middle school dropout rates in rural China. You see the problem?

There is a lot of geographical/class variance in China and numbers we're discussing can easily be dominated by geographical factors. If you take an underlying population with IQ of 100, smartest 20% move to cities and the remaining 80% remain peasants, you will find an average rural IQ of 95 and an average urban IQ of 121 - a huge effect that will easily swamp any influences of Flynn effect, innate intelligence, role of the writing system, etc. etc.

RKU1 said...

Actually, most peasant societies---Africa being the most extreme case---have never been close to their Malthusian limits, and population was kept in check by all sorts of other factors, which provide the main selective pressure.  I think traditional China was quite unusual in that regard, and the extremely sophisticated nature of its rural economy also put a strong premium on intelligence.

(other comment) Furthermore, if you examine the studies in Lynn's table, it seems very difficult to locate an East Asian case---across quite a number of years and countries---in which the scores are much below 100, and an enormous number of cases where they are well above.  Meanwhile, across the far wealthier European countries, a good fraction of all the studies show results considerably below 100.  It seems implausible that fraud would be so skewed in that way.

As a partial confirmation of the very impressive Chinese IQ numbers, I recall that American IQ researchers (Terman?)were a bit surprised  a century ago to discover that the children of totally impoverished Chinese immigrants had  IQ scores at or perhaps even a bit above the white average.  

LaurentMelchiorTellier said...

I agree with esmith that Chinese data perhaps are exceptionally prone to "nationalistic" data-corruption, and any aggrandizing oddities should be treated with skepticism. 

But the Chinese data do not stand alone: the Japanese, Korean, Taiwanese, Singaporean, Hong Kongese, etc datasets do show very remarkable stability / no-Flynn.

The "affuent" European countries (Austria, Australia, West Germany, Netherlands, etc) show the same pattern of stability / no-Flynn, whereas "rising from poverty" SE Euros (Greece, East Germany, etc) are rising. I suppose this is sort of analogous to how NE Asia contrasts with South East Asia, etc.

*
I'm honestly sort of discombobulated about this, this flies against what I thought I knew about Flynn. And the Wealth'tions link I showed you (plus the other data I've seen) have shown a typical, slow "rise" pattern in both EA and NWE. 

esmith said...

It's not necessarily even nationalistic data corruption. It's just that it's very hard to do sociology properly in a society like the one in China today.

For example, if you do a sociological study that measures a nationwide IQ in the United States, it's well understood how you need to sample the population to get meaningful data. The population is relatively homogeneous and everything is relatively static. You know where to go to sample the rural population, and, even if you don't, it's not a big deal because the rural population is all but nonexistent (total employment in agriculture is something like 0.5% of adult population). This last part is true for most developed countries.

The story is totally different in China. The population is stratified on many levels, there are constant flows in all directions, and the obvious places where you might want to start (large cities) are actually outliers, because, unlike the developed countries, China and India have significant numbers of people employed in agriculture (China: 38%, India: 52%, as per CIA World Factbook.)

Points that we have in places like Taiwan and Hong Kong are more meaningful, and useful to some extent, but we shouldn't jump to evolutionary conclusions, because populations of these places were probably undergoing selection due to flows at the city:country boundaries for a long time.

LaurentMelchiorTellier said...

At the time of the 15 point gap mentioned, the difference was 87 and 102, for NI and IA, respectively.

My understanding is that IA are not the "smartest of all Americans", just similar to British A, German A, etc.

esmith said...

I have to disagree with this vehemently. Malthusian limits are in place in most preindustrial societies close to 100% of the time. The primary factor keeping population in check may not always be hunger, it may be wars, epidemics, or infanticide, but it is nearly universal for such societies to stay above 6 births/woman, and possibly close to the physiological limit of ~10 births/woman. The environment can't accommodate the expansion at that rate, so there's always something that constrains the population.

Before 1800, there were two notable cases (that I know of) where Malthusian constraints were temporarily relaxed for multimillion human populations. #1 is the Black Death: the sharp collapse in population of Europe that took around 100 years to compensate. #2 is the European repopulation of Americas: immigrants to the United States would reproduce at the natural rate, but there was a lot of empty land where all the excess children would end up moving to.

I sincerely believe that coastal and islander EA societies have mean IQs above 100. But those are only a small part of total population. It's not so much fraud as attention bias.

I haven't seen any studies of IQs of Chinese immigrants to the US. There are two distinct groups of Chinese immigrants. Immigrants coming since 1970's have been primarily originating from self-selected regions such as large coastal cities (Shanghai, Beijing) and islands (Hong Kong, Taiwan). They tend to rank higher than average white Americans on all intelligence & socioeconomic scales. The second group is the descendants of people who came here in the 19th century. These remain minorities in this country in every sense of the word.

steve hsu said...

In many years of travel to/in China, and conversations with Chinese people worldwide, I have never met one who believed in a rural/urban or interior/coastal gap in (genetic) potential for intelligence. When I bring up such a possibility people just find it highly implausible. (Note these people I am talking about are generally quite open to HBD in general.) A common response I get about this is that anyone familiar with the history of China realizes that there has been an enormous amount of internal migration and population mixing over the centuries (and they don't believe in the systematic selection effects you mention, with urbanites being drawn from the smarter rural population; at least, they don't think it's a big effect). Recall that during the cultural revolution many urban intellectuals were sent to the countryside to work alongside peasants. These people, even after close observation, do not tend to believe in any systematic difference in potential for intelligence.

esmith said...

That's interesting to know. I may be wrong. But denying something does not necessarily make it false (after all, quite a few, if no most residents of Texas deny global warming :) ). There are significant differences in educational attainment by province. It's only reasonable to expect that there are differences in genetic intelligence.

esmith said...

Let's look at neighboring countries.

To the south, Lynn gives 78 for Nepal. I'd expect a similar value in Tibet and for Tibetans in other provinces.

To the west, numbers are hard to find, and Lynn has to engage in
creative averaging. I did find one very old study that estimated the IQ of
Kyrgyzstan in the 70's and described illiterate Uzbek peasants as "incapable of abstract thought". Kyrgyzstan came in dead last on every scale in 2009 PISA. Kazakhstan came in 7th from the end.

To the north, numbers are even harder to find. Mongolia is a third world country. Lynn tries to explain it away by blaming Communist influence, but it was hardly a developed country even before Communists got there (at the time when Mongolians overthrew the Qing rule in 1911, residents wore mostly nomadic, the largest city had the population of 25,000, and the literacy rate was 5%.) IQ studies of Mongolian population are almost nonexistent. There is a recent article that claims to have measured the mean IQ of rural Mongolian children and it was 120, but I have to write that off as incredible. Mongolia does not participate in any international scholastic achievement studies.

Kyrgyz people, Uzbek people, Kazakh people, Mongolians and Uighurs all have the same origins and similar cultures (the biggest difference is that Mongolians never converted into Islam). The last three also share the dominant Y-DNA haplogroup. I wouldn't expect to see large differences in intelligence.

In essence, all credible datapoints for non-Han ethnic groups in the interior of China indicate mean IQ's below 90, if not below 80.

Coupled with above-100 scores in Korea and Taiwan, I have to conclude that either there's a gradient that runs from the interior to the coast of China, or there is a cultural factor that kept Han Chinese thoroughly intermixed with each other but not intermixed at all with other ethnic groups.

LaurentMelchiorTellier said...

Good post, e. Cladistically, East Asians are closer to Europeans than they are to Thai and other SE Asians, whose offshoots make up the largest minorities in China; we tend to forget how distant the clades within China are. There's a *lot* of variation within China, I think of China as being about as diverse as the entirety of Europe. Of course, most of the treatment of IQ differences I can find in English are all about Tibet (by most accounts, 10-15 point gap to Han), but that signal has a lot of environmental contamination. My sister actually did some work in Tibet with iodine and iron deficiency, which is severe.

I can understand why many Chinese feel the way Steve describes.. the media here does not encourage feelings of division. 

steve hsu said...

I don't know enough about this to have a strong opinion (high confidence level), but I thought it worth reporting what I learned from asking around.

SekeMutema said...

While the derivation and writing of ennglish is alphabetic, I would say that the reading of english is ideographic once you achieve fluent literacy. On a personal level when I read I see the word not the letters, therefore the word itself has become a super ideogram. The often mentioned 50% redudancy in english spelling is proof of this.

steve hudson said...

Since the comments are on ideograms and IQ, perhaps  I have something of interest for this audience.  I have developed a system of ideograms for English that I call neoideograms, or "ids".  Disgusted with the overcomplexity, antiquatedness, and nonsensicality of many of the kanji (chinese characters) used in Japanese, I decided to give a try developing my own system that is simple, contemporary, and efficient. On a base of 20-30 elements from Chinese characters that I liked (why reinvent the wheel?) and a like number of elements from Egyptian hieroglyphics, some symbolic logic, some computer related symbols, etc., but mostly images from everyday life, I've built a system a workable system that I've transcribed books into.  It is highly educational in that it illustrates Latin and Greek roots -- it is a great way to learn English much, much better. I have what I call "etymids" -- etymologically based ideograms, and "directids" that illustrate the ideas more directly. I've applied them to German with much success, so they have some potential as an ideographic Esperanto. I have a dictionary of them that covers 5k words or so, but since Latin and Greek roots are illustrated and over 90% of English words of 2 syllables or more are from Latin and Greek, many more words are easily covered with this system. I'll be coming out with books on them. I expect them to be of interest to homeschoolers as they are so engaging, different, and worthwhile. They are highly creative, each word and idea a puzzle as to how to represent it graphically, with no correct answer. Graphic images engage more of the brain, so they are definitely IQ boosters. Anyone interested can contact me at neoideographer@gmail.com.  I'm especially interested in anyone in the Seattle area who is interested and might want to help edit and develop. 

steve hudson said...

Since the comments are on ideograms and IQ, perhaps  I have something of interest for this audience.  I have developed a system of ideograms for English that I call neoideograms, or "ids".  Disgusted with the overcomplexity, antiquatedness, and nonsensicality of many of the kanji (chinese characters) used in Japanese, I decided to give a try developing my own system that is simple, contemporary, and efficient. On a base of 20-30 elements from Chinese characters that I liked (why reinvent the wheel?) and a like number of elements from Egyptian hieroglyphics, some symbolic logic, some computer related symbols, etc., but mostly images from everyday life, I've built a system a workable system that I've transcribed books into.  It is highly educational in that it illustrates Latin and Greek roots -- it is a great way to learn English much, much better. I have what I call "etymids" -- etymologically based ideograms, and "directids" that illustrate the ideas more directly. I've applied them to German with much success, so they have some potential as an ideographic Esperanto. I have a dictionary of them that covers 5k words or so, but since Latin and Greek roots are illustrated and over 90% of English words of 2 syllables or more are from Latin and Greek, many more words are easily covered with this system. I'll be coming out with books on them. I expect them to be of interest to homeschoolers as they are so engaging, different, and worthwhile. They are highly creative, each word and idea a puzzle as to how to represent it graphically, with no correct answer. Graphic images engage more of the brain, so they are definitely IQ boosters. Anyone interested can contact me at neoideographer@gmail.com.  I'm especially interested in anyone in the Seattle area who is interested and might want to help edit and develop. 

Blog Archive

Labels