« The future of blogging | Main | FDA Delay »
IQ and the Wealth of Nations
How many more times will someone suggest this book in the comments section of this blog? I like this book and I think it offers a real contribution. Nonetheless I feel no need to suggest it in the comments sections of other peoples' blogs.
I do not treat this book as foundational because of personal experience. I've spent much time in one rural Mexican village, San Agustin Oapan, and spent much time chatting with the people there. They are extremely smart, have an excellent sense of humor, and are never boring. And that's in their second language, Spanish.
I'm also sure they if you gave them an IQ test, they would do miserably. In fact I can't think of any written test -- no matter how simple -- they could pass. They simply don't have experience with that kind of exercise.
When it comes to understanding the properties of different corn varieties, catching fish in the river, mending torn amate paper, sketching a landscape from memory, or gossiping about the neighbors, they are awesome.
Some of us like to think that intelligence is mostly one-dimensional, but at best this is true only within well-defined peer groups of broadly similar people. If you gave Juan Camilo a test on predicting rainfall he would crush me like a bug.
OK, maybe I hang out with a select group within the village. But still, there you have it. Terrible IQ scores (if they could even take the test), real smarts.
So why should I think this book is the key to understanding economic underdevelopment?
Addendum: I am sorry there have been too many nasty comments, so I have taken the comments down. They aren't deleted forever, I like to think that I will have time to pick out the bad ones and put the thread back up. I do understand that most of you (and not just on one side of the debate) are capable of discussing this topic with the appropriate tone.
Posted by Tyler Cowen on July 18, 2007 at 05:51 AM in Books | Permalink
Comments
Some people do what works. How these individuals stumbled upon their successful rules, I, for one, do not know. One's Modus Operandi might be more advanced, but carelessly suggesting IQ might be the reason for it surely disproves that thesis.
Posted by: RRE at Jul 18, 2007 6:01:49 AM
If high IQs are in part an artifact of development, which is what I think you are suggesting, (familiarity with written tests, etc.) then isn't there a reverse causation thing going on here? Or perhaps a feedback/multiplier process.
Although the book, which I haven't read, ought to deal with that rather obvious point.
If the authors are suggesting that raw intelligence - 'real smarts' as opposed to measured IQ - explains development, then, contrary to your observation that people in poor countries have plenty of real smarts, they must be arguing that people in poor countries are in fact less smart. Which, if nothing else, is stunningly politically incorrect (and, my instinct would be, plain incorrect).
If they are arguing that developing the kind of intelligence that also leads to high IQ scores explains a development, that's a significantly different argument - one that would also make your point about real smarts in Mexican villages besides the point.
Which is it?
If they do argue that people in poor countries are less smart (as distinct from lower IQ measures), how do they explain it? Nutrition? Hopefully not genes.
Nutrition has the potential to generate a feedback process, so development leads to better fed smarter people, leads to more development etc.
Posted by: Luis Enrique at Jul 18, 2007 6:52:46 AM
I've seen some people who treat addiction complain about the official diagnosis criteria. They say that classifying addictions in the same way as diseases tempts people to take what is a description of behavior (something like "has attempted to quit, interferes negatively with social life and work, development of tolerance") and use it as an explanation of behavior ("He's that way because he's addicted").
Doesn't IQ tests suffer from the same problem? We know bloody little about what causes what correlations there are, but people go around acting as if IQ is an answer to something.
There is a thread over at crooked timber on it, if anyone really feels like engaging the people who would like to prove their intelligence with a number rather than with achievements...
Posted by: Harald Korneliussen at Jul 18, 2007 6:54:44 AM
Tyler states that he likes the book and thinks it makes a contribution. He never actually says that he read it. I certainly have not read it. But assuming that the book marshals evidence and makes testable statements, Tyler seems to be dismissing these with his personal, n=1, experience in Mexico.
Posted by: mc at Jul 18, 2007 7:08:55 AM
I would be interested in reading it. I'm pretty sure that I'm intelligent enough to read it, but I can't afford to buy it.
Perhaps it's at the public library . . .
Posted by: john Mark van Rozendaal at Jul 18, 2007 7:32:56 AM
They are extremely smart, have an excellent sense of humor, and are never boring. And that's in their second language, Spanish.
That's all well and good, but the topic of the book is the correlation between IQ and wealth. So, the relevant question is, are the people of San Agustin Oapan wealthy?
A correlation between IQ and wealth may exist independently of whether or not there are other types of intelligence.
Posted by: Franz at Jul 18, 2007 7:57:05 AM
Poverty and wealth are measured in a way that gives high numbers to the results of massive capital investment. So of course these very smart people would have low IQs. They acquired their skill sets in a low capital environment. IQ measures some of the skill sets that are advantageous for massive capital investment.
Posted by: Huggy at Jul 18, 2007 8:01:49 AM
Perhaps I'm missing something, but I'm not sure I see the controversy in saying a country's average score on IQ tests is directly related to its national wealth. I think the fact calling it an IQ test , which is a misnomer, is the problem since no one seems able to definitively say what intelligence is in the first place. Leaving that aside and given the relatively consistent results of multiple applications of the test, we can assume the test is measuring something and leave it at that. Whether it's a certain way of thinking, a certain body of knowledge, or what have you the end result still seems to be that increasing whatever the test is measuring leads to a more economically well off society.
That's what people should be focusing on, not a nebulous philosophical argument on the nature of intelligence and whether or not one body of knowledge (say natural lore) is as valuable as another (say formal math and language skills).
Posted by: Dan Chituc at Jul 18, 2007 8:03:50 AM
It is puzzling that anyone should say that IQ can only be measured by written tests. For decades, intelligence tests for illiterates have been used massively, all over the world. I believe that fact is mentioned in the cited book. Also, it is irrational to say that intelligence is a meaningless concept and then to say that some Mexican villagers are as intelligent as anyone else. Is intelligence quantifiable or isn't it?
Oh, and if they are so "smart", how come they're not rich?
Posted by: Robert Speirs at Jul 18, 2007 8:15:56 AM
@Dan:
"Whether it's a certain way of thinking, a certain body of knowledge, or what have you the end result still seems to be that increasing whatever the test is measuring leads to a more economically well off society."
Or the other way around, of course. An economically well off society increases what the test is measuring.
@Robert:
"Oh, and if they are so "smart", how come they're not rich?"
Pff.. why is America richer than Russia, even though football is the former's national sport and chess the latter's? Are russians really that stupid, even though the play chess very well? Or this causality between intelligence and national income like the emperor's new clothes: non-existant.
Posted by: JSK at Jul 18, 2007 8:44:24 AM
It's not just for illiterates... My testing for the gifted program was entirely non-written. This was in 1979. Moreover, the type of testing that was done would have been extremely difficult to do on a written exam.
Whatever it is that IQ measures, a written IQ test would (cheaply) measure something quite different than what I was measured against.
Having said that, there is a well-documented rise in US national IQ over the last hundred years which to my knowlege is yet to be explained, especially since our educational standards have steadily declined over the same period.
Posted by: Nathan Zook at Jul 18, 2007 8:49:20 AM
Thanks Tyler, personal experience is and should be an important part of our understanding of the world.
It would be nice to hear more about your travel experiences on the blog. "my favorite things" columns only take us so far.
Posted by: michael vassar at Jul 18, 2007 8:54:11 AM
Thanks Tyler, personal experience is and should be an important part of our understanding of the world.
It would be nice to hear more about your travel experiences on the blog. "my favorite things" columns only take us so far.
Posted by: michael vassar at Jul 18, 2007 8:54:18 AM
Totally agree with Tyler. For more of the same, see Guns, Germs, and Steel. A quick quote:
"From the very beginning of my work with New Guineans, they impressed me as being on the average more intelligent, more alert, more expressive, and more interested in things and people around them than the average European or American is."
Diamond presents excellent arguments for ancient geographical causes for different levels of development. Concrete factors like the extreme isolation of Australian aborigines are much more important than the results of a culturally biased test.
Posted by: Spergler at Jul 18, 2007 8:56:27 AM
If they do argue that people in poor countries are less smart (as distinct from lower IQ measures), how do they explain it? Nutrition? Hopefully not genes.
They do argue that it's genes:
The significance of the high heritability of intelligence is that it implies that the differences in intelligence between the peoples of different nations are likely to have a genetic basis
(Via http://www.gnxp.com/MT2/archives/001463.html)
Posted by: JewishAtheist at Jul 18, 2007 9:23:16 AM
They are extremely smart, have an excellent sense of humor, and are never boring. And that's in their second language, Spanish.
What's their first language, Nahuatl?
Posted by: Joshua Holmes at Jul 18, 2007 9:36:12 AM
@JSK
"Or the other way around, of course. An economically well off society increases what the test is measuring."
That's always the question in any kind of a statistical study, are we looking at correlation of causation? I haven't read the book (though it's certainly on the reading list now), but I'd imagine if the issue isn't dealt with it's certainly the next logical step. Simply on the face of things, I'm inclined to think that increasing what's traditionally called IQ can't help but drive economic success as people have more opportunities opened to them, but it's certainly possible it could go the other way.
@JewishAtheist
It's a real shame to hear they took the genetic route by way of explanation. Besides the wildly inappropriate connotations blaming poverty on inherent inferiority raises, the simple fact is it should be clear to any student of humanity and history that genetic inferiority is patently false. Now an extremely promising work is tainted with the assumed, if not actual, racist overtones.
Posted by: Dan chituc at Jul 18, 2007 9:52:28 AM
I think Tyler (it's you? check? OK) is right on when he says "[we] like to think intelligence is mostly one-dimensional, but at best this is true only within well-defined peer groups of broadly similar people". I know that by the standards of a traditional farmer or a hunter I would be considered a bumbling idiot, and it's even quite possible that this would hold true even if I had their background (in other words, it may be affected by genes).
But who are we to say that our kind of intelligence is the only important one? In the news, I now see that Norman Borlaug, the father of the green revolution and Nobel peace price recipient, was awarded yet another prize, from Bush this time. It's not surprising, they say he may have saved a billion people from starvation. The thing is, I read his bio on wikipedia, and what do I see?
He grew up on a farm. He failed the entrance exam to the University of Minnesota, but did somehow secure a place anyway. (The reason he could apply in the first place was a depression era social program!).
Posted by: Harald Korneliussen at Jul 18, 2007 10:03:04 AM
Most people from other cultures would not do well on a written IQ test. It reminds me of Stephen Jay Gould's book The Mismeasure of a Man. Some new European immigrants were IQ tested when they got off the boat on coming to America. Of course, many didn't speak English. Some who didn't do well were taken for sterilization. Later it was proved that this process was used to obviously weed out certain groups of immigrants.
Posted by: Scott Miller at Jul 18, 2007 10:07:11 AM
I do not treat this book as foundational because of personal experience.
Correct, you ignore certain science when it runs up against your initial biases. Similarly, global warming must not be happening because my feet are cold right now.
I've spent much time in one rural Mexican village, San Agustin Oapan, and spent much time chatting with the people there. They are extremely smart, have an excellent sense of humor, and are never boring.
None of these things are inconsistent with lower average IQ scores. Humans are, as a species, very smart. This does not mean individual differences in intelligence are artificial or meaningless.
I'm also sure they if you gave them an IQ test, they would do miserably. In fact I can't think of any written test -- no matter how simple -- they could pass. They simply don't have experience with that kind of exercise.
The implied argument here is that the tests are biased against certain 'undeveloped'/'minority' populations. That we haven't measured the same underlying suite of abilities that we have in 'developed' cultural populations or that they do not carry the same economic implications. But this is an empirical claim with methodological avenues open to investigating it. It just so happens those empirical investigations do not support your insinuations. As you might know if you actually read the three relevant books in question or paid attention to their arguments on test validity and reliability which directly address this issue.
When it comes to understanding the properties of different corn varieties, catching fish in the river, mending torn amate paper, sketching a landscape from memory, or gossiping about the neighbors, they are awesome.
See above - you are not at all addressing how individual differences apply to learning these skill sets and others, or how this applies to economic outcomes. You are failing to understand and/or engage the arguments you are purporting to challenge.
Some of us like to think that intelligence is mostly one-dimensional...
Correction: there are some people who actually take the time to learn about the rich science of intelligence, and there are others who brush the entire science aside because they believe they already know what that science says and disagree with this straw knowledge on principle.
So why should I think this book is the key to understanding economic underdevelopment?
The books and the paper by your fellow George Mason colleague, Garett Jones, both provide convincing arguments for why IQ needs to be taken seriously for understanding economic development. It's one of the more robust predictors of economic outcomes, the best human capital measure, we have a lot of world data, it doesn't appear to be simply a proxy or result of another simple environmental variable ("education", "money"), and it predicts economic outcomes just as successfully for individuals and groups of individuals.
There is no "one" variable, and none are probably necessary or sufficient, but ignoring IQ is missing a mighty important piece of the global (and national) inequality picture.
Perhaps your readers would stop pestering you if they got a sense you were taking the research more seriously than exceedingly glib and wrong-headed posts like this suggest you do. Or if this same irritating bad faith approach to the research wasn't so ubiquitous.
Posted by: Jason Malloy at Jul 18, 2007 10:15:19 AM
Bravo on this post---I've been waiting for a well-known blogger to make a statement like this (in contrast to say Posner, who seems to put a ridiculous weight on IQ in his various blog posts).
Posted by: Kenji at Jul 18, 2007 10:25:54 AM
I can't believe that commentators on this blog supports primitive, racist measures of both intelligence and wealth. I thought we had moved on about 100 years ago.
Posted by: Gina at Jul 18, 2007 10:27:04 AM
What's remarkable about the stalwart defenders of IQ = intelligence, is how their own arguments seem to cast doubt on their thesis ... unless they're willing to concede that they themselves score poorly.
Any such concessions?
Posted by: Anderson at Jul 18, 2007 11:23:04 AM
Oh, and if they are so "smart", how come they're not rich?
Bwahahahahaha!
Posted by: fustercluck at Jul 18, 2007 11:25:42 AM
The post shows the massive disconnect between popular ideas on IQ testing and the reality. Ninety percent of academic psychologists think that IQ tests are valid and measure intelligence, while the popular press (and some bloggers) airily dismiss decades of science because it doesn't fit their prejudices. It's as if a union member refused to believe in the benefits of free trade.
Posted by: Dennis Mangan at Jul 18, 2007 11:29:39 AM
Anderson,
You state that the IQ = intelligence defenders "own arguments seem to cast doubt on their thesis".
How so? I don't see anything in your post offering any proof of your assertion.
Posted by: Roy at Jul 18, 2007 11:29:48 AM
Seriously though. Tyler's remarks on Mexican villagers point to a larger problem among the cognitive elites. They spend so much time with other smart people that, on the rare occasion they do interact with people of average or low intelligence, they are surprised that they can hold a conversation with them. But most people in the world can talk, tell stories etc, this is not a sign of IQ, only of general human-level intelligence.
As JP Rushton has pointed out, this wrongly leads people to not accept the validity of low African IQ's, because of their 'winning personality'.
Posted by: adrian at Jul 18, 2007 11:40:26 AM
The problem is that, though Jason Malloy's point may be true - that IQ test results are correlated with (causes of? who knows?) material wealth - the zealots further extrapolate that a high IQ score is the bar by which we should set immigration (or other social agenda) standards.
This country would be NOWHERE with a population of nuclear physicists and nothing but. A healthy economy needs people who measure up intelligently in terms that Tyler mentions - people who might dismally fail such an IQ test.
Posted by: fustercluck at Jul 18, 2007 11:44:54 AM
I'm pretty sure nuclear physicists are capable of running a cach register at Walmart or picking lettuce off the ground and putting it into a cart.
And even if they weren't, setting a high IQ bar for immigrants won't leave the US with a population of only physicists. There will still be over 150 million Americans with a 2 digit IQ left to pick lettuce.
Posted by: Roy at Jul 18, 2007 11:57:30 AM
I can't believe that commentators on this blog supports primitive, racist measures of both intelligence and wealth. I thought we had moved on about 100 years ago.
Nice try, but your shock, dismay, and namecalling don't constitute an argument. Did you have something substantive to add?
the zealots further extrapolate that a high IQ score is the bar by which we should set immigration
Actually, I haven't seen this proposed explicitly. Most restrictionists seem to favor either simply lower numbers of immigrants period, or else credential/skill based evaluation. In any case I would expect that if someone did take this approach, setting the bar at the US average of a hundred rather than 'a high IQ score' would be quite sufficient.
Tyler and Jared Diamond both play the same trick in finding a bunch of nice things to say about intellectual capabilities of the third worlders they visit. It's OK to make comparisons as long as they show the Mexicans/New Guineans/whoever in a favorable light. It would in the long run be more helpful to these people to emphasize the role of education, abstract thought, and mathematics in creating a wealthy society; but that would be eurocentric and perhaps racist. Better to coo condescendingly about their fantastic fish-catching and paper-mending skills, as if they were three years old.
Posted by: bbartlog at Jul 18, 2007 12:29:14 PM
I can't speak for anyone else, but I'd think those mentioning how intelligent people in underdeveloped countries actually are should take this kind of study as a sign for optimism, not as a sign the idea of IQ is inherently useless or racist. The entire reasoning behind the viewpoint everyone is "intelligent" is that all people are basically the same and all equally capable of intellectual success. This study at the very least suggests that all we have to do (or a major part of our strategy going forward) is to help people in underdeveloped countries bridge the gap between a high "intelligence" and a high IQ as measured by the test.
I honestly fail to see how saying "group X that isn't succeeding is lacking the intellectual skills and training needed" is automatically racist. There's certainly the possibility that this thinking can lead into an assumption of inferiority or superiority, but the way to fight that is to again look at the facts and discount that possibility, not to try and discredit this entire system for studying reality.
Posted by: Dan Chituc at Jul 18, 2007 12:45:45 PM
I have had experiences similar to Tyler's everywhere -- Washington County, Maine, the slums of DC, rural backwaters of Texas, among the woodchucks of Vermont. People who live at subsistence level and who never read can often impress.
Posted by: Kent Guida at Jul 18, 2007 12:57:13 PM
Tyler wrote: "So why should I think this book is the key to understanding economic underdevelopment?"
Perhaps because planting corn, catching fish, sketching a landscape or gossiping about neighbors are hardly activities designed to enhance economic development? Subsistence agriculture, no matter how nobly endured, is not a sign of intelligence nor an indicator of success.
Posted by: Nathan Park at Jul 18, 2007 1:04:08 PM
Re: Zook:
Having said that, there is a well-documented rise in US national IQ over the last hundred years which to my knowlege is yet to be explained, especially since our educational standards have steadily declined over the same period.
The Flynn effect, yes. I don't think it's certain that it's actually continued to apply for the generation born in the 1990s -- tests in other industrialised nations (most recently in Great Britain) have apparently indicated that the Flynn effect has reversed, and average IQ scores are going down. The same thing may be occuring in the United States. No idea whether these tests have disaggregated the natives, though.
Regarding standards of education -- our educational standards for good comprehensive educations may have declined somewhat, but the average educational expectations now are, I think, rather higher than they were in 1907. In that we now expect everyone, including the very poor, to have some, and to be able to read and write. Not just the toffs and the bourgeoisie anymore. The decline in educational standards across the entire population, if there has been one (and I think there has) is much more recent, probably dating from the 80s or 90s or so.
Posted by: Taeyoung at Jul 18, 2007 1:13:38 PM
Perhaps instead of shouting down people who present theories people find objectionable we should let peer review determine what's valid.
"What Dr. Lahn told his audience was that genetic changes over the past several thousand years might be linked to brain size and intelligence. He flashed maps that showed the changes had taken hold and spread widely in Europe, Asia and the Americas, but weren't common in sub-Saharan Africa."
http://online.wsj.com/public/article/SB115040765329081636-T5DQ4jvnwqOdVvsP_XSVG_lvgik_20060628.html?mod=blogs
We will never know if Dr. Lahn's research is valid because the thought police have shut him up and he is moving on to other areas of research.
Posted by: noman at Jul 18, 2007 1:15:55 PM
Dear Dan chituc,
your post makes me nostalgic about the atmosphere of intellectual freedom that prevailed under the Inquisition.
Pope Urban VIII at least knew Galileo's work and allowed him to write about the Copernican theory as long as he treated it as a mathematical hypothesis.
May I be as bold as to suggest that your moving sentences (like "it's a real shame to hear they took the genetic route by way of explanation")may bring tears to our eyes but are a bit short on facts and long on moral posturing?
Your devoted Attila S.
Posted by: Attila Smith at Jul 18, 2007 1:31:12 PM
Tyler, after reading this post and comments, I have a post title request: should you judge a blogger by his commentators?
Looking forward to it.
P.S I'm a loyal MR reader.
Posted by: seer at Jul 18, 2007 2:29:04 PM
Apparently some people hear have never heard of culture neutral tests like Raven's Progressive Matrices, which even illiterates can take. I find it especially funny that people are still citing Stephen Jay Gould and his mismeasure.
Intelligence is not merely caused by prosperity, as it was detected in poor far east countries like China long before they began to prosper. The reason Russia and China both underperformed for their IQs is that they had communist governments. Russian and Chinese immigrants tend to do quite well abroad. Good examples of how communism can produce drastically different results for basically the same people are East vs West Germany, North vs South Korea and Hong Kong vs Mao's China.
This post from Tyler was very weak. No data, just warm fuzzy talk about fishing, corn and gossip among peasants, none of which has an bearing on "IQ and the Wealth of Nations".
Posted by: TGGP at Jul 18, 2007 2:38:06 PM
If anybody out there actually wants to learn about the book, my 2002 review covers its strength and weaknesses:
http://www.vdare.com/Sailer/wealth_of_nations.htm
In 2004, I put all its natioanl average IQ scores in a convenient table:
http://www.isteve.com/IQ_Table.htm
Also, see my discussion of the table here:
http://www.vdare.com/sailer/lynn_and_flynn.htm
Lynn has published two follow-up books. Jason Malloy's massive review on GNXP is here:
http://www.gnxp.com/blog/2006/02/world-of-difference-richard-lynn-maps.php
And my shorter review, with lots of pretty graphs, showing changes in relative IQ by nation over time, is here:
http://www.vdare.com/sailer/060423_lynn.htm
I've also written a couple of article about how to raise IQs in the Third World through adding iodine and iron to staple foods:
http://www.vdare.com/Sailer/national_iq.htm
http://www.vdare.com/sailer/copenhagen.htm
Posted by: Steve Sailer at Jul 18, 2007 3:08:24 PM
Robert Lucas has a famous saying about growth rates between countries: "The consequences for human welfare involved in questions like these are simply staggering: Once one starts to think about them it is hard to think of anything else." Yet despite that bold proclamation, economists still don't know why some countries are rich and others persistently poor. People seem to be coming around to the view that technological advance is the key (e.g. Greg Clark's new book); yet what allows technological advance to occur? It's not unreasonable to think that IQ (or something like it) could be some part of the answer. I think it would be somewhat embarassing for the economic profession if it turns out that the most important question in economics is answered by psychologists instead of economists...
Posted by: tc at Jul 18, 2007 3:28:36 PM
Not that it matters, but I would be most interested in reading Tyler's response to Jason Malloy's comments.
Posted by: Chip Smith at Jul 18, 2007 3:36:12 PM
I find this debate somewhat amusing. On one side, we have moral posturing by people who obviously haven't read the scientific literature on IQ, g, intelligence, etc. (and don't seem to want to know because it might upset them). This group also has a deeply flawed/one-sided view of the history of IQ testing (most likely derived from Stephen J. Gould, who, to put it as politely as I can, basically lied about IQ testing and, for what it's worth, taught a course at Harvard called "Science as a Social Weapon" - not agenda driven? Sure). For instance, all of the assertions that Scott Miller parrots from Gould's Mismeasure of Man are false.
On the other side, there are people who have read the literature and are willing to patiently explain it. Bravo.
Here, as always, the 800 pound gorilla in the corner is not that there are group differences in IQ, or that these differences accurately predict, or should I say are a reflection of, group differences in real world performance, behavior and outcomes in many domains of human achievement (FWIT, they do), but that these group differences may have a genetic component. There is quite a lot of evidence indicating that genes do play a substantial role in intelligence differences between people, but we are just in the beginning stages of understanding the exact mechanisms by which this works. (Take a look at Gray and Thompson's Neurobiology of Intelligence in Nature Neuroscience Reviews June 2004 and the book Origin of Mind by Geary (2005) for a detailed discussion.) It is currently not clear to what extent (if at all) that racial/ethnic group difference in psychometric intelligence are due to genetic differences, however, most specialists in the relevant fields seem to think that genes do or could play a role in accounting for some of the observed phenotypic differences in measured intelligence and behavior.
Many people viscerally want to reject out of hand the possibility that there could be genetic differences between different human groups that influence talent and temperament. Some even want to reject the idea that there is even such a thing as "races" or "population groups." (Most of these people may be familiar with some form of "Lewontin's fallacy - for a good rebuttal of this, see A.W.F. Edwards article "Human Genetic Diversity" in BioEssays 25.8). However, recent genetic data shows that human population groups do indeed cluster genetically into groups that roughly correspond with traditional racial classifications used by physical anthropologists, i.e., Eurasians/Caucasians (encompassing Europeans, Middle Easterners, North Africans and South Asians), East Asians, Amerindians, Sub-Saharan Africans, and Oceanic (Australian Aborigines, Melanesians). These groups are not platonic and there are obviously fuzzy boundries (e.g., the Turkic peoples of Central Asia are a grading of East Asian into Eurasian and Polynesians and Micronesians are a grading of East Asian and Melanesian). There is also evidence that since Human populations started to seperate from each other and move into different environments about 50,000-60,000 years ago, they have been under strong selective pressure and have genetically diverged from each other somewhat. (See Nicholas Wade's NYTimes article: Humans Spread Globally, Evolved Locally from 6/26/07 for a good summary of this, or check the source articles he mentions for greater understanding). This certainly opens the door to the possibility of average genetic differences in talents and temperament between population groups, but, at present, we do not know if this is the case.
Additionally, here is the abstract of Henry Harpendings presentation to the 2007 meeting of the Assoc. of American Physical Anthropologists:
Understanding human races: the retreat of neutralism.
Henry Harpending
"Discussion and debate about human races has been dominated for decades by neutral theory and statistics. Since this literature never posed a real question, it has never produced an answer. Lewontin's 1972 paper with its claim that a value of 1/8 of a statistic like Fst is “small” and that this means that human race differences are insignificant is a staple of our textbooks. Recently geneticists have had a closer look and pointed out that Fst of 1/8 describes differences among sets of half sibs and few claim that half sibs are insignificantly related. Anthony Edwards has shown that the significance of differences is in the correlation structure of a large number of traits, again denying the Lewontin assertion that human differences are small. Alan Templeton in 1998 claimed that human races were less differentiated that races of some other large mammals, but he compared human nuclear DNA statistics with statistics from mtDNA in the other species. An appropriate comparison shows that human are more, not less, differentiated than other large mammal species. Since neutral differences are a passive record of demographic history they are not very significant for issues of functional biology. Newly available data sources allow us to study the natural selection of race differences instead of their drift. It appears that there is a lot of ongoing evolution in our species and the loci under strong selection on different continents only partially overlap. Human race differences may be increasing rapidly."
Posted by: Philly Guy at Jul 18, 2007 3:55:23 PM
I don't know much about this stuff, but Jason Malloy's comment seems a lot better than Tyler extremely dismissive post. Why do we all have such strong feelings on things we know so little about?
"The problem is that, though Jason Malloy's point may be true - that IQ test results are correlated with (causes of? who knows?) material wealth - the zealots further extrapolate that a high IQ score is the bar by which we should set immigration (or other social agenda) standards."
Isn't this beside the point. Aren't we discussing whether the hypotheses in the book are true or not? If they are, wouldn't you at least like to know it so you could base your own arguments in fact?
Posted by: josh at Jul 18, 2007 4:05:39 PM
The IQ test could be taken from pre school level, so there are IQ tests which does not contain any text For cross country studies there are two 'fair' tests ( they are most used in references for the book ) One was already mentioned Raven's Progressive matrices another one is Cattell Culture Fair test.
There is one interesting book which explains about view on 'one' intelligence.
"The g factor" by Arthur R. Jensen. Of cause there are views that there are multiple intelligences but to some extent they could live peacefully with Jensen's view. And what is more important progressive matrices test is mostly g-loaded - so the 'measure' should be correct.
Posted by: Sergey Kurdakov at Jul 18, 2007 4:05:52 PM
I'm sympathetic with both sides here. On the one hand, IQ seems to be a real measure of something, and that something seems to correlate to other somethings. On the other hand, IQ ain't everything, smart people can be awfully dumb as well as mess the world up, people who don't score well on IQ tests can show all kinds of gifts, etc ...
So, FWIW, I sort the problem out this way for myself.
Instead of falling into the trap of arguing what's intelligence and what's not, I use a meta-category of "talents."
In other words: there are tons of different kinds of talents, from "character" to "a feel for dogs" to "a nice singing voice" to "intuitiveness" to .... Well, to what I call "IQ-style intelligence."
In other words, why fight over whether "having an amazing instinctive rapport with dogs" is a form of "intelligence" or not? Seems dumb. Why not say that "having an amazing instinctive rapport with dogs" is one kind of talent, and "having a dazzling IQ-style intelligence" is another one?
That way you get the best of both worlds: people are, yes, multidimensional and do have many differetn attributes and characteristics; and, yes, there is such a thing as IQ-style intelligence and it may even be a pretty important thing.
A and B, not A or B.
Posted by: MIchael Blowhard at Jul 18, 2007 4:08:52 PM
Isn't this the point? The mental skills that many hunter gatherer or village type societies possess are not the skills that will lead to success in a modern society. If I'm stuck in desert, having a !Kung Bushman as my companion would undoubtedly increase my survival chances several fold. If you took that same guy to New York City, that guy is going to be lost. You could probably bring him as a child and he still wouldn't do very well because natural selection has honed his brain for an entirely different environment. My ancestors, however, have lived in settled communities for thousands of years.
Posted by: KingM at Jul 18, 2007 4:24:52 PM
@ Attila Smith
As I freely admitted, I haven't read the book yet (though I absolutely intend to in the near future), but even given the fact I don't know the entirety of the material in the book I can't say I agree with the light you're trying to paint my objections in. It IS irresponsible to attribute lower performance on an IQ test to systematic inferiority of a racial or geographic group unless (and note this unless) there is compelling evidence to suggest that link above other possible explanations. I certainly don't consider the fact that an underdeveloped country is scoring lower on IQ tests to be evidence that they are genetically less intelligent, and I doubt most people would either.
If the book does provide evidence to back up such a claim, or cites other works that provide that argument, then I'll be front and center apologizing for having jumped the gun in judging their claim and I'll be happy to discuss the issue based on the merits of that evidence. I just happen to find it doubtful that a book devoted to the analysis of national IQ tests also happened to find the time and space to build a compelling case for a conclusive genetic inferiority in intelligence among certain racial groups.
I'm not saying that this issue is taboo and can't be studied, but I am saying given the history of such claims being made without basis and used to extremely distasteful ends, there is a burden on people discussing the issue to be sure of their facts before they make claims. That should be a given in all areas of intellectual discussion, so I don't think it's at all an undue burden on research to make sure that's the case here.
Posted by: Dan Chituc at Jul 18, 2007 4:27:42 PM
Maybe I'm wildly off base, but it seems to me the problem is that people on both sides of the debate view IQ as a value judgement. "Intelligence" is a loaded word that probably shouldn't be used. I think it is fair to say that IQ measures more or less the ability of a person to function succesfully in Western society. I think it is fine to state there are different types of intelligence - if you view intelligence as a measure of a groups succesful adaptation to its environment. I'm sure that many high IQ people from the US would fail miserably if they had to live the life of that Mexican village. What the "IQ and Wealth of Nations" really suggests is that the children and grand-children and great grandchildren of those Americans would probably also be less successful at that life style, in the same way descendents of those Mexican villagers don't do well in our environment. Genes will tell.
Posted by: vanya at Jul 18, 2007 4:32:10 PM
Civilizations have have been around for over 5000 years and during that time the location of the most prosperous has not been in the countries that are now developed. For the most part there was little movement of population until recent times. Any theory that tries to connect economic success of nations to genetics through IQ testing needs to account for the fact that Northern Europe was a backwater until 500 years ago and during much of human history places that are now third world countries were the centers of civilizations. The correlations between the results of IQ testing and national income need a lot more study to determine which is the cause and which is the effect. From what I know of IQ testing, it represents a composite number measuring many abilities, which is why people with the same IQ excel in different fields. Which of the abilities are most valuable is determined by the nature of the society, for example visual memory is important in learning to read and also in a world without maps so you can find your way home, but maybe maps were invented by people compensating for the lack of this skill.
Posted by: joan at Jul 18, 2007 4:40:46 PM
Can someone point us to a summary of Lynn's methodology.
More precisely, how does he control for:
1. Type of government (e.g. socialist vs. capitalist vs. tribal)
2. Political stability
3. Opportunities for education (this is very different from IQ)
4. Geographical advantages
and maybe other factors.
Without adequate control for these or similar factors the book may very well be useless.
Posted by: Alex at Jul 18, 2007 4:48:32 PM
Cowen is mixing (at least) two things.
Do the villagers according to him actually have strong ability for abstract thinking, and are only not well measured by potential IQ tests because of lack of experience? Or o they have lower g, but compensate with a bunch of other soft qualities?
Not clear. If Tyler Cowen honestly believes in the first hypothesis it should be easy to test. There are PLENTY of IQ tests that are not written. IQ tests in third world countries are by the way mostly done by the educated middle class in the cities, not illiterate villagers.
Like many others Cowen confuses a scientific explanation (low average IQ) with a personal attack on his friends in Mexico. He shouldn’t. IQ is not a measure of human worth. Hitler probably had a reasonably high IQ.
However if Cowen indeed admits that a group are good at gossiping and catching fish, but not particularly good at things the market actually values, such as processing abstract information, he has not dispelled the IQ-GDP link.
Lastly Tyler Cowen is making the same feel-good-but-silly mistake Diamond makes in Guns. Fine, you are impressed by how much more than you the natives know about their environment. But the relevant question is how much time Cowen or Diamond would need to match and supersede the natives at “predicting rain”, identifying plants in their surrounding or “mending torn amate paper”.
If you lived there your entire life, or even 6 months in their village, would the average high-IQ individual honestly do worse than Yali and Juan Camilo? Would Juan and other Mexicans do as well as Americans in IQ tests and intellectually demanding (and from the market demanded) tasts if they went to American schools and became familiar with paper and pen tests?
So why aren’t they?
Posted by: Tino at Jul 18, 2007 4:57:56 PM
Before speculating that "the book may very well be useless," try reading the articles on the book that I cited above.
Posted by: Steve Sailer at Jul 18, 2007 4:58:30 PM
As I freely admitted, I haven't read the book yet (though I absolutely intend to in the near future), but even given the fact I don't know the entirety of the material in the book I can't say I agree with the light you're trying to paint my objections in. It IS irresponsible to attribute lower performance on an IQ test to systematic inferiority of a racial or geographic group unless (and note this unless) there is compelling evidence to suggest that link above other possible explanations. I certainly don't consider the fact that an underdeveloped country is scoring lower on IQ tests to be evidence that they are genetically less intelligent, and I doubt most people would either.
It's been a while since I read the book, but if you look into the research on intelligence testing, there is some extremely compelling evidence that the differences in intelligence observed between certain racial groups is substantially genetic.
Take whites and blacks. There is a fifteen point gap in average IQ scores between these two populations that consistently shows up in every developed country where the two groups are tested, from America to Britain to Brazil. In America, a metastudy (google: Minnesota Scarr Adoption Study) found that while black children adopted as infants by middle class white parents scored higher than the black average as young children (though not as high as their white adoptive brothers and sisters), those gains eroded as they grew older, and by adulthood, their IQs had regressed to the black mean.
Why should this be? Studies of twins who have been seperated at birth show that, while the twin raised in the better household often had a higher IQ during childhood, the twins' respective IQs approached parity as they aged. This suggests that environmental influences on IQ wear off as we get older, and as adults, we are left more or less with whatever IQ is set by our genes. If you look at the Minnesota Scarr Adoption study with this knowledge, it really does seem that black children get a "boost" from being raised by whites, but that the boost is artificial and temporary. As adults, they do no better on intelligence tests, on average, than other blacks.
So if the fifteen point gap between blacks and whites is genetic here in the U.S., we should expect 15 points of whatever gap is present between black nations and white nations to also be genetic, since whites in Europe share the same general gene pool as whites in the States and blacks in West Africa share the same general gene pool as do blacks in the States. If I remember the book's findings correctly, the gap between Europe and sub-Saharan Africa was more along the lines of 20-25 points. So it stands to reason that 15 points of that difference is genetic, and the other 5-10 points is due to severe rates of malnutrition, vitamin deficiencies, and God knows what else.
In regard to the differences between other racial groups, there hasn't been as much testing, so it's more difficult to say. Personally, I suspect that the 5 point difference between East Asians and whites (favoring East Asians) is probably genetic, because in every white country that East Asians move to, from the U.S. to Britain to Italy to Russia, they consistently outperform the natives. But I wouldn't take a firm stance on that, just because I haven't seen the same quantity of evidence.
Anyway, the information is out there. You just have to have a strong enough stomach to go through it. Personally, I sympathize with people who are resistant to the idea that these differences exist, because a lot of crap has been done to people in the name of real and perceived racial differences. But if the science says these differences are there, what can you do?
Posted by: Marc at Jul 18, 2007 5:35:00 PM
Sailer,
I've browsed through your links -- they seem more geared towards evaluating the validity of the IQ results with some speculation on what the IQ vs. national income correlation implies.
That's not what I asked for though.
I've asked for what controls does Lynn use for type of government, opportunity for education and the like. The only relevant information I found on this in your links was a claim that GDP has a 0.63 correlation with an index of economic freedom.
I still don't know the answer to this basic question: do they do multivariate analysis with GDP regressed on IQ, Index of economic freedom, geographic factors etc. -- or do they just regress GDP separately with each of these variables?
Given that the correlation between IQ and GDP (0.7) is similar to the correlation between the index of economic freedom and GDP (0.63), does Lynn do any work at all trying to disentangle what is the real causal factor of high GDP or is it just speculation?
I didn't say that Lynn's book was useless -- just that without more detailed analysis, the discovery of a correlation between IQ and GDP is useless.
I still want to know details about Lynn's analysis.
Posted by: Alex at Jul 18, 2007 6:07:01 PM
Professor Cowen & his supporters make a rather curious implicit assumption:
There is no population differences in intelligence.
I cannot take anything you say on the topic seriously until you present a credible evolutionary theory supporting this claim. Can you? If so, please share as I am rather curious and have wondered for years...
Also, recent empirical evidence suggesting substantial recent human evolution, including in neuronal function must be dismissed. I suggest you start with thrashing this rather 'racist' pnas article from 2005
http://www.pnas.org/cgi/reprint/103/1/135?maxtoshow=&HITS=10&hits=10&RESULTFORMAT=&fulltext=darwinian+selection+landscape&searchid=1&FIRSTINDEX=0&resourcetype=HWCIT
Good luck!
Posted by: BW at Jul 18, 2007 6:23:31 PM
Tommy
"Because the requisite IQ needed to build and maintain civilization back, say, during the days of the ancient Sumerians, Egyptians, or Romans was much lower than it is today."
Maybe - but given what we now know about the speed with which evolution can occur it is possible that people were very different back then, historically there could've IQ population peaks followed by inevitable dysgenic effects caused by migrations, war etc, and on and on.
Posted by: adrian at Jul 18, 2007 6:25:13 PM
I would be most interested in reading Tyler's response to Jason Malloy's comments.
If past history is any guide, you will be disappointed.
Posted by: RJ at Jul 18, 2007 6:25:20 PM
Shouldn't we move beyond self-reported "race" to actual genetic heritage and admixture (haplogroup type, etc.) in studying IQ, populations, and heredity? I've read that 30% of white-identified Americans have recent african genetic admixture, and I suspect at least an equal percentage of black-identified Americans have recent european genetic admixture. For example, I think something on the order of 25% of black-identified men in the United States have a European Y chromosome, and something like 15% of black-identified people in the United States have European mitochondrial DNA.
Posted by: Hopefully Anonymous at Jul 18, 2007 6:43:19 PM
Alex:
A lot of the factors you mention (good governance etc.) are partially or wholly a function of IQ, and should not be controlled for.
Multivariate regressions are not very accurate in my opinion, but since you asked for it: Yes, they do uphold the predictive power of IQ as a determinant of growth.
http://www.siue.edu/~garjone/JonesSchneslides.doc
The history of civilisation is consistent with IQ differences, especially assuming that selection for it took of a few thousand years ago. China, ancient Greece and Rome are all high IQ areas. Mesopotamia, Egypt and Indus might well have been above average 7000 years ago.
Invading more advanced neighbours and claiming their achievements does not count (the so called Muslim civilisation). For all the hype the Americas only managed a stone-age level civilisation, without literacy or advanced tools. Same for Africa and Oceania.
Posted by: Tino at Jul 18, 2007 6:50:28 PM
Hopefully Anonymous:
It is a question of degree of relation, not "on drop of blood". The seperation holds, and mixed groups tend to score in between. Certainly this should be taken into account in Mexico.
Posted by: Tino at Jul 18, 2007 6:53:52 PM
Regarding objections raised that russians or chinese should be just as rich as americans if their IQs are similar, refer to here:
http://rpongett.phpwebhosting.com/gdp.html
which gets excellent GDP predictability using 3 variables: IQ, economic freedom, and oil wealth. The methodology is overly complicated (the guy should be using log-income instead of income, I think) but the relationship is there.
As an aside, I wonder if Tyler posts these troll-bait entries *expecting* that the well-informed will come in and beat up the facile arguments in them to a pulp. It's as if he desparately wants the truth to be known, but desparately doesn't want to be seen as believing it! What other explanation could there be for repeated blog posts, featuring much more simple-minded arguments than the average post quality, with open comments?
Posted by: pwyll at Jul 18, 2007 6:58:39 PM
Yes, I think pwyll might be right.
Posted by: Steve Sailer at Jul 18, 2007 7:16:49 PM
Tino,
The claim that economic freedom is caused by high IQ is a highly dubious claim. There are above average IQ people that still believe in socialism and stupid people that oppose it.
Political institutions depend to a high degree on historical contingency, intellectual fashions and the like.
At any rate, even assuming that you're right, the causal chain of high IQ --> economic freedom --> economic prosperity has quite different policy implications that a more direct high IQ --> economic prosperity.
For instance, it focuses on reforming governmental institutions as a solution to poverty rather that claiming that in order to advance economically you should watch your diet.
Multivariate analysis can have problems, but at least it can be used to make more sensible claims about causation. Given that both IQ and economic freedom have similar correlations with GDP when regressed separately, no conclusion can be drawn about the causal factor without further analysis.
Posted by: Alex at Jul 18, 2007 7:25:05 PM
Dear Professor Tyler,
although I happen not to agree with you on today's post, I want to tell you my admiration for the gentlemanly way you post comments that attack your point of view.
I am also impressed by the erudition and intelligence (oops, the "I" word...) of your readers .
Blogs, alas, are not like that in Smugistan, my country.
Yours gratefully, Attila S.
Posted by: Attila Smith at Jul 18, 2007 8:14:45 PM
Let's just cut the bullshit and move straight to the master race/eugenics part of the discussion, shall we?
It's the pink elephant in the room at which many of you are winking.
Posted by: fustercluck at Jul 18, 2007 8:42:21 PM
Hopefully Anonymous,
The self report rates are fairly commensurate with genetic tests. The average self reported white in America has 2% African ancestry. Obviously, for the most part, that does not make much of an impact on the phenotypes of white Americans.
Posted by: Dareano at Jul 18, 2007 8:45:50 PM
And now come the myriad protestations...
Posted by: fustercluck at Jul 18, 2007 8:46:49 PM
"Because the requisite IQ needed to build and maintain civilization back, say, during the days of the ancient Sumerians, Egyptians, or Romans was much lower than it is today. Climate and geography were much more important determinants of where civilization would arise back then."
It has always seemed to me that it would take higher IQ or more creative people to build a civilization, invent writing, and create institutions etc.than it does to learn about what has been inherited. All humans everywhere lived as hunter gathers until about 10,000 years ago when farming started independently in many places around the world. Any selection or adaptation to live in organized communities could have started no sooner than that and would have been at work in many locations. Here is a best guess based on DNA tracking of the populating of the earth.
http://www.bradshawfoundation.com/journey/timeline.swf
Any theory about inherited characteristics must at a minimum account for the correct movement patterns and time scales. It is clear that there has been selection in different population for disease resistance on time scales of 1000's of years, and so it is possible that there has been some selection for different skills and abilities for living in cities. However the Vikings were among the last people in the world to adopt a "civilized" life style and yet their descendant are among the most successful economically. Lacking a explaimation for inherited IQ effects on historical wealth of nations, I think at least as likely reason for the correlation of IQ and income is that high GDP causes high IQ.
Posted by: joan at Jul 18, 2007 8:51:39 PM
Sure fustercluck, why not. I mean, it's not like you can refute comments made from posters like Jason Malloy. Whenever this topic comes up one side presents well thought out arguments backed by a ton of research and the self satisfied and smug commentators simply retort, "bububuracist".
Also no researach shows that 'the master race' has the hishest average IQ. Odd...
Posted by: Dareano at Jul 18, 2007 8:57:24 PM
Sure fustercluck, why not. I mean, it's not like you can refute comments made from posters like Jason Malloy. Whenever this topic comes up one side presents well thought out arguments backed by a ton of research and the self satisfied and smug commentators simply retort, "bububuracist".
Also no researach shows that 'the master race' has the hishest average IQ. Odd...
Posted by: Dareano at Jul 18, 2007 8:58:25 PM
The nature vs. nurture arguments about IQ can be a distraction because what we know is that differences in average national IQ are going to be around for a long time. If the gap suddenly disappeared in all the babies being born tomorrow for some magic reason, the gap among the workforce wouldn't begin to shrink until 2025 and wouldn't disappear until 2072.
So, the current realities demand far more study than they've gotten from the economics profession. I appreciate Prof. Cowen's complicated method for calling this to the attention of his fellow economists.
Posted by: Steve Sailer at Jul 18, 2007 9:44:49 PM
The relative gaps among groups have been fairly stable for several generations. East may have picked up a few points on everybody else over the last 40 or 50 years, and Maoris in New Zealand may have done the same (although that's disputed), but otherwise, remarkable relative stability is the norm.
One of Lynn's recent books, for instance, lists 620 IQ studies of different groups going back, in a few cases, to the first quarter of the 20th Century. I've created a graph showing that there has been no overlap of average scores among Japanese (23 studies), Hispanics in America (17 studies), and Australian Aborigines (17 studies).
The Japanese have consistently scored somewhere around 105 on a scale where white American are pegged at 100, Hispanics in the U.S. at about 90 (which, by the way, is roughly the world average), and Australian Aborigines at very low levels.
http://www.vdare.com/sailer/060423_lynn.htm
I don't know what causes differences in average IQ, but the evidence is overwhelming that they were stable enough over the 1950-2000 era to be used in studies, and, with almost as much assurance, that they will be stable enough over, say, the 2000-2030 era to be important to study for the future.
Posted by: Steve Sailer at Jul 18, 2007 9:49:19 PM
I think you don't read very well. (Please don't breed.)
I'm conceding that there might in fact be correlation between wealth building and IQ - I won't go so far as to admit causation but for argument's sake, let's say that's true.
Now what? Is there something that should be done with that information? Is it outrageous to point out that theories like this have driven nefarious agendas in the past, and suggest that this line of thought would be used for those same purposes now and into the future?
Amazing how much time and energy certain people spend in proving theories only to play the dumb card at the conclusion.
Posted by: fustercluck at Jul 18, 2007 9:52:18 PM
I'll respond to Jason in CAPS:
"...you ignore certain science when it runs up against your initial biases. Similarly, global warming must not be happening because my feet are cold right now.
THAT'S STARTING WITH AN AD HOMINEM ATTACK, PLUS IT IS ATTEMPTING GUILT BY ASSOCIATION; GLOBAL WARMING IS CLEARLY REAL.
I've spent much time in one rural Mexican village, San Agustin Oapan, and spent much time chatting with the people there. They are extremely smart, have an excellent sense of humor, and are never boring.
None of these things are inconsistent with lower average IQ scores. Humans are, as a species, very smart. This does not mean individual differences in intelligence are artificial or meaningless.
WE DON'T KNOW WHAT TO MAKE OF THE CORRELATION BETWEEN NATIONAL INCOME AND IQ. NONE OF YOUR POINTS HERE REBUT THIS.
I'm also sure they if you gave them an IQ test, they would do miserably. In fact I can't think of any written test -- no matter how simple -- they could pass. They simply don't have experience with that kind of exercise.
The implied argument here is that the tests are biased against certain 'undeveloped'/'minority' populations. That we haven't measured the same underlying suite of abilities that we have in 'developed' cultural populations or that they do not carry the same economic implications. But this is an empirical claim with methodological avenues open to investigating it. It just so happens those empirical investigations do not support your insinuations. As you might know if you actually read the three relevant books in question or paid attention to their arguments on test validity and reliability which directly address this issue.
THERE IS MORE AD HOMINEM HERE. IT IS FAIR THAT YOU CITE LITERATURE BUT THE PROFESSIONAL CONSENSUS AS A WHOLE DOES NOT GO AGAINST MY POINTS.
When it comes to understanding the properties of different corn varieties, catching fish in the river, mending torn amate paper, sketching a landscape from memory, or gossiping about the neighbors, they are awesome.
See above - you are not at all addressing how individual differences apply to learning these skill sets and others, or how this applies to economic outcomes. You are failing to understand and/or engage the arguments you are purporting to challenge.
I'M NOT DOUBTING THAT SOME PEOPLE ARE SMARTER THAN OTHERS.
Some of us like to think that intelligence is mostly one-dimensional...
Correction: there are some people who actually take the time to learn about the rich science of intelligence, and there are others who brush the entire science aside because they believe they already know what that science says and disagree with this straw knowledge on principle.
THIS IS MORE AD HOMINEM. MY ORIGINAL POST ALREADY NOTED THAT THE ONE-DIMENSIONALITY OF INTELLIGENCE HOLDS MORE CLEARLY FOR WELL-DEFINED PEER GROUPS, AND LESS SO ACROSS CULTURES. THE LITERATURE ON NEURODIVERSITY FINDS SIGNIFICANT DEVIATIONS FROM UNI-DIMENSIONALITY OF INTELLIGENCE. IT IS FAIR TO SAY THE MATTER IS NOT SETTLED BUT AGAIN THIS IS MORE ATTACK THAN SUBSTANCE.
So why should I think this book is the key to understanding economic underdevelopment?
The books and the paper by your fellow George Mason colleague, Garett Jones, both provide convincing arguments for why IQ needs to be taken seriously for understanding economic development. It's one of the more robust predictors of economic outcomes, the best human capital measure, we have a lot of world data, it doesn't appear to be simply a proxy or result of another simple environmental variable ("education", "money"), and it predicts economic outcomes just as successfully for individuals and groups of individuals.
I'M A BIG FAN OF GARRETT'S AND I WAS CHAIR OF THE HIRING COMMITTEE WHEN WE HIRED HIM. I'VE READ ALL HIS MAJOR PAPERS. HE DOESN'T SAY WHAT MANY OF YOU SAY HE DOES, AND HE HOLDS HIS PRONOUNCEMENTS TO STANDARDS OF GREAT CARE.
There is no "one" variable, and none are probably necessary or sufficient, but ignoring IQ is missing a mighty important piece of the global (and national) inequality picture.
Perhaps your readers would stop pestering you if they got a sense you were taking the research more seriously than exceedingly glib and wrong-headed posts like this suggest you do. Or if this same irritating bad faith approach to the research wasn't so ubiquitous.
DO NOTE, IN ADDITION TO THE POINTS ABOVE, THAT I SAY CLEARLY THAT THE BOOK MAKE SOME VALUABLE CONTRIBUTIONS. I SIMPLY THINK THAT IQ IS FAR MORE ENDOGENOUS THAN DOES THE AUTHOR. FOR INSTANCE SEE MY PREVIOUS POSTS ON THE FLYNN EFFECT, USE GOOGLE.
I AGREE THERE IS LOTS ABOUT IQ WE DON'T KNOW. MANY OF THE COMMENTATORS ON THIS POST ARE NOT ADVANCING THE DISCUSSION. READERS SHOULD JUDGE FOR THEMSELVES.
Posted by: Tyler Cowen at Jul 18, 2007 9:57:53 PM
Dareano at Jul 18, 2007 8:45:50 PM, Why talk about white-identified americans and african admixture in terms of the "average white american", when 30% of white-identified americans have traceable recent african admixture, and 70% of white-identified americans don't have traceable recent african admixture? Those seem like two discrete populations, and I don't see the utility of averaging them together if one is trying to determine the relations of populations, actual genetic ancestry, and IQ. It seem to argue to me for moving away from self-reporting of social race and towards genetic ancestry and admixture tests in doing these type studies.
Posted by: Hopefully Anonymous at Jul 18, 2007 10:47:23 PM
I find Tyler's position to be closer to Darwin than the genetic IQ group. Darwin believed that evolution is such a complicated and random process that it escapes man ability to contemplate or mirror it in one's mind.
The genetic IQ group believe they have found THE FACTOR that has propelled man through the centuries. Tyler and Darwin are much more humble and therefore open to the likelihood of competing theories. To me, the IQ group come across as creationist.
Posted by: thehova at Jul 18, 2007 11:02:34 PM
Tyler,
Thomas Sowell pointed out years ago that IQ tests that call for abstract reasoning won't necessarily do a good job of discerning the full genetic potential of people who grow up in Third World cultures where abstract reasoning is ignored. Sowell cited the example of two 17-year-old African boys who made witty fun of an IQ test question calling for abstract reasoning. They had a real talent for the "vivacious gossip" you value highly.
But, as Sowell went on to point out, the IQ test was also measuring their economic potential correctly. A 17-year-old who despises abstract reasoning has a limited future in a modern technological economy.
And that's what we see: average national IQ scores have strong predictive power for economics.
Posted by: Steve Sailer at Jul 18, 2007 11:11:46 PM
Hopefully Anonymous:
Your facts are not quite correct. There was an article in Scientific American a few years ago dealing with racial admixture in the US. Basically, African Americans have about 20-25% European ancestry on average (i.e., this is the portion of there genes that are from Europeans as opposed to Sub-Saharan Africans), but this varies from about 10% in the South-East to about 40% in the Mid West and Pacific Northwest. Nearly all African Americans have some European ancestry.
With regard to Non-hispanic white Americans, about 30% of self identified whites have some non-European ancestry, though the purportion of non-European ancestry is not nearly as large as the percentage of African American ancestry that is of European origin. Furthermore, the non-European ancestry in white Americans is overwhelmingly Native American, not African. Most of this mixing occured pre-20th Century, generally in frontier regions, which changed over time. As noted by someone above, the percentage of African American ancestry in whites is negligible (less than 1 or 2%). This is an effect of the one-drop rule in American society where anyone with even a little African descent is deemed "black." Additionally, the proportion of Native American ancestry in African Americans is very small.
With respect to Mexicans, a little under 10% of them are of European ancestry (mostly Spanish) and tend to form the elite that runs Mexico (rather poorly). Roughly another 10% are Amerindians. The vast bulk, however, are Mestizo. Mitochondrial DNA shows that they are virtually all Amerindian on thier maternal lines. Y chromosome haplyotypes show that their paternal ancestry has a large portion of European (Spanish) ancestry (perhaps the majority, but I'm doing this from memory and it was a while ago when I read the article). They also have a substantial Amerindian element in their paternal lineages and a very small Subsaharan African element.
With respect to your point about using genetic ancestry and admixture tests for psychometric and economic studies, it is valid, but largely unnecessary for two reasons. First, in the American context, studies reveal that 99.9% of people tested using genetic markers cluster with their self identified "race." See e.g., Rosenberg, Pritchard, Weber et al. (2002). Second, ancestry per se doesn't really matter when we are looking at differences between self identified ethnic groups. Cultural self identification with a group is more important. It doesn't really matter what proportion of African American ancestry is European or African or what proportion of Mexican ancestry is white or Amerindian. What really matters are the current differences between self identified groups and the degree of their intractability (regardless of whether the reason for intractibility is genetic or not), as Steve Sailer points out. We still don't know to what extent genes play a role in average group differences in talent and temperament (though as I noted above, given the large amount of recent differential natural selection in various human population groups, genetic differences in talent and temperament are not implausible, as noted by Steven Pinker in his 2006 Edge.org question: what's your dangerous idea?). To the extent such genes are identified in the future, it only really matters how they are distributed within self identified cultural/ethnic groups, not what continents their ancestors came from. Depending on the degree of selective pressure, meaningful genetic change can occur quite quickly in a population and a population can therefore be quite different from its parent population(s) in heavily selected traits.
Posted by: PhillyGuy at Jul 18, 2007 11:25:52 PM
Dan Chituc said:
I'm not saying that this issue is taboo and can't be studied, but I am saying given the history of such claims being made without basis and used to extremely distasteful ends, there is a burden on people discussing the issue to be sure of their facts before they make claims. That should be a given in all areas of intellectual discussion, so I don't think it's at all an undue burden on research to make sure that's the case here.
That's just great. Let's shut down scientific inquiry and discussion because of fears of the resurrection of Nazi Germany. How about being a grown up and discussing the issues and not trying to shut down debate? The Nazis are long gone, and they're not coming back. It's safe to come out from under the bed and discuss this topic.
Posted by: Jumpyg at Jul 18, 2007 11:26:05 PM
Is there any reason not to call IQ-test-taking talent "abstract reasoning power" instead of "intelligence"?
It seems to me that at least half the heat that this topic generates has to do with terminology. One side is saying "THIS and only this is intelligence," while the other side is saying "nonsense, there are lots of ways of being smart or talented."
So why persist in insisting on the terminology when it rubs people the wrong way and creates lots of unnecessary resistance?
Posted by: MIchael Blowhard at Jul 18, 2007 11:29:46 PM
"Is there any reason not to call IQ-test-taking talent "abstract reasoning power" instead of "intelligence"? "
Good point Michael. I believe that's why Herrnstein and Murray chose to use the term "cognitive ability" instead of intelligence in the Bell Curve. Heck, we could even refer to it as developed cognitive ability and remain agnostic the role of genetics with respect to group differences. Unfortunately, whatever their causes, the obsevered differences in "developed cognitive ability" do seem to be real as far as their real world economic effects and, as Steve Sailer noted above, seem to have changed little in relative terms in the last 50 years and there apparent intractibility makes it seem unlikely that they will change in the near future.
Posted by: PhillyGuy at Jul 18, 2007 11:39:28 PM
I'm also sure they if you gave them an IQ test, they would do miserably. In fact I can't think of any written test -- no matter how simple -- they could pass. They simply don't have experience with that kind of exercise.
Aren't there some visio-spatial IQ tests that require little or no writing? How would the villagers do on one of those?
Posted by: Peter at Jul 18, 2007 11:46:34 PM
If IQ is the "key to understanding economic underdevelopment" how did the South Koreans get so much more IQ than the North Koreans?
Posted by: aem5 at Jul 19, 2007 1:15:28 AM
The race-IQ gap is NOT entirely genetic. Africans in Africa have much lower IQs on average than African-Americans (even though the former as immigrants are more likely to hold white collar desk jobs than whites, likely due to selection). This cannot be explained by genetic admixture. Gene Expression has the math here.
Steve Sailer has data on how white "blacks" are and how black "whites" are in the United States here. There is a really cool triangle plot of African, European and Amerindian ancestry among hispanic and carribean groups here.
Posted by: TGGP at Jul 19, 2007 1:22:28 AM
Regarding the question of the importance of IQ for economists, the reason is that they often control for various factors to determine how much of an influence other factors have. You might do some regressions showing education to be very important for growth, and then you would think that in order for a poor country to develop it needs to start educating its people. However, education is often correlated with IQ, and controlling for IQ helps to get a better picture of the gains that can be gotten from education, which for most developing countries would still be significant, but will not turn them into Asian/Celtic "tigers". Bryan Caplan at first thought that education was the most important thing for "thinking like an economist", but when he factored in IQ it turned out THAT was the most important. He made the point I just made above here, which I apparently forgot about but remembered part of his argument.
Posted by: TGGP at Jul 19, 2007 1:30:41 AM
The thing I find odd in the discussion so far is the assumption made that national boundaries coincide in some important way with specific gene pools. The measures of GDP used are national and necessarily reflect the pooled effect of all within the national boundaries. For a measure of this nature to correlate strongly with a phenotype with significant levels of heritability would require that the national boundaries approximate genetic divisions fairly accurately. After all, there are significant differences in the growth rate within the larger areas containing the countries discussed. If the genetic differences did not map closely with the national boundaries the correlations should break down very quickly.
Generally people assume that what is naively constucted as race maps to distinct genetic populations. The general idea of race divides humans into groups that correspond to continent of origin. Genetic studies examining genetic variation directly indicate this an inappropriate and inaccurate view. Populations within each continent show significant diversity. In particular Africa contains about 50% of the genetic diversity found within the species. There appear to be groups that can be isolated based on genetic markers, but there are multiple groups within each continent. There appears to be about as much difference among groups on a single continent as there is between groups from different continents.
Given this information, I think it requires a naive outlook (or excessively cynical, the two are hard for me to distinguish) to make the argument that heritable differences in IQ are reliable markers of national or continental differences in anything, especially economic development.
Sorry, that is a little turgid, but my point of view for what its worth.
Posted by: iam at Jul 19, 2007 1:38:52 AM
Right, Richard Lynn has since adopted my point that the roughly 15 point IQ gap seen between the average African and the average African-American is solid evidence for a sizable non-genetic influence on IQ differences. Black Americans share about 80% of their ancestry with West Africans, but their 15 point advantage in average IQ over Africans is perhaps five times larger than a simple genetic model would account for. So, it appears that the African environment tends to depress IQ below its genetic potential.
Something similar may be true for Tyler's Mexican village as well.
But, of course, that doesn't mean that IQ tests are inaccurate at estimating the economic potential of people living in IQ-depressing environments.
According to UNICEF's Micronutrient Initiative, two proven and cost-effective ways to prevent medical syndromes that lower national average IQ is fortifying staple foods with iodine and iron. We've been doing this in the U.S. since the 1930s. Putting iodine in salt has largely eliminated here the disease of cretinism.
Posted by: Steve Sailer at Jul 19, 2007 2:33:42 AM
"The thing I find odd in the discussion so far is the assumption made that national boundaries coincide in some important way with specific gene pools."
Right, so Richard Lynn published a subsequent book after "IQ and the Wealth of Nations" recounting 620 studies focused upon separate racial groups. You can read Jason Malloy's summary here:
http://www.gnxp.com/blog/2006/02/world-of-difference-richard-lynn-maps.php
Posted by: Steve Sailer at Jul 19, 2007 2:40:22 AM
This discussion would be more interesting if the IQ test was renamed a "Wealth Predictor" test.
Posted by: Russell Nelson at Jul 19, 2007 2:55:34 AM
"Putting iodine in salt has largely eliminated here the disease of cretinism."
I need to tell some acquaintance to eat more salt
Posted by: Luis Enrique at Jul 19, 2007 4:09:38 AM
Tyler strikes me as an honest, sunday-school kinda guy, the Ned Flanders of economics, sharing both his naive optimism and faith in the supernatural. He believes what he says.
Posted by: adrian at Jul 19, 2007 5:17:36 AM
Cowen believes in his warm fuzzy theories. He is besieged enough by the IQ-arguments on his comment pages to feel he needs a defence. Unfortunately his refutation is not particularly convincing.
Beside using the word Ad Hominen 3 times, Cowen didn't really respond to Jason Malloy.
1. “WE DON'T KNOW WHAT TO MAKE OF THE CORRELATION BETWEEN NATIONAL INCOME AND IQ. NONE OF YOUR POINTS HERE REBUT THIS.”
Strawman. He wasn’t rebutting any general argument about the level of certainty, he was addressing a very specific claims Tyler Cowen made about the Mexican villagers. (The sense of humor and not being boring are consistent with low IQ). Let us simply note that Cowen conspicuously failed to respond to Malloy.
2.
“THE PROFESSIONAL CONSENSUS AS A WHOLE DOES NOT GO AGAINST MY POINTS.”
The professional consensus among whom? Anthropologists and sociologist or actual intelligence researchers? If you are referring to the later you are wrong. The professional consensus is that the low IQ levels measured among some population groups is *not* due to a lack of familiarity with “paper and pen” tests.
A multitude of other tests (reactions tests etc.) have shown the same result.
Let’s repeat Mallows question. Does Cowen honestly believe there is any test of abstract reasoning that the villagers will excel at?
3. “THE LITERATURE ON NEURODIVERSITY FINDS SIGNIFICANT DEVIATIONS FROM UNI-DIMENSIONALITY OF INTELLIGENCE. IT IS FAIR TO SAY THE MATTER IS NOT SETTLED BUT AGAIN THIS IS MORE ATTACK THAN SUBSTANCE.”
Of course intelligence is not *completely* uni-dimensional. Some smart people are relatively better at analytical thinking, language or visiospacial. But this is quite irrelevant, the point is that there is a
strong correlation between general intelligence and specific cognitive abilities.
We hardly need a perfect relation. Even if various mental functions are only to some degree a function of g different averages across groups lead to large differences in economic outcome.
The only argument would be asymmetry, that the Mexican villagers have some mental abilities that actually exceed Asians/Europeans. But Cowen does not offer even a hint of evidence for this. Instead he points to things that clearly are not intelligence (local knowledge), and to personality traits.
More importantly he fails to make a strong argument how the good traits of the villagers can be transformed into GDP. This is after all the debate. “IQ and the Wealth of Nations”, not “IQ and how much cosmopolitan professors enjoy their travels”.
4. “I SIMPLY THINK THAT IQ IS FAR MORE ENDOGENOUS THAN DOES THE AUTHOR.”
Maybe, I am certainly open to this. But where is the evidence? The Mexican village has nothing to do with this. The Flynn effect is NOT about changed relative positions, just a higher average for all.
The Bahrain/Qatar/Saudi example offers some natural experiment where wealth, nutrition and education levels skyrocket, whereas IQ remains the same.
5. “I AGREE THERE IS LOTS ABOUT IQ WE DON'T KNOW. MANY OF THE COMMENTATORS ON THIS POST ARE NOT ADVANCING THE DISCUSSION.”
Right… How could explaining the links between test-accuracy and outcome, describing the historical development or pointing to fallacies possibly be “advancing the discussion”?
Malloy and Sailor are not ‘advancing the discussion’, since they are refuting Tyler Cowens ideological positions with facts and logic.
Posted by: Tino at Jul 19, 2007 5:51:05 AM
Tyler and Darwin are much more humble and therefore open to the likelihood of competing theories.
Just weird.
Posted by: Kismet at Jul 19, 2007 7:39:10 AM
This discussion would be more interesting if the IQ test was renamed a "Wealth Predictor" test.
IQ is actually a better predictor of income than wealth (on an individual, not country level). I guess smart people like spending lots of money or something. Also, a lot of smart people seek the admiration of other smart people within academia or open-source programming and other such things. As I argued at Caplan's post about nerds vs jocks, a lot of jock qualities actually help to earn high incomes (this is discussed in Tom Wolfe's books). The IQ test is not designed to be a wealth predictor test and does not work as well for predicting wealth as it does for predicting lots of other stuff.
Posted by: TGGP at Jul 19, 2007 8:31:04 AM
As an uninformed observer, I'm glad to see Prof. Cowen responding to comments. Thanks for your time, Tyler.
Posted by: josh at Jul 19, 2007 8:54:32 AM
“The IQ test is not designed to be a wealth predictor test and does not work as well for predicting wealth as it does for predicting lots of other stuff.”
Wrong. IQ predicts wealth pretty well, especially taking into account the huge measurement issue (value of human capital, small firms etc.), and the fact that you tend to measure at one point of time, not across the life. Smart people earn much more, and obviously tend to consume most of this over their lifetime. Just accumulating wealth would be pretty stupid.
The myth that IQ does not predict wealth comes from studies that control for factors such as education (largely function of IQ) and race.
http://www.iapsych.com/articles/zagorsky2007ip.pdf
Median income young baby boomers by IQ test score
<75 $15,020
85 $27,700
100 $36,826
115 $45,675
>125 $55,555
Median net worth of young baby boomers by IQ test score
<75 $5775
85 $24,250
100 $57,550
115 $94,500
>125 $133,250
Pretty big difference if you ask me, between 130.000$ and 6.000$. But hey, a clever researcher can make any relation disappear by controlling for enough factors. After all, isn’t that the job of scholars? To, hide, ignore and dismiss unwelcome facts?
Posted by: Tino at Jul 19, 2007 10:39:21 AM
You'll have to pardon me leaping in just to go after two posts from yesterday afternoon.
I think that Marc's comment (5:35) ignores an alternative, and equally compelling explanation about IQ differences: social norms, group behaviour, whatever you care to call it. I've heard the same study about cross-race adoption used as an example of American blacks succumbing to pressure about "acting white," an aregument which would be initially mitigated by growing up in a white household and reinforced by the decrease in scores when American blacks are reminded of racial differences prior to taking cognitive tests.
Tino's comment at 6:50, though, sounds off-base. There have been no major population movements which would account for the dilution (catastrophic wipe-out sounds more like it) of the above-average populations which catalyzed civilisation in Egypt and Mesopotamia on the scale which IQ&WN demands. Mesopotamia is a weaker argument, since cities were sometimes destroyed en masse and mass migrations occurred (but only constituted a large minority in the best of times, demographically speaking).
Egypt, however, hasn't had a population shift since the Sea Peoples attacked. Throughout its history, the population of the region has never significantly changed. Unless you argue along the lines that Black Athena is an understatement, that the Book of Exodus represents the literary retelling of utterly massive displacement of the original population, or that the Arab r