Perhaps what aggravates me the most about the position of Andrew Sullivan is how he, and others who endorse the Bell Curve argument, express that endorsement by essentially lying about what the argument says. To read Sullivan, or the myriad people who have popped up in my comments, one might think that these arguments offer minor distinctions between the intelligence of the black population and the intelligence of the white population, that we're talking nickels and dimes here. This is flatly untrue. Both the Bell Curve and the larger suite of arguments about race and IQ that Sullivan and others are endorsing say that the black population is significantly less intelligent than the white population. The Bell Curve argues that the average white person has an IQ that is more than a standard deviation higher than the average black person. Since the publication of that book, Charles Murray and those like him have endorsed the view that sub-Saharan Africans have an average IQ better than two standard deviations lower than the average white American. (See, for example, the notorious Rushton-Jensen article, co-authored by the president of the explicitly racist and eugenicist Pioneer Society.) In other words, they believe that the difference in intelligence between the average white American and the average sub-Saharan African is the same as or larger than the difference in intelligence between the average sub-Saharan African and someone who suffers from Down Syndrome. These are not fine distinctions.
Sullivan wrote "No one is arguing that 'that black people are dumber than white,' just that the distribution of IQ is slightly different among different racial populations." If you take nothing else away from me, ever, take this: this is wrong. To say that a standard deviation of difference represents a "slight" difference is simply untrue. To say that the two full standard deviations separating sub-Saharan Africans from white Americans, asserted by the race science crowd, is merely a slightly different distribution is to engage in some truly mendacious wordplay, or to betray a lack of even elementary statistical education.
Does accepting these premises equate with arguing that black people are dumber than white people? I would suggest that it does. As I said in a previous post, I don't doubt that people are accurately reporting IQ data for different populations. What I doubt is that intelligence is a quantifiable phenomenon; that IQ is a meaningful proxy for it; that IQ tests are free of systematic bias and data corruption; and that these differences can be responsibly asserted to be the product of heredity and not environmental and other factors. But if you do accept these premises, I can't see how there is any meaningful way you can deny that statement. After all, the Bell Curve's central argument is precisely that intelligence is real, measurable, accurately quantified with IQ and IQ testing, largely heritable, and that black people have low IQs.
(How long ago did Sullivan read the Bell Curve? I have it sitting in front of me. The data is right there. I can't understand how Sullivan can believe that he's arguing the same thing as the book and yet still call these slight differences in distribution.)
This divide, between the pride with which people assert their independence and honesty on this issue, and the way in which they relate the arguments of race science in the most anodyne and minimized way-- that's what bothers me the most. It's the hypocrisy in patting yourself on the back for facing "harsh truths" and then failing to accurately reflect what those "truths" you're endorsing actually say.
In the post I linked to above, Coates talks about his community.
I have lived in the black community virtually my entire life. I went to black public schools. I went to a black university. I have spent a third of my life with a black woman. When I wake up in the morning, black people are the first thing I see. My black mother and father hurled books at me. My black Howard professors shot down my dumb theories. My black book editor parses through my long unwieldy thoughts. My black wife reads my first drafts. In very literal terms, what you read here everyday is representation of the collective brain-power of a black community.If I know the rhythms of blogging, this episode will soon draw to a close and everyone will part as friends. I must insist on pointing this out: if the argument of the Bell Curve and attendant views is correct, the majority of the people Coates has described here are likely of significantly below average intelligence. That's what the argument says, that a significant majority of black people have below average IQs and that these IQs accurately reflect their intelligence. Sure, some race science types might say that Coates is likely to run with an above-average crowd. But if you expand the ranks of people far enough, and statistics are true, and the Bell Curve is true, we're talking about a group of people-- the people this man loves and admires-- who are largely made up of the unintelligent. That's what the book says. And you, reading at home, the black people you know and work with and socialize with? The Bell Curve says that you can expect a significant majority of them to be significantly below average intelligence. That's the text of the book. It's right there, in black and white.
Posts
13 comments:
I currently teach in sub-Saharan Africa and my students and fellow faculty are considerably more intelligent than those I encountered working in Asia. I believe Asians are supposed to be even smarter than White People according to the Bell Curve. After living in Africa for a year now, I am convinced that most of what most people in the US think they know about Africa is just plain wrong. That goes doubly for the idea that Africa's problems stem from a lack of human capital. The raw material in terms of cognitive ability of the students at the University of Ghana is as high as any university in the US. I would put my best students up against those from Harvard, Yale, and Stanford any day of the week.
J. Otto Pohl - I suspect you were intending to help, but generalizing about the intelligence of your students is just as cringe-inducing.
Also, I never thought of lack of human capital as a statement about innate ability, but rather as a lack of those with the right education and training (for a whole host of complicated reasons).
Freddie,
I can't stand Sully, so I won't even begin to defend him, but in general, only half of what you say here is correct. Specifically, this:
"Both the Bell Curve and the larger suite of arguments about race and IQ that Sullivan and others are endorsing say that the black population is significantly less intelligent than the white population. The Bell Curve argues that the average white person has an IQ that is more than a standard deviation higher than the average black person. Since the publication of that book, Charles Murray and those like him have endorsed the view that sub-Saharan Africans have an average IQ better than two standard deviations lower than the average white American."
The question then becomes, what does this mean? Does it mean that "black people are dumber than white"? In a crude sense yes - but we also know that since average white IQ is close to one standard below an average Jew, we can say that "white people are dumber than Jews".
But you go off the rails when you claim that the people TNC "loves and admires" are "unintelligent". As you yourself admit, TNC could easily be hanging out on the right-hang side of the black IQ curve and furthermore most of us go through life loving and admiring some folks who aren't that intelligent (e.g. people in my family come to mind, like my mother's father who probably had a below-average IQ but worked hard and was a wonderful father/grandfather). Intelligence isn't wisdom, or love, or patience (this blogger links to a study that says "The ability to delay gratification at age four is twice as good a predictor of later SAT scores as IQ.) You get the idea.
Three considerable posts on this tired topic? Someone said it before but it's worth repeating. Of course well-heeled white Americans do better on a test that they created, what's surprising is that it's by such a small amount.
I wonder who's smarter, blacks or whites? To even posit the question is racist. Sullivan does it like most intelligent men before him, all in the name of science.
Hi Freddie,
Here's one thing I think about.
It is at least conceptually possible that the domain of people who are black -who self-identify as black- are, on average, less intelligent than whites.
If you were interested in finding evidence that this were true or false, you need to go out and get data.
So what data -not exactly, just in rough outline- would show that blacks, or some population P, were less intelligent than some population P'.
I don't think we have data that does show blacks are less intelligent, but I worry in the back of my mind that I am being too conservative in how I look at that evidence. That sort of conservativism is the mistake global warming skeptics make. You can get suck skeptics to realize their mistake by asking them what evidence that shows that there is anthropogenic global warming occurs would look like. The skeptics never have an answer, showing they are too picky in how they evaluate evidence for the theory they are biased against.
Are we doing that when we are skeptical about the conclusion that there is a partially racially determined difference in IQ? How do you know we're not?
I think we're not, but I have trouble answering this question. I hoped to hear your thoughts.
Sorry. I forgot the central premise of my problem: If you believe that there is no possible evidence that would count as evidence that X is true when the question of whether X is true is an empirical question, then you are being too skeptical about X. You are so skeptical about X that nothing could convince you, which would only be acceptable in cases where X was known to be false a priori. (Presumably, questions about racial intelligence should be settled empirically.)
I hadn't realized the numbers were so stark. While it seems extremely prudent and fair to more or less dismiss out of hand the arguments of traditional racial intelligence theorists such as Murray et al. for very good reasons -- the notions that race is a constructed social designation and American 'racial' groups aare in fact only social groups; that intelligence isn't actually any unified quantity in the human and that in any case it's absurd to suggest that it would be revealed by crude tests; or a view of genetics that takes a view of gene expression that is properly contingent on the effects of environment -- nevertheless, it would seem to me tha tthe data demand that an honest person claiming to attend to it give some account of what it reflects, if anything. I appreciate this post because it does give a sense of the number of conceptual assertions that are required to arrive at what appears to be the normative stance, which I think is what drives Sullivan's poorly executed attempts to take up this issue, namely to completely dismiss the data as without import. In my view, that is a hard place to arrive at in view of what the data say. IQ tests may not reveal anything about intelligence; intelligence may not be heritable through genes; intelligence may not even be a real quantity in humans. Additionally, the concept of racial groups may not be an analytically valid category for comparisons of IQ scores or any other measurement of human traits, either because they are more defined by socialization than by race per se, or because they are simply arbitrary and illegitimately essentialist, and so one could argue that it was never even legitimate to collect and compile this data in this way on the first place - that doing so was motivated by a scientifically invalid hypothesis form the start.
One can say all these things and more about this way of conceptualizing the meaning of this data, but nevertheless the data exist, and by Freddie's own account, they are fairly gobsmacking. Even if the results we see are produced entirely by a combination of a fundamentally culturally biased tool for assessing ability and the effects of culturally accumulated centuries worth of economic, cultural, intellectual, and physical oppression, it's still apparently the case that American self-identified African-Americans score much worse on IQ tests than whites. One can say anything one wants about this, but it does seem to me that, once you've taken up the topic, if one considers oneself any kind of a serious thinker, one ought to try to figure out what there is to rightly say about it, and say it.
What it seems to me one shouldn't do, if one considers oneself a serious thinker, would be to talk a lot about this matter but finally arrive at an endorsement of what seems to be the consensus view in the academy (admittedly, that assessment being in part based on the reporting of a few academic acquaintances of Sullivan himself) that, really, the way to talk about this data is not to talk about it, to pretend that it doesn't exist.
I don't consider myself a serious thinker, but my impression is that Freddie does consider himself one, or considers himself to be trying to be one. But I'm not clear whether, in these three posts in which he has certainly told us how we should not think about the data, Freddie has in fact told us how he thinks we should think about it, or at any rate how he does.
You say the data are striking, Freddie, and there are certainly many ways that they can be explained or brought into question. But is there one of or one combination of the possibilities that you think is the way to approach the question, Freddie? Do you have a theory of the case here? Or not? Should people be trying to develop a theory of the case regarding the data you cite, Freddie? Or should they not be?
Shorter me: I certainly don't feel for myself any need to try to think through how to think about these data (which are old now and probably should not be updated for the reason I suggest above - merely because we have older data that was arguably collected based on racist premises doesn't mean we should think we need to collect updated data using the same flawed premises. But it is the case that the data that were collected still exist.)
But neither do I feel the need to offer definitive accounts of why people who are attempting to interpret these data are doing so in wrong ways. I plead incompetence. But if one is saying there is a wrong way, then one is saying there is a way not to be wrong, even if it is adopting the most fundamental view, namely to say the move to interpret the data is wrong, because the concepts that were necessary for the very collection of the data were illegitimate, ill-defined, or invalid for such use. It seems to me you need to say something or other is the right way to approach the data and why, even if that is not approaching it, if you want to say that something else is the wrong way to approach it.
It seems to me what you have said in these posts has been this: 'This approach is wrong. It might be wrong because of this, or it might be wrong because of that, I'm not sure. But it's wrong.' I think most sensible people intuitively agree with you, but most people don't do it for serious-thinker kinds of reasons. If you want to claim to, I contend you need to figure out those reasons, and say what they are, and you also need to then - to the best of your ability as a non-expert but still-serious empirical thinker - say as much as you feel you can about what the right way to approach these data is that is implied by or at least consistent with your serious-thinker-typer reasons for saying that what has been going on is wrong. But perhaps you disagree that you need to do any of this in order to maintain a claim to be a serious thinker.
Even shorter me. Suppose we could definitively, absolutely show that genetic heredity has absolutely nothing to do with these data - that African Americans are conclusively, absolutely not inherently any less intelligent than white Americans as a statistical matter. Would that be the last and only thing you would want to say about the data showing lower group scores on certain tests for the one group compared to the other? Would there not be anything else that you'd like to aver about them by way of composing an overall analytic position with respect to these findings?
Andrew Sullivan is obviously a bigot, but I've never been impressed with the notions that we can't quantify intelligence or that it's too complex to be reduced to genetics.
Taken to the extreme, we would have to conclude that apparent differences in intelligence between humans and chimps cannot be ascribed to our respective genomes.
All things considered, it would be surprising if there were not some degree of inter-population variation in average intelligence, just as there is for phenotypic traits.
Interesting post over at Ddecline of the Empire noting that studies of human skulls have shown that the brain capacity of "primitive" man was significantly higher than modern man. So...many of our ideas about intelligence may be untrue and based on biases and prejudices.
what a lot of words to say nothing. nice touch ending with a total non sequitur - as they say on the internet, HERP DERP.
"That's what the argument says, that a significant majority of black people have below average IQs and that these IQs accurately reflect their intelligence."
a significant majority? AMG you are so fulla crap.
that is not true. the argument is that there is a statistically significant between group difference. this difference is MUCH SMALLER than within group differences, just like in gender.
Which is why black male TNC has an order of magnitude more cognitive ability than say....white male you. and i, white female person peg at least 2std above white male Andrew Sullivan.
;)
Post a Comment