What We Mean When We Say ‘Mean Congenital Race Differences’

Ta-Nehisi Coates writes:

What We Mean When We Say ‘Race Is a Social Construct’

Our notion of what constitutes “white” and what constitutes “black” is a product of social context. It is utterly impossible to look at the delineation of a “Southern race” and not see the Civil War, the creation of an “Irish race” and not think of Cromwell’s ethnic cleansing, the creation of a “Jewish race” and not see anti-Semitism. There is no fixed sense of “whiteness” or “blackness,” not even today. It is quite common for whites to point out that Barack Obama isn’t really “black” but “half-white.” One wonders if they would say this if Barack Obama were a notorious drug-lord.

When the liberal says “race is a social construct,” he is not being a soft-headed dolt; he is speaking an historical truth. We do not go around testing the “Irish race” for intelligence or the “Southern race” for “hot-headedness.” These reasons are social. It is no more legitimate to ask “Is the black race dumber than then white race?” than it is to ask “Is the Jewish race thriftier than the Arab race?”

Why is it that “Race Is a Social Construct” is taken as a serious argument against the proposition that there are mean genetically conditioned differences between sub-populations called races?

Generally, why is social constructionism thought to be inconsistent with between group hereditarianism? Ta-Nehisi Coates writes: “It is no more legitimate to ask “Is the black race dumber than then white race?” than it is to ask “Is the Jewish race thriftier than the Arab race?”

Well, of course it is no less legitimate. This is why Eric Turkheimer, for example, opposes the study of race differences in IQ:

Why Race Science is Objectionable

If I may address my fellow Jews for a moment, consider this. How would you feel about a line of research into the question of whether Jews have a genetic tendency to be more concerned with money than other groups? Nothing anti-semitic, mind you, just a rational investigation of the scientific evidence. It wouldn’t be difficult to measure interest in money and materialism, and it wouldn’t surprise me if as an empirical matter Jews scored a little higher on the resulting test than other groups. As a behavioral geneticist I can assure you without reservation that the trait would be heritable, and, if anyone bothered to take the time to find out, specific genes would have small associations with it. Of course, this research program has already been carried out, at least to the extent the relevant technology was available in 1939. While we are at it we could open a whole scientific institute for the scientific study of racial stereotypes, and finally pull together the evidence on sneaky Japanese, drunken Irish, unintelligent Poles, overemotional women and lazy Italians. [Emphasis added.]

But the possibility of congenital Jewish-Arab differences in thriftiness does not make the possibility of congenital African-European American differences in IQ any more or less likely.

But this doesn’t address the problem with the thesis. It is fundamentally flawed. Social constructionism, per se, in no way implies non hereditarianism. This is #1 on my race fallacy list.

Were you to arbitrarily socially construct groups and then select out groups for which there were appreciable differences in some highly heritable trait, it would be more likely than not that the between group difference in that trait would be partially genetically conditioned. We can try creating such groups. In the case of color: Argentinians versus Colombians, North Hemispherains versus South Hemispherians; Theravada Buddhists versus Mahayana Buddhists, people with curly hair versus people with straight hair in the global context, rich versus poor Mexicans. Of course, in some instances the differences (between the pairs of arbitrarily socially constructed groups) would be completely unrelated to genetics. And whether the majority of the pairs would exhibit genetic differences would depend on the precise heritability estimate of the trait. But clearly, social constructionism, per se, is not inconsistent with between group hereditarianism. So when it comes to “race” what is the argument?

The argument against probable racial differences in some highly heritable trait must be grounded in the claim that race is not an arbitrarily constructed group. (Imagine if people decided their race in the way in which they decided their occupation. Maybe this is what is done! How do we know?) It must be claimed that there is something about the race delineation criteria which makes improbable genetic differences in the specific trait under question.

Generally, almost everyone on the other side has this backwards. The social construction of race tends to increase the number of possible avenues by which genetic differences can arise — and so increase the a priori probability that there are such differences, given within population differences and knowing nothing else. Take “Hispanics”, a very socially constructed group, as an example. “Hispanic” is a US construction. It refers to those individuals of Latin American Ancestry who migrated to the US and who continue to self identify as Hispanics. In terms of Human Capital, this is an unrepresentative group of all Latin Americans. And it is known that for this group there is negatively selected ethnic attrition. One plus one adds up to two if aptitude is non trivially heritable within the Latin American and US Hispanic populations.

So congenital race differences are questionable, to the extent that they are, not because race is an arbitrary social construct but because race is delineated in terms of something specific — in terms of geographic ancestry — and because the idea that the traits in question would congenitally vary by this variable — geographic ancestry — is questionable. Is it? Sure. This is why race differences are not assumed to be like social class differences — and how much more socially constructed can you get then social class — and so not assumed to be partially genetically conditioned. This nonsense that social constructionism implies non hereditarianism has to go.

When the liberal says “race is a social construct,” he is not being a soft-headed dolt; he is speaking an historical truth. We do not go around testing the “Irish race” for intelligence or the “Southern race” for “hot-headedness.” These reasons are social.

Let’s try out the magic reasoning:

Genius and mental retardation is social constructs, so … whala … mean IQ differences are only “social.”
Male and Female is a social construct, so … whala … mean morphological differences are only “social.”
Disease and Health is a social construct, so … whala …mean outcome differences are only “social.”

The Ta-Nehisi Coates who wrote this piece fits the empty Black suit stereotype. He also argues:

In a world where Kevin Garnett, Harold Ford, and Halle Berry all check “black” on the census, even the argument that racial labels refer to natural differences in physical traits doesn’t hold up….If you’re a math guy, perhaps your instinct is to point out the problems in the interpretation of the data. My instinct is to point out that your entire experiment proceeds from a basic flaw — no coherent, fixed definition of race actually exists.

Would he make the same argument against mean congenital differences in color? The real question is: Why is this logic so contextual? One thinks of Levin (2002)

In these terms, generalizations about race have clear meanings: To say the races differ genetically, for instance, is to say that humans whose ancestors were born in different parts of the world tend to have different genotypes. This hypothesis may be true or false and of course must be tested empirically, but it is perfectly well formed. Practical imperatives about race are similarly well formed: When advocates of affirmative action demand set-asides and numerical goals for Blacks, they have in mind individuals whose ancestors were brought to the United States from Africa. It would have been interesting to have seen Yee et al. (1993) criticize affirmative action on the grounds that “racial minority” is ill defined, and that, for the same reason, laws banning racial discrimination are unconstitutionally vague.

When I google scholar “achievement African American Black”, I get > 22K hits. Examples include:


Somehow individuals are being delineated by race. And no one seems to have a problem with it, except when discussion turns to mean genetic differences. Environmentalists have little problem operationalizing and researching race when it comes to their hypothesized environmental differences. Why, suddenly when it comes to hypothesized heritable differences? The hidden assumption seems to be that there is something unique about heritable differences. That researching or discussing genetic differences, unlike environmental differences, presupposes uniquely precise concepts. It doesn’t, of course. You can arbitrary socially construct any two groups measure various differences between them and then investigate to what extent the differences have a genetic basis. When it comes it race, it’s not as if no one investigates socially important genetic differences:


So the “problem” only exists when it comes to behavioral traits.

Really, the more interesting question is: Why do so many find these pseudo arguments so compelling?

This entry was posted in Uncategorized. Bookmark the permalink.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s