Created: 2000-03-05
Last Modified: 2000-03-05
Go to
Brad De Long's Home Page

Teaching | Writing | Career | Politics | Book Reviews | Information Economy | Economists | Multimedia | Students | Fine Print | Other | My Jobs

The IQ Debate

J. Bradford DeLong



Juicy details, please.

My Comment:

Not many juicy ones, I'm afraid. It was at the time of the [Bell Curve's] launch. The AEI had a lunch for the press and others. I used my status as Deputy Assistant Secretary of the Treasury to get a ticket.

If I recall correctly (but a bunch of different meetings are jumbled together in my mind), during the question period I asked (a) if it was not true that differences in measured IQ accounted for a negligible proportion of income inequality once one controlled for a multi-dimensional measure of parents' socio-economic-status and for education, and (b) if it was not true that the only way Herrnstein and Murray could even claim that measured IQ was an important cause of anything was by removing the variables--education, more than one dimension of parents' ses--that everyone else thought were the important causes.

I went on to assert (c) that Herrnstein and Murray's claim that once you control for IQ that there is no sign of wage discrimination against African-Americans was a blatant lie (they had a regression in which predicted black and white wages were equal when the right-hand-side variables were set at their means among whites, but in which predicted black wages were... somewhere between 10 and 20% lower... when the right-hand-side variables were set at their means among blacks). And then I asserted (d) that since Murray was maintaining (i) that measured IQ was determined by genes--barely affected by differences in children's environments--and (ii) that measured IQ in the U.S.--in spite of a "genetic endowment" that from Murray's perspective had been "deteriorating"--had been growing at 0.3 points per year for the past-half century, the only conclusion I could draw was that Murray was certainly not one of the cognitive elite.

At that point I was told that my participation was not wanted...


Brad DeLong


> :::scratching head::: I can't help wondering how much of the no-connection
>between intelligence and race position is just an emotional reaction to
>former ideas of strict links between them. The fact that these last ones
>have been disproved does not make those first ones proved.

My Comment:

There have only been 2000 or so generations since homo sapiens sapiens headed north and south out of East Africa or the Middle East or wherever we finally evolved.

Suppose that two diverging subpopulations headed into two different regions, one of which gave no survival advantage to the amount of higher intelligence conveyed by a (dominant) gene let's call [High IQ} and the other of which headed into a region in which having [High IQ] gave you an extra 1 in a thousand chance of surviving to breed...

Suppose further that both subpopulations started with a population frequency of [High IQ] of 1/20 percent--so that 0.1 percent of each population started with a [High IQ gene].

Then after 2000 generations, the proportion of the neutral selection pressure population with [High IQ] is still 0.1%, and the proportion of of the high-intelligence selection pressure population is... 0.73743%

We are a very young--and interbreeding--subspecies. 2000 generations--40,000 years--is not very long for natural selection to work without truly enormous amounts of differential selection pressure on different non-interbreeding populations.

In America there is a lot of intellectual garbage lying around--leftovers from arguments (back before 1860) that slavery was an institution designed to make the simple, childlike African-Americans better off, arguments (from 1880-1940 or so) that segregation was an institution designed to make the simple, childlike African-Americans better off, arguments (from 1900-1960 or so) that it was a waste to spend money on schools for African-Americans because everyone knew they weren't as smart, and arguments (from 1970-today) that large gaps in wealth and economic welfare between European-Americans and African-Americans reflect not the legacy of past discrimination but instead the lower native intelligence of African-Americans.

For example, you can go look at _The Bell Curve_, and find Charles Murray arguing four things:

--intelligence test scores are hardly affected at all by differences in the environments in which people are raised.
--the average intelligence test score of Americans has been rising for virtually the entire past century at about 0.3 IQ points per year.
--the (large: 15 IQ point) difference between immigrants from Poland and Italy and native-born white Americans around 1900 was due to differences in their environments.
--the (large 15 IQ point) difference between African-Americans and white Americans today is due to differences in their genes.

Now if intelligence test scores are hardly affected at all by differences in environments, then Polish-Americans and Italian-Americans have no business closing the average test score gap vis-a-vis native-born white Americans. If intelligence test scores are hardly affected at all by differences in environments, then there is no way that intelligence scores can have risen by perhaps 30 IQ points over the past century.

Conversely, if intelligence test scores have been trending upward--as they have--and if the grandchildren of immigrants from Italy and Poland have the same test scores as the grandchildren of WASPS--as they do--then no one has any business at all saying that intelligence test scores are hardly affected by differences in environments.

And if test scores are affected by differences in environments--and if other big gaps between ethnic group average scores have gone away as environments have become equalized--then no one has any business at all presuming that today's differentials are anything other than environmental.

The conclusion I reach from all of this is that Murray and Herrnstein are certainly not members of the... shall we say cognitive elite? And that anyone who wants to leap from the fact of large average test score differences today to the presumption that they have some genetic basis--that their reasoning has roots not in evidence but in American racism.

(Unless, of course, they also want to leap to the conclusion that grandchildren of Italian and Polish immigrants to the United States also have a genetic intelligence deficit--or to the conclusion that grandchildren of such immigrants would have a genetic intelligence deficit had aliens not been using effector technology to monkey with their genes at the fertilized ovum stage: then I would call them not racist but consistent--and insane.)

Brad DeLong

Sign up for Brad Delong's (general) mailing list

Go to related links...

Add a comment on this page...

Read other people's comments on this webpage

Professor of Economics J. Bradford DeLong, 601 Evans Hall, #3880
University of California at Berkeley
Berkeley, CA 94720-3880
(510) 643-4027 phone (510) 642-6615 fax

This document:

Search This Website