Charles Murray’s Human Diversity contains little on the genetic basis of racial differences in average intelligence; it is clear Murray doesn’t want to be the subject of another moral panic like that which greeted The Bell Curve. He merely mentions that it is “tough” to defend the belief that “ethnic differences in IQ are meaningless,” (206) and explains why in a long endnote (416–18). Here he points out that attributing the Black-White gap in America (or other Western countries) to “racism” predicts that Blacks would score higher in all-Black countries; in fact, scores are uniformly lower in Black Africa and Haiti than among for American Blacks. If you then appeal to “the legacy of colonial racism” (416), you must explain how colonialism affects IQ. The most plausible suggestion is through parental SES. To test this hypothesis, one must adjust scores for parental SES. Murray notes that “this has been done frequently,” and the literature “consistently shows that doing so diminishes the size of the B/W difference by about a third.” In other words, two-thirds of the gap cannot be accounted for in this way. Moreover, most studies indicate that the B/W difference increases as parental SES rises; in other words, higher parental SES is associated with a rise in Black IQ, but with an even bigger rise in White IQ.
Other explanations offered for the race gap include Blacks’ relative unfamiliarity with standard English, the administering of tests by White rather than by Black teachers, or Blacks’ lack of motivation to work hard on tests which “clearly reflect White values.” In response, Murray cites the consensus statement “Intelligence: Knowns and Unknowns” published in the February 1996 issue of the APA flagship journal American Psychologist in response to the public controversy surrounding The Bell Curve. The eleven experts reported that controlled experiments have revealed no substantial contribution to the racial gap from any of these causes (although they may play a role in particular cases). The statement also notes the high predictive value of tests for academic performance.
Since that statement was issued, experimental evidence has been produced indicating that the racial gap is “effectively eliminated” when Black and White students are “tested on the basis of newly learned information.” The difficulty here is that such tests inherently measure short-term memory as much as, or more than, IQ; the gap disappears because the test is no longer so g loaded.
Murray acknowledges the possibility of arguing that the role of bias is too broad to be captured by any assessment of language and predictive ability. He calls this “the ‘background radiation’ theory of racism’s effect on IQ.” (418) This perspective would make “racism” an occult but omnipresent reality not unlike what Africans call “bad juju.” As Murray says, such a perspective cannot be refuted with data, since it conceives all data as vitiated a priori by racism; in other words, it is an unfalsifiable metaphysical commitment.