Review published in Personality & Individual Differences 26, 767-774, April 1999

A review of:
Robert J. Sternberg & Elena Grigorenko (eds.) (1997). Intelligence, Heredity and Environment. Cambridge, UK : Cambridge University Press.

For the past decade, the only serious worry for the London School has been what might happen if critics of g and h² ever actually read the writings of Hans Eysenck, Art Jensen, Tom Bouchard and Sandra Scarr. From the mid-1980's, new twin and adoption studies were delivering impressive evidence of g's substantial heritability. Even Sir Cyril Burt was being vindicated. Yet could there possibly be some flaw?
       It became clear in 1987 from Intelligence and Education (Oxford : Clarendon) that such eminent critics of IQ as Maurice Schiff and Richard Lewontin had simply not read what the London School had to say. These critics did not realize that Eysenck and Jensen made precisely the allowance of IQ being 10-15% environmental for which Schiff's own adoption study provided further evidence. So what would happen if they and such tenured psychologists as Stephen Ceci, Howard Gardner and Douglas Wahlsten actually examined the hereditarian case? Might they actually find errors that had escaped attention because of their previous preference for dissociation?
      In the present volume, the feat of inducing reading has been achieved. The book supplies within the same covers the views of leading hereditarians, the replies of leading sceptics about g, and an amusing summing-up by Earl Hunt. Although Douglas Dorfman and Leon Kamin and the important non-psychologists Stephen J. Gould and Steven Rose do not appear for the sceptics, readers can have every reasonable assurance that academic opposition to London School views is unlikely to get stronger than the arguments advanced in these pages. This is an important publishing achievement, which is appropriately topped by a sensational revelation of how social-environmentalists have begun to shift their ground.
      Sadly, the hereditarians -- who bat first -- are not really on their best form. They have done it all so often before; but no-one reads it, so they become jaded. Sandra Scarr contents herself with old (though good) data; Art Jensen declines to use the breakthrough study by Phillipps (1993) showing that MZ twins, because of sharing the same placenta, actually draw apart from each other as one of them wins the competition for maternal blood supply; and the Colorado team is, as ever, so mesmerized by psychogenetic formulae that it will persuade precisely no-one. Fortunately, Tom Bouchard, does an excellent job of exposing the vacuity of latter-day opposition to heritability estimates. In particular, he refutes the claim that MZ twins become similar because they are treated according to their appearance. (Appearance has precisely no correlation with IQ.) John Loehlin and his Texan co-workers furnish the interesting statistic that adoptees reared together actually correlate negatively for IQ (at -.09, compared to +.24 in biological siblings).
      By contrast, however, the sceptics are a positive shambles. Their favourite gambits are as follows: nature and nurture cannot be separated; Bouchard's identical twins were only separated for twenty years; "the deprivation experiment can tell us nothing about the role of genes because it varies only experience"; genes have their effect in interaction with the environment; the h² for IQ might be different in different environmental 'reaction-ranges'; the h² might be different if new environments realized more of a person's 'genetic potential'; everything is necessarily cultural. These mouthings are variously platitudinous, incoherent, unfalsifiable, tired (seldom advancing beyond Herrman and Hogben's (1932/3) speculative reservations) or even well in line with what hereditarians themselves have demonstrated (as to how a child's genes help create its environment). Still more striking is the complete failure of the sceptics to explain any percentage of IQ variance whatsoever. Instead of seeing themselves as offering a competing, social-environmentalist theory that can handle the data, or some fraction of it, the sceptics simply have nothing to propose of any systematic kind. Instead, their point or hope is merely that everything might be so complex and inextricable and fast-changing that science will never grasp it. Hunt finally summarizes matters: the sceptics have humanitarian and hypothetical reservations about the hereditarian case, but they have no alternative to it. As Hunt says, "If we insist on treating genetic and cultural explanations of intelligence as a stomping match, then the behavioral geneticists are the stompers and the proponents of cultural effects are the stompees."
      Of course, cognoscenti will find some valid points being made by the sceptics. Naturally, science can only talk of the heritability of IQ across the range of environments that is studied -- undoubtedly somewhat restricted when all homes have been selected as suitable by adoption agencies. (At the same time, modern twin studies often involve a less than full population range of IQs because low-IQ twin volunteers are hard to find. This effect means that h² is certainly under-estimated.) But the big surprise comes as social-environmentalist pack leader Stephen Ceci develops his (entirely non-predictive) 'bio-ecological model.' Quite simply, while surrounding his concession with waffle and bluster, Ceci acknowledges (p. 303) "the important role that genetics plays in intellectual development." This is a far cry from Leon Kamin's (1974, p. 1) proposition: "There exist no data which should lead a prudent man to accept the hypothesis that I.Q. test scores are in any degree heritable." Plainly, within academia, the nature-nurture debate about IQ variation is over bar the shouting. Now the hereditarian victory must be translated into public acceptance and into liberal forms of educational and eugenic advance.

CHRIS BRAND

Edinburgh, ii 1998

References

HERRMAN, L. & HOGBEN, L., 1932/3, 'The intellectual resemblance of twins.' Proceedings of the Royal Society of Edinburgh 53, 105-129.

KAMIN, L. J. (1974). The Science and Politics of IQ. Potomac, MY : Lawrence Erlbaum Associates.

PHILLIPS, D. I. W. (1993). 'Twin studies in medical research: can they tell us whether diseases are genetically determined?' Lancet 341, 8851, 17 iv, 1008-1009.

Review published in Personality & Individual Differences 26, 767-774, April 1999

For more on the psychology and politics of human individual and group differences, including the weekly William McDougall NewsLetter, go to <http://www.crispian.demon.co.uk>.