Book Reviews The g Factor: The Science of Mental Ability Arthur R. Jensen Westport, CT: Greenwood, 1998, 700pp. US$39.95 cloth. ISBN 0-2759-6103-6. Greenwood Publishing, 88 Post Rd. West, Box 5007, Westport, CT 06881, USA. J. Philippe Rushton University of Western Ontario, Canada Few scientists have effects or laws named after them. Arthur lates of g (excluding race), its heritability, and its practical Jensen's name is listed in a number of dictionaries as an predictive power. The fact that psychometric g has many "ism!" The Random House and Webster's Unabridged Dic- physical correlates proves that it is not just a methodological tionaries contain the following entry: artifact. Among biological variables, g loads on heritability coefficients determined from twin studies and inbreeding Jen-sen-ism (jen'se niz'em), n. the theory that an depression scores calculated in children born from cousin- individual's IQ is largely due to heredity, including marriages. g is also related to brain size measured by Mag- racial heritage. [1965-1970]; after Arthur R. Jensen netic Resonance Imaging (MRI), brain evoked potentials, (born 1923), U.S. educational psychologist, who pro- and intracellular brain pH levels. It (g) is a product of human posed such a theory; see -ism]—Jen'sen-ist, Jen'sen- evolution and is also found in nonhuman animals. ite', n., adj. Despite these caveats, The Bell Curve affair allows one to safely predict that The g Factor's coverage of race will The "theory" attributed to Jensen has, in fact, been around strike many as of central importance. All the issues Jensen since the time of Francis Galton (1822-1911), whose He- raised in 1969 are still with us today. Indeed, much of the reditary Genius (1869) predated by exactly one century opposition to IQ testing and heritability would probably Jensen's famous Harvard Educational Review article that disappear if it were not for the stubborn and unwelcome fact led him to be labeled a "hereditarian." The dictionary defi- that, despite extensive well-funded programs of interven- nition cannot be overly derided, however, as Jensen's (1969) tion, the Black-White difference refuses to go quietly into review of the evidence that IQ is heritable and that genetic the night. Chapter 11 of The g Factor fully documents that, factors are involved in the Black-White IQ gap had enor- on average, the American Black population scores below the mous impact. White population by about 1.2 standard deviations, equiva- Jensenism, one of the great heresies of twentieth-century lent to 18 IQ points. (This magnitude of difference gives a science, is partly responsible for getting the Darwinian-Gal- median overlap of less than 15%, meaning that less than 15% tonian paradigm back on track in differential psychology of the Black population exceeds the White average of 50%). after it had been derailed in the behavioral sciences for at The difference between Blacks and Whites in average IQ least a generation following World War II. In a brilliant scores has scarcely changed over the past 80 years (despite 40-year career that has earned him a place among the most some claims that the gap is narrowing) and can be observed frequently cited figures in contemporary psychology, Arthur as early as three years of age. Controlling for overall socio- Jensen has systematically researched and extended Charles economic level only reduces the mean difference by 4 IQ Spearman's (1927) seminal concept of g, the general factor points. Culture-fair tests tend to give Blacks slightly lower of intelligence. The g Factor is an awesome and monumental scores, on the average, than more conventional tests, as do exposition of the case for the reality of g. It does not draw nonverbal tests compared with verbal tests, and abstract back from its most controversial conclusions—that the av- reasoning tests compared with tests of acquired knowledge. erage differences in IQ found between Blacks and Whites On average, Blacks also score 1 standard deviation below has a substantial hereditary component, and that this differ- Whites in academic achievement throughout the period from ence has important societal consequences. grades 1 through 12 (and also considerably below all other However, The g Factor is not about race, as such. The disadvantaged minorities tested—Puerto Rican, Mexican- first five chapters deal with the intellectual history of the American, and American Indian). discovery of g and various models of how to conceptualize intelligence. Other chapters deal with the biological cone- International IQ Distribution J. Philippe Rushton is Professor of Psychology at the University of Western Ontario, London, Ontario N6A 5C2, Canada (E-mail: Inspired by "Jensenism," researchers like Richard Lynn and [email protected]). His most recent book is Race, Evolution, Philip E. Vernon not only pushed the envelope, but extended and Behavior (1995, Transaction). the "outside of the envelope" and made the race-IQ debate 230 Politics and the Life Sciences September 1998 Book Reviews international in scope with their findings that East Asians although the nearly 1 standard deviation difference between average higher on tests of mental ability than do Whites, Whites and East Indians showed no correlation between g whereas Caribbeans ( and especially Africans) average loadings and standardized mean differences, the 2 standard lower. East Asians, measured in North America and in deviation difference between Whites and Blacks showed a Pacific Rim countries, typically average IQs in the range of correlation of +.62. 101 to 111. Caucasoid populations in North America, Spearmans hypothesis even applies to the g factor ex- Europe, and Australasia typically have average IQs from 85 tracted from performance on elementary cognitive tasks. In to 115 with an overall mean of 100. African populations some of these studies, 9- to 12-year-olds are asked to decide living south of the Sahara, in North America, in the Carib- which of several lights is illuminated and move their hand bean, and in Britain typically have mean IQs from 70 to 90. to press a button that turns that light off. All children can (Blacks in sub-Saharan Africa score about 2 standard devia- perform the tasks in less than one second, but children with tions [approximately 30 IQ points] below the mean of higher IQ scores perform faster than do those with lower Whites on nonverbal tests.) scores, and White children, on average, perform faster than Black children. The correlations between the g loadings of these types of reaction-time tasks and the Black-White dif- Spearman's Hypothesis ferences range from +.70 to +.81. Jensen also applied Spearmans hypothesis to East Asian- But the 18-point IQ difference between American Blacks White comparisons using these same reaction-time meas- and Whites is only an average. On some subtests the Black- ures. The direction of the correlation is opposite to that in White difference is smaller and on other subtests the Black- the Black-White studies, indicating that, on average, East White difference is larger. Black-White differences are Asians score higher in g than do Whites. No one so far seems markedly smaller on tests of rote learning and short-term to have looked at East Asian-White differences on conven- memory than on tests of reasoning and those requiring tional psychometric tests as a function of their g loadings. transformation of the input. For example, on the Forward From the study just mentioned, however, Jensens prediction Digit Span Test, in which people are asked to recall a series is clear: One should fmd the reverse of Spearmans hypothe- of digits in the same order as that in which they were sis for Black-White differences. presented, Black-White differences are quite small, but on the Backward Digit Span Test, in which people recall a series of digits in the reverse order to that in which they were Are Race Differences Heritable? presented, they are quite large. One day, while rereading Spearmans (1927) The Abilities of Man, Jensen tells us that Chapter 12 presents Jensens technical arguments for why he noted the suggestion (which appears on page 379) that he believes that race differences are about 50% heritable. He Black-White differences on various tests are a function of emphasizes the fact that it is precisely those components of each tests g loading. Here, Jensen thought, was the essential intelligence tests that are most heritable and that most relate phenomenon that would explain, in much broader, more to brain size which most profoundly differentiate Blacks fundamental terms, the specific psychometric phenomenon from Whites. Thus, Black-White differences on 11 subtests that gave rise to the variation in the Black-White average of the Wechsler Intelligence Scale for Children are predicted differences. by the amount of inbreeding depression on the same 11 The g Factor summarizes the results of numerous inves- subtest scores from Japan (r = +.48). The inbreeding predic- tigations of Spearman's hypothesis on a wide variety of tion was a sufficiently robust predictor to overcome gener- psychometric tests administered to large representative sam- alization from the Japanese in Japan to Blacks and Whites ples of Whites and Blacks. Chapter 11, for example, de- in the United States. There really is no non-genetic explana- scribes the results from 17 independent data sets on a total tion for the inbreeding effect and its ability to predict Black- of nearly 45,000 Blacks and 245,000 Whites derived from White differences in scores on IQ tests. 171 psychometric tests. g loadings consistently predict the The g Factor also cites the evidence of transracial adop- magnitude of the Black-White difference (r = +.63). Spear- tion studies. Three studies have been carried out of Korean man's hypothesis is borne out even among three-year-olds and Vietnamese children adopted into White American and administered eight subtests of the Stanford-Binet.