Book Review: The Bell Curve: Intelligence and Class Structure in American Life. By Richard J. Herrnstein & Charles Murray. Pp. 845. (New York : The Free Press, 1994). £25.

This is a large, moderately priced, and well-written book, expounding the view that cognitive ability is what is principally responsible for success in life. By “Cognitive Ability” is meant General Intelligence, that is to say Spearman’s g factor not analysed into particular factors as could be done. The book is open to criticism for its failure to take this into account, true though it may be that what is measured by ordinary I.Q. testing as general intelligence is probably what usually matters. It is taken for granted, although not really argued, that some substantial part of human intelligence is genetically inherited and so determined at conception. But of course some of the cognitive ability as finally realized is affected by environmental factors operating after birth, such as social and economic conditions and nutrition etc., and especially by education and upbringing. And it is this that is also responsible for what is finally achieved, and not simply genetical inheritance.

There is plenty of evidence, which can scarcely be contested, that human populations show a good deal of variance in intelligence as revealed by I.Q. scores, and that this is normally distributed (the “Bell Curve” of the book’s title), with a Standard Deviation (SD) of about 15 I.Q. points. This means that someone whose I.Q. is 115 is in the 84th percentile, in the top 16% of the population and so quite bright and suited for university education, and with an IQ of 130, two SD’s from the mean and at the 98th percentile is very bright indeed. But about 70% of the population, who are within one SD either side of the mean, although not especially intelligent, are solid hard-working citizens well able to support themselves in a modern society. With an I.Q. below 85, at the 16th percentile, people are beginning to be a bit dull although still able to earn their keep in humble but nevertheless useful occupations, whereas below I.Q. 70 at the 2nd percentile they are beginning to be mentally deficient and may have to be supported on welfare by the rest of the community.

The book is in two sections, the first being concerned almost exclusively with white men in the U.S.A. (to avoid complicating the calculations women are mostly excluded, although the implication is that the same conclusions would apply to them). But so far as the men are concerned there is evidence, illustrated at length with graphs and tables taken mostly from official and semi-official sources, that the “cognitive elite” do the best in almost every way. They not only nowadays often have a university education, giving them entry into the best and most highly paid occupations, but all the way down in non-professional employment, to even the lowest paid jobs, those with the highest I.Q.’s are always near the top, both financially and otherwise. Furthermore, since nowadays most people can make the best of the cognitive abilities that they do possess, by for example getting a good education at public expense whether or not they can afford it themselves, a stratified caste system is being evolved. The highly intelligent, who get the best jobs, only meet and marry others like themselves, and have little contact with anyone lower down the socio-economic scale, let alone with the unintelligent underclass they have to support by taxation. And at all levels those whose cognitive abilities are above, or below, their social and economic position can gain promotion, or less easily demotion, so that cognitive abilities get sorted out by social class. That was not so a few generations ago, when there were always quite a lot of intelligent people lower down the social scale who would have great difficulty in bettering themselves, and of course some stupid ones at the top.

Since it is supposed to be a land of opportunity, where everyone should rise as high as they can regardless of social origins, this would be an unexpected and unwanted result of the social mobility they rightly pride themselves on in the U.S.A. But it would be so only if some significant part of cognitive ability was genetically heritable. Where environmental factors are also important, as they probably are, there will always be some people lower down who have difficulty in doing as well as they think they should, and will therefore be discontented and may turn to crime. The problem is that low cognitive ability has a self-fulfilling effect on the social environment: unintelligent people provide bad conditions for themselves and their families, so that even the more intelligent of their offspring are relatively disadvantaged from the start.

These conclusions are far from being politically correct, but there is much worse to follow in the second section, concerned with ethnic differences in relation to I.Q. and to social inequality. There now seems little doubt that African-Americans appear on average to have significantly lower cognitive ability than do the European-Americans, with a mean I.Q. score of about 85, more than one SD below the whites. East Asian immigrants, Chinese, Japanese, and Vietnamese etc., on the other hand, are doing rather better than the whites, being on average perhaps marginally below in abstract reasoning but well ahead on tests for spatial relations. The “Latinos” (Mexicans, Puerto Ricans, etc.) are between the rest of the whites and the blacks, as are the native American Indians.

Many of the blacks live in bad social conditions, with welfare dependency, illegitimacy, and crime all too common, which seems to be associated with low cognitive ability, since this underclass includes many relatively dull whites and Latinos as well, with I.Q.’s below 85. And of course there are plenty of intelligent blacks whose I.Q.’s are above the white average, and who are doing very well for themselves. These may form a relatively small minority of the black population, but there are still hundreds of thousands if not millions of them. The East Asians, whose cognitive ability is usually high, often prosper, even if they have only recently arrived as refugees.

But some of the blacks certainly do constitute a problem, and the question is whether their low cognitive ability is due to the bad conditions of their lives in the underclass, or whether the bad conditions result from low cognitive ability. A bit of both, probably, but if their upbringing is to any significant extent responsible for their poor performance in I.Q. tests, it should be possible to correct this, by better welfare and education and so on. And later on in life by “affirmative action”, with quotas for various forms of employment as now often required by law, and for entry to higher education. This means that black people are to be given preference over whites whose abilities may appear to be the same or better, on the assumption that appearances are wrong, and that given the chance the blacks should do as well as the whites.

It has to be admitted that affirmative action in the U.S.A. has so far had results which are not encouraging. The Bell Curve produces plenty of evidence, which no doubt can and will be contested, to show that where blacks have been promoted above whites with similar I.Q. scores their performance has often been less than satisfactory. For example, in universities where there is now a firm policy for affirmative action in admissions, the blacks have higher drop-out rates and seem to do significantly less well in the final examinations. For these reasons Murray & Herrnstein are not in favour of affirmative action, principally because it represents a misuse of limited resources, but also because the whites, feeling that they have been rejected for reasons of colour, may become discontented - which doesn’t make for good race relations. Whereas blacks who find themselves unable to cope with the demands of the positions to which they have been promoted also become discontented, and if they fail they are likely to attribute this to racial prejudice rather than to their own incompetence. And of course blacks who do have the intelligence to do well, as many do, may be under-valued by those who feel that they have got to where they are by affirmative action rather than their own abilities. The Bell Curve therefore comes out strongly against affirmative action, especially where it is legally enforced, attributing its failure to the fact that lower cognitive ability, and therefore lower potential for success, is largely determined genetically and so is not to be corrected by social engineering after birth. And you can’t get much more politically incorrect than that!

The book is written in a persuasive style, supported by copious evidence, much of it no doubt selected to prove the point. But is what it says in fact correct? Needless to say, most of the book reviews, especially in the U.S.A., have been very unfavourable. But these tend to attack the good faith of the sources for its conclusions, without really showing why the conclusions must be wrong. For example, a review of over 4,500 words in the New York Review of Books (1 December 1994), entitled “The Tainted Sources of the Bell Curve”, is largely devoted to an attack on The Mankind Quarterly, which is a serious academic journal, and on some of its contributors supposed to be fascists who favour apartheid etc. And Leon Kamin, in the Scientific American (February 1995), takes much the same line. This may or may not be justified, but one would like to know exactly where their facts are supposed to be wrong.

There have of course been some sensibly critical reviews, notably by S. J. Gould in The New Yorker (28 November 1994), and by S. Blinkhorn in Nature (1 December 1994), although I have seen none that are in any way favourable. But the book is an important one which, having already sold 400,000 copies, must be regarded as forming a real reaction against the politically correct orthodoxy now prevalent in the U.S.A. and elsewhere. Its arguments can scarcely be refuted in short reviews, but a full-scale book in reply is said to be planned for later on in the year. In the meantime one can only say that The Bell Curve deals with some serious problems in a provocatively controversial fashion, and that it will repay serious critical attention, perhaps most of all by those who don’t like its conclusions.

C. B. Goodhart