This would seem to be the promised reply to The Bell Curve, by Richard J. Herrnstein and Charles Murray, 1994 (reviewed in the Newsletter of June 1995), by what the blurb rather optimistically describes as "a group of the country’s most distinguished intellectuals [which] dismantles the alleged scientific foundations and criticizes the alarming public policy conclusions of this incendiary book".

In fact, it consists of 20 shortish chapters (the longest being 22 pages, and twelve of 10 pages or fewer), of which all but 6 are expanded versions of articles appearing in the New Republic on 31 October 1994. The 20 authors, who are all American, include one highly reputable evolutionary biologist (Gould) and six professors of sociology, law and history etc., the rest being mostly journalists.

These are all attacks upon The Bell Curve from a politically correct point of view. The most weighty, and all the more effective for being quite temperately expressed, is the article by Stephen J. Gould reprinted from The New Yorker of 28 November 1994, which criticises the supposed "general factor" for intelligence, "g", arrived at long ago by Spearman with his mathematical technique of Factor Analysis. This purports to show that, although "intelligence" does indeed comprise a number of quite different factors, for mathematical, logical, linguistic and other abilities, these are (with perhaps a few exceptions) correlated with each other, so that a single figure for an individual’s IQ score is indicative of something which really does exist, namely the general level of cognitive ability.

The argument that this is not so is a difficult and controversial one, developed at greater length by Gould, and others, in for example his book The Mismeasure of Man (1981, reprinted 1992). This may seem contrary to what is generally believed (which of course doesn’t mean it must be wrong), that some people are destined from the start to be more intelligent, and therefore more successful, than others. The Bell Curve however simply assumes that this must be so and also that some significant part of human intelligence, being genetically heritable, is unaffected by environmental factors operating after conception. And that this is taken for granted, rather than being positively argued for, is a weakness in the book which Professor Gould is well able to exploit. But the argument, in this article at any rate, is really only destructive, putting nothing in place of what it destroys. And anyway, it cannot be assumed that there really is no genetically inherited component in whatever it is that IQ tests purport to measure as general intelligence.

The Bell Curve is in two different parts, the first being concerned with the emergence of a "cognitive elite" of higher intelligence, and the effect of this on social behaviour, including welfare dependency, poverty, crtime and so on. This takes it almost for granted not only that there really are important differences in intelligence, but also that these are to some significant extent genetically heritable. That would be disputed by Gould, but there are in fact good reasons to think that this must be so, discussed in my own article in the Newsletter of March 1995.

The second and rather longer part of the book deals with what appear to be ethnic differences in cognitive ability, especially with regard to the "blacks" in the United States, and it is this that concerns almost all the other contributors to The Bell Curve Wars. There is no question that the mean IQ of people of African origin in America is at present about 15 points below those who come from Europe, with the "Latinos" of central America somewhere between the two, and immigrants from Asia being perhaps rather above the whites. But of course if the mean IQ score for the black population is one standard deviation below that for the whites, as it does seem to be, this still means that a substantial minority of blacks will be scoring better than the majority of whites. Nobody is suggesting that the blacks have some inherent deficiency in cognitive ability which puts them below all the whites, quite apart from the fact that a substantial part of the genetical inheritance of the "blacks" in America does anyway derive from the whites, and probably vice versa though to a considerably less extent.

It is also true that many black Americans nowadays live in conditions which are much less favourable than for most of the whites. And since the IQ scores actually achieved are known to be affected by environmental factors including early upbringing and education, that must account for at least part of the observed differences between the races. So the question is how much of this is due to the environment, and how much to genetical inheritance. Herrnstein & Murray believe that genetical inheritance has an important part to play here. But however persuasive their arguments may appear to be, supported by copious statistics, they are certainly not conclusive, and it remains quite possible that the whole of the difference may be due to the deprived conditions in which many of the blacks have to live.

This has important practical consequences, since if the apparently lower average intelligence of the blacks is indeed due wholly to the conditions of poverty under which many of them live, it ought to be possible to correct this in a single generation, by improved welfare for the depressed population, and by "affirmative action" and "head start" policies in education, and positive discrimination in employment. That this has already been tried with rather disappointing results, is one of the main arguments against the differences being wholly environmental in origin: but that cannot be regarded as conclusive.

If however significant genetically heritable differences really do exist at present, it would take many generations of selective breeding to produce any real changes. This would mean that many of the blacks, together with the less intelligent whites and latinos, would indeed form a "cognitive underclass", dependent upon the rest of the community for its support. Herrnstein and Murray think that this may already be happening in America, for which they are vigorously attacked by most of the contributors to The Bell Curve Wars who take the line that, whether or not this conclusion can be supported by objective statistics, it is morally and politically unacceptable. That may well be so, but it would be better to examine critically the evidence upon which Herrnstein and Murray rely, rather than simply to reject their conclusions. Otherwise it may look like conceding that their statistical evidence is unarguable, which of course is not the case.

Whether you like it or not, The Bell Curve represents a serious study of important matters, and if its conclusions are to be refuted this will have to be done by an equally serious study of all the evidence, including where appropriate the collection of new statistics. That will require a substantial book by a single author, or at the most two or three, properly equipped with the necessary statistical and scientific expertise. The present collection of short pieces of varying quality, many of them little more than journalism, really won’t do at all. The Bell Curve, however disturbing ("incendiary" is hardly the word), deserved a serious critical study which, it is to be hoped, will eventually be forthcoming.

Dr. C. B. Goodhart