| Galton Institute Home Page | March 1995 Newsletter Contents | Newsletter Index |
Although to most evolutionary biologists it may seem obvious that some significant component of human intelligence must be genetically heritable (the “hereditarian” view), it doesn’t necessarily follow that this must be so. It certainly isn’t something to be taken for granted, since many psychologists, educationalists, and teachers, are “environmentalists” and believe that intelligence develops only after birth as a result of environmental influences. In progressive circles it is politically incorrect, and indeed élitist, to suppose that intelligence and educability could be determined genetically at conception, with some people destined from the start to do better than others.
First of all, the evolutionary argument. Over the past 3˝ million years the brains of our hominid ancestors have more than quadrupled in size. The earliest australopithecine brain was not much larger than a modern chimpanzee’s, little more than 400 cc. From then onwards there has been a more or less steady increase in brain volume through Homo habilis and Homo erectus to modern Homo sapiens, which averages not much under 1800 cc in some populations. And stone artefacts indicate that human material culture has been increasing in complexity in step with the increase in brain volume. But the brain is a physical character, which must have evolved through natural selection, with individuals having the largest brains leaving the most descendants.
Natural selection however can only work in populations showing variance in the genetical heritability of the character being selected for. It follows that, if human intelligence has evolved through increasing brain size, as it surely must have done, at least some part of the variance in intelligence in the populations upon which selection has been operating must have been genetically heritable. Q. E. D.
However, although some of human intellectual achievement probably does depend upon genetically heritable intelligence, an important part of our cultural inheritance is transmitted by teaching and learning from one generation to the next. Insofar as this also could be said to be inherited, it provides an example of the inheritance of characters acquired during the lifetime of each individual, and it is subject to selection as better ideas and practices replace what went before. But there is no question of genetical heritability here, nor of natural selection as this is ordinarily understood by biologists, which must have been responsible for the increase in brain volume and also, presumably, in the intelligence associated with it.
So the question is, how much of human intellectual achievement nowadays is due to innate intelligence, and how much is acquired during the lifetime of each individual ? Extreme environmentalists, while probably conceding the validity of the evolutionary argument in the past, would say that, since nowadays cultural inheritance has almost completely taken over, the genetically heritable component of intelligence was now of little or no significance, implying that human biological as opposed to cultural evolution had come to its end. Hereditarians, on the other hand, while accepting the importance of cultural evolution, would believe that genetically heritable intelligence was still responsible for a significant part of the observed variance in intellectual achievement.
So it is important to discover whether there really is some significant genetically inherited component in human intelligence at the present time. This is not easily ascertained, but one approach is through the “Regression to the Mean” shown by polygenic characteristics, where offspring are closer to the mean for the whole population than are the parents. For example, human stature is known to be highly heritable, but although tall people have tall children these are usually shorter than their parents. And conversely, short people usually have rather taller children, both regressing upwards or downwards towards the mean. The same effect is seen with IQ, highly intelligent parents having children with scores somewhat lower than their own, while the children of unintelligent people, though none too bright themselves, often do better than their parents. This is consistent with IQ scores (and presumably the intelligence they are supposed to measure) being genetically heritable. If that were not so, one would have expected that the children would follow their parents, both being subjected to more or less the same environmental conditions. It is true that highly intelligent parents may not always provide the best environment for their children, but it seems unlikely that unintelligent people should actually do better in that respect. Although still somewhat controversial, there is subtle genetical argument here, the force of which is not always appreciated by environmentalists. However, it isn’t easy to see how it can be refuted.
There is anyway a quite different way of attacking this problem which has the advantage of producing an estimate of what proportion of intelligence is genetically inherited, as opposed to being due solely to environmental factors operating after conception. This is in the twins studies, pioneered by the late Sir Cyril Burt in the 1920’s and 1930’s. Burt claimed to have collected over 50 sets of identical (monozygotic) and fraternal (dizygotic) twins, who had been separated soon after birth and brought up by different foster parents. And it appeared that the separated monozygotic twins not only resembled each other in appearance but also had closely similar IQ scores, whereas the separated dizygotic twins showed no such resemblances. Since monozygotic twins are genetically identical, whereas fraternal twins are no more alike than they are to their other brothers and sisters, these differences must be due to the genetically inherited component of their IQ scores, which Burt estimated to have been in excess of 50%. However, 5 years after Burt’s death in 1971, serious questions about the validity, and indeed the honesty, of these twins studies were raised, originally in the Sunday Times and later confirmed and extended by the late Prof. L. S. Hearnshaw in the authorized biography, Cyril Burt Psychologist (1979). By that time Burt’s views on the heritability of intelligence, and especially his advocacy of the 11+ examination for selective Grammar Schools, was regarded as politically incorrect by many educational psychologists and other progressive people, who hastened to provide further evidence of supposed fabrication of data and other scientific fraud. This was derived from work published in old age (Burt was 88 when he died) but based upon what he had done long before in the 1920’s and 1930’s, and admittedly it was often muddled and badly presented, quite apart from any question of deliberate dishonesty. These allegation have however now been devastatingly answered in two books, The Burt Affair by Robert Joynson (1986) and Science, Ideology and the Media: The Cyril Burt Scandal by the late Ronald Fletcher (1991). And even the British Psychological Society, which in March 1980 had organized a special symposium “to assess the impact of Burt’s falsifications”, on 24 February 1992 rather grudgingly issued a statement which concluded “Council considers that it is now inappropriate for the Society as such to seek to express a fresh opinion about whether or not the allegations directed at Burt are true. Moreover, in the light of greater experience, the British Psychological Society no longer has a corporate view on the truth of allegations concerning Burt”. But this received much less publicity than did the original allegations of fraud.
Most unfortunately, on the advice of an eminent academic psychologist many of Burt’s numerous papers, which are known to have included his original notes on the twins studies, were incinerated soon after his death. Of course, it could well be that pioneering work done 60 years and more ago would be judged inadequate by modern standards but, since Burt himself was in no way responsible for the destruction of the evidence, he ought to be given the benefit of the doubt as to what it might have revealed. However, in principle studying the IQ scores of separated twins is a valid way of assessing the contributions of genetical inheritance and environmental factors respectively to the development of human intelligence. And this work has recently been repeated by others, who have found that up to over 70% of the variance in IQ scores was associated with genetic variation, close to Burt’s original estimate.
So the conclusion must be that there is indeed some substantial genetically heritable component in human intelligence as measured by IQ testing, which is irrevocably determined at conception. That has various important implications, which will be considered in a subsequent article.
C B Goodhart