Steve Connor: This is not a black and white issue

There is, perhaps, no brew quite so heady as that which mixes race, genes and intelligence. The history of the 20th century is replete with examples of how people with deep-seated racist views have misappropriated the science of genetics to justify their belief in the superiority of one race over another.

Jim Watson, the co-discoverer of the DNA double helix, is no racist. I have met him several times and believe that his words, as quoted in a Sunday newspaper last weekend, are more of a reflection of a scientist who wants open and honest debate than someone who genuinely believes that whites are superior to blacks.

Nevertheless, Watson's remarks implying that black Africans are less intelligent than white westerners have been seen as offensive and racist. He appears to justify his position by citing the results of IQ tests on different racial groups. He also invokes Darwinian natural selection by saying: "There is no firm reason to anticipate that the intellectual capacities of peoples geographically separated in their evolution should prove to have evolved identically. Our wanting to reserve equal powers of reason as some universal heritage of humanity will not be enough to make it so."

Watson, who arrives in Britain today to promote his new book, Avoid Boring People, has some explaining to do. He is a controversialist who enjoys making mischief. He is also a supreme master in the art of science communication so it will be interesting to see how he extricates himself from his latest polemical outburst.

He will no doubt cite, once again, the results of IQ tests indicating that black Americans are less intelligent than white Americans. This is ancient stuff, but few of us actually know what it means. For a start, people who are good at IQ tests are merely demonstrating little more than they are good at IQ tests. Intelligence per se is a more difficult entity to quantify.

In fact, it is Chinese Americans who appear do best in "non-verbal" IQ tests. Studies going back 20 years or more also show that Asian Americans – who in 1987 comprised just 2 per cent of the US population – are staggeringly overrepresented in American universities: 14 per cent at Harvard, 16 per cent at Stanford, 20 per cent at MIT and 21 per cent at Caltech. One psychologist in the 1960s prophetically described Chinese Americans as the "natural aristocracy" because of their overachievement in practically all things intellectual.

So the story is not black and white. Leaving aside the usefulness of IQ tests and what they measure, the question is whether the difference between the races is down to genetics or whether it is due to cultural upbringing. Just because skin colour and race are genetically determined, is it reasonable to suppose that IQ difference between the races has also got something to do with genes?

At the heart of this question is the subject of heritability, a scientific measure of how much of a trait is due to genetics (nature) and how much is due to environment (nurture). Studies of identical twins, who have the same genes, reared apart suggest that the heritability of IQ is 70 per cent, meaning that 70 per cent of the variation in IQ is due to genetics.

There is a widespread assumption that because IQ has such a large heritability, then the differences in IQ between the races is largely due to genes and not environment, culture or upbringing. This would be wrong. For a start, heritability has a major, inherent problem. It is a mathematical ratio and as such it applies only to the population in question. It can only be used to analyse variation within that particular group, and cannot be used to compare differences between groups.

In other words, heritability cannot be applied to any other group, other than the one in question. Heritability in whites cannot, therefore, be used to explain heritability differences between whites and blacks, or even between two populations of the same racial group separated in time.

People now do much better in IQ tests than 30, 40 or 50 years ago. The reason for the effect is not clear. What is clear is that the improvement cannot be genetic. Evolution does not work that fast. Black American children also do worse in IQ tests as they get older. Again, this fall in IQ is not genetic, and must have something to do with upbringing and environment, possibly interacting with genes.

And even something that is completely heritable does not mean that it is immutable and fixed. Height, for example, is highly heritable, yet until recently the middle classes were significantly taller than lower social classes, simply because they had a better diet.

Another example is the genetic disorder phenylketonuria, which is 100 per cent heritable. Children with the defective genes, left untreated, become educationally subnormal. However, a simple change to their diet at birth rectifies the problem – a supreme example of the power of nurture over nature.

Jim Watson knows all this, so it is a bit of a mystery as to why he raised the issue in such simplistic terms – unless of course he really believes that we are slaves to our DNA.

s.connor@independent.co.uk