One of the most incendiary topics in all of psychology is race differences in intelligence. Blacks score worse than whites do on intelligence tests, about 15 points worse to be exact, and psychologists have known this since the 1910's. Every few decades the public is reminded of this fact. It happened in 1969 with Arthur Jensen's famous article in the Harvard Educational Review, in 1994 with the publication of The Bell Curve, and in 2013 with the the discovery of Jason Richwine's dissertation. Each time these facts are brought to light a media shit storm ensues. Researchers who think that blacks score lower on intelligence tests than whites because of genetic differences between blacks and whites, so called hereditarians, are often labeled racists by the media. And the hereditarians insist that they are just being good scientists. They're telling the public the facts, even if the public doesn't want to acknowledge them. The truth is that both camps are in error. Many hereditarian researchers are not racist. But they are wrong. The facts do not support the view that genetics has anything with do with racial differences in intelligence. To demonstrate this I am going to analyze the common argument in favor of hereditarianism, show why they are wrong, and then demonstrate that we have good reason to suspect that the environment is what accounts for the black white IQ gap.
Many hereditarians start by pointing out that many studies shown that differences between individuals in intelligence are substantially due to differences in genetics. But individual differences being caused by genetic differences does not entail that group differences are. Consider two gardens of corn. Within each garden, the differences in height between corn stalks is mostly due to genetic differences between stalks. But the stalks in one garden are, on average, much taller than the stalks in the other garden. And this has nothing to do with genetics, it is because one garden gets much more sunlight than the other. Thus, the heritability of individual differences tells us nothing about the heritability of group differences.
Hereditarians also often note that many obvious environmental theories fail. For instance, blacks from rich families have lower IQ's than whites from poor families. So theories that center around poverty and education probably fall flat. But there are many more facets to environment than education and money. Nutrition, pre natal environment, and one's culture, are some obvious choices. Further more, what this argument boils down to is that environmentalists often have trouble proving which particular environmental variables cause the black white IQ gap. But the same can be said of hereditarians. They have no idea which specific genes cause the black white IQ gap and until they do they can't blame environmentalists for not knowing the specific environmental factors that cause the black white IQ gap.
Many hereditarians also point to trans-racial adoption studies to support there view. In these studies, members of one race are adopted by members of another race. The idea is that we can use these studies to see if difference in home life are what cause racial differences in intelligence. For instance, in the Minnesota trans-racial adoption study black children were adopted into white families. In late adolescence, it was found that blacks had lower IQ's than half black's who had lower IQ's than whites. The hereditarians frequently point to this study as a vindication of their views. But this studies sample size was very small. There were only 21 blacks. And these studies can only control for family environment. These black kids still grew up in a society that treated them a certain way because they were black. And they probably learned to expect certain things of themselves because they were black. Not to mention all the abnormality of dealing with growing up as a someone with adopted parents of another race.
And other studies have suggested different results. For instance, a study in the mid 1980's tracked one set of blacks that were adopted by black parents and another set who were adopted by white parents. It found that that the black children adopted by white parents ended up being about 15 IQ points brighter than those adopted by black parents. 15 IQ points is also the gap in IQ between whites and blacks.
Decades before the Minnesota trans-racial adoption study, a study in England was carried out in which children were placed in a highly enriched nursery environment aimed at helping the children learn well. This environment, which reflected an institutional environmental setting outside of the home, was applied to all children equally regardless of race. It produced some interesting results. As it turns out, the black children ended up having higher IQ's than the white children. Hereditarians often respond to this study by saying that the children were young, only 4 or 5, when their IQ's were measured. This is seen as problematic because genetics counts for very little of individual differences in IQ at such a young age. But this reply doesn't meet close scrutiny. For one thing, mainstream estimates state that about half of IQ variation is heritable even in childhood. And besides, by the age of 5 the vast majority of the black/white IQ gap is already present. So if genes don't matter much at age 5, so much worse for explanations that rely on genes.
The next hereditarian argument will take a little bit of background knowledge to understand and refute. People that do well on math tests also tend to do well on writing tests and tests of vocabulary. In fact, the scores people get on all the common intelligence subtests correlate with each-other. Researchers refer to a variable, G, for general intelligence, which explains this overlap in scores. Some subtests correlate with general intelligence better than others. This is called a tests g-loading. The higher the g-loading the more it correlates with general intelligence. It has also been found that those subtests which are the most influenced by genes are also the most g-loaded. Finally, it has been found that the subtests that blacks and whites have the biggest differences in scores on are also the most g-loaded and therefore the most heritable. This suggests that the black white IQ difference is probably heritable too since it correlates with the heritability of paticular tests.
Key to this argument then is the estimates of the heritability of difference IQ subtests. And just how do researchers come up with heritability estimates? The study designs are actually pretty simple and open to some pretty obvious flaws. The simplest of studies note that a trait under consideration, say intelligence, runs in families just like genes do and computes heritability based on the extent to which the trait runs in families. But this is obviously unsound: environments run in families too. You and your siblings probably had much more similar environments growing up than you and your second cousin did. So the fact that you and your siblings are probably more alike in terms of intelligence than you and your second cousin are hardly implies that genes have anything to do with it. In a slightly more sophisticated study design researchers compare identical and fraternal twins. Identical twins share essentially all of their genes and fraternal twins share half of their genes. So, the thinking goes, if identical twins are more similar in-terms of intelligence than fraternal twins are then it must be because identical twins are more similar genetically than fraternal twins are. Researchers use the difference between the similarity in identical and fraternal twins to compute heritability. Again, the problems here are obvious. People treat identical twins more similarly than they do fraternal twins. For instance, in one study identical twins were found to be over 3 times as likely as fraternal twins to report that they felt that them and their twin were “brought up as a unit” and 9 times as likely to have been confused for their twin. So this approach won't work either.
Finally, there is the gold standard of behavioral genetics research: the twin adoption studies. In these studies researchers look at how similar twins are when they are raised in different homes via adoption. These studies show that identical twins reared apart are more similar in terms of intelligence than fraternal twins are. Surely, this difference must be due to their genes! Once again, this design has serious flaws. If carried out perfectly, fraternal twins will still often have more different environments than identical twins will. This is for several reasons. For one, fraternal twins are often not of the same sex! One is a boy and one is a girl. This cannot happen in the case of identical twins. Obviously, this will bias the results to find more similarly among identical twins. Secondly, identical twins will usually have shared a placenta while in the womb. This means that they will have had a much more similar prenatal environment than fraternal twins. And this prenatal period, in which so much development occurs, could be highly important for complex traits like intelligence. And yet, the methods of behavioral genetics completely over look this key factor. Further still, these twin methods are never carried out the way they are planned. The twins often grew up for years with each-other before being separated. And when they are separated it is often to highly similar upper middle class homes. And many times they are adopted by members of the same families or people who live in the same city! These factors could produce the similarity of identical twins in terms of intelligence just as easily as genes could. Thus, behavioral geneticists don't have any sound way of measuring the heritability of an IQ subtest. So it is true that blacks and whites differ most with regards to tests that highly correlate with general intelligence. But we cannot use this fact to infer that blacks and whites differ most on those tests which are the most heritable because the estimates of heritability produced by behavior geneticists are methodologically unsound.
Another argument commonly used by hereditarians is based on a phenomena called “regression to the mean”. Regression to the mean happens when an unusually high or low value is followed by a value closer to the mean of a data set. In the case of intelligence, this would mean that really smart and really stupid people will have kids with intelligence levels that are closer to average then their own. This is a well documented fact. And, as it happens, smart blacks tend to have less intelligent kids than smart whites. Thus, blacks tend to regress to lower means than whites do. Hereditarians often note that is what would be expected on a genetic model of the difference due to genetic recombination. What they fail to note, however, is that an environmental explanation can just as easily explain this. If a black person gets lucky and avoids those environmental influences which normally depress black IQ, why on earth should we expect that their children will be just as lucky? The only reason why you would expect this is if you thought that the black white IQ gap was caused by something as simple as money which richer and smarter black parents could easily pass on to their children. But we've already discarded some simplistic explanations. Regression to the mean is consistent with a genetic or an environmental hypothesis. It doesn't tip the scale one way or another.
Another common hereditarian argument concerns brain size. Brain size has a modest correlation with intelligence meaning that people with larger brains tend to be a little smarter. And blacks have, on average, smaller brains than whites. Finally, differences in brain size among the general population are highly heritable. So brain differences between the races are probably due to genes and probably cause some of the difference in intelligence. This argument is flawed in a number of respects. First, as we have already reviewed, heritability estimates are founded on bad methods. Secondly, even if they were sound the heritability of individual differences tells us nothing about the heritability of group differences. And thirdly, two groups can differ in brain size but not differ in intelligence. For instance, men have larger brains than women but have the same average IQ. So we know that brain size differences don't always lead to intelligence differences.
The final hereditarian argument to consider runs like this: the races evolved in extremely different contexts. This caused them to evolve differences in almost all physical traits from skin color to wrist size to height. Why, then, should we not expect evolution to have effected the psychologies of the races in the same way? Isn't is just wishful thinking to propose that evolution essentially stopped at the neck? This argument is pretty good so far as arm chair speculation goes. But it isn't anything more than that and data doesn't back it up. What it proposes is that we should expect the races to have evolved different psychologies because they evolved somewhat different bodies. What this ignores is that we know, in general, that the races are the same psychologically. For instance, a huge meta analysis of studies on dozens of facets of personality found that the races were basically the same in the vast majority of personality measures used. Thus, armchair evolutionary story telling aside, we have good evidence for the view that the races developed psychologies much more similar than their bodies.
Now that we have seen that the common arguments for hereditarianism are in error, lets look at some arguments against hereditarianism. The first comes from studies of the IQ's of children from different combinations of black and white parents. According to the hereditarian view point, if you are of mixed race whether it is your mother or your father who is black should make no difference. But this isn't what is found. Quite to the contrary, studies have found that children of black mothers and white fathers score 9 IQ points lower than children of white mothers and black fathers. On an environmental view this is easy to explain: mothers matter more for child development than fathers do and the parenting practices of white mothers may be more conducive to intelligence than those of black mothers. This finding is much harder to explain on a hereditarian view.
Next, lets look at studies on racial ancestry. According to the hereditarian view, the more white genes a black person has the smarter they should be. This is not what is found. The evidence begins with two studies of racial admixture as measured by blood group analysis. Both of these studies found that white admixture did not significantly predict IQ. Similarly, self reported white ancestry among blacks has been found to not correlate with IQ. Even studies of mulattos, those who are half black and half white, have produced mixed results. The hereditarians may point to skin color as a measure of racial ancestery which does correlate with IQ. But the fact that visible markers of racial ancestry correlate with IQ while ones not so easily seen, but measured by self reported ancestry and blood group analysis, do not, suggests that the correlation between skin color and IQ is the result of differential treatment based on skin color, by ones self or others, rather than being directly the result of genetic differences.
We have now seen that the arguments in favor of hereditarianism fail while the arguments against it succeed. Now lets say a few things about what environmental factors might influence group differences in intelligence. The first thing to note is that we know beyond a shadow of a doubt that the environment has a strong enough influence on intelligence to account for the difference. In fact, the IQ's of blacks in America today are already higher than those of white Americans in the early 1940's. But the fact is that we don't know which specific environmental factors influence the black white IQ gap. It's probably a large number of factors which may include nutrition, prenatal environment, infant health, differences between black and white culture (especially concerning parenting style), and the effects of stereotypes on blacks.
In summary, there is no good reason to think that genetics plays an important role in the black white IQ gap and very good reason to think that it does not. We know that the environment is capable of producing the gap but we don't know which specific variables account for it yet. We do know that money and education are not sufficient to explain the gap and that a more subtle approach to environmental influences is therefore called for. Hopefully this will characterize future research as we come to realize what specific environmental factors account for the black white IQ gap.
