EPILOGUE
THE STRONG HEREDITARIAN view of intelligence holds that intelligence is mostly a matter of genes. You are only going to be as smart as your genes allow, and nothing much in the environment—neither the way you are brought up in your family nor the kind of schools you go to—is going to significantly change that. Many if not most scientists who study intelligence and many if not most laypeople in the United States believe this. Such thinking is extremely unfortunate for the individual, because it implies that hard work can produce little in the way of improvement to “real” intelligence. It is a disaster for public policy because it implies that educational interventions are doomed to failure. Fortunately, the view is quite wrong. And here is how we know.
There is no fixed value for the heritability of intelligence. It differs from one population living in one set of circumstances, to another population in another set of circumstances. If the environment is highly favorable to the growth of intelligence, then the heritability of intelligence is indeed fairly high—perhaps as high as 70 percent. This is the situation that exists for the upper middle class in developed countries. The environments they create promote intelligence, and one family differs little from another in that respect. At the limiting extreme of identical environments for everyone in a given group, the only factor that can influence differences in intelligence is genetics. The upper middle class comes close enough to that situation that heredity for that group can be very important in determining differences in intelligence.
But if the environment is highly variable—differing greatly between individual families—then the environment is going to play the major role in differences in intelligence between individuals. And this is the situation for the poor. For them, only 10 percent of the variation in intelligence is driven by heredity, which means that improving the environment of children born into poor families could have a big effect on intelligence. And in fact if you raise a poor child in an upper-middle-class household, the expected value added is at least 12 IQ points and may be as high as 18 IQ points. The effect on academic achievement is also very great—at least half a standard deviation and under some circumstances as much as a full standard deviation.
Entirely aside from the degree to which heritability is important for one group or another in the population, heritability places no limits whatsoever on modifiability—for anybody. The height of people in developed countries has increased greatly in recent generations—and this increase has nothing to do with genetics.
There has been a similarly striking increase in IQ over the past century. The scores on IQ tests have increased by more than 18 points in the last sixty years and probably by the better part of two standard deviations (30 points) over the past hundred years. And on the Raven Progressive Matrices—the test heralded for decades as being a culture-free measure of true intelligence—the gain has been two standard deviations in less than sixty years.
Why have the gains occurred? It is simple at base: the schools and the culture have changed radically in such a way as to affect scores on many of the subtests of IQ tests. Parents and schools increasingly teach children how to categorize objects and events in taxonomic terms suitable for scientific analysis. The media teach children the ways of the world—why policemen wear uniforms, why street addresses are numbered in order, and why people pay taxes—resulting in higher scores on comprehension subtests of IQ tests. Improvements on the Raven matrices—and in the fluid intelligence underlying performance on it—can be traced at least in part to the ever-more geometric and analytic ways of teaching arithmetic over recent decades, and possibly in part to computer games. A few years ago McDonald’s was including in its Happy Meals mazes that were more difficult than the mazes in an IQ test for gifted children!
And then there is the fact that people are receiving a lot more education than ever before. In a century the mean number of years of schooling has gone from seven to fourteen. Since a year of school adds as much to IQ scores as two years of age, it would be astonishing if IQ had not changed radically over that period.
How much of the IQ gains can be called “real” intelligence gains? I can say several things about that. First of all, it is out of the question that the great-grandchildren of people who were ten years old in 1910 are two standard deviations smarter, if we define intelligence broadly as the “ability to reason, plan, solve problems, think abstractly, comprehend complex ideas…and learn from experience.” On the other hand, people today really are more intellectually capable than their forebears. Children who can tell you why street addresses are numbered consecutively are smarter in some sense than children who can’t. Being able to think about the similarities of objects and events in taxonomic terms is a real advantage. Heuristics for reasoning, such as procedures for hypothesis testing, are part of the curriculum at every level of school, and they can be applied to everyday problems. Planning and choosing are two aspects of intelligence that have been increased by virtue of widespread knowledge of probability theory and cost-benefit reasoning.
Since school makes children smarter, there is no doubt that better schools can make them smarter still. Although vouchers, charter schools, whole-school interventions, and teacher certification or higher academic degrees do not reliably improve education, other factors do—and some matter a great deal. Teachers differ a lot in quality, and so finding ways to improve the quality of teaching could make a great difference. If we could replace the bottom 5 percent of teachers every year with average-quality teachers, the level of children’s academic performance would increase hugely in just a few years. Use of computer-assisted forms of teaching can produce huge gains in the rate of learning, and some types of cooperative learning are highly effective. And recall the Herrnstein demonstration with an intensive program in Venezuela that radically improved the problem-solving skills of ordinary junior high school students. It also raised their IQ scores by a nontrivial amount—5 points on a typical test of multiple problem-solving skills.
The received opinion about the relationship between social class and intelligence is that intelligence, which is largely inherited, drives social class. Smarter people have better genes so they are destined to rise in society, whereas less smart people have worse genes so they are destined to fall. It is true that intelligence is partially heritable, and more intelligent people on average will be of a higher social class in virtue of their greater inherited intelligence. But I believe that the role of genetic inheritance in determining social class is fairly small. The difference between the average IQ of the children of the lower third of the socioeconomic status (SES) distribution and the average IQ of the children of the upper third is about 10 points. We know that some of this is due to biological but not genetic factors, including exercise, breast-feeding, and exposure to alcohol or cigarette smoke, as well as hazardous chemicals and pollution. And some of it is due to the disruption in schools of lower-SES children and to the fact that peers are pulling intelligence mostly in a down direction. We also know that socialization in lower-SES homes is not optimal for developing either IQ or school readiness. Moreover, a child born into roughly the bottom sixth of the SES distribution will have an IQ 12 to 18 points higher if raised by parents from roughly the top quarter of the SES distribution. All of this does not leave much room for genes in the social-class equation. I do not doubt that genes play a role, but I would be surprised to find that the differences in inherent genetic potential of the social classes are very great. Certainly much if not most of the 10 points separating the average of the children of the lower third and the average of the children of the upper third is environmental in origin.
For the race difference in IQ, we can be confident that genes play no role at all. Most of the evidence offered for a genetic component to the race difference is indirect and readily refuted. Virtually all of the direct evidence, which is due mostly to the natural experiment resulting from the fact that American “blacks” range from being completely African to largely European in heritage, indicates no genetic difference at all with respect to IQ. And the difference between the races in both IQ and academic achievement is being reduced at the rate of about one-third of a standard deviation per generation. The IQ of the average black is now greater than that of the average white in 1950.
The No Child Left Behind Act demands that the difference in academic achievement between the classes and between the races be erased in half a generation by the schools alone. This is absurd. It ignores the fact that class and race differences begin in early infancy and have as much to do with economic factors and neighborhood and cultural differences as with schools.
That is the bad news about gap reduction. The good news is that big improvements in IQ and academic achievement for lower-SES and minority children are possible. And we know at least the outlines of what those improvements look like. Half-measures have been tried and are not going to make a lot of difference. We need intensive early childhood education for the poor, and we need home visitation to teach parents how to encourage intellectual development. Such efforts can produce huge immediate gains in IQ and enormous long-term gains in academic achievement and occupational attainment. Highly ambitious elementary, junior high, and high school programs can also produce massive gains in academic achievement. And a variety of simple, cost-free interventions, including, most notably, simply convincing students that their intelligence is under their control to a substantial extent, can make a big difference to academic achievement.
Believing that intelligence is under your control—and having parents who demand achievement—can do wonders. At any rate that has been true for Asians and Jews. There is no reliable evidence of a genetic difference in intelligence between people of East Asian descent and people of European descent. In fact, there is little difference in intelligence between the two groups as measured by IQ tests. Some evidence indicates that East Asians start school with lower IQs than do white Americans. After a few years of school this difference seems to disappear. But the academic achievement of East Asians—especially in math and the sciences, where effort counts for a lot—is light-years beyond that of European Americans. Americans of East Asian extraction also differ little in IQ from European Americans. In any case, the academic achievement and occupational attainment of Asian Americans exceed by a great amount what they “should” be accomplishing given their IQs. The explanation for the Asian/Western gap lies in hard work and persistence.
Jewish culture undoubtedly has similarly beneficial effects. Jewish values emphasize accomplishment in general and intellectual attainment in particular. Differences between Jews and non-Jews in intellectual accomplishment at the highest levels are very great. A genetic explanation for this is not required inasmuch as even greater differences have occurred for Arabs and Chinese versus Europeans in the Middle Ages, for differences between European countries at various points since the Middle Ages (with reversals occurring between Italy and England and with movement from savagery to sagacity in scarcely two centuries in Scotland), and for regional differences in the United States. We are left with an IQ difference of two-thirds to a standard deviation between Jews and non-Jews. At least some of this difference is surely cultural in origin.
Finally, there is much that we can do to increase the intelligence and academic achievement of ourselves and our children. Everything from the biological (exercise and avoidance of smoking and drinking for pregnant women, and breast-feeding for newborns) to the didactic (teaching categorization, following good tutoring principles) can make a difference to intelligence.
We can now shake off the yoke of hereditarianism in all of our thinking about intelligence. Believing that our intelligence is substantially under our control won’t make us smart by itself. But it’s a good start.