NOTES ON IQ
IQ tests and scores are not in fact crucial to our thesis, but they are useful. Intellectual accomplishment is all that really matters: If people routinely won Nobel Prizes in physics with low IQ scores, or, for that matter, routinely aced calculus exams but flunked the IQ test, we'd junk the IQ tests. But that doesn't happen: IQis an imperfect but useful measure of intelligence.
You'll frequently hear that we don't really know what intelligence is, that we don't know how to measure it, that IQtests are biased, and that IQ scores don't predict anything, or that they don't predict anything outside of school. Often these complaints are salted with personal anecdotes about some acquaintance that had a high IQscore but was lazy, crazy, or suffered from unforgivable personal hygiene. And in recent years, other forms of intelligence have become all the rage. Daniel Gole- man has written of "emotional intelligence" and "social intelligence," pointing out how they can help to predict job success and personal happiness. And other forms of intelligence have been proposed. In his 1993 book, Howard Gardner suggested that there are many types.28 But the data hardly support these attempts to complexify cognitive testing. The supposed special kinds of intelligence don't predict anything useful or, when theydo, predict only to the extent that they are correlated with general intelligence.
Yet IQ tests work in the sense that they predict performance. They were originally developed in order to predict how well children would do in school, and they do an excellent job of that. They also have moderate to high predictive power on many other questions, such as job performance, health, risk of accidental death, income, and other characteristics that may be less obvious, such as susceptibility to Alzheimer's disease. To make our position perfectly clear, we'd like to emphasize that saying IQscores have some predictive power is not the same thing as saying that they determine everything.
Of course, exceptions don't make trends disappear. Muggsy Bogues may have played in the NBA while being 5 feet 3 inches tall, and there may be numerous individuals who are 6 feet 8 inches but were so clumsy on their high-school basketball teams that they sat on the bench the entire season. But in general, height still matters in basketball. It's not the only thing that matters, it doesn't absolutely determine success, but on average it makes a lot of difference. The same can be said of IQ^For most life events it's not as important as height is in basketball, but it's fairly important. Nor are IQtests biased: They predict academic performance with the same accuracy in different ethnic groups.29
Moreover, IQis highly heritable. What this means is that an individual's IQis partially determined by genetic factors, so that it tends to be more similar to that of his or her parents and siblings than a randomly chosen person's IQwould be. Siblings with the same biological parents have similar IQs even when they are raised separately, whereas adopted siblings don't, even when raised together.
The same is true of height: Tall people tend to have taller- than-average children. In fact, IQin adulthood is about as heritable as height. IQin childhood, on the other hand, is less heritable and more susceptible to environmental influences. These effects of environment on the measured IQs of children, which disappear at or after puberty, are the basis for claims that IQcan be improved by interventions like Head Start.
Nongenetic factors also influence IQ,but for the most part, the ones that matter are not the ones people thought would matter. Prenatal care, breastfeeding, nutrition, access to early education, Mozart in the womb, and oat bran all have little or no effect. Surprisingly, the way in which a family raises children seems to have no effect on adult IQ. This argues against some popular environmental explanations for high intelligence among the Ashkenazi Jews—in particular, the notion that Jewish mothers have a special way of rearing children that boosts IQ.
ASHKENAZI PSYCHOMETRICS
As noted earlier, Ashkenazi Jews have the highest average IQof any ethnic group for which there are reliable data. Just how much higher is it? Many studies have found that it is 0.75 to 1.0 standard deviations above the general European average, corresponding to an IQof 112-115.30 Another, more recent study concluded that the advantage is slightly less, only half a standard deviation.31 While the difference between the Ashkenazim and other northern Europeans in average IQ may not seem large, it leads to a large difference in the proportion of the two populations with very high IQs.32 For example, if the mean northern European IQis 100, the mean Ashkenazi IQis 110, and the
standard deviation in both populations is 15, then the number of northern Europeans with IQs greater than 140 should be 4 per 1,000, whereas 23 per 1,000 Ashkenazim should exceed the same threshold, about a sixfold difference. This is a general statistical effect, not something that happens only with IQ.
The fact that Ashkenazi Jews have high IQs on average and corresponding high academic ability has long been known. In 1900 in London, Jews took a disproportionate number of academic prizes and scholarships in spite of their poverty.33 In the 1920s, a survey of IQ_scores in three London schools with mixed Jewish and non-Jewish student bodies—one prosperous, one poor, and one very poor—showed that Jewish students, on average, had higher IQs than their schoolmates in each of the groups. The differences between Jews and non-Jews were all slightly less than one standard deviation, and the students at the poorest Jewish school in London had IQscores equal to the overall city mean of non-Jewish children.34
That study, though conducted in 1928, is still important today because it contradicts a widely cited misrepresentation of a paper authored by researcher Henry Goddard in 1917.35 Goddard gave IQtests to people suspected of being retarded and found that the tests identified retarded Jews as well as retarded people of other groups. The psychologist Leon Kamin reported in 1974 that Goddard had found that Jews had low IQ_ scores. Kamin's reason for citing the study was to claim that Goddard and other IQresearchers of the 1920s had been biased against Jews and other minority groups. This erroneous analysis was picked up by many authors, including the well-known Harvard evolutionary biologist Stephen Jay Gould, who used it as evidence of the unreliability of IQtesting.36 Gould seems tohave believed that a popular impression that Jews had low IQs contributed to the passage of the Immigration Act of 1924, which was aimed at restricting immigration from southern and eastern Europe. However, by 1922 Jews already made up more than a fifth of Harvard undergraduates, and the Ivy League was already instituting admissions policies aimed at limiting Jewish admissions (the infamous "Jewish quotas"), which involved placing less emphasis on academic merit. If some people in the 1920s had the impression that Jews had low IQs, that impression cannot have been widely shared. The 1928 study of IQ_ in three London schools shows that in fact there were already researchers in the West who were noticing that Jews seemed to have higher IQs, on average, than the members of other groups.
But the achievements of Ashkenazi Jews are certainly not confined to IQscores. They have an unusual ability profile when it comes to some other forms of testing as well. They have high verbal and mathematics scores on other types of standardized tests, though their visuospatial abilities—that is, their ability to rotate three-dimensional objects in their minds, for example— are typically somewhat lower, by about half a standard deviation, than the European average. The Ashkenazi pattern of success corresponds to this ability distribution—great success in mathematics and literature, more typical results in representational painting, sculpture, and architecture.
It is noteworthy that non-Ashkenazi Jews do not have high average IQscores. Nor are they overrepresented in cognitively demanding fields like medicine, law, and academics. In Israel, Ashkenazi Jews, on average, score 14 points higher than Oriental Jews, almost a full standard deviation, which is 15 or 16 points on most IQ_tests.37That difference means that the average non-Ashkenazi Jew in Israel would have an IQscore that would be at the 20th percentile among the Ashkenazim. Academic accomplishment in the two groups seems to vary in the same way, even among those born and raised in Israel: Third- generation Ashkenazi Jews in I
srael are 2.5 to 3 times more likely to have graduated from college than third-generation Mizrahi Jews, for example (the ancestors of the Mizrahim moved to Israel from Asia and North Africa).38
THE ASHKENAZI MUTATIONS
The best-known of the genetic diseases disproportionately affecting Ashkenazi Jews are Tay-Sachs disease, Gaucher's disease, and the breast-cancer mutations BRCA1 and BRCA2, but there are a number of others, such as Niemann-Pick disease, Canavan disease, and familial dysautonomia. Some of these cause neurological problems. And they're unusually common among Ashkenazi Jews—so common that they constitute an enduring puzzle in human genetics.
In principle, absent some special cause, genetic diseases like these should be rare. New mutations, some of which have bad effects, appear in every generation, but those that cause death or reduced fertility should be disappearing with every generation. Any particular harmful mutation should be rare; however, one in every twenty-five Ashkenazi Jews carries a copy of the Tay- Sachs mutation, which kills homozygotes in early childhood. This is an alarming rate.
The mutations that so frequently affect Ashkenazi Jews are mysterious in another way. Many of them fall into two categories or clusters involving particular metabolic pathways: Theyaffect the same biological subsystem. Imagine a fat biochemistry textbook, where each page describes a different function or condition in human biochemistry: Most of the Ashkenazi diseases would be described on just two of those pages. The two most important genetic disease clusters among the Ashkenazim are the sphingolipid storage disorders (Tay-Sachs disease; Gaucher's disease; Niemann-Pick disease; and mucolipidosis, type IV) and the disorders of DNA repair (BRCA1 and BRCA2; Fanconi anemia, type C; and Bloom syndrome).
What is the explanation of this odd pattern? We know of only two mechanisms that can create high frequencies of dangerous, even lethal mutations: genetic drift in a bottleneck or natural selection.
THE BOTTLENECK HYPOTHESIS
Most medical geneticists believe that these common Ashkenazi genetic diseases are a product of population bottlenecks. A population bottleneck occurs when a population goes through a period in which it is quite small. This often happens in the founding of a population. In a bottleneck, gene frequencies change almost randomly; just as you can get unrepresentative results (different from 50-50) when you flip a coin just a few times, or when you poll 20 people rather than 1,000, in a population bottleneck you get random changes that can affect large portions of the population in question.
When we say that a population is "small," we generally mean a few hundred individuals at most. Europe, for example, did not go through a bottleneck following the Black Death in the Middle Ages. The plague may have killed off half the population of Europe, but 40 million survivors is not a small number. It left plenty of genetic variation intact.
If a few lethal mutations became common in a bottleneck and a dramatic population expansion followed, we would see a large population with a surprising number of genetic diseases— diseases that were rare in most other populations. This has certainly happened in some cases. Amish communities that had very few original founders have seen this effect with their high incidence of several specific genetic disorders. It also happened in Pingelap, a Pacific island that was devastated by a typhoon around 1775, leaving about twenty survivors. Today almost 10 percent of the islanders suffer a form of severe color blindness.
Genetic diseases made common in a bottleneck are the product of chance, however, so there is no particular tendency for them to fall into a few metabolic pathways: They're scattered all over the biochemistry book, not concentrated on a few pages.
Our knowledge of human genetics has expanded rapidly over the past few years, and we now have good estimates of the total number of human genes (about 22,000) and the number of genes in different functional categories—in particular, the number involved in sphingolipid metabolism (108). We looked at twenty-one genetic diseases among the Ashkenazi and calculated the probability of finding four that affect sphingolipid metabolism, assuming randomness, in a given population. That probability was very low, less than 1 in 100,000. That can't be a coincidence.
We can say some other things about a population that has recently passed through a tight bottleneck. There would be overall genetic changes: reduced genetic variety in nuclear genes, increased genetic linkage, and increased genetic differences fromother populations. All of these properties are measurable, and none have occurred among the Ashkenazi Jews.39
Finally, a genetic bottleneck would not increase a population's intelligence. If it was severe enough, it would almost certainly decrease intelligence as moderately deleterious genes became common.
Therefore, although bottlenecks can explain high frequencies of genetic disease in some cases, the bottleneck hypothesis cannot possibly explain the genetic data and the spectrum of genetic disease observed among the Ashkenazi Jews.
NATURAL SELECTION
The alternative explanation is natural selection, the process by which some alleles cause increased reproduction in their bearers. Some gene variants have favorable effects in a given environment (in this case, the physical and social environment experienced by the Ashkenazi Jews during the Middle Ages), so that people with those variants have more children, on average, than others in that population. Those variants gradually become more common, ultimately leading to significant changes. In some cases, certain gene variants can have positive effects in individuals with one copy, and negative effects, such as disease, in individuals with two copies—the people with one copy have a "heterozygote advantage." As we have discussed earlier, the most famous example is sickle-cell anemia, where a mutation causing a very dangerous form of anemia in those carrying two copies has risen to high frequency in some parts of the world. There are a number of other malaria defenses of this sort that are expensive in terms of human health.
Clearly, natural selection can sustain quite high frequencies of serious, even lethal genetic diseases in some circumstances. Just as clearly, it can lead to a set of common mutations that cluster in a few metabolic pathways. This has happened with the malaria defenses: Many affect the hemoglobin molecule (sickle-cell, hemoglobin C, hemoglobin E, alpha- and beta-thalassemia), whereas others, such as G6PD deficiency or glycophorin C, affect other aspects of the red blood cell. Since the malaria parasite attacks red cells, this pattern is easy to understand.
Heterozygote advantage isn't confined to genetic defenses against malaria; it can also occur in other cases where certain traits are favored by selection. It seems that the key to such cases is that there has been strong selection (carriers have a big advantage) applied over a relatively short time period. Over longer periods, mutations with fewer side effects eventually occur and win out. The fact that heterozygote advantage can favor other traits is important, because we think that most of the characteristic Ashkenazi mutations are not defenses against infectious disease. One reason is that these mutations do not exist in neighboring populations—often literally people living across the street—that must have been exposed to very similar diseases. Instead, we think that the Ashkenazi mutations have something to do with Ashkenazi intelligence, and that they arose because of the unique natural-selection pressures the members of this group faced in their role as financiers in the European Middle Ages.
We see a clear example of heterozygote advantage in a trait other than disease resistance in whippets, a breed of dog similar to a small greyhound. Some whippets carry a mutated version of myostatin, a gene that limits muscle development. Dogs withthis mutation grow more muscle. Whippets with one copy are faster, on average, than other whippets and often win races.40 Those with two copies, called "bully whippets," are extremely muscular, have muscle spasms, and are not competitive as racers. One copy gives an advantage in a particular ability; two copies actually have a negative effect on the same ability.
Now, in order for natural selection to work in this way, the population has to be genetically isolated from its neighbors; otherwise it cannot be
come different. Admixture dilutes the effects of natural selection and can stop it in its tracks. You might compare it to boiling down soup while continually adding
Whippets with 0, 1, and 2 copies of the myostatin mutation
water—you won't get anywhere that way. As it happens, the Ashkenazi Jews were genetically isolated during the Middle Ages: not because of the Pacific Ocean, as happened with Pin- gelap, but owing to social reasons—internal rules against intermarriage combined with external prejudice.
For most of that period, both intermarriage with non-Jews and conversion to Judaism were very rare. That's certainly what the historical record indicates, but we can check that record using genetics. If we look at alleles that are clearly from the Middle East, we see that they account for a substantial fraction of Ashkenazi ancestry today: at least 50 percent, according to our analysis. This shows strong limits on the rate of genetic admixture, since even 2 percent mixing per generation over the past 2,000 years would have caused the Ashkenazim to become almost completely (80 percent) European. A continuing process of admixture (as opposed to a lot of admixture in the early stages) interferes the most with ongoing natural selection, but even if we make that pessimistic assumption, it looks as if the admixture rate was under 1 percent per generation—low enough to allow for the sort of natural selection we're suggesting. In fact, early European admixture might even have furthered the selective process, since even low levels of admixture can be an important source of beneficial alleles. More generally, the position of Israel at a natural crossroads, subject to invasion by Romans, Greeks, Persians, Babylonians, Assyrians, and Egyptians, may have resulted in unusually high genetic variety, which would also have furthered selection.
The 10,000 Year Explosion Page 19