Dna: The Secret of Life

by Watson, James

  MZ twinning occurs globally in about four of every thousand pregnancies and seems to be nothing more than a random accident. DZ twinning, on the other hand, may run in families, and varies from population to population: a group in Nigeria tops the list with DZ twins accounting for forty pregnancies per thousand, while there are only three per thousand in Japan.

  The basic premise of the standard form of twin study is that both members of a same-sex pair of twins, whether DZ or MZ, are raised the same way (i.e., receive similar "nurtures"). Suppose we are interested in a simply measurable characteristic like height. If DZ1 and DZ2 were both raised on the same diet of food, love, and so on, any difference in height between them would be attributable to some combined effect of genetic differences and whatever subtle differences of nurture may have crept in (for instance, DZ1 always finishes her milk; DZ2 never does). But if we follow the same program with MZl and MZ2, the fact that these twins are genetically identical eliminates genetic variation as a factor; any differences in height must be a function of only those subtle environmental differences. All things being equal, MZ twins will then tend to be more similar in height than DZ twins, and the extent to which this is true gives us a measure of how much genetic factors influence height. Similarly, the extent to which MZ twins have more similar IQs than DZ twins reflects the effect of genetic variation on IQ.

  This kind of analysis is also applicable to the inheritance of genetic diseases. We say that twins are "concordant" if they both have the disease. An increase in concordance when we look from DZ twins to MZ twins would support the claim of a strong genetic basis to the disease: for example, DZ twins are 25 percent concordant for late-onset diabetes (if one twin has it, then there is a one in four chance that the other does too), whereas MZ twins are 95 percent concordant for the disease (if one twin has it, then nineteen out of twenty times the other does too). The conclusion: late-onset diabetes has a strong genetic component. Even here, though, the environment plays an obvious role: if it were not so, we would see 100 percent concordance in MZ twins.

  A long-standing criticism of this kind of twin study addresses methodology: MZ twins tend to be treated more alike by their parents than DZ twins. Parents sometimes make a virtual fetish of identicalness: often, for instance, MZ twins are even dressed exactly alike, a habit some weirdly carry into adulthood. This is a legitimate criticism insofar as the more pronounced similarity of MZ twins (as compared with DZs) is interpreted as evidence of genetic influence when in fact it could simply be a reflection of the more precisely similar nurtures shared by MZs. And here is a further wrinkle in the same problem: how do we tell whether a pair of same-sex twins is DZ or MZ? "It's easy," you say. "Just look at them." Wrong. In a small but significant proportion of cases, parents mistake their same-sex DZ twins for MZ twins (and thus tend to subject them to the supersimilar nurture routine – the same frilly pink frock for each); and conversely a small proportion of parents with MZ twins wrongly take them for DZs (dressing one in frilly pink and the other in bright green). Fortunately, DNA fingerprinting techniques have rescued twin studies from this comedy of errors. The test can determine for sure whether the pair are indeed as they were supposed to be, whether DZ or MZ. The mistaken-identity groups then serve as the perfect experimental control in the analysis: for example, height difference in DZ twins cannot be put down to differences in nurture if the parents were raising them as MZs.

  Perhaps no form of twin study holds more popular appeal than analysis of MZs separated at birth. In such cases, the rearing environments are often very different, and so marked similarities are attributable to what the twins have in common: genes. It makes good copy: you see reports of MZs separated at birth who, it turns out, both have red velvet sofas and dogs called Ernest. Striking though these similarities may be, however, chances are they are mere coincidences. There is almost assuredly no gene coding for the red velvet upholstery preference, or for impulses in dog naming. Statistically, if you list a thousand attributes – make and model of car, favorite TV show, etc. – of any two people, you will inevitably find ones that overlap, but in the press these are the ones that get reported, usually in the Believe It or Not column. My coauthor and I both drive Volvo station wagons and appreciate a cocktail or two, but we are most definitely not related.

  Popular or no, twin studies have had a checkered history. Part of the ill repute stems from the controversy surrounding Sir Cyril Burt, the distinguished British psychologist who did much to establish the use of twins in studies of the genetics of IQ. After his death in 1971, a detailed examination of his work suggested that some of it was fraudulent – Sir Cyril, some alleged, was not above inventing a few twins from time to time if he needed to bolster sample sizes. The truth of these charges is still debated, but one thing is undeniable: the episode cast a shadow of suspicion over not only twin studies but all attempts to understand the genetic basis of intelligence. In fact, the combination of the Burt affair and hair-trigger political sensitivity to the topic have in effect stifled research by cutting the flow of grant money. No money, no research. Tom Bouchard at the University of Minnesota, a distinguished scientist whose massive 1990 survey of twins reared apart redefined twin studies, had such difficulty raising funds that he was forced to go cap in hand to a right-wing organization that supports behavioral genetics to further its own dubious political agenda. Founded in 1937, the Pioneer Fund counts among its early luminaries Harry Laughlin, the chicken geneticist we encountered in chapter 1 who turned his attentions to humans and entered the vanguard of American scientific racism. The fund's charter was "race betterment with special reference to the people of the United States." That legitimate researchers like Bouchard should be faced with a choice between seeking such a tainted sponsor or seeing their work perish represents a staggering indictment of the federal agencies that fund scientific research. Tax dollars are being allocated according to political rather than scientific merit.

  Bouchard's Minnesota twins study revealed that a host battery of personality traits – as measured using standardized psychological tests – were substantially affected by genes. In fact, more than 50 percent of the variability observed in a range of characteristics – the tendency to be religious, to name one – was typically caused by underlying variation in genes. Bouchard concluded that one's upbringing has surprisingly little effect upon personality: "On multiple measures of personality and temperament, occupational and leisure-time interests, and social attitudes, MZ twins reared apart are about as similar as MZ twins reared together." In other words, when it comes to measurable components of personality, nature seems to trump nurture. This lack of impact of upbringing on personality development has even Bouchard scratching his head. Upbringing has little effect, and yet the data still show the environment's considerable effect: MZ twins raised apart are as similar to each other as those raised together, but there are nevertheless differences in both cases between members of a pair. Could there be an aspect of environment distinguishable from upbringing? One suggestion is that variation in prenatal experience, the life of a fetus in utero, may be important; even small differences at this early developmental stage – when, after all, the brain is being assembled – may have a significant impact on who we become. Even MZ twins may find themselves in very different uterine settings courtesy of the natural whims of implantation – the lodging of the embryo in the wall of the uterus – and the development of the placenta. The popular belief that all MZ twins share a single placenta (and therefore have similar uterine environments) is wrong: 25 percent of MZ pairs have separate placentas. Studies have shown that such twins differ more from each other than do pairs who have shared a placenta.

  The elephant in the living room of all twin studies is the genetics of intelligence. How much of our smarts is determined by our genes? Everyday experience suffices to prove there is a lot of variation out there. While teaching at Harvard, I became intimately acquainted with the familiar pattern: in any population, there are a few who really aren't too b
right, a few who are alarmingly smart, and a vast majority who are middling. The fact that the setting was Harvard, where the population had been preselected in favor of intelligence, makes no difference: the same proportions hold whatever the group. This "bell curve" distribution of course can describe just about any trait that varies in humans: most of us are medium tall, but there are a few super-tall and a few super-short among us. But when used to describe variations in human intelligence the bell curve has demonstrated powers to raise a dust storm of objection. The reason is that in a land of equal opportunity, where we are each free to advance as far as our wits will carry us, intelligence is a trait with profound socioeconomic implications: the measure of it is predictive of how one will fare in life. And so in this matter the nature/nurture debate becomes entangled with the noble aspirations of our meritocratic society. But given the complex interplay of the two factors, how can we reliably judge their respective weights? Smart parents not only pass on smart genes; they also tend to rear their children in ways that foster intellectual growth, thus confounding the effects of genes and environment. This is the reason careful twin studies are so valuable in permitting us to analyze the constituents of intelligence.

  Bouchard's study and earlier ones as well have found that as much as 70 percent of the variation in IQ is attributable to corresponding genetic variation: a strong argument for the primacy of nature over nurture. But does this really mean that our intellectual fate is largely sealed by our genes – that education (even our own free will) has little to do with who we are? Not at all. As with all traits, it is nice to be blessed with favorable genes, but there is much that nurture can do to influence the standing of any individual, at least in the bell curve's vast midland, where variations in social circumstance are mainly determined.

  Take the case of one group within Japan, the Buraku. They are the descendants of Japanese who by feudal custom had once been condemned to perform society's "unclean" tasks, like slaughtering animals. Despite the modernization of Japanese society, the Buraku remain impoverished and marginalized outsiders, scoring on average ten to fifteen IQ points lower than the national Japanese mean. Are they genetically inferior, or is their IQ simply a reflection of their lowly status in Japan? It would seem to be the latter: Buraku who have immigrated to the United States, where they are indistinguishable from other Japanese Americans, have shown an increase in IQ and over time the fifteen-point gap with their fellows in the homeland disappears. Education matters.

  In 1994, Charles Murray and Richard Herrnstein published The Bell Curve, arguing that, despite the well-established effect of education, the discrepancies in the average IQ scores of different races may themselves be attributable to genes. It was a profoundly controversial claim, but not as simpleminded as many have supposed. Murray and Herrnstein understood that the combined observations of a genetic basis to IQ and of differences in average IQ among groups do not lead directly to the conclusion that genes are responsible for the intergroup differences. Imagine sowing the seeds of a particular plant species in which height varies genetically. Put one set of seeds in a tray with high-grade soil, and another in a tray of poor soil: in both trays we see variation in height; some individuals are taller than others – as expected, given genetic variation. But we also observe that the average height for plants in the tray of poor soil is less than the average for those in the tray of rich earth. The environment, in the form of soil quality, has affected the plants. While genetics is the dominant factor in determining height differences among plants within a tray – all other factors being equal – genetics has nothing to do with the differences seen between the two trays.

  Does this same argument apply to African Americans, who lag behind other Americans in measures of IQ? Since poverty rates among African Americans are relatively high, with a large proportion of individuals finding themselves rooted in the relatively poor educational soil of the inner city, environment surely does contribute to their underperformance on IQ tests. Murray and Herrnstein's point, however, was that the discrepancy was so great that environment likely couldn't explain it all. Similarly, environmental factors alone may not account for why, globally, Asians, have on average higher IQs than other racial groups. The idea of measurable variations in average intelligence among ethnic groups is not one, I admit, I want to live with. But though The Bell Curve's claims remain questionable, we should not allow political anxieties to keep us from looking into them further.

  There is perhaps no more heartening proof of the role of environment in human intelligence than the Flynn effect, the worldwide phenomenon of upwardly trending IQ, named for the New Zealand psychologist who first described it. Since the early years of the twentieth century, gains have ranged between nine and twenty points per generation in the United States, Britain, and other industrialized nations for which reliable data-sets are available. With our knowledge of evolutionary processes, we can be sure of one thing: we are not seeing wholesale genetic change in the global population. No, these changes must be recognized as largely the fruits of improvement in overall standards both of education and of health and nutrition. Other factors as yet not understood doubtless play a role, but the Flynn effect serves nicely to make the point that even a trait whose variation is largely determined by genetic differences is in the end significantly malleable. We are not mere puppets upon whose strings our genes alone tug.

  The finding that there is a substantial genetic component to our behavior should not surprise us; indeed, it would be far more surprising if this were not the case. We are products of evolution: among our ancestors, natural selection indubitably exerted a strong influence over all traits that have figured in our survival. The human hand, with its marvelous opposable thumb, is the product of natural selection. In the past, therefore, there must have been varying forms of the hand, with natural selection favoring the version we have today by promoting the spread of the genetic variants underlying it; in this way, evolution ensured that every member of our species would be endowed with this supremely valuable asset.

  Behavior, too, has been critical to human survival, and therefore sternly governed by natural selection. Presumably our enthusiasm for fatty and sweet foods evolved this way. Our ancestors were ever pressed to meet their nutritional requirements; therefore the propensity to take full advantage of all energy-rich foods whenever any became available was of huge benefit. Natural selection would have favored any genetic variations that ensured a sweet tooth since those with it survived better. Today those same genes are the scourge of everyone who struggles to keep off the weight in parts of the world with abundant food sources: what was adaptive in our ancestors is now maladaptive.

  Ours is a strikingly social species; it is logical, therefore, to infer that natural selection once favored genetic adaptations facilitating social interaction. Not only would gestures, like smiling, have evolved as a means of signaling one's state of mind to other members of the group, but presumably there would also have been strong selective pressures in favor of psychological adaptations permitting one to judge the intentions of others. Social groups are prey to parasitism; there are always individuals who seek to benefit from membership without contributing to the general good. The capacity to detect such freeloaders is vital to the success of a cooperative social dynamic. And though we are no longer hovering in small groups around one fire roasting the communal supper, our gifts for sensing one another's moods and motivations may nevertheless come from those early phases of our development as a social species.

  Since the publication of E. O. Wilson's Sociobiology in 1975, evolutionary approaches to understanding human behavior have themselves evolved, giving rise to the modern discipline of evolutionary psychology. In this field the search is for the common denominators of our behavior – human nature, the characteristics shared by all of us, whether New Guinea Highlander or Parisienne – which we seek to understand, trait by trait in relation to some past adaptive advantage conferred by each (see Plate 61 & 62). Some such correlation
s are simple and relatively uncontroversial: the grasping reflex of a newborn, for instance, strong enough that a baby can use its hands and feet to suspend its full body weight, is presumably a legacy from the time when the ability to cling to a hirsute mother was important for infant survival.

  Evolutionary psychology does not, however, limit its scope to such mundane faculties. Is the relatively low representation of women in the mathematical sciences worldwide a universal fact of culture, or might eons of evolution have selected male and female brains for different purposes? Can we understand in strictly Darwinian terms the tendency of older men to marry younger women? With a teenager likely to produce more children than a thirty-five-year-old, might such men be seen as succumbing to the power of evolutionary hardwiring that urges each of them to maximize the number of his offspring? Similarly, do younger women go for wealthy older men because natural selection has operated in the past to favor such a preference: a powerful male with plenty of resources? For now any answers to these questions are mainly conjectural. As we discover more of the genes underpinning behavior, however, I am confident that evolutionary psychology will migrate from its current position on the fringes of anthropology to the very heart of the discipline.