No Two Alike

by Judith Rich Harris

  Evolutionary biologists and network theorists refer to them as “self-organized systems.” A school of fish, for instance, is a self-organized system. The school moves around in a coordinated way that makes it look as though it has a leader, but there is no leader. Each fish is simply doing what a fish of that species does: responding to environmental cues (moving toward food or away from danger) and, at the same time, taking account of the position of its neighbors. The pattern they form—so pleasing to the eye—results from the fact that the distance between them remains fairly constant as they swim this way and that. To maintain the pattern only two rules are needed: don’t get too close and don’t get too far away. The definitions of “too close” and “too far” are quite precise and vary by species.32

  A school of fish is a social organization maintained by stereotyped behavior. Natural selection acts upon individual fish to produce the stereotyped behavior; in this way, evolution can produce patterns that are larger than the individual fish. Natural selection shapes the patterns by tuning the rules that produce the stereotyped behaviors. Evolutionary biologists call this “multilevel selection”: the fitness of the individual fish depends not only on how well it does relative to other fish, but also on how well its school does relative to other schools.33

  The individual fish in a school do not recognize one another; they do not need to. But the individual chickens in a pecking order do recognize one another. If you form a new flock by putting together ten unacquainted hens or roosters, a power struggle will begin immediately. A pair of birds will peck at each other, or threaten to do so, until one concedes defeat, thus establishing a dominant-submissive relationship between them. Once that has been established there is no longer any need to fight: the one with lower status simply gives way to the one with higher status, and peace reigns. The flock has become a smoothly functioning superorganism. Dominance hierarchies are beneficial not only to the winners (who thereafter get first access to food and mates) but also to the losers (who thereafter get beaten up less often and less severely). Both winners and losers get the advantages of being members of a group. An isolated bird is a sitting duck, so to speak, for predators.34

  The pecking order in chickens depends on their recognizing each other and remembering their past encounters, but their memories are nowhere near as good as that of the proverbial elephant, the one who never forgets. Chickens get mixed up if their flock numbers more than ten (dominance hierarchies are unstable in larger flocks), and a flockmate that takes a vacation is soon forgotten. If you remove a chicken from the flock and put it back in a week, it will resume its place in the hierarchy, but after a three-week absence it will have to reestablish its rank all over again.35 So a chicken’s mental lexicon for chickens doesn’t contain many pages and the ink fades quickly. But who would have expected a chicken to have a mechanism for recognizing individuals and behaving appropriately to them?

  Now let’s take another step downwards in brain size. How about paper wasps? Yes, paper wasps, which live in colonies and construct communal nests (those papery gray things you sometimes see attached to tree branches or the eaves of houses) have dominance hierarchies. Colonies are founded anew each spring by females that survive the winter; several females that overwinter in the same hideout will cooperate in starting a new colony. But the cooperation is at first of a belligerent sort: in the early days they interact aggressively with one another. Soon a dominance hierarchy emerges and there is a sharp reduction in fighting.

  Among paper wasps, the payoff for being the alpha female is considerable: she gets to be the chief egg layer of the colony. The others lay an occasional egg, but Alpha tolerates no nonsense: if she spies an egg that isn’t hers, she eats it. Before long the ovaries of the other wasps recede and they stop trying. They become workers in the nest, helping to rear Alpha’s children. They stay because a paper wasp can’t survive on its own and because there is always the possibility that Alpha might die and they can move up in the hierarchy (in which case their ovaries will grow back). Also, there is a good chance that Alpha is their sister, so the young they help to rear might be their nieces and nephews.36

  Believe it or not, there is evidence that paper wasps are capable of some sort of limited recognition of individual nestmates.37 But recognition, strictly speaking, isn’t necessary: dominance hierarchies in this species may be established and maintained by means of feedback loops.

  Many biological systems use feedback loops to accomplish a goal. The homeostatic mechanisms that control body temperature in warm-blooded creatures make use of negative feedback loops: too much of something triggers a process that reduces it. Too much warmth triggers reactions such as sweating that reduce body temperature. The two commandments of fish schools—don’t get too close and don’t get too far away—are negative feedback rules. But the commandment that formed the school in the first place—if you see a bunch of fish of your kind, join them—is a positive feedback rule. In positive feedback loops, the rich get richer: having a lot of something triggers a process that increases it.

  Here’s how positive feedback loops could produce dominance hierarchies in paper wasps. Assume that winning an aggressive encounter causes some change—hormonal, perhaps—in the winner. This change signals her status to potential rivals and thereby makes her more likely to prevail in future showdowns.38 By behaving in a certain way and/or emitting the right kind of pheromones, she exudes the wasp equivalent of self-confidence. The wasps below her in the dominance hierarchy might be responding to these signals, rather than remembering what happened the last time they made the mistake of dissing Alpha.

  Positive feedback loops of this sort are not restricted to species with wasp-size brains. They are observable in animals that are clearly capable of remembering every member of their group—wolves and monkeys, for example. According to the evolutionary biologist E. O. Wilson,

  The identity of the leading male in a wolf pack is unmistakable from the way he holds his head, ears, and tail, and the confident, face-forward manner in which he approaches other members of his group. He controls his subordinates in the great majority of encounters without any display of overt hostility…. Similarly, the dominant rhesus [monkey]male maintains an elaborate posture signifying his rank: head and tail up, testicles lowered, body movements slow and deliberate and accompanied by unhesitating but measured scrutiny of other monkeys that cross his field of view.39

  The description rings true; we all know guys like that. There are positive feedback loops in humans, too.

  The unhesitating scrutiny with which the alpha male looks at the other monkeys in his group is a mark of his status. In the hierarchical power domain, looking directly at another is a challenge. If two individuals—monkeys, apes, or humans—happen to make eye contact, the lower-ranked one indicates submission by looking down or away. If he maintains eye contact, he’s responding to the challenge by issuing a challenge of his own.40

  In the previous chapter I described a mental module that Simon Baron-Cohen called an “eye-direction detector” and which I will rename, for compactness, the “gaze detector.” The gaze detector tells you where someone is looking—in particular, whether someone is looking at you. What you do with that information depends on what kind of relationship you have with the other person—or, to put it in Bugental’s terms, which domain of social life is currently in play. In the hierarchical power domain, prolonged eye contact means “I challenge you.” In the mating domain, it means “I love you.” In the first case it can lead to a fight; in the second, to sex.

  Briefer glances also have different meanings in the two domains. In the mating domain they are used in flirting.41 In the hierarchical domain, they are a sign of what the ethologist Michael Chance has called the “attention structure” of a group. The alpha male in a primate group is not distinguished by how many glances he directs toward the others but by how many glances he receives: he receives by far the most. In general, high-ranked individuals are looked at more than those in lower
ranks. Because the lower-ranked individual has to yield to a higher-ranked one, he has to keep track of where his superiors are and what they are doing. If these glances should happen to result in eye contact, he quickly averts his eyes.42

  Rank in primate groups is not a simple matter; several factors are involved in determining an individual’s status. Males and females generally have separate hierarchies, with all or most of the males able to dominate all or most of the females. Another important factor is kinship. An individual who has high-status relatives in the group can call on their help in a dispute with the members of other families.

  It pays to belong to a powerful family. A recent study of baboons living in a game reserve in Botswana showed that these monkeys are keenly aware both of family connections and of power hierarchies within their troop. The researchers made auditory tapes of the baboons’ noisy altercations and then pieced together sequences that sounded (to a baboon) as if one animal was asserting dominance and the other was yielding. The baboons who heard these phony sequences appeared to be more disturbed or puzzled (judging by how long they looked in the direction of the sounds) by sequences that involved status reversals—either a low-status animal getting the best of a high-status one from the same family, or an animal from a low-status family getting the best of one from a high-status family. But they were clearly more troubled by the latter type of reversal. The researchers concluded that baboons classify other baboons both by individual rank and by family, and that they understand that changes in the rank ordering of families are more disruptive.43

  The results of this experiment also show that baboons can recognize individual members of their troop by the sound of their voices as well as by sight. Either form of recognition will lead them to the lexicon page for that particular baboon—a page that contains, not only information about their own kinship with that individual, but also information about its kinship with others in their troop.

  Now we come to the chimpanzee. Dominance in this species depends not only on physical power and family connections, but also on alliances between nonrelatives. When two chimpanzees come to blows, explained the Dutch primatologist Frans de Waal,

  A third ape may decide to enter the conflict and side with one of them. The result is a coalition of two against one. In many cases the conflict extends still further, and larger coalitions are formed…. Chimpanzees act selectively when intervening in a conflict between other members of the group. All the group members have their own personal likes and dislikes which dictate how they act. The choices they make are biased choices, which generally remain constant over the years. This does not mean to say that relationships in the group do not change; indeed, this is the most fascinating aspect of chimpanzee coalitions. Why should C, who has supported A against B for years, gradually begin to support B against A?44

  Something like that happened in the chimpanzee colony in the Arnhem Zoo in the Netherlands, where de Waal spent several years observing the primates. When he arrived at the zoo, the alpha male, A, was a dignified old chimpanzee named Yeroen, still physically powerful but beginning to show signs of age. B was Luit, a little younger but equally big and strong, and C was Nikkie, the youngest and most boisterous of the three, still gaining in size and strength.

  For more than two years, A was the top banana. Then B began to challenge his supremacy. C sided with B and their coalition was successful in overthrowing A, so that both B and C became dominant over A. But now that A was acting submissive to C, C began a campaign against B and obtained A’s cooperation. With A’s backing, C toppled B and became the alpha male.

  Such changes in status are accompanied by changes in behavior and demeanor. The alpha male looks very self-confident as long as he’s the alpha male but, observed de Waal, “As soon as his position is seriously threatened, the self-confidence may disappear completely.”45 The other chimpanzees are well aware of what is going on. Each adult member of the colony has to know who’s currently on top, who’s plotting a coup, and who’s supporting each contender. The pages in a chimpanzee’s lexicon have to be kept up-to-date.

  Earlier I said that the dominance hierarchy is the result of pairwise interactions between individuals. That is true despite the existence of coalitions of the sort de Waal described. Chimpanzee A may be dominant over B either because he is bigger and stronger or because he has the support of C. In either case the relationship between A and B is one of dominance and submission.

  Dominance hierarchies have also been observed in groups of human children. The cross-cultural psychologist Carolyn Edwards studied the multi-age play groups of traditional small-village societies and reported,

  Older children respond to others younger and smaller than themselves by establishing dominance over them…. Accordingly, a pecking-order of size and strength consistently emerges in multi-age groups, certainly in the multi-age play group of siblings, half-siblings, and courtyard cousins found in most of our samples.46

  In societies where children go to school and children’s groups consist mostly of nonrelatives of roughly the same age, individuals who are taller or stronger than their agemates are likely to have higher status, especially in boys’ groups. Researchers have studied children in nursery schools and found attention structures similar to those reported in other primates: high-status children are looked at more.47

  But in humans—even in boys—the attention structure is not identical to the dominance hierarchy. The child with the highest position in the attention structure is not necessarily the largest or most aggressive; he or she may instead be an organizer or initiator—one who thinks up interesting games and persuades others to participate in them.48 A child can have high status in a group without being at the top of the dominance hierarchy. In humans—I’ll return to this point in a later chapter—status is complex and multidimensional.

  The dominance hierarchy of a group depends on who’s in it at the moment. If A is absent, the remaining members may simply move up a rung or, if coalitions are involved, there may be other repercussions: without A’s support, B may lose his dominance over C. When a group splits up into two smaller groups or two groups coalesce, an individual’s rank is likely to change. In ancestral times, when human families foraged together in small, temporary groups, a child’s position in the dominance hierarchy might shift upwards or downwards whenever the adults decided to make a switch. A boy who could, for a while, dominate all the other boys in the play group might be quickly demoted when a family with a larger, stronger boy joined up. If children in ancestral times sooner or later became acquainted with all the other children in their clan, they must also have learned which ones they could dominate and which could dominate them.

  As de Waal observed, the behaviors associated with being the alpha male depend on remaining the alpha male; the ex-alpha has lost, not only his status, but also his air of self-confidence. Whether one behaves in a dominant or submissive fashion depends on one’s rank in a particular group at a particular time. That is why a child who has been dominated for years by an older sibling can become a dominant member of a group of agemates—or, for that matter, of any group that does not happen to include the older sibling.49

  The relationship system is a discriminator, not a generalizer—a splitter, not a lumper.50 Its purpose is to make fine distinctions among individuals, even between identical twins, so that the system’s owner can behave appropriately toward each individual. Obligations must be repaid, duplicity remembered, compatible companions sought out, obnoxious ones avoided, those with higher status deferred to. Since information keeps coming in, the system has to distinguish between newer and older data and to favor the new: What have you done for me lately?

  The information we collect on individuals is put to use in our dealings with them. Thus, parents behave differently toward each of their children. Thus, children behave differently toward older siblings than to younger ones, and differently toward peers than to siblings. Thus, a man’s relationship with his boss is nothing like his relations
hip with his father. Thus, a woman who falls in love with a man is not sexually attracted to his identical twin.51

  Humans build mental “working models,” not just of their relationships with their parents, but for every individual with whom they interact. On the basis of a brief encounter, we create a new page in our lexicon, and whatever we have so far learned about that person is recorded there. The fundamental attribution error—our tendency to make a judgment about someone’s personality on the basis of a small, perhaps atypical, sample of behavior—is a result of the way the lexicon works. Since its purpose is to tell us how to act if we should encounter this individual again, it makes the best guess on the basis of whatever information it has. The best guess is that this individual will behave in the future the way he or she behaved in the past.52

  Managing relationships is Job 1 for a baby. The relationship system is ready to go, straight out of the starting gate. Human infants are prepared to begin assembling a lexicon of people as soon as they draw their first breath.

  8

  The Socialization System

  ANIMALS ADAPTED by evolution to a group lifestyle are among the most successful on earth, but there is a drawback to this kind of adaptation: the members of these species generally can’t survive on their own. Their lives, along with their chances of reproducing, are dependent on remaining members of the group. Hence the paper wasp at the bottom of the dominance hierarchy remains in the nest and the pecked-upon chicken remains in its flock.