No Contest: The Case Against Competition

by Alfie Kohn

  And what of those who already incline toward competition? Why do they persist in their ways? Kelley and Stahelski found that whereas cooperative individuals realistically perceive that some people are cooperative like themselves while others are competitive, competitive individuals believe that virtually everyone else is also competitive.57 Thus we have another self-fulfilling prophecy: competitive people (falsely) assume that all others share their orientation—and, indeed, those who declare most vociferously that “it’s a dog-eat-dog world out there” usually are responsible for more than their share of canine consumption—which impels them to redouble their own competitive orientation. This finding has been replicated by several other researchers.58

  Both the desirability of competition and the strategies for implementing it, then, are taught to us from our earliest days. This orientation then reproduces itself. What is peculiar, as Riesman pointed out, is that we should have to contrive such a thorough program of socialization if competition really were part of human nature. (Typically it is those who argue most strenuously for the latter who are, at the same time, vigorously promoting this training process.) Far more plausible is the hypothesis that all of this training is not superfluous: we compete only because we have learned to do so.

  In order to prove this hypothesis correct rather than simply arguing for its plausibility, one would need some evidence that children can be taught to cooperate. If they take to this approach—learn it easily, enjoy it, continue using it—it would be evident that competition is not inevitable. Longitudinal studies of this kind have not been performed, to my knowledge—unless one regards cross-cultural data as naturalistic evidence along these lines. However, some work in early childhood education comes close.

  A few years ago, Gerald Sagotsky and his colleagues at Adelphi University successfully trained 118 pairs of first- through third-grade students to cooperate in a series of classroom games. This was done through a combination of direct instruction and modeling (having them watch someone else be reinforced for engaging in the desired behavior). About seven weeks later, a new experimenter tried a new game with these children and found a significant retention and generalization of the cooperative approach, particularly among the older children. “Overall, the study indicates that a relatively brief and straightforward intervention can effectively train cooperation.”59 Similar generalization has been reported on the part of fifth graders60 and, in still another study, third graders.61 Earlier and more primitive research found that children continued behaving cooperatively even when reinforcement was briefly withdrawn.62 More recently, Aronson not only had remarkable success with a cooperative learning technique but also discovered that teachers continued to use it years after the experiment was over.63 Deutsch reported that adults, too, taught themselves to cooperate when the game they were playing rewarded such behavior.64

  A somewhat more anecdotal account of this effect is provided by David N. Campbell, who reports on his visit to an unnamed British elementary school. When another American teacher on the tour asked these children who was the smartest among them, they “didn’t know what he was talking about. They had evidently never thought about it. . . . There were no put-backs, grades, tests, gold stars. All stories and drawings were displayed on the walls. Children were not placed in failure situations, forced to prove themselves, to read at ‘grade level’ every week.” When he returned home, Campbell resolved to make his own classroom less competitive.

  It required only about three weeks for the changes to emerge [he continues]. The first was an end [to] the destruction of others’ work. Later a spirit of cooperation and help began to be common. Finally there was what I look for as the real measure of success: children talking freely to every adult and stranger who walks in, leading them by the hand to see projects and explaining their activities, no longer afraid, suspicious, or turned inward. Such changed attitudes developed because we stopped labeling and rank-ordering.65

  Some of the most interesting work with recreational cooperation has been done by Terry Orlick at the University of Ottawa. After leading children from preschool age to second grade in cooperative games, he found a threefold to fourfold increase in the incidence of cooperative behavior when the children were later left to play by themselves. Control groups, meanwhile, tended to become more competitive as the year progressed.66 Orlick also found that children reported being happier playing cooperative games: “Given the choice, two thirds of the nine- and ten-year-old boys and all of the girls would prefer to play games where neither side loses rather than games where one side wins and the other side loses.”67 In the classroom, meanwhile, 65 percent of a group of sixth graders said they preferred a cooperative learning structure. The total sample here was divided evenly between those who tended to attribute their successes or failures to themselves (“internalizers”) and those who attributed consequences to fate or other people (“externalizers”). Significantly, a majority of both groups preferred cooperative learning.68 As of 1984, a pair of researchers were able to cite seven studies showing a preference for cooperative over competitive or independent experiences.69

  I am aware of no studies that found a preference for competition over cooperation—providing the subjects had experienced the latter in some fashion. This qualification is critical: it is not unusual for people to say they prefer to compete but then to change their minds when they see at first hand what it is like to learn or work or play in an environment that does not require winners and losers.70 This, incidentally, has been found to be true of college students, too.71 The evidence seems persuasive, then, that children can learn to cooperate and, when given a choice, seem to prefer this cooperative arrangement.

  The evidence is far less clear, however, on the question of how old children must be in order to learn to cooperate. The traditional position, following Piaget’s developmental schedule, is that children cannot cooperate or compete in any meaningful way until they are about six or seven years old and have reached the “operational” stage.72 This position is compatible with the usual movement of children toward increasing competitiveness. What precedes competition, it is argued, is not positive cooperation but merely the absence of competition (two very different things)—or, for that matter, the absence of any sophisticated goal-directed activity. The ability to cooperate and to compete are said to develop around the same time.73

  The fact that very young children do not compete, then, is not terribly significant. Young children do not have body hair either, but that doesn’t mean that body hair is a learned phenomenon. On the other hand, the fact that children normally become more competitive as they grow older74 also means very little because most children have not been exposed to models of cooperative interaction.

  Yet not everyone is agreed that young children cannot cooperate. Spontaneous pro-social behavior takes place in infants, as we saw before.75 Anna Freud observed nineteen-month-old toddlers building a tower together, taking turns adding blocks.76 This could mean that Piaget and his followers are wrong. But even if we insist that such collaborative or helping behaviors do not qualify as genuine cooperation, there is evidence that children of only four or five—past the toddler stage but not yet “operational”—can indeed be taught to cooperate. In fact, Orlick has found “young children to be most receptive to cooperative ventures and cooperative challenges.” He proposes that this is true because “the younger the child, the less time he has had in the competitive mainstream of our society and therefore the more willing he is to accept cooperative games.”77 Millard C. Madsen’s research seems to support Orlick’s view, at least within this country. In presenting an experimental problem that required a cooperative solution, he found that “more younger than older children are successful in solving the problem in such a way as to maximize reward.”78 Sagotsky, on the other hand, found that older children (seven- or eight-year-olds) learned to cooperate more readily than younger ones, which is what Piagetian theory would predict.79 But all of these researchers agree—and the whol
e of this section confirms—that cooperation can be learned; thus, competition is by no means inevitable.

  LIFE IN OTHER CULTURES

  Even in the era of the “global village,” as Marshall McLuhan characterized the interdependence of the modern world, ethnocentrism remains. It is present in two forms: as a value judgment that unfamiliar cultural practices are ipso facto inferior (“if it’s different, it’s worse”), and as an empirical belief that what is present in one’s own land must be universal (“if it’s here, it’s everywhere”). I am concerned with the latter, and specifically with the assumption, usually unstated, that because competition is so pervasive in these parts it must be pervasive everywhere. If this assumption is wrong, then competition is learned and it is not inevitable.

  In chapter 1, I suggested that the United States appears to be uniquely competitive. This observation also has been made by researchers who have observed other cultures and/or tested their inhabitants. Anthropologists Beatrice and John Whiting recorded the frequency of such acts as touching, reprimanding, offering help, insulting, and so on, in six cultures, one of which was in the United States (a small New England town). As a proportion of total acts observed, the latter scored lowest on offering help.80 Another researcher compared the way people in four Scandinavian countries, England, and the United States described child-rearing orientations and norms of masculinity in their respective nations. On both subjects, she found a far greater emphasis on competition in the United States.81 In experiments with Anglo-American and Mexican children, two psychologists discovered that the former “tended to remain in conflict even when to do so prevented them from getting as many toys as possible”—a tendency that was rightly characterized as irrational competition. When permitted to do so, these children also took away another child’s toy even though they had nothing to gain from this act.82

  These data concerning the competitiveness of Americans* are particularly useful because they are being contrasted with findings from other cultures. Let us now examine some of these other cultures in greater detail by turning to the reports of anthropologists and other cross-cultural observers. The United States, as we shall see, is appreciably more competitive than many other cultures; in fact, some cul tures appear to be entirely noncompetitive.

  Consider first the case of primitive cultures, which many people (relying on caricatured images of growling cavemen) associate with fierce competition. As with ideas about the animal world, this perception is largely mistaken. Prehistoric people actually were remarkably cooperative, and in fact may have distinguished themselves from other primates precisely by virtue of the extent of their cooperativeness. A growing number of anthropologists are concluding that cooperation—not brain size or the use of tools, and certainly not aggressiveness—defined the first humans.83 The distinguished biologist George Edgin Pugh wrote:

  Primitive human societies differ greatly from other primates in the amount of cooperation that is achieved. Within primitive human societies “sharing” is a way of life. . . . The sharing is not limited to food, but extends to all types of resources. The practical result is that scarce resources are shared within the societies approximately in proportion to need.84

  Through the medium of kinship, early humans developed cooperative arrangements that, according to Marshall Sahlins, were apparently mandated by virtue of the conditions of life. In his words, “The emerging human primate, in a life-and-death economic struggle with nature, could not afford the luxury of a social struggle. Co-operation, not competition, was essential. . . . Hobbes’s famous fantasy of a war of ‘all against all’ in the natural state could not be further from the truth.”85 Substantiation for this conclusion is provided by extant hunter-gatherer societies, such as the Congo pygmies, Kalahari Bushmen, Australian Aborigines, and the Waoranis of the Amazon, all of which are overwhelmingly cooperative.86

  The fact that some of these cultures still exist today invites a wider consideration of contemporary noncompetitive societies that cast our own competitiveness into sharp relief. It was Margaret Mead and her associates who first attended to this characteristic: Cooperation and Competition Among Primitive Peoples (1937) described several such cultures in some detail. These include:

  THE ZUÑI INDIANS—“The orientations of all institutions, with little exception, to a basic principle of cooperative, nonindividualistic behavior is the pattern of Zuñi culture.”87 Possession of material goods is not seen as desirable; wealth circulates freely, and there is therefore no competition in the economic sphere. The major recreation cum religious ritual is a ceremonial four-mile footrace. Anyone can participate, the winner receives no special recognition, and his name is not even announced. In fact, someone who has consistently won is prevented from running.88

  THE IROQUOIS INDIANS—“Beyond the degree of cooperation required to achieve the greatest efficiency in production, there was found, especially in agricultural activity, cooperation for the purpose of experiencing the pleasures of group work.”89

  THE BATHONGA—“Bathonga society is highly cooperative within the bounds of the village, and in all other social and economic relations it is essentially noncompetitive. . . . In economic organization, in technology, in social relations little range is given to any expression of competition.”90

  On the basis of a dozen cultural studies that comprise the book, Mead writes that.

  the most basic conclusion which comes out of this research [is] that competitive and cooperative behavior on the part of individual members of a society is fundamentally conditioned by the total social emphasis of that society, that the goals for which individuals will work are culturally determined and are not the response of the organism to an external, culturally undefined situation.91

  Cross-cultural research in this vein has continued and expanded in the half century since Mead’s study was published. This research substantiates her conclusion by documenting significant differences in competitiveness among people of different cultures. Consider:

  BLACKFOOT INDIAN children cooperated far more effectively than urban Canadian children in a series of experimental games, and this occurred whether they were being rewarded collectively or individually.92

  ISRAELI KIBBUTZ children cooperated far more effectively than urban Israeli children, the latter being “unable to stop their irrational competition . . . even though they obviously realized . . . [this was] not paying off for any of them.” What’s more, the kibbutzniks spontaneously arranged to divide their prizes equally among the members.93 In later experiments that replicated this result, “Kibbutz groups characteristically demonstrated a high degree of organization . . . in sharp contrast to the unrestrained and unorganized tug of war that was typical of the city groups.”94

  KIKUYU children from Kenya cooperated more effectively than American children at an experimental game.95

  RURAL MEXICAN children were more cooperative than Mexican-Americans, who were in turn more cooperative than Anglo-Americans. The latter, as noted above, tended to compete even when the situation was arranged to reward cooperation and tended to take away another’s toy for sheer spite twice as often as did the Mexican children.96 Americans often missed the cooperative solution, “spontaneously declaring], ‘This game is too hard,’ or ‘No one can win.’ When asked after the experiment how they might have gotten some toys, competitive subjects most often responded, ‘If I could play alone,’ or ‘If I could move more than once [not take turns moving].’”97 Among the reasons proposed for this significant cultural disparity is the fact that “rural Mexican mothers tend to reinforce their children noncontingently, rewarding them whether they succeed or fail, whereas Anglo-American mothers tend to reinforce their children as a rigid function of the child’s achievement.”98 The idea that children should be accepted and loved unconditionally—rather than in proportion to the number of others they have beaten at something—is a very peculiar idea to many Americans.

  THE MIXTECANS OF JUXTLAHUACA, MEXICO, “regard envy and competitiveness as
a minor crime.”99

  THE TANGU OF NEW GUINEA eschew competitive games, preferring one called taketak in which two teams spin tops. The objective of the game is to reach an exact draw.100

  THE INUIT OF CANADA live with virtually no competitive structures. Their recreation, like their economic system, is cooperative.101

  AUSTRALIAN ABORIGINES display a marked preference for “cooperative action,” and one experiment found that they cooperated just as readily with members of other tribes as with their own tribe members.102

  NORWEGIANS almost always responded in kind to cooperative behavior in an experimental game, whereas American subjects did so only about half the time.103

  JAPANESE education is said to be far less competitive than ours. “The Japanese have always been inventive in devising ways of avoiding direct competition,” Ruth Benedict reported in the mid-1940s. “Their elementary schools minimize it beyond what Americans would think possible.”104 As of 1980, this was still the case: “Teachers . . . try to create balanced groups composed of people with diverse abilities, and they encourage the students to help each other.”105 And from yet another source, in 1982: “Challenging problems are posed to entire classes, whose members are encouraged to talk to, and to help, one another and are allowed to make mistakes; at times, older children visit the classrooms and aid the younger ones.”106