by Alfie Kohn
Faced with such evidence, proponents of competition can no longer use catharsis to justify the aggressiveness of sports. Their last refuge is, as usual, the myth of “human nature.” Michael Novak, for instance, asserts that “the human animal is a warlike animal” and that sports merely “dramatize conflict.”47 But whether or not we are unavoidably aggressive—and the data suggest that we are not—one cannot argue in good faith that sports merely dramatize conflict. The studies demonstrate that athletic competition not only fails to reduce aggression, as catharsis theory would predict, but actually encourages it. This is not really surprising given that sport represents a kind of circumscribed warfare—something pointed out not only by such critics as George Orwell, who called it “war minus the shooting,”48 but also by generals: It was Wellington who said that the battle of Waterloo was “won on the playing fields of Harrow and Eton.” It was Douglas MacArthur who said: “Upon the fields of friendly strife are sown the seeds that, upon other fields, on other days, will bear the fruits of victory.” And it was Eisenhower who said that “the true mission of American sports is to prepare young people for war.”49 The point is not that athletes will rush to enlist, but that athletic competition both consists in and promotes warlike aggression.
There have been numerous anecdotal and experimental accounts of the relationship between violence and sports, but probably the most famous investigation was the series of studies conducted between 1949 and 1954 by Muzafer Sherif and his colleagues. In the so-called “Robbers’ Cave” experiment (named after the area of Oklahoma where it took place), the researchers took a group of normal eleven- and twelve-year-old boys at a Boy Scout camp and divided them into two teams. These teams, the Rattlers and the Eagles, lived for three weeks in separate cabins and were pitted against each other in such competitive games as baseball, football, and tug-of-war—with prizes for the winning team. The hypothesis was that situations where one group could be successful only at the expense of the other (i.e., competition) would promote generalized hostility and aggressive acts. This is exactly what happened. The boys began taunting and insulting each other, in some cases turning against good friends who were now on the opposing team. They burned each other’s banners, planned raids, threw food, and attacked each other after the games and at night. It is important to realize that there was no difference between the members of the two teams; they were in all respects homogeneous. Only the fact of structural competition can account for this hostility.50
The practice of dividing children into teams for a series of competitive encounters is still common in summer camps. These teams often are identified by colors, and the affair is aptly known as “Color War.” It is a dreadful spectacle, a study in humiliation and rage, which I witnessed over several years as a camp counselor. Even very young children understand that the only thing that matters during the War is the relative standing of their team. Everything must be sacrificed for the Blues or Whites, and fervent loyalties develop as soon as the arbitrary team assignments are announced. Erstwhile friends on the other side are now met with a coldness that often erupts into nastiness. In the camp where I worked, the competition extended beyond sports: writing cheers and participating in a sort of quiz bowl insured that nonathletic youngsters, too, could use their skills in rivalrous fashion. (Indeed, one regularly finds hostility boiling up in chess matches, interscholastic debates, and any other sort of recreational competition one cares to mention. Athletes have no patent on aggression.)
Competition does not promote aggression merely on the part of participants. Fan violence is a frequent companion to sports, from high school students pelting the opposing team’s bus with rocks to the death of three hundred soccer fans in a 1964 brawl in Peru. In 1971, sixty-six people died in similar fashion in Glasgow; in 1985, it was thirty-eight in Brussels. Looting and rioting regularly occur in U.S. cities following a hometown victory in the Super Bowl or World Series. After each such incident, pundits and political leaders scratch their heads and try to imagine what could have caused such “senseless” behavior. The Brussels riot, begun by Liverpool youths, produced hypotheses ranging from alcohol to the British character. The one cause that was not considered was the effect of competition itself.51 In any case, the frequency of such behavior on the part of fans again disproves the catharsis theory. “There are so many cases of spectators becoming violent as a result of an emotionally pitched game,” says Terry Orlick, “that we have to wonder why the notion persists that the viewers will lessen their aggressive inclinations by seeing the game. Clearly someone forgot to tell these fans that watching highly competitive or aggressive sports is supposed to subdue their aggressive tendencies.”52
***
It would be a mistake to confine a discussion of competitively inspired aggression to sports. Games, after all, are supposed to matter less than the rest of life; they are offered as something playful and fun. In other arenas, where competition is in deadly earnest, there may be fewer displays of brute violence but there is at least as much hostility. Joseph Wax’s reflections on education are worth quoting at length;
One must marvel at the intellectual quality of a teacher who can’t understand why children assault one another in the hallway, playground, and city street, when in the classroom the highest accolades are reserved for those who have beaten their peers. In many subtle and some not so subtle ways, teachers demonstrate that what children learn means less than that they triumph over their classmates. Is this not assault? . . . Classroom defeat is only the pebble that creates widening ripples of hostility. It is self-perpetuating. It is reinforced by peer censure, parental disapproval, and loss of self-concept. If the classroom is a model, and if that classroom models competition, assault in the hallways should surprise no one.53
We are back to the inescapable fact of generalization: what we are taught is appropriate in one situation does not remain confined there. Competition spreads to other arenas, assumes other forms, and along with it—as part of a package deal—goes aggression. One interesting bit of evidence here relies on an empirical tool known as the F scale that was devised by the social theorist Theodor Adorno and his colleagues some years ago. People whose answers to certain questions led to their being classified as high on this scale were said to be aggressive as well as destructive, cynical, authoritarian in their thinking, and enamored of power and “toughness,” among other things.54 Harold Kelley and Anthony Stahelski, in an influential 1970 article based on Prisoner’s Dilemma research, concluded that “persons low on the F scale and similar personality measures are like the cooperative persons identified in the present game research [while high scorers on the F scale] seem to correspond to the competitive type.”55
I cite this correlational finding because it pertains to long-term personality patterns, something about which (at least with respect to competition) we have little data. As we have already seen, though, there is good evidence of a causal link between competition and aggression. Some have explained this evidence in terms of the theory that aggression results from frustration. Originally, the connection was proposed in just such simple terms;56 today, the paradigm has been refined. Many theorists propose that competition generates a high level of arousal, meaning that we may not immediately become aggressive as a result, but that we are predisposed to respond in this way if we are then frustrated by something. Of course most of us encounter frustrations all the time, so the competition-aggression link remains fairly solid.
This theory neatly explains why losers become aggressive, but it raises the question of why winners (or fans of winning teams) do so as well. In an experiment with five- and six-year-olds, Janice Nelson and her associates found that “success as well as failure in competition produced consistent increases in aggression, as compared with the effects of noncompetitive play,” although failure made the children more aggressive.57 Another study, which first involved watching an adult be aggressive, discovered that the boys who won a subsequent competition were more aggressive than those who fai
led.58 Such findings suggest two possibilities: (1) competition generates aggression as a result of something other than its capacity to frustrate participants, or (2) even winning is not enough to eradicate the frustrating elements of competition. The authors of the second study prefer the latter hypothesis: “Whether successful or not, competition is considered a frustrating experience because of the threat of defeat and the unpredictability of the outcome.”59 In any case, it is quite clear that the hostile encounter called competition—on the playing field and in other contexts, for both participants and spectators—leads us to become more aggressive.
RX: COOPERATION
Competition is the worst possible arrangement as far as relationship is concerned. There may be drawbacks to working or playing or learning independently, but at least this does not place us in the position of viewing each other as enemies. Happily, though, we do not have to settle for the lesser evil of isolation. There is a third approach—cooperation—which has a powerful positive effect on relationship. Chapter 3 showed that working together is more productive. Now I want to emphasize a somewhat less surprising but equally consequential finding: when we cooperate, we are inclined to like each other more.
Even in a mercilessly competitive society, there are pockets of cooperative activity—enough, at least, so that each of us knows what it is to work with others to paint a room, prepare a report, cook a meal. To remember such experiences is to know that cooperation encourages us to view our collaborators favorably; it is to understand how cooperation teaches us, more broadly, the value of relationship. Cooperation means that the success of each participant is linked to that of every other. This structure tends to lead to mutual assistance and support, which, in turn, predisposes cooperators to feel an affinity for one another. At the very least, cooperation offers an opporunity to interact positively (which independent effort does not and which competition actively discourages); at the most, it provides an irresistible inducement to do so.
There is by now an impressive pile of studies substantiating these conclusions, but their findings seem so predictable that I will review them only briefly. David and Roger Johnson conducted a precise statistical analysis of ninety-eight studies on this issue published between 1944 and 1982. Their collective finding:
Cooperative experiences promote more positive relationships among individuals from different ethnic backgrounds, between handicapped and nonhandicapped individuals, and more homogeneous individuals than do [the alternative arrangements:] cooperation with intergroup competition, interpersonal competition, and individualistic experiences.60
As of 1985, the Johnsons themselves had conducted thirty-seven studies of interpersonal attraction under different learning arrangements. Thirty-five of them clearly showed that cooperation promoted greater attraction, while the results were mixed in the other two.61
“Interpersonal attraction” is a catchall phrase that stands for quite a number of positive effects. Among them:
• ENCOURAGEMENT GIVEN: Research by Brenda Bryant discovered that children who worked cooperatively “more actively encouraged and supported the self-enhancement of others than children from a competitive or individualistic learning environment.”62
• ENCOURAGEMENT RECEIVED: Other studies have shown that students perceive such encouragement and support from their peers in a cooperative setup.63
• SENSITIVITY: Gillian King and Richard Sorrentino found more “sensitivity to the needs of others” displayed in cooperation; their experimental subjects also reported that cooperative situations were much more pleasant than competitive situations.64
• OTHER-ORIENTATION: Children can more readily move beyond a self-centered orientation and begin to take others into account once they are placed in a cooperative environment, reported Emmy Pepitone and her colleagues.65
• PERSPECTIVE-TAKING: Cooperation, as I noted earlier in this chapter, is by far the best way to promote perspective-taking. People tend to see things from the other person’s point of view when they are working with, rather than against, each other.
• COMMUNICATION is improved through cooperation. Deutsch’s classic experiment with undergraduates showed that when students worked cooperatively, “more ideas were verbalized, and members were more attentive to one another. . . . They had fewer difficulties in communicating with or understanding others.” (With competition, Deutsch adds, communication tends to be “unreliable and impoverished.”)66
• TRUST: Even when individuals have unequal power (as represented in an experimental game), they tend to trust each other in a cooperative structure. Trust is virtually absent under competition.67
Some of the most exciting research in this field concerns the effect of cooperation on students who are different from one another. It says quite a bit if any two people will come to like each other as a result of having cooperated on something. But it says far more if the same thing happens with people of different abilities or ethnic backgrounds. Research from the 1940s and 1950s suggested that “interaction within a cooperative context [was] a major determinant of whether cross-ethnic contact produced positive attitudes and relationships,” and later work has confirmed this.68 Simply bringing together children from different backgrounds does not produce harmony, as a chorus of bitter voices never tires of pointing out. But instead of simply indicting desegregation (or, worse, abandoning it), it seems more sensible to investigate what happens to students once they are in the same classroom. Whereas competition creates an atmosphere of hostility and does nothing to overcome differences, cooperation builds bridges. Its capacity for encouraging positive regard is no less potent when the cooperators are from different backgrounds, as the studies show quite clearly.69 And this effect is not limited to class projects: once cooperative learning brings children together, they continue to enjoy spending time with each other. After the lessons are over, students in cooperative classrooms socialize more heterogeneously than those in competitive or independent learning situations.70 In all, cooperation points the way to what child development researcher Urie Bronfenbrenner once called a “curriculum of caring.”
Implicit in all of this is a truth that many people may find disconcerting. How we feel about other people is often put down, rather vaguely, to “chemistry”—not a subject for social science research, but a beguiling mystery to be wondered at by poets and songwriters. Even if we are willing to scrutinize patterns in our relationships, we tend to see them as a function of other people’s behavior and personality. (We like people who are likable.) But the implication of what I have been reporting in this chapter is that our attitudes about others are largely dependent on the structure in which we interact. I do not mean to suggest that an individual’s actions are irrelevant to how I regard her. Neither do I believe that science can (or should) eliminate all mystery from what happens between two people—or reduce the uniqueness of each relationship to generalizable laws. Still, there is no escaping the fact that the texture of our relationships depends to a significant degree on the context in which we come to know each other. I will look very different to someone for whom I am a rival than to someone for whom I am a partner. Of course, not all competitors will be lifelong enemies, just as not all cooperators will develop enduring friendships. But a predisposition toward hostility or attraction undeniably develops as a result of the structure under which we deal with one another. That is what the evidence—and, if we think about it, our own experience—demonstrates.
WE VS. THEY
The benefits of cooperation are so compelling that it seems even a competitive society must take notice. But how can such a society allow its members to taste the fruits of working together while still socializing them to keep the larger competitive framework? The answer is intragroup cooperation with intergroup competition. Work with some others, but do so in order to defeat everyone else. This is about as close as some of us ever get to genuine cooperation. I would argue that it represents a rather unsatisfying compromise—better than competition at every level
but not as good as competition at no level. The conventional wisdom in a competitive culture, by contrast, is that intergroup competition heightens—or perhaps is even required for—cooperation within the group. Whatever is to be gained by cooperating depends on working to defeat a common enemy. Every “We” needs a “They.”
Is this true? Recall that when studies found performance was enhanced by intragroup cooperation, people asked whether this effect actually relied on intergroup competition. The answer was unequivocally no ([>]). Such competition either was irrelevant to the cooperation-generated productivity or actually diminished it. The same seems to be true here. Intergroup competition is not necessary for intragroup cooperation to draw people together. Robert Dunn and Morton Goldman set out to test this question with college students in 1966. They found, first, that intergroup competition led subjects to view those from other groups negatively, and, second, that intergroup competition was not helpful in developing feelings of acceptance within each group. Such rivalry, they concluded, “may not only be unnecessary, but may do social harm through the intergroup tensions that it arouses.”71 The Johnsons, too, paid attention to this issue when they analyzed the results of scores of studies on interpersonal attraction. They found that subjects tended to like each other more where cooperation did not take place in the context of intergroup competition.72 Finally, a 1985 study confirmed that
intergroup cooperation promoted more positive cross-ethnic and crosssex relationships than did intergroup competition. These results dis-confirm the position that competition among groups leads to attraction among collaborators . . . and they provide some support for the position that the more pervasive the cooperation, the greater the interpersonal attraction.73