A New Look at the Classic Robbers Cave Experiment

A sports tournament led boys to something like inter-tribal war.

Posted Dec 09, 2009

[Social media counts reset to zero on this post.]

I begin with a research story, a true one.

In the early 1950s, the social psychologist Muzafer Sharif and his colleagues conducted a now-classic experiment, on intergroup conflict and resolution of conflict, with 11- and 12-year-old boys at a summer camp in Oklahoma's Robbers Cave Park.[1] Sherif's procedure involved three phases:

(1) He began by dividing the boys, by a random procedure, into two distinct groups, who slept in different parts of the camp and were given separate sets of chores and activities, so they could develop a sense of group identity.

(2) Then he established conditions designed to induce hostility between the two groups. (Experiments of this sort could be done in the 1950s--a time before the era of research ethics review boards, and a time, before cell phones, when parents did not feel compelled to check up on their camping kids. The boys did not know they were participants in an experiment; they thought that they had been invited to take part in a regular camping experience.)

(3) Once the groups were sufficiently hostile, he tried various methods to reduce the hostility.

The famous result of the experiment--repeated in most introductory psychology textbooks, including my own--was that hostilities were best reduced by establishing superordinate goals, defined as goals that were desired by both groups and could be achieved best through intergroup cooperation. For example, to create one such goal the researchers staged a breakdown in the camp's water supply. In response to this crisis, the boys temporarily forgot their differences and worked cooperatively to explore the mile-long water line and find the break. With each such cooperative adventure, hostilities between the groups abated, and by the end of a series of such adventures the boys were arranging many friendly cross-group interactions on their own initiative.

Sherif's focus in this experiment was on ways to reduce intergroup hostility, but my focus here is on his method for creating the hostility, something not generally discussed in the textbooks. His procedure was remarkably simple. In phase two he invited the two groups of boys to compete with one another in a tournament involving a series of competitive games--including several games of baseball, touch football, and tug of war--all refereed by the camp staff. The members of the winning team would receive prizes, such as pocketknives, that were much valued by the boys. Formal sports conducted for prizes--that was how Sherif and is colleagues generated animosity between the groups. It apparently worked like a charm, not just in this experiment, but also in others that Sherif and his colleagues had conducted earlier.

As the series of games progressed, the two groups became increasingly antagonistic. Initial good sportsmanship gave way gradually to name-calling, harassment, accusations of cheating, and cheating in retaliation. As the hostilities mounted, they spread to camp life outside of the games as well as in the games. Even though the boys all came from the same background (white, Protestant, middle class) and had been divided into groups by a purely random procedure, they began to think of the boys in the other group as very different from themselves--as dirty cheaters who needed to be taught a lesson. Serious fistfights broke out on several occasions. Raids were conducted on the cabin of the opposing group. Some boys carried socks with stones in them, to use as weapons "if necessary." One group pulled down and burned the other group's flag. Many of the boys declared a desire not to eat meals in the same mess hall with the other group; and joint meals, when held, became battlegrounds where boys hurled insults and sometimes food at members of the other group. What at first was a peaceful camping experience turned gradually into something verging on intertribal warfare, all created by a series of formal sporting events.

Formal sports occupy a precarious space between play and reality

Let's step back momentarily from this experiment and reflect a bit on boys' play in general.

Much of boys' play involves mock battles. In some cases the battles lie purely in the realm of fantasy. The boys collaboratively create the battle scenes, decide who will play which parts, and, as they go along, decide who is wounded, or dies, or is resurrected. Some people, who don't understand boys' play, mistake such play for violence and try to stop it, especially when it is acted out in a vigorous, rough-and-tumble manner. But it isn't violence; it's play. We should think of those players not as warriors but as junior improvisational Shakespeares. They are using their imaginations to create and stage dramatic, emotion-inspiring stories. Play of this sort is non-competitive as well as nonviolent. No score is kept; nobody wins or loses; all are just acting out parts. There are also no fixed teams in play of this sort. If the play involves pretend armies, the players arrange the armies differently for each bout of play. Such play does not create enemies; rather, it cements friendships.

A step removed from such fantasy battles is the informal play of team games such as baseball, soccer, and basketball--games that are referred to as "sports" when played formally. These games, too, can be thought of as mock battles. There are two teams (armies), who invade one another's territory, defend their own territory from invaders, and strive to conquer one another, all ritualized by the rules of the game. By "informal" play of these games, I mean that the games are organized entirely by the players and have no obvious consequences outside of the game context. There are no trophies or prizes, no official records of victories or losses kept from one game to the next, no fans who praise winners or disparage losers. These games may be classed as "competitive," but they are really, at most, only pseudo-competitive. A score may be kept, and the players may cheer happily each time their team scores, but, in the end, nobody cares who won. The "losers" go home just as happy as the "winners." These games, too, cement friendships and do not create enemies. I wrote about the valuable lessons learned in play of this sort in my post of Nov. 11, 2009.

If the boys in Sherif's experiments had played informal games of baseball, touch football, and tug of war, rather than formal ones, I doubt that hostilities would have resulted. With no prizes or acknowledgments of victories and losses from outside authorities, the players would have focused more on having fun and less on winning. With no adult referee, the players would have had to cooperate to establish the ground rules for each game and judge consensually when rules had or had not been broken. They would have had to argue out and negotiate their differences. Cheating and name calling, if they went too far, would destroy the fun and end the game. Players who weren't having fun would quit, so the only way to keep the game going would be to play in ways designed to ensure that everyone had fun. Boys everywhere know how to do that. In fact, it is reasonable to suppose that such informal games, if they occurred, would have brought the two groups of boys closer together because of the cooperation required, much like searching for the break in the water line.

Fantasy battles and informal sports are pure play, and pure play creates friendships, not enemies. Formal sports are not pure play, and therefore they have the capacity, under some conditions, to create enemies. Formal sports lie outside of the realm of pure play because they are controlled by officials who are not themselves players and because they have clear out-of-game consequences, in such forms as prizes or praise for victory. (See Nov. 19, 2008, post on the definition of play.) In formal sports it is not as clear as it is in informal sports that the battle is merely a pretend battle.

Formal sports occupy a space somewhere between play and reality, and, depending on a wide array of factors, a formal game can shift more toward one than the other. When the balance shifts too far toward reality, a defeat is a real defeat, not a pretend one, and those defeated may begin to perceive the other team as real enemies. Sherif and his colleagues apparently found a formula for setting up formal sports in a manner that quickly moved from play to real battles.

I plan to continue this topic in my next post, next week, with an examination of some more recent research studies having to do with the effects of competitive team sports on the moral values and behaviors of the participants. The effects appear to depend very much on whether the players are led to focus primarily on winning or primarily on playful enjoyment and on development of their own skills.


See new book, Free to Learn

[1] Sherif, M., Harvey, O. J., White, B. J., Hood, W. E., & Sherif, C. S. (1961). Intergroup conflict and cooperation: The Robbers Cave experiment. Norman: University of Oklahoma Book Exchange.

More Posts