The Robbers Cave experiment

What it says

Muzafer Sherif wanted to know where hostility between groups comes from. His answer was that it needs no bad people, no old grudge and no difference between the groups at all. Two groups competing for something only one can have is enough.

The study ran in three stages. First each group formed on its own, with a name, a flag and its own rules. Then the two were put into a tournament with prizes only one side could win, and the friction came fast. Last, the staff set problems that neither group could solve alone, and the hostility faded.

The idea the study is cited for is realistic conflict theory. Groups turn on each other when their interests really clash, and they stop when they need each other. Sherif's word for a need of that kind was a superordinate goal: one above both groups, wanted by each and out of reach of either alone.

The three stages, and the one that did not work

What the staff arranged at each step, and what the boys did in answer.

1. Forming the groups

   Each cabin spent about a week on its own, hiking, swimming and cooking. The boys named themselves the Rattlers and the Eagles, made flags and settled who led.

2. The tournament

   Baseball, tug of war and tent pitching, with medals and pocket knives for the winners. Name calling came first, then a burned flag, then raids on the other cabin.

3. Contact that changed nothing

   The staff put the two groups together for meals and a film. The meals ended in food fights. Being in the same room did nothing on its own.

4. Goals that needed everyone

   The camp water supply stopped, and finding the blocked valve took both groups. A truck that would not start had to be pulled by every boy on one rope. After a few of these the raids stopped, and the boys chose to ride home on one bus.

Where it came from

Sherif ran the study in 1954 at Robbers Cave State Park in Oklahoma, with his wife Carolyn Sherif and three colleagues. The full report came out in 1961, and his short account of the superordinate goal in 1958.

The boys were eleven year olds from the Oklahoma City area, chosen for being ordinary: steady homes, average grades, no trouble at school. They did not know they were in a study. The staff were the researchers, and the camp was the laboratory.

What holds up, and what does not

The central result has held. Competition for something scarce produces hostility between groups, and a goal both groups need and neither can reach alone reduces it. Gordon Allport argued the same year, in The Nature of Prejudice, that contact between groups works when they cooperate toward a shared end. The work on contact since has kept that condition.

Two things weaken it. Henri Tajfel showed in 1970 that schoolboys sorted into groups on a trivial basis, with nothing to compete for, still favored their own side when sharing out points. So a clash of interests is enough for hostility, but it is not needed for taking sides. And Gina Perry's 2018 reading of the study's archives found more of the hand of the staff than the report let on. The staff planned the frictions and stepped in to keep them going. An earlier attempt in 1953 had gone differently. The boys saw through parts of it and turned on the staff rather than each other, and that study was called off and never fully reported.

What it does not mean

It does not show that groups are bound to fight, or that people in a group lose their judgment. The boys fought when the camp was set up so that one side's win was the other side's loss, and they stopped when it was set up differently. The lesson is about the setup. It also does not say that any shared task will do. The water and the truck mattered to both groups and could not be managed by one. A joint task nobody cares about is not a superordinate goal.

Check yourself

Three questions on what is above. Each answer says why.

1. Pick the statement that describes what finally reduced the hostility between the two groups.

   1. Putting the groups in the same room for meals and a film
   2. Tasks that both groups needed done and neither could do alone
   3. Telling the boys they were being studied
   Show the answer

   Answer: B. Meals and a film together ended in food fights. The water supply and the stalled truck needed every boy on the job, and after a few of those the raids stopped.

2. What is a superordinate goal, in Sherif's sense?

   1. A goal that matters to both groups and that neither can reach by itself
   2. The prize the winning group takes home from a tournament
   3. A rule the staff set for both cabins to follow
   Show the answer

   Answer: A. The word means a goal above the groups: both want it and each needs the other to get it. The tournament prize was the opposite, a goal only one side could win.

3. Tajfel's 1970 study sorted boys into groups on a trivial basis and gave them nothing to compete for. What did it add to the Robbers Cave finding?

   1. That groups with no clash of interests never favor their own side
   2. That a clash of interests is enough for hostility but not needed for taking sides
   3. That boys of that age cannot form groups on their own
   Show the answer

   Answer: B. The sorted boys still favored their own members when sharing out points. Competition explains the raids at Robbers Cave, but people take sides with far less than that.

Sources

1. Sherif, M., Harvey, O. J., White, B. J., Hood, W. R., and Sherif, C. W. (1961). Intergroup Conflict and Cooperation: The Robbers Cave Experiment. University of Oklahoma Book Exchange.
2. Sherif, M. (1958). Superordinate goals in the reduction of intergroup conflict. American Journal of Sociology, 63(4).
3. Allport, G. W. (1954). The Nature of Prejudice. Addison-Wesley.
4. Tajfel, H. (1970). Experiments in intergroup discrimination. Scientific American, 223(5).
5. Perry, G. (2018). The Lost Boys: Inside Muzafer Sherif's Robbers Cave Experiment. Scribe.

Practice it on Learn

More theories

• The Milgram experiment

  The Milgram obedience experiment: what the 1963 study did, the 65 percent figure, how the variations changed it, and what later work says it shows.

• Social identity theory

  Social identity theory explained: the three steps, the minimal group experiments behind it, and what it does and does not say about prejudice.

• The Stanford prison experiment

  The Stanford prison experiment: what happened in the basement in 1971, the demand characteristics, and what it does not prove.