In 1948, three years after World War II ended, psychologist Muzafer Sherif brought 22 eleven-year-old boys to Robbers Cave State Park in Oklahoma, divided them into two groups. Ran them through deliberately engineered conflict—competitive games where one team's gain was the other's loss.
The boys developed real antagonism, raided each other's cabins, then worked together to solve a water shortage and fix a broken food truck. The conflict dissolved, and some asked to ride home in the same bus.
This experiment shaped fifty years of how we think we can fix human division. But Sherif stopped asking what mattered—what happened when the boys went home to families, neighborhoods. Schools organized around real differences in income, geography, and access.
In 1968, after the assassination of Martin Luther King Jr. , diversity-training programs began using Sherif's method. They bring people across racial lines into a room, give them a shared task, and expect reconciliation. Fifty years of corporate programs, school initiatives, and university workshops have followed the same arc. The training ends, people return to organizations structured around promotion, hiring. Resources distributed along the same lines that created the original conflict. We keep rerunning Robbers Cave and then acting surprised when the effect doesn't survive contact with actual competition.
We keep rerunning Robbers Cave and then acting surprised when the effect doesn't survive contact with actual competition.
”The mechanism Sherif proved is real, but it only works when the groups have no genuine competing interests. Apply this to your own organization—notice which problems you solve by bringing people together and which ones immediately reappear the moment the room clears and resource allocation resumes. That gap between the collaborative moment and the competitive structure it sits inside is where you're actually working.