Diffusion is Predictable: Testing Probability Models for Laws of Interaction

ranged as replicas of commonplace social situations conducive to cultural production. Since cultural phenomena were actually brought about in the experiments and not merely simulated, the miniature histories appear to be operational replicas. The matter of these particular replicas working outside of the laboratory has not been established. But the experiments do raise questions as to the actual workings of culture in history. Our most disturbing discovery was the creativity of isolated groups. The stimulation and novelty of experience in open social settings did not increase cultural creativity in the experiments. The experimental demonstration of the coincidence of social closure with cultural invention poses the problem of the extent of their association in history. There are plausible examples of socially restricted creativity in schools of art and thought, in golden ages and recurrent clusterings of all sorts of distinguished achievement. But the great example of mobile and cosmopolitan urban populations having made the most of cultural change runs counter to our experimental results. We should hesitate to extend these results without qualification to a general interpretation of all cultural accomplishment. However, our results do give reason to question any bland presumption that social mobility leads inevitably to cultural creativity. We should now be more careful to search for explicit circumstances under which either isolation or mobility may contribute to the production of culture.