We have been here before. In the 1970s, the catastrophe theory of René Thom seemed to promise an understanding of how sudden changes in society might be provoked by small effects. This initiative atrophied rather quickly, since Thom’s phenomenological and qualitative theory did not really offer fundamental explanations and mechanisms for the processes it described. Chaos theory, which matured in the 1980s, has so far proved rather more robust, supplying insights into how complicated and ever-changing (“dynamical”) systems rapidly cease to be precisely predictable even if their initial states are known in great detail. Chaos theory has been advocated as a model for market economics, and its notion of stable dynamical states called attractors seemed to provide some explanation for why certain modes of social behavior or organization remain immune to small perturbations. But this theory has not delivered anything remotely resembling a science of society.