“When I speak of 'cycles,' I am referring to lengthy intervals of relative homogeneity, if not in the resolving of problems, than at least with respect to the consistency of their capacity to productively irritate.”