Abstract
This article summarizes a presentation made to an ESRC-sponsored workshop on social science methodological innovation. First, the notion of methodological innovation is explored on the way to an argument that innovation in social sciences occurs at a conceptual level rather than at the level of techniques for data gathering. Four historical examples are briefly described to exemplify the argument: Zadeh's fuzzy set theory, Goffman's presentation of self, Bartlett's schema, and Glaser and Strauss' grounded theory. In response to questions posed by the organizers, it is noted that the four examples were produced by individuals with a considerable time lag before uptake of their concepts in the diffusion curve. The article ends with speculation on the possible nature of a social science center for conceptual innovation.