Abstract

This paper examines Ian Hacking's arguments in favor of entity realism. It shows that his examples from science do not support his realism. Furthermore, his proposed criterion of experimental use is neither sufficient nor necessary for conferring a privileged status on his preferred unobservables. Nonetheless his insight is genuine; it may be most profitably seen as part of a more general effort to create a space for a new form of scientific and philosophical certainty, one that does not require foundations.