Abstract

It is customarily thought that in addition to the class of observed phenomena there is a larger class of observable phenomena. For a theory to be empirically adequate, it must be true on this larger class. It is denied that there is such a thing as the class observable over and above observed phenomena. This does not entail that empirical adequacy reduces to agreement with just the observed facts. Observability is a feature of abstract items in the models of theories, and thus conditions their empirical import. This doctrine of observability yields a brand of empiricism that a realist might live with