Abstract

What is the empirical content of a theory? If a theory is identified with one of its linguistic formulations, the only available answers allow for no non‐trivial distinction between empirical and non‐empirical content. The restriction of such a formulated theory to a narrow ‘observational’ vocabulary is not a description of the observable part of the world but a hobbled and hamstrung description of its entire domain, still with non‐empirical implications. Viewing a theory as identified through the family of its models––the structures it makes available for modelling the phenomena––yields a new approach. The distinctions so made are illustrated with Newton's physics, absolute versus relative motion, nineteenth‐ century ether theory of electromagnetism, and quantum mechanics. A hermeneutic circle in the interpretation is noted, and the theory‐independence of the observable/unobservable distinction maintained.