Carnap on theoretical terms: Structuralism without metaphysics

Abstract: Both realists and instrumentalists have found it difficult to understand (much less accept) Carnap’s developed view on theoretical terms, which attempts to stake out a neutral position between realism and instrumentalism. I argue that Carnap’s mature conception of a scientific theory as the conjunction of its Ramsey sentence and Carnap sentence can indeed achieve this neutral position. To see this, however, we need to see why the Newman problem raised in the context of recent work on structural realism is no problem for Carnap’s conception; and we also need to locate Carnap’s work on theoretical terms within his wider program of Wissenschaftslogik or the logic of science. Carnap’s distinctive approach to theoretical terms dates from beginnings of his semantic period in the years following the publication of The Logical Syntax of Language (1934). Particularly important, in this context, is his monograph Foundations of Logic and Mathematics (1939), where the partial interpretation view of theoretical terms first emerges. This view is further articulated in “The Methodological Character of Theoretical Concepts ” (1956), and it is then connected with Carnap’s use of the Ramsey sentence in “Beobachtungssprache und Theoretische Sprache ” (1958)—through the mediation, as Stathis Psillos first documented, of Hempel’s “The Theoretician’s Dilemma ” (1958).i Carnap continued to work on the Ramsey sentence representation of theories (and the closely related representation using Hilbert’s ε-operator) throughout the 2 late 1950s and early 1960s, culminating in Philosophical Foundations of Physics in 196