CAUSAL THEORY OF REFERENCE AND THE PROPER NAMES PUZZLE

Abstract

This paper revisits Michael Devitt’s causal theory of names through the classical Mark Twain / Samuel Clemens case to evaluate its ability to explain differences in informativeness and rational behaviour without resorting to Cartesian assumptions. Devitt originally claimed that distinct causal designation chains account for both meaning and informativeness. However, in his later work he rejects the idea that a theory of meaning must address epistemic matters, arguing that informativeness stems not from semantic differences but from the “law of identity” and surface dissimilarity between names. The paper challenges this separation by showing that semantic content—as causal mode of reference—cannot be detached from the subject’s cognitive perspective if we are to make sense of reasoning and rational action. Through a sequence of inferential tests involving co-referential and non- co-referential names (Mark Twain / Samuel Clemens, London / Londres, Liebknecht [Wilhelm/Karl]), it is shown that meaning must align with the subject’s point of view to explain rational behaviour. When semantic content diverges from cognitive access, inference and reasoning become unintelligible. The upshot is that Devitt’s notion of linguistic competence as knowledge-how rather than knowledge-that does not eliminate the need for epistemic engagement within a theory of meaning. The paper concludes that informativeness and meaning are interdependent: items of whose sameness or difference the rational subject can be unaware cannot perform the role of meaning.