Fodor argues for two theses that can be characterized as forming a dilemma: On one side, if we endorse a functional theory of concepts, we explain Frege cases in semantic terms but we cannot avoid semantic holism. On the other side, if we endorse a causal/informational approach, we avoid semantic holism but we cannot explain Frege cases in semantic terms. Fodor (e.g., 1994, 1998 y 2008) avoids the dilemma claiming that we can give a syntactic, non-semantic explanation of Frege cases. In this paper, I want to offer an alternative answer to the fodorian dilemma. I will offer a semantic explanation of Frege cases that incorporates both causal and functional elements. I will claim that the cognitive content of a concept (the kind of content apparently motivated by Frege cases) is the set of causal/informational contents of the representations that figure in its functional role. I believe that individuating representations in functional roles by their causal content helps to avoid holism (avoiding the process of ramsification typically used to individuate roles) and that identifying cognitive or narrow content with causal contents makes possible to endorse a causal theory while avoiding referentialism (whose rejection is motivated by Frege cases).
Causal semantics; Functionalism; Cognitive content; Holism; Referentialism.