Eric Lormand, University of Michigan

My work on mental content is broadly within recent "naturalistic" frameworks. Most other such theories begin by trying to account for reference in terms of naturalistic mind-world relations. I think this strategy is unlikely to succeed directly, because the class of representational states is too heterogeneous, and because the strategy gives too little prominence to richly varied within-mind factors in the determination of reference. Indeed, the main effect of the strategy has been to bolster the counterexample industry.  By contrast, I begin with a theory of semantic structure in terms of naturalistic relations within the mind.

The key idea is that an individual mental state may have very many contents all at once, so that the entire set of contents may be holistic while each content is not, and so that some contents of simple ideas may be simple while others are complex. With these steps, a naturalized theory of reference need only be concerned with the simple contents of simple ideas—the semantic "atoms"—easing the development of a viable theory, which need not apply to all ideas in one swoop. The theory of semantic structure is neutral among the leading naturalistic theories of reference (in terms of causation, correlation, teleology, asymmetric dependence, etc.), but I hope to develop a specific naturalized theory of reference that benefits from the semantic-structure headstart. In the spirit of the multiple-meanings view, I think that ideas have multiple reference conditions, among which some are determined by (historically-based) teleology, and some are determined by (synchronically-based) asymmetric dependence. I also want to show how the overall theory should apply to linguistic meaning in addition to mental content (see section 4 of "PSHAW!").

The basic reason that contents are holistically interconnected is that inferential dispositions are holistically interconnected (and that inferential role is a component of content). Another issue about cognitive architecture and inferential holism that relates directly to content is the distinction (if any) between perception and cognition. This is important for distinguishing among various "modules" or "systems" relative to which something is a semantic "atom" (see section 4 of "Atomist").