Abstract

The purpose of this dissertation is to come-up with a theory of mental content as a causally efficacious property. It is claimed that such a notion of mental content needs to satisfy seven desiderata: mental content has to be a causal-nomic property, that is, one which is subsumed by causal-laws. mental content has to determine causal processes which are isomorphic to its semantic liaisons. mental content has to be an atomistic or molecularistic property: it cannot be a holistic property. mental content has to be individuated at an empirically adequate level of grain. That is, the theory has to account for the empirical possibility of informational differences between coextensive terms , and for the empirical possibility of theoretical substitutions of terms that are neither coextensive nor co-referential . mental content has to be a representational property that is able to encode information about objects and states of affairs in the world. mental content has to be realized by a supervenience base which could endow the content-bearing states of physical Twin-earthlings with distinct causal powers. mental content must allow for the possibility of error or misrepresentation. ;I argue that a theory of Modal Response Dependence Informational Content is the theory which can satisfy desiderata A--G. As informational content is constituted by nomic relations to the distinct instantiations of properties in the world, it is nomic, atomistic, representational, and can satisfy the supervenience thesis by reference to relationally individuated brain-states. It also satisfies the isomorphism condition as content is identified with causal relations to objects and events in the world. ;Construed in terms of informational relations, the theory of content becomes a species of response-dependence theories of content, with content relativized to selective responses of the organism to objects and events in its environment. However, if limited to actual responses, those informational relations turn out to be either too coarse or too fine to ground mental content. It also threatens the account on error. It is therefore suggested that content be determined by reference to possible selective responses of the organism, with the modality ranging over the set of cognitively possible worlds. I argue that such a theory can then satisfy the remaining desiderata on grain and error