The neuronal structure of the cortex has usually been perceived as
radically di erent from the semantic structure on which cognition is
defined. This paper, in a line with other papers (Werning, 2001,
2003a, 2004; Maye, 2003; Maye & Werning, 2004), however, argues that
despite of these prima facie differences, semantic structure can be
reduced to neural structure if one assumes that oscillatory neural
networks be an appropriate model of neural reality. Several
neurobiological data on neural synchronization support this assumption
(Singer & Gray, 1995). In the context of this paper, semantic
structures are assumed to subserve two purposes. First, its elements
evaluate linguistic expressions semantically and can thus be regarded
as the meanings of those expressions. Second, its elements are
themselves semantically evaluable with respect to external
content. The linguistic expression dog , e.g., has as meaning the
concept DOG, which is some internal state of the cognitive system
whatever it may reduce to neurally. This internal state is itself a
representation and must thus have external content. Its external
content will here be identified with the set of dogs and is thus
identical to the denotation of the predicate dog . Since
representations are supposed to co-vary reliably with what they
represent (Fodor, 1992), the latter identification requires that
internal states reliable co-vary with the denotations of those
expressions whose meanings they are.