Representationism, Realism and the Redundancy of `Mentalese'

Publisher Website
Google Scholar
Abstract
Representationism as a theory of knowledge cannot avoid collapsing into solipsism, cannot distinguish between true and false beliefs without contradicting its own basic premise and cannot give any objective ground for inference. Fodor's postulated `mentalese' is shown to suffer these defects; further, its nativism cannot bridge the supposed gap between sensory particulars and cognitive universals. His arguments for its necessity for natural language learning are problematic. Direct realism avoids these unnecessary difficulties by postulating a native perceptual ability to identify kinds of complex object, and finds the grounds of inference in the perceived structure of the world. Cognitive science can have useful applications if freed of representationism.

This publication has 4 references indexed in Scilit: