Connectionism and the Specter of Representationalism

Abstract

Jerry Fodor and Zenon Pylyshyn’s 1988 attack on connectionism is both well known and widely discussed.¹ It is meant to shore up support for the classical computationalist view of cognition, in the face of recent connectionist challenges. Less well known is their 1981 attack on the direct realism of J. J. Gibson.² It is meant to provide a similar sort of support, by sustaining the representational realism to which the classical computationalist view is committed.

Conscious, deliberate thinking is both exhausting and infrequent, a last resort to be appealed to only when all habitual capacities have failed.

Anthony Quinton “The Problem of Perception”

As one acquires a certain skill [including a perceptual skill], one does not internalize the conscious inferences and make them faster, but rather the rules underlying the skill become embodied in the agent’s physiological and cognitive structures-thus one conserves mental effort.

Aaron Ben-Zeev “Can Non-pure Perception Be Direct?”

I would like to thank Panayot Butchvarov, David K. Henderson, and John Tienson for their helpful comments on an earlier draft of this paper.