ICS Architectures, Classical Constituents, and Systematicity
Thursday, February 20, 2003, 12:00pm - 07:00pm
Rutgers University, Department of Philosophy
Brian P. McLaughlin
Department of Philosophy
The central questions of the philosophy of mind are the nature of mental phenomena, and how mental phenomena fit into the causal structure of reality. The computational theory of mind aims to answer these questions. Subtleties aside, the central tenet of the theory is that a mind is a computer. According to the theory, mental states and events enter into causal relations via operations of the computer. The main aim of the theory is to say what kind of computer�what kind of computational mechanism�a mind is. Computationalists hold, of course, that the functional architecture of the computing system that grounds our mental abilities resides in our brains. There is, however, no consensus as to what even the general character of that architecture is. The two dominant research paradigms are the symbol system paradigm and the connectionist (or neural-network) paradigm. They differ in what kind of computer the mind is assumed to be, and thus in the kinds of functional architectures explored. The symbol system paradigm presupposes that the mind is a kind of automatic formal system or system of such systems, while the connectionist paradigm presupposes that it is a system of connectionist networks.
One of the leading challenges to the view that the architecture of the mind is connectionist is to explain how a connectionist architecture could explain the systematicity and productivity of thought without implementing a symbolic��language of thought��architecture. Paul Smolensky has proposed a kind of architecture that he calls an �integrated connectionist, symbolic architecture� (ICS architecture), and which he maintains can do precisely that. The talk will introduce the main ideas of the computational theory of mind, and then focus mainly of a critical examination of Smolensky�s claim to have answered the challenge mentioned above.