The Modularity Theses. The main versions, the problems, and what might be done (Modularitetsteserna: De huvudsakliga versionerna, problemen och vad som skulle kunna göras)

Abstract

This paper is about mental architecture. Its main purpose is to examine claims that the internal organisation of the mind is modular either to a modest extent or to a massive extent, respectively. It also includes a summary of the pointed critique of modularity, according to which “mental modules”, if such there be, are most likely few and far between. According the original version of the modularity thesis, put forward by Jerry A. Fodor in 1983, modules are responsible for the link between mind and world: they all have to do with various aspects of perception and sensation, but do not figure in higher cognition. Fodor’s modules all exhibit “all or most” features found on a list of nine. This list is very controversial. Peter Carruthers, who arguably is the main proponent of a massive version of the thesis, wishes to include higher cognition in his theory of modular architecture and therefore modifies the list by interpreting some of the features differently, as well as by shortening the list as such. However, while massive versions are popular amongst evolutionary psychologists and AI researchers, controversy remains – not only concerning the (new) list, but also the range of the thesis. Much of the critique reflects the old divide between rationalist and empiricist philosophy, or more pointedly, the question of whether the capacities of the human mind are mostly innate or mostly learned. Jesse J. Prinz belongs to the latter camp and rejects both modularity theses, arguing that they are largely mistaken about the workings of the mind, and in some respects are completely unfounded. The paper ends with my own proposal for a middle-ground theory attempting to reconcile these positions, since I believe each has something important to bring to the table.