Modestly Modular vs. Massively Modular Approaches to Phonology

Abstract

This thesis considers the extent to which phonology (that is, the phonological processor) can be considered a module of the mind. It is divided into two parts. In the first, an approach of 'modest' modularity owing to Fodor (1983) is explored. In the second, the 'massive' modularity model, due to evolutionary psychologists in general, but Caruthers (2006a) in particular, is examined. Whilst for Fodor (1983, 2000) the mind is only modular around its periphery (i.e. only its input and output systems are modules), for massive modularists the mind is modular through and through, up to and including its central capacities. The two authors, therefore, by extension differ in their definitions of modularity: Fodor (1983, 2000) sees 'informational encapsulation' as being essential to modularity, whereas for Carruthers (2006) domain specificity is much more important. The thesis concludes that whether phonology is a module or not then depends on the definition of modularity, for although a substance-free phonology which has no phonetic grounding could count as strong evidence for the informational encapsulation (and therefore the modularity) of phonology by Fodor's (1983) standards, some aphasiology data has shown that semantic treatments can remediate phonological word finding difficulties in aphasia, which would be indicative that phonology is not domain-specific, and therefore amodular in the terms of massive modularists like Carruthers (2006a).1 In order to answer whether phonology is modular, then, we must first define, once and for all, what modularity (and indeed phonology) means. Until then, the debate remains, and so does my resolve to settle it.