Some consequences of Lexical Phonology

Phonological theory in recent years can be said to have undergone a ‘modularisation’ in several respects. The formal theory is no longer expected to explain everything about phonology by itself: generalisations about phonological change which previously were used to motivate constraints on abstractness or opacity have turned out to make more sense as effects of real-time language acquisition and use. Secondly, phonological representations have become multi-tiered arrays, and much that seemed problematic about the application of rules has resolved itself in terms of properties of these arrays. Lastly, phonology itself is seen as applying both within the lexicon to the output of each morphological process, and to the output of the syntsactic component. The lexicon, moreover, may itself be organised into a hierarchy of levels, each constituting a quasi-autonomous morphological and phonological domain. In this paper I propose to investigate some consequences of this third kind of modularisation, the approach which has come to be known as LEXICAL PHONOLOGY.