A reanalysis is presented of Smith's (1973) description of a developing phonological system, appealing to nonlinear phonological theory and underspecification. The revision furnishes a more satisfactory account of the linguistic facts, judged against linguistic criteria. It also provides the first sound theoretical basis for a distinction between input lexical representations and output lexical representations in child phonology, since this distinction is seen to be a consequence of the linguistic analysis itself. The paper discusses the interrelationship between linguistic analysis and psycholinguistic theory construction.
|Number of pages||36|
|State||Published - Jan 1986|