Formal and Empirical Arguments Concerning Phonological Acquisition

This paper draws on the generative literature in phonological acquisition, as well as on the work of phoneticians and psycholinguists, in an attempt to propose a unified view of the acquisition of phonological competence. We start with an examination of the Optimality Theoretic (OT) account of the well-known comprehension/production dilemma in child language presented in Smolensky (1996ab, henceforth S). We will argue that S’s model encounters two serious difficulties — the first concerning his proposed parsing algorithm and the second concerning the issue of the learnability of underlying forms. We offer alternative parsing algorithms and examine their implications for learnability and the initial ranking of OT constraints. Finally, we propose, based on evidence from a variety of sources, that the resolution of the comprehension/production dilemma lies not in phonological domain (linguistic competence), but rather in the domain of implementation of linguistic knowledge (performance). With a revision of certain aspects of the OT model for children’s phonologies and of learnability theory in phonology, the paper attempts to contribute both to research on OT and to the study of phonological acquisition generally. It is not our aim here to argue for or against OT approaches to phonology. We will not hesitate, however, to point out flaws in the application of OT in the current literature. We hope that the improvements in the application of OT which we propose, as well as mention of points in which the theory seems to fare worse than alternative approaches, will lead to the kind of progress in the fields of phonology and acquisition that transcends the provinciality of particular frameworks.