Dynamic Articulatory Phonology and the Supervision of Speech Production

ABSTRACT Articulatory Phonology unifies the domains of phonetics and phonology, linking utterance planning and execution by common units of control. It links with the Task Dynamic Model of speech production, forming a smooth data pathway from the most abstract level to the physical level of articulatory configuration. This paper reviews the need for refinements to the model and proposes task supervision to explain some data previously overlooked. ARTICULATORY PHONOLOGY Articulatory Phonology was proposed by Browman and Goldstein [1] partly as an attempt to unify phonetics and phonology by treating them as ‘low and high dimensional descriptions of a single system’ [2]. They come together by the idea that the constraints of the physical system underlie the phonological system, and by making the units of control at the planning level the same as those at the physical level; planning and execution are seen as more closely related than in other theories of phonetics and phonology.The plan of an utterance is formatted as a