Velar Movement and Timing: Evaluation of a Model for Binary Control

Abstract

Cinefluorographic analyses of the articulatory movements of two speakers were used to evaluate a binary model of velar control. To obtain precise timing data for movements of the tongue and velum, radiopaque markers were attached to these structures and then tracked by cinefluorography. It is argued that the binary control model does not survive rigorous empirical test because (1) it makes some predictions that are inappropriate for American English, and (2) the confirmation criteria used in previous reports were applied inconsistently. We suggest that if binary control models are to be used in making predictions about the timing of articulations, then the binary feature values should be interpreted by means of a systematic phonetics or an elaborated speech production model. We further suggest that sequences of articulatory movements may be organized in a hierarchical fashion, such that neuromotor instructions from the motor command system often are issued simultaneously for the tongue, velum, lips, and jaw, even though the commands may not always apply to the same phonetic segment.