The kinematics of prominence in American English

Speech units under prominence present longer, larger, and faster constriction gestures than their non-prominent counterparts. However, whether there are discrete degrees of prominence, and if so how many, has yet to be discovered, partly because the contribution of the information structure in marking prominence is unclear. Here, we employ electromagnetic articulography (EMA) to examine the effect of focus (unfocused, de-accented, broad focus, narrow focus, and contrastive focus) on the kinematics (position, duration, velocity, and stiffness) of the consonant gestures of the test words’ stressed syllable. Test words varied in terms of length (one, two, three, and four syllables) and position of stress (first, second, or third syllable). Results from eight native speakers of American English indicate that prominent gestures become longer, larger, and faster in a way that consistently reflects three degrees of prominence. Nonetheless, kinematic dimensions differ in the number of degrees of prominence they distinguish, ranging from two levels (stiffness) to three (position and velocity) and four levels (duration). All dimensions distinguish contrastive focus from unfocused, while de-accented tends to be grouped with unfocused, and narrow focus with either contrastive or broad focus. A hierarchy of prominence is proposed taking into consideration categorical versus gradient distinctions. [Work supported by NSF.]