While examinations of the distinction between phonetics (gradient sound patterns) and phonology (categorical sound patterns) in spoken languages have increased our understanding of linguistic patterns, further understanding can be gained by examining this distinction in sign language. The current proposal outlines a program of research that will probe the distinction between gradient and categorical patterns of handshape variation in American Sign Language (ASL).
Through a series of production and perception experiments, the researchers will quantify the degree to which handshape coarticulation (gradient overlap of gestures) and assimilation (categorical feature spreading) are present in ASL. The production experiments will yield kinematic data on ASL handshapes as produced in different linguistic environments by native ASL signers. These data will be subjected to repeated measures analyses of variance (ANOVA's) to determine the extent to which handshape variation occurs in ASL and the factors that significantly affect such variation. The temporal correlation of the handshape gesture with respect to the movement gesture of a sign will also be analyzed to determine whether the data are best understood as the product of an assimilatory or a coarticulatory process. Finally, the production data will be used to create the stimuli for the perception experiments. Native and near native signers of ASL will be perform forced-choice identification tasks. Through the perception experiments the researchers will be able to determine the extent to which handshape neutralization occurs, thus gaining evidence as to whether the handshape variation patterns should be classified as assimilation or coarticulation.
Similarities found in the articulatory patterns of speech and sign can be used to extend theories of gestural overlap and feature spreading beyond the oral-audio modality and into the manual-visual modality, providing a broader theory of coarticulation and assimilation. Similarities in the production of sign and speech suggest that phenomenon such as coarticulation do not depend on the availability of a speech module. Instead commonalities in the production of sign and speech may reflect general principles of motor control, such as motor economy, that are common to all motor systems.
