Community Level Differences
in the Perception and Production of Coarticulatory Speech
Individual speakers can differ in the extent to which they overlap articulatory gestures. Similarly, listeners can differ in how much they rely perceptually on coarticulatory information. In this presentation, I will explore the ways in which this kind of community level variation is structured at the level of the individual. Specifically, I will investigate the extent to which the coarticulatory production patterns of an individual correspond to that individual’s reliance on coarticulatory information during speech perception. Most theories of speech production and perception (including theories of sound change) assume a link between the perception and production repertoires of individuals, although there is limited evidence to date for the existence of such a link.
The focus in this presentation will be on the perception and production of anticipatory nasalization. In the studies that will be discussed, nasal airflow was used to identify the onset of anticipatory nasalization during speech (at what point during the vowel does nasal airflow initiate in the production of a word like scent), and eye-tracking was used to measure the perceptual reliance on anticipatory nasalization (when presented with an auditory stimulus scent, do listeners fixate on the target scent rather than the competitor set based on the anticipatory nasalization during the vowel, or do they wait for the disambiguating information contained in the consonant following the vowel).
I will discuss two studies that explore this phenomenon, in American English and in Afrikaans, respectively. These two languages are similar in terms of the observed inter-speaker variation in the reliance on anticipatory nasalization during both production and perception. The social structure of the variation is different between the two speech communities, however. In American English, the extent of nasalization is socially relatively unmarked, while the degree of nasalization is strongly correlated with the difference between so-called “White Afrikaans” and the socio-ethnic minority variety of the language known as “Kleurling Afrikaans”. I will show that there is evidence for the correlation between perception and production repertoires of speakers in both American English and in Afrikaans, and explore some of the differences that arise from the different social embeddedness of anticipatory nasalization in these two speech communities.