Previous
Next
Dr. Matt Carlson
March 29, 2024
9:00 am
Foster Auditorium, Pattee and Paterno Library

Dr. Matt Carlson

“You Can’t Hear a Phoneme: Insights on the Nature of Phonetic (and Phonological) Category Representations From L2 Speech Learning”

Matthew Carlson, Ph.D.
Associate Professor, Penn State
Department of Spanish, Italian, and Portuguese

Matthew Carlson’s goals in this talk are twofold. The first is quite straightforwardly to remind us that the perception of L2 speech sound categories per se probably contributes little to the difficulties faced by learners in expressing themselves or understanding others in the L2. For example, difficulty distinguishing minimal pairs is unlikely to surface outside of laboratory experiments using isolated words, either for the L2 listener or for their interlocutors. Where real impacts do occur, these seem more likely to stem from speed of lexical and morphosyntactic processing, lexical knowledge, or social attitudes surrounding accentedness and nativeness.

His second goal is to shift our focus from the sources of (supposed) difficulty onto what L2 users actually learn, and thereby to seek new insight into the nature of phonetic category structure in general. Here I will focus on L2 speech perception. The logic begins by acknowledging humans’ exquisite sensitivity to acoustic detail, which persists throughout life, not least in our ability to learn new sociophonetic variation, recognize the speech of specific individuals, and so forth. Where experience with multiple languages shapes learning therefore has little to do with sensitivity to detail, but rather it concerns what listeners do with acoustic detail. He will present a series of perception experiments that draw our attention to experience-dependent differences in how listeners represent acoustic detail at various levels of abstraction, but he will then argue that it is more useful to consider these abstract representations not as an inventory of phonetic categories or phonemes, but as a repertoire of possible paths that processing might take after hearing some speech. Under this view, phonetic and phonological categories do not resemble objects so much as they do crossroads, which can be reached from multiple directions, and from which certain other crossroads might be accessed.