Broadly speaking, my research is concerned with the social meaning of phonetic variation. I am particularly interested in theories of indexicality, as all of my work seeks to offer precise analyses of the ways in which sound patterns connect to the social world. In doing so, I have drawn on a range of data types and data collection methods (fieldwork, acoustic analysis, articulatory investigations, perceptual techniques), each offering different insights into what linguistic variants mean.
A question underlying most of my work is how the social meanings of linguistic features (whether highly conventionalized and durative or surfacing only ephemerally in moments of talk) influence their patterns of distribution. This influence may be evident in the speech of single individuals as they navigate the different arenas of their lives or across communities as sound change occurs.
Much of my research has focused on language, gender, and sexuality. I have argued for an approach that treats the phenomenon of “sounding gay” as a context-dependent interactional achievement. Linguistic features that might be recruited in the construction of sexuality serve a variety of functions, some of which relate more directly to the interactional context than identity projection per se.
Sociophonetics and Affect
At present, my primary project examines the connection between sociophonetic variation and affect. This works builds off a central tenet of third-wave approaches to variation (and their precursors): that the “choice” of a linguistic variant is rooted in the dynamics of interaction. While a number of qualitative approaches to variation have elucidated the many ways that linguistic variants can serve interactional functions, incorporating these insights into a quantitative paradigm has posed a significant challenge. As a result, the field of variation has yet to construct a theory that treats interaction as fundamental.
My most recent projects establish a strong connection between sociophonetic practices and one particular dimension of interaction: the expression of affect. This work attempts to capture affect across (non-phonetic) modalities, through the way that interactants position and move their bodies and through the sentiment encoded in the lexical/semantic content of their speech.
My research on embodied affect draws on computer vision techniques (all developed and implemented by doctoral student and frequent collaborator, Rob Voigt) to code for whether speakers are smiling, how much they are moving their bodies, and the extent to which they tilt their heads. Data for these projects have been collected in the Interactional Sociophonetics Laboratory, a space with the acoustical specifications of a sound recording booth (to facilitate the collection of high-quality audio data), that is staged like a living room (to encourage less self-conscious talk), with inconspicuous video cameras (to enable analyses of the visual channel).
Results of these projects (conducted in collaboration with Patrick Callier, Katherine Hilton, and Rob Voigt) reveal that speakers produce more shifted variants of the Western Vowel Shift (lowering of the front lax vowels and fronting of the back vowels) when they are smiling and moving more. Speakers are also more likely to phonate in creaky voice when they are moving less and not smiling.
At a minimum, these results indicate that more explanatory models of variation are possible if interactional factors (like the expression of affect) are incorporated into analysis. They also highlight the value of conceptualizing regional varieties in affective terms.
Voices of California
A second current project is a collaborative endeavor between students and faculty in my department called Voices of California (VOC). Each year, we visit to a lesser-studied region in California and interview residents about their community and life histories. As of 2018, we have conducted around 800 interviews in Merced, Redding, Bakersfield, Sacramento, Salinas, Humboldt County, and Redlands. The VOC corpus serves as an archive about language and life in these communities, as well as a dataset for investigating linguistic variation. So far, I have conducted acoustic analyses on the area’s vowel system (with Annette D’Onofrio, Penny Eckert, Seung Kyung Kim, Teresa Pratt, and Janneke Van Hofwegen), fronting and retraction of /s/ (with Janneke Van Hofwegen), degree of voicing in stop closures, velarization of onset /l/, creaky voice quality (with Patrick Callier), and phrase-final lengthening (with Jeremy Calder, Penny Eckert, and Julia Fine), all in collaboration with graduate students, postdoctoral scholars, and faculty members at Stanford.