Anderson, A., Bader, M., Bard, E. G., Boyle, E., Doherty, G., Garrod, S., Isard, S., Kowtko, J., McAllister, J., Miller, J., Sotillo, C., Thompson, H. S., & Weinert, R. 1991. The HCRC Map Task Corpus. Language and Speech, 34(4), 351–366.
Brown, G., Anderson, A., Yule, G., & Shillcock, R. 1983. Teaching talk: Strategies for production and assessment. Cambridge University Press.

Bucholtz, M., & Hall, K. 2016. Embodied sociolinguistics. In N. Coupland (Ed.), Sociolinguistics: Theoretical debates. 173–197. Cambridge University Press.
Ekman, P. 1992. Facial Expressions of Emotion: New Findings, New Questions. Psychological Science, 3(1), 34–38.
Gudi, A., Tasli, H. E., den Uyl, T., & Maroulis, A. 2015. Deep learning based FACS Action Unit occurrence and
intensity estimation. 11th IEEE International Conference and Workshops on Automatic Face and Gesture
Recognition, 6, 1–5.
Kiesling, S. 2018. Masculine stances and the linguistics of affect: On masculine ease. NORMA: International Journal for Masculinity Studies, 13(3–4), 191–212.
Podesva, R. 2021. The role of the body in language change. In L. Hall-Lew, E. Moore, & R. Podesva (Eds.), Social meaning and linguistic variation: Theoretical foundations (pp. 363–381). Cambridge University Press.
Pratt, T. 2023. Affect in sociolinguistic style. Language in Society, 52(1), 1–26.
Siedlecka, E., & Denson, T. 2019. Experimental Methods for Inducing Basic Emotions: A Qualitative Review. Emotion Review, 11(1), 87–97.

Speech is an embodied phenomenon: the body is central both to the production and perception of spoken language. Despite this, sociolinguistic research has tended to overlook the body (Bucholtz & Hall 2016), treating language as an autonomous system. Because of this, we lack an understanding of how bodily processes condition social patterns of language use. To address this gap, we investigate the phonetic consequences of embodied emotion, testing whether adopting a bodily posture that is linked to the expression of affect results in systematic and predictable acoustic outcomes that, in turn, feed processes of socially meaningful variation. Doing so allows us to explore the possibility that sociolinguistic variation may arise from acts of embodied stance-taking. If such a link between the body, affect, and speech is found, this would provide novel insight into the actuation and diffusion of language variation and change (Podesva 2021; Pratt 2023).
Our discussion is based on an examination of correlations between specific facial expressions and linguistic variability during a collaborative map task (Brown et al. 1983; Anderson et al. 1991). In our experiments, participants are paired with a confederate who, unbeknownst to participants, has been hired by the research team to vary the degree of cooperative behavior in the completion of the task. This allows us to induce different emotional responses in participants (e.g., frustration and distancing for uncooperative conditions vs satisfaction and interpersonal alignment in cooperative ones), and so identify the bodily and linguistic correlates of different emotional states (Siedlecka & Denson 2019). Map task interactions are recorded with high quality video cameras and lapel microphones, using specialist software that synchronizes audio and video channels and automatically detects movement in 468 key points in the face (Gudi et al. 2015). This allows us to link acoustic variation with specific emotion-linked changes in facial expression.
Experiments are conducted in Danish (Copenhagen) and Swiss German (Bern). We focus on these languages because they have similar vowel inventories, but differ in how vocalic variation is evaluated. While raising front vowels in Copenhagen Danish, for example, carries prestige, vowel lowering is more prestigious in Bernese German. The languages thus provide a useful case study where, despite their systemic similarities, sociolinguistic ideologies predict opposing trajectories of stylistic variation. This allows us to examine how embodied affect and social meaning interact to condition patterns of acoustic variation in spoken interaction.
Testing of 64 participants (32 in each language) is ongoing, with data collection scheduled for completion in early 2024. Analyses will model the extent to which movement of specific facial action units (e.g. lip corner, jaw; Ekman 1992) correlates with formant frequencies of target vowels, and whether those patterns are consistent across languages. Results will reveal whether a significant correlation exists between embodied affect and sociolinguistic variation. The findings will provide the first controlled examination of recent proposals regarding the role of affect in conditioning variability (Kiesling 2018; Pratt 2023) and help us to understand the relationship between speech, the body, and the way variation acquires social meaning.