Conventional beauty and linguistic accommodation
LINK | Young, N.J., Britain, D. & Leemann, A. (2022). A blueprint for using deepfakes in sociolinguistic matched-guise experiments. Proceedings of INTERSPEECH 2022, Incheon, Korea, 5268–5272.
with Drs. Adrian Leemann and David Britain
Funding body: Swiss National Science Foundation Spark Grant
Center for the Study of Language and Society, University of Bern
Are physically attractive people the leaders of linguistic change? This research project will investigate the influence of conventional beauty on the transmission of linguistic innovations in a community.
When linguistic innovations – specifically those related to sound – spread throughout a community, the conduits are the momentary interactions between two speakers. In these situations, interlocutors will typically align (or dis-align) with one another’s speech. This process is referred to as "accommodation". The accumulation of speech-accommodation events facilitates the cognitive embedding of certain innovations, which solidifies their (re)production among speakers in a community. That solidified production among incrementally growing numbers of speakers results in community-wide change.
Research has shown that social factors can influence whether two interlocutors accommodate each other’s speech – that is to say, whether they produce phonetic features more close to one another’s features. To name a few, these social factors include nationality, race, social status, dialect attitude, and relationship strength. However, no study has examined the effects of physical attractiveness, despite the fact that research has shown it to be influential in many other domains (e.g., purchasing decisions or teacher assessments of students).
We propose a new experimental approach that will examine the effects of conventional beauty on the person to person transmission of a new feature in London English: the "high rising terminal" (HRT). HRT involves ending a declarative statement with a question-like fundamental-frequency contour (e.g., https://www.youtube.com/watch?v=z756L_CkakU). Research has shown that fundamental frequency is especially sensitive to accommodation, which makes HRT a befitting linguistic variable to test.
Our experiment will record the baseline speech of a participant telling a short story that s/he has devised. We then show a video guise to the participant (randomized as ‘conventionally beautiful’ or ‘conventionally normal’, male or female, with HRT or without HRT). We then have the participant tell the same story again. We then measure the difference in HRT usage between story 1 and story 2. Eight video guises will be created, and we will recruit 240 participants for this study.
This project has the potential to make an important theoretical contribution. Fields like Psychology and Economics have identified conventional beauty to be an important social factor for predicting a number of material outcomes. Meanwhile, Linguistics has more or less ignored the effects conventional beauty might have on transmitting change, which means the project would offer a much-needed theoretical update.