CAREER: Sociolinguistic Development in Pre- and Early Adolescents

Michigan State University, East Lansing MI

Investigators

Abstract

Sociolinguistic variation concerns how language is influenced by language users. One of the major mysteries of sociolinguistic variation and change is how young speakers first acquire and then subsequently use language in such a way that generational change is pushed forward. Answering this question is important because it illuminates how broader community level linguistic patterns intersect and are influenced by an individual’s linguistic patterns. Leveraging a novel audio diary method developed by the investigator, a large corpus of adolescent speech is created for an educational cohort of students. Acoustic analysis of this corpus is performed to test hypotheses regarding language change, including whether expansion of linguistic patterns is a feature of all speakers in a cohort or whether it is driven by a handful of speakers. The project supports education by connecting middle- and high-school students to the field of linguistics through partnerships with teachers and through their inclusion in the longitudinal corpus. Lesson plans about linguistics for adolescents are developed and undergraduate, graduate, and postdoctoral trainees receive training in linguistic analysis. This project also benefits society by addressing the increasing need for language scientists in fields like generative AI and the broader concern of understanding language differences. Further, the project provides substantial community outreach in the context of ongoing documentation of language, and it creates a publicly available data archive. Previous research has shown that community-level language change proceeds like a logistic curve, where, as birth year increases, so does a speaker’s rate of new variants. In addition, research has shown that the most advanced speakers at any given point in time are late-adolescents, meaning that while children acquire their parents’ production, their production begins to deviate in the pre-teen and teenage years, as teenagers incrementally use new variants. It is theorized that pre- and early-adolescents accomplish this incrementation by unconsciously modeling the distribution of variants from their surrounding speech community, calibrating an age-based vector from it, and then leveraging these different variants in their own stylistic practices. To test this hypothesis, this project leverages an existing longitudinal initiative that has collected self-recorded audio diaries for many years and expands it to obtain data from speakers spanning the range of adolescence. In doing so, this project provides one of the first comprehensive data sets consisting of regular, longitudinal data across an entire cohort of speakers. Acoustic analysis of the corpus, focusing on formant values from vowels, is used to test the hypotheses regarding individual- and community-level language change, thus advancing sociolinguistic theory. This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.