Lifespan change, generational differences and the progression of sound change: phonetic analyses of /s/-retraction in real and apparent time

• Applicant: Dr. Mary Stevens
• Subject Area: General and Comparative Linguistics, Experimental Linguistics, Typology, Non-European Languages ; Applied Linguistics, Computational Linguistics ; Individual Linguistics, Historical Linguistics
• Term: since 2017
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 387170477

Historically documented sound changes are typically thought to be driven primarily by the addition of innovative speakers (e.g. younger generations) to a population, with pronunciation norms remaining relatively stable over the adult lifespans of its individual members. On the other hand, there is empirical evidence that adult speakers’ pronunciation targets can shift incrementally over real time in the direction of historically documented sound changes. This suggests that sound change might occur not only when younger generations acquire a more innovative system but also when individual members of the population become more innovative over time. However, the relationship of such linguistic plasticity during adulthood to the kinds of permanent sound changes that can affect whole languages remains to be shown. The proposed project renewal investigates /s/-retraction in two Australian English-speaking communities: one large and linguistically diverse, the other small and linguistically homogeneous. It seeks to understand whether – and to what extent – adult speakers might contribute to population-level sound change and whether pronunciation norms are more stable in large, linguistically diverse populations or small, homogenous populations. The proposed project renewal will expand an existing dataset to enable comparison between two separate cohorts of middle-aged adults (recorded a decade apart) and between speakers of different ages. This expanded dataset will enable a first empirical investigation into the extent to which community-level sound changes might be driven by changes within individual lifespans or across generations. Longitudinal data will be obtained by simulating interaction amongst members of the two communities using an agent-based model that has been shown to reliably predict longitudinal changes observed in real speakers. The outcome of the agent-based model will be used to assess whether population diversity impacts the rate and/or magnitude of sound change in individuals and at the group level.

• DFG Programme: Research Grants
• International Connection: Australia
• Cooperation Partners: Dr. Kathleen Jepson; Dr. Debbie Loakes