Language change in multidimensional space

What is it about?

Our study demonstrates the use of multivariate analyses on big datasets to identify sociolects in a speaker community by finding clusters of language variables that are used distinctively by speakers grouped by different social factors. Grouping of individual language variation into distinct sociolects may drive the formation of new dialects and languages, so our study develops important statistical tools to linking the ‘micro-level’ processes of language change within a speaker community to the ‘macro-level’ outcomes of language divergence. We demonstrate the power of the multivariate analyses to identify sociolects using the Gurindji Kriol dataset that is of an unprecedented size in terms of the number of language variables, speakers, and social factors. We find solid evidence for the distinct use of the language across generations in the speaker community, which indicates the importance of peers in language change and the possible role of the sociolinguistic salience of variables in shaping this change.

Featured Image