When a dialect undergoes phonological change, it results in a different cognitive representation of language - and yet not only are different generations able to understand each other, they often don't even realize a dramatic change has occurred! My dissertation research focuses on one aspect of phonological change, by asking how structural differences are represented and produced by individual speakers. My findings suggest that during phonological change, speakers born during the community's transition generation learn and produce both phonological systems, suggesting that phonological systems act as a linguistic variable and that phonological change propagates through grammatical competition, akin to syntactic change.

Large corpora of sound files is, for me, one of the most exciting aspects of being a sociophonologist today. Big data sets allow us to ask more nuanced questions about sound change and variation. I've worked with the Origins of New Zealand English (ONZE) corpus, applying automatic vowel extraction and normalization methods (FAVE) to enable better comparisons across speakers in that corpus. I am also involved in collecting and analyzing data for the Philadelphia Neighborhood Corpus (PNC), and regularizing data in the PNC to make it more accessible to folk less experienced with code. I was also one of the project managers for the Influence of Higher Education on Local Phonology (IHELP) corpus.

As part of the Philadelphia Neighborhood Corpus project, I conducted a series of ethnographic sociolinguistic interviews in a neighborhood in Philadelphia. White male participants produced high rates of (TH)-fronting (a feature of African American English), which I analyze as borrowed from African American English as a second-order indexical variant signalling street orientation. This data is a classic example of linguistic diffusion in adult speech, in which speakers borrow a phonological feature but simplify its phonological conditioning.

Artificial language games enable researchers to investigate aspects of language learning, dialect contact, and signalling convergence that would otherwise be impossible to test in real world situations. In collaboration with Gareth Roberts, we conducted artificial language experiments to test hypotheses that emerged from my ethnographic work on dialect contact.