Much of my research deals with the variability inherent in language. I am particularly interested in how language-internal factors influence sound change and what this tells us about how variation is represented in speakers’ grammars. Alongside my doctoral research into northern English velar nasals, I am involved in a project investigating diachronic and synchronic frequency effects in Manchester /t/-glottalling, conducted alongside Ricardo Bermúdez-Otero, Maciej Baranowski and Danielle Turton, which engages with important issues of phonological representation.
I’m also interested in articulatory phonetics, and in particular the mapping between acoustics and articulation. I am currently involved in an ongoing research project (alongside Stephen Nichols) using ultrasound tongue imaging, lip camera recording and electromagnetic articulography to investigate /s/-retraction in British English, which explores how different articulatory mechanisms contribute to the development of this sound change. Read about it here and here!
I also work a lot with large corpora of Twitter data as an innovative tool to investigate regional variation. I’ve worked on regional patterns of ‘dialect writing’ on Twitter to explore how users portray their spoken dialect on social media, and I’ve also delivered workshops on how to collect and analyse Twitter data for linguistic research (you can access materials from the latest one here).
I love data visualisation, and this goes hand in hand with my interest in dialectology and regional variation. I’ve been involved in a cool project with Laurel MacKenzie and Danielle Turton mapping regional variation throughout the UK and comparing these contemporary patterns of variation with earlier dialect surveys to track linguistic change in the dialects of the British Isles. I’ve put together an interactive dialect map website, which you can view here, and we’ve also published some of our initial results in the Journal of Linguistic Geography.
As a frequent user of Forced Alignment software like FAVE, I have looked into the possibility of using such methodological tools to automate the detection of sociolinguistic variation. I’ve presented on this topic at NWAV (read about it here and here!) and I’m currently pursuing further ways to increase efficiency in variationist analysis and improve the accuracy of speech~text alignment.