I’m interested in (socio)phonetic variation and sound change, and their implications for theories of phonological knowledge and the phonetics-phonology interface. I especially enjoy examining the articulatory and acoustic detail of sound changes in progress, to refine our insight into what is behind such changes, and also where and how the variation involved should be located in theories of speakers' knowledge of sound structure. Slightly more succinctly, I'd say I'm a LabPhonologist with variationist tendencies.
Work
My most recent research is on the use of creaky voice (or vocal fry) in the L2 English and L1 Dutch by speakers in a multilingual academic community (an undergraduate campus college) where English is the lingua franca. It's joint research with my colleagues Hielke Vriesendorp, Yosiane White and Hugo Quené. Our initial findings (that these Dutch speakers use creaky voice to roughly the same extent as L1 English speakers have been reported to in other studies, and that it's speaker-dependent, rather than language-dependent) have been published in the latest ICPhS proceedings volume.
I hope to also soon return to and write up the results of a study on voicing in Dutch, in particular that related to past tense allomorphy. Dutch speakers sometimes select a prescriptively incorrect allomorph for the regular (‘weak’) past tense (see Ernestus & Baayen 2001, 2003, 2004). Examining the production and perception of the past tense forms, Patrycja Strycharczuk and I found many mismatches between the rule-based predicted realisations and those actually produced, as well as considerable phonetic overlap in the production of voicing in these forms, and effects of lexical neighbourhood and frequency.
Patrycja Strycharczuk and I also collaborated on an ultrasound study of so-called derived geminates (and the potential resulting degemination) involving Dutch r. The resulting article was published in Journal of Phonetics (2018).
My PhD thesis (2015) was on the large-scale phonetic variation found with Dutch r. In it, I develop a model of progressive sound change to explain the origins, development and current status of Dutch r variation. To untangle the geographical, social and linguistic factors involved, I collected and analysed data from some 400 speakers (~20.000 tokens) in the Netherlands and Flanders. With Jim Scobbie at " Queen Margaret University (Edinburgh) I also collected a small-scale corpus of articulatory (ultrasound) data on Dutch coda approximant r, an innovative variant. The results of this study appeared as a book chapter (2010, see Publications and Presentations), and they are also discussed in my thesis.