i consider myself a phonologist, and within that field i am mainly concerned with prosodic phenomena such as
word stress and accent. my neverending subject of interest in that respect is modern greek because of its highly complex lexical
stress system. i am furthermore working on sri lankan malay prosody (which, as my colleague
sebastian nordhoff and i argue,
has no word stress at all) and ancient greek accent (together with
the learnability of phonological phenomena in combination with lexical information is my main subject of research. in my dissertation i developed an approach to the learning of underlying representations in line with the grammar, couched in a stochastic, bidirectional version of optimality theory developed by paul boersma. the theoretical framework i feel most at home in is optimality theory, although i start toying around with the use of harmonic grammar and its application to learnability issues since my stay at umass amherst (thanks to joe pater). in my one-year stay at umass amherst (financed by a rubicon fellowship of the nwo), i expanded my approach to segmentation problems as in e.g. tibetan numerals, and cases of allomorphy (e.g. as in english indeterminate articles and french liaison). together with silke hamann i modeled experimental findings on the L2 acquisition of phonotactic restrictions.