My research applies phonetic methodologies to examine segmental and suprasegmental aspects of language, including understudied varieties as well as the speech of multilinguals. I test phonological descriptions of sound systems by conducting experiments, both to measure production and to determine what features of production are used by listeners in perception. My early work focused on f0, both lexical and sentential. My MA thesis measured the f0 and voice quality correlates of the tone system of Burmese, and my dissertation examined production and perception of the lexical pitch accent of the Trøndersk dialect of East Norwegian, thereby informing the phonological analysis of the contrast. I also examined how the lexical pitch accents are affected by contrastive focus, thereby teasing apart lexical pitch from the sentential and pragmatic pitch correlates. I conducted perception experiments with resynthesised stimuli to determine which aspects of production were used as perceptual cues to differentiate the accents.
I also collaborated on research into the tone system of Teotepec Chatino, an indigenous language of Mexico, described as having 13 lexical tones. I have also conducted research on the influence of linguistic experience on the perception of dental and alveolar stops (contrastive in Irish English) by speakers of Irish English and American English.
I recently published work on the intonation of Lebanese Arabic, examining phrase-final intonation as well as the interaction of tonal alignment with vowel length and syllable structure in this variety of Arabic. The goal of this whole project was to provide a detailed description of prosody in the Lebanese Arabic of Beirut, in terms of intonation as well as how this interacts with vowel length, and finally, to determine how this fits into the typological literature on intonation and text-tune alignment as well as linguistic variation.
At UTEP, I started to examine the production of /l/ sounds in different syllable positions and in words of different stress patterns, in both English and Spanish. Our goal was to investigate velarisation patterns (a typical American English pattern) among balanced bilingual speakers in the El Paso region. We found that these speakers showed different patterns in their two languages, with the English /l/ being more velarised overall, and an effect of syllable position only in English and not transferred to Spanish.
With a number of graduate students at AUB, I investigated the production of stops by monolingual speakers of Lebanese Arabic as well as bilingual speakers (L2 English). We are investigating the effects of L1 to L2 transfer, as well as examining to what extent highly proficient bilinguals may have L2 to L1 transfer effects, and also whether codeswitching affects their production.
As a postdoctoral researcher at the University of Graz, Austria (2015-16), I worked with Ineke Mennen using previously recorded data to examine the acoustic correlates of lexical stress in Welsh and Welsh English. Research on understudied varieties tends to involve multilingual speakers, thereby introducing possible influences of transfer and the emergence of contact varieties, which deserve a focus in the linguistic literature. I recently published collaborative work on the variety of Armenian spoken by a minority population in Lebanon, where we found that the production of obstruents differs from what has been described in the phonological literature. Traditionally, Western Armenian has been described as contrasting voiced stops and affricates with voiceless aspirated ones but no phonetic research had been conducted. In our novel study, we observed that voiceless stops and affricates are in fact unaspirated. Since these speakers all have Arabic as an L2, we then examined recordings of Western Armenian speakers from the US who do not speak Arabic, to tease apart influences of different L2s. We found that speakers in both groups produced voicing patterns that were in line with the majority language, that is, in Lebanon, speakers of Armenian had a voiced vs voiceless unaspirated contrast, while speakers in the US had a voiceless unaspirated vs aspirated contrast.
In sum, my research focuses on how predictions made by phonological descriptions - particularly of understudied varieties - are borne out phonetically, as well as informing typological literature by providing phonetic research on lesser-examined varieties. This naturally involves multilingual speakers so integrates an examination of transfer and code-switching, when present, as well how linguistic experience shapes processing. My research combines rigorous experimental methods and a deep interest in cross-linguistic and cross-varietal investigation.