Research
Research
I am a phonologist interested in how sound systems maintain, lose, and reorganize contrasts. My work focuses on prosodic structure and sound change, especially how phonetic cues such as duration and F0 become organized into phonological systems such as tone, quantity, tonal accent, moraic structure, and metrical contrasts.
A central question in my research is why some prosodic systems are possible but typologically rare. I ask whether such systems reflect universal grammatical constraints, or whether they arise as low-probability outcomes of production, perception, learning, and diachronic transmission.
Primary research direction
Prosodic Structure and Cue Reorganization
This is the central direction of my current research. I investigate how duration, F0, and prosodic structure interact in the emergence of tone, quantity, tonal accent, and metrical contrasts. I am especially interested in cases where prosodic systems appear to reverse expected cue relations or maintain unusually dense contrasts.
A related part of this work asks how foot structure can organize syllable-internal positional asymmetries. I am especially interested in cases where different subsyllabic constituents follow different dimensions of metrical structure: rime and moraic licensing may track foot headhood, while onset licensing may track foot-edge position.
One idea that motivates this work is that phonetic cues do not simply replace one another during sound change. Instead, cues can be reorganized. For example, duration may first condition the emergence of pitch movement, while later tonal categories may develop their own durational implementation. Such cases provide a way to ask how phonetic cue relations become prosodic grammar.
Modeling and mechanism
Modeling Contrast Stability
This line of work asks why some phonological contrasts persist, merge, or reorganize over time. I use computational models and experimental evidence to study how production biases, perceptual ambiguity, lexical structure, and functional load shape the stability of contrasts.
My current work in this area focuses on Cantonese tone near-merger. More broadly, I ask how production and perception can become misaligned, and how such misalignment affects the long-term dynamics of contrast maintenance, merger, and re-stabilization.
Empirical domain
Tone and Laryngeal Phonology in Southeast Asia
Another strand of my work investigates the interaction of tone, voicing, F0, and prosodic domains in Southeast Asian languages. These projects ask how laryngeal and tonal cues are represented, how they interact with prosodic structure, and how they can be reanalyzed through sound change.
This strand also includes earlier work on non-native lexical tone imitation and computational evaluation. In that work, machine-learning classifiers trained on native tone productions were used to evaluate how non-native speakers reproduce Thai tonal categories, linking tone phonetics, second-language speech, and computational methods.
Secondary research strand
Quantitative Morphology and Lexical Structure
In addition to my work in phonology, I study lexical structure and word formation using corpus and quantitative methods. This work asks how morphological systems organize semantic relationships, productivity, and lexical family structure.
My current project in this area investigates Khmer word formation, comparing historical prefixation with modern compounding. The project asks whether so-called fossilized prefixes are semantically random or whether they retain measurable structure, and how this compares to productive compounding.
Methods
Methods
Across these projects, I combine formal phonological analysis with quantitative and experimental methods. My work uses acoustic analysis, corpus methods, statistical modeling, artificial-language learning, and computational simulations of sound change. I treat these methods as tools for answering phonological questions: how contrasts are represented, how they change, and why some systems are common while others are possible but rare.