The interaction of Phonology with Morphosyntax

Much of my work seeks to investigate the extent to which the phonological component of grammar interacts with morphology and syntax, and how to best model this interaction. I rely on data from a diverse set of understudied African languages such as Guébie (Kru), Amharic (Ethio-Semitic), Atchan (Kwa), and Moro (Kordofanian) to explore these questions. My dissertation, for example, focused on process morphology such as scalar tone shift and vowel replacement in Guébie, asking whether morphology, phonology, or the specific interactions between the two modules, are responsible for morphologically conditioned phonological alternations, and proposing a model integrating components of Distributed Morphology and morpheme-specific constraint-based phonology to account for these phenomena. Since then, together with Peter Jenks and Sharon Inkelas, I have been developing a framework which accounts for a wide range of morphologically and lexically specific phonological phenomena by applying cophonologies at syntactically determined phase boundaries, Cophonologies by Phase. Cophonologies by Phase makes specific, testable predictions about the domain of application of long-distance phonological processes.

In addition to my work on Guébie morphology and phonology, I am interested in cross-linguistic prosodic patterns. For example, I have described and analyzed syllable weight and stress systems of Amharic (Ethio-Semitic), and shown that reduplication in Amharic, while apparently weight-dependent, actually targets the stressed syllables at the stem-level of analysis. Such a system depends on cyclic phonological evaluation, where the stem is evaluated first, before the word as a whole.

I currently hold an NSF-DEL grant which funds field work trips for me and my students to Côte d'Ivoire to continue investigating what Guébie morphophonology can teach us about the interaction between phonology and morphosyntax in general. Specifically, I continue to explore the domains, syntactic or prosodic, within which phonological processes apply in Guébie.


Much of my theoretical research relies on data collected through fieldwork abroad or locally. Since 2013, I have been working with the Guébie community (Kru) in southwest Côte d'Ivoire to document and describe the grammar of the language. I have led several field trips to Gnagbodougnoa, Côte d'Ivoire, most recently with a team of students in the summer of 2019, funded by an NSF-DEL grant. Together with my students, an Ivoirian linguist, and Guébie native speakers, I am working to record a collection of Guébie texts and histories, many of which are available in the Guébie collection of the California Language Archive. Additionally, I have recorded and transcribed hundreds of hours of Guébie word lists and targeted elictation tasks. This is important because prior to this project, the Guébie language was neither documented nor described. My work in the village of Gnagbodougnoa since 2014 has given me the opportunity to explore the phonology, morphology, and syntax of the language, in addition to learning to speak the language and experience Guébie culture first hand. As a highly tonal, mostly monosyllabic language, Guébie morphology tends to involve systematic phonological changes to verbal and nominal roots, rather than concatenation of roots with affixes. This characteristic of Guébie and other Kru languages is thoroughly investigated in my dissertation and subsequent work. I plan to continue documenting and describing the Guébie language and exploring its linguistic properties, at the same time working with the community to develop communinty-oriented language resources as desired. I am currently writing a grammar of the language.

Documenting Guébie is also important for understanding the internal classification of Kru languages. Kru languages as a family are understudied, and the relationship between individual Kru languages is not well understood. Thorough documentation of Kru languages will contribute to a clearer picture of the Kru language-family tree.

In addition to my work with Guébie, I have worked to document other languages spoken in Abidjan, Côte d'Ivoire, namely Atchan (Kwa), Lobi (Gur), and Nouchi (an urban contact language). From the US, I have worked with speakers of a number of other African languages on smaller-scale documentation projects, including Moro (Kordofanian), Dafing (Mande), Nobiin (Nilotic), Ga (Kwa), and Amharic (Ethio-Semitic).

Computational tools for Language Documentation

As a linguistic fieldworker, the lack of technologically advanced language documentation tools is quite apparent to me. I am working to fill this gap by creating online database structures useful for my own work, but also available to other linguists. For example, I co-developed an application with a software engineer, a novel online, searchable, data-entry site called TwistedTongues, which I use to maintain the Guébie (Kru) corpus, and as a database tool in Field Methods courses. TwistedTongues is available to for use to anyone, and a manual is availalble here. I have also created a readable, searchable, publicly available database of Moro (Kordofanian), which is used by community members and linguists alike. The Moro database code is available on GitHub.

Areal linguistic features of Africa (ALFA)

The distribution of linguistic features can tell us much about the interaction of linguistic communities, and the historical relationship between languages and people groups. Together with a team of linguists at UC Berkeley, Princeton University, and the University of Delaware, I am working to create a database of linguistic features, syntactic, morphological, and phonological, in languages throughout Africa. I have been involved specifically in investigating the distribution of certain tonal and word order features in Africa. Current work focuses on so-called STAMP morphs in the Macro-Sudan Belt. As part of this project (ALFA), I have collaborated with linguists and computer scientists to create an online mapping software, which plots the languages in our database on a map, pulling latitude and longitude, as well as language family information from glottolog. The result is a straightforward way to visualize the distribution of linguistic features.