Bruce Hayes                 Bruce P. Hayes                      

Theresa M. and Henry P. Biggs Centennial Term Chair in Linguistics [what's this?]

Dept. of Linguistics
Los Angeles CA 90095-1543

Image of my email address

Office hours for Fall 2022:  Monday 1-12, Thursday 4-5.

My office is 2101G Campbell Hall. I can also Zoom with you with advance notice.

Visit the Gallery of Wug-Shaped Curves. These are data plots that look like wugs, based on data from several fields of linguistics. Paper is now out in Annual Review of Linguistics.



Research/Downloadable Papers

Summary of current work  

In our current research, my collaborators and I approach a single phenomenon with three methods in parallel: (i) data analysis in the classical tradition of generative grammar, using rules and constraints; (ii) experimentation, to assess productivity and generality of phonological knowledge, (iii) modeling: machine-implemented algorithms, incorporating elements of phonological theory, learn the grammar through examination of a data corpus. The idea is to study not just the data pattern of the language, but to determine more precisely what the native speaker knows and demonstrate through modeling how she might come to know it. These goals have always been central to generative linguistics; advances in both theory and technology now help us address them more directly.

On the more formal side, I'm involved in efforts to assess models of variation in language by looking at mathematical properties of the patterns they generate. To my knowledge, these patterns represent a (possible) class of linguistic universals not previously studied. In the papers listed below, see Hayes (2022), Zuraw and Hayes (2017),  and Hayes (2017).

I also work from time to time on the analysis of poetic meter, where many of the tools of formal phonology have proven helpful.  Here is a brief summary of this work.

My CV can be accessed here.

Currently active projects: (a) modeling how infants learn English suffixes without knowing English; with Canaan Breiss, Megha Sundara, and Mark Johnson. (b) learning underlying representations, with a blend of math and phonological theory; with Yang Wang. (c) Simultaneous learning of syntax and phrasally-active phonological constraints, with Tim Hunter and Canaan Breiss. (d) A squib with Aaron Kaplan pointing out some intriguing quirks of Noisy Harmonic Grammar; (e) A guide for users to MaxEnt phonology, with a spreadsheet emphasis.



Talk handouts and slides

Handouts whose content is in the papers listed above are mostly not included here.  Some of these talks you can watch as web-posted video.