Constraint/feature inference

Collaborator: Kevin McMullin

The traditional view that phonological constraints are stated over featurally defined natural classes of segments has been challenged on the basis of both experimental evidence and learnability issues. I investigate what alternatives might exist to natural class-based constraints, as well as how such constraints could be inferred from naturalistic data. I have proposed segment classes based on quasi-cliques, sets of related segments based on a similarly named concept from graph theory, and shown their usefulness in moving beyond the explanatory power of natural classes. More recently, my interest in learning segment classes has expanded to learning classes that can constitute the "tiers" of tier-based strictly local languages.

Papers and presentations

Allen, Blake. Evaluating distance-based alternatives to natural class-based constraints. Poster presentation. Annual Meeting on Phonology 2014. Cambridge, Massachusetts: September 21, 2014.

(See also papers and presentations on my TSL project page.)

Online resources

Constraint and prediction files, plus testing data from Hayes & Wilson (2008)