Ceske Budejovice11

Radically substance-free phonology

Sylvia Blaho

In this course, we will investigate the formal properties of phonological representation and computation, based on my 2008 dissertation (http://ling.auf.net/lingBuzz/000672). The starting point of the approach taken is that these can and should be investigated independently of the effect that extraphonological factors, most notably phonetics, have on the shape of individual phonologies.

We'll discuss different formal aspects of phonological representations, and consider a model using privative indexical features that can freely enter into feature geometrical dependency relations with one another. We'll compare this model to some of its predecessors, most notably, Hale & Reiss's binary feature model, Dresher, Rice et al.'s contrast-based model, Element Theory used in Government Phonology (cf. Harris), and the Parallel Structures Model of Morén.

The representations proposed here are integrated with an Optimality Theoretical model of computation, and constraint schemas governing featural interactions are discussed. The working of the model is illustrated by three case studies: Slovak sandhi voicing, Hungarian voicing assimilation, and Pasiego Spanish vowel harmony.

Here are the handouts: 1, 2, 3, 4, 5