Course Description
Opacity is a central notion of generative phonology and its interfaces with historical linguistics, language acquisition, and learnability, and figures prominently in debates concerning rule- and constraint-based approaches to phonology. The goals of this seminar are to provide an introduction to phonological opacity, review properties of core examples, and discuss two ongoing research threads. The first thread aims to formally characterize how processes interact, delimiting the typology of and defining the precise formal relationships between pairwise interactions. The second thread articulates two related ideas. The first is to develop purely extensional characterizations of opacity, independent of grammatical framework. The second is that the proverbial challenge of opacity for constraint-based approaches is due to specific, often implicit constraint assumptions. Making these assumptions fully explicit allows proposals that have been developed to accommodate opacity to be systematized within a coherent formal framework. Both threads focus on better understanding the formal underpinnings of opacity, rather than on accounting for specific opaque patterns through specific tricks within specific frameworks. The course presupposes a working familiarity with rule-based and constraint-based phonological formalisms; by the end of the course, participants should be able to dive into both the classical and more recent literature on phonological opacity.
Area Tags: Phonology, Linguistic Frameworks, Learning Theory, Mathematical Linguistics
(Session 1) Monday/Thursday 9:00am – 10:20am
Location: ILC S140
Instructors: Giorgio Magri & Eric Bakovic
Giorgio Magri studied philosophy and mathermatics at the University of Milano before completing his PhD in Linguistics at the Massachusetts Institute of Technology in 2009. Since 2012, Magri is a permanent researcher at the French CNRS, based in Paris. Magri is currently on leave as a visiting scholar at MIT where he is teaching a seminar on models of probabilistic phonology. Magri has worked on semantics and then on learnability in constraint-based phonology. Recently, his interests in phonology have extened to typological analysis of probabilistic sound patterns and he works on constraint-based models of probablilistic phonology.
Eric Bakovi? is a Professor of Linguistics at UC San Diego, with a BA from UC Santa Cruz and a PhD from Rutgers University. His research interests lie in phonological theory generally, and specifically in formal models of interaction between phonological generalizations. His recent collaborative work investigates the outer limits of the set of possible phonological patterns (with Adam McCollum, Anna Mai, and Eric Meinhardt), specifies an algorithm for equalizing the typological predictions made by Harmonic Grammar and Optimality Theory (with Anna Mai), and defines a formal typology of process interactions and their relations to one another (with Lev Blumenfeld).
