Colloquium

Title: Building and using every tool in the box to understand obviation and voice in Ojibwe
Christopher Hammerly, Linguistics, University of British Columbia

 

Abstract:

The goal of this talk is to give an overview of my lab’s recent efforts to create computational tools and resources, as well as use experimental methods, in order to document and understand various properties of Anishinaabemowin (AKA Ojibwe) grammar in both first speakers and learners. The aim of this body of work is not only to inform our general knowledge of human language and cognition, but also to directly and positively impact revitalization work. I will start by introducing the central grammatical systems of obviation and voice. Obviation organizes nouns according to prominence: the noun “in the spotlight” of the discourse is encoded as proximate, while all others are relegated as obviative. In turn, direct/inverse voice indicates argument structure relations in transitive predicates, fixing whether the proximate noun is the proto-agent (direct voice) or theme (inverse voice). Together, these systems are central to communicating basic sentences like “The dog is chasing the cat”, but they are also difficult for teachers to communicate and for learners to acquire. Using evidence from eye tracking studies with both L1 and L2 speakers, statistical patterns from original corpus studies (aided by computational tools including finite state transducers and constraint grammars), traditional fieldwork, and the theoretical framework of Optimality Theory, I will develop both a description and theoretical account of the complex patterns of obviation and voice in both fluent speakers and learners of Ojibwe.