MRPG: Word Domains in Athabaskan

Swarthmore College, Swarthmore PA

Investigators

Abstract

The purpose of this planning grant is to provide infrastructure for long-term research on Athabaskan languages, with a focus on Tahltan, a critically endangered language of northwest British Columbia. A set of data collection, organization and analysis activities are proposed to study the structure of verbs. Verb forms are fundamental in the study of Athabaskan languages because they provide crucial information about the structure and interpretation of sentences (i.e., the syntax and semantics). They are also important in the description of sound structure (or the phonology) because the different domains in an Athabaskan verb represent different sound systems. Verbs are thus fundamental to subsequent documentation of Tahltan, i.e., dictionaries, texts, grammars, and theoretical hypothesis testing, all of which fall within the scope of this long-term research program. Verbs in Athabaskan languages are made up of a stem, which contributes the principal meaning of the word, and two classes of prefixes, namely the disjunct and conjunct prefixes. The disjunct prefixes form a class with stems because they are lexical items and have the full range of phonological structures. These properties distinguish disjunct prefixes and stems from the conjunct prefixes, which are not lexical items and have a highly restricted set of phonological structures. One chief aim is to give a natural account of the differences between these two morphological classes, for both the lexical/non-lexical distinction and their different phonologies. A set of hypotheses are developed in Optimality Theory that explain this correlation in terms the adherence of different morpheme types to their underlying representation. A host of predictions are made concerning the range of possible phonological structures and the types of processes expected in each word domain. The research plan begins with the study of Navajo verbs, both to evaluate the optimality theoretic analysis in a well-documented Athabaskan language, and as a way to provide a framework for primary linguistic documentation of Tahltan. Existing sound recordings of Tahltan will be digitized, and the fieldnotes and word lists associated with them will be incorporated in a database of large corpora that can be searched electronically. This database will clarify specific research questions and provide a guideline for additional data collection. Three trips in the field are planned to confirm the data and generalizations of previous work and to supplement the study of verb forms. The resulting relational database and grammatical sketches will provide the first complete description of verbs and materials for testing specific hypotheses about the morphological and phonological structures of verb forms. In addition, the lexical database will make possible a variety of other research and teaching activities that extend beyond the scope of verb structure.