Date of Award


Document Type


Degree Name

Doctor of Philosophy (PhD)


Geography and Anthropology

First Advisor

M. Jill Brody

Second Advisor

M. Jane Collins


The literature concerning language, law, and power has traditionally focused on the linguistic strategies legal professionals employ for the purpose of controlling the speech of their clients and/or witnesses. However, newer research, stemming from gender-based theories of how power and/or powerlessness is manifest in language, promotes linguistic analyses from the perspective of the disempowered as opposed to the empowered, Such studies claim that speakers' election or avoidance of certain discourse features (e.g., hedges, hypercorrect forms, polite address forms, overlap, quotatives and gestures, etc.) and responses-types (e.g., declarative-context questions, copy responses, fragmented versus narrative responses, etc.) reflects their perceived power as a function of social characteristics such as status and gender. Taking excerpts of cross-examination sequences retranscribed from the court transcript of a single criminal trial as data, this work focuses on discerning evidence of power and/or powerlessness in the language of five witnesses. Methodology includes using discourse analytic methods including investigation of the patterning of linguistic features, the application of quantitative criteria established in the literature, and in-depth consideration of the discourse context. Results of both quantitative methods and context-sensitive analyses of the data are found to combine effectively in characterizing witnesses' speaking styles as being powerful or powerless. Witnesses' language evidences features of speech that have traditionally been designated as powerful. Additionally, in this study, witness language is presented as being proactive rather than as being invariably passive, and the choice of power-linked features by witnesses is shown to be motivated by individual objectives which supersede gender, social status, and other factors that are generally credited in the literature as predominantly influencing the presence or absence of power-linked features in witness language.