Department of Linguistics, California State University. Standards of Evidence, Ethnographic Methods and Analysis Abstract/Summary
Drawing on the author's own research on Corsican bilingual schools, this paper explores the relationship between ethnographic research practices and sound standards of evidence for analytical claims. The presentation will consider time, breadth, intensity, and systematicity. I hope to illustrate how sustained and systematic observations over time make it possible for the ethnographer to assess behaviors/events at a specific moment in light of histories of interaction, particulars of relationships between people, histories of individual practice, explicitly expressed philosophies or ideals—not to mention histories that go beyond the ethnographic moment (for example, personal and professional trajectories and patterns of socialization or apprenticeship; intergroup relations etc.). Time depth also allows the ethnographer to make empirically grounded assessments of whether a particular utterance or event or interaction is widespread/habitual/representative vs. relatively unusual—either for an individual, or for a collectivity. Both the unusual and the habitual can be telling examples, but they are telling in different ways, and it is crucial to be able to make appropriate distinctions. The second them, breadth of ethnographic research, clearly relates to an ideal that can only ever be partially realized, particularly in complex societies: that the ethnographer's goal should be to try to come to a grasp of the wholeness of the lives of the people being studied. That means attempting to observe those people in multiple contexts, or to observe multiple aspects of a particular setting (for example, an institutional one). The impossibility of doing this fully is one of the constant frustrations of the ethnographic project, but it is one that should be an enduring reminder of the limits of any individual ethnographer's understanding. Finally, I want to emphasize that though there are many emergent and contingent dimensions of ethnographic research, ethnographic observations, interviews and other data collection can be held to standards of rigorous, systematic research design.
Example: in a forthcoming article1, I analyze the distribution of pedagogical and social functions across Corsican and French in a French literacy lesson for 5 and 6 year olds. The analysis shows that the teacher's use of French is confined almost exclusively to voicing the written text, and that most significant pedagogical work is done in Corsican. I interpret this as a conscious effort to position Corsican as a legitimate language, and to counterbalance the dominance of French in the society at large and in the children's daily lives. At the same time, I note that the this counterbalancing is not done in a way that squeezes French out of the lesson; on the contrary, it is there in significant proportions and is the language of formal literacy. I interpret this as one of the ways in which the teacher acknowledges children's French competencies and makes them matter in the classroom. My general point is that we can view the teacher's language choice as a form of stance-taking that simultaneously positions her vis-a-vis the two languages of the classroom and positions those languages vis-a-vis each other. In this analysis, I draw on the following ethnographic sources of evidence, which depend crucially on observations of the same people, over time, in a variety of contexts that are situated historically, socially and politically.