Abstract
Concept of "social role" is an integral part of social sciences since the 1930s up to this day, however since the 1960s it has been criticised from various positions. Ethnomethodology and conversation analysis (EM/CA), which inform the perspective of my paper, approach the topic in characteristic way.
Social role is re-specified as a member phenomenon based on categorization work, which is employed during interaction as a device and resource in accomplishing different interactional tasks. In other words, for EM/CA roles are not something given, presupposed, described and revealed by the social scientist, but a social accomplishment, which is being continually confirmed and sustained during the interaction; the fact that people in society "play different roles" is a natural knowledge of the members and they manage it more or less artfully.
My contribution aims to find out how the EM/CA findings in this area can be expanded by a secondary analysis of oral history interviews, specifically the recordings from USC Shoah Foundation's Visual History Archive (http://sfi.usc.edu). Roles (as membership categories) are used by the oral history interview participants as interactional devices on two levels: interactional and narrative.
In case of the interactional level, we speak of membership categories and categorization related to the roles of participants of the currently ongoing oral history interview. The interview itself is a membership categorization device, which is composed of role-derived pair categories "interviewer" and "narrator", but also the category of "listeners" (or "spectators").
In case of the narrative level, we speak only of the various past roles of the interviewed person. My paper will provide illustrations for both levels, consider their relations, similarities and differences, and mention some limitations of the data.