Some Problematic Aspects of Communication in Cross Racial Research in the United States

I don't know whether the interview fragment reproduced above actually occurred. I have no difficulty in believing that it could have. My informant claimed that it actually occurred during an interview in the Bay Area-with a white, working class respond ent. Social scientists have long been aware of the risks in attempt ing to find out things from people simply by asking them. There is some question whether most past attempts to resolve problems in the ethnography of interrogation have been sufficiently successful. More particularly, how much confidence can we have in some research based on powerful statistical analyses of responses from large numbers of respondents? There is, moreover, reason to be skeptical even of some findings which result from intensive work with small numbers of informants; this is true even in research where there is reason to believe that the researchers have gained the confidence of their informants. In two recent papers I have attempted: (1) to suggest to sociologists some of the ways in which language data (particularly speech behavior) can help sociologists to ask better questions-and obtain better answers-about such things as social conflict and social control (Grimshaw, 1969b) and; (2) to outline some of the ways in which the failure to take language and speech behavior into account can invalidate social research which might otherwise be very valuable (Grimshaw, 1969a). The second article (on the