Reflection: Beyond sociological variables

Note that Brown and Levinson’s book was published in a sociolinguistics series. It is perhaps not surprising then that the methodological flavor of the dominant sociolinguistics paradigm, that of Labov, with its emphasis on quantification had affected subsequent politeness studies (and we should note the role of studies in social psychology, which traditionally

emphasize quantification). Numerous researchers began administering questionnaires (a favorite though not the only methodology) to quantify the kind of politeness strategies used by people of different relative power, social distance and so on (see Spencer-Oatey 1996, for many references). In general, the studies confirmed Brown and Levinson’s predictions for power and to a lesser extent social distance, but generally avoided testing ranking. However, the basis of these studies is now being questioned. Spencer-Oatey (1996) demonstrated that researchers varied widely in what is understood by power or social distance. In fact, these variables were subsuming other independent variables. Baxter (1984), for example, showed that the attitude affect (i.e. whether there is liking or disliking between participants) was getting muddled up with social distance, despite the fact that it is an independent variable. More fundamentally, research on social situations and context generally has moved on. Social values, it is argued, are not static but dynamic, and they are not given values (i.e. known by participants) but negotiated in interaction (e.g. I may start by assuming that somebody is more powerful than me but re-evaluate that in the course of an interaction). To be fair to Brown and Levinson, they did acknowledge this vision, stating that values on their variables “are not intended as sociologists’ ratings of actual power, distance, etc., but only as actors’ assumptions of such ratings, assumed to be mutually assumed, at least within certain limits” (1987: 74–76; original emphasis). But they did not back this vision up with a suitable methodology (one which is more qualitative in nature and thus able to handle the complexity), and certainly subsequent researchers chose to ignore it.