Sociolinguistics and its metalinguistic paradox: a plea for history

Academic disciplines in the humanities and social sciences seek to develop metalanguages that have analytical purchase across contexts, domains, cultures and language zones. Sociolinguistics for example seeks to make... [ view full abstract ]

Academic disciplines in the humanities and social sciences seek to develop metalanguages that have analytical purchase across contexts, domains, cultures and language zones. Sociolinguistics for example seeks to make socially-relevant generalizations, and to that end it requires methodologies and terminologies that transcend a particular domain or social context. And yet, sociolinguistics is committed to respect for locally generated meanings and categories, and pluralistic understandings of human communication. Paradoxically, increasing success in one direction seems to imply greater failure in the other.

One way to understand the dilemma this paradox represents is in terms of an opposition between a scientific or methodologically objective sociolinguistics as opposed to a discursive, reflexive or (loosely) ‘postmodern’ one. The origins of post-war scientific sociolinguistics lie in the attempt to merge dialectology and historical linguistics (as in the classic 1968 paper by Weinreich, Labov and Herzog, ‘Empirical foundations for a theory of language change’), and secondly in the introduction of class as a social dimension to which observable linguistic phenomena are held to correspond. Both these innovations constituted critiques directed at nineteenth century historical linguistics, specifically the Neogrammarian model of language change, and were ‘sociolinguistic’ in the sense that one could, it was argued, study language change ‘on the hoof’, plotting it against social parameters such as age, race, gender and class. In a different (but for the purposes of this discussion analogous way), the methodologies of ethnomethodology and CA involve a methodologically empowered observer. The presumption is that locally generated, interactional meanings are available to the investigator based on observable features of interaction.

The dilemma is however most acute for discursive or reflexive sociolinguistics, since it does not see itself as a purely observational science, and is shaped by social constructionism. One possible endpoint to this ideologically-driven constructionism is skepticism, that is, a position in which the observer doubts the status even of their own experientially based reports of the social reality they have set out to capture. In highly contested domains of social action, or in dealing with radically disempowered communities, the outsider may need to cede terminological priority to the insider, and the politics of social description in such cases is constantly threatened by the charge of appropriation or misrepresentation. Insiders who take an ethnographic stance face a complex of related problems involving self-positioning, inauthenticity, and representativity.

This complex of problems cannot be solved in any straightforward sense. However the issues sketched above are explored through two case studies. The first concerns the term ‘mother tongue’ and the assumptions that underlie post-Renaissance vernacularism; the second is the sex-gender distinction. The argument is made that sociolinguistics needs a deeper and more nuanced awareness both of its own disciplinary history and of history itself, and that it must abandon its quest for an ideal analytical metalanguage.