Haig: gender as a linguistic category 1 Manuscript of a paper held at the Workshop on Work in Progress: Kurdish Gender Studies. Katzow, 23-25.09.2000. This paper has been published in a German translation as: Haig, G. 2004. Das Genussystem in der kurdischen Sprache: strukturelle und soziolinguistische Aspekte. In: Hajo, S., Borck, C., Savelsberg, E., Şukriye, D. (eds.) Gender in Kurdistan und der Diaspora. Münster: Lit, 33-58. The gender system in the Kurdish language: structural and sociolinguistic aspects 1 Geoffrey Haig 1 Introduction Gender can be investigated as (at least) a biological, a social, and a linguistic category, but there is seldom a simple one-to-one matching of categories at different ontological levels. The focus of this paper lies squarely on gender as a linguistic category. To avoid false expectations, I should make it clear that in linguistics, the term "gender" has a rather specialised meaning which cannot simply be equated with the various readings of "gender" in other social sciences―see §1. Thus the bulk of this paper is concerned with what may appear to be a somewhat abstract discussion of formal grammatical structures. Nevertheless, I believe that the prerequisite for examining broader issues of gender and language as they relate to Kurdish is the correct analysis of grammatical gender in Kurdish, and the main objective of this paper is to propose such an analysis, focussing on the phenomenon of common gender in the Kurmanjî, or Northern, dialect of Kurdish (§3). In §5 I will turn to some possible avenues for interdisciplinary research in the fields of Kurdish gender studies, where language issues may provide valuable insights into the social category of gender. The discussion remains of necessity very brief, because as far as I am aware, there has been as yet no linguistic research on Kurdish that specifically targets gender 1 I am very grateful to Friederike Braun, Michael Chyet, and Michiel Leezenberg for critical comments on earlier versions of this paper. The sole responsibility for the remaining errors is of course my own.