Women at Class Crossroads: A Critical Reply to Erikson and Goldthorpe's Note

Today's conjugal family presents some significant problems for empirical studies. Class analysts treating the family can choose between a) a neo-conventional approach using the spouse with the highest socioeconomic position as the determinant for family classification; b) an analysis based on individuals or c) an analysis based upon both spouses' positions in the world of work. Our most important objection to the conventional approach is that it only considers the contribution of one partner. In Erikson and Goldthorpe's work families are generally classified according to the male position, while economically active women are usually excluded in the analysis of the class position of conjugal families. The conventional approach constitutes both an empirical and a theoretical handicap for dealing with the issue of the family in class analysis. For instance, the conventional classification offers no option for studying class related conflicts within the family over such issues as children's education. Further, with a focus limited to the family level, conventionalists cannot hope to deal with the articulation of gender at the societal level in the class system. This articulation, as critics such as Stan worth (1984) and Marshall et al. (1988) note, has consequences for the distribution 'of lifechances, class formation and class action among both men and women alike' (Marshall et al. 1988:73). In our questioning of the conventional approach, we hope to demonstrate first that women's wage labour influences both their own, their spouse's and their family's life situation. Further we note that with higher female employment frequency and continuing gender segmentation of the labour market, one obtains a number of families with cross-class experience. Our second major point is that these partnerships differ in many crucial ways from class homogeneous families. Thus, the spouse's class is germane to any analysis of families in the class system. Analyses which ignore one partner are based on assumptions which are difficult to defend either theoretically or empirically. Erikson and Goldthorpe accuse us of not clearly supporting either model (b) or (c) nor suggesting some new model. We look for a class analysis where individuals occupy positions in mediating instances such as the family, unions, interest organisations and political parties (1987a:395). Such institutions provide at least a potential base for class action. Thus for us the conflict is not about the use of individuals versus families. We consider individuals as people stamped by their experience in families, as well as in other situations. We have approached our analysis by attempting to present individuals in the appropriate context for the problem under consideration, meaning that at times we consider individuals, and at times look at individuals as members of specific sorts of family or other constellations such as unions. We do not accept the criticism that we have