ly, analysts may define workers in relation to the means of production, but the path from objective position in the relations of production to political consciousness and identity is twisted and can lead to multiple outcomes. The peril of overlooking the process through which class is socially constructed can perhaps be seen most clearly in South Africa, where white workers used access to the ballot box to guarantee their privileged position toward black workers. White workers rallied under the banner: Workers of the world unite and fight for a White South Africa.32 The point is not simply that race can weaken class cohesion. Indeed, phrasing the relationship between race and class in this manner assumes the causal priority of class over race. Conceptualizing race in this manner-as something that divides classes-prevents scholars from integrating race more meaningfully into their analyses. The basis for excluding racially subordinated groups from the franchise was race, not class, and a conceptualization of race that assigns it importance only insofar as it divides classes is unlikely to provide much illumination about the processes through which these groups are excluded and later included into the democratic polity. Perhaps the worst result of Rueschemeyer, Stephens, and Stephens' use of race is that, by acknowledging its importance only when it overlaps with class, they doubly exclude African-American women, once by defeminizing class and twice by defeminizing race. Gender and Race as Categories of Analysis While defining democracy as universal adult male suffrage limits the scope of the variables considered as relevant causal factors, there is a more fundamental point about how analytical categories are constructed. Rueschemeyer, Stephens, and Stephens could grant that their theory explains only a small part of the story of democratization and that separate studies are needed to explain women's incorporation. Ultimately, Rueschemeyer, Stephens, and Stephens lament that analysis of female suffrage would require another comparative since there is little correlation between the extension of suffrage across class and across lines.33 Their analytical error is to assume that, since the group being incorporated in the study is men, they can avoid dealing with gender. As Terrell Carver has observed, gender is not a synonym for women.34 Gender, defined as the social organization of sexual difference, includes both men and women. The absence of women in the definition of the dependent variable, then, does not mean that played no role in episodes of democratization that enfranchised men. As the work of feminist theorists has shown, played an intimate role in defining who counted as a citizen; therefore infused democratic institutions from the beginning, since the abstract individual-the basis for citizenshipwas male.35 The abstract individual of liberal theory was from the start deeply imbued with and race, and certain categories of persons were not defined as
Read full abstract