Sister Carrie, who opposes the sentimental idealism of the supposed hero, Ames, in portraits of immobile worlds dominated by rhetorics of social mobility (An American Tragedy [1925]), Dreiser reveals the prevailing social theory of his era to be the ideological handmaid to a basically unjust capitalist system. Dreiser's resistance to the naturalist assumptions embedded in liberal social theory brings us to the final set of literary examples to be considered: two writers who first embraced the practical potential of social science, and ended up more critical of its assumptions than any of the authors so far discussed. Yet however critical they became, W. E. B. Du Bois and Gertrude Stein remained attached to social science in ways that informed the works that are of concern to the history of the novel-The Souls of Black Folk (1903) and The Making of Americans (1906-8).
W. E. B. Du Bois and Gertrude Stein share the position of social marginals, as well as the experience of social scientific training. Both -204- were also self-exiles from American society: Du Bois settled in Ghana at the end of his life, Stein moved to Paris before she was thirty. Perhaps the most significant similarity is that both studied with William James and were heavily influenced by his pragmatist social science.
For Du Bois and Stein, typecasting was not an inevitable process but a political activity. Both saw the damaging effects of typecasting on their respective social groups and believed that greater control over their group's representation would extend its social possibility. They sought out the role of the expert cataloger of modern social life as a means of remedy and instruction.
As two writers who were personally implicated in questions of social difference and drawn to the promise of liberal social science, Du Bois and Stein represent powerful confrontations with the central intellectual concerns of their era: the seductive potential of categories and types, the social scientific conflation of knowledge and uniformity, individualism versus collectivism as competing ideals, the role of literature in relation to social science. They are distinctive, and crucial to our exploration, in having recognized the pivotal role that social science played in the modern era. While they were critical of this role, they also pursued it. This ambivalence toward the posture of social scientific expertise is built into the narrative personae of their two major novelistic works.
Of all the literary authors discussed so far, Du Bois is unique in actively combining sociological and literary methods. As a student of history and sociology at Harvard in the 1890s (with two years of study in Germany), Du Bois was drawn to the potential of this new discipline for arbitrating the problem of race in America. He more often found, however, that sociology was a symptom of the problem rather than a solution to it. Even the most enlightened of sociologists, W. I. Thomas, in a 1904 article, 'The Psychology of Race-Prejudice' (American Journal of Sociology), came perilously close to calling racial prejudice inherent. And F. H. Giddings's concept of 'the consciousness of kind,' which held that the sense of community inevitably diminished with the increase of racial and ethnic differences, was used to justify turn-of-the century schemes for the deportation of African Americans.
Du Bois's earliest training was in history and economics, culmi-205- nating in his dissertation The Suppression of the African Slave-Trade 1638–1870 (1896). The book's bias is historicist, that is, Du Bois focuses on the historical genesis of slavery in order to redress the condition of African Americans in his contemporary society. Like Billy Budd, Suppression is most drawn to the closing decades of the eighteenth century: Melville's post-Revolutionary era of consolidation matches Du Bois's post-Revolutionary America, the moment of enlightenment that managed to entrench the most oppressive of slave systems. 'There never was a time in the history of America,' wrote Du Bois, 'when the system had a slighter economic, political, and moral justification than in 1787, and yet with this real, existent, growing evil before their eyes, a bargain largely of dollars and cents was allowed to open the highway that led straight to the Civil War.' By delineating the economic considerations that consistently overshadowed the moral question of slavery, Suppression embodies the weight of historical memory that tempered Du Bois's faith in social instrumentalism. Any program of social action had to contend with the historical process that had created and still informed African American possibility.
Given Du Bois's career-long interest in the theoretical problem of racial difference and its relationship to conceptions of social evolution, it seems appropriate that his entry into social science was through history. Du Bois's historical approach is consistent with the methods of the era's classic sociological theorists, for whom sociological analysis required a broad command of different cultures as well as historical periods. Like Spencer, Durkheim, and Weber, Du Bois is interested in the transformation of societies, as well as in the persistence of certain ideas and habits over time. Du Bois differs from these analysts in attending to the ways in which social evolution occurs and also fails to occur as a consequence of deliberate social policy. Du Bois likewise departs from an essentially static evolutionary script (favored by Spencer and Durkheim) that projects a normative pattern of development and evaluates all populations according to that pattern. The culmination of Du Bois's training as a social scientist was his classic anatomy of African American society in Philadelphia.
The central drama of The Philadelphia Negro (1899) lies in Du Bois's effort to strike a balance between assessing the collective con-206- dition of Philadelphia's African Americans and distinguishing the various strata of that community, with their different relationships to American norms and values. His study contains the seeds of his growing dissatisfaction with social science while it lays the groundwork for the problem that would plague his career: how could a commitment to a collective African American destiny be accommodated to the promise of individual assimilation and progress dividing that collectivity? Sociological theories of stratification together with his continuing absorption in Spencerian ideas formed the unsettling core of Du Bois's method. His turn away from sociology following The Philadelphia Negro may have had as much to do with the ways in which it magnified an emerging contradiction in his own thought as with the limitations he saw in the discipline itself. In practicing sociology, he adopted the dominant sociological trajectory of his era: the supplanting of basically conservative, essentialist notions about human potential with a liberal ideal that emphasized assimilation and training. This new ideal, however, retained a fundamental tie to the essentialist view, in upholding a belief in 'the survival of the fittest.' The superior elements of any social group, went the argument, would inevitably rise and prosper. Given this sociological climate, it is not surprising that an outgrowth of Du Bois's Philadelphia study was his first conceptualization of 'the talented tenth,' an attempt to distinguish the best 'strata' of the African American race.
Du Bois's adaptation of these sociological principles for African American Philadelphia was timely, given a prevailing racial ideology of two nations, one white, one black, that relentlessly homogenized African Americans. Against this biological fiat of racial homogeneity, Du Bois set another biological fiat implicitly condoned by social science, which emphasized inherent differences of talent within each group. Du Bois thus used Social Darwinist ideas to challenge a prevailing racial ideology.
The irony is that his struggle against a white conspiracy that intentionally muffles African American achievements was mirrored by the response to his book. Through the reception (or more accurately, nonreception) of The Philadelphia Negro by the sociological profession-which took over half a century to confer its 'classic' status-Du Bois experienced firsthand the limits upon all African Americans. Du Bois's failure to gain a hearing as a sociologist sig-207- naled the failed promise of the discipline's liberal assumptions. In declaring his next major study, The Souls of Black Folk (1903), a work of 'faith and passion,' he seemed to be deliberately distancing himself from the rational agenda of The Philadelphia Negro.
In The Souls of Black Folk, Du Bois undertakes the imaginative reconstruction of the territory he covered in his sociological classic. His desire to gain control over the representation of African Americans will not be accomplished through the straitjacket of sociological method, he implies, but requires a more literary technique. If Philadelphia undertakes the work of social description, Souls undertakes the work of social change. Philadelphia selects among preexisting African American types assigned by the dominant society, while Souls surveys all the available African American types and, finding them wanting, begins to recover the powers of self-identification for African Americans themselves.
The two books are best seen as companion pieces, which need to be read together in order to understand their deepest implications. The striving Philadelphian bent on self-improvement joins the collectivity of African American souls. The insistence of Souls on plurality suggests Du Bois's new attitude toward the liberal individualism of sociology-it has never represented a true possibility for African Americans. Souls dismisses the claim that African Americans are individuals, the 'fittest' capable of assimilating and rising like any 'immigrant' group. What was deplored in Philadelphia, the
