In the past few years “equality-versus-difference” has been used as a shorthand to characterize conflicting feminist positions and political strategies. Those who argue that sexual difference ought to be an irrelevant consideration in schools, employment, the courts, and the legislature are put in the equality category. Those who insist that appeals on behalf of women ought to be made in terms of the needs, interests, and characteristics common to women as a group are placed in the difference category.1 In the clashes over the superiority of one or another of these strategies, feminists have invoked history, philosophy, and morality and they have devised new classificatory labels: cultural feminism, liberal feminism, feminist separatism, and the like.2 Recently the debate about equality and difference has been used to analyze the Sears case, the sex-discrimination suit brought against the retailing giant Sears, Roebuck & Company by the Equal Employment Opportunity Commission (EEOC) in 1978, in which historians Alice Kessler-Harris and Rosalind Rosenberg testified on opposite sides.
There have been many articles written on the Sears case, among them a recent one by Ruth Milkman. Milkman insists that we attend to the political context of seemingly timeless principles: “We ignore the political dimensions of the equality-versus-difference debate at our peril, especially in a period of conservative reaction like the present.” She concludes:
As long as this is the political context in which we find ourselves, feminist scholars must be aware of the real danger that arguments about “difference” or “women’s culture” will be put to uses other than those for which they were originally developed. That does not mean we must abandon these arguments or the intellectual terrain they have opened up; it does mean that we must be self-conscious in our formulations, keeping firmly in view the ways in which our work can be exploited politically.3
Milkman’s carefully nuanced formulation implies that equality is our safest course, but she also is reluctant to reject difference entirely. She feels a need to choose a side, but which side is the problem. Milkman’s ambivalence is an example of what the legal theorist Martha Minow has labeled in another context “the difference dilemma.” Ignoring difference in the case of subordinated groups, Minow points out, “leaves in place a faulty neutrality,” while focusing on difference can underscore the stigma of deviance. “Both focusing on and ignoring difference risk recreating it. This is the dilemma of difference.”4 What is required, Minow suggests, is a new way of thinking about difference and this involves rejecting the idea that equality-versus-difference constitutes an opposition. Instead of framing analyses and strategies as if such binary pairs were timeless and true, we need to ask how the dichotomous pairing of equality and difference itself works. Instead of remaining within the terms of existing political discourse, we need to subject those terms to critical examination. Until we understand how the concepts work to constrain and construct specific meanings, we cannot make them work for us.
A close look at the evidence in the Sears case suggests that equality-versus-difference may not accurately depict the opposing sides. During testimony, most of the arguments against equality and for difference were, in fact, made by the Sears lawyers or Rosalind Rosenberg. They constructed an opponent against whom they asserted that men and women differed, that “fundamental differences”—the result of culture or long-standing patterns of socialization—led to women’s presumed lack of interest in commission sales jobs. In order to make their own claim that sexual difference and not discrimination could explain the hiring patterns of Sears, the Sears defense attributed to the EEOC an assumption that no one had made in those terms—that women and men had identical interests.5 Alice Kessler-Harris did not argue that women were the same as men; instead she used a variety of strategies to challenge Rosenberg’s assertions. First, she argued that historical evidence suggested far more variety in the jobs women actually took than Rosenberg assumed. Second, she maintained that economic considerations usually offset the effects of socialization in women’s attitudes to employment. Wages were an incentive to take new, demanding, or atypical positions. And third, she pointed out that, historically, job segregation by sex was the consequence of employer preferences, not employee choices. The question of women’s choices could not be resolved, Kessler-Harris suggested, when the hiring process itself predetermined the outcome, imposing generalized gendered criteria that were not necessarily relevant to the work at hand. The debate joined then not around equality-versus-difference but around the relevance of general ideas of sexual difference in a specific context.6
To make the case for employer discrimination EEOC lawyers cited obviously biased job applicant questionnaires and statements by personnel officers, but they had no individuals to testify that they had experienced discrimination. Kessler-Harris referred to past patterns of sexual segregation in the job market as the product of employer choices, but mostly she invoked history to break down Rosenberg’s contention that women as a group differed consistently in the details of their behavior from men, instead insisting that variety characterized female job choices (as it did male job choices), that it made no sense in this case to talk about women as a uniform group. She defined equality to mean a presumption that women and men might have an equal interest in sales commission jobs. She did not claim that women and men, by definition, had such an equal interest. Rather Kessler-Harris and the EEOC called into question the relevance for hiring decisions of generalizations about the necessarily antithetical behaviors of women and men. The EEOC argued that hiring at Sears reflected inaccurate and inapplicable notions of sexual difference; Sears argued that “fundamental” differences between the sexes (and not its own actions) explained the gender imbalances in its labor force.
The Sears case was complicated by the fact that almost all of the evidence offered was statistical. The testimony of the historians, therefore, could only be inferential at best. Neither historian had much information about what had actually happened at Sears; rather each of them sought to explain small statistical disparities by reference to gross generalizations about the entire history of working women. Moreover, in a kind of parody of positivism they were forced to swear to the truth or falsehood of these generalizations that had been developed for purposes other than legal contestation and they were forced to treat their interpretive premises as matters of fact. Reading the cross-examination of Kessler-Harris is revealing in this respect. Each of her carefully nuanced explanations of women’s work history was forced into a reductive assertion by the Sears lawyers’ insistence that she answer questions only by saying yes or no. Similarly, Rosalind Rosenberg’s rebuttal to Kessler-Harris eschewed the historian’s subtle contextual reading of evidence and sought instead to impose a test of absolute consistency. She juxtaposed Kessler-Harris’s testimony in the trial to her earlier published work in an effort to show that Kessler-Harris had misled the court.7 (In court, Kessler-Harris argued that history showed a profusion of job choices by women; her book suggested that women “preferred” work that could be combined with domestic responsibilities.)8 Outside the courtroom, however, the disparities of the Kessler-Harris argument could also be explained in other ways: in relationship to a labor history that had typically excluded women, it might make sense to overgeneralize about women’s experience, emphasizing difference in order to demonstrate that the universal term “worker” was really a male reference that could not account for all aspects of women’s job experiences. In relationship to an employer who sought to justify discrimination by reference to sexual difference, it made more sense to deny the totalizing effects of difference by stressing instead the diversity and complexity of women’s behavior and motivation. In the first case, difference served a positive function, unveiling the inequity hidden in a presumably neutral term; in the second case, difference served a negative purpose, justifying what Kessler-Harris believed to be unequal treatment. Although the inconsistency might have been avoided with a more self-conscious analysis of “the difference dilemma,” Kessler-Harris’s different positions were quite legitimately different emphases for different contexts; only in a courtroom could they be taken as proof of bad faith.9
The exacting demands of the courtroom for consistency and “truth” point up the profound difficulties of arguing about difference. Although the testimony of the historians had to explain only a relatively small statistical disparity in the numbers of women and men hired for full-time commission sales jobs, the explanations that were preferred were totalizing and categorical.10 In cross-examination, Kessler-Harris’s multiple interpretations were found to be contradictory and confusing, while the judge praised Rosenberg for her coherence and lucidity.11 In part, that was because Rosenberg held to a tight model that unproblematically linked socialization to individual choice; in part it was because her descriptions of gender differenees accorded with prevailing normative views. Kessler-Harris, in contrast, had trouble finding a simple and singular model that would at once acknowledge difference and refuse it as an acceptable explanation for the Sears employment pattern. So she fell into great difficulty maintaining consistency in the face of hostile questioning. On the one hand, she seemed to be assuming that economic opportunism equally affected men and women (and thus that men and women were the same). How then explain the gender differences her own work had identified? On the other hand, she was tarred (by Rosenberg) with the brush of subversion, for implying that all employers might have some interest in sex-typing the labor force, for deducing from her own (presumably Marxist) theory a “conspiratorial” conclusion about Sears’s behavior.12 If the patterns of discrimination that Kessler-Harris alluded to were real, after all, one of their effects might well be the kind of difference Rosenberg pointed out. Caught within the framework of Rosenberg’s categorical use of historical evidence, Kessler-Harris and her lawyers relied on an essentially negative strategy, offering details designed to complicate and undercut Rosenberg’s assertions. Kessler-Harris neither challenged the theoretical shortcomings of Rosenberg’s socialization model nor offered an alternative model of her own. That would have required, I think, either fully developing the case for employer discrimination or insisting more completely on the “differences” line of argument by exposing the “equality-versus-difference” formulation as an illusion.
In the end the most nuanced arguments of Kessler-Harris were rejected as contradictory or inapplicable and the judge decided in Sears’s favor, repeating the defense argument that “an assumption of equal interest was unfounded” because of the differences between women and men.13 Not only was the EEOC position rejected, but Sears’s hiring policies were implicitly endorsed. According to the judge, since difference was real and fundamental, it could explain statistical variations in Sears’s hiring. Discrimination was redefined as simply the recognition of “natural” difference (however historically or culturally produced), fitting in nicely with the logic of Reagan conservatism. Difference was substituted for inequality, the appropriate antithesis of equality, becoming inequality’s explanation and legitimation. The judge’s decision illustrates a process literary scholar Naomi Schor has described in another context: it “essentializes difference and naturalizes social inequity.”14
The Sears case offers a sobering lesson in the operation of a discursive, that is a political, field. It provides insight not only into the manipulation of concepts and definitions but also into the implementation and justification of institutional and political relationships. References to categorical differences between women and men set the terms within which Sears defended its policies and the EEOC challenged them. Equality-versus-difference was the intellectual trap within which historians argued not about tiny statistical disparities in Sears employment practices but about the normative behavior of men versus women. Although we might conclude that the balance of power was against the EEOC by the time the case was heard and that, therefore, its outcome was inevitable, part of the Reagan plan to reverse affirmative action programs of the 1970s, we still need to articulate a critique of what happened that can inform the next round of political encounter. How should that position be conceptualized?
When equality and difference are paired dichotomously, they structure an impossible choice. If one opts for equality, one is forced to accept the notion that difference is antithetical to it. If one opts for difference, one admits that equality is unattainable. That, in a sense, is the dilemma apparent in Ruth Milkman’s statement cited at the outset of this essay. Feminists cannot give up “difference”; it has been our most creative analytic tool. We cannot give up equality, at least as long as we want to speak to the principles and values of a democratic political system. But it makes no sense for the feminist movement to let its arguments be forced into preexisting categories, its political disputes be characterized by a dichotomy we did not invent. How then do we recognize and use notions of sexual difference and yet make arguments for equality? The only response is a double one: the unmasking of the power relationship constructed by posing equality as the antithesis of difference, and the refusal of its consequent dichotomous construction of political choices.
Equality-versus-difference cannot structure choices for feminist politics; the oppositional pairing misrepresents the relationship of both terms. Equality, in the political theory of rights that lies behind the claims of excluded groups for justice, means the ignoring of differences between individuals for a particular purpose or in a particular context. Michael Walzer puts it this way:
The root meaning of equality is negative; egalitarianism in its origins is an abolitionist politics. It aims at eliminating not all differences, but a particular set of differences, and a different set in different times and places.15
This presumes a social agreement to consider obviously different people as equivalent (not identical) for a stated purpose. In this usage, the opposite of equality is inequality or inequivalence, the noncommensurability of individuals or groups in certain circumstances, for certain purposes. Thus, for purposes of democratic citizenship the measure of equivalence has been, at different times, independence, or ownership of property, or race, or sex. The political notion of equality thus includes, indeed depends on, an acknowledgment of the existence of difference. Demands for equality have rested on implicit and usually unrecognized arguments from difference; if individuals or groups were identical or the same there would be no need to ask for equality. Equality, therefore, might well be defined as deliberate indifference to specified differences.
The antithesis of difference in most usages is sameness or identity. But even here the contrast and the context must be specified. There is nothing self-evident or transcendent about difference, even if the fact of difference—sexual difference, for example—seems apparent to the naked eye. The questions always ought to be, What qualities or aspects are being compared? What is the nature of the comparison? How is the meaning of difference being constructed? Yet in the Sears testimony and in some debates among feminists, (sexual) difference is assumed to be an immutable fact, its meaning inherent in the categories male and female. The Sears lawyers put it this way: “The reasonableness of the EEOC’s a priori assumptions of male/female sameness with respect to preferences, interest, and qualifications is…the crux of the issue.”16 The point of the EEOC challenge, however, was never sameness but the irrelevance of categorical differences.
The opposition men-versus-women, as Rosenberg employed it, asserted the incomparability of the sexes and, though history and socialization were the explanatory factors, these resonated with categorical distinctions inferred from the facts of bodily difference. When the opposition men-versus-women is invoked, as it was in the Sears case, it refers a specific issue (the small statistical discrepancy between women and men hired for commission sales jobs) back to a general principle (the “fundamental” differences between women and men). The differences within each group that might apply to this particular situation—the fact, for example, that some women might choose “aggressive” or “risk-taking” jobs, or that some women might prefer high- to low-paying positions—were excluded by definition in the antithesis between the two groups. The irony is, of course, that the statistical case required only a small percentage of women’s behaviors to be explained. Yet the historical testimony argued categorically about “women.” It thus became difficult to argue (as the EEOC and Kessler-Harris tried to) that, within the female category, women typically exhibit and participate in all sorts of “male” behaviors, that socialization is a complex process that does not yield uniform choices. To make the argument would have required a direct attack on categorical thinking about gender. For the generalized opposition male-versus-female serves to obscure the differences among women in behavior, character, desire, subjectivity, sexuality, gender identification, and historical experience. Kessler-Harris’s insistence on the specificity (and historically variable aspect) of women’s actions seemed like so many special cases, so many ultimately irrelevant exceptions, because she did not present a fully theorized alternative to Rosenberg’s insistence on the primacy of sexual difference.
The alternative to the binary construction of sexual difference is not sameness, identity, or androgyny. By subsuming women into a general “human” category, we lose the specificity of female diversity and women’s experiences; we are back, in other words, to the days when “Man’s” story was supposed to be everyone’s story, when women were “hidden from history,” when the feminine served as the negative counterpoint, the “Other,” for the construction of positive masculine identity. It is not sameness or identity between women and men that we want to claim, but a more complicated historically variable diversity than is permitted by insisting that male and female are antithetical categories, a diversity that is also differently expressed for different purposes in different contexts. In effect the duality this opposition creates draws one line of difference, invests it with biological explanations, and then treats each side of the opposition as a unitary phenomenon. Everything within each category (male or female) is assumed to be the same; hence differences within either category are suppressed. In addition, the relationship between the categories is presented as one between positive (male) and negative (female) poles. In contrast, our goal is to see not only differences between the sexes but the way these work to repress differences within gender groups. The sameness constructed on each side of the binary opposition hides the multiple play of differences and maintains their irrelevance and invisibility.
Placing equality and difference in antithetical relationship has, then, a double effect. It denies the way in which difference has long figured in political notions of equality and it suggests that sameness is the only ground on which equality can be claimed. It thus puts feminists in an impossible position, for as long as we argue within the terms of discourse set up by this opposition we grant the current conservative premise that since women cannot be identical to men in all respects, they cannot expect to be equal to them. The only alternative, it seems to me, is to refuse to oppose equality to difference and insist continually on differences—differences as the condition of individual and collective identities, differences as the constant challenge to the fixing of those identities, history as the repeated illustration of the play of differences, differences as the very meaning of equality itself.
Alice Kessler-Harris’s experience in the Sears case shows, however, that the assertion of differences in the face of gender categories is not a sufficient strategy. What is required in addition is an analysis of fixed gender categories as normative statements that organize cultural understandings of sexual difference. This means that we must open to scrutiny the terms “men” and “women” as they are used to define one another in particular contexts—workplaces, for example. The history of women’s work needs to be retold from this perspective as part of the story of the creation of a gendered work force. In the nineteenth century, for example, certain concepts of male skill rested on a contrast with female labor (by definition unskilled). The organization and reorganization of work processes was accomplished by reference to the gender attributes of workers, rather than to issues of training, education, or social class. And wage differentials between the sexes were attributed to fundamentally different family roles that preceded (rather than followed from) employment arrangements. In all these processes the meaning of “worker” was established through a contrast between the presumably natural qualities of men and women. If we write the history of women’s work by gathering data that describe the activities, needs, interests, and culture of “women workers,” we leave in place the naturalized contrast and reify a fixed categorical difference between women and men. We start the story, in order words, too late, by uncritically accepting a gendered category (the “woman worker”) that itself needs investigation because its meaning is relative to its history.
If in our histories we relativize the categories man and woman, of course, it means we must also recognize the contingent and specific nature of our political claims. Political strategies then will rest on analyses of the utility of certain arguments in certain discursive contexts, without, however, invoking absolute qualities for women or men. There are moments when it makes sense for mothers to demand consideration for their social role, and contexts within which motherhood is irrelevant to women’s behavior; but to maintain that womanhood is motherhood is to obscure the differences that make choice possible. There are moments when it makes sense to demand a re-evaluation of the status of what has been socially constructed as women’s work (“comparable worth” strategies are the current example), and contexts within which it makes much more sense to prepare women for entry into “nontraditional” jobs; but to maintain that femininity predisposes women to certain (nurturing) jobs or (collaborative) styles of work is to naturalize complex economic and social processes and, once again, to obscure the differences that have characterized women’s occupational histories. An insistence on differences undercuts the tendency to absolutist and, in the case of sexual difference, essentialist categories. It does not deny the existence of gender difference, but it does suggest that its meanings are always relative to particular constructions in specified contexts. In contrast, absolutist categorizations of difference end up always enforcing normative rules.
It is surely not easy to formulate a “deconstructive” political strategy in the face of powerful tendencies that construct the world in binary terms. Yet there seems to me no other choice. Perhaps as we learn to think this way solutions will become more readily apparent. Perhaps the theoretical and historical work we do can prepare the ground. Certainly we can take heart from the history of feminism, which is full of illustrations of refusals of simple dichotomies and attempts instead to demonstrate that equality requires the recognition and inclusion of differences.17 Indeed, one way historians could contribute to a genuine rethinking of these concepts is to stop writing the history of feminism as a story of oscillations between demands for equality and affirmations of difference. This approach inadvertently strengthens the hold of the binary construction, establishing it as somewhat inevitable by giving it a long history.18 When looked at closely, in fact, the arguments of feminists do not usually fall into these neat compartments; they are instead attempts to reconcile theories of equal rights with cultural concepts of sexual difference, to question the validity of normative constructions of gender in the light of the existence of behaviors and qualities that contradict the rules, to point up rather than resolve conditions of contradiction, to articulate a political identity for women without conforming to existing stereotypes about them.
In histories of feminism and in feminist political strategies there needs to be at once attention to the operations of difference and an insistence on differences, but not a simple substitution of multiple for binary difference, for it is not a happy pluralism we ought to invoke. The resolution of the “difference dilemma” comes neither from ignoring nor embracing difference as it is normatively constituted. Instead it seems to me that the critical feminist position must always involve two moves: the first, systematic criticism of the operations of categorical difference, exposure of the kinds of exclusions and inclusions—the hierarchies—it constructs, and a refusal of their ultimate “truth.” A refusal, however, not in the name of an equality that implies sameness or identity but rather (and this is the second move) of an equality that rests on differences—differences that confound, disrupt, and render ambiguous the meaning of any fixed binary opposition. To do anything else is to buy into the political argument that sameness is a requirement for equality, an untenable position for feminists (and historians) who know that power is constructed on, and so must be challenged from, the ground of difference.