Human Studies 16: 143-162, 1993. © 1993 KluwerAcademic Publishers. Printed in the Netherlands.

Moral voices, moral selves: About getting it right in moral theory SUSAN HEKMAN Department of Political Science, University of Texas at Arlington, Arlington, TX 76019

1. Introduction

In 1982 Carol Gilligan published an empirical analysis of the decisionmaking processes of a sample of girls and young women confronted with real-life moral dilemmas. Gilligan claimed that these girls and women articulated their moral situations in a "different" moral voice, a voice distinct from the moral voice defined in previous studies that involved primarily male subjects. Nearly a decade later the moral and epistemological ramifications of Gilligan's study are still being explored. A wide array of claims have been made on the basis of Gilligan's analysis. Moral philosophers, feminist theorists and moral psychologists have asserted, variously, that Gilligan's "different voice" represents a significant departure in moral theory, that it is empirically unjustified, that it establishes the superiority of a distinctively feminine moral voice, and that it provides a complement to the unitary voice of moral theories that have dominated moral philosophy in the west. Some of these claims are made explicitly in In a Different Voice or in Gilligan's subsequent work, while others are implicit in it. Some contradict Gilligan's analysis or try to supercede it. The different claims that allegedly arise from her work, furthermore, are not consistent. In the following I will sort through several of these claims and examine them from an epistemological perspective. My larger goal, however, it to do more than make some sense of this extensive literature, as difficult as that task is. My thesis is that most of the claims made by Gilligan and her commentators are not sufficiently radical. The claims fall under two headings: f'n-st, the contention that the unitary moral theory of the Western tradition should be replaced by a dualistic theory, or, second, that the justice perspective in moral theory should be replaced with the more superior care perspective. The problem with these claims is that neither challenges the epistemology that charac-

144 terizes Westem moral theory, an epistemology that attempts to def'me one true moral theory. What I would like to suggest, in contrast to these claims, is that if we take seriously Gilligan' s understanding of the creation of moral voices, voices that arise from relational, discursively constituted subjects, then much more radical implications flow from her work. In her analysis of how moral voices are constituted Gilligan wants to reduce the moral voices to two, masculine and feminine, and, furthermore, to fit her theory into the epistemological space of Western moral theories. In the following I will argue that neither of these moves is epistemologically viable. If it is the case that subjects are constituted through discursive relationships and, further, that central to the constitution of the subject is the creation of a moral voice, then the moral epistemology that flows from this entails a radical alteration of existing moral theories. It follows that moral voices will be as diverse as the subjects that produce them, that many moral voices must be defined, not only two. It also follows that it is impossible to define one true moral theory based on the def'mition of one or even two true moral voices. In short, I will argue that the relational self that Gilligan describes effectively deconstructs the metanarrative of moral theory that has dominated Western thought since its inception. Gilligan's theory cannot, as she and many of her commentators suggest, be accomodated by a reworkhag of the epistemology of existing moral theory, but, rather, demands a new epistemological space for the moral.

2. Care and justice In In a Different Voice Gilligan presents a thesis that is, on the surface,

quite simple. She argues that the women and girls that she studied articulate their moral dilemmas in a voice that differs from that of the men and boys studied by her colleague Lawrence Kohlberg and, before him, Piaget. She argues that the girls and women articulate what she defmes as a "care" perspective, while the boys and men that Kohlberg studied articulate a "justice" perspective. I The relationship that she posits between the masculine justice orientation and the feminine care orientation, however, is far from simple. In the course of her book she suggests at least three possible relationships between the two perspectives. Some of these are overlapping, s o m e contradictory. All have implications both for moral theory and for the social scientific methodology that Gilligan employs. The fh-st relationship between justice and care that GiUigan posits is that the two orientations are separate but equal. Kohlberg's stage sequence of moral development relegated the care perspective that Gilligan identifies in women's moral voices to a lower level of moral development. Gilligan's

145 principle thesis is that the care perspective represents a separate but equal pattem of moral development. Her researches constitute an attempt to rescue the care perspective from Kohlberg's inferior categorization by positing a different, yet equally valid developmental pattem for women. In In a Different Voice Gilligan (1982:26) states this aim very succinctly: Adding a new line of interpretation, based on the imagery of the girls' thought, makes it possible not only to see development where previously development was not discerned, but also to consider differences in the understanding of relationships without scaling these differences from better to worse. Although Gilligan (1982:14) concedes that it is difficult to say "different" without saying better or worse, she does not deviate from the claim that the voice of care is both distinct from and equal to that of justice. In her subsequent empirical work Gilligan has expanded the separate but equal perspective in terms of what she calls the "focus phenomenon" (1987a:28). In these studies Gilligan and her colleagues (1989:xviii) have found that the majority of those studied describe moral conflict from either a justice or a care perspective. Gilligan concludes that justice and care are not opposites, but, rather, different ways of organizing moral problems. At one point she utilizes the duck/rabbit figure to make her point. Just as it is impossible for an observer to see the figure as both a duck and a rabbit simultaneously, she (1987a:22) argues that moral problems can be seen as problems either of care or of justice, but not as both at once. What we must do, she claims, is educate both moral voices and abandon the notion of a single fight answer to moral questions. She (1987b:88) hopes that this will lead to ways of speaking about differences that do not lead to "invidious comparisons" between justice and care. It is tempting to move from Gilligan's argument to the claim that different kinds of moral problems are appropriate to care reasoning, others to justice reasoning. It is significant that Gilligan herself does not make this move. Several of her commentators, however, have done so. John Hardwig (1984), for example, argues that the justice orientation is appropriate to the public sphere, the care orientation to the private. Virginia Held (1987) argues that Gilligan's position entails "domain relativism," and thus a pluralistic, tolerant moral theory that is richly diverse rather than unitary and hierarchical. 2 The separate but equal thesis, however, either in the version posited by Gilligan or that advanced by Hardwig and Held, creates dangerous pitfalls for feminist theory. The particular separation of moral domains suggested by these approaches reifies the public/private distinction that has defined the inferior role of women in Westem thought. The hope that, at this point in history, woman's "private morality" will be elevated to

146 equal status with man's public domain seems at best naive. Even though Gilligan avoids the association of care with the private realm and justice with the public, her argument that the separate but equal position she advocates will eschew "invidious comparisons" is unrealistic. The difference that she describes reifies the inferior status of women central to western history and epistemology. The "invidious comparisons" she wants to eschew are inseparable from that epistemology. A second possible relationship between justice and care that runs through Gilligan's work is that of complementarity. At the end of In a Different Voice Gilligan (1982:174) states that in a mature person the perspectives of justice and care converge, resulting in a "dialogue between fairness and care." In a more recent work she describes this dialogue in terms of a "double fugue" in which the themes of justice and care are woven into a harmonious whole (Gilligan, Rogers and Brown 1990:321). Several of the essays in her edited book, Making Connections, emphasize that the girls that were studied integrate considerations of justice and care in their definitions of moral problems (Bernstein and Gilligan, 1990; Stem, 1990). What this entails is that justice and care are not separate, but intertwined. In making this argument Gilligan specifically relates the two moral perspectives to conceptions of self: the care perspective entails a relational self, the justice perspective a separate one. Gilligan argues that these two conceptions of self are likewise intertwined. Unlike many of her commentators Gilligan does not separate the relational and autonomous selves, defining them as different modes of being. Rather, she argues that they are two aspects of human existence: one can only experience self in the context of relationship with others and, conversely, one can only experience relationship by differentiating other from self (Gilligan, Rogers and Brown, 1990:328). Gilligan's complementarity theme raises more problems than it solves. First of all it contradicts the separate but equal thesis she simultaneously advances: complementarity entails convergence, not separation. More importantly, it once more associates women with the caring, relational aspects of moral life, men with the autonomous and separate, yet denies the hierarchy that this move necessarily entails within the confines of Western epistemology. Gilligan's claim that the two moral perspectives are complementary is, of course, far from unique. The understanding of woman as complement to man goes all the way back to Eve's creation as Adam's helpmate. Throughout Western history woman's moral perspective has been understood as an inherently inferior supplement to man's moral principles. Gilligan is here taking a familiar understanding of woman's voice and trying to raise it from inferiority to equality. Once more her effort is noble but futile; complements are always inferior to that which they complement. 3

147 An important aspect of Gilligan's complementarity thesis is a claim that raises even more difficult epistemological questions. Gilligan asserts that the justice perspective is incomplete without the addition of the care perspective and, thus, that the addition of this perspective brings us closer to an understanding of the "truth" of human moral development. This effort to fred the "truth" of moral development by "completing" the justice perspective with that of care is a theme that runs strongly through Gilligan's discussion in In a Different Voice. At the outset she claims that her aim is to "yield a more encompassing view of the lives of both sexes" (1982:4) by looking for the "truth of woman's experience" (1982:62). But it is important to note that Gilligan's search for "truth" is not limited to women's experience alone. She (1982:98) wants to claim that the care perspective describes a truth characteristic of the human life cycle: To admit the truth of the women' s perspective to the conception of moral development is to recognize for both sexes the importance throughout life of the connection between self and other, the universality of the need for compassion and care. The concept of the separate self and of moral principles uncompromised by the constraints of reality is an adolescent ideal, the elaborately wrought philosophy of a Stephen Daedalus whose flight we know to be in jeopardy. Furthermore, she (1982:173) wants to claim that her work provides "a new perspective on relationships that changes the basic constructs of interpretation." Thus the "marriage" she (1982:174) proposes between men's and women's developmental patterns "could lead to a changed understanding of human development and a more generative view of human life." The themes of universality and truth that are introduced in In a Different Voice have become stronger in Gilligan's subsequent work. She now claims that there are two universal dimensions of early childhood relationships, inequality and attachment, and that these experiences produce the dual perspectives of justice and care. Because girls are more attuned to attachment the care perspective is more developed in women; the opposite is true of men. Her point, however, is that both perspectives are rooted in universal truths of human existence (Gilligan and Wiggins, 1988:114-16). Gilligan is not shy about the revolutionary implications of her theory. She claims that it refutes Piaget's claim that egocentrism is inherent, identifying it instead as the result of boys' childhood experience (Gilligan and Wiggins, 1988:136). Her theory challenges the definition of "human" as separate, autonomous and egocentric that has dominated not only moral psychology, but moral philosophy as well (1987a:31). Finally, she (1987b:93, 77) claims that her work replaces the unitary theory of moral development with a dualistic

148 theory resulting in a new, and better, "map of development." In the next section I will argue that Gilligan's pursuit of the universal truth of human moral development in these passages contradicts other aspects of her theory. At this point, however, I would like to make the narrower point that the position that Gilligan is adopting in these passages is what Sandra Harding has labelled "feminist empiricism." Feminist empiricists do not challenge the basic structure of the masculine tradition of science, but, rather, attempt to correct or complete it (Harding 1987). They argue that masculine scientific approaches have been incomplete because they have ignored the "truth" of feminine experience. Their solution is to add the dimension of feminine experience to the masculine account and, thus, with a more complete analysis, achieve the male scientist's goal: truth and objectivity. This accurately describes Gilligan's goal as she states it in these passages. She does not challenge Kohlberg's methodology outfight, but merely argues that it is incomplete. More importantly, she does not challenge the goal of Kohlberg's work, uncovering the "truth" of human experience. On the contrary, she adopts truth and objectivity as her goals as well and faults Kohlberg for failing to attain them. In all of her work, Gilligan only hints at a third possible relationship between justice and care: the superiority of care to justice. In 1985 at a conference on "Women and Moral Theory" GiUigan (1987a:32) stated: "The promise of joining women and moral theory lies in the fact that human survival, in the late twentieth century, may depend less on formal agreement than on human connection." As far as I know this is the only time that Gilligan has privileged the care perspective over that of justice. In all of her other work she has argued that the implication of her analyses is to institute a dual conception of moral theory in which justice and care are equally valid. Furthermore, she has argued that although the care perspective is dominant in women, it is not exclusively defined by gender. But although Gilligan is, with this one exception, scrupulous in her assertion of the equality of justice and care, the argument that care is superior to justice is worth noting because it has been one of the major consequences of Gilligan's work. Given Gilligan's hesitancy on this issue, it is ironic that one of the most striking results of her researches has been the development of the argument, first, that the care perspective is essentially feminine and, second, that it is superior to the masculine-defined justice perspective. Sarah Ruddick (1989) and Nel Noddings (1984) in particular have argued that the caring, connected moral perspective of women is superior to the abstract, formalistic morality of men. Frequently citing Gilligan's work in support of their arguments, Ruddick, Noddings and other feminist theorists have asserted that the justice perspective of men has been instrumental in creating the domination,

149 racial discrimination and rape of nature that characterizes the modern world. The advantages and disadvantages of this approach to moral theory have been extensively discussed in contemporary feminist theory. I will only make two brief points here that are relevant to my argument. First, the discourse of care advanced by these theorists does nothing to displace the monistic, hierarchical and universalistic epistemology of the masculinist theories they attack. Rather, these theorists simply reverse the hierarchy implicit in the masculinist theories by arguing that the morality that they privilege is, in actuality, inferior. Second, the care perspective as it is presented by these theorists is inconsistent with the over-all tenor of Gilligan's approach. Unlike Ruddick and Noddings, Gilligan seeks to advance a dualistic rather than a monistic theory. Her goal is to develop a theory that attempts to probe the truth of the human condition, a truth that encompasses both justice and care.

3. Truth and objectivity The different perspectives on the relationship between care and justice that Gilligan offers raise a number of epistemological problems. In this section I will focus on one of those problems, a problem that has been the focus of much of the criticism of In a Different Voice. A major component of Gilligan's argument is her assertion that her approach captures the "truth" of human development more adequately than her rivals. She claims that her studies are more "objective" because they encompass the truth of both women's and men's moral experiences. Many of Gilligan's critics, both moral psychologists and moral theorists, understand Gilligan's claim to be essentially an empirical one. Hence, subsequent to the publication of In a Different Voice, a heated debate has raged over the empirical accuracy of Gilligan's claim, that is, whether the existence of a different voice can be scientifically documented.4 Kohlberg, the principal target of her attack, treats Gilligan's challenge as essentially an empirical claim, arguing that, as an empirical psychologist, he (1982:528) is open to "new data and new findings" despite his misgivings about Gilligan's approach. Several of Gilligan's critics fault her work on the grounds of its failure to conform to standard social scientific methodology. Some of these criticize the "bias" in her account, resulting in, as one critic puts it, "social science at sea without an anchor" (Nails 1983:664). Others cite the paucity of evidence that Gilligan offers to establish her position (Eisenberg and Lennon, 1983; Walker 1984). In some of her rejoinders to these critics, furthermore, Gilligan meets these criticisms by defending the empirical support for her theory. For example, in a recent work she defends her central claim, the

150 existence of sex differences in empathy, citing the larger male prison population as evidence (Gilligan and Wiggins, 1988:113). In 1986 Signs published a review symposium on In a Different Voice. This review is instructive for a number of reasons. First, it reveals that even within the bastion of feminist theory and in a journal that has been on the forefront of the feminist critique of science, feminist empiricism was, at least in this instance, the dominant discourse. In her introduction to the review the editor asks whether the dialogue on In a Different Voice constitutes a break from "the male-dominated tradition of confrontational debate" (Kerber et al., 1986:304). The answer to this question must be a resounding "no." Had the editor asked whether the review represents a departure from the androcentric standards of science, the answer would also be "no." Most of the authors of the review articles argue that the evidence does not support Gilligan's claim of a different voice and that "claims about these differences should be subjected to the empirical tests that are the basis of social science" (Greeno and Macoby, 1986:316). The second notable aspect of this review is Gilligan's reply. In a forceful argument that effectively destroys her critics' positions, Gilligan asserts that her claim in In a Different Voice was not statistical or empirical but interpretive. She asserts that her point against Kohlberg was not that men and women differ on his scale but, rather, that his scale only fits men's experiences. Even more significantly she (1986:328) argues that "There is no data independent of theory, no observations not made from a perspective. Data alone do not tell us anything; they do not speak, but are interpreted by people." Although none of her critics noted it, this same position is clearly stated at the beginning of In a Different Voice: (1982:6) "Then the presumed neutrality of science, like that of language itself, gives way to the recognition that the categories of knowledge are human constructions." Anyone familiar with recent discussions in the philosophy o f the natural and social sciences will recognize the origin of Gilligan's argument here. She is utilizing current critiques of social scientific methodology to argue against the presumption of a theory-free observation language for social inquiry. On the basis of these critiques she argues that Kohlberg's theory is "biased" in that he begins with the assumption of what constitutes moral behavior and then proceeds to find data that fit his theory. She is further arguing that women's moral "underdevelopment" is the result of the bias inherent in Kohlberg's theory, that is, that there is a discrepancy between his theory and the data of women's moral experience (1987b:78). All of these assertions are consistent with the critique of social scientific "objectivity" that has flourished in recent decades, a critique that, significantly, has been central to feminist critiques of science. What does not

151 follow from this critique, however, is Gilligan's claim, both in In a Different Voice and her subsequent work, that her researches lead to the "truth" of human moral development. This claim is inconsistent with the critiqueof-objectivity position that she uses to attack Kohlberg and reveals a significant contradiction in her work. Despite the existence of this contradiction, however, Gilligan's critique of Kohlberg's moral theory has significant implications. It follows from Gilligan's critique of Kohlberg that it is impossible to "objectively assess" moral development patterns because anyone who attempts to do so will bring to that investigation assumptions of what constitutes "morality." It is particularly easy to criticize Kohlberg along these lines because, although he (1982:525) claims that his goal is "the rational reconstruction of the deep structure of a morally educated adult's moral judgments and intuitions," it is quite clear that what he designates as "moral" is a pattern of thought that has been identified as moral in a very long tradition of Western moral theory. Thus Kohlberg' s theory of moral development obviously determines his assessment of the "data" he "discovers." But other conclusions follow from Gilligan's critique that she does not discuss, ff it is the case that all theories of moral development are human constructions informed by preconceptions, then this is true of her theory as well. Gilligan's theory of the moral development of women is as much "biased" by her preconceptions as is that of Kohlberg. It further follows that, despite her claims to the contrary, Gilligan is not "discovering" the "truth" of women's experiences, but, rather, interpreting the data in light of her own preconceptions as to what constitutes the "moral." Instead of condemning Gilligan for this contradiction in her work, however, it is more fruitful to examine exactly what is at issue here. Kohlberg's definition of the moral realm is a quite literal translation of a tradition of moral theory rooted in a discourse that was first articulated by the Greeks. In this discourse moral reasoning based on abstract, formal principles is defined as the "true" and "highest" form of moral thought. In this discourse considerations that involve particularistic, relative and perspectival elements are declared to be inferior or, more precisely, outside the moral realm altogether. What Gilligan has done is to examine these particulartistic, relative elements and, in opposition to the Westem moral tradition in general and Kohlberg in particular, to declare them to be within the sphere of the "moral." She extends her heresy by declaring this realm to be on an equal footing with abstract, formalistic moral reasoning. In making this declaration she is, undoubtedly, making a significant conceptual breakthrough. But it is not simply the case that she is imposing her own personal definiton of the "moral" on the data she collects. Rather, she is documenting a self-understanding that many women possess. In our culture

152 women are expected to exhibit caring and nurturing qualities, while men are expected to exhibit qualities associated with abstract considerations of justice. By defining the nurturing qualities of women as a legitimate moral sphere Gilligan is simultaneously opposing the tradition of Western moral theory and tapping into understandings of the gendered division of moral labor that is a product of that tradition. She is not discovering the "truth" of moral development any more than Kohlberg is. What she. is doing is labelling as "moral" the concerns that women are supposed to and, indeed, do have in this culture. It is for this reason that the "different voice" has resonated so strongly for so many women that have read Gilligan's work. Gilligan is correct in her assertion that statistical "proof" of the different voice is irrelevant. 5 What is relevant is that she is documenting a selfunderstanding that is common to most women in our society.

4. Discourses of morality What I am arguing, then, is that the epistemological import of Gilligan's work is not, as she claims, that she is adding to or correcting the masculine moral theories of theorists such as Kohlberg, or that she is discovering t h e "truth" of human moral development, but, rather, that she is offering a new discourse of morality that redefines what is to count as "moral." In opposition to the unitary moral tradition that Kohlberg is articulating, a tradition in which only abstract, formalistic thought counts as moral, Gilligan claims that she is advancing a dual conception of moral discourse in which both considerations of care and those of justice are defined as equally valid moral domains. In advancing this claim, Gilligan is accutely aware of the strength of the moral tradition against which she is arguing. In her work and that of her associates she cites several instances in which adolescent girls are reluctant to describe the ethical dilemmas that they confront as "moral problems" because they have been told that "morality" is the sphere of abstract rules and obligations, not a realm concerned with the particularity of relationships. One of Gilligan's associates notes that because what is morally imperative for these girls is not considered "moral" by society, the result is a loss of trust in their judgment and moral integrity (Brown 1990:107). Gilligan (1990:26) puts it this way: "The wind of tradition blowing through women is a chill wind because it brings a message of exclusion." What I would like to suggest, however, is that Gilligan's assertion of a dualistic moral theory is not consistent with the nature of her challenge to masculine moral theory. Although she is clear in her claim that the Western tradition of morality excludes the moral concerns traditionally associated

153 with women, she goes on to argue that we can add this "different voice" to that moral tradition as a separate (or compatible) but equal component. There are two compelling sets of arguments that reveal this position to be both epistemologically confused and theoretically undesirable. First, the epistemological space defined by the Western moral tradition does not allow for the addition of an element that is equal to the standard of universal, absolute, formal principles that defines that tradition. This is particularly the case with Gilligan's theory because she proposes the addition of "feminine" elements of particularity and connection that contradict the very basis of the masculinist tradition. What I am arguing here is similar to Foucault's argument with regard to the human sciences in The Order of Things (1971). Foucault asserts that the epistemological space of modernity leaves no room for the human sciences. My assertion is that the same is true of Gilligan's theory relative to the western moral tradition: the episterne of moral theory explicitly prohibits the introduction of a rival element and, most particularly, a rival element with radically different epistemological assumptions. It follows that Gilligan's advocacy of a dualistic, non-hierarchical moral theory is simply not viable. My second argument is that if we assume that Gilligan's theory effectively deconstructs the unitary, abstract principles of masculine moral theory, then we are left with a much more radical conclusion than Gilligan herself advances. Gilligan claims that the "different" moral voice that women articulate is constituted by the relational experience of girls in childhood. Less explicitly she claims that men's more abstract and autonomous moral voice is also a result of childhood relationships. But if moral voices are a result of particular relational experiences, an expression of the subject that is constituted from these experiences, then it must follow that there are many different moral voices, not only two. The forces that constitute subjectivity are not encompassed by gender alone. Race, class, culture, and historical locatedness are also constitutive factors. The moral theory that is implicit in Gilligan's work demands that we must confront the problem of identifying and legitimizing the variety of moral voices that are linked to diversely constituted subjects. What I am suggesting, in other words, is that Gilligan's theory leads us into a radically new epsitemological space, but that space is not, as she envisions, one in which two identifiable and familiar voices interact, but, rather, one in which a plurality of voices are heard. Gilligan and her commentators have much to say about the tradition of moral theory that they oppose and the alternatives that they propose. I will attempt to establish both of the above points through an analysis of these discussions. Gilligan casts her argument in opposition to the work of Kohlberg, but it is evident, not least of all to Gilligan herself, that Kohlberg is far from original, that he is a loyal son of the western moral tradition.

154 That the moral discourse of modemity has been dominated by formal, abstract principles, particularly as they were expressed by Kant, needs no argument here. Kohlberg relies heavily on one of the most prominent twentieth century representatives of the formalistic tradition, John Rawls. In A Theory of Justice Rawls presents a theory of moral development that is divided into three stages, the morality of authority, association and principles. The morality of authority which, Rawls stresses, is temporary, is based on love. The morality of principles, in contrast, is based on abstract "first principles" that rise above "contingencies" (1971:456). Kohlberg's stage sequence is a direct reflection of Rawls' theory. Kohlberg posits three levels of moral development, pre-conventional, conventional and postconventional, with two stages within each level. The culmination of his sequence is Stage 6 which he characterizes as a morality of individual principles or conscience (1984:xxix). Kohlberg makes many claims for these stages. First, he (1981:178) claims that they are scientific and empirical. Second, he claims that they are universal. Although he concedes that moral philosophy is not itself universal, he is adamant in his assertion that the basic moral principles are. Anyone who attains Stages 5 and 6 in any culture, he (1981:98) asserts, will attempt to formulate universal principles. Third, he claims that the stages are evolutionary both culturally and individually: the two highest stages are absent in preliterate cultures, and within particular cultures individuals progress to higher stages of moral development as they mature. His fourth claim is a negative one: he asserts that he does not mean to imply that the individuals who attain the higher stages are more moral. His claim, rather, is that the universal principles of the higher stages are more adequate, that they are more likely to resolve moral problems (1984:331). Unfortunately for the consistency of his theory, this claim seems to fly in the face of his additional assertion that Stage 6 is "what it means to judge morally. If you want to play the moral game, if you want to make decisions which anyone could agree upon in resolving social conflicts, Stage 6 is it" (1981:172). This perspective on Kohlberg's work reveals the liability of his methodological presuppositions. Kohlberg begins with a precise conception of what morality essentially is - Stage 6 - and sets out to gather evidence to support his theory. He finds this evidence for a reason that Gilligan's researches clarified: educated people in this culture have been taught that "morality" consists of a set of abstract, universal principles. In defending the empirical evidence for his thesis Kohlberg is undaunted by his findings that only 5% of adult Americans test out at Stage 6, that these are people with extensive philosophical training, and that t h e majority of adult Americans (and nearly all women) are at the conventional level of morality (1981:100, 146, 192). Many critics of Kohlberg have attacked him for his

155 elitism, sexism and ethnocentrism. These charges are accurate, but few of these critics have gone on to examine the methodological and epistemological presuppositions that lead him to his conclusions. "Moral judgments," Kohlberg (1981:170) declares, "unlike judgments of prudence or aesthetics, tend to be universal, inclusive, consistent and grounded on objective, impersonal or ideal grounds." It follows that other kinds of judgments will be declared not to be moral from Kohlberg's perspective. The girls that Gilligan interviews are hesitant to call their relational problems "moral" because they, too, have been taught that "morality" is the sphere of formalistic principles. The problem here is not that Kohlberg is "biased" and that this leads to his sexism and elitism. Gilligan is just as "biased" when she wants to extend the concept "moral" to the non-principled, relational, particularistic realm that has been associated with women. What Kohlberg has done is simply to defme the "moral game" as it has been articulated in the western tradition and then confirmed his theory by interviewing men and boys who were educated in the rules of that game. Given Kohlberg's perspective, it is not surprising that he is unimpressed by Gilligan' s researches. He dismisses her work in a manner consistent with the moral tradition he espouses. Gilligan, he claims, is dealing with "special" relationships, those of the family. He (1984:338) concedes that his studies did not include the full domain of moral development and "welcomes" Gilligan's attempt to add to that domain. He (1984:232) proposes a conception of the dimensions of morality with the personal at one end, impartial justice at the other. He (1984:348) further asserts that as women enter men's occupations, they, too, will adopt the morality of justice over care. Kohlberg's "refutation" of Gilligan here fits her theory quite neatly into his traditional perspective: women's care relationships deal with private, "special," and hence inferior concerns; as women move out of this realm they, like men, will recognize the superiority of the justice perspective. Some commentators on the Gilligan-Kohlberg controversy have argued that the dilemma that it poses can be resolved by looking at an alternative tradition in moral theory, one that opposes the universalism of the masculinist tradition. The theorists most often cited in the attempt to establish an alternative tradition are Simone Weil and Iris Murdoch. 6 Both Weil and Murdoch reject the unitary, abstract formalism of Kantian ethics and instead propose a contextual morality that is very similar to Gilligan's ethic of care. Weil argues against the tradition of individualism and rights that has defined moral thought in the western tradition. As an alternative she proposes a redefinition of justice as love and a separation of rights and justice. She (1977:325, 334) claims that "Justice consists in seeing that no harm is done to men" and that rights have no connection with love. Well

156 argues that central to the moral experience is the practice of a form of "attention." Iris Murdoch employs this concept of attention in her theory, making it the centerpiece of her account. Morality, Murdoch asserts, is constituted by "attention" to individuals. Against the dominant moral tradition Murdoch argues that morality is not usually a matter of choice that a rational, autonomous individual freely exercises through the application of abstract principles. Rather, moral choice is historically conditioned by the lives we lead and the loving connections we establish with other people (1970:37-8). Many of the themes that Gilligan takes up in her description of the ethic of care are also expressed by Murdoch. Just as Gilligan attacked the presumption of the autonomous self, Murdoch attacks the Cartesian concept of self and argues that it offers an inadequate description of morality. Morality, she (1970:56-66) claims, is not an act of will, but the practice of attention tied to love. She (1970:28,76, 87) criticizes the attempt to make morality a branch of science and logic, arguing that a better model would be that of art, which, unlike science, presents us with truthful images of the human condition. She (1970:71) concludes The Sovereignty of Good by arguing that moral philosophy needs a new "more realistic, less romantic terminology if it is to rescue thought about human destiny from a scientifically minded empiricism which is not equipped to deal with real problems." Strictly speaking, Murdoch and Gilligan diverge on this point. Murdoch advocates replacing the justice tradition with the care perspective while Gilligan argues for a non-hierarchical dualism. A moral theorist more congruent with Gilligan's position is Lawrence Blum. Blum argues that the Kantian ethical tradition must be supplemented by the moral value of altruism. He (1980:5-6) argues that "The good we do our friends cannot be expressed within Kantian categories" because altruistic emotions are not grounded in universal moral principles. In a long and complex argument Blum seeks to establish what would seem to be an obvious point: that friendship is a legitimate moral phenomenon and that the caring love that we bestow on our friends is a moral good. Blum is careful not to deny Kantian ethics; his aim is rather to assert that there are other types of moral goodness. Although praising Murdoch for her emphasis on the particularity of morals, Blum argues that her view is incomplete; it must be supplemented by the Kantian, impartial view. Morality, he claims, has different spheres; from Blum's point of view Kant and Murdoch are simply expressing those different spheres (1986:360). Weil, Murdoch and Blum all lend philosophical sophistication to Gilligan's advocacy of a different voice. Advocates of the philosophies of Weil and Murdoch argue that they supply precisely what Gilligan's theory lacks, the articulation of an alternative moral tradition. The argument for the

157 existence of an alternative moral tradition, however, is difficult to sustain. Both Weil and Murdoch argue against what they themselves recognise is the dominant tradition in moral theory. Their voices do not constitute a recognised alternative, however, but, rather, amount to only faint voices of protest. Both are marginal figures in moral philosophy and both, not coincidentally, are women. Their position is more accurately described by Foucault's term "subjugated knowledges." The very existence of this knowledge establishes the power of the hegemonic discourse against which they argue. Another objection to the existence of an alternative moral tradition is indicated by Blum. He (1988:482) notes that Kohlberg dismisses the care perspective as a junior partner in the justice/care union he proposes. What both Gilligan and Blum are claiming is that the private sphere of personal, particularistic relationships is not, as the justice tradition dictates, an amoral sphere but, rather, a legitimate moral realm equal to the abstract, formalisfic sphere of justice and rights. The problem with this claim, however, is precisely the problem that Blum identified in Kohlberg's redefined theory. The logic of the justice tradition excludes the addition of particularistic, relativistic elements. If these elements are added they must, following the logic of that tradition, be relegated to an inferior status. Care cannot be added as an equal partner with justice because, as Kohlberg puts it so clearly, Stage 6 is simply what morality is. The justice tradition is epistemologically exclusive. It cannot incorporate the addition of a perspective that deconstructs its epistemological foundations. 7 In addition to the advocacy of an alternative moral tradition, contemporary moral theorists have suggested another solution t o the problems posed by Gilligan's discussion of the ethic of care: redefining justice in terms of care. This tactic was suggested by Weil whose separation of justice and rights was predicated on the definition of justice as love. This approach has been more extensively developed in two recent contributions to feminist political theory. In Justice, Gender and the Family Susan Moller Okin (1989:15) argues that the distinction between justice and care is overdrawn and that the best theorizing about justice includes care. What Okin tries to do in her book is nothing short of the transformation of the public/private distinction that restricts justice to the public sphere. Arguing that if families are unjust, we cannot create a just polity, Okin asserts that we must evaluate the justice of families in our society. Although she concedes that we do not now associate justice with intimacy, she argues that we would gain much from doing so (1989:135). Iris Young adopts a similar approach, but emphasizes a different aspect of the problem. Attacking Gilligan's separation of justice and care, Young asserts that feminists who retain the public/private distinction in moral theory fail to challenge impartiality in

158 the public sphere (1990:97). Against this, Young argues that impartiality is an impossible ideal that masks the domination of the hegemonic group. A better ideal, she (1990:112) argues, is public fairness in a context of heterogeneity and partial discourse. 8 The approaches of Okin and Young, however, like that of the other theorists discussed above, fail to assess the radical implications suggested by Gilligan's work. Okin and Young attempt to redefine the relationship between justice and care, but like the other theorists, they attempt to do so within the epistemological space that defines traditional moral theory. None of these approaches offers a radical challenge to that epistemology, an epistemology that defines the space of the "moral" within our tradition. They are attempts at redefinition, not transformation. More importantly, none challenges the notion that there is one "correct" approach to moral questions and that moral theory is unitary, or, in the case of Gilligan and Blum, a dualistic unity. All of these approaches are tyring to "get it right" without challenging the epistemological assumptions that compel them to look for the one right answer. 9 The search for the right answer, a new and "better" moral theory, however, will always be a futile one. First, it will run up against the solid wall of tradition that has already decreed the right answer to the question of morals (abstract principles) and allows for no other correct answers to this question. Second, attempts to find the right answer, however it is defined, will perpetuate the epistemology that grounds the tradition of moral theory these solutions purport to attack, an epistemology of absolute truths. What is required, on the contrary, is an approach that displaces that epistemology and, specifically, abandons the search for a single right answer. Such an approach is entailed by Gilligan' s understanding of the relational origins of moral voices, but she fails to draw out the implications of it in her work. Gilligan argues that two kinds of subjects evolve out of childhood experiences, masculine subjects whose moral voices speak in terms of separation and abstract principles and feminine subjects whose moral voices resonate with considerations of care and connection. There are several radical moves in this formulation of the self and morality. By deconstructing the absolutism of the justice tradition, Gilligan has moved us into a new epistemological space. What her work suggests is that discourses of morality have internal criteria of rightness and development, and that the unitary moral universe that has been presupposed in the Western tradition must be displaced. This entails a profound change in our understanding of the rules of the moral language game; it means that we cannot fit morality back into the epistemological space from which it has escaped. What Gilligan does not explore, however, is the question of whether there are more than two moral voices. 10 If Gilligan is correct in her assertion that

159 moral voices are the product of the experiences through which the subject is constituted, then it would surely follow that there are more than the two subjects she posits. Although Gilligan's discussion of the constitution of masculine and feminine subjects is profound and insightful, she misses the opportunity to assess the different kinds of masculine and feminine subjects that will arise from different childhood experiences, differences caused by the influence of class, race and culture. These forces, along with gender, will exert a profound effect on the constitution of the subject and, hence, on the moral voices that these subjects speak. Gilligan began by examining the "different" moral voice of women in our society. Her researches have spawned an examination of the masculine bias of moral theory and the valuation of an ethic of care defined in gender terms. What I am arguing here is that a very different line of investigation is suggested by Gilligan's researches: an exploration of the constitution of multiple moral voices. Gilligan's work effectively deconstructs the metanarrative of traditional moral theory. What is entailed by the rejection of the metanarrative of truth has been extensively explored in recent discussions of epistemology. What is entailed by the rejection of the metanarrative of morality and the positing of a heterogeneity of moral voices and moral selves, however, has been less extensively explored. Bernard Williams (1985:173) argues that we would be better off without the "peculiar institution" of morality. Iris Murdoch (1970:44) asserts that the abstract reasoning that characterizes traditional moral theory is a special case of moral reasoning, not the paradigm for all moral thinking. But we have to go much beyond this to explore the parameters of this approach to moral questions. Moral notions, like the subjects who hold them, are contextual and relational. They are constituted by the discourses or language games of the societies in which those subjects live. Gilligan claims that there are two moral voices constituted by the different childhood experiences of girls and boys. But this approach is too simple. It ignores the fact that masculinity and femininty are constituted differently in different cultures; racial and class differences also result in a variety of constructions of masculine and feminine subjects. It follows that there are many different moral voices even within a given society; moral standards will vary respective to these moral voices. Critics of this approach to morality will argue that this obviates the possibility of moral judgments. But although this position does eliminate reference to a moral metanarrative, it does not entail arbitrariness. It entails, rather, that our definition of the moral and the patterns of our moral thinking are, like all other aspects of our lives, constituted by discursive formations. The result is not moral anarchy, but, rather, an attention to the particularity of moral notions and the internal dynamics that constitute them. That this is a very different approach to morality is undeni-

160 able. That it is necessary to account for the different moral voices in particular human societies seems unavoidable.

No~s 1. Gilligan is careful to avoid the gendered connection of these two perspectives, denying that the care perspective is exclusively feminine and the justice perspective exclusively masculine. But in her studies the gendered association of the perspectives is rarely challenged. 2. Flanagan and Jackson (1987) make a similar point. 3. See Walker (1984) for a discussion of complementarity and opposition. It is interesting that John Stuart Mill also posits the complementarity of male and female moral voices. His formulation, however, also fails to relieve the inferiority of the female voice. 4. Among others, see Eisenbery and Lennon (1983), Walker (1984), Broughton (1983), Nails (1983), Kohlberg (1982), Kerber et aL (1986), and Baumrind (1986). 5. See Friedman (1987) for a discussion of the irrelevance of statistical proof of the different voice. Deborah Tannen (1990) documents how the empathetic voice of women structures their self-understanding. 6. Annette Baler (1987) also cites Hume's moral theory as an example of this alternative tradition. 7. Two other theorists are frequently mentioned in this context: MacIntyre and Frankena. MacIntyre's theory of virtue is cited as compatible with Gilligan's ethics of care. But MacIntyre's communitarian ethics (1984) is profoundly hostile to women. Frankena (1963) is cited as advocating two moral axes, benevolence and justice. But for Frankena both benevolence and justice are defined as abstract moral principles. 8. See Benson (1990) for a redefinition of free agency along these lines. 9. See Tronto (1987) for another attempt to find the right answer, in this case by more securely grounding the ethic of care. 10. See Blum (1988) for a discussion of more than two moral voices.

References Baler, A. (1987). Hume: The women's moral theorist? In E. Kittay and D. Meyers (Eds.), Women and moral theory, 37-55. Totowa, NJ: Rowman and Littlefield. Baumrind, D. (1986). Sex differences in moral reasoning: Response to Walker's [1984] conclusion that there are none. Child Development 57:511-521. Benson, P. (1990). Feminist second thoughts about free agency. Hypatia 5(3):47-64. Bernstein, E. and Gilligan, C. (1990). Unfairness and not listening: Converging themes in Emma Willard girls' development. In C. Gilligan, N. Lyons and T. Hanmer (Eds.), Making connections, 147-161. Cambridge, MA: Harvard University Press. Blum, L. (1980). Friendship, altruism and morality. Boston: Routledge and Kegan Paul.

161 Blum, L. (1986). Iris Murdoch and the domain of the moral. Philosophical Studies 50:343-367. Blum, L. (1988). Gilligan and Kohlberg: Implications for moral theory. Ethics 98(3):472-491. Broughton, J. (1983). Women's rationality and men's virtues: A critique of gender dualism in Gilligan's theory of moral development. Social Research 50(3):597-642. Brown, L. (1990). When is a moral problem not a moral problem? In C. Gilligan, N. Lyons and T. Hanmer (Eds.), Making connections, 88-109. Cambridge, MA: Harvard University Press. Eisenberg, N. and Lennon, R. (1983). Sex differences in empathy and related capacities; Psychological Bulletin 94(1): 100-131. Flanagan, O. and Jackson, K. (1987). Justice, care and gender: The KohlbergGilligan debate revisited. Ethics 97:622-637. Foucault, M. (1971). The order of things. New York: Random House. Frankena, W. K. (1963). Ethics. Englewood Cliffs, NJ: Prentice-Hall. Friedman, M. (1987). Care and context in moral reasoning. In E. Kittay and D. Meyers (Eds.), Women and moral theory, 190-204. Totowa, NJ: Rowman and Littlefield. Gilligan, C. (1982). In a different voice. Cambridge, MA: Harvard University Press. Gilligan, C. (1986). Reply by Carol Gilligan. Signs 11(2):324-333. Gilligan, C. (1987a). Moral orientation and moral development. In E. Kittay and D. Meyers (Eds.), Women and moral theory, 19-33. Totowa, NJ: Rowman and Littlefield. Gilligan, C. (1987b). Remapping development: The power of divergent data. In C. Farnham (Ed.), The impact of feminist research in the academy, 77-94. Bloomington, IN: Indiana University Press. Gilligan, C. (1989). Adolescent development reconsidered. In C. Gilligan et al. (Eds.), Mapping the moral domain, vii-xxxix. Cambridge, MA: Harvard University Press. Gilligan, C: (1990). Preface. In C. Gilligan, N. Lyons and T. Hanmer (Eds.), Making connections, 6-29. Cambridge, MA: Harvard University Press. Gilligan, C. and Wiggins, G. (1988). The origins of morality in early childhood relationships. In C. Gilligan et al. (Eds.), Mapping the moral domain. Cambridge, MA: Harvard University Press. Gilligan, C., Rogers, A. and Brown, L. (1990). Epilogue. In C. Gilligan and T. Hanmer (Eds.), Making connections, 314-334. Cambridge, MA: Harvard University Press. Greeno, C. and Maccoby, E. (1986). How different is the "different voice"? Signs 11(2):310-316. Harding, S. (1986). The science question in feminism. Ithaca, NY: Cornell University Press. Hardwig, J. (1984). Should women think in terms of rights? Ethics 94(3):441-455. Held, V. (1987). Feminism and moral theory. In E. Kittay and D. Meyers (Eds.), Women and moral theory, 111-128. Totowa, NJ: Rowman and Littlefield. Kerber, L. et al. (1986). On In a different voice: An interdisciplinary forum. Signs 11(2):304-333. Kohlberg, L. (1981). The philosophy of moral development. San Francisco: Harper and Row,

162 Kohlberg, L. (1982). A reply to Owen Flanagan and some comments on the PukaGoodpaster exchange. Ethics 92(3):513-528. Kohlberg, L. (1984). The psychology of moral development: Essays on moral development, vol. 2. San Francisco: Harper and Row. Mclntyre, A. (1984). After virtue. 2nd ed. Notre Dame, IN: University of Notre Dame Press. Murdoch, I. (1970). The sovereignty of good. London: Routledge and Kegan Paul. Nails, D. (1983). Social-scientific sexism: Gilligan's mismeasure of man. Social Research 50(3):643--664. Noddings, N. (1984). Caring. Berkeley, CA: University of California Press. Okin, S. M. (1989). Justice, gender and the family. New York: Basic. Rawls, J. (1971). A theory of justice. Cambridge, MA: Harvard University Press. Ruddick, S. (1989). Maternal thinking. Boston: Beacon Press. Stem, L. (1990). Conceptions of separation and connection in female adolescents. In C. Gilligan, N. Lyons and T. Hanmer (Eds.), Making connections, 73-87. Cambridge, MA: Harvard University Press. Tannen, D. (1990). You just don't understand: Women and men in conversatio. New York: William Morrow. Tronto, J. (1987). Women's morality: Beyond gender differences to a theory of care. Signs 12(4):644-663. Walker, L. (1984). Sex differences in the development of moral reasoning: A critical review. Child Development 55:677-691. Weil, S. (1977). The Simone Weil reader. Ed. G. Fanichas. New York: David McKay. Williams, B. (1985). Ethics and the limits of philosophy. Cambridge, MA: Harvard University Press. Young, I. M. (1990). Justice and the politics of difference. Princeton, NJ: Princeton University Press.