Forthcoming in the APA Newsletter on Feminism and Philosophy (Spring 2000)
©February 28, 2000 Sally Haslanger

Sally Haslanger
MIT Department of Linguistics and Philosophy
shaslang@mit.edu

I. INTRODUCTION

Unlike feminist ethics, or feminist political philosophy, or even feminist epistemology and philosophy of science, feminist metaphysics cannot be said (yet!) to have standing as a full-fledged sub-discipline of either philosophy or feminist theory. Although one can find both undergraduate and graduate courses devoted to the other sub-fields just mentioned, a course in feminist metaphysics is a rare find; and there are few professional philosophers who would consider listing in their areas of specialization both feminist theory and metaphysics. There are many reasons for this, some having to do with academic politics, e.g., women have not broken into the ranks of metaphysicians in anything like the numbers that can now be found in ethics or political philosophy, and some having to do with tensions between the methods and topics of standard feminist projects and standard metaphysical projects, e.g., feminism is typically taken to be a normative enterprise whereas metaphysics is not.

However, it would be a mistake to conclude from the apparent absence of feminist metaphysics in the structure of the discipline that there is no feminist engagement with metaphysical issues. This is partly because feminist metaphysics often gets done in the context of feminist ethics or feminist epistemology, since these fields take up questions concerning the metaphysics of the self, of autonomy, and of value. However, even beyond these (relatively) disciplinary discussions, there is feminist metaphysics being done under the general rubric of feminist theory, i.e., in the rather open-ended interdisciplinary discussion that serves as the theoretical wing of the effort to end sexist oppression. In this interdisciplinary discussion the topics are not organized in philosophically canonical ways, and the metaphysical issues are often intertwined with other issues, making it hard to see what metaphysical implications, if any, there might be.

In both the (relatively) disciplinary and interdisciplinary work, one can find a kind of engaged metaphysical inquiry that potentially offers more than the luxurious pleasure of thinking one's way through an abstract metaphysical puzzle. Don't get me wrong. I think this luxurious sort of intellectual pleasure is a truly wonderful thing, and wish that everyone so-inclined could have the opportunity to enjoy it; but there is another, and I daresay deeper, intellectual satisfaction that can be found in thinking through puzzles that stand as barriers to personal or social transformation. Some of these puzzles rely on metaphysical presuppositions, and it is one task of feminist metaphysics to identify and question them. Of course, questioning metaphysical presuppositions and offering conceptual alternatives is a tiny part of what feminists need to do to bring about a more just world; but it is, nonetheless, work that needs to be done.

II. QUESTIONS

What are the guiding questions that motivate and organize work in feminist metaphysics? Here are a few: How have metaphysical claims about what there is (and what there is not) supported sexism? Are there metaphysical assumptions or patterns of inference that feminists should challenge? What assumptions or patterns of inference should feminists endorse? Replies to these questions have offered critiques and reconstructions of concepts for thinking about, e.g., the self, sex, sexuality, mind and body, nature, essence, and identity. Feminists have also questioned whether metaphysics is a legitimate form of inquiry at all, raising epistemological questions about, e.g., foundationalist assumptions implicit in metaphysical inquiry (Haslanger, 2000). Due to limitations of space, I will focus the former set of issues, mentioning methodological and epistemological questions only in passing.

To begin an overview of feminist metaphysics in this century, it is helpful to return to Simone de Beauvoir's classic work The Second Sex (de Beauvoir 1989). Two of her most famous claims appear to have profound metaphysical implications: "One is not born, but rather becomes, a woman," (de Beauvoir 1989, 267) and "He is the Subject, he is the Absolute--she is the Other" (de Beauvoir 1989, xxviii). There is disagreement about how to interpret both claims, yet to many the former serves as the slogan for the view that gender is socially constructed, and the latter identifies the content of feminine construction as all that is opposed to the masculine, the masculine also being what counts as the subject or self. Three interconnected themes prominent in feminist metaphysics emerge here: (i) the social construction of gender (and other categories), (ii) the relational nature of the self (and other categories), (iii) the dangers of dualistic thinking.

III. SOCIAL CONSTRUCTION

In claiming that one is not born a woman, Beauvoir was not suggesting that one is never born with female body parts; rather, her concern was that possession of female (or male) body parts, in and of itself, does not imply how one could or should be socially situated. In spite of this, societies, for the most part, reserve for females certain social roles, norms, and activities that disadvantage them in relation to males, casting the differences as necessary because natural (de Beauvoir 1989, Ch.1). If it is recognized, as Beauvoir urged, that women's and men's roles are a product of social rather than natural forces, this opens up the possibility that the roles could be and so should be made more equitable through social change.

This theme--that social hierarchies are sustained through myths of their natural basis--has prompted a tremendous amount of work on the construction of gender in particular (MacKinnon 1989), but also on the construction of other "naturalized" social categories such as race (Williams 1991; Appiah 1996) and in a somewhat different way, sexuality (Butler 1990; Butler 1993). Research in history, anthropology, literature, and sociology has chronicled the various mechanisms by which gender (and other such categories) is enforced, and research in psychology and biology has further loosened the ties between body types and social roles. Having witnessed the power of naturalizing "myths," feminists tend to be wary of any suggestion that a category is "natural" or that what's "natural" should dictate how we organize ourselves socially. However, there are several different de-naturalizing projects that are often mistaken for each other that engage different sorts of metaphysical issues.

Consider a claim that the category Ø is not a natural category, but is "socially constructed." There are several different versions of this claim; let me distinguish three here. The first version claims that the concept Ø being employed, e.g., the concept of the self, or the body, or of nature, is meaningful for us because it serves a certain social function; the implication is that although the classification may seem "natural," i.e., may seem to be the only or obvious way to capture what's objectively true, there are other (perhaps equally, or even more objective) conceptions of the self or the body that might be more useful and less politically problematic (Haraway 1988; Haraway 1991; Gatens 1996; Anderson 1995; Haslanger 1995). It is often emphasized in this context that the world isn't dictating that we think in the terms we do, e.g., we tend to assume that selves are rational choosers, but does this conception of the self serve certain interests? Are there alternative conceptions that would do better justice to the phenomena we're trying to understand? More generally, do our concepts track "natural" kinds? Must legitimate concepts track "natural" kinds? What is a "natural" kind, and are there any?

The second version of the claim that Ø is not a natural category is not concerned with the objectivity or "naturalness" of our conceptual repertoire, but is concerned with the claim that individuals are members of certain categories naturally, or by nature.[1] E.g., in saying that the category "woman" is a product of social forces, one might mean that individual women are not women "naturally" or "by nature," but come to be so through a social process. (Frye 1983: Scott 1986) The same might be said of selves, or families, or nations, e.g., families aren't "natural" entities--groups of people don't count as families simply by virtue of their natural properties--but are entities constituted by social processes such as marriage, adoption, and culturally variable kinship structures. This move allows us to raise questions about why the social processes giving rise to membership in the category are what they are, whether they are just, and how they could be changed. More generally, what approach should we take to social ontology: to what extent are social kinds and social entities "real"? Can facts about social entities such as families or nations all be reduced to facts about the individual members? Are social facts more malleable than natural facts?

The third version of the claim that Ø is not a natural category but is "socially constructed" echoes themes of the previous versions. However, rather than viewing membership in a category in terms of (pre-existing) individuals satisfying our concepts (or not), the idea is that use of our concept Ø "constructs" individual Ø's or, more generally, our conceptual framework "constructs" all the individuals that we can meaningfully think or speak about. Feminists have suggested, for example, that the world as we know it is "constructed" in this sense from a conceptual framework that is problematically male or masculine (MacKinnon 1989), or problematically heterosexist (Butler 1993). The task, then, is to reconstruct our frameworks to be less sexist and heterosexist, so we can construct a more just world.

There are many different interpretations of this somewhat mysterious idea of "construction." Some have drawn on speech-act theory; some of the issues overlap with contemporary discussions of projectivism about secondary qualities and value; and of course there are parallels with Kant. It is a matter of ongoing discussion what exactly we are to make of this third version of the claim that our categories are socially constructed (Hacking 1986; Haraway 1989; Butler 1993). But there is no doubt that it raises a host of metaphysical and epistemological questions --some new and some familiar--concerning the mind's engagement with the world.

IV. RELATIONS

The previous section outlined ways in which feminists have problematized the idea that particular categories are "natural." Similarly, feminists have problematized the idea that particular categories are intrinsic or non-relational. The critical charge, stated very generally, is that dominant frameworks for representing the world, especially the social world, purport to classify things on the basis of intrinsic properties when in fact the classifications are (or should be) crucially dependent on relational properties.[2]

There are two forms of this critique, and correspondingly, two kinds of response. On the first form, the charge is that dominant frameworks misrepresent their subject matter by ignoring important relational aspects of what they purport to be talking about. For example, feminists have long charged that philosophical conceptions of the self, e.g., the conception of the independent rational self-regulator, are framed in grotesquely atomistic terms, ignoring our inevitable and valuable dependence on each other. In response, feminists have urged us to recognize and revalue the complexity of subjectivity not addressed in models of rational agency, and to incorporate in our understanding of the self facts about the realities of human dependence and interdependence for which women have been primarily responsible. (Meyers 1997; Meyers forthcoming)

The second form of such critique also alleges that the dominant frameworks misrepresent their subject matter by obscuring what's relational. However, the goal is not to capture and revalue the background relations. In the cases in question, the charge is that although the system of classification appears to be sorting individuals on the basis of intrinsic properties, in fact there are invidious relations that are being masked by these appearances. (Flax 1986, 199-202) Just as there are reasons why dominant frameworks construct myths about what's natural to justify subordinating practices, likewise they construct myths about what's intrinsic.

Consider again Beauvoir's claim that "He is the Subject, he is the Absolute--she is the Other." Part of what is at stake in Beauvoir's conception of women as Other is the idea that our conceptions of gender and of the self are implicitly relational, e.g., although it may seem that we can define what it is to be a woman without reference to men, in fact we cannot (Wittig 1992; MacKinnon 1989; Haslanger 1993). To be a woman is to stand in a complex set of social (and hierarchical) relations to men (mutatis mutandis for men). And to be a Subject is to stand in a complex set of social relations to some group of Others.

These particular claims of Beauvoir's are, of course, controversial and would need further argument to be made plausible; but the claims are less important than the general idea that relations, especially social relations, are sometimes obscured by our ordinary frameworks for thinking of things. This is of special interest to feminists (and antiracists) for reasons linked to those we have for questioning the representation of a category as "natural." Begin with a background assumption that social life cannot help but accommodate what's "natural." We then can contribute to some category's appearing "natural" by supposing that the basis for membership in the category is intrinsic (thus obscuring the social relations that are the real basis for membership). In this context, pressure to change or abolish the category seems unreasonable.

These critiques raise invite us to ask: how should we re-conceptualize the self and other parts of our social ontology? What is the relation between intrinsicness and naturalness? On what basis can we claim that one framework is "masking" another?

V. DUALISMS

In the previous section I outlined a project of "uncovering" relations in apparently non-relational frameworks. In the sort of cases I had in mind what's "uncovered" are concrete social relations, e.g., relations of sexual subordination. However, Beauvoir's claims about Subject and Other point to additional insights not yet explored.

In saying that "He is the Subject, he is the Absolute--She is the Other," part of Beauvoir's point is that although it may appear that our distinction between subjects and non-subjects is a purely descriptive demarcation of a specific category of substances (selves), in fact, the distinction in use is normative and non-substantive. Begin with the issue of substances: one of the traditional characteristics of substances is that substances do not have opposites, i.e., there is no opposite of horse (non-horse does not count as an opposite). This is in contrast to many qualities: pale/tan, long/short, inside/outside. One way of explicating Beauvoir's suggestion is that once we look at the conditions for subjecthood, we see that there is an opposite to being a subject: subjects are, for example, free and autonomous persons, and the opposite of a free and autonomous person is someone unfree, in her terms, someone condemned to immanence. Moreover, it is not only the case that being a subject has an opposite, but that the opposition in question carries normative weight--so much so that the devalued side of the opposition (the Other) is denied reality in its own terms: what it is to be Other just is to be opposite to the Subject.

Again the feminist project is one of unmasking certain ordinary assumptions about our classifications of things: the category of Subject is not--ontologically speaking--what it may seem. More specifically, categories that appear to be descriptive may in fact be functioning normatively; and categories that appear to be substantive, may in fact be functioning as one end of a qualitative spectrum. Although Beauvoir's example has us focus on the notion of a subject or self; feminists have explored the same form of argument with other notions, notably, sex, gender, and race.

There are two significant consequences of this sort of analysis. First, with substances, it is standardly supposed that you are a member of the kind or not and there is no middle ground: you are a horse or you aren't. (Because there is no opposite or contrary to horse, the only negative option a contradictory.) Again we can contrast this with other opposites: there is a middle ground between pale/tan, long/short, inside/outside; and some things avoid the opposition altogether, e.g., my coffee mug is neither pale nor tan. Casting a category as substantial, then, limits the available categories for classification. For example, suppose we understand 'male' substantively. If males are a substance kind, then everything is either male or not-male, with no middle-ground. But if 'not-male' actually functions as a way of picking out females, then it would seem, in practice, that everything must be either male or female, and there can be no space for genuine categories of people who are intersexed, or other-sexed, or for refusing to sex people at all. One strategy, then, for undermining the idea that a category is substantival is to highlight the multiplicity of individuals and categories "in between" the primary category and its implicit opposite. Category proliferation--the generation of a continuum or genuinely "mixed" categories--can loosen the grip of substantival assumptions. (Butler 1987; Lugones 1994; Haraway 1988; Zack 1995)

Second, in the case of substance kinds, those things that are not in the kind don't themselves form a kind of their own. They are what's "left over." The class of all things that are not-horses includes computers, stars, dust, basketballs, people, etc. So, if we elide 'not man' and 'woman', then women are not read as a kind. As Marilyn Frye puts it,

"When woman is defined as not-man, she is cast into the infinite undifferentiated plenum...[this partly explains why] many men can so naturally speak in parallel constructions of their cars and their women, and say things like, "It's my house, my wife, and my money, and the government can't tell me what to do about any of it." It also illuminates the fact that women are so easily associated with disorder, chaos, irrationality, and impurity....There are no categories in not-man; it is a buzzing booming confusion." (Frye 1996, 1000).

Frye's strategy is not to challenge the substantive status of the man-kind by proliferation, but to challenge its hegemony in the space of persons. So she proposes the construction of a woman-kind that is defined in its own terms, not simply by opposition to men. (Also, Schor and Weed 1994) She argues, among other things, that this will require a recognition of real differences not only between men and women, but among women.[3]

This barely scratches the surface of feminist discussion of the dualisms that guide our thinking, both in philosophy and common sense. These include mind/body, reason/emotion, nature/culture, freedom/necessity, agent/patient. It does, I hope, provide some introduction to the feminist issues that arise in thinking about classification, substances, dichotomy, and the potential political import of ontology.

VI. CONCLUSION

There are a couple of over-arching questions that seem to me worth raising now with a (brief) overview in place. The outline I've offered suggests that in a number of different ways feminist are keen to "unmask" or "uncover" or "demythologize" certain aspects of our ordinary (and philosophical) thinking. Where ordinarily we take ourselves to be dealing with an ontology of substances, natural things, intrinsic properties, we're in fact dealing with an ontology of social things, relations, and non-substantive (and often normative) kinds. But what is the relationship between these sorts of "unmasking" projects and projects that count as part of "mainstream" philosophy, or more specifically, "mainstream" metaphysics? So much analytic metaphysics consists in "reconstructions" of our ordinary concepts; a significant amount of it is unabashedly "revisionary." So along those lines, feminist metaphysics would seem to fit right in. But of course it doesn't fit right in. Clearly feminist metaphysics differs from the mainstream in its subject matter and background assumptions. But is it more than that? Is feminist metaphysics just "mainstream" metaphysics directed at different issues, or is there a deep difference? And if there is a deep difference, what exactly is it?[4]

REFERENCES

Anderson, Elizabeth. 1995. "Knowledge, Human Interests, and Objectivity in Feminist Epistemology." Philosophical Topics 23, no. 2 (Fall): 27-58.

Appiah, K. A. 1996. "Race, Culture, Identity: Misunderstood Connections." In K. A. Appiah and A. Gutmann, ed., Color Conscious: The Political Morality of Race. Princeton: Princeton University Press.

Butler, Judith. 1987. "Variations on Sex and Gender." In Feminism as Critique, ed., S. Benhabib and D. Cornell. Minneapolis: Univ. of Minnesota Press, pp. 128-142.

________. 1990. Gender Trouble: Feminism and the Subversion of Identity. New York: Routledge.

________. 1993. Bodies That Matter: On the Discursive Limits of "Sex". New York: Routledge.

de Beauvoir, Simone. 1989. The Second Sex, trans. H. M. Parshley. New York: Vintage Books.

Flax, Jane. 1986. "Gender as a Problem: In and For Feminist Theory." American Studies/Amerika Studien 31:2 (June): 193-213.

Fraser, Nancy and Linda Nicholson. 1990. "Social Criticism without Philosophy: An Encounter Between Feminism and Post-modernism." In Feminism/Postmodernism, ed., L. Nicholson. New York: Routledge, pp. 19-38.

Frye, Marilyn. 1983. The Politics of Reality: Essays in Feminist Theory. New York: The Crossing Press.

________. 1996. "The Necessity of Differences: Constructing a Positive Category of Women." Signs 21, no. 4 (Summer): 991-1010.

Gatens, Moira. 1996. Imaginary Bodies. New York: Routledge, 1996.

Hacking, Ian. 1986. "Making Up People." In Reconstructing Individualism, ed., M. Heller et al. Stanford: Stanford Univ. Press.

Haraway, Donna. 1988. "Situated Knowledges: The Science Question in Feminism and the Privilege of Partial Perspective." Feminist Studies 14, no. 3 (Fall): 575-600.

________. 1989. "Introduction." In her Primate Visions. New York: Routledge, 1989, pp. 1-15.

________. 1991. "A Cyborg Manifesto: Science, Technology and Socialist Feminism in the Late Twentieth Century." In her Simians, Cyborgs and Women: The Reinvention of Nature. New York: Routledge, pp. 149-181.

Haslanger, Sally. 1993. "On Being Objective and Being Objectified." In A Mind of One's Own, ed., L. Antony and C. Witt. Boulder, CO: Westview, pp. 85-125.

________. 1995.