Printer Friendly

From autonomy to agency: feminist perspectives on self-direction.

In this Essay, I will consider the ways that the liberal norm of autonomy has been modified--or in the terms of this symposium, "reconstructed"--by its encounters with contemporary feminist theory. It remains an open question whether the feminist influences I describe will, in fact, "reconstruct" liberal autonomy, or transform it into something else altogether. The normative goals toward which I hope to turn this exploration, however, make the "reconstructive" label attractive. As a feminist legal scholar, I have become interested in the topic of women's agency: its persistence even under circumstances of oppression, the roles it plays in facilitating resistance to oppression, and the forms it takes at the current time or might take under less oppressive circumstances. My goal is to use law to highlight and foster women's agency as part of the larger task of using law to combat women's oppression. The legal tools I employ rely on a range of liberal assumptions, including assumptions about the effects of discrimination or oppression and-the nature and formation of the self or subject. Thus, whether the liberal autonomy I will discuss becomes altered incontrovertibly by its encounters with structuralist or poststructuralist feminism, it may be reconditioned by liberal norms through its insertion into the legal system, making the reconstitutive metaphor appropriate.

Autonomy is both a characteristic of a human subject and a quality that inheres in particular acts or choices. In Section One, I describe the ways in which autonomy has been discussed, first by two leading liberal theorists, Joel Feinberg and Gerald Dworkin, and second, by Diana Meyers, a liberal feminist critic of traditional liberal accounts. I argue that liberal theorists have developed their conceptions of autonomy at some distance from those attributes of human subjects, such as emotional or relational interdependence and strong gender-related socialization, that bear heavily on the lives of women. Liberal feminists such as Meyers recognize the need for a closer relationship between the conceptual and the empirical in defining autonomy, and introduce an account of gender socialization that highlights the interrelatedness of most women's lives.(1) Meyers' work, however, also retains elements of the individualist orientation of liberalism: it neglects the pervasiveness, variousness, and potential coerciveness of gender and other forms of socialization and it understates the extent to which the development and exercise of autonomy is frequently a collective enterprise rather than an individual enterprise. In the last portion of this section, I turn to a group of feminist theories that help to articulate my own view. These theories reject individualist assumptions and describe a pervasive, plural social construction of the subject, in the context of intersecting power inequalities. These latter accounts, however, do not deny the possibility of self-definition or self-direction among women or others similarly constituted. They reinterpret and resituate these attributes in the context of a distinct understanding of the formation of the subject. In this essay, I acknowledge this reinterpretation by using the term "agency" rather than "autonomy."

In Section Two, I examine the features of "agency" as they have emerged in feminist accounts of women's self-determination. Like many liberal theorists, I assert that agency manifests itself in various forms of self-definition and self-direction. Like Meyers, I highlight, as agency, conduct or competencies that may not have been recognized as agency in the past.(2) However, because I understand agency to emerge in a context of group-based oppressions, I focus on examples of agency that have been obscured by liberal analysis: those aimed at resistance to subordination or oppression. Moreover, in contrast to liberal accounts, I describe such agency as emerging through collective action as well as individual self-reflection, and being directed toward cultural and political, as well as individual targets.

In Section Three, I consider what legal strategies might be used to highlight and facilitate these forms of agency, and what changes in traditional accounts of the legal subject may be required in order to accommodate this conception of agency.

I. AUTONOMY AND WOMEN'S LIVES

Legal scholars often invoke a version of the liberal self when they seek to describe the paradigmatic legal subject. This self is formed prior to social interaction, is not fundamentally shaped by group membership or affiliation, and is capable of autonomous choice.(3) This account may be understood as a shorthand that serves to distinguish the liberal legal self from the subject in other forms of legal and political theory; it also mutes the differences in power or social circumstances, from which liberal legal theory tends to abstract. In the work of liberal theorists who have sought to describe the self, matters are more complicated. These theorists acknowledge that the individual has a history, is subject to the influence of various others, and is subject to obligations and commitments that can constrain his choices.(4) These entanglements, however, are not central to whom liberal theorists understand the individual to be.(5) Moreover, the empirical question whether individuals can and dc successfully extricate themselves from such influences does not appear to have been central in arriving at a conception of autonomy.(6)

In this section, I examine descriptions of the self, capable of autonomy, that appear in works by Joel Feinberg and Gerald Dworkin. Though these accounts differ in the way they describe the effects that influences such as relational commitments or socialization have on the lives of individuals, their treatment ultimately assigns such influences a peripheral rather than a constitutive role in shaping a conception of autonomy.

A. The Autonomous Self and Autonomous Decisionmaking

Joel Feinberg has advanced one of the most thoughtful and comprehensive accounts of the liberal self appropriately described as "autonomous."(7) This account is perhaps best known for its description of the qualities that inhere in one who is autonomous: qualities such as moral authenticity, integrity, and distinct self-identity.(8) These qualities also provide a kind of overview of the self in whom they inhere.

The autonomous individual or self described by Feinberg strives to maintain his rational self-direction in a world populated by impinging judgments and entangling commitments, complicated by uncertain vision and weakness of the will.(9) He works to maintain and operate consistently a moral system that is his own, i.e., one that reflects careful adjustment and internalization of available norms, is true to itself without being inattentive to the needs of others, and resists moral pressure or control by others as well as moral weakness or abdication by oneself.(10) This self is not radically disconnected from social influence or free of relationships with or commitment to others.(11) In fact, Feinberg takes pains to recharacterize accounts that may have overstated the disconnectedness of the autonomous self.(12) In discussing the extent to which autonomous persons are self-created, Feinberg acknowledges both the social influences that help to form one's rudimentary character and the parental influences that help "implant" the potential for authenticity.(13) Similarly, Feinberg stresses that the "moral independence" that characterizes autonomy should not be read to require "an uncommitted person, maximally independent of the demands of others."(14) He continues, "we should conceive of de facto autonomy in such a way that it is not diminished by voluntary commitments, at least below a reasonable threshold."(15)

In important ways, though, these provisos form the exceptions rather than the rule. They create the fine detailing on a portrait of the self that is marked by strong boundaries that are theoretically possible to maintain against the claims and incursion of others. Of the dozen or so characteristics that distinguish the autonomous individual, most highlight forms of self-directedness and distinguish them from the condition of being subject to the influence of, or confined within relationships to, others.(16) Of the characteristic of authenticity, for example, Feinberg explains that the autonomous person "is not merely the mouthpiece of other persons or forces. Rather, his tastes, opinions, ideals, goals, values, and preferences are all authentically his."(17) Of the characteristic of distinct self-identity, Feinberg writes that the autonomous person "is not exhaustively defined by his relations to any particular other. For example, he may protest that he is not content to be known and described merely as the former husband of some movie star...."(18) This self is, in its fullest manifestations, free of externally-imposed decisionmaking.(19) It also is able to impose judicious limits on the claims that are made on its self-directed decisionmaking by others.(20)

Finally, this self is not systematically confined by differentials in power or circumstance that can shape internal processes of judgment.(21) In one passage, Feinberg explains how luck may play a role in the capacity for self-government: "I do not govern myself if you overpower me by brute force and wrongfully impose your will on mine, or if illness throws me into a febrile stupor, delirium, or coma, or if poverty reduces me to abject dependence on the assistance of others."(22) It is worth noting, however, that such luck confines more by limiting external opportunities than by shaping internal perceptions. It also seems closer to the random lot of individuals than to a predictable outcome of particular group membership. Perhaps more importantly, circumstantial "luck" affects few in a prohibitive way, and is less determinative in its impact than excellence of character. "In normal circumstances," Feinberg concludes, "opportunity is more or less available for most people; the autonomous person is the one who makes the most of it."(23)

Even accounts that entail a more frontal acknowledgment of mutual dependence,(24) social influence, or the disparate opportunities of groups echo this individuated emphasis. Gerald Dworkin begins The Concept of Autonomy by noting that certain widely-observable facts appear to challenge the most highly-individuated notions of autonomy: "We know that all individuals have a history. They develop socially and psychologically in a given environment with a set of biological endowments. They mature slowly and are, therefore, heavily influenced by parents, peers, and culture."(25) A viable concept of autonomy must acknowledge these kinds of influences. Dworkin's view combines two elements: authenticity and procedural independence. Authenticity is achieved when an individual's judgments emanate from a motivational structure that he identifies as his own.(26) "It is only when a person identifies with the influences that motivate him, assimilates them to himself ... that these influences are to be identified as `his,'" Dworkin explains.(27) "If, on the contrary, a person resents being motivated in certain ways, is alienated from these influences ... then these influences ... are not viewed by him as `his.'"(28) The second element, procedural independence, is achieved when the individual's identification with his goals "is not itself influenced in ways which make the process of identification in some way alien to the individual."(29)

Dworkin's account seeks to acknowledge the ways that autonomy-experiencing selves may be socialized and situated. Not only are certain forms of socialization an explicit part of his vision. But also, and more importantly, the possibility of being motivated to the point of action by influences that are external, indeed alien, to oneself suggests a considerable degree of interrelation between the self and its surrounding environment. In important respects, however, Dworkin's discussion presses these attributes of the individual subtly to the margin. Social influence is most salient during particular phases of life, such as childhood and immaturity.(30) The opinions and practices of others may shape only minimally individuals who have become mature. To acknowledge an influence that is more than de minimus risks calling the autonomy of a given actor into question.(31) Alien motivational structures are precisely that: distinct from the internal motivational structure of the actor.(32) They are discussed by reference to problems such as drug addiction,(33) which suggests an unequivocal subjugation of the will, and which readers may be inclined to distance from their own experience.

These features of Dworkin's and Feinberg's theories need not mean that they exclude the possibility of systematic social influence, or that they believe it is routinely possible to avoid or transcend such influence. On the contrary, these features may suggest that deciding how likely autonomy is to be achieved, and by whom, simply is not the primary focus of these theorists' inquiry.(34) Their goal may instead be to define with conceptual clarity an ideal state for whatever subjects are able to attain it.(35) In any case, however, assessing the barriers that relational interdependence and socialization present to autonomy and assessing the role they should play in defining autonomy do not figure importantly in mainstream liberal theories of autonomy.

B. Self and Subject in Feminist Theory

1. Feminist Revisions of Liberal Autonomy

These questions have, however, been addressed by feminist theorists working in the liberal tradition. These theorists have insisted that autonomy be defined through closer observation of the lives of human subjects, particularly those subjects who are women.(36) This analysis has resulted in the questioning of some features of the liberal conception as developed by theorists such as Feinberg and Dworkin. A thoughtful and provocative example of work within this genre is Diana Meyers's essay, Personal Autonomy and the Paradox of Feminine Socialization.(37) This essay suggests the breadth and flexibility of liberal understandings of autonomy, as well as the need for movement beyond certain liberal assumptions.(38)

Meyers takes as her focus the life of the "traditional woman": one who is strongly socialized to feminine norms and devotes herself primarily to the care of her family.(39) This woman is marked by several characteristics that would seem likely to disqualify her as an autonomous agent, according to the assumptions of mainstream liberal theory.(40) Her role entangles her indefinitely in relational demands that often supersede her own needs for quiet, privacy, and self-development.(41) Moreover, she has `chosen' this role in large part because of an early and comprehensive socialization that defines this role as appropriate and valuable to her because she is a woman.(42)

In one respect, the magnitude of her constraint, the traditional woman may be distinctive. Yet when Meyers analyzes more carefully the elements that would seem to threaten the traditional woman's autonomy, she finds them everywhere. Virtually all workers are economically dependent on their employers; many workers' labor is motivated by strong affective commitments that tie them to sometimes self-abnegating work; and almost everyone enacts elements of childhood socialization, including femininity, masculinity, and other characteristics.(43)

The prevalence of these characteristics in the lives of many women, as well as their presence in the larger population, leads Meyers to conclude that these qualities should be regarded as restricting or qualifying autonomy, rather than extinguishing it.(44) She develops this view by identifying in the life of the traditional woman three kinds of partial or limited autonomy. The first is "narrowly programmatic autonomy," in which the actor may be unable to answer autonomously the broad question, "how should I live my life?," but may be able to achieve autonomy in responding to that question at more restricted or concrete levels.(45) The second is "episodic autonomy," in which the actor is able to act autonomously in relation to some choices or decisions but not others.(46) The third is "partial access autonomy," in which the actor has access to some, though not all, authentic or unconditioned attributes of self that may form the basis of autonomous decision.(47) Meyers concludes that these notions of limited autonomy point to an understanding of autonomy as a kind of competency, or a set of "introspective, imaginative, reasoning, and volitional skills" that makes it possible to act in a self-aware and self-directed fashion.(48) The traditional woman's role, in many cases, "curtails the development of autonomy competency."(49) Thus, she is capable of achieving some measure(s) of autonomy, but not fully-developed autonomy.(50)

Meyers's account reflects several characteristics that mainstream liberal conceptions of autonomy seem to lack. Most crucially, her account suggests a stronger relation between the conceptual and the empirical in defining autonomy.(51) Dworkin and Feinberg are not unconcerned with the concrete details of human lives, yet they do not dwell on the capacity of the average person to attain the attributes that comprise autonomy or on what it means for their definitions that many do not.(52) They are most concerned with describing what autonomy should be, assuming some human beings are able to achieve it, and with distinguishing conduct or attributes that come within this characterization from those that do not.(53) While they believe that these understandings cannot neglect widely shared features of human experience, such as childhood socialization, they address these features either as general background conditions or, in extreme examples, as limiting cases.(54) They do not describe such features in rich sociological detail, nor do they attempt to modify their conceptions to account for the differential occurrence of these features in the human population.(55)

Meyers's approach is different. In characterizing autonomy, she is concerned with whether it can be achieved and with which groups prevailing definitions tend to exclude.(56) She also is concerned with the traditional woman both as a figure whose distinctive characteristics have been neglected, and as a figure who manifests in pronounced form attributes that a wider variety of women and men share.(57) Many feminists have identified caregiving as an attribute or practice to which a wide range of women are socialized.(58) Women may feel called upon to put the interests of others first or define themselves as someone's mother or someone's wife, regardless of whether they can claim other identities or choose to work outside the home.(59) In addition, Meyers points out, the economic dependence, affective entanglement, and socialization evinced by traditional women extend to many men.(60) Meyers aims to encompass such groups within discussions of autonomy without suggesting that the self-direction they exercise is the full autonomy of which human beings may be capable.

She achieves this goal through the notion of partial or qualified autonomy and through the larger view of autonomy as competence.(61) The designation of partial autonomy is not a patronizing or empty gesture; it is an alternative conception that reveals autonomy to be capable of more complex, varied definition than previously has been thought.(62) It takes partial or qualified forms of autonomy not simply as deficient--though they are clearly deficient in some respects--but as illustrative of the continuum on which autonomy and heteronomy can be found, and of the ways that autonomy is developed as a competence.(63)

This vision of autonomy as a multifaceted, refinable competence is the final distinguishing feature of Meyers's account. Few mainstream accounts treat autonomy as an "on-off' switch. Feinberg's analysis of the attributes of the autonomous individual, for example, suggests that some people achieve autonomy more fully than others. Meyers, however, makes this understanding explicit and central to her conception through her elaboration of partial autonomy.(64) Moreover, in her understanding of autonomy as a competence, she explains how differences and gradations may come into being, while illuminating the range of skills or capacities that are required for full autonomy.(65) Finally, this conception of autonomy performs a valuable function in dignifying those who exercise it, even in part, and in suggesting the possibility of change or growth in autonomy competence. Identifying a series of way stations between complete subjection and full sovereignty over oneself elaborates the condition of the partially autonomous without stigmatizing that condition or describing it as fixed. She who is partially autonomous today retains the potential for full(er) autonomy tomorrow, through the development, under the proper circumstances, of autonomy competence.

2. Critiquing Liberal Feminist Revisions

While Meyers's work begins the modification of liberal notions of autonomy, it does not take us far enough. She demands a closer relation between the conceptualization of autonomy and the sociological observation of the subjects who exercise it,(66) yet her own account neglects important feminist insights about the character and formation of those subjects. Furthermore, while she presents feminine socialization as a challenge to mainstream theories of autonomy,(67) her challenge does not capture the breadth, complexity, or contingency of this process of social construction. Finally, Meyers does not adequately evoke the contexts of differential power in which this socialization operates. This failure has important implications for the ways she defines the exercise of autonomy.

Although Meyers presents socialization as a force that demands revision of liberal accounts of autonomy, her account of women's socialization is surprisingly narrow. For Meyers, women's distinctive, constrained position stems almost exclusively from an ideology of separate "spheres."(68) Women are socialized from childhood to favor care-giving, self-sacrifice, and the satisfactions of domesticity, while men are encouraged to sample the range of more public opportunities that their greater independence allows.(69) A major flaw in this account is that it neglects the range of other ways in which women (and men) may be socialized and the distinct though equally comprehensive ways their autonomy can be affected.

Institutional or cultural androcentrism, for example, also may shape women. Some feminists have argued that women's inequality is reflected in, and perpetuated by, a tendency to value practices, norms, and characteristics socially understood as male, and to place these attributes, as well as the men who manifest them, at the center of social and institutional arrangements.(70) Women socialized in such contexts may accept the subordination of their opinions or aspirations to others, in contexts that have nothing to do with childrearing, caregiving or biological dependence.(71)

Practices of sexualized dominance may also shape women. Feminists such as Catharine MacKinnon have described women, paradigmatically, as being beaten, raped, and subjected to harassment and pornography.(72) More insidiously, these practices of violation extend beyond individual perpetrators to laws and linguistic constructions(73) that characterize public harms as private matters, describe violations as perpetrators' exercise of civil rights, or hold women responsible for the injuries they suffer.(74)

Like socialization to the domestic sphere, these forms of feminine social construction have a range of implications for women's autonomy. They may radically constrain the range of women's choices by making them afraid to live, work, or walk in particular areas, or reluctant to engage in particular practices or voice particular views.(75) They also may be punishing and pervasive enough to stunt women's tastes, values, and conceptions of themselves. Women who are routinely marginalized or objectified may come to see themselves as less worthy or capable human beings whose role is to support, or even service, men. These forms of socialization, like Meyers' socialization to domesticity, can prevent women from formulating or embracing any personal system of values. MacKinnon notes, "when you are powerless, you don't just speak differently. A lot, you don't speak.... [Y]ou are deprived of a life out of which articulation might come ... prevented from having anything to say."(76) Through these forms of socialization, a woman also may end up embracing socially-enforced norms to such a degree that it becomes impossible for a woman (or anyone else) to be sure whether her norms have "internal" or "external" sources.(77)

Even this broader catalogue of sexist influences is flawed, in that it, like Meyers's account, appears to suggest that feminine socialization affects all groups of women in similar ways. Many recent feminist theorists, though, have made clear that patterns of gender socialization intersect with, and may be influenced by race, class, sexual orientation, and variations of time, place, and political context.(78) These complex, contingent, intersectional patterns create a vast mosaic of socially-engendered practices and responses that are difficult to generalize about or to predict in advance. These complex patterns of socialization are also implemented by a far broader range of instrumentalities than Meyers tends to acknowledge. Meyers focuses on childhood socialization and education, as well as interpersonal pressures to conform.(79) Structuralists, such as MacKinnon, and the feminist poststructuralists who have followed her, point not only to these comparatively obvious and benign means, but also to stark disciplinary practices, including stigmatization, harassment, and riolence,(80) and to the pervasive force of linguistic construction,(81) These insights mean that socialization affects far more people, in a far more complicated manner, than Meyers suggests, making partial autonomy less the exception than the rule. They also mean that earlier liberal efforts to distinguish "internal" and "external" influences on individual choice become not just difficult, but virtually unintelligible.

Structuralist and poststructuralist feminists also underscore a dimension of power that is largely absent from Meyers's account.(82) Meyers describes women's socialization as comparatively disadvantageous for the exercise of autonomy.(83) She does not, however, stress that it also produces systematic inequalities, nor does she probe the question of whose interests such inequalities or such socialization serve. Later feminists do not describe simply undifferentiated, undirected `social construction,' but also intersecting practices of construction shaped by more powerful groups that tend to socialize less powerful groups to various forms of subordination.(84) This element of political context is crucial, as it provides a starting point for certain dimensions of constraint and autonomy that Meyers' account neglects. It suggests that there are forms of autonomy that are best understood as resistance, not to individual attempts at interference, but to group-based attempts at disempowerment. It reveals a collective dimension to the development and exercise of autonomy that Meyers fails to develop adequately.

Consistent with its liberal premises, Meyers's account of autonomy is, in most respects, an individualized conception. Although she acknowledges that autonomy is developed in a social context, sometimes through explicit collaboration with friends and others,(85) autonomy competency is honed through a process of individual self-reflection and the development of self-awareness and critical judgment.(86) Moreover, the purposes to which autonomy is directed are largely the individual's own. When that individual must confront or engage others in order to implement her autonomously chosen program, it is most often with nonspecific individuals or with the paradigmatic liberal threat to autonomy: the agent of the state.(87) Meyers does not acknowledge the extent to which the development of one's self-awareness, or self-definition, may be crucially dependent on political or cultural exchange and collaboration with others. She does not stress that the commitments that structure one's exercise of autonomy may be formulated through collective exchange and action. Nor does she explain that the group-based resistance that can constitute autonomy in a context of unequal power may be directed at a variety of cultural practices, institutional arrangements, and officers of the state. It is possible, however, to formulate a feminist theory that underscores the collective dimension of self-definition and self-direction, and that highlights the processes of complex social construction in a field of unequal power that Meyers' theory largely neglects. In the following section, I develop one such theory.

II. FROM AUTONOMY TO AGENCY

Feminist theorists have offered a series of insights about women's circumstances and the social formation of human subjects that have challenged many aspects of liberal autonomy. Liberal feminists such as Meyers have challenged the attenuated connection between theoretical descriptions of autonomy and nuanced observation of the lives of those who exercise it.(88) Using the phenomenon of feminine socialization, they have highlighted the difficulty of drawing many of the distinctions central to traditional liberal accounts, such as the distinction between internal and external sources of direction for particular acts or choices.(89) In fact, liberal feminists such as Meyers have problematized, or at least relativized, the entire process of distinguishing autonomy from heteronomy by arguing that there are many forms of partial autonomy, and that autonomy should be conceived as a competence developed over time through a process of exercise and self-reflection.(90)

Subsequent feminists, including structuralists and poststructuralists, have introduced further challenges. They argue that the socialization that Meyers sees as sufficiently influential to give rise to the category of partial autonomy is, in fact, more pervasive, complex, and variable than even she suggests.(91) Not only the most traditional groups of women, but virtually everyone is subject to formation by social norms, images, and practices. Unlike the socialization to domesticity highlighted by Meyers, however, these influences are multiple, specific to particular contexts, and capable of shaping and intersecting with each other in innumerable, unpredictable combinations.(92) The variety and pervasiveness of these influences makes the question of distinguishing internal from external direction not simply difficult, but almost unintelligible. Moreover, this complex socialization operates in a field of power relations that has not been highlighted either by feminist liberal theorists or traditional liberal theorists of autonomy. This means that practices contributing to socialization may operate not simply on individuals, but on individuals as members of groups, and that reinterpreting and resisting these practices or socially-assigned meanings may be one way of exercising self-direction. This political context also suggests that the processes of self-definition and self-assertion that have been characterized as autonomy may be more collective than liberal theorists have suggested, both in their genesis and in the targets of their operation.

These differences in assumptions, including the insistence on a closer fit between the conceptual and the empirical in defining autonomy, the acknowledgment of myriad ways in which social construction alters the conceptualization of autonomy, and the recognition of a context of unequal power relations that animates and gives collective character to many forms of self-direction, give "autonomy" a different meaning than it has had in liberal philosophy. I will acknowledge this difference by using the term "agency" to characterize self-definition and self-direction under this distinct conceptual framework. Given these assumptions, the way in which human subjects exercise agency will be the focus of the following section. I will explore two dimensions of agency: one which I will refer to as self-definition, and another which I will refer to as self-direction.(93) In exploring each dimension, I will emphasize the collective aspect of the genesis or exercise of agency, the meaning of agency as resistance, the range of institutional and social targets in relation to which agency may be exercised, and other features that distinguish this account from its liberal predecessors.

A. Agency as Self-Definition

Many philosophers focus on an aspect of autonomy that they describe as self-definition. Self-definition may be described as determining how one conceives of oneself in terms of the goals one wants to achieve and the kind of person, with particular values and attributes, one considers oneself to be.(94) In many forms of liberal philosophy, this process involves distinguishing one's own values, or sense of oneself, from those that may be externally imposed; that is, values one gets, in the words of Robert Young, "secondhand."(95) In the work of feminist liberals such as Meyers, distinguishing between autonomy and heteronomy in the area of self-definition is more complicated. An identity defined wholly in relation to the needs of others, or a sense of self that mirrors precisely the norms of a conspicuous form of socialization, may raise a kind of red flag where autonomy is concerned. We also must consider, however, whether the person exhibits any of the forms of partial autonomy that Meyers describes and whether she manifests, or her life provides her with opportunities to develop, the plural competencies that make self-definition and the other manifestations of autonomy possible.

The concept of agency that I draw from recent feminist work differs from the foregoing accounts in several ways. The first difference lies in the effect of complex social formation. The socialization that liberals view as hindering autonomy is sufficiently comp]ex and pervasive that it cannot simply be transcended. Agency must operate within and in relation to this socialization.(96) Self-definition does not occur through a process of excavating the pre-social self or disentangling oneself from social influences. Most feminist theorists of complex social construction would regard this as futile.(97) Self-definition occurs, first, by becoming aware of the way that one's self, and one's self-conception, are socially constituted.(98) A woman may become aware, for example, that images or attitudes she has regarding her body, her competence to perform certain tasks, or her strength or vulnerability in relation to others, are shaped by norms that describe these matters at least partly as a function of gender. Developing this awareness does not permit her to transcend these socially conditioned visions of self, but it allows her greater room in which to affirm, reinterpret, resist, or partially replace them. When she becomes aware of them not solely as attitudes that she holds, but as norms embedded in social institutions or practices that are transmitted to her and to others, she may begin to compare them with other feelings or intuitions that she has about herself. Though she does not have recourse to some complete, pre-social self that can be uncovered, she may draw on moments of insight that arise from her reflection on her experience, or attitudes she holds that are shaped by other social infiuences. This process of reflection and comparison, which is facilitated by her awareness of certain self-conceptions as socially shaped, may allow her to identify more strongly with certain images and strive for greater distance from others.(99)

The second distinctive feature of feminist "agency" is that there is a political dimension to this process of recognizing and reflecting on the influence of social norms. Social or cultural norms that embody negative judgments about women's bodies, women's competence, or women's power in relation to others are not mere coincidence. They are a product of, and a means by which, women's oppression is perpetuated in particular settings.(100) These norms do not simply make it more difficult for women to develop independent self-conceptions. While this is a critique that might emerge from a largely liberal account, such as that of Meyers, a vision of complex social construction problematizes the notion of an "`independent" self-conception. These norms make it more difficult for women to develop a positive self-conception that permits them to take part in a range of satisfying choices or projects.(101) Members of more privileged groups, in contrast, may benefit from norms that are enabling and that cause them to see themselves as attractive, competent, and capable in relation to others, permitting them to undertake a wide range of choices and projects.

The complexity of social influence, however, means that one's capacity for self-definition is not simply a function of belonging to one empowered or disempowered group. As subjects with a range of attributes that are assigned innumerable meanings in a particular culture, our ability to define ourselves in positive and authorizing ways may be assisted by some of these meanings, and undermined by others. A woman of color, for example, may be disenabled by social meanings assigned to her capability or competence by virtue of her race or gender, but authorized by the centrality or normality conferred by her heterosexuality, or by positive meanings assigned to her race and gender attributes by a smaller group, such as her church or a circle of friends. She may be empowered by some of the meanings assigned to black womanhood (e.g., strong matriarchal head of household) and disempowered by others (e.g., dependent single parent, or emasculating threat to the traditional family form).

This multiplicity of norms and the political context in which it occurs has several consequences. Developing a self-conception that is authorizing, and at least partially free of the negative conceptions that are part of the complicated process of groupbased subordination, is comparatively difficult for members of disempowered groups. It is far from impossible, however, because identities and the social meanings assigned to them are complex. The intersection of a variety of group-based norms may prove authorizing for an individual woman even if she is not precisely aware of the social sources of her authorization and constraint. More importantly, however, once she becomes aware of the influence of some group-based norms on her self-definition, she may be able to use or even create others to support a more enabling self-conception and to resist the influence of more dominant disenabling norms.(102)

This point leads to the final difference between feminist conceptions of agency and liberal conceptions of autonomy in regard to self-definition. The conception of agency I articulate here acknowledges the collective aspect of the process of self-definition to a far greater degree.(103) This is true first of the process by which one becomes aware of social influences on one's self-conception. As feminist writing on "consciousness-raising" has made clear, it may be only through conversation with others who have confronted similar feelings that a woman becomes aware that her self-conception does not simply reflect her own shortcomings, but is a function of views and expectations that are instilled socially.(104) The collective dimension of self-definition is even clearer in that phase that involves creating or mobilizing competing self-conceptions for purposes of resisting prevalent negative imagery. Patricia Hill Collins has written eloquently of this process as it occurs in the lives of black women. This process may begin with flashes of dissonance:(105) potent, if transitory, perceptions that the images that reflect dominant conceptions of black women do not accord with individual women's own senses of their lives, mothering, or sexuality.(106) These moments of insight may draw on, or be supported by, oppositional or dissonant cultural images of women's lives. Collins, Hortense Spillers,(107) and others describe black women as developing alternative visions of their sexuality, or of their control in a largely hostile social universe, by drawing on the images of female blues singers, or the lyrics of black women's music.(108) Women may also develop their own images by altering or recontextualizing dominant images of women.(109) As Collins suggests, the generation of these competing images requires collectivity.(110) Moreover, the appropriation of a collective vision, as a means through which individual women can see themselves differently, requires women to join or at least expose themselves to the operation of that collectivity.(111) In a variety of formal settings, informal groups, and images of connection,(112) women act together to develop alternative self-definitions, elaborate dissonant messages in cultural sources, and provide an affirming audience for emerging self-conceptions.(113) These collective forms of agency expand upon the liberal view that self-definition is achieved through solitary self-reflection or occasional conversation with a friend.

B. Agency as Self-Direction

A second aspect of autonomy described by liberal philosophers consists in the direction of one's own course, including the identification of particular goals and the implementation of particular projects and lifeplans. Liberals focus on the ability to formulate goals and plans that are one's own, as opposed to the products of the influence of others, and to follow through on these without distraction, dissuasion, or weakness of' will.(114) For Meyers, the analysis is much the same, except that the interference, distraction, or dissuasion occasioned by female socialization may be transferred from the outside to the internal mental processes of the individual female subject.(115) The distinctive features of a constructivist feminist conception of agency produce a greater set of changes in these understandings.

When social influences are pervasive, and pervasively internalized, charting a course that is distinctively one's own may be more difficult to separate, both empirically and conceptually, from pursuing a course that arises from socialization.(116) Some of the distinguishing features of agency in the context of self-definition may help identify agency here as well. The first is the important role played by recognition of social influence.(117) Awareness of the ways in which social formation may shape one's goals, or the ways in which one pursues them, may provide room in which to affirm, reject, or modify some of these means and ends. When a woman realizes, for example, that her reluctance to undertake physically demanding work may stem less from her own physical incapacity than from a set of social norms that allocate jobs involving physical demands or danger to men, it may be easier for her to contemplate pursuing a blue collar trade.

A second common theme is that barriers to self-direction may emerge as part of a larger system of social and political inequality.(118) Obstacles that appear, as Feinberg notes, as a matter of "luck" may hinder achievement of one's goals: hindrances also may be traceable to people with individualized motives. But a feminist account of agency supplies the insight that people may be hindered by obstacles that bear on them as members of a disempowered group, whether these obstacles are imposed from without, instilled within, or both. Such obstacles may consist of norms that constrain choices or deprecate abilities; or they may consist of practices that intimidate, belittle, or degrade. The woman deterred from blue collar work by social norms marking such work as masculine also may find, when she attempts to undertake such work, that she is met by hostile or sexualized language, physical or sexual assault, equipment sabotage, and inadequate mentoring and supervision. Each of these forms of interference may prevent a woman from exploring or achieving a range of goals, and each is imposed upon her specifically because she is a woman.(119) Opposition to these politically-based obstacles is a final distinctive feature of agency as self-direction. I will focus on this feature at length here because it has received comparatively little treatment in the liberal philosophical literature.

In setting forth a kind of typology of responses to group-based obstacles to self-direction, I hope to underscore three things. The first is the range of responses that reflect agency as self-direction, some of which may not be recognized as such, and some of which even may be mistaken as acquiescence. By identifying the subtle and overt forms in which resistant agency manifests itself, I hope to offer a more complete descriptive account of wornen's agency, and extend our notions of the kinds of conduct that should be understood as resistance. The second point I will underscore is the extent to which many forms of agency as self-direction require collaboration with others. In many accounts of liberal autonomy, other people function primarily as obstacles to the realization of an individual's plans or goals. While others may assist in an intermittent and episodic way, the individual is the primary source of his own lifeplans and the primary instrument for achieving them. In the account below, participation with others is sometimes necessary to conceive particular goals in the first place, and it is often necessary in order to achieve them. This need for collective action arises from the context of inequality in which these efforts at self-direction occur. To the extent that barriers to particular choices or lifeplans arise from systematic, institutionalized inequality,(120) collective action may be necessary in order to resist or transform these entrenched barriers.(121) To the extent that internalization of these barriers makes it difficult even to conceive particular goals or plans, it may be necessary to work with others in order to loosen the hold of dominant forms of socialization.(122) The final point I hope to emphasize below is the range of targets at which women's self-directive resistance may be aimed. This is particularly true of those forms of agentic resistance I will call "transformative."(123) Many feminist constructivists describe what Kirstie McClure has called the "diffusion of [the] political"(124): that the power to produce political inequalities emanates not only from the institutions of the political system per se, but from social, cultural, and linguistic practices.(125) Resistance that seeks to transform these inequalities therefore may be directed not simply at governmental institutions, but at social interactions, cultural representations, or uses of language as well.

1. Resistant Self-Direction

Some forms of self-direction occur in contexts where social or political transformation is not the goal of the resistance itself. Women who resist in these ways may not be seeking to transform society in any systematic sense, but simply to pursue their own choices and plans in contexts where doing so evokes serious gender-based challenge. A woman who chooses traditionally masculine work may be met with harassment from her colleagues. A woman who exhibits independence within, or seeks to extricate herself from, an abusive relationship may be met with domestic violence. The ways in which women respond to these challenges often constitute a form of self-direction or an effort to negotiate gender-based obstacles in order to achieve their larger goals.(126) These responses, however, do not always involve explicit confrontation; therefore, they are not always recognized as forms of resistance or as manifestations of agency. Two examples may help make the range of such resistant self-direction clear.

Psychologist Louise Fitzgerald has studied the responses of women subjected to sexual harassment. Her work suggests that women only rarely object directly to the perpetrator or report the practice to supervisors.(127) Fitzgerald observes, however, that women are not passive in the face of harassment. Most avail themselves of a repertoire of forms of indirect resistance that signal their distaste for the conduct and their desire that it cease.(128) Women may change the subject, use humor to diffuse the impact of the treatment or comment, leave the room when offensive conduct occurs, or avoid the perpetrator.(129) The muted, indirect character of such response has made it difficult for many observers to recognize it as resistance.(130) These moves are not directly confrontative; they may not serve, in many cases, to end the harassing conduct.(131) They do, however, seek to control or mitigate the effect this conduct has on the woman who is its target, and to permit her to function in the workplace in which it occurs.(132) This suggests that they are legitimate forms of self-direction that she exercises on her own behalf. In some cases, the muted character of such response may identify it as a form of "partial access autonomy," the expression of a woman whose socialization to feminine passivity prevents her from tapping the sources of more forthright resistance. In many cases, the muted or indirect character of such resistance reflects the calculated or instinctive response of a woman who seeks to protect her livelihood in the face of a behavioral obstacle.

Martha Mahoney's work on battered women conveys a similar message. Mahoney has argued that the legal images of battered women that have emerged from strategies of battered women's self-defense often depict them as pathologically passive in the face of their partners' abuse.(133) Her work demonstrates, though, that the reality of most battered women's lives is far more complex.(134) Women may be unable to bring a battering relationship to a rapid close, but they may assert themselves in a variety of ways that contribute to their security--and to that of their children--and equip them ultimately to end the relationship.(135) They may act to protect their children from abuse and otherwise secure their well-being.(136) They may amass funds, information, and support that ultimately help them to escape the relationship.(137) They may strategize actively in the face of separation abuse, which exposes women to increased violence when they try to leave, and succeed at separating after multiple attempts.(138)

These aspects of battered women's self-direction need to be recognized as a form of agency. They permit battered women to protect their children, to preserve specific portions of their lives, and in some cases, to exit their abusive relationships. These examples of battered women's self-direction also reflect agency's variable, context-specific, nonunitary character.(139) A woman's self-assertion may be prominent in some contexts of her life and virtually absent in others. Examined across time, or in different areas of a woman's life, it may present a multi-faceted picture of her ability to direct her course.(140) These characteristics may be attributable in part to the greater external and internal constraints imposed uppon women and other members of disempowered groups, but they also reflect the response of a subject who is formed by a complex array of social influences. The way these numerous, variable factors intersect in a particular time and place may determine her capabilities for self-direction in that context. This intersection not only bears on, but may, indeed, help her to form her own trajectory of self-assertion.(141) The integrity or coherence associated with feminist notions of agency may thus be less stringent and more context-based than the conceptions of autonomy advanced by liberal theorists.

2. Transformative Self-Direction

Other forms of self-direction occur in a context in which a broader awareness of group-based oppression has made social transformation the actor's primary goal. This explicit resistance to oppression, which expresses itself in a variety of forms, is a kind of agency that has not been explored frequently, even in work on members of disempowered groups such as women.

Transformative agency may be reflected in resistance aimed at formal political or legal institutions by individuals or groups. Catharine MacKinnon's essay, Linda's Life and Andrea's Work, offers a powerful example of this form of agency.(142) In this essay, MacKinnon traces the paths of three women, Linda Marchiano, Andrea Dworkin, and herself, in their movement for the legal regulation of pornography.(143) These women, MacKinnon argues, have raised their voices in ways that are not only dissonant, in relation to the powerful, constructing force of the pornography industry, but unexpected, given the patterns of oppression they describe.(144) One of the primary effects of pornographic representations, MacKinnon argues, is to construct women as sexual objects, incapable of expressing themselves in their own voice, or manifesting a point of view.(145) For Linda Marchiano to have survived the coerced filming of "Deep Throat," and raised her voice against its objectifications, is a wholly unexpected act of self-assertion.(146) Andrea Dworkin's theorization of the male domination of society and state in a context where these forces combine to render her perspective unintelligible, and Catharine MacKinnon's creation of claims for legal relief in a legal system that had constructed harassment as harmless flirtation and pornography as a civil right, are equally startling and unexplained.(147)

A glimpse into the operation of institutions, or a vivid experiential insight into their own constraint, may move such women to grasp a vision of themselves or an explanation of their lives in particular settings that is contrary to the images propounded by dominant social arrangements. They may then be encouraged, by the sheer force of their insight, the sense of power conferred by other group-based attributes, the support of other women, or the receptiveness or the resistance of their first "outside" audiences, to act on this insight to alter the arrangements in which they find themselves. Both the ability to glimpse an explanation rendered unintelligible by existing practices and the ability to act on that insight to change those practices are forms of agency that emerge, often dramatically, in women's lives.(148) These forms of agency reflect inevitable gaps in the structures that produce oppression, as well as the assertion of a multiply-constructed will: they occur in what we might conceive as the conceptual space between social influence and social determination.

Self-reflection and self-assertion, however, need not produce action aimed at governmental institutions. Some forms of self-assertion may be aimed at mobilizing groups of women, or at altering social norms that have constrained them. Efforts to unionize secretarial and clerical workers on university campuses,(149) or exotic dancers in urban areas,(150) reflect another example of transformative agency aimed at group-based mobilization, rather than at governmental action, and the consequent alteration of private institutions and social understandings. Such efforts challenged popular conceptions of groups consistently devalued under dominant gender-based norms. Moreover, they used collective action to secure better wages and conditions for the labor they performed. The "Take Back the Night" movement was an effort by women to confront the sexualized domination that objectively endangered and subjectively terrorized them.(151) The women involved sought to reclaim through collective, symbolic action the times and spaces from which danger and fear had sequestered them, and to implement practical strategies to enhance their security.(152)

Still other forms of transformative agency have been aimed at disrupting cultural productions, or small-scale social practices, that contribute to women's oppression. Acts performed by individual women in bedrooms, on buses, in commercial exchanges, or through visual arts of "fan 'zines" may be understood as challenges to oppressive arrangements. This view is based on the premise that disrupting the uncontested primacy of a particular image or understanding may encourage observers to question their assumptions. In so doing, it may create credibility for competing images and understandings.(153) Mundane interactions or stark acts of resistance may create these disruptions. The refusal even to pretend to be amused by a sexist joke that produces solidarity or complicity among its listeners is an example of a small-scale disruption. Aggressive, noncompliant behavior by the target of a rape attempt, designed to disrupt the rapist's "script" of male aggression and female vulnerability, is an example of a larger and riskier disruption.(154)

Many of the individual acts that seek transformation through disruption of dominant discourses occur in the realm of the sexual. The assertion of women's sexual subjectivity can be a form of resistance in and of itself,(155) given many feminists' convictions that practices of sexualized domination, Victorian moralism, repressive feminism, or a combination thereof, have tended to erase or eradicate women's sexual subjectivity.(156) Women may attempt to disrupt dominant sexual discourses through the creation of women's erotica(157) or through individual practices of self-presentation.(158) Particular strategies may include role reversals, juxtaposition of conflicting images, or other alterations of dominant imagery that call assumptions about gender identity into question.(159) They may include projections of individual women's conceptions of sexuality that are more idiosyncratic in that they involve neither dominant nor even identifiable oppositional imagery.(160)

Because these efforts to disrupt social or cultural practices are premised on an extremely diffused sphere of the political, those involved do not always view collective action as necessary to create the disruption that catalyzes change.(161) Some of these efforts have been undertaken collectively, however. They represent a form of group-based agency that can shape cultural practices to an arguably broader extent.(162) The "Riot Grrls" movement, which seeks to shape cultural imagery through the proliferation of a kind of grunge rock music and lyrics, and advocates a distinctive brand of self-acceptance and self-assertion through the dissemination of "fan `zines,"(163) exemplifies this collective effort to produce change through cultural intervention.

III. AGENCY AND LAW

If one is persuaded, as an initial matter, that the forms of agency highlighted above accurately describe and normatively benefit the lives of women, one question that follows--particularly if one is a legal scholar--is how such understandings might be brought to the field of law. Legal institutions are not, of course the only institutions that might be used to foster agency. In fact, the social constructivist assumption that law is simply one factor among many producing oppression suggests that this might not even be the most important place to start. Yet legal rules can bring the coercive arm of the state to bear on practices perpetuating oppression; moreover, they represent a discrete source of social meaning on which it is possible to focus in a coherent way. What might law do to foster the forms of agency discussed in this Essay? And what forms of mutual accommodation would need to occur between the feminist notions of agency developed here and the premises of a primarily liberal legal system? In the following section, I discuss two kinds of approaches to bringing a feminist conception of agency to the field of law. The first approach looks at the matter instrumentally: how might we use the law to foster agency of these types? I will examine this question looking at both the explicit, prescriptive role of legal rules and the subtler, constitutive role of legal ideas and images. The second, which I will discuss briefly, is more broadly revisionary. It assumes that feminist theories of agency endeavor not simply to recharacterize women's lives, but to change the way we conceptualize the human subject and its formation. It asks what central assumptions of liberal legal theory and related doctrine would have to be modified in order to reflect this view of human subjects before the law.

A. Instrumental Approaches

The law might be used in a range of different ways to facilitate or enhance women's agency. Most of these approaches reflect one of two different strategies. First, legal actors can draw on the explicit prescriptive, or prohibitive, power of legal rules. This strategy may first seem paradoxical, for when legal actors assume a conventional liberal notion of autonomy, legal regulation is rarely conceived as a promising means to vindicate it. When self-direction is understood as acting according to one's internal metric(s) of value, legal actors have concluded that the best way to foster such self-direction is to back off: in short, to leave particular areas unregulated and subject to the operation of individual choice. Consider, for example, the regulation of pornography. For many years the call for legal regulation of pornography was made by feminists who sought to secure women from organized, sexualized oppression;(164) feminists and others who were concerned primarily with women's sexual expression or agency did not favor regulatory solutions but rather advocated deregulation, sometimes as a backdrop for such proposals as boycotts or proliferation of feminist erotica.(165) However, as women's self-direction, sexual and otherwise, has come to be understood as a multiple socially-influenced phenomenon that takes place in a context of shifting and unequal power relations, some theorists have begun to argue that legal regulation in this area may help to foster agency.

Although feminists interested in sexual agency have remained concerned that regulation of pornographic materials would reinforce the judgmental norms that have confined and stigmatized women's expressions of sexuality, some have argued that regulation of related sexually oppressive practices might actually help women to exercise agency in the area of sexuality. Susan Keller has argued, for example, that women and men might feel freer to experiment with the interpretation and use of a range of sexual representations if they knew that women had not been coerced in making them. She thus proposes to use the MacKinnon-Dworkin provision prohibiting the coercion of actors and models to facilitate such experimentation and expression.(166) I have proposed looking at the other end of the production process: deploying the MacKinnon-Dworkin provision that targets the use of pornography in a subsequent sexual assault might also fuel sexual experimentation and improvisation by decoupling pornography from coercion in yet another way.(167) Carlin Meyer, who opposes the use of the MacKinnon-Dworkin provision, has argued that feminists should use law to restrict the power of media conglomerates to dominate the production of culture--in particular, cultural images of women that fuel their sexualization.(168) All these legal strategies use the power of prohibitory legal rules to foster women's agency in the domaain of sexuality and self-presentation.

Another approach would draw not on the prohibitory power of legal rules, but on the constitutive role of legal ideas and images. Legal depictions are part of a larger set of political and cultural depictions that help to shape our sense--as women and as human being more generally--of who we are.(169) An awareness of this effect in articulating rules and requirements that bear on women's agency may help to shape popular conceptions of women's capacity for self-direction. For example, in the discussion of resistant or partial agency above, I noted that observers often miss this agency when it emerges because it does not conform to their more confrontational expectations about what it means to resist oppression. This misperception often has been costly for women, because when observers fail to see the response they expect, they conclude that women are passive in the face of oppression. This leads many observers, including some legal actors, to assume that women are either weak, wholly compromised figures who can be treated paternalistically, or inadequately assertive individuals who should be compelled by the use of legal incentives to defend their own rights.(170) Were legal rules, instead, to highlight the constrained forms of resistance that women already undertake, this might help to relieve women from the burden of these misrepresentations. To the extent legal rulings shape broader social understandings, this use of law might also help observers to understand the various ways that oppression conditions women's responses, and the unrealistic character of the expectation that women generally will respond confrontationally to gender-specific injuries.

Drawing on the constitutive or representational role of law is not always easy, or even possible, because courts formulate legal rules to resolve particular cases (or, more broadly, give social direction), not to represent human subjects in particular ways. Yet feminist advocates who are alert to representational issues may help shape legal rules in ways that serve both litigants and the larger groups of which they are a part. Legal doctrine in many areas involving gender-specific injuries requires plaintiffs to demonstrate that they resisted the sexualized treatment;(171) such rules may provide one avenue for challenging dominant understandings through legal imagery. Adapting the showings required, to reflect women's less overtly confrontational responses, could be an efficacious means of underscoring the character of women's response and relieving women plaintiffs from what have sometimes become unduly onerous burdens in litigation.

In the sexual harassment area, for example, claimants are required to demonstrate that the behavior of which they complain was unwelcome. Courts have vacillated between a stringent requirement that women demonstrate a contemporaneous objection to a perpetrator or a supervisor, which often results in the dismissal of the claim, and a fluid requirement that the court simply rely on the plaintiffs credibility in asserting that the conduct was unwelcome, which permits plaintiffs to recover more readily but may fuel popular perceptions that plaintiffs are passive in relation to harassing treatment. The research by Fitzgerald and others suggesting that women employ a repertoire of indirect responses to harassment may be useful in this context. If courts were to accept evidence that women changed the subject, left the room, or avoided the harasser as demonstrating the unwelcomeness of the conduct, this would highlight the strategies women most frequently use to respond to harassment; moreover, it would be unlikely to impede recovery because empirical evidence indicates such responses are comparatively widespread. A few courts have recently taken this route;(172) a more fully theorized explanation of why such evidence is illuminating might persuade others to follow suit.

B. Reconstructing Liberal Assumptions

It may be possible to do more, however, than simply use the law to highlight certain instances of agency as self-direction or resistance. We might instead, or in addition, seek to reconstruct the largely-liberal legal system to comprehend and respond to a human subject whose self-definition and self-direction have more the quality of feminist agency than of liberal autonomy. If we sought to pursue this broader goal, what premises or assumptions would need to be modified? The possibilities are numerous, but two suggestions should provide a sense of the scope of the task.

First, the legal system would need more fully to acknowledge, and to represent, the social formation of human subjects. There are predictable boundaries on the extent to which such change could occur: there is only a limited amount of complexity in social formation that courts might be expected to acknowledge.(173) Moreover, a view that went so far as to posit social determinism might threaten the central legal premise that individuals should be held responsible for the consequences of their (illegal) actions.(174) Yet such factors might best be characterized as outside limits; neither militates decisively against a greater recognition that human beings are formed in their preferences, abilities, and capacities to respond to coercion, by material circumstances, and relationships or affiliations with others. Although the courts have moved away from this recognition in areas such as equal protection,(175) they have taken more promising approaches in other fields, including some common law and statutory, areas.(176) A crucial part of acknowledging the effects of social situatedness would be to highlight the effects of inequalities of power.

Second, the legal system would need to acknowledge variation and contingency, not only in the agency of any given subject, but in other human attributes as well. Because liberals assume that the salient attributes of the subject are formed prior to social interaction, liberal legal decisionmakers have been slow to comprehend the differences that social context may make in a subject's characteristics or responses. With respect to agency, for example, subjects are expected to be unambivalently assertive or thoroughgoingly submissive, wholly compromised or fully resilient; courts may respond with confusion or resistance when subjects manifest (or advocates highlight) greater context-based variability. A woman who suffers a sexualized injury, but is nonetheless self-possessed or even assertive, may evoke perplexity or disbelief from a liberal judiciary.(177) Some of this confusion may stem from impatience with claims of gender-specific injury, and a related desire to cabin the claimant within the polar stereotypes of maladjusted complaint or wounded femininity. But some judicial confusion may stem from the expectation of a kind of characterological unity in the subject that tends to be fostered by liberalism. For scholars and advocates to insist on variability in individual legal subjects, and on the manifestation in different contexts of divergent characteristics or responses, may help to overcome this expectation and to help legal decisionmakers accommodate a more complex subject.

IV. CONCLUSION

Many transformational steps may need to be taken before the legal system can embrace a subject whose engagements with others reflect the notions of agency described above. But as scholars and advocates begin to highlight both these forms of situated self-assertion and the circumstances of authorization and constraint that give rise to them, we will come closer to describing the self-direction not only of women, but of socially-embedded, differentially-empowered human beings.

(1.) See DIANA T. MEYERS, SELF, SOCIETY, AND PERSONAL CHOICE 152-55 (1989) [hereinafter MEYERS, SELF, SOCIETY, AND PERSONAL CHOICE]; Diana T. Meyers, Personal Autonomy and the Paradox of Feminine Socialization, 84 J. PHIL. 619, 621-24 (1987) [hereinafter Meyers, Personal Autonomy].

(2.) Cf. Meyers, Personal Autonomy, supra note 1, at 619-20 (observing that some feminists have embraced "distinctively feminine styles of agency").

(3.) See, e.g., James Boyle, Is Subjectivity Possible? The Postmodern Subject in Legal Theory, 62 U. COLO. L. REV. 489 (1991); Tracy E. Higgins, Democracy and Feminism, 110 HARV. L. REV. 1657, 1659 (1997).

(4.) See Joel Feinberg, Autonomy, in THE INNER CITADEL: ESSAYS ON INDIVIDUAL AUTONOMY 27, 45 (John Christman ed., 1989) [hereinafter THE INNER CITADEL].

(5.) See id. at 33-34.

(6.) See generally id. at 33 (discussing the extent to which an autonomous person is "self-made" as only a "modest" point in developing a theory of autonomy).

(7.) See id. at 30-43.

(8.) See id.

(9.) See id. at 45.

(10.) See id. at 36-38.

(11.) See id. at 33-34.

(12.) See id.

(13.) Feinberg notes:
   The autonomous person is often thought of as a "self-made man."... Even his
   character as authentic cannot be entirely the product of his own doing. To
   suppose otherwise is to conceive of authenticity in such an exalted way
   that its criteria can never be satisfied.... [A] person must already
   possess at least a rudimentary character before he can hope to choose a new
   one. The other side of that point is that if a child needs to "learn to be
   authentic," it must be the case that he is not already authentic when he
   starts. There can be no magical ex nihilo creation of the habit of rational
   reflection. Some principles, and especially the commitment to reasonable
   self-criticism itself, must be "implanted" in a child if she is to have a
   reasonable opportunity of playing a part in the direction of her own
   growth.


Id. at 33 (footnote omitted).

(14.) Id. at 38 (footnote omitted). Feinberg explains:
   [It] is hard to imagine such a person with the moral virtues that thrive on
   involvement--compassion, loyalty, cooperativeness, engagement, trust. If we
   think of autonomy as de facto independence simply, then the uncommitted
   person is an autonomy-hoarder, who scores high on our scale. But if we
   think of autonomy as the name of a condition which is itself admirable, a
   kind of ideal condition, then the uncommitted person is subject to demerits
   on his score. He is clearly no paragon.


Id.

(15.) Id.

(16.) See id. at 30-43.

(17.) Id. at 32. Feinberg continues, "[o]ne way of being inauthentic, so understood, is to be a habitual and uncritical conformist who receives his signals from some group whose good opinion he needs, or from unknown tastemakers in the advertising agencies and public relations firms." Id.

(18.) Id. at 31-32.

(19.) See id. at 32.

(20.) See id. at 32-34.

(21.) See id. at 31.

(22.) Id.

(23.) Id.

(24.) In one thoughtful feminist account of autonomy and interdependence in liberal philosophy, for example, Feinberg and Dworkin are described as examples of liberal theorists who strive to account; for human (and, in many cases, female) attributes such as interdependence. See Marilyn Friedman, Autonomy and Social Relationships: Rethinking the Feminist Critique, in FEMINISTS RETHINK THE SELF 40, 47-50 (Diana Tietjens Meyers ed., 1997).

(25.) Gerald Dworkin, The Concept of Autonomy, in THE INNER CITADEL, supra note 4, at 54, 58. Dworkin adds that a proper view of autonomy should not conflict "with emotional ties to others, with commitments to causes, with authority, tradition, expertise, leadership, and so forth." Id. at 59.

(26.) See id. at 60.

(27.) Id.

(28.) Id. at 60.

(29.) Id. at 61.

(30.) See id. at 58.

(31.) See id. at 60.

(32.) See id. at 59.

(33.) Dworkin discusses this example as a paradigm introduced in the work of Harry Frankfurt. See id. at 60-61 (citing Harry G. Frankfurt, Freedom of the Will and the Concept of a Person, in THE INNER CITADEL, supra note 4, at 63).

(34.) In a question and answer session following the presentation of this paper at the symposium at William & Mary, Dworkin distinguished these types of inquiry as "empirical questions" that were not his primary focus. The relative marginality of such questions of application may be inferred even from a brief text such as The Concept of Autonomy. Although Dworkin takes pains, for example, to make clear that "procedural independence" is a required element of autonomy, he treats only in passing the thorny "empirical" question of whether any given actor has, in fact, come by his "own" motivational structure in a way that reflects freedom from strong external influences, and he appears to resolve the question by recourse to traditional categories such as deception or coercion. See Dworkin, supra note 25, at 59-60.

(35.) One might approach this difference in emphasis by reference to Richard Fallon's distinction between "descriptive" and "ascriptive" autonomy. See Richard H. Fallon, Jr., Two Senses of Autonomy, 46 STAN. L. REV. 875, 877-78 (1994). "Descriptive" autonomy "refers to people's actual condition and signifies the extent to which they are meaningfully self-governed in a universe shaped by causal forces." Id. at 877. "Ascriptive" autonomy "represents the purported metaphysical foundation of people's capacity and also their right to make and act on their own decisions." Id. at 878. My focus in this Essay, and in my critique of Dworkin and Feinberg, is clearly on descriptive autonomy. Dworkin's and Feinberg's work may contain strains of both but in their movement away from the actual empirical condition of different groups of human beings, they press toward the ascriptive.

(36.) See, e.g., Meyers, Personal Autonomy, supra note 1.

(37.) See id

(38.) See id.

(39.) See id. at 621-24.

(40.) Cf. THOMAS E. HILL, JR., Servility and Self-Respect, in AUTONOMY AND SELF-RESPECT 4, 4 (1991) (arguing that servility, such as that manifested by the traditional woman, displays a lack of self-respect).

(41.) See Meyers, Personal Autonomy, supra note 1, at 623-24.

(42.) See id.

(43.) See id. at 622-24.

(44.) See id. at 624.

(45.) See id. at 624-25.

(46.) See id. at 625.

(47.) See id. at 625-26. Meyers describes, as an example of partial access autonomy, a fundamentalist Christian wife and mother, who is not able to access those parts of herself that would permit her to oppose her childhood socialization, but is able to access those elements of unconditioned will that permit her to assert herself against a public school teacher who proposes to teach her children theories of evolution rather than creation. See id.

(48.) Id. at 627.

(49.) Id.

(50.) See id. at 627-28.

(51.) See id. at 619-28.

(52.) See Dworkin, supra note 25, at 54-62; Feinberg, supra note 4, at 27-49.

(53.) See Dworkin, supra note 25, at 54-62; Feinberg, supra note 4, at 27-49.

(54.) See Dworkin, supra note 25, at 54-62; Feinberg, supra note 4, at 27-49.

(55.) See Dworkin, supra note 25, at 54-62; Feinberg, supra note 4, at 27-49.

(56.) See Meyers, Personal Autonomy, supra note 1, at 619-20, 624-28.

(57.) See id. at 621-24.

(58.) The variety of feminist theorists who espouse this view do not necessarily agree with each other about its sources or implications for women's lives. Compare, e.g., MARTHA ALBERTSON FINEMAN, THE NEUTERED MOTHER, THE SEXUAL FAMILY AND OTHER TWENTIETH CENTURY TRAGEDIES (1995) (delineating a social constructivist account of women as caregivers) with Robin West, Jurisprudence and Gender, 55 U. CHI. L. REV. 1 (1988) (discussing a biologically-based account of women as interconnected with others).

(59.) See Meyers, Personal Autonomy, supra note 1, at 623-24.

(60.) See id. at 622.

(61.) See id. at 626-28.

(62.) See id. at 626.

(63.) See id. at 626-27.

(64.) See id. at 625-26.

(65.) See id. at 626-28.

(66.) See id. at 620-21.

(67.) See id. at 621-24.

(68.) See id. at 622. This critique applies not only to this Essay--which focuses in particular on women's subjection through their socialization to domesticity--but also to MEYERS, SELF, SOCIETY, AND PERSONAL CHOICE, supra note 1, her book-length exploration of the implications of socialization for liberal theories of autonomy. See generally id. at 141-55 (discussing de Beauvoir and female socialization).

(69.) See MEYERS, SELF, SOCIETY, AND PERSONAL CHOICE, supra note 1, at 141-55.

(70.) For examples of feminism grounded in a critique of androcentrism, see SANDRA LIPSITZ BEM, THE LENSES OF GENDER: TRANSFORMING THE DEBATE ON SEXUAL INEQUALITY (1993); Christine A. Littleton, Reconstructing Sexual Equality, 75 CAL. L. REV. 1279 (1987).

(71.) See generally BEM, supra note 70, at 40-42 (noting that "androcentrism is the privileging of male experience and the `otherizing' of female experience").

(72.) See CATHARINE A. MACKINNON, FEMINISM UNMODIFIED: DISCOURSES ON LIFE AND LAW (1987) [hereinafter MACKINNON, FEMINISM UNMODIFIED]; CATHARINE A. MACKINNON, TOWARD A FEMINIST THEORY OF THE STATE (1989) [hereinafter MACKINNON, TOWARD A FEMINIST THEORY].

(73.) For a fascinating discussion of the role of language in perpetuating women's sexual subordination, see Judith Butler, Contingent Foundations: Feminism and the Question of Postmodernism," in FEMINIST CONTENTIONS 35 (1995). Butler analyzes an expression employed by one of the defense attorneys in the New Bedford gang-rape trial, who asked what the complaining witness, a woman with a boyfriend at home, was doing "running around the streets getting raped." Id. at 52. The phrase, Butler argues, characterizes women as constrained and passive in relation to specific acts of sexual aggression ("getting raped"), but nonetheless responsible for bringing such acts on themselves through their irresponsible or provocative behavior ("running around getting raped"). See id. at 52-53.

(74.) This range of problems is introduced in CATHARINE A. MACKINNON, Difference and Dominance, in FEMINISM UNMODIFIED, supra note 72, at 32.

(75.) In Homelessness and the Issue of Freedom, Jeremy Waldron describes the ability to undertake particular actions in particular places as a form of freedom. See JEREMY WALDRON, Homelessness and the Issue of Freedom, in LIBERAL RIGHTS: COLLECTED PAPERS 1981-1991, at 309 (1993). It is not clear whether this constraint would, according to Dworkin's distinction, constitute simply a denial of freedom (the ability to act) or also a deprivation of autonomy (the ability to act on values that one identities as one's own). See Dworkin, supra note 25, at 59-61. Presumably, if the constraint made it difficult or impossible to implement values that were important to one's moral scheme, it might restrict autonomy as well.

(76.) CATHERINE A. MACKINNON, Difference and Dominance, in FEMINISM UNMODIFIED, supra note 72, at 39.

(77.) See Meyers, Personal Autonomy, supra note 1, at 628 (noting that many feminists "are justifiably suspicious of the professed fulfillment of many traditional women").

(78.) See, e.g., KATHRYN PYNE ADDELSON, Autonomy and Respect, in IMPURE THOUGHTS: ESSAYS ON PHILOSOPHY, FEMINISM & ETHICS 212, 218-19 (1991).

(79.) See MEYERS, SELF, SOCIETY, AND PERSONAL CHOICE, supra note 1, at 141-71.

(80.) For a fascinating description of sexual harassment as a disciplinary practice, see Katherine M. Franke, What's Wrong With Sexual Harassment?, 49 STAN. L. REV. 691 (1997).

(81.) See CATHARINE A. MACKINNON, Desire and Power, in FEMINISM UNMODIFIED, supra note 72, at 46, 55.

(82.) For a feminist philosophical account that makes this point about Meyers, although through a different analytic framework than those I refer to above, see ADDELSON, supra note 78, at 218-20.

(83.) See MEYERS, SELF, SOCIETY, AND PERSONAL CHOICE, supra note 1, at 156.

(84.) See ADDELSON, supra note 78, at 218-20.

(85.) See MEYERS, SELF, SOCIETY, AND PERSONAL CHOICE, supra note 1, at 135-40. For another liberal feminist account that stresses the participation of others and the influence of social context in the development of autonomy, see Marilyn Friedman, Self-Rule in a Social Context: Autonomy from a Feminist Perspective, in FREEDOM, EQUALITY, AND SOCIAL CHANGE 158 (Creighton Peden & James P. Sterba eds., 1989).

(86.) See MEYERS, SELF, SOCIETY, AND PERSONAL CHOICE, supra note 1, at 42-58 (elaborating on the skills that are required for the development of autonomy competency).

(87.) See, e.g., Meyers, Personal Autonomy, supra note 1, at 626 (providing the example of a traditional mother defending her creationist education of her children against evolution-wielding teachers in the public schools).

(88.) Cf. id. at 619 (noting the inadequacy in understanding personal autonomy simply as doing what one wants to do without undue interference).

(89.) Cf. id. at 628 (arguing that a woman's decision to embrace the traditional roles of housekeeping and parenting could be a completely autonomous choice rather than the result of an external source).

(90.) See id. at 626-28.

(91.) See ADDELSON, supra note 78, at 218-20 (discussing the "interactionist model" as a better way to investigate the relationship between the traditional woman and social structure than Meyers' socialization model).

(92.) See id.

(93.) Some liberal theorists of autonomy, including feminists such as Meyers, also describe a dimension of autonomy that is identified as "self-discovery." I will not use this term because it seems to connote the uncovering of a self that is somehow prior to social influence. I acknowledge, however, that even in the context of systematic social construction, there may be a process of becoming aware of what one experiences as characteristic of, or appropriate to, oneself. In the discussion below, I will discuss this as part of self-definition.

(94.) Cf. MEYERS, SELF, SOCIETY, AND PERSONAL CHOICE, supra note 1, at 43-45 (discussing self-definition as a necessary component of autonomy that is described as doing what one really wants to do).

(95.) Robert Young, Autonomy and the "Inner Self," in THE INNER CITADEL, supra note 4, at 77, 78.

(96.) See generally ADDELSON, supra note 78, at 220 (noting that "we cannot discuss personal autonomy without discussing the social and political structure").

(97.) See, e.g., id.

(98.) See MEYERS, SELF, SOCIETY, AND PERSONAL CHOICE, supra note 1, at 43, 45.

(99.) See id. at 27-28.

(100.) See Kathryn Abrams, Sex Wars Redux: Agency and Coercion in Feminist Legal Theory, 95 COLUM. L. REV. 304, 309 (1995) (noting that dominance feminists characterize women's sexuality primarily as a construct of oppressive forces).

(101.) Cf. id. at 354 (explaining the difficulty in resisting the forces that oppress women).

(102.) See MACKINNON, TOWARD A FEMINIST THEORY, supra note 72, at 91-92 (illustrating the effect of women's consciousness-raising groups in developing a knowledge that combats the experience of social inequality).

(103.) See id. at 83-84 (recognizing that feminist consciousness-raising involves collective social being as opposed to individual or subjective ideas).

(104.) For an interesting discussion of this dimension of consciousness raising, see id. at 85-105.

(105.) Collins argues that this effort also is born of necessity: "Unlike white women's images attached to the cult of true womanhood, the controlling images applied to Black women are so uniformly negative that they almost necessitate resistance if Black women are to have any positive self-images." PATRICIA HILL COLLINS, BLACK FEMINIST THOUGHT: KNOWLEDGE, CONSCIOUSNESS, AND THE POLITICS OF EMPOWERMENT 95 (1991).

(106.) See id.

(107.) For one version of Spillers's account, see Hortense J. Spillers, Intersticies: A Small Drama of Words, in PLEASURE AND DANGER: EXPLORING FEMALE, SEXUALITY 73 (Carole S. Vance ed., 2d ed. 1992).

(108.) Collins comments in particular on the demand for Respect articulated by Aretha Franklin, and the valuation of independence reflected in Billie Holiday's God Bless the Child Who's Got His Own. See COLLINS, supra note 105, at 108-10.

(109.) Katherine Franke and Susan Keller find this strategy in the work of the singer Madonna. See Katherine Franke, Cunning Stunts: From Hegemony to Desire A Review of Madonna's Sex, 20 N.Y.U. REV. L. & SOC. CHANGE 549, 552 (1993-94); Susan Etta Keller, Review Essay, 18 W. ST. U. L. REV. 463 (1990) (reviewing JUSTIFY MY LOVE (Warner Reprise Video 1990) and ANDREW ROSS, NO RESPECT: INTELLECTUALS AND POPULAR CULTURE (1989)).

(110.) See COLLINS, supra note 105, at 88-89.

(111.) See id.

(112.) As Collins points out, histories or images of women's solidarity with each other have helped women to situate themselves within sub-communities whose norms fuel more positive self-conceptions. Collins describes the bonds between women articulated in fiction such as Toni Morrison's Sula. See id. at 97. Similarly, Adrienne Rich evokes the "lesbian continuum"--a term designating the variety of ways that women have affiliated with and supported each other--in response to what she views as the socially dominant imagery, which compels both heterosexuality and women's subordination to men. See ADRIENNE CECILE RICH, Compulsory Heterosexuality and Lesbian Existence, in BLOOD, BREAD, AND POETRY: SELECTED PROSE 1979-1985, at 23, 51 (1986). These histories and images of women's solidarity also have served as another source of dissonant imagery, as they have located women within social relations that reflect reciprocity and support, rather than consistent hierarchy and subordination.

(113.) See COLLINS, supra note 105, at 96-99, 103-04.

(114.) See generally MACKINNON, TOWARD A FEMINIST THEORY, supra note 72, at 46-47 (discussing "voluntarism" notions of liberalism).

(115.) See Meyers, Personal Autonomy, supra note 1, at 626-27.

(116.) Cf. id. at 627-28 (illustrating that a housewife who is internalized with a certain concept of womanliness will find it difficult to reconcile autonomous life plans).

(117.) See, e.g., id. at 621-22 (arguing that early childhood socialization confines women to certain roles).

(118.) See generally MACKINNON, TOWARD A FEMINIST THEORY, supra note 72, at 94-95 (noting how women have been subordinated in society).

(119.) It is not merely biological femaleness, but also female gender (i.e., the range of social characteristics associated with femaleness) that may be penalized or stigmatized under the system of gender inequality described above. For a useful account of how sexual harassment may penalize men who display socially female (or otherwise non-masculine) characteristics, see Franke, supra note 80, at 737-40.

(120.) See Meyers, Personal Autonomy, supra note 1, at 620-24.

(121.) See MACKINNON, TOWARD A FEMINIST THEORY, supra note 72, at 84-91.

(122.) See id. at 86-87 (examining the need for women to have collective groups apart from men to raise self-consciousness without the pitfall of retreating into a socialized role).

(123.) See infra notes 142-63 and accompanying text.

(124.) Kirstie McClure, On the Subject of Rights: Pluralism, Plurality and Political Identity, in DIMENSIONS OF RADICAL DEMOCRACY: PLURALISM, CITIZENSHIP, COMMUNITY 108, 123 (Chantal Mouffe ed., 1992) (referring to "the diffusion of political sites across the surface of the social itself").

(125.) See id.

(126.) See generally Louise F. Fitzgerald et al., Why Didn't She Just Report Him? The Psychological and Legal Implications of Women's Responses to Sexual Harassment, 51 J. SOC. ISSUES 117, 118-21 (1995) (discussing the different ways women respond to sexual harassment and their reasons for doing so).

(127.) See id. at 121.

(128.) See id. at 119-21.

(129.) See id. at 119-20.

(130.) Susan Estrich has critiqued this prevalent understanding of the nature of resistance as reflecting the norms of a schoolyard fight. See Susan Estrich, Rape, 95 YALE L.J. 1087, 1105 (1986).

(131.) See generally Fitzgerald et al., supra note 126, at 119-20 (including non-confrontational respondents among the 26% that "did nothing")

(132.) See id. at 122-23 (explaining that many women do not report harassment out of fear of retaliation).

(133.) See Martha R. Mahoney, Legal Images of Battered Women: Redefining the Issue of Separation, 90 MICH. L. REV. 1, 18 (1991). Mahoney argues that this sometimes is attributed to the framing of the defense itself and sometimes attributable to the ways in which it is interpreted by legal decision-makers. See id. at 40-41.

(134.) See id. at 24-34.

(135.) See id. at 19-24.

(136.) Mahoney argues that children have an equivocal effect on women's response in the context of battering. They may make women more determined to undertake the act of self-assertion needed to separate. See id. at 66 (offering a narrative of a wornan who finally left when she believed her partner would harm their baby). They also may make women more reluctant to leave. See id. at 21 (offering a narrative of a woman reluctant to leave because her abuser was, or had been, a good father to their children). Children also may make them more cautious in planning for separation. See id. at 19-24.

(137.) See id. at 21-23, 75-76 (noting examples of abused women who turned to support groups, family, and the courts to manage the abuse).

(138.) See id. at 61-68.

(139.) See, e.g., id. at 24 (noting that the circumstances particular to each woman's life will affect the choices she makes).

(140.) Cf. Meyers, Personal Autonomy, supra note 1, at 625-26 (observing, by way of example, that autonomy must be gauged along numerous dimensions and may vary in different contexts).

(141.) An interesting example of this interaction of external influences and the development of one's self conception is offered in Vicki Schultz, Women "Before" the Law: Judicial Stories About Women, Work, and Sex Segregation on the Job, in FEMINISTS THEORIZE THE POLITICAL 297 (Judith Butler & Joan W. Scott eds., 1992). Schultz examines the professional paths of women who have left more traditionally female categories of employment to enter blue collar trades, and observes that this movement is often conditioned importantly by the attitudes of management toward women in nontraditional work and the initiatives employers undertake to facilitate the transition. See id. at 316-17.

(142.) See CATHARINE MACKINNON, Linda's Life and Andrea's Work, in FEMINISM UNMODIFIED, supra note 72, at 127-33.

(143.) See id.

(144.) See id. at 131-33.

(145.) See id. at 130.

(146.) See id. at 131-32.

(147.) See id. at 127-33. Indeed, the most problematic part of MacKinnon's essay is her inability, within a theory that depicts women as systematically dominated, to explain how these dissonant and demanding voices have emerged. It is noteworthy that MacKinnon does not attempt such an explanation--indeed, to my mind, it would be quite difficult--but simply presents her audience with the apparent contradiction as a kind of tribute to Dworkin and Marchiano. See id. at 131-33. Given a slightly different theory--one that describes the forces producing women's oppression as less unitary, more variable in character, and more contingent in their effects on any particular woman -- such an explanation would not be difficult to frame. I offer a number of such explanations in the next paragraph.

(148.) Cf. id. at 131 (commenting that the work of the three women--speaking out against pornography--cannot be explained as anything other than amazing, given the system of pornography).

(149.) See, e.g., TONI GILPIN ET AL., ON STRIKE FOR RESPECT: THE CLERICAL AND TECHNICAL WORKERS' STRIKE AT YALE UNIVERSITY 1984-85 18-32 (1995).

(150.) See Carla Hall, Exotic Dancers Stage Union Drive Labor, L.A. TIMES, Aug. 29, 1996, at A3, available in 1996 WL 11639198.

(151.) See Laura Lederer, Introduction to TAKE BACK THE NIGHT: WOMEN ON PORNOGRAPHY 15, 19 (Laura Lederer ed., 1980).

(152.) See Frances Elisabeth Olsen, Feminism in Central and Eastern Europe: Risks and Possibilities of American Engagement, 106 YALE L.J. 2215, 2255 n.229 (1997) (describing the informal curfew placed on women by danger and fear); see also TAKE BACK THE NIGHT: WOMEN ON PORNOGRAPHY (Laura Lederer ed., 1980) (including a number of essays written in the fight against pornography).

(153.) See, e.g., Keller, supra note 109, at 469-70.

(154.) See Sharon Marcus, Fighting Bodies, Fighting Words: A Theory and Politics of Rape Prevention, in FEMINISTS THEORIZE THE POLITICAL, 385, 396-97 (Judith Butler & Joan W. Scott eds., 1992).

(155.) See Franke, supra note 109, at 551-53 (providing the basis for concluding that there is political significance in Madonna's assertion that she is in control, quite apart from the substantive nature of her sexual representations).

(156.) Thus, women's sexual self-assertion could, in theory, be a remedy for women's sexual oppression as it is understood either by dominance feminists or by pro-sex (e.g., sex radical) feminists. For a comprehensive examination of these two feminist approaches to women's sexuality, see Abrams, supra note 100, at 304. Such disruptive resistance, however, is not endorsed in the leading dominance-based account, that of Catharine MacKinnon. This is because, as a structuralist account, which describes all social and institutional influences as pressing in the single direction of sexualized domination, and as an account which reflects a more traditionally confined understanding of the political realm (i.e., the preferred remedy is an appeal to law), MacKinnon's theory provides neither a reliable explanation for how such dissonant sexual self-presentation could emerge, nor any ground for believing that individual action could make much difference in women's condition.

(157.) See Keller, supra note 109, at 468; Carlin Meyer, Sex, Sin, and Women's Liberation: Against Porn-Suppression, 72 TEX. L. REV. 1097, 1151-52 (1994).

(158.) See DUNCAN KENNEDY, SEXY DRESSING ETC. 167-70 (1993).

(159.) See Keller, supra note 109, at 463-66.

(160.) See generally DOROTHY ALLISON, SKIN (1994) (critiquing cliched aspects of most fictional literature depicting sexuality, and describing effort to write about sex in ways that it had not been written about before).

(161.) Cf. JANE J. MANSBRIDGE, WHY WE LOST THE ERA 129-33 (1986) (describing the benefits and drawbacks of individual action in the context of the women's movement behind the ERA).

(162.) See MACKINNON, TOWARD A FEMINIST THEORY, supra note 72, at 83-105.

(163.) See Julia O'Malley, You Go, Grrl: Philosophy Kicks, ANCHORAGE DAILY NEWS, Mar. 1, 1996, at Fl, available in 1996 WL 11930642.

(164.) See ANDREA DWORKIN & CATHARINE MAcKINNON, PORNOGRAPHY AND CIVIL RIGHTS: A NEW DAY FOR WOMEN'S EQUALITY 33-34 (1988).

(165.) See, e.g., Brief Amicus Curiae of Feminist Anti-Censorship Task Force et al., American Booksellers v. Hudnut, 771 F.2d 323 (7th Cir. 1985) (No. 8403147), reprinted in 21 U. MICH. J.L. REFORM 76 (1987-88).

(166.) See Susan Etta Keller, Viewing and Doing: Complicating Pornography's Meaning, 81 GEO. L.J. 2195, 2237-38, 2239-42 (1993).

(167.) See Abrams, supra note 100, at 339 n.136.

(168.) See Carlin Meyer, Media, Markets and Women's Liberation: Enlarging the Feminist Agenda (1998) (unpublished manuscript, on file with author).

(169.) For a discussion of the distinction between prescriptive or prohibitory, and constitutive roles of law, see Kathryn Abrams, The Constitution of Women, 48 ALA. L. REV. 861, 861-64 (1997).

(170.) I have discussed this premise and the dichotomous interpretations that emerge from it in previous works. See Abrams, supra note 100; Kathryn Abrams, Songs of Innocence and Experience: Dominance Feminism in the University, 103 YALE L.J. 1533, 1552-60 (1994) (reviewing KATIE ROIPHE, THE MORNING AFTER: SEX, FEAR AND FEMINISM ON CAMPUS (1993)).

(171.) Much interesting work on this requirement, both in the context of rape and in the context of sexual harassment, has been done by Susan Estrich. See Estrich, supra note 130, at 1121-22; Susan Estrich, Sex at Work, 43 STAN. L. REV. 813, 826-33 (1991).

(172.) See Chamberlin v. 101 Realty, 915 F.2d 777, 783-84 (1st Cir 1990); Stockett v. Tolin, 791 F. Supp. 1536, 1543 (S.D. Fla. 1992).

(173.) For a discussion of the difficulties that courts have experienced in accommodating complexity in the characterization of human subjects, see Kathryn Abrams, Title VII and the Complex Female Subject, 92 MICH. L. REV. 247c, (1994).

(174.) For a discussion of the tension between this premise an] some efforts to shape defenses based on strong social construction in the criminal law, see Anne M. Coughlin, Excusing Women, 82 CAL. L. REV. 1 (1994).

(175.) See, e.g., Kathryn Abrams, Equality and Impasse: Mobilizing Group-Based Perspectives in an Era of Group-Blindness, in REDEFINING EQUALITY 13, 13-28 (Neal Devins & Davison M. Douglas eds., 1998).

(176.) The Title VII claim for sexual harassment is, perhaps not surprisingly, one of the doctrinal areas in which courts have most clearly acknowledged the effects of social influence on human subjects. Perhaps the best known example is the debate over whether the "reasonable person" or the "reasonable woman" provides the proper frame of reference from which to assess the pervasiveness of alleged sexual harassment. Compare Ellison v. Brady, 924 F.2d 872, 879-81 (9th Cir. 1991) (endorsing "reasonable woman" standard), with Oncale v. Sundowner Offshore Servs., Inc., 118 S. Ct. 998, 1003 (1998) (endorsing "reasonable person in the plaintiff's position, considering `all the circumstances'" (quoting Harris v. Forklift Sys., Inc., 510 U.S. 17, 23 (1993))). Yet there are other examples, such as a recent case in which the U.S. Court of Appeals for the Ninth Circuit acknowledged that exposure to systematic sexual harassment in the workplace may have an impact on the relationship between a female sexual harassment claimant and her male lawyer. See Stoll v. Runyon, No. 97-17398, 1999 WL 12998, at *5 (9th Cir. Jan. 15, 1999).

However, acknowledgment of the effects of social context on individual self-definition and self-direction is not limited to the sexual harassment area. The ongoing tobacco litigation, in which individual claimants and states have sought to hold tobacco companies liable for the injuries sustained by longtime smokers, provides an example of an entirely distinct context in which courts have been asked to consider the possibility of systematic social influence--from withholding of scientific evidence to misleading advertising--on a decision our legal system, and our society, historically have been wont to see as the product of "individual choice."

(177.) This response may well have been the basis of the "serious psychological injury" requirement imposed by many federal courts of appeals in sexual harassment cases. It took a trenchant, majority opinion by Justice O'Connor in Harris v. Forklift Sys., Inc., 510 U.S. 17 (1993), arguing that a plaintiff need not reach the point of having a "nervous breakdown" before she was entitled to claim a Title VII violation, see id. at 22, before the law consistently reflected the fact that a woman could be resilient in the face of sexualized conduct, and still have suffered an actionable injury.

KATHRYN ABRAMS, Professor of Law and Acting Director of Women's Studies, Cornell University. My thanks to Steve Garvey and Steve Shiffrin for their helpful thoughts on an earlier draft. I am also grateful to participants in the "Reconstructing Liberalism" conference, to faculty workshop participants at the DePaul University College of Law and the University of Michigan Law School for their comments, and to Stephanie Sechler and Ali Nathan for exceptional research assistance.
COPYRIGHT 1999 College of William and Mary, Marshall Wythe School of Law
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1999 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Abrams, Kathryn
Publication:William and Mary Law Review
Geographic Code:1USA
Date:Mar 1, 1999
Words:15955
Previous Article:Reflections on coercing privacy.
Next Article:Autonomy and agency.
Topics:

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters