Some of the most interesting and important work on freedom in recent years has been produced by feminist theorists (e.g., Friedman Reference Friedman2003; Hirschmann Reference Hirschmann2003; Nedelsky Reference Nedelsky2011; Zerilli Reference Zerilli2005; and see Anker, forthcoming). Feminist theory has also generated important insights about the character and conditions of human agency, the intersectional nature of human identities, and the dynamics of political action that richly enhance our understanding of freedom (Ackerly and Attanasi Reference Ackerly and Attanasi2009; Alcoff Reference Alcoff2006; Anzaldua Reference Anzaldua2007; Bartky Reference Bartky1990; Beltrán Reference Beltrán2010; Benhabib Reference Benhabib1992; Brown Reference Brown1995; Butler Reference Butler2004, Reference Butler2006; Crenshaw Reference Crenshaw1989; Hill-Collins Reference Hill-Collins2005; Love Reference Love2007; Mohanty Reference Mohanty2003; Young Reference Young2005). Yet feminist perspectives on freedom—like others in political theory today (e.g., Berlin Reference Berlin1990; Pettit Reference Pettit2000; Villa Reference Villa2008)—are limited by their tendency to conceive freedom in a monistic way. Whether freedom means the exercise of personal choice (Hirschmann Reference Hirschmann2003), or inaugural acts of world making (Arendt Reference Arendt1958; Zerilli Reference Zerilli2005), or noninterference (Berlin Reference Berlin1990), or nondomination (Pettit Reference Pettit2000), or something else, theorists generally describe freedom as being just one thing. Yet the experiences that we have of freedom come in different forms, and sometimes the different kinds of freedom that we have, or that we seek, conflict with one another. Nor is it obvious that these conflicts could ever be fully reconciled, even under ideal conditions. Our freedom today is complex: We are at once free and unfree, in ways that are multiple and often conflicting.

This essay brings together feminist insights about the nonsovereign character of human agency and the intersectional quality of human identities in order to sketch a new, pluralist approach to freedom. A plural view of freedom helps make sense of the complex, frequently contradictory, experiences of women and other marginalized groups in ostensibly free, democratic societies today, where the conditions that support freedom in some forms coexist with deep and entrenched obstacles to freedom in other forms. A plural approach illuminates the subtle, informal dynamics that compromise freedom in ways that are systematic and often invisible, but it also recognizes that freedom sometimes survives the conditions that compromise it. To theorize freedom in plural terms cuts against conventional narratives about freedom in political theory, feminist and otherwise. But doing so is the only way to make sense of the complexity of freedom today—and the only way to make real progress in advancing it for women and other marginalized groups.

Although agency has traditionally been associated with rational autonomy and located in the exercise of a sovereign will, thus equating agency with personal choice or control over one's action, recent work by feminist theorists has disputed this view. Agency is neither reducible to the experience of control nor located exclusively within the individual; it is a potent but nonsovereign experience, to use the Arendtian language that theorists such as Zerilli have adopted (Arendt Reference Arendt1958, 234–36; Zerilli Reference Zerilli2005, 19–20).Footnote ¹ My own view of nonsovereign agency follows Arendt and Zerilli in distinguishing agency from control and in treating agency as a socially distributed phenomenon, rather than an exclusively internal capacity of the individual. Yet my view makes individual identity far more central to agency than either Arendt or Zerilli would allow (Krause Reference Krause2011).

As I use the term, agency refers to the affirmation of one's subjective existence, or identity, through concrete action in the world. To be an agent is to affect the world in ways that concretely manifest who you are, to see yourself and be seen by others in the effects you have, to recognize your deeds as being in some sense your own. Understood this way, agency arises at the interface of identity and efficacy. Without a reflexive sense of self that is antecedent to any given action, one cannot have the experience of seeing oneself in one's deed. Hence there is no agency without identity. By the same token, there can be no agency without effects. Agency, in contrast to mere willing (or dreaming), involves actually having an impact on the world. And many of the effects that we have are themselves affected by the social interpretations given to our actions by others. What we can accomplish in any particular instance depends partly on what other people think we are doing, and on the background of social meaning against which their interpretations arise; individual agency depends on social uptake. Thus, agency is at once a subjective and an intersubjective phenomenon; it emerges out of the communicative exchanges, background meanings, social interpretations, personal intentions, self-understandings, and bodily encounters through which one's identity is manifest in one's deeds (Krause Reference Krause2011).

Insofar as agency is socially distributed in this way, it will be subject to the same dynamics of intersectionality that feminists have shown to shape individual identities (Ackerly and Attanasi Reference Ackerly and Attanasi2009; Alcoff Reference Alcoff2006; Anzaldua Reference Anzaldua2007; Beltrán Reference Beltrán2010; Crenshaw Reference Crenshaw1989; Hill-Collins Reference Hill-Collins2005; Love Reference Love2007; Mohanty Reference Mohanty2003). How others interpret our actions, and hence the effects we have on the world, will vary with the different degrees and kinds of social uptake that are available to us, which themselves will vary with the multiple, intersecting strands of our identities. Systematic social inequalities powerfully affect social uptake, disabling individual agency and compromising freedom in patterned ways for members of marginalized groups. Yet because identities are intersectional, any one individual's agency will be subject to multiple dynamics of uptake, vulnerable to diverse failures but also potent in sometimes unexpected (and sometimes undesirable) ways. Each of us will be both free and unfree as our agency is considered along multiple dimensions, or seen from the standpoints of different aspects of our identities with the different trajectories of uptake they make possible or disallow, given the prevailing background of social norms and meanings.

Moreover, we will be free and unfree in qualitatively different kinds of ways. Freedom in the most general sense is the collection of background conditions and relations that make it possible for individuals to exercise agency, or that support the affirmation of our subjective existences through concrete action in the world. What the nonsovereign, intersectional view of agency suggests is that these conditions are multiple and complex. Hence, no one account of freedom will be adequate. For instance, because agency is a socially distributed phenomenon and not simply an interior function of the individual will, the “negative liberty” model of freedom as noninterference will be incomplete—as feminists have long understood. Freedom is not something that the individual can do by herself, at least not so long as she is living in society with others. It is true that interference by the state and by other individuals and groups can indeed impede agency and hamper freedom. Negative liberty is therefore a valuable principle for ordering social and political life, but only up to a point. We cannot achieve freedom by simply leaving people alone to do their own things.

Nor is the neorepublican ideal of nondomination sufficient. In contrast to noninterference, the principle of nondomination protects us not just against actual interference but also against relationships in which certain people have the power to arbitrarily interfere, even if they do not actually exercise this power (Pettit Reference Pettit2000). Domination certainly does obstruct the affirmation of one's subjective existence because under conditions of domination, one's action is made to affirm someone else's subjective existence rather than one's own. The paradigm case of domination, after all, is slavery, in which one person functions as the tool of another's will. Yet the paradigm of slavery does not capture all the ways in which we can be unfree. The activists who rioted outside the Stonewall Inn in 1969, for instance, were not resisting anything quite like slavery. Gay people have been shamed, discriminated against, and physically attacked, but our position in society has never been reducible to the position of a slave relative to his or her master.

The unfreedom of gay people is not captured by the logic of domination. It is perhaps better expressed in the language of oppression, where the impediments to agency are more like bars of imprisonment than chains of slavery. In contrast to domination, which can be exercised by one individual upon another, oppression is a collective phenomenon and applies to persons only in view of their membership in a particular group. And whereas domination requires intentionality and depends on one person's ability to control another, oppression rests on impersonal, systematic patterns of privilege and prejudice that no one really controls and that are often unintentional. Rather than exploiting some persons for the purposes of others, oppression works through social norms and internalized habits to confine individuals within boundaries that suffocate and disfigure, systematically undercutting their ability to be in and affect the world in ways that manifest their subjective existence. Liberation from these bonds is an important aspect of coming out, which brings the freedom to affirm one's full subjectivity in the world, rather than just a fragment of it. Freedom as nonoppression has long been a part of the political aspirations of marginalized groups.

Freedom as nonoppression also is incomplete, however. Sometimes the affirmation of one's subjective existence through concrete action involves changing the world and transforming the self in much the same ways that Arendt (and with her Zerilli) describes (Arendt Reference Arendt1958; Zerilli Reference Zerilli2005). Remaking the world together with others is a powerful way of experiencing freedom, and the Arendtians are right to remind us that in this process, we ourselves are often changed (Zerilli Reference Zerilli2005, 25, 65, 71, 98). Yet agency is not always and only about novelty or about remaking the world. Sometimes agency is just about being in the world as one is—and as the world is. To equate freedom exclusively with inaugural world making or the transformation of the self is to marginalize many actual exertions of human agency and lived experiences of freedom.

So none of these familiar ideals captures the whole of what freedom means or has meant to people who seek it. Indeed, no one conception of freedom can capture the whole of it. When we understand human agency in all its nonsovereign, intersectional complexity, we can see that there are different ways to be free, and that the different ways in which we can be free may not always coexist easily with one another. Freedom as nonoppression, for instance, may require the fostering of social solidarities and mutual recognition in ways that conflict with freedom as noninterference.

Likewise, one way to make sense of the debate about women who take the veil is to recognize that the issue involves—for the individual woman herself—different forms of freedom in conflict with one another. For some women, taking the veil is an act of resistance to Western imperialism (Hirschmann Reference Hirschmann2003). As such, it represents an instance of freedom as world making, in which women stand up against the forces of political aggression, economic exploitation, and liberal secularism. Working together with others in their communities, they mean to inaugurate a new way of life. Wearing the veil is a mark of their participation in this collective activity of world making; it affirms their agency. And even when women choose to take the veil for reasons that have nothing to do with freedom as world making—because they personally wish to live a pious life, or to honor the traditions of their native land—to affirm their right to do so is to protect freedom as noninterference. If we care about freedom in this form, we should respect their choices. Yet from the standpoint of freedom as nondomination, taking the veil symbolizes women's subordination to the authority of a patriarchal order in which they lack an equal share in power and can be instrumentalized for the purposes of others. And to the extent that the veil embodies women's systematically secondary social status and their confinement, it is also at odds with freedom as nonoppression. The question, then, is not whether women who take the veil are free but, rather, which forms of freedom are instantiated by taking the veil and which ones are compromised.

The veiling debate is difficult to resolve without remainder because the debate is not simply a conflict between freedom and other values, in which those who are committed to freedom have an obvious line to defend. Instead, it is (at least in part) a conflict between different kinds of freedom, none of them reducible to any other, and all of them valuable. The plurality of freedom means that at least some of these conflicts will admit no perfect resolution. Understanding freedom's plural character can help protect us from making policy decisions that blindly sacrifice some freedoms to others. The pluralist view also guards against the complacency that so often accompanies life in ostensibly free societies such as our own, which regularly let the achievement of some kinds of freedom obscure from view the dearth of freedom in other forms. And it can inhibit the tendency to think that having perfected the art of freedom, we are entitled to impose it on others around the world. Finally, by emphasizing the multiple experiences that constitute freedom and the various conditions that sustain it, a pluralist view encourages us to be open to new articulations of freedom. It suggests that the meaning of freedom is open-ended because responsive to changes in the human condition and in our mutual understanding of one another's lives, and hence that it cannot be settled once and for all.

So, feminist perspectives on freedom would do well to start from a nonsovereign, intersectional model of human agency. If we want to understand freedom at the margins of democratic societies, then we need to grasp the ways that agency is both individuated, or tied to individual identity, and socially distributed. And we need to appreciate the diverse, intersecting effects on agency that our multidimensional identities have. We will also need to come to terms with the plural character of freedom, the fact that freedom is not one thing but has many faces, which means that serving freedom may sometimes require conflicting things of us. And this means that freedom is rarely an all-or-nothing experience—we are more and less free, free and unfree along multiple dimensions—which suggests that freedom can never be perfectly complete. Still, by reorienting our theories of freedom in nonsovereign, pluralist directions, we are better equipped to understand and overcome the many obstacles to freedom that still plague women and other marginalized groups in so many places today, including ostensibly free, democratic societies like our own.