The Perils of Individualism

The modern concept of the individual as we know it is largely a product of the social developments and resulting intellectual transformations of the Enlightenment. One commonly associates the Enlightenment with the ideas of Descartes, Newton, Leibniz, de Condorcet, Berkeley, Hume, Montesquieu, Locke, Spinoza, Rousseau, and Kant. But, it is also a period of scientific discovery, economic expansion, increasing urbanization, rising population, improving communications, and striking political change extending from the Glorious Revolution of 1688 in England to the American and French Revolutions of the late eighteenth century (Taylor 1989). These changes encompass the emergence of what Jürgen Habermas (1989) famously identifies as a genuine public sphere. This period also sees the rather abrupt and dramatic birth in the 1760s of a recognizably modern consumer society (Winter 2002). The resultant conception of the individual is as a rational, self‐directing agent. In science, it is represented as the knowing subject who masters the world through reason. In politics, it is the free, sovereign individual in the “state of nature” who comes together with others to form a Social Compact. In economics, it is the choosing subject who acts rationally to maximize his or her self‐interest.

Today, it is increasingly apparent that the modern conception of the individual as a rational, self‐directing agent operates as a mythology that organizes and distorts science, economics, politics, and religion. In science, we celebrate the heroic work of the individual “genius” that pushes the boundaries of knowledge—often conferring on such “stars” outsized rewards, both social and economic. But scientific discovery is actually a process that unfolds in a community. It is the prior work of the field that lays the groundwork (whether positive or negative) for any particular breakthrough. And the significance of a discovery or invention often depends on what others in the field have found or seen—each innovation altering the possibilities inherent in all related innovations (Root‐Bernstein 1989). Sometimes, significant inventions change the field in ways that remain beyond the inventor's own comprehension. Joseph Priestly was the first to isolate oxygen. But, it was Antoine Laurent Lavoisier who showed that Priestly had actually discovered oxygen, and not “dephlogisticated air” as Priestly claimed to his dying day (Kuhn 1970). Similarly, Max Planck employed the idea of quantizing the electromagnetic energy to develop what we now call quantum theory, but he explicitly denied that the energy in atoms was quantized into discrete packets. It was Einstein who insisted that quantization was more than a mathematical formalism and who drew out some of its implications (leading to the photoelectric cell familiar in many electronic sensing devices) (Kuhn 1978).

In economics, the belief in the efficiency and reliability of markets that was presumed to follow from the rational actor model has been devastated by the spectacular financial meltdown of 2008. At the highest levels of government, it was assumed as an article of faith that rational market actors would not extend risky subprime loans or invest in exotic (later toxic) derivatives unless it were in their long‐term interests. The determination, which followed ineluctably from the economic model, was that “the self‐interest of lending institutions” would lead to decisions that “protect shareholders’ equity.” When it did not turn out that way, no less a figure than Alan Greenspan was forced publicly to admit that he had “found a flaw” and was “in a state of shocked disbelief.” As he conceded before Congress: “The whole intellectual edifice collapsed” (Andrews 2008).

In politics, the idea of individual autonomy has been deployed by the Right to undermine the fundamental idea of the State as an institution of collective self‐governance. This is evident in the health care debate where opponents of change quickly and effectively mobilize the argument that any national health care program will deprive Americans of “choice.” More perniciously, it appears in the arguments that portray government as an alien entity rather than as an agent of our collective will and characterize taxes as a kind of “theft” rather than the contribution that citizens make to provide for their general welfare.

Commodification and Social Fragmentation

In religion, the myth of the autonomous individual has worked a similarly dramatic transformation. As Vincent Jude Miller (2003) argues in his thoughtful book, Consuming Religion, religion in contemporary American society has largely been commodified. No longer understood in terms of a community of believers, religion is increasingly seen as an array of beliefs and practices from which one picks and chooses. Americans now freely mix and match spiritual ideas; it is more and more common to find Catholics who believe in reincarnation or who idolize the late Pope but reject almost every aspect of Church doctrine with which he was identified. At a recent conference I attended, one of the panelists actually introduced herself as a “Quathlic.” It is now common to hear parents who conceive their role as one of exposing their children to various religions’ forms of religious life so that they can make a more informed choice. Popular books on the afterlife offer images of heaven, sans harps and fluffy clouds, as a pleasant, affluent neighborhood populated by family and friends enjoying all the familiar accoutrements of modern‐day consumer life (Roof 2001). Commodification is evident, as well, in the deliberate cultivation and use of media celebrity by such figures as John Paul II, Mother Teresa, and the Dali Lama (Miller 2003).

Because it constitutes the unspoken backdrop of the argument that follows, I want to take a moment to say something about what I mean by consumerism. Consumer culture is something more and deeper than the materialist values and shopping habits of contemporary Western society. Consumerism is a form of life with a particular ontology that functions (as an ontology must function) to enable some ways of being and inhibit or preclude others. As an ontology, it reconfigures within its own logic whatever set of practices and values it comes in contact with.

What is that logic? The essential element of consumerism is the commodity form. The commodity presents itself to us abstracted from its context. This is true in two senses. First, the commodity is presented to me without any indication of its mode or conditions of production: When I select a head of lettuce at my local supermarket, I am blissfully unaware of the misery of the largely migrant workers in California or Florida, many of whom may be illegal immigrants without the full protection of the law working under exploitative conditions, who toiled in the hot sun to pick the lettuce that will ultimately appear in my salad. The lettuce presents itself to me as fresh and green, without any indication of its origin. Second, the commodity presents itself to me as an item of intrinsic value—a high‐definition television, a beautiful sports car—abstracted from the elaborate social system of exchange and interaction (markets, monetization, broadcast networks, the highway system, etc.) that in fact determines its value. The ontology of commodification, in other words, fundamentally depends on processes of abstraction and atomization.

It is this abstracted and atomized form of engagement that is everywhere reproduced in consumer culture. This ontology is embedded directly in product design, reproducing itself in social fragmentation. Consider the suburban single‐family home. Instead of living among extended family, the modern nuclear family lives in a free‐standing dwelling surrounded by similar such dwellings all occupied by two and, usually, only two generations of people from essentially the same socioeconomic class (and, all too often, race). There are no poor people in this neighborhood (nor, usually, the very wealthy), few single people, and but a few older couples. No longer are elderly parents likely to be taken in and cared for; they are outsourced first to retirement communities and, later, to nursing homes. Children grow up with little of the day‐to‐day intergenerational contact and instruction that once maintained religious, ethnic, and other traditions. Indeed, American homes have been getting larger and larger over the past three decades so that children increasingly have their own rooms, often with their own televisions or play consoles, to which they can retire. The sit‐down family dinner is now a rarity in suburban American homes, having been supplanted by individual, microwaved dinners, and pizza delivery. What is practiced here is a mode of engagement with the world in which choice among prepackaged commodities—television programs, computer games, frozen dinners, and other fast‐food products—rather than interaction with others is the central mode of engagement with the world. What is produced in this social matrix is the very antithesis of a democratic politics.

The Concept of Democracy

What is it that we speak of when we criticize institutional arrangements (e.g., judicial review or single‐member, winner‐take‐all election districts) or the implications of a social practice (e.g., consumerism) as insufficiently democratic? The concept of democracy is both obscure—an essentially contested concept (Gallie 1956)—and, at the same time, something about which we have robust, often confident intuitions.

At a minimum, democracy signifies popular sovereignty and majority rule—that is, government by and for the people. In an article entitled Democracy and Equality, Robert Post (2005, 143) argues that it is not sufficient to identify democracy with popular sovereignty and majority rule because the former “is a normative idea,” while the latter “are descriptive terms that refer to particular decision making procedures.” ¹ It is perfectly possible, he points out, for “the People” to opt for a tyrannical form of government or a majority of the electorate to vote for particular antidemocratic measures. Thus, he concludes, democracy must be defined in terms of autonomy understood in the Kantian sense of being governed only by those rules one gives oneself. A democratic society, then, is one that engages in collective self‐governance. ²

The conventional understanding of democracy, in other words, depends upon a conceptual vocabulary of self, agency, and autonomy. The problems with this conventional understanding are first, that it relies on a set of foundational assumptions that are theoretically untenable; second, that it leads to normatively undesirable conclusions; and, third, that it is pragmatically inadequate, indeed, counterproductive (as evidenced by our attempts at transporting democracy to cultures very different than our own). Both to understand and exercise democracy, we will have to take a very different theoretical tack and ask what kinds of practices and institutions would be necessary to construct ourselves in its image. The strategy, in other words, is to take the insight about social construction seriously—if not, indeed, foundationally—and explore who we would have to become and how we might get there if we would live together democratically.

Collective Self‐Governance: Institutional and Individualist Difficulties

Much of the debate about democratic self‐governance concerns how collective decision making can best be expressed or how it should be mediated to protect other norms. The more profound problem, however, is what Frank Michelman (1999, 15) calls “the institutional difficulty.” How can we reconcile the ideal of collective self‐governance with the practical reality of dissensus? For, unless a system requires unanimity, ³ the dissenting minority will—almost by definition—experience the rule of the winning majority as something other than “self” governance. I say “almost by definition” because, on one hand, to accede to the majority's rule is to comply with a rule other than the one the dissenters would give to themselves but, on the other, under the right conditions of identification the dissenters might nonetheless accept an otherwise disagreeable rule as their own.

The most common strategy for bridging this gap and resolving the institutional difficulty is to posit conditions of participation sufficiently fair, inclusive, and effective to induce the minority to identify the result as their own. In its strongest form, this would occur, Michelman (1988, 1526–27) explains, when the terms of discourse are sufficient not just to obtain consensus, but also to enable those initially in the minority to experience their conversion as fully voluntary and uncoerced. In a weaker form, Post (1995, 273–84) advocates a view of “responsive democracy” in which the conditions of discourse are such that the dissenters could maintain “a warranted sense of autonomous and effective contribution, through public discourse, to the process of creating the social order in which they live” (Post 2005, 144).

There can, of course, be other bases for this kind of legitimating identification with the decisions of a polity with which one might otherwise disagree. I will consider some others below. The important point here is that neither version of discursive legitimation is really credible. The strong version would require unanimity for, in the face of imperfect agreement, all attempts at collective self‐governance must conclude in prescription under the threat of coercion (Winter 1991a, 970–71). The weak version presents formidable empirical problems: under conditions of modern life, very few people can even credibly—let alone, justifiably—believe that they have an effective voice in creating the social order in which they live. The most that could plausibly be claimed is that in liberal democracies everyone has a formally guaranteed right to contribute to the public discourse that shapes the social order—though even that is waning as it now will often be restricted to a barricaded “free speech zone” miles away from the policymakers one seeks to influence (Archibold 2001; Kifner 2004).

Michelman (1999, 8) is right: understood as collective self‐government, democracy is a “necessarily compromised” ideal that presents itself to us as “necessarily damaged goods.” Still, for all his commendable realism, Michelman understates the case. The institutional difficulty struggles with the problem of how self‐governance can be collective, but, it takes for granted that self‐governance can be individual. Indeed, it takes for granted (and, as we shall see in a moment, it can fairly be said that Michelman takes for granted) that there are “selves” out there capable of self‐governance.

To put it differently, the conceptual vocabulary of democracy depends on concepts of the self, agency, and autonomy made problematic—if not, indeed, obsolete—by disciplines ranging from psychoanalysis through poststructuralism and contemporary cognitive science. When the “self” is itself understood as socially constructed, what does it mean to talk about self‐governance? What can autonomy mean once we have come to recognize our “choosing” selves as the products of historical contingencies, ongoing social practices, and other such exogenous factors? To paraphrase Michelman (1999, 12–13), “who or what is the agent or subject of the self‐government we have in mind—who or what is the self whose government by him of himself, or by her of herself, or by it of itself, is the government we care about?” Let us call this “the individualist difficulty.”

Michelman addresses this problem of the contingency of the self in his discussion of liberal constitutional rights, where he attempts to finesse the question. Liberal individualists, he says (1999, 66–67), can acknowledge that we are “socially situated beings, enmeshed in institutions, cultures, vocabularies, relationships and groups, dependent on them for identity.” They can even concede the contingency of the faiths and commitments “that a person never chose but simply has, from birth or from accident.” All that is necessary is “the view that a person's life will be massively lacking in value if it does not acquire enough of its shape and direction … from aims, attitudes, commitments, and pursuits that the person willingly and consciously adopts as her own.”

It is an appealing vision that resonates with us because we are, after all, constituted as liberal individual subjects. But, it has several problems. The first, most immediate problem is its liberal parochialism: there are communities, both historical and contemporary, in which a person's life is considered massively lacking in value if it does not acquire its shape and direction from aims, attitudes, commitments, and pursuits given from elsewhere—God, tradition, filial or patriotic obligation—and assumed as a matter of duty. From a Kantian or liberal individualist perspective, that person still acts autonomously as long as he or she chooses to adopt those duties as his or her own. But such a response is not a resolution of the problem of liberal parochialism so much as reformulation of the conflict in its own, “voluntarist” terms. For, within the worldview of such communities, it is the acceptance of and surrender to exogenous higher authority that is the essence of virtue. Indeed, within such worldviews, the acceptance of and surrender to higher authority is often understood as the very definition of authenticity. In contrast, the idea that one gets to choose whether to accept the authority of God, tradition, etc. is itself likely to be seen as a sign of hubris. ⁴

The liberal redescription, in other words, responds to a very real problem of incommensurabilty with an act of misrecognition. One might, of course, take the liberal account to be more accurate or true than the insider's phenomenological understanding (which would then be a form of false consciousness). But that would be particularly self‐contradictory for liberalism with its core commitments of tolerance and respect for human dignity. And, in fact, liberal democratic polities are systematically confounded when confronted by the claims of communities whom it cannot understand except on its own liberal terms. This problem is apparent in the context of Native Americans’ religion‐based land claims, which are systematically transformed in a way that simultaneously distorts and disadvantages them (Williams and Williams 1991; Winter 1991a).

The second problem with Michelman's chastened reformulation of liberal autonomy is that it does not answer the individualist difficulty so much as assume it away. He invokes our common sense that what gives our lives value is our capacity to take conscious responsibility for or ownership of its socially situated shape. Here, Michelman assumes each of the principal points in contention: (1) that it is the operative moment of choice—rather than, say, sacrifice, or suffering or the value of one's contribution—that gives human life value and nobility; (2) that choice is uncomplicated and meaningfully possible; and, relatedly, (3) that the subjective experience of choice is genuine and not an act of rationalization, bad faith, or neurotic self‐delusion. When the ego, in Freud's (1996, 353) famous phrase, is not even master in its own house, it is hard to know when, how, and if my conscious choices are “authentic” let alone autonomous.

This is a corrosive, even upsetting, critique. Not surprisingly, it tends to mobilize mechanisms of denial and resistance. Because I am going to lay it aside for a few moments, I want to dispel any misunderstandings I may have created. I am not saying that the past is a prison house that statically and deterministically reproduces itself. Nor am I saying that personal transformation is impossible. For one thing, I would be the last person to make that claim having been reared in an intensely insular tradition of the sort I referred to earlier and in which I no longer participate. More to the point, as I have argued elsewhere, once we understand the cognitive mechanisms of social reproduction it becomes clear that social reproduction is necessarily dynamic, imaginative, and adaptive (Winter 1991b). Social reproduction, therefore, is never really static even in the most rigid of traditions. Nor am I denying that we can distinguish between more or less autonomous modes of being. (Here, ontology recapitulates phylogeny: What is true of democracy is also true of autonomy.) I am arguing, rather, that we cannot effectively talk about these issues in the Romantic vocabulary to which Michelman adverts. I am arguing that, if we are to nurture—perhaps it would be better to say preserve—democracy, we will need to rethink it free from these classic conceptions of the self, agency, and autonomy.

There is a second reason why the conceptual dependence of democracy on classic notions of autonomy is problematic. The conventional understanding of democracy as popular sovereignty and collective self‐government carries with it (both historically and conceptually) strong notions of equality. This is evident in rhetoric of the founding period: the Declaration of Independence's pronouncement that “all men are created equal”; the French revolutionary's slogan liberté, egalité, fraternité—freedom, equality, brotherhood; and the American revolutionary desire to live “by no man's leave.” It is apparent as well—though more problematically—in the Anglo‐American rule of law ideal. Notwithstanding the evident hypocrisy in disenfranchising broad swaths of the population—women, slaves, ethnic minorities—exhibited by most historical democracies, the idea of democracy as the rule of the people carries with it strong leveling notions. ⁵ “Democracy,” John Quincy Adams declared, “has no monuments. It strikes no medals. It bears the head of no man on a coin. Its true essence is iconoclasm” (Moltmann 1999, 45). Conceptually, the democratic ideal implies strong egalitarianism because hierarchy is antithetical to self‐rule: the subordinated do live by some other man's leave; they live under conditions of heteronomy—which historically (and ironically) also became a reason for excluding some (women, slaves, etc.) from participation in the polity. On the collective level, democracy requires equal justice under law. No one is above the law because everyone must be governed by the same law, the law that the people collectively give to themselves.

This intimate, conceptual connection between democracy and equality is an important part of the resonance and historical power of the egalitarian ideal. The successive democratization of American society—the abolitionist movement, the universal suffrage movement, the public school movement, the women's suffrage movement, the labor movement, the progressive movement, the New Deal, the civil rights movement, the women's movement, the gay‐lesbian liberation movement—has been the result of egalitarian assaults on the various forms of social, political, and material subordination. Even the word “democratize” connotes the intimate conceptual connection between democracy and equality: to democratize a practice, resource, or capacity is to make it more fully available to everyone, without regard to status, on terms of equality.

At the same time, however, the intimate connection between democracy and equality creates a conceptual (as well as practical) problem for democratic theory: to the extent that democracy presupposes liberal autonomy (understood as freedom of decision and action in certain critical spheres of personal and public life), it also presupposes inequality. This is easy enough to see with respect to private property: the autonomy to acquire property entails both the right to accumulate it and, of course, to exclude others from using or otherwise appropriating it. But much the same dynamic is true of aspects of liberal autonomy, such as free speech, that we do not typically view as a zero‐sum game. Thus, Post (2005, 149–51) argues that attempts to legislate equality in public discourse—prohibitions on racist and other forms of hate speech, regulation of corporate speech, or limitations on campaign expenditures—conflict both with the primary democratic commitment to autonomy and, more acutely, with “the logic of democratic legitimacy” that requires the autonomy of all citizens to participate in public discourse in order to maintain their identification with the process of creating the social order in which they live.

Democratic theory resolves the conflict between autonomy and equality by recasting the latter as formal equality. As Post (2005, 148–49) writes, the “one person, one vote” principle “signifies that each person is to be regarded as formally equal to every other in the influence that their agency can contribute to public decisions.” In the context of public discourse, “the relevant equality of agency inheres in the liberty to express oneself in the manner of one's choice.” He thus concludes that “many forms of equality associated with notions of distributive justice or fairness might actually be inconsistent with democracy.” Democratic equality, he wryly notes, “can easily be experienced as thin and formal.”

For those who hold strong egalitarian principles, however, Post hopefully suggests that all may not be lost. Strong egalitarianism may yet have a place to the extent it enhances democracy. Imagine, he says, a group of citizens with the autonomy to participate in public discourse but who are so destitute, marginalized, and stigmatized by the majority that they are alienated from the polity and can no longer identify with its decisions as their own. In that case, democracy requires rectification of those inequalities: not because it is fair and right to do so, but because it is necessary to restore the sense of identification needed for democratic legitimacy. “The distinction is significant,” Post (2005, 152–53) says, because it means that: “Democracy does not require the full rectification of these inequities, but only the rectification necessary to maintain democratic legitimacy.”

Many will find this conclusion discomfiting. But, it is not the result of some mistake; it is the necessary consequence of Post's theoretical approach. To start with a rigorous (and monochromatic) definition of democracy and then proceed to its implications is, inevitably, to distort the concept precisely because it is abstracted from the social context of concrete historical practice that gives it meaning.

Here, the process of abstraction renders the concept of democracy defective and dysfunctional. It is defective because it turns democracy into the enemy of equality. When the conventional understanding of democracy is made dependent on an abstract conception of autonomy as exercised by individuals who are themselves viewed in abstraction from their constitutive social context and formation, it should hardly surprise that the corresponding notion of equality is a formal one that is every bit as much emptied of all substantive content. Famously, formal equality both conduces to and underwrites substantive inequality as we saw with respect to property and free speech. The law, in all its majesty, prohibits the rich as well as the poor from sleeping under the bridges of Paris. It affords exactly the same right to try to obtain social goods (such as property and free speech) to all. It is just that some will be much better situated to do so than others.

The process of abstraction is, also, particularly dysfunctional because it both devalues and dilutes the concept of equality. It devalues equality because it relegates it to a mere means: it robs equality of any independent normative force, making it relevant—that is, tolerated—only to the extent that it operates as a tool of legitimation. And it dilutes equality, as we have just seen, by making legitimation rather than fairness the measure of its scope.

It is this same distorting process of abstraction that plagues Michelman's argument: although he acknowledges the contingent, socially situated nature of the self and its commitment and aims, he nonetheless treats that self in abstraction from that context. He even says so: doubts about subjectivity and agency “do not stop liberals from making precisely such attributions, virtually automatically, at the times when they focus thought on matter of political arrangement” (Michelman 1999, 67). On Michelman's (1999, 66–67) account, the self is constituted by institutions, cultures, vocabularies, relationships, and groups; it is, he says, “dependent on them for identity.” Yet, he has the self somehow stepping outside itself to view the aims, attitudes, and commitments that are constitutive of its identity and decide whether to affirm them. Who is this self, emptied of all content, that stands outside itself to decide whether to adopt itself as its “true” self? It is, of course, none other than the autonomous self of pure will that Michelman purported to disclaim when he acknowledged the social constitution and situatedness of the self.

No, the self that steps outside itself would have to be the very same situated self—that is, the one constituted by the very aims, values, and commitments that it is deciding whether to jettison or keep. We can better see the circularity here by replacing the inapt, though deeply conventional spatial metaphor with the equally conventional one of reflection. Just as I eye myself in the mirror with all the subjectivity of my already extant self‐image—which depending on who I am may be unrealistically self‐flattering or witheringly self‐critical—the process of self‐reflection cannot be innocent, untainted, or (in the Kantian sense) autonomous.

But that does not mean that self‐reflection is impossible, just difficult. Much the same can be said of self‐governance. It is not a metaphysical capacity that individuals just “have,” but one that can be developed or destroyed in the process of coming to maturity. So, too, it can be nurtured or suffocated by the social practices available to the self as it elaborates its life in the relationships, groups, and activities through which it comes to understand and define itself. In short, the higher order processes of reflection and self‐governance—the preconditions or, at least, correlates of any meaningful notion of autonomy—are themselves socially constructed, socially situated capacities that we cannot (and cannot afford to) take for granted.

Sexuality and Self‐Governance

What would it mean to rethink democracy, autonomy, and self‐governance from a perspective that takes the subject as truly social? How can concepts such as democracy and autonomy even make sense without the classical conceptions of agency and “self”?

In this section, I propose to explore that question by considering a seemingly unrelated question buried in the previous discussion: what is the relationship between Kantian autonomy (understood as being governed only by those rules one gives oneself) and liberal autonomy (understood as freedom of decision and action in certain critical spheres of personal and public life)? On one level, the two conceptions of autonomy cohere or, at least, overlap. Both presuppose a sphere protected from the direction of others (particularly the State) in which the individual engages in self‐determination. But, the two are not synonymous. Within that protected domain, an individual may decide to act in ways that, though free of coercion by others, would not be autonomous in the Kantian sense because governed by passion rather than by a rule one has given to oneself.

Kantian and liberal autonomy are, thus, maximally distinct in the sphere of sexual autonomy that has been so central to modern constitutional law. Our examination of the concept of autonomy in that context will help us to uncover a perhaps hidden and, in some sense, counterintuitive connection between sexual autonomy and political self‐governance. Ultimately, this will lead us to a reconceptualized notion of autonomy as—paradoxically—shared, social, and dependent. By interrogating the link between sexuality and self‐governance, I hope to show that “autonomy” can be nourished only by attending to the interactions between a situated individual and some larger, interdependent set of social relations.

From the right to use contraception upheld by the Supreme Court in Griswold v. Connecticut (1965) and Eisenstadt v. Baird (1972), through a woman's right to terminate a pregnancy recognized in Roe v. Wade (1973) and Planned Parenthood v. Casey (1992), to constitutional protection for same‐sex intimacy in Lawrence v. Texas (2003), questions of sexual autonomy have held center stage in American constitutional law over the past fifty years. These, of course, are the preoccupations (and battlegrounds) of a society in which sexual autonomy is perceived by many (but not everyone) as a fundamental aspect of human flourishing. It is not hard to imagine how Kant might react to this cultural development. For Kant, it is rationality that defines what it means to be human; correspondingly, to be autonomous is to act only pursuant to reasons or ends that, upon critical reflection, one adopts as one's own. The passions, in general, and sexuality, in particular, threaten autonomy. Worse, sex for Kant is by its very nature immoral: as he says in The Metaphysics of Morals, “the natural use that one sex makes of the other's sexual organs is enjoyment, for which one gives itself up to the other. In this act, a human being makes himself into a thing, which conflicts with the right of humanity in his own person” (Kant 1996, 62). Sex, moreover, violates the categorical imperative because it makes of the other a mere means: “sexuality is not an inclination which one human being has for another as such, but is an inclination for the sex of another. …[O]nly her sex is the object of his desires” (Kant 1963, 164). Accordingly, sex for Kant can be redeemed neither by consent ⁶ nor by love. ⁷ It can be redeemed only by marriage because, in marriage, each accords the other reciprocal rights over the entirety of one's person (167).

As Irving Singer (1984, 382) notes with wry understatement, it is not likely that sexuality (and, we might add, intimacy) was for Kant “a cultivated or pervasive enjoyment.” Kant's understanding of sexuality is philosophically reductive, psychologically unsophisticated, and emotionally constricted. Singer (379) observes that, for Kant, sex in itself is just an appetite that “involves no desire for the welfare of its object.” But, this is not sexuality in its fully human form. It is instead sexuality in the form of perversion that, as Baudrillard maintains, grounds the narcissistic phenomenology of consumerism. As he defines it, sexual perversion is:

the inability to grasp the other, except discontinuously, as the object of desire in its total individual singularity: the other is transformed into a paradigm of various erotic parts of the body. …  This woman is no longer a woman, but a sexual organ, breasts, belly, thighs, voice, or face. …  From that point onwards she becomes an “object,” constituting a series whose different terms are ranked by desire, whose real signified is no longer at all the person loved, but the subject himself in his narcissistic subjectivity … turning amorous relations into a discourse with himself. (Baudrillard 1990, 53)

Eros, as Plato suggests in the Symposium (1956, 79–82), is a sexual desire that attaches to a person. “By awakening us to the living presence of someone else,”Singer (1984, 382) writes, “sexuality can enable us to treat this other being as just the person he or she happens to be. . . .”“Sex,” he continues, can “be seen as an instinctual agency by which persons respond to one another through their bodies.” This last point, I want to emphasize, is neither figurative nor merely descriptive. Some of you—I hope all of you—have had the experience during intimacy of that moment in which you could actually feel your lover's thoughts through his or her body. “Human life,” as Merleau‐Ponty (1964a, 227) explains, is “through and through mental and corporeal, always based upon the body and always (even in its most carnal modes) interested in relationships between persons.”

We can now see that the mind/body dichotomy is the axis around which so many of these otherwise disparate questions revolve. The problem of autonomy, like the problem of sexuality, is for Kant a question of mind and reason ruling over the body and its passions. ⁸ But “[d]esire and knowledge (body and mind),” Samuel Delaney (1999, 168) insists, “are not fundamental oppositions; rather they are intricately imbricated and mutually constitutive aspects of the social.” The mind/body dualism in all its varied forms—subject and object, reason and passion, freedom and constraint, public and private, internal and external—is, as Merleau‐Ponty (1964a, 226) suggests and as Kant's misapprehension of sexuality confirms, “tenable only this side of a certain point of misery and danger.”

Yet, it is precisely these dualisms—public reason and private interest, internal and external, subject and object—that we engage over and over again in our discourse about democracy. Thus, the problem of republican theory is how to produce citizens who will act with civic virtue in the public interest (the res publica), putting aside self‐interest and the interest of faction. The problem of pluralist democracy (Michelman's “institutional difficulty”) is how citizens can identify with and accept as their own the aggregated decisions of their fellow citizens even when their own interests or strongly held views are being sacrificed on the altar of majority rule. In much the same vein, the individualist difficulty arises for Michelman precisely because, as explained in the previous section, he treats the social as an external precondition that the subject can choose to reject or affirm as her own. As Merleau‐Ponty (1962, 362) explains: “It is as false to place ourselves in society as an object among other objects, as it is to place society within ourselves as an object of thought, and in both cases the mistake lies in treating the social as an object.”

The second, closely related motif concerns the distorting ontological power of liberal individualism. This is peculiarly clear in the constitutional right to privacy cases. In Eisenstadt (1972, 453), the second of the Court's contraception cases, Justice Brennan declared that: “If the right of privacy means anything, it is the right of the individual, married or single, to be free from unwarranted governmental intrusion into matters so fundamentally affecting a person as the decision whether to bear or beget a child.” The first thing to note about this statement is that it characterizes sex and procreation as matters of individual autonomy when they are, ineluctably, activities we engage in with others. ⁹ Indeed, on reflection, Justice Brennan's characterization of sexual autonomy as an individual right takes on a surreal quality: after all, when one is alone one does not need a contraceptive.

The second thing to note about Justice Brennan's assertion is that it is not quite true. The decisional autonomy in matters of procreation that is protected by the Constitution is not, like other constitutional rights, something that belongs equally to all individuals. Consider a couple who conceive an unplanned child and disagree on whether to have it. If the man wants the baby but the woman does not, there is of course no way he can legally require her to bring the child to term. Conversely, if the woman decides to have the baby and the man does not want children, not only is there no legal way he can stop her but she can enlist the courts to declare his paternity and require him to support the child. Whatever may be said about this legal state of affairs—and, given the structure of gendered power relations in which we live, I would say that this asymmetry makes good sense—the one thing that should be clear is that it is decidedly not about an abstract, formally equal individual right to autonomy in matters pertaining to whether to bear or beget a child.

Taken together, these two observations crystallize my penultimate point. Sexual autonomy (whether understood as conforming my sexual behavior to only those moral rules I adopt as my own or, more simply, as freedom of action and decision with respect to my sexuality) can only be exercised within an interdependent, highly constrained, and already constructed social world. (As we shall see shortly, the same is true for autonomy generally.) In Merleauvian terms, it is not so much that my sexual autonomy can only be expressed in a social situation as that it can only be expressed through that situation. “The fact is,”Delaney (1999, 193–94) explains,

The “freedom” to “be”“gay” without any of the existing gay institutions or without other institutions that can take up and fulfill like functions … means nothing. Many gay institutions … have grown up outside the knowledge of much of the straight world. But these institutions have grown up very much within our society, not outside it. They have been restrained on every side. … They do not propagate insanely in some extrasocial and unconstrained “outside/beyond,” apart from any concept of social responsibility. … The freedom to “be”“gay” without the freedom to choose to partake of these institutions is just as meaningless as the freedom to “be”“Jewish” when, say, any given Jewish ritual, text, or cultural practice is outlawed; it is as meaningless as the freedom to “be”“black” in a world where black music, literature, culture, language, food, and churches and all the social practices that have been generated through the process of black historical exclusion were suddenly suppressed. I say this not because a sexual preference is in any necessary way identical to a race, or for that matter identical to a religion. … I say it because all three are complex social constructs, and thus do not come into being without their attendant constructed institutions.

Notice that, once we take the social seriously as a constitutive dimension of being, ¹⁰ questions of the form “what is X?”, “what does X mean?”, and “what does it mean to be X?” are converted into a question of “what are the constitutive contexts and practices that make X meaningful?” Moreover, from the perspective of this transformation, the former questions are now revealed as the abstractions and reifications that they are. Thus, Delaney marks “gay,”“Jewish,”“black,” and “freedom” in scare quotes (as, indeed, I did earlier in this paper with “autonomy”) to designate their quality as fictive objects of thought. To personalize the point, I am not Jewish because of parentage, ethnicity, race, or some other essentialized quality. I am Jewish either because I participate in the community, rituals, and practices of my childhood (which I no longer do) or (though, for one who does participate, this would be “and”) because (to paraphrase Michelman (1988, 1523)) Jewish history and upbringing are “constitutively present” for me “as language, culture, worldview, and political memory.”

Which brings me to the point of the argument. To ask “what is democracy?”, “what does autonomy mean?”, and “what does it mean to be self‐governing?” is really to ask “what are the constitutive contexts and practices that make democracy, autonomy, and self‐governance meaningful?” (Winter 2002). Curiously, someone at the Bush Department of State almost got it. Its website contained the following: “Democracy is indeed a set of ideas and principles about freedom, but it also consists of a set of practices and procedures that have been molded through a long, often tortuous history. In short, democracy is the institutionalization of freedom” (United States Department of State 2006). It went on to list all the usual candidates such as popular sovereignty, majority rule, minority rights, fair elections, and the rule of law together with such democratic values as tolerance, pluralism, pragmatism, cooperation, and compromise. What it failed to indicate, of course, is how one could short circuit that “tortuous history” to transport and implant those practices straightaway in a foreign cultural context. What it failed to explain is how a society that does not already happen to enjoy values such as tolerance, pluralism, pragmatism, cooperation, and compromise might go about shaping citizens who do.

As Michael Sandel (1996, 124–200) points out, from Jeffersonian agrarian democracy through the Knights of Labor and the industrial democracy movement in the early twentieth century, one of the central debates in U.S. politics was over what economic arrangements would best produce citizens capable of self‐governance. He notes, and as I have argued elsewhere (Winter 2005), the development of a consumer society—a process that was already in train and the subject of republican critique and condemnation by the late eighteenth century (Breen 2004, 299, 330–31; Sandel 1996, 145–50)—has bred citizen‐subjects who are passive and disaffected or, as Hannah Arendt (1968, 141–42) might put it, who abdicate the burdens of self‐government to the State and view themselves as “mere clients.” But Lizabeth Cohen (2003, 409) is also correct when, in response to Sandel, she points out that we are not going to reverse the socio‐cultural developments of the last century no matter how normatively desirable that might be. The question that haunts us is how, in the context of the modern world, we could meaningfully nurture and develop citizens with a capacity for self‐governance.

It would be fair at this juncture to ask what I mean by “self‐governance.” For now, let me offer the following working definition adapted from Arendt (1958, 175–247): self‐governance is the ability to exercise initiative with respect to one's fate. Note five things about this way of thinking about self‐governance:

First, it recognizes self‐governance as a relative capacity both in the sense that it may be possessed in various degrees (i.e., a person may have a greater or lesser capacity for self‐governance or may be more able to take initiative in some areas of life than in others) and in the sense that as humans our control over our lives is always partial and contingent. It may be dependent on luck, circumstance, good health, and a thousand other fortuities.

Second, self‐governance entails both agency and responsibility. To take initiative with respect to one's fate is necessarily to presuppose that the world will (at least to some degree) be responsive to one's actions, that one's actions have consequences, that one's actions therefore matter, and that one takes responsibility both for oneself and for what one's actions mean to others. To put it another way, a self‐governing person is one who understands that the world is response‐able to one's initiatives and, therefore, that one bears responsibility for one's actions.

Third, as I observed in the close of the previous section, self‐governance is a situated, developmental capacity: it can be nurtured or destroyed in the process of coming to maturity; so, too, it can be facilitated or suffocated by the social practices available to the self as it elaborates its life in the relationships, groups, and activities through which it comes to understand and define itself.

Fourth, self‐governance is multidimensional and incomplete. It is multidimensional because personal and collective self‐governance is not merely a single phenomenon as it is applied to the individual and to the group. If I am hungry, I can teach myself to cook. If I need to get to another city for a conference, I can go online and find the best flight at the cheapest fare. (These examples may seem too simplistic but, as we shall see in a moment, that is only because we take them for granted.) If I want to stop global warming, I am going need to do a lot more than just take the initiative to replace my incandescent light bulbs with the more planet‐friendly, squiggly kind. Indeed, even personal self‐governance, as we saw in the case of sexuality, will often require the assistance or participation of others; in other words, it will often require one to engage in politics. Accordingly, the capacity for self‐governance is incomplete because the move along the continuum from personal to collective self‐governance requires additional capacities such as the ability to empathize, identify, coordinate, and cooperate with others.

Fifth, it follows closely that personal and collective self‐governance are mutually constitutive and reinforcing, just as appreciated by the classical republican theorists. Some capacity to engage in personal self‐governance is necessary for me to exercise the initiative to engage with others in order to influence our shared fate; successful participation in collective self‐governance reinforces my sense of competence and ability to control my fate. Note that this way of understanding self‐governance harmonizes individual and collective self‐governance rather than pits them against one another as in Michelman's “institutional difficulty.”

Delaney (1999, 188) argues that there is something fundamentally wrong with the traditional view “that places pleasure and the body in fundamental opposition to some notion of a legally constrained social responsibility” and that we need to work toward “a discourse that sees that pleasure and the body are constitutive elements of the social as much as are law and responsibility.” It is easy to see why the traditional view might see sexuality as uniquely threatening to collective self‐governance. Sex is the sphere of passion, subjectivity, privacy, and self‐interest in its purest form: physical pleasure. It is, thus, the antithesis of the sphere of public reason in which we govern ourselves only by those rules that, after critical reflection and deliberation, we legislate to ourselves. From within this traditional view, constitutional protection of sexual autonomy is especially problematic and carries with it a peculiar burden of justification. In his recent book on constitutional theory, for example, James Fleming (2006, 93) justifies the constitutional right of privacy by assimilating it to deliberative democracy. Thus, he argues that it “reserve[s] to persons the power to deliberate about and decide how to live their own lives, with respect to certain matters unusually important for such personal self‐governance.”

Fleming's formulation rests on the conventional, but nonetheless questionable assumption that the crux of the privacy right lies in deliberation and personal choice rather than, say, the psychological importance of sexual and familial intimacy or the social importance of particular enclaves of intimate privacy. State prohibition of same‐sex intimacy, for example, would be no less an assault on human dignity (and perhaps more), if it were to turn out that sexual orientation is a function of genetic and developmental factors rather than a matter of “lifestyle choice.” Moreover, as we saw above, the conventional characterization of the right of privacy as a matter of personal choice is descriptively incorrect. Not everyone has a constitutional right to choose whether to be a parent. Moreover, once one does parent a child, not everyone has the same rights with respect to “personal decisions relating to marriage, procreation, contraception, family relationships, child rearing, and education” ( Lawrence 2003, 574). In Michael H. v. Gerald D. (1989), for example, a divided Court upheld California's statutory preference for the integrity of the nuclear family against the claims of a biological father seeking to maintain a relationship with the daughter he conceived with a married woman who was separated at the time, but who later returned to her husband. In both situations, individual choice takes a back seat to the claims of particular social enclaves of privacy. ¹¹

But Fleming is nonetheless right about one thing: sexual autonomy and personal self‐governance are profoundly related. It is an important area of emotional and psychological life where I must learn how to take initiative with respect to my well‐being and do so in concert with others. Under modern social conditions in the West, it is the social domain in which teenagers and very young adults get their first real taste of freedom as they explore their sexuality outside the supervision of parents and most social institutions. Indeed, in the earliest stages of adulthood, sexuality is the domain in which we learn to be responsive and responsible to the other. The successful negotiation of sexuality and, ultimately, intimacy requires one to develop skills and values such as empathy, negotiation, compromise, cooperation, recognition of and respect for the other. Conversely, the pathologies of intimacy—narcissism, manipulation, exploitation, etc.—are also pathologies of collective self‐governance. The social practices of sexual autonomy are an important arena in which, under contemporary conditions, we can develop our capacities for self‐governance.

Of course, sexuality is neither the first nor the only domain in which we learn responsiveness and responsibility to an Other. That begins already in infancy. But, this merely underscores (in a Freudian or neo‐Freudian, object relations sort of way) the continuity between early childhood development and later emotional, psychological, and sexual maturity. Interestingly, one of the devastating effects of childhood sexual abuse is the peculiar way in which it destroys the capacity for agency. A few years ago, I was having lunch with a friend and former colleague and I asked after his former lover, with whom I knew he was still in contact. “Funny you should ask,” he said, “I just spoke to Tom the other day. He's practicing with a large law firm in D.C. and he called me all in a panic because he's going on his first business trip and he realized that he had no idea how to get a ticket, how to get himself to the airport, etc.” I asked my friend whether Tom had been sexually abused as a child. “Oh, my god,” he exclaimed, “how did you know?”

For those who hold a deracinated, Kantian view of autonomy the idea that self‐governance is a product of such intimate dependencies must seem maximally dissonant. But, it will be no surprise that the child–parent and sexual‐romantic relations are two of the primary sites in which we seek recognition and establish identity. It is perfectly normal if not clichéd to say that a critical part of what we moderns expect in our intimate sexual relationships is someone who “sees me as I really am.” One of the important psychological goods that intimacy confers is the comfort and confidence that, again to use the vernacular, allows me “to be me.” The Kantian view of rational agency as the defining characteristic of humanity yields a particularly deracinated view of autonomy as a matter of rational self‐legislation. But a more realistic, humanistic view that understands that we are, necessarily, social beings yields a view of freedom and self‐governance as situated phenomena in which an active human intelligence is able to establish its autonomy on the very ground of its dependence. ¹² In Merleau‐Ponty's (1962, 456) lovely phrase, “without the roots which it thrusts into the world, it would not be freedom at all.”