Sex, Culture, and Justice: The Limits of Choice (14 page)

26. Ibid., 80.

27. Ibid., 79.

1. For example, Hirschmann claims that, in the case of domestic violence, ‘‘The first level . . .—social construction as domination and ideological misrepresentation—is the most evident’’ (ibid., 114).
   

   be understood not as an objective feature of humanity, or of woman- or manhood, but as the
   normative
   principles of equality and autonomy. Materialized social norms can be criticized as ideological misrepresen- tations only from a standpoint of universal normative principles that do not depend on social norms for their validity.
   

   Normative principles such as freedom and equality should not be understood as objective facts about the essential nature of human be- ings. Describing someone as free, or as equal with others, can be un- derstood in two ways: as a description of how that person is actually treated in the context in question, or as a claim about how that person
   ought
   to be treated. For example, describing a black slave in colonial America as free and equal is either an incorrect statement of fact (on the first understanding) or a normative statement that slavery should be abolished (on the second). It is not particularly helpful to under- stand the claim that the slave is free and equal as referring to some objective fact about the essential nature of humanity, that humans are naturally, metaphysically, or unavoidably free and equal. Not only is it unclear what such a statement would mean, its substantive implica- tions are in any case that people
   ought to be treated
   as free and equal. Normative principles can be universal without being essentialist: de- scribing how people ought to be treated, rather than how they inevita- bly or essentially are.
   

   Finally, Hirschmann outlines the third level of social construction: ‘‘the discursive construction of social meaning.’’ It is this level which, according to Hirschmann, invokes ‘‘postmodern’’ thought, particularly that of Foucault.
   29
   ‘‘At this level,’’ Hirschmann writes, ‘‘construction of reality takes root in our very language, where it establishes the parame- ters for understanding, defining, and communicating about reality, about who women are, what we are doing, what we desire.’’
   30
   In other words, social construction affects our ability to communicate and con- ceptualize by affecting language. Without language, Hirschmann maintains, we cannot have ideas, and cannot be subjects: ‘‘We can only be the kinds of persons that our context and language allow.’’
   31
   This level of social construction is needed, Hirschmann suggests, to avoid
   

2. Hirschmann cites thinkers such as Foucault, Derrida, Lyotard, Spivak, and Butler in
   The Subject of Liberty
   (see page 81)
   .
   It is Foucault, though, who receives her most detailed analysis.
   

30. Ibid., 80.

31. Ibid., 81.

the implication of the first two levels that ‘‘men are the constructors, women the constructed, just as, for Marx, capitalists are the construc- tors, the proletariat are the constructed.’’
32

As we have seen, Bourdieu in particular explicitly emphasizes that his account of domination does not imply such a crude social analysis of hierarchy, and so Hirschmann’s desire to avoid such an implication is well founded. However, it is not clear that such an implication either is entailed by the first two levels, or would be countered by the social construction of language. If a powerful group wishes consciously to dominate a powerless group, and construct it as inferior, it can use the social construction of language as a tool, inventing and using deroga- tory terms for the members of the powerless group—or, indeed, flat- tering terms for members of that group who display desired character- istics. Similarly, the powerless group can consciously and intentionally attempt to resignify that same language. Thus, the term ‘‘queer’’ was consciously used and entrenched for a long time as a tool of homopho- bia. However, it has been consciously ‘‘reclaimed’’ by some members of the homosexual community as a means to fight its oppression, so that ‘‘queer politics’’ now indicates a radical endorsement of homosex- ual equality. Socially constructed language, then, can be a tool for either top-down oppression or more diffuse, capillary forms of power and resistance.

While Hirschmann’s discussion does highlight the key elements of social construction, and helps us to think through the concept in the context of gender, I suggest a return to the basic idea behind Hirsch- mann’s account: that social construction has effects on both the
options
that are available to be chosen, and on the
preferences and beliefs
that lead an individual to choose one option rather than another.
33
In other words, we can ask what it is that an individual is able to choose, and we can ask whether and in what sense she actually, actively makes an autonomous choice.

Liberalism and Universalism

How, then, can liberals respond to the charge that their attempts at developing universal principles are rendered invalid by the social con-

1. Ibid.
   

2. Hirschmann terms these two effects ‘‘the external structures of patriarchy and the inner selves of women’’ (ibid., ix).
   

1. struction I am asking them to acknowledge? As we have seen, some liberals attempt to escape this problem by stepping back from claims to universality. Political liberals such as John Rawls, for example, claim that their liberal principles need not be universal since they need not apply to all aspects of people’s lives (or to the lives of people who do not live in liberal societies). Rather, liberal principles apply to political life only, and moreover, they can be accepted as appropriate political principles by all reasonable people within a liberal society. If this over- lapping consensus pertains, there is no need for controversial claims about universality. Hence Rawls is keen in both
   Political Liberalism
   and
   The Law of Peoples
   to distance himself from any claims that liberalism might be a universal doctrine. In
   Political Liberalism
   the distancing oc- curs when liberalism is relegated to the political sphere and not the comprehensive; in
   The Law of Peoples
   liberalism is further restricted to societies that are already liberal.
   

   I suggested earlier that, rather than escaping the problem of univer- sality, the strategy of political liberalism is actually particularly vulnera- ble to criticisms such as Butler’s. Because it attempts to portray itself as a doctrine based on no comprehensive conception of the good, a doctrine to which all can agree without being coerced or accepting power-laden discourse on substantive nonpolitical values, it is particu- larly vulnerable to accounts of social construction that question the possibility of such neutral universality. However, Rawls’s strategy of restricting liberalism to already liberal societies undermines its norma- tive force. Indeed, this strategy conflicts with what Brian Barry sees as liberalism’s very purpose. As he puts it, ‘‘The point of liberalism is that it is universalistic. . . . The liberal position is clear. Nobody, anywhere in the world, should be denied liberal protections against injustice and oppression.’’
   34
   

   There is something of a tension, then, between the claim that liber- alism needs to recognize its own particularity and the liberal desire to make universal egalitarian claims. Rawls recognizes that comprehen- sive liberalism is not universal, but even political liberalism rests on substantive premises. Not everyone does, can, or will agree with those premises or with the liberal conclusions, and finding liberal arguments persuasive may have much (though not everything) to do with being brought up within a liberal society. Political liberalism may appear to
   

2. Brian Barry,
   Culture and Equality,
   138.
   

   be universal in origin (since it is derived from an overlapping consen- sus) and (nearly) universal in application (‘‘nearly’’ in that the exclusion of unreasonable groups from the domain of deliberation may mean that they are excluded from the domain of liberal freedom and equal- ity). In fact, political liberalism is particular in origin and particular in application, even within politically liberal societies. It is particular in origin because not everyone has or will or can agree with liberal values. As Will Kymlicka argues, ‘‘Non-liberal minorities . . . want internal restrictions that take precedence over individual rights. Rawls’s politi- cal liberalism is as hostile to that demand as Mill’s comprehensive liberalism. The fact that Rawls’s theory is less comprehensive does not make his theory more sympathetic to the demands of non-liberal mi- norities.’’
   35
   Similarly, Charles Taylor claims, ‘‘Liberalism is not a possi- ble meeting ground for all cultures, but is the political expression of one range of cultures, and quite incompatible with other ranges. . . . [L]iberalism can’t and shouldn’t clam complete cultural neutrality. Lib- eralism is also a fighting creed. The hospitable variant I espouse, as well as the most rigid forms, has to draw the line.’’
   36
   Liberalism must be particular in origin. It does not start from universal premises, or from a position outside social influence. It cannot be derived from every alternative conception of the good. It is not the real meaning behind all other, apparently illiberal, texts, religions, or cultures. Liber- alism is a substantive, power-laden, non-neutral, situated doctrine that clashes, fundamentally and sometimes irreconcilably, with other doc- trines.
   

   Moreover,
   political
   liberalism at least is particular in application. As Okin argues, even those groups which are deemed reasonable are al- lowed to engage in illiberal practices outside the political sphere. As such, some individuals within those groups (often women) do not ade- quately benefit from liberal freedom and equality. Such groups may claim liberal acceptance, since they comply with liberal justice in pub- lic. However, this public compliance allows them to reject those self- same values in the private sphere—the sphere that most undermines the equality of the most vulnerable individuals.
   37
   

   If we reject political liberalism, two main options remain for restor-
   

3. Kymlicka,
   Multicultural Citizenship,
   164.
   

4. Taylor, ‘‘The Politics of Recognition,’’ 63.
   

5. Okin, ‘‘Is Multiculturalism Bad for Women?’’
   

   ing consistency between universalism and particularism. The first is to accept that liberalism is a particular doctrine and, as a result, to restrict its application to liberal societies—or more precisely, to culturally ho- mogenous liberal societies. Liberalism becomes openly particular in both origin and application. This option constructs liberalism as a sub- stantive theory of the good like any other, and relegates its ethical status to that of moral relativism, an outcome that is problematic on several levels. At the most simple and practical level, there are very few homog- enous liberal societies, and so very few places for such a liberalism to apply. Few liberals would be content to limit their horizons so drasti- cally. Moreover, a political doctrine such as liberalism must be compat- ible with some account of political obligation. That is to say, it must be able to give a reason why those who do not agree with liberal laws or institutions must nevertheless obey them, and cannot claim the sover- eignty of self-interest. Furthermore, as a normative theory, liberalism needs to be able to claim that some positions are flawed, and to have some grounds for making such judgments.
   

   An alternative option is for liberals to recognize that their doctrine can be particular in origin without renouncing all claims to be univer- sal in application. Liberalism can appear caught in a series of paradoxi- cal dilemmas concerning the scope of its fundamental principles and the contradictions of universality and pluralism. These dilemmas take the form of the question ‘‘Should we apply X to those who reject X?’’ Thus we may ask, ‘‘Should we tolerate doctrines that preach intoler- ance?’’ or ‘‘Do we treat people equally if we allow them to treat them- selves and each other unequally?’’ or ‘‘Do we violate people’s right to choose their own way of life if we argue that a chosen life is better than an unchosen one?’’ John Gray argues that liberalism has ‘‘two faces,’’ with one face answering ‘‘no’’ to these sorts of questions and the other answering ‘‘yes.’’
   38
   Similarly, William Galston identifies two versions of liberalism and defends the version that would answer ‘‘yes’’ (so that, for example, liberals should tolerate intolerance).
   39
   In Part Two I dis- cuss similar questions that arise when considering whether the liberal concern for choice and autonomy imply that people should be able to choose nonautonomous lives. In general, I answer ‘‘no’’ to these sorts of questions. It seems to me to be a peculiar denial of the liberal point
   

6. John Gray,
   Two Faces of Liberalism.
   

7. William Galston, ‘‘Two Concepts of Liberalism.’’
   

   of view to answer ‘‘yes’’ to them, almost to say that being liberal re- quires one to be ashamed of liberalism and to attempt to suppress its values.
   40
   Liberalism seems to entail letting nonliberals win—or, as the old joke goes, a liberal is someone who can’t take her own side in an argument.
   41
   

   In defense of my position it is worth reflecting on the relationship between equality and universalism. It is in the nature of liberal equality that it cannot be denied to people on the basis of characteristics such as gender, race, or culture. Equality is not equality if it allows women to be rendered systematically inferior, if it does not recognize the equal moral worth of those in other countries, or if it fails to treat members of some cultures and religions as individuals worthy of autonomy, for example, by failing to protect the rights of women. Indeed, the con- cepts of equality and universality are strongly linked. It is
   possible
   to have nonuniversal conceptions of equality.
   42
   For example, the Declara- tion of Independence of the United States stated in 1776 that ‘‘all men are created equal’’ but, at the time, that equality applied only to white males. Thus nonuniversal equality applies only to a particular group of people, between whom conditions are equal. However, the more a conception of equality is nonuniversal—that is, the smaller and more specific is the group—the more it is a description of one level of a hierarchy, rather than a statement of equality. ‘‘All white men are equal (and not black people or women)’’ is another way of saying ‘‘society is stratified along lines of gender and race.’’ In other words, society is characterized by inequality. Nonuniversal equality is hierarchy. It fol- lows, then, that a strong commitment to equality is
   pro tanto
   a commit- ment to universalism.
   

8. Susan Mendus poses a similar question in her
   Impartiality in Moral and Political Phi- losophy:
   she asks whether the liberal ideas of skepticism and reasonable pluralism are self- undermining since they imply their own uncertainty or contingency. See especially 18–22.
   

9. This point, that the alternative view of liberalism seems to entail letting nonliberals win, echoes what Susan Mendus describes as the paradox of toleration, ‘‘which involves explaining how the tolerator might think it good to tolerate that which is morally wrong.’’ Mendus believes that the paradox can be solved by pointing to the liberal belief in autonomy: ‘‘We ought to tolerate what is morally wrong because it is part of being an autonomous agent that one should be allowed to do what is morally wrong.’’ However, the paradox remains if the ‘‘morally wrong’’ thing that we are allowing the agent to do in the name of autonomy is to follow a nonautonomous life (
   Toleration and the Limits of Liberalism,
   20, 161).
   

10. It is also possible to have a nonegalitarian conception of universality, if one’s concep- tion of universality were something like ‘‘everyone should adhere to the rules of this particu- lar hierarchical system.’’
    

    Joseph Raz considers the concept of universality. He shows, through a variety of examples, that no set of formal conditions can encapsulate the intuitive meaning of the concept, for any proposed conditions have counterintuitive results. As such, he concludes, ‘‘The view that values are universal, as a commonly expressed view, seems to me to be a
    substantive moral view.
    . . . It is a product of the moral struggle for the rejection of certain false value distinctions: the rejection of the special privileges of the aristocracy, and the evaluative beliefs which under- pinned them in pre-enlightenment Europe, the rejection of racism and sexism, and the like.’’
    43
    This statement echoes Butler’s position: any suggested definition of the universal contains within it substantive moral claims about the kinds of people who should count within it, and the kinds of attributes which are normatively relevant. Raz is correct to make such a statement. However, his conclusion that liberals should settle for a minimal, formal understanding of the concept of the univer- sal is unsatisfactory, as his counterexamples show.
    44
    Rather, liberals should recognize the connection between the concepts of equality and universality: insofar as liberalism is a theory of equality and not hierar- chy, it should be universally applied. Paradigmatically, liberal equality denies the moral relevance of ascriptive characteristics such as gender and race, and asserts that the individual is the correct unit of moral analysis. As such, liberal values of freedom, autonomy, and equal worth should be accorded to individuals regardless of their ascriptive characteristics, and regardless of their position within a group hierar-
    

11. Joseph Raz,
    Value, Respect and Attachment,
    58.
    

12. Raz suggests four possible criteria for a value to be universal. The first is that ‘‘the conditions for [the value’s] application can be stated without use of singular references, that is, without any reference to place or time, or to any named individual, etc.’’ The second adds the requirement that ‘‘in principle, [the value] can be instantiated in any place and at any time’’ (
    Value, Respect and Attachment,
    54). Neither of these definitions of universality is suffi- cient, according to Raz, since they would allow gendered values, such as feminine virtues, to count as universal. The third is that ‘‘a value is universal only if, if at least some people can display it, then it is in principle possible for every individual person to display it’’ (56). Raz dismisses this one as well, since everyone could in principle have a sex change, again render- ing gendered values universal. Finally, Raz suggests a definition according to which ‘‘only rights which everyone enjoys, not merely ones which everyone can in principle enjoy, are universal.’’ He rejects this suggestion as well, although his reason for doing so is odd: ‘‘Not everyone is funny or amusing, but we are, I assume, happy to regard this as a universal evaluative property’’ (58). It is not quite clear in what way humor is a universal property—on the contrary, people seem to have very different ideas of what constitutes humor or amuse- ment. That problem aside, the point of relevance here is that Raz is correct to dismiss at least the first two definitions, and possibly the third as well, as insufficiently substantive concep- tions of universality.
    

    chy. The substantive normative principles that underlie the liberal com- mitment to equality also suggest a commitment to universality.
    

    Richard Rorty suggests that liberalism has an inherent tendency to universalize, but claims that this tendency need not rest on a commit- ment to universal origins or to neutral or objective premises. His ap- proach of ‘‘ironism’’ is a version of social construction, according to which we must recognize that our beliefs are ‘‘caused by nothing deeper than contingent historical circumstance,’’
    45
    and that our own ‘‘final vocabularies,’’ or the sets of words we use to justify our basic beliefs, cannot themselves be given any noncircular justification. More- over, an ironist feels that there is no sense in which her own final vocabulary is more objective or true than the final vocabulary of others, since she finds admirable qualities in alternative final vocabularies, and since she rejects the very idea that a final vocabulary could ever be ‘‘true’’ or ‘‘objective.’’
    46
    This combination of views leads liberal ironists such as Rorty to conclude: ‘‘We cannot look back behind the processes of socialization which convinced us twentieth-century liberals of the validity of [liberal values] and appeal to something which is more ‘‘real’’ or less ephemeral than the historical contingencies which brought those processes into existence.
    We
    have to start from where
    we
    are.’’
    47
    

    Despite this radical rejection of objective foundations for liberalism, Rorty retains a commitment to universality. I suggested earlier that a commitment to equality entails a corresponding commitment to uni- versality. Rorty does not engage with the concept of equality as such; for him, it is the avoidance of cruelty that defines liberalism. However, he also maintains that liberalism contains within itself the tendency to universalize, a tendency that in no way conflicts with its particular ori- gins but rather flows from it. This is because what might be thought of as the ‘‘ethnocentrism’’ of a liberalism that can only be particular ‘‘is the ethnocentrism of a ‘we’ (‘we liberals’) which is dedicated to enlarg- ing itself, to creating an ever larger and more variegated
    ethnos.
    It is the ‘we’ of the people who have been brought up to distrust ethnocen- trism.’’
    48
    In other words, even a commitment to liberal values that is self-consciously contingent can be compatible with the wish to univer-