While I believe that all contemporary social movements are amenable to analysis in these terms, I am going to focus on the feminist movement to make my point. Several interesting discussions of the relevance of Habermas’s dualistic social theory to the contemporary women’s movement have already appeared.14 In the most comprehensive article on the subject of feminist critique of dualistic social theory, Nancy Fraser argues that, far from facilitating an understanding of feminism, Habermas’s dualistic social theory (especially his distinction between system and lifeworld), is not only “gender — blind,” but also “in important respects androcentric and ideological.”15 Fraser proposes a far more radical critique of dualistic social theory than the one I am proposing.
Fraser rejects Habermas’s system/lifeworld distinction, arguing that there is no meaningful way to differentiate categorically between the spheres of paid and unpaid labor, between the family and the “official” economy.16 Indeed, she argues that there is no warrant for assuming that a system-integrated organization of childrearing would be any more pathological than that of other work. This response, however, misses the real thrust of the distinction between system and social integration.
While Habermas, in his more Marxist moments, does try to distinguish between symbolic and material reproductive processes, the heart of the theory rests on the far more important distinction between modes of action coordination and not on the substantive elements of action itself. The claim, in short, which Fraser has not disproved, is that there is a fundamental difference between processes (cultural reproduction, social integration, socialization), social relations, and institutions in which the weight of coordination must be communicative and those that can be “media-steered” without distortion, such as markets or bureaucracies. This is not because labor or creative/productive activity takes place only in the second domain, but because meaning, norms, and identities cannot be maintained, reinterpreted, or created through functional substitutes for the coordinating accomplishments of communicative interaction. The heart of the difference between
formally organized sets of social relations (subsystems) and others lies in the tendency of the former to neutralize the normative background of informal, customary, or morally regulated contexts of action that are tied to validity claims and to substitute for these contexts of interaction that are generated by positive law and are “media-steered.”17 The latter are coordinated by media that operate through linguistic codes; these codes, however, relieve actors of the necessity of mutually agreeing on the definition of the situation involved in every relevant interaction, thereby bypassing (or rendering impossible) the reference to normative validity claims. Meanings, norms, and identities are not created in such contexts but are used (or reinforced) for systemic purposes.
Viewing the family as an economic system would thus entail either a wholesale embrace of systems theory18 (thereby rendering it immune to the kind of normative critique Fraser wants to make) or a misunderstanding of what a system is in Habermas’s theory: a formally organized, media-steered set of social relations. If one intends to challenge the meanings, norms, and identities that are constitutive of gender inequality, then this is the wrong tack to take. The systems-theoretic approach obliterates the very dimension in which these are created and reproduced. Although families do perform economic functions, although they can be and are functionalized by the imperatives of the economic or the administrative subsystem, although there are strategic interactions within them as well as exchanges of services and labor for money or support, and although these are distributed along gender lines, families are not thereby economic systems. They are neither formally organized nor media-steered. By the same token, they cannot be described as administrative systems even though they are certainly imbued with power relations.19
The work performed by women within the family is unrecognized, unremunerated, and uncompensated, and it therefore disadvantages women even in the “official” labor market (reinforcing the image of dependency on a male “breadwinner”). Nevertheless it is unhelpful to describe childrearing as being just like the rest of social labor. The fact that it can and has been partially transferred to day-care centers or nurseries and remunerated does not mean that it can be formally organized in the way that other work can be, nor that it is either desirable or possible to transfer childrearing in its entirety to system-integrated institutional settings. Communicative coordination of interaction remains at the heart of childrearing and nurturing, as any parent, childcare worker, or nursery school teacher knows. Unless one is advocating the total institutionalization of preschool children and the total commodification of childrearing as the sole alternative to being raised by full-time mothers, then one must assume that children come home at some time of day—at which point they require attention and nurturing. Moreover, nurseries,
day-care centers, and schools are themselves institutions within civil society. They have their economic and bureaucratic side, of course, but when organizational or economic requirements outweigh the communicative tasks of nurturing and teaching, they subvert the raison d’etre of the institutions and have pathological consequences (unnurtured and untaught children).
While it is certainly conceivable that more household tasks can migrate from the home to the market, surely there is and ought to be a limit to this. I do not agree with the notion that all creative, productive, or reproductive activities should necessarily take the form of wage labor. Even when they do, this does not mean that the institutional frameworks in which these activities occur can be analyzed as economic systems. Only on the misleading assumption that all “social labor” is equivalent and thus equally amenable to or distorted by system integration could one consider primary socialization and nurturing in the same light as all other work. Only, in short, if one construes families simply as sites of unpaid, socially necessary labor time could the differences between interfamilial relations and social relations of production disappear. But this sort of assumption has been criticized by many feminists for overextending the categories of the Marxian critique of capitalism to issues they were not constructed to address.20
If one is willing to grant that a modern economy requires that some forms of labor be commodified and formally organized, the central question for critical theory is to how to distinguish the sorts of activities that should be left to the market mechanism or formally organized from those that should not be. There are two distinct issues here. For example, feminist critiques of “surrogacy contracts” challenge the appropriateness of exchanging babies for money (commodification) and treating pregnancy and childbirth on the model of a labor contract. Marketization in such cases would seem to distort the woman’s relation to her body, herself, and her child, and it is not necessary to explicate this intuition on the basis of naturalist or essentialist arguments.21 The idea of the communicative infrastructure of the social relations of civil society suffices to account for the distortions that arise from delivering these relations over to the market. And while day care and schooling involve paid labor (marketization of teachers’ or child-care workers’ services), this does not mean that these activities can or ought to be formally organized. They do not have the same form, purpose, or meaning as other wage labor. The public and private institutions in which child care and teaching take place are core components of civil society, despite the fact that the professional services involved are remunerated. In short, some criterion is necessary for assessing whether or not commodification or formal organization would have implications with respect to certain forms of activity or interactions that are unacceptable and unnecessary in a modern society. My theory of civil society offers a good start in this direction.
Instead of attempting to render the roles of worker and childrearer compatible by assimilating the latter to the former, an analysis that proceeds from the system/lifeworld distinction would lead one to challenge the gender subtext of both roles while insisting on their difference. Modernization has already involved the migration of work (including education) from the home to the market. But surely a large part of a specifically feminist solution to the double burden of the working mother, to the subordination and insecurity attached to the homemaker role and to labor market inequities, must entail the degendering of the childrearing, nurturing, and homemaker roles, along with a fight against the gendered division of labor in the workplace. Wages for housework and child care would only reinforce its gendered character and lock women even more strongly into low-paid service jobs. The domestic “division of labor” clearly entails a power relation based in part on women’s economic dependency that deprives women of real choice and of equal voice in the distribution of such tasks; it both derives from and reinforces their inferior position in the labor market.22 It is this relation that must be challenged.
But this approach does not rest on the strained analogy between families and economic systems, nor between childrearing and other productive labor. Instead, it involves a challenge to the patriarchal norms that define families and attach genders to household and other roles. Indeed, the very possibility of articulating and challenging the ways in which the modem capitalist economy and the equally modern nuclear family intersect (through gendered roles) presupposes their differentiation. Changes in the identity, normative conception, and internal role structure of families would not alter the fact that interfamilial relations including childrearing must be communicatively coordinated. Quite the contrary. One could not even criticize the contemporary family as unjust, as deformed by the unequal distribution of money, power, and asymmetric gender relations, if one did not presuppose its communicative infrastructure.23
The distinction between conventional and postconventional orientations captures a key dimension of power in existing gender norms. The forms that male dominance takes in the patriarchal nuclear family and the ways in which it structures job categories (and client relations in the welfare state) and the corresponding gender identities are modern in the descriptive, historical sense.24 But they are neither rational nor modern in the normative sense—that is, in the way Habermas uses these words. The norms underpinning male dominance are an example of traditionalism par excellence; that is, they are based on a conventional normative “consensus” frozen and perpetuated by relations of power and inequality that lead to all sorts of pathologies in the lifeworld. The traditionalist attitude toward de facto norms based on such a consensus does not mean that the relevant norms are
lingering forms of premodern status inequalities. It does mean that they are sealed off from critique and traditionalized, as it were. Indeed, they are based on a selectively rationalized civil society, and it is precisely the blockages to its further modernization in the normative sense that Habermas’s theory tries to articulate. Moreover, the differentiation of the subsystems of economy and state from the lifeworld is a precondition for releasing the cultural potentials of modernity and for freeing communicative interaction from ritualistically reproducing sacralized, conventional norms. The lifeworld cannot be internally differentiated, the institutions of civil society cannot be modernized, subjectivity cannot be decentered, and roles cannot be challenged unless communicative interaction is unburdened from the task of coordinating all areas of life.
Nevertheless, there is more to male dominance than even a modern brand of traditionalism, and Fraser does a real service by signaling a missing dimension in Habermas’s analysis of power. It is misleading to restrict the term “power” to hierarchically structured relations in bureaucratic settings without providing another term to articulate asymmetric social relations in other institutions. One would do better to distinguish among different kinds of power or, rather, among various codes of power and modes of the operation of power. Otherwise, one is left without the means to conceptualize the differential ability to impose norms, define identities, and silence alternative interpretations of femininity, masculinity, and needs. Traditionalism results from this ability but does not account for it. It is important that we know how the various forms of power operate in the construction of gender, how they permeate socialization processes, and how the norms and identities generated in civil society intersect with functioning of power as a medium in bureaucratic settings.
This would involve an analysis of power relations that is supplemental rather than antithetical to the conception of power as a coordinating medium. I have argued that formal organization is a precondition (and hence a mark of identification) of the construction of the autonomous subsystem of power.25 It is a necessary prerequisite for power to function as a steering medium (and to be institutionalized as such). But it is neither the only mode in which power operates nor its only code. As many have pointed out, power generated outside of formal rules exists within organizations; power relations exist before the historical emergence of the medium of power, and power relations are operative in contexts that are not formally organized.26
Power can be defined generally as the transfer of selectivity (the ability to determine what can be done and said). Power operates through the conditioning of expectations (and of expectations of expectations), linking relatively preferred and relatively rejected combinations of alternatives of at least two persons.27 This transfer presupposes both the availability of negative
sanctions and a code (or several codes) of power. Many but not all codes of power incorporate forms of inequality that distinguish among individuals as higher and lower, superior and inferior.
In one of its guises, within formally organized contexts, power operates as a steering medium that can then be extended outward to functionalize relations and institutions of civil society that are not themselves formally organized and thereby achieve administrative goals.28 As such, the medium of power uncouples the coordination of action from consensus formation in language and neutralizes the responsibility of participants in the interaction.29 What counts here is not the presence of a rigid bureaucratic hierarchy or structure of domination in the sense of a clear chain of command30 but the formalization of the action context such that abstract rules and impersonal roles (be they offices or functions) become at least the official channel (one among several) through which power (selection of what can or cannot be said or done) flows. Thus, the binary schematization of interactions in sets of formal codes (especially legal/illegal) produces an objectivating attitude toward the action situation, an abstraction from concrete persons, and a certain automatic quality to the continuation of the interaction.31
Power does not operate only as a steering medium.32 There are, of course, power relations within institutional settings that are not formally organized and thus lack a necessary condition for the anchoring of the medium of power. Here, too, power operates through “binary codes” that transfer selectivity, expedite communication, and avoid the risks of dissention so long as they are not challenged. But these “codes” have a different structure from those attached to steering media in formally organized contexts. Most important, they do not fully replace ordinary language in its coordinating function; instead, they involve second-order processes of consensus formation in language. Nor do they involve depersonalized social relations. Habermas has analyzed prestige and moral authority in this way, distinguishing these “generalized forms of communication” from steering media. Prestige and moral authority can motivate action or compliance, but the validity claims underlying them can also be challenged; and if these do not survive critique, their normative basis and their power to motivate collapse. Moreover, moral authority and prestige remain strongly attached to particular persons and contexts.33
It is reasonable to assume that the list of “generalized forms of communication” could be expanded to include status, authority, and gender.34 Moreover, in line with Habermas’s distinction between normative and communicative action, I should distinguish between forms that allow communicative thematization and questioning up to a fixed point (such as traditional authority) and forms that are so constructed as to allow in principle for unrestricted thematization, questioning, and even criticism. It is also
possible for the structure of a generalized form of communication to change, for example, from traditional to democratic authority, from status to merit, or from one conception of gender to another.
I maintain that gender is a generalized form of communication or, rather, the code of such communication. Existing gender codes, even if historically changing and in this sense hardly traditional, are so constructed as to stop questioning at a supposedly unchallengeable meaning complex that is defined as “natural.” That power operates through gender codes, reducing the free selectivity of some and expanding that of others, is the most important and paradigmatic core of any theory that might be labeled feminist. Gender is not another steering medium, but rather the set of codes in and through which power operates. Outside formal organizations (where it can serve as a secondary code of the power medium), gender continues to displace ordinary language communication and facilitate the operation of power. However, the codification of gender does not fully uncouple interaction from the lifeworld context of shared cultural knowledge, valid norms, and responsible motivations. Gender norms and identities are based ultimately on the intersubjective recognition of cognitive and normative validity claims. While conventional understandings of gender also reduce the expenditure of interpretive energy and the risks attending mutual understanding, their ability to motivate action and compliance is still linked to the alternatives of agreement or failed consensus.35 This “relief effect” is not neutral in relation to the intersubjective recognition of norms, identities, or meanings.
Of course, the peculiar power of conventional interpretations in this domain lies in the fact that the meanings and norms at stake are bound up with identities that are transmitted through primary socialization and reinforced in secondary socialization processes throughout one’s adult life. Power operating in the code of gender delimits not only what one understands as natural/unnatural, natural/cultural, male/female, feminine/mascu — line, attractive/unattractive, and appropriate/inappropriate sexual objects and aims, but also constructs the meaning of bodies and operates upon them. Gender norms and identities are, in addition, reinforced by direct or indirect, positive or negative sanctions that can (but need not) be linked to unequal access to money and power in the form of media. They must therefore be challenged on two fronts: The conventional gender codes of power must be dissolved by actors who take the responsibility for creating new meanings and new interpretations into their own hands, while inequities in the distribution of money and power must be contested.
It is in this sense that gender identity links the public and private domains of civil society to each other and to the economy and the state administration.36 Viewing gender as a generalized form of communication, a power
code distinct from but reinforced by the media of money and power generated in the subsystems, gives us a rich theoretical framework for articulating the public/private distinction in gender terms.
The most significant flaw in Habermas’s work is his failure to consider the gendered character of roles of worker and citizen that emerged along with the differentiation of the market economy and the modern state from the life- world. Feminist historians have documented the parallel construction of the roles of housewife and mother and the restriction of women to these roles (as nurturer), as the flip side of the transition from the family economy to the capitalist mode of production and the replacement of autocratic/monarchic with republican/liberal forms of constitutionalism.37 As wage labor became dominant, the role of wage worker came to be understood as a gendered, male role, while the family was constructed to be a private sphere, the domain of women, in which no “real” work was done. The same holds true for the republican conception of the citizen-soldier, which by definition excluded women.38 It is no accident that as the roles of male breadwinner and male citizen crystallized, a cult of domesticity emerged to provide the ideological components of the new wife and mother role. Of course, a father role also developed, but this was an empty role, another name for the breadwinner. Thus, as a generalized medium of communication, gendered power relations have been built into all of the roles developed in (a selectively rationalized) modem society.39
It should be obvious that this reconstruction of the gender subtext of the institutional articulation of modern capitalist societies into sets of public and private relations does not undermine the dualistic social theory I have been defending. Rather, it presupposes the argument that the lifeworld “reacts in a characteristic way” to the emergence of the economic and state subsystems by internally differentiating itself into the public and private spheres of civil society, into sets of institutions oriented to cultural transmission, social integration, socialization, and individuation.40 In my analysis of civil society, the acquisition of actionable civil rights, however selective and problematic these may be, institutionalizes the public and private spheres of civil society and subjects the economy and the state to its norms. The norms at issue here are, of course, not the ones Fraser has in mind when she appropriates the conception of the multidirectional character of influence among the various public and private spheres of classical capitalism. Patriarchal gender norms are hardly “freedom-guaranteeing,” and they have justified the exclusion of women from those rights and norms that were. By implication, the gender norms that shape the key social roles mediating among institutions must be subject to critique and replaced by nonpatriarchal identities and roles.
The same holds true for the welfare state systems Fraser analyzes. I have argued that the norms of civil and political society continue to exert influence on the economy and state through the mediating institutions of political and
economic society. The “receptors” for societal influence in these spheres are, however, restricted in scope and highly selective with respect to which norms they mobilize or reinforce. Patriarchal gender norms are certainly among the latter, and they structure the roles and policies put in place by many welfare reforms. Since these norms (already backed up by inequalities in money and power) constitute women as dependents, it is not surprising that they comprise the bulk of those who become clients. The key question today is not whether but which lifeworld norms will be decisive.”41
The colonization thesis highlights the problems associated with the opposite direction of interchange: the penetration by the media of money and power (and formal organization) into the communicative infrastructure of everyday life. This tends to reify and deplete nonrenewable cultural resources that are needed to maintain and create personal and collective identities. This includes the resources that are needed to create nonpatriarchal norms in the lifeworld and to develop the solidary associations and active participation that would help them assert their influence on the subsystems.
Habermas’s sketchy but extremely suggestive analysis of the new forms of juridification utilized by welfare states highlights the ambiguities involved in the double process of interchange between system and lifeworld. On the one hand, juridification in the domain of the family involves the extension of basic legal principles to women and children who were formerly denied legal personhood by law under the doctrine of couverture (in Anglo-American countries at least). In other words, egalitarian principles replace patriarchal norms in the form of rights—of child against parents, of wife against husband, etc. Such new rights tend to dismantle the position of the paterfamilias in favor of a more equal distribution of competences and entitlements among family members. The direction of influence here clearly flows from civil society to the state, involving a choice of norms. It is these norms that are reinforced in civil society by the state as the end result of law making.
On the other hand, if the structure of juridification involves administrative and judicial controls that do not merely supplement socially integrated contexts with legal institutions but replace these by the operation of the medium of law, as is often the case under welfare law, then emancipation in the family is achieved at the cost of a new type of possible bond.42 Experts (judges or therapists) become the adjudicators of the new rights and the conflicts around them. They intervene with their juridical or administrative means into social relations that become formalized, dissociated, and reconstructed as individualized cases to be handled administratively or juridically like any other set of adversary relations. Formal, individualizing, and hence universalizing judgments that cannot deal with contextual complexities disem- power clients by preempting their capacities to participate actively in finding solutions to their problems. It is thus the medium of law itself that violates the communicative structures of the sphere that has been juridified in this way. This form of juridification goes beyond the external legal codification of rights. The administrative penetration of civil society it entails preempts the development of procedures for settling conflicts that are appropriate to structures of action oriented by mutual understanding. It blocks the emergence of discursive processes of will formation and consensus-oriented procedures of negotiation and decision making. It also necessarily abstracts away from the specific context, conditions, relations, and needs of each individual “case.” It is precisely the disempowering effects of this sort of decontextualized, individualizing, and formalistic decision-making that feminist analysts of recent reforms in family law have described and criticized in some detail.43
Debate and confusion over the meaning and desirability of seeking rights in this domain have permeated the feminist discussion. I believe that the distinction between law as institution and law as medium, and the colonization thesis are helpful here. A theory of civil society constructed along these lines allows one to conceptualize all of the important aspects that makes the new “rights” so ambiguous. From this perspective, it becomes clear that the ambivalence of feminists vis-a-vis “equal rights” legislation in this domain is based on a real dilemma: The acquisition of formal equality through means and techniques that abstract away from particular contexts, level differences, and block the creation of egalitarian social relations within civil society is an ambiguous gain indeed. In a context not only of substantive inequality (the old Marxist insight) but also of contested and fragile identities, such means will either generate new dependencies or foster the resuscitation of the old patriarchal norms as a defense against the disintegrative side effects of state penetration. Traditional patriarchal forms of life have become formally dele- gitimated by the new rights for women and children, but the client/expert relations that proliferate in civil society via the medium of law neither abolish substantive inequalities in power or voice nor facilitate the creation of new meanings, identities, and norms. In effect, the new vertical relations between the legal subject and the judge or social worker substitute for the horizontal communicative interaction needed to generate new solidarities, egalitarian norms, and ways of life to replace the old ones. Consequently, autonomous processes of collective empowerment and the creation of nonpatriarchal identities in civil society are blocked.44
It would be extremely misleading, however, to assume that all welfare state reforms have the same structure or logic. Surely legal reforms that secure the freedom of wage workers to organize unions and bargain collectively, that protect them from being fired for such collective action, and that secure worker representation on company boards differ in kind from means — tested grants to single-parent households and from social services that “instruct” clients on how to function properly as childrearers and responsi
ble providers according to some preconceived model.45 The difference between these types of reforms is not fully captured by reference to the genders (or, for that matter, to the race) of the people they target. In addition to stating that women are the objects of one type of reform, men of the other, one ought to be able to say what it is about the reforms themselves that make some enabling and others debilitating.
Dualistic social theory allows one to do just this. The former set of reforms, unlike the latter, do not create isolated clients of a state bureaucracy, but rather empower individuals to act together collectively, to develop new solidarities, and to achieve a greater balance of power relations because they are addressed to an area that is already formally organized.46 Such reforms create “receptors” in the economic subsystem for the influence of the norms and modes of action of civil society by putting procedures for discursive conflict resolution into place, thereby asserting control of the latter over the former without dedifferentiating them. The second type of reform does the reverse: It brings the full force of administrative agencies into areas that are not and should not be formally organized. This threatens the communicative infrastructure and autonomy of civil society and undermines the capacities of “beneficiaries” to act for themselves or to settle conflicts discursively. Yet one certainly would not want to argue that juridification, regulation, or monetary benefits in civil society by definition humiliate or disempower those whom they are meant to benefit. The question that arises is not whether juridification (the creation of new rights) or state intervention (the granting of new benefits) should occur in civil society, but which kind of legal rights, administrative relations, or monetary benefits ought to be established. Considering that women are the prime targets/beneficiaries of welfare in this domain, surely such a question is not “askew” of feminist concerns.47
A feminist version of the critique of the welfare state must involve its reflexive continuation.48 Thus, the decolonization of civil society and its modernization (in the sense of replacing conventionally held patriarchal norms with communicatively achieved norms) are both feminist projects. So, too, is the development of egalitarian institutions that can influence the administrative and economic systems. The first project would permit juridification only in forms that empower actors in civil society without subjecting them to administrative control. The second would dissolve male domination in both public and private institutions. The third would entail structural reforms in economic and political society to make them receptive and complementary to the new identities and the newly democratized, egalitarian institutions of civil society.49