Professional Resources HIERARCHIES, JOBS, BODIES: A Theory of Gendered Organizations JOAN ACKER University of Oregon and Arbetslivscentrum, Stockholm
In spite of feminist recognition that hierarchical organizations are an important location of male dominance, most feminists writing about organizations assume that organizational structure is gender neutral. This article argues that organizational structure is not gender neutral; on the contrary, assumptions about gender underlie the documents and contracts used to construct organizations and to provide the common sense ground for theorizing about them. Their gendered nature is partly masked through obscuring the embodied nature of work. Abstract jobs and hierarchies, common concepts in organizational thinking, assume a disembodied and universal worker. This worker is actually a man; men's bodies, sexuality, and relationships to procreation and paid work are subsumed in the image of the worker. Images of men's bodies and masculinity pervade organizational processes, marginalizing women and contributing to the maintenance of gender segregation in organizations. The positing of gender-neutral and disembodied organizational structures and work relations is part of the larger strategy of control in industrial capitalist societies, which, at least partly, are built upon a deeply embedded substructure of gender difference. AUTHOR'S NOTE: Presented at the American Sociological Association Annual Meetings, Chicago, August 1987. I wish to thank Judith Lorber, Pat Martin, and Ronnie Steinberg who contributed a great deal to this article through their careful and insightful comments and suggestions. Conversations with Harriet Holler, Carole Paleman, and Dorothy Smith also helped my thinking. REPRINT REQUESTS: Joan Acker, Department of Sociology, University of Oregon, Eugene, OR 97403. Most of us spend most of our days in work organizations that are almost always dominated by men. The most powerful organizational positions are almost entirely occupied by men, with the exception of the occasional biological female who acts as a social man (Sorenson 1984). Power at the national and world level is located in all-male enclaves at the pinnacle of large state and economic organizations. These facts are not news, although sociologists paid no attention to them until feminism came along to point out the problematic nature of the obvious (Acker and Van Houten 1974; Moss Kanter 1975, 1977). Writers on organizations and organizational theory now include some consideration of women and gender (Clegg and Dunkerley 1980; Mills 1988; Morgan 1986), but their treatment is usually cursory, and male domination is, on the whole, not analyzed and not explained (Hearn and Parkin 1983). Among feminist social scientists there are some outstanding contributions on women and organizations, such as the work of Moss Kanter (1977), Feldberg and Glenn (1979), MacKinnon (1979), and Ferguson (1984). In addition, there have been theoretical and empirical investigations of particular aspects of organizational structure and process (Izraeli 1983; Martin 1985), and women's situations have been studied using traditional organizational ideas (Dexter 1985; Wallace 1982). Moreover, the very rich literature, popular and scholarly, on women and work contains much material on work organizations. However, most of this new knowledge has not been brought together in a systematic feminist theory of organizations. A systematic theory of gender and organizations is needed for a number of reasons. First, the gender segregation of work, including divisions between paid and unpaid work, is partly created through organizational practices. Second, and related to gender segregation, income and status inequality between women and men is also partly created in organizational processes; understanding these processes is necessary for understanding gender inequality. Third, organizations are one arena in which widely disseminated cultural images of gender are invented and reproduced. Knowledge of cultural production is important for understanding gender construction (Hearn and Parkin 1987). Fourth, some aspects of individual gender identity, perhaps particularly masculinity, are also products of organizational processes and pressures. Fifth, an important feminist project is to make large-scale organizations more democratic and more supportive of humane goals. In this article, I begin by speculating about why feminist scholars have not debated organizational theory. I then look briefly at how those feminist scholars who have paid attention to organizations have conceptualized them. In the main part of the article, I examine organizations as gendered processes in which both gender and sexuality have been obscured through a gender- neutral, asexual discourse, and suggest some of the ways that gender, the body, and sexuality are part of the processes of control in work organizations. Finally, I point to some directions for feminist theory about this ubiquitous human invention. WHY SO LITTLE FEMINIST DEBATE ON ORGANIZATIONS? The early radical feminist critique of sexism denounced bureaucracy and hierarchy as male-created and male-dominated structures of control that oppress women. The easiest answer to the "why so little debate" question is that the link between masculinity and organizational power was so obvious that no debate was needed. However, experiences in the feminist movement suggest that the questions are not exhausted by recognizing male power. Part of the feminist project was to create nonhierarchical, egalitarian organizations that would demonstrate the possibilities of nonpatriarchal ways of working (Gould 1979; Martin 1990). Although many feminist organizations survived, few retained this radical-democratic form (Martin 1990). Others succumbed to the same sorts of pressures that have undermined other utopian experiments with alternative work forms (Newman 1980), yet analyses of feminist efforts to create alternative organizations (Freeman 1975; Gould 1979) were not followed by debates about the feasibility of nonpatriarchal, nonhierarchical organization or the relationship of organizations and gender. Perhaps one of the reasons was that the reality was embarrassing; women failing to cooperate with each other, taking power and using it in oppressive ways, creating their own structures of status and reward were at odds with other images of women as nurturing and supportive. Another reason for feminist theorists' scant attention to conceptualizing organizations probably lies in the nature of the concepts and models at hand. As Dorothy Smith (1979) has argued, the available discourses on organizations, the way that organizational sociology is defined as an area or domain "is grounded in the working worlds and relations of men, whose experience and interests arise in the course of and in relation to participation in the ruling apparatus of this society" (p. 148). Concepts developed to answer managerial questions, such as how to achieve organizational efficiency, were irrelevant to feminist questions, such as why women are always concentrated at the bottom of organizational structures. Critical perspectives on organizations, with the notable exception of some of the studies of the labor process (Braverman 1974; Knights and Willmott 1985), although focusing on control, power, exploitation, and how these relations might be changed, have ignored women and have been insensitive to the implications of gender for their own goals. The active debate on work democracy, the area of organizational exploration closest to feminist concerns about oppressive structures, has been almost untouched by feminist insights (Rothschild 1987; Rothschild-Whitt, 1979). For example, Carole Pateman's influential book, Participation and Democratic Theory (1970), critical in shaping the discussions on democratic organization in the 1970s, did not consider women or gender. More recently, Pateman (1983a, 1983b, 1988) has examined the fundamental ideas of democracy from a feminist perspective, and other feminist political scientists have criticized theories of democracy (Eisenstein 1981), but on the whole, their work is isolated from the main discourse on work organization and democracy. Empirical research on work democracy has also ignored women and gender. For example, in the 1980s, many male Swedish researchers saw little relation between questions of democracy and gender equality (Acker 1982), with a few exceptions (Fry 1986). Other examples are studies of Mondragon, a community in the Spanish Basque country, which is probably the most famous attempt at democratic ownership, control, and organization. Until Sally Hacker's feminist study (1987), researchers who went to Mondragon to see this model of work democracy failed to note the situation of women and asked no questions about gender. In sum, the absence of women and gender from theoretical and empirical studies about work democracy provided little material for feminist theorizing. Another impediment to feminist theorizing is that the available discourses conceptualize organizations as gender neutral. Both traditional and critical approaches to organizations originate in the male, abstract intellectual domain (Smith 1988) and take as reality the world as seen from that standpoint. As a relational phenomenon, gender is difficult to see when only the masculine is present. Since men in organizations take their behavior and perspectives to represent the human, organizational structures and processes are theorized as gender neutral. When it is acknowledged that women and men are affected differently by organizations, it is argued that gendered attitudes and behavior are brought into (and contaminate) essentially gender- neutral structures. This view of organizations separates structures from the people in them. Current theories of organization also ignore sexuality. Certainly, a gender-neutral structure is also asexual. If sexuality is a core component of the production of gender identity, gender images, and gender inequality, organizational theory that is blind to sexuality does not immediately offer avenues into the comprehension of gender domination (Hearn and Parkin 1983, 1987). Catharine MacKinnon's (1982) compelling argument that sexual domination of women is embedded within legal organizations has not to date become part of mainstream discussions. Rather, behaviors such as sexual harassment are viewed as deviations of gendered actors, not, as MacKinnon (1979) might argue, as components of organizational structure. FEMINIST ANALYSES OF ORGANIZATIONS The treatment of women and gender most assimilated into the literature on organizations is Rosabeth Kanter's Men and Women of the Corporation (1977). Moss Kanter sets out to show that gender differences in organizational behavior are due to structure rather than to characteristics of women and men as individuals (1977, 291-92). She argues that the problems women have in large organizations are consequences of their structural placement, crowded in dead-end jobs at the bottom and exposed as tokens at the top. Gender enters the picture through organizational roles that "carry characteristic images of the kinds of people that should occupy them" (p. 250). Here, Moss Kanter recognizes the presence of gender in early models of organizations A "masculine ethic" of rationality and reason can be identified in the early image of managers. This "masculine ethic" elevates the traits assumed to belong to men with educational advantages to necessities for effective organizations: a tough-minded approach to problems; analytic abilities to abstract and plan; a capacity to set aside personal, emotional considerations in the interests of task accomplishment; a cognitive superiority in problem- solving and decision making. (1974, 43) Identifying the central problem of seeming gender neutrality, Moss Kanter observes: "While organizations were being defined as sex-neutral machines, masculine principles were dominating their authority structures" (1977, 46). In spite of these insights, organizational structure, not gender, is the focus of Moss Kanter's analysis. In posing the argument as structure or gender, Moss Kanter also implicitly posits gender as standing outside of structure, and she fails to follow up her own observations about masculinity and organizations (1977, 22). Moss Kanter's analysis of the effects of organizational position applies as well to men in low-status positions. Her analysis of the effect of numbers, or the situation of the "token" worker, applies also to men as minorities in women-predominant organizations, but fails to account for gender differences in the situation of the token. In contrast to the token woman, White men in women-dominated workplaces are likely to be positively evaluated and to be rapidly promoted to positions of greater authority. The specificity of male dominance is absent in Moss Kanter's argument, even though she presents a great deal of material that illuminates gender and male dominance. Another approach, using Moss Kanter's insights but building on the theoretical work of Hartmann (1976), is the argument that organizations have a dual structure, bureaucracy and patriarchy (Ressner 1987). Ressner argues that bureaucracy has its own dynamic, and gender enters through patriarchy, a more or less autonomous structure, that exists alongside the bureaucratic structure. The analysis of two hierarchies facilitates and clarifies the discussion of women's experiences of discrimination, exclusion, segregation, and low wages. However, this approach has all the problems of two systems theories of women's oppression (Young l981; see also Acker l988): the central theory of bureaucratic or organizational structure is unexamined, and patriarchy is added to allow the theorist to deal with women. Like Moss Kanter, Ressner's approach implicitly accepts the assumption of mainstream organizational theory that organizations are gender-neutral social phenomena. Ferguson, in The Feminist Case Against Bureaucracy (1984), develops a radical feminist critique of bureaucracy as an organization of oppressive male power, arguing that it is both mystified and constructed through an abstract discourse on rationality, rules, and procedures. Thus, in contrast to the implicit arguments of Moss Kanter and Ressner, Ferguson views bureaucracy itself as a construction of male domination. In response to this overwhelming organization of power, bureaucrats, workers, and clients are all "feminized," as they develop ways of managing their powerlessness that at the same time perpetuate their dependence. Ferguson argues further that feminist discourse, rooted in women's experiences of caring and nurturing outside bureaucracy's control, provides a ground for opposition to bureaucracy and for the development of alternative ways of organizing society. However, there are problems with Ferguson's theoretical formulation. Her argument that feminization is a metaphor for bureaucratization not only uses a stereotype of femininity as oppressed, weak, and passive, but also, by equating the experience of male and female clients, women workers, and male bureaucrats, obscures the specificity of women's experiences and the connections between masculinity and power (Brown 1984; see also Martin 1987; Mitchell 1986; Ressner 1986). Ferguson builds on Foucault's (1979) analysis of power as widely diffused and constituted through discourse, and the problems in her analysis have their origin in Foucault, who also fails to place gender in his analysis of power. What results is a disembodied, and consequently gender-neutral, bureaucracy as the oppressor. That is, of course, not a new vision of bureaucracy, but it is one in which gender enters only as analogy, rather than as a complex component of processes of control and domination. In sum, some of the best feminist attempts to theorize about gender and organizations have been trapped within the constraints of definitions of the theoretical domain that cast organizations as gender neutral and asexual. These theories take us only part of the way to understanding how deeply embedded gender is in organizations. There is ample empirical evidence: We know now that gender segregation is an amazingly persistent pattern and that the gender identity of jobs and occupations is repeatedly reproduced, often in new forms (Bielby and Baron 1987; Reskin and Roos 1987; Strobet and Arnold 1987). The reconstruction of gender segregation is an integral part of the dynamic of technological and organizational change (Cockburn 1983, 1985; Hacker 1981). Individual men and particular groups of men do not always win in these processes, but masculinity always seems to symbolize self-respect for men at the bottom and power for men at the top, while confirming for both their gender's superiority. Theories that posit organization and bureaucracy as gender neutral cannot adequately account for this continual gendered structuring. We need different theoretical strategies that examine organizations as gendered processes in which sexuality also plays a part. ORGANIZATION AS GENDERED PROCESSES The idea that social structure and social processes are gendered has slowly emerged in diverse areas of feminist discourse. Feminists have elaborated gender as a concept to mean more than a socially constructed, binary identity and image. This turn to gender as an analytic category (Connell 1987; Harding 1986; Scott 1986) is an attempt to find new avenues into the dense and complicated problem of explaining the extraordinary persistence through history and across societies of the subordination of women. Scott, for example, defines gender as follows: "The core of the definition rests on an integral connection between two propositions; gender is a constitutive element of social relationships based on perceived differences between the sexes, and gender is a primary way of signifying relationships of power" (1986, 1067). New approaches to the study of waged work, particularly studies of the labor process, see organizations as gendered, not as gender neutral (Cockburn 1985; Game and Pringle 1984; Knights and Willmott 1985; Phillips and Taylor 1986; Sorenson 1984) and conceptualize organizations as one of the locations of the inextricably intertwined production of both gender and class relations. Examining class and gender (Acker 1988), I have argued that class is constructed through gender and that class relations are always gendered. The structure of the labor market, relations in the workplace, the control of the work process, and the underlying wage relation are always affected by symbols of gender, processes of gender identity, and material inequalities between women and men. These processes are complexly related to and powerfully support the reproduction of the class structure. Here, I will focus on the interface of gender and organizations, assuming the simultaneous presence of class relations. To say that an organization, or any other analytic unit, is gendered means that advantage and disadvantage, exploitation and control, action and emotion, meaning and identity, are patterned through and in terms of a distinction between male and female, masculine and feminine. Gender is not an addition to ongoing processes, conceived as gender neutral. Rather, it is an integral part of those processes, which cannot be properly understood without an analysis of gender (Connell 1987; West and Zimmerman 1987). Gendering occurs in at least five interacting processes (cf. Scott 1986) that, although analytically distinct, are, in practice, parts of the same reality. First is the construction of divisions along lines of gender - divisions of labor, of allowed behaviors, of locations in physical space, of power, including the institutionalized means of maintaining the divisions in the structures of labor markets, the family, the state. Such divisions in work organizations are well documented (e.g., Moss Kanter 1977) as well as often obvious to casual observers. Although there are great variations in the patterns and extent of gender division, men are almost always in the highest positions of organizational power. Managers' decisions often initiate gender divisions (Cohn 1985), and organizational practices maintain them - although they also take on new forms with changes in technology and the labor process. For example, Cynthia Cockburn (1983, 1985) has shown how the introduction of new technology in a number of industries was accompanied by a reorganization, but not abolition, of the gendered division of labor that left the technology in men's control and maintained the definition of skilled work as men's work and unskilled work as women's work. Second is the construction of symbols and images that explain, express, reinforce, or sometimes oppose those divisions. These have many sources or forms in language, ideology, popular and high culture, dress, the press, television. For example, as Moss Kanter (1975), among others, has noted, the image of the top manager or the business leader is an image of successful, forceful masculinity (see also Lipman-Blumen 1980). In Cockburn's studies, men workers' images of masculinity linked their gender with their technical skills; the possibility that women might also obtain such skills represented a threat to that masculinity. The third set of processes that produce gendered social structures, including organizations, are interactions between women and men, women and women, men and men, including all those patterns that enact dominance and submission. For example, conversation analysis shows how gender differences in interruptions, turn taking, and setting the topic of discussion recreate gender inequality in the flow of ordinary talk (West and Zimmerman 1983). Although much of this research has used experimental groups, qualitative accounts of organizational life record the same phenomena: Men are the actors, women the emotional support (Hochschild 1983). Fourth, these processes help to produce gendered components of individual identity, which may include consciousness of the existence of the other three aspects of gender, such as, in organizations, choice of appropriate work, language use, clothing, and presentation of self as a gendered member of an organization (Reskin and Roos 1987). Finally, gender is implicated in the fundamental, ongoing processes of creating and conceptualizing social structures. Gender is obviously a basic constitutive element in family and kinship, but, less obviously, it helps to frame the underlying relations of other structures, including complex organizations. Gender is a constitutive element in organizational logic, or the underlying assumptions and practices that construct most contemporary work organizations (Clegg and Dunkerley 1980). Organizational logic appears to be gender neutral; gender-neutral theories of bureaucracy and organizations employ and give expression to this logic. However, underlying both academic theories and practical guides for managers is a gendered substructure that is reproduced daily in practical work activities and, some what less frequently, in the writings of organizational theorists. (cf. Smith 1988) Organizational logic has material forms in written work rules, labor contracts, managerial directives, and other documentary tools for running large organizations, including systems of job evaluation widely used in the comparable-worth strategy of feminists. Job evaluation is accomplished through the use and interpretation of documents that describe jobs and how they are to be evaluated. These documents contain symbolic indicators of structure; the ways that they are interpreted and talked about in the process of job evaluation reveals the underlying organizational logic. I base the following theoretical discussion on my observations of organizational logic in action in the job-evaluation component of a comparable-worth project (Acker 1987, 1989, 1990). Job evaluation is a management tool used in every industrial country, capitalist and socialist, to rationalize the organizational hierarchy and to help in setting equitable wages (International Labour Office 1986). Although there are many different systems of job evaluation, the underlying rationales are similar enough so that the observation of one system can provide a window into a common organizational mode of thinking and practice. In job evaluation, the content of jobs is described and jobs are compared on criteria of knowledge, skill, complexity, effort, and working conditions. The particular system I observed was built incrementally over many years to reflect the assessment of managers about the job components for which they were willing to pay. Thus today this system can be taken as composed of residues of these judgments, which arc a set of decision rules that, when followed, reproduce managerial values. But these rules are also the imagery out of which managers construct and reconstruct their organizations. The rules of job evaluation, which help to determine pay differences between jobs, are not simply a compilation of managers' values or sets of beliefs, but are the underlying logic or organization that provides at least part of the blueprint for its structure. Every time that job evaluation is used, that structure is created or reinforced. Job evaluation evaluates jobs, not their incumbents. The job is the basic unit in a work organization's hierarchy, a description of a set of tasks, competencies, and responsibilities represented as a position on an organizational chart. A job is separate from people. It is an empty slot, a reification that must continually be reconstructed, for positions exist only as scraps of paper until people fill them. The rationale for evaluating jobs as devoid of actual workers reveals further the organizational logic - the intent is to assess the characteristics of the job, not of their incumbents who may vary in skill, industriousness, and commitment. Human beings are to be motivated, managed, and chosen to fit the job. The job exists as a thing apart. Every job has a place in the hierarchy, another essential element in organizational logic. Hierarchies, like jobs, are devoid of actual workers and based on abstract differentiations. Hierarchy is taken for granted, only its particular form is at issue. Job evaluation is based on the assumption that workers in general see hierarchy as an acceptable principle, and the final test of the evaluation of any particular job is whether its place in the hierarchy looks reasonable. The ranking of jobs within an organization must make sense to managers, but it is also important that most workers accept the ranking as just if the system of evaluation is to contribute to orderly working relationships. Organizational logic assumes a congruence between responsibility, job complexity, and hierarchical position. For example, a lower-level position, the level of most jobs filled predominantly by women, must have equally low levels of complexity and responsibility. Complexity and responsibility are defined in terms of managerial and professional tasks. The child-care worker's responsibility for other human beings or the complexity facing the secretary who serves six different, temperamental bosses can only be minimally counted if the congruence between position level, responsibility, and complexity is to be preserved. In addition, the logic holds that two jobs at different hierarchical levels cannot be responsible for the same outcome; as a consequence, for example, tasks delegated to a secretary by a manager will not raise her hierarchical level because such tasks are still his responsibility, even though she has the practical responsibility to see that they are done. Levels of skill, complexity, and responsibility, all used in constructing hierarchy, are conceptualized as existing independently of any concrete worker. In organizational logic, both jobs and hierarchies are abstract categories that have no occupants, no human bodies, no gender. However, an abstract job can exist, can be transformed into a concrete instance, only if there is a worker. In organizational logic, filling the abstract job is a disembodied worker who exists only for the work. Such a hypothetical worker cannot have other imperatives of existence that impinge upon the job. At the very least, outside imperatives cannot be included within the definition of the job. Too many obligations outside the boundaries of the job would make a worker unsuited for the position. The closest the disembodied worker doing the abstract job comes to a real worker is the male worker whose life centers on his full-time, life-long job, while his wife or another woman takes care of his personal needs and his children. While the realities of life in industrial capitalism never allowed all men to live out this ideal, it was the goal for labor unions and the image of the worker in social and economic theory. The woman worker, assumed to have legitimate obligations other than those required by the job, did not fit with the abstract job. The concept "a job" is thus implicitly a gendered concept, even though organizational logic presents it as gender neutral. "A job" already contains the gender-based division of labor and the separation between the public and the private sphere. The concept of "a job" assumes a particular gendered organization of domestic life and social production. It is an example of what Dorothy Smith has called "the gender subtext of the rational and impersonal" (1988, 4). Hierarchies are gendered because they also are constructed on these underlying assumptions: Those who are committed to paid employment are "naturally" more suited to responsibility and authority; those who must divide their commitments are in the lower ranks. In addition, principles of hierarchy, as exemplified in most existing job-evaluation systems, have been derived from already existing gendered structures. The best-known systems were developed by management consultants working with managers to build methods of consistently evaluating jobs and rationalizing pay and job classifications. For example, all managers with similar levels of responsibility in the firm should have similar pay. Job-evaluation systems were intended to reflect the values of managers and to produce a believable ranking of jobs based on those values. Such rankings would not deviate substantially from rankings already in place that contain gender typing and gender segregation of jobs and the clustering of women workers in the lowest and the worst-paid jobs. The concrete value judgments that constitute conventional job evaluation are designed to replicate such structures (Acker 1989). Replication is achieved in many ways; for example, skills in managing money, more often found in men's than in women's jobs, frequently receive more points than skills in dealing with clients or human relations skills, more often found in women's than in men's jobs (Steinberg and Haignere 1987). The gender-neutral status of "a job" and of the organizational theories of which it is a part depend upon the assumption that the worker is abstract, disembodied, although in actuality both the concept of "a job" and real workers are deeply gendered and "bodied." Carole Pateman (1986), in a discussion of women and political theory, similarly points out that the most fundamental abstraction in the concept of liberal individualism is "the abstraction of the 'individual' from the body. In order for the individual to appear in liberal theory as a universal figure, who represents anyone and everyone, the individual must be disembodied" (p. 8). If the individual were not abstracted from bodily attributes, it would be clear that the individual represents one sex and one gender, not a universal being. The political fiction of the universal "individual" or "citizen," fundamental to ideas of democracy and contract, excluded women, judging them lacking in the capacities necessary for participation in civil society. Although women now have the rights of citizens in democratic states, they still stand in an ambiguous relationship to the universal individual who is "constructed from a male body so that his identity is always masculine" (Pateman 1988, 223). The worker with "a job" is the same universal "individual" who in actual social reality is a man. The concept of a universal worker excludes and marginalizes women who cannot, almost by definition, achieve the qualities of a real worker because to do so is to become like a man. ORGANIZATIONAL CONTROL, GENDER, AND THE BODY The abstract, bodiless worker, who occupies the abstract, gender-neutral job has no sexuality, no emotions, and does not procreate. The absence of sexuality, emotionality, and procreation in organizational logic and organizational theory is an additional element that both obscures and helps to reproduce the underlying gender relations. New work on sexuality in organizations (Hearn and Parkin 1987), often indebted to Foucault (1979), suggests that this silence on sexuality may have historical roots in the development of large, all-male organizations that are the primary locations of societal power (Connell 1987). The history of modern organizations includes, among other processes, the suppression of sexuality in the interests of organization and the conceptual exclusion of the body as a concrete living whole (Burrell 1984, 1987; Hearn and Parkin 1987; Morgan 1986). In a review of historical evidence on sexuality in early modern organizations, Burrell (1984, 98) suggests that "the suppression of sexuality is one of the first tasks the bureaucracy sets itself." Long before the emergence of the very large factory of the nineteenth century, other large organizations, such as armies and monasteries, which had allowed certain kinds of limited participation of women, were more and more excluding women and attempting to banish sexuality in the interests of control of members and the organization's activities (Burrell 1984, 1987; Hacker and Hacker 1987). Active sexuality was the enemy of orderly procedures, and excluding women from certain areas of activity may have been, at least in part, a way to control sexuality. As Burrell (1984) points out, the exclusion of women did not eliminate homosexuality, which has always been an element in the life of large all-male organizations, particularly if members spend all of their time in the organization. Insistence on heterosexuality or celibacy were ways to control homosexuality. But heterosexuality had to be practiced outside the organization, whether it was an army or a capitalist workplace. Thus the attempts to banish sexuality from the workplace were part of the wider process that differentiated the home, the location of legitimate sexual activity, from the place of capitalist production. The concept of the disembodied job symbolizes this separation of work and sexuality. Similarly, there is no place within the disembodied job or the gender- neutral organization for other "bodied" processes, such as human reproduction (Rothman 1989) or the free expression of emotions (Hochschild 1983). Sexuality, procreation, and emotions all intrude upon and disrupt the ideal functioning of the organization, which tries to control such interferences. However, as argued above, the abstract worker is actually a man, and it is the man's body, its sexuality, minimal responsibility in procreation, and conventional control of emotions that pervades work and organizational processes. Women's bodies - female sexuality, their ability to procreate and their pregnancy, breast-feeding, and child care, menstruation, and mythic "emotionality" - are suspect, stigmatized, and used as grounds for control and exclusion. The ranking of women's jobs is often justified on the basis of women's identification with childbearing and domestic life. They are devalued because women are assumed to be unable to conform to the demands of the abstract job. Gender segregation at work is also sometimes openly justified by the necessity to control sexuality, and women may be barred from types of work, such as skilled blue-collar work or top management, where most workers are men, on the grounds that potentially disruptive sexual liaisons should be avoided (Lorber 1984). On the other hand, the gendered definition of some jobs "includes sexualization of the woman worker as a part of the job" (MacKinnon 1979, 18). Those are often jobs that serve men, such as secretaries, or a largely male public (Hochschild 1983). The maintenance of gendered hierarchy is achieved partly through such often-tacit controls based on arguments about women's reproduction, emotionality, and sexuality, helping to legitimate the organizational structures created through abstract, intellectualized techniques. More overt controls, such as sexual harassment, relegating childbearing women to lower- level mobility tracks, and penalizing (or rewarding) their emotion management also conform to and reinforce hierarchy. MacKinnon (1979), on the basis of an extensive analysis of legal cases, argues that the willingness to tolerate sexual harassment is often a condition of the job, both a consequence and a cause of gender hierarchy. While women's bodies are ruled out of order, or sexualized and objectified, in work organizations, men's bodies are not. Indeed, male sexual imagery pervades organizational metaphors and language, helping to give form to work activities (see Hearn and Parkin 1987, for an extended discussion). For example, the military and the male world of sports are considered valuable training for organizational success and provide images for teamwork, campaigns, and tough competition. The symbolic expression of male sexuality may be used as a means of control over male workers, too, allowed or even encouraged with in the bounds of the work situation to create cohesion or alleviate stress (Collinson 1988; Hearn and Parkin 1987). Management approval of pornographic pictures in the locker room or support for all-male work and play groups where casual talk is about sexual exploits or sports are examples. These symbolic expressions of male dominance also act as significant controls over women in work organizations because they are per se excluded from the informal bonding men produce with the "body talk" of sex and sports. Symbolically, a certain kind of male heterosexual sexuality plays an important part in legitimating organizational power. Connell (1987) calls this hegemonic masculinity, emphasizing that it is formed around dominance over women and in opposition to other masculinities, although its exact content changes as historical conditions change. Currently, hegemonic masculinity is typified by the image of the strong, technically competent, authoritative leader who is sexually potent and attractive, has a family, and has his emotions under control. Images of male sexual function and patriarchal paternalism may also be embedded in notions of what the manager does when he leads his organization (Calas and Smircich 1989). Women's bodies cannot be adapted to hegemonic masculinity; to function at the top of male hierarchies requires that women render irrelevant everything that makes them women. The image of the masculine organizational leader could be expanded, without altering its basic elements, to include other qualities also needed, according to many management experts, in contemporary organizations, such as flexibility and sensitivity to the capacities and needs of subordinates. Such qualities are not necessarily the symbolic monopoly of women. For example, the wise and experienced coach is empathetic and supportive to his individual players and flexibly leads his team against devious opposition tactics to victory. The connections between organizational power and men's sexuality may be even more deeply embedded in organizational processes. Sally Hacker (1989) argues that eroticism and technology have common roots in human sensual pleasure and that for the engineer or the skilled worker, and probably for many other kinds of workers, there is a powerful erotic element in work processes. The pleasures of technology, Hacker continues, become harnessed to domination, and passion becomes directed toward power over nature, the machine, and other people, particularly women, in the work hierarchy. Hacker believes that men lose a great deal in this transformation of the erotic into domination, but they also win in other ways. For example, many men gain economically from the organizational gender hierarchy. As Crompton and Jones (1984) point out, men's career opportunities in white-collar work depend on the barriers that deny those opportunities to women. If the mass of female clerical workers were able to compete with men in such work, promotion probabilities for men would be drastically reduced. Class relations as well as gender relations are reproduced in organizations. Critical, but nonfeminist, perspectives on work organizations argue that rational-technical systems for organizing work, such as job classification and evaluation systems and detailed specification of how work is to be done, are parts of pervasive systems of control that help to maintain class relations (Edwards 1979). The abstract "job," devoid of a human body, is a basic unit in such systems of control. The positing of a job as an abstract category, separate from the worker, is an essential move in creating jobs as mechanisms of compulsion and control over work processes. Rational- technical, ostensibly gender-neutral, control systems are built upon and conceal a gendered substructure (Smith 1988) in which men's bodies fill the abstract jobs. Use of such abstract systems continually reproduces the underlying gender assumptions and the subordinated or excluded place of women. Gender processes, including the manipulation and management of women's and men's sexuality, procreation, and emotion, are part of the control processes of organizations, maintaining not only gender stratification but contributing also to maintaining class and, possibly, race and ethnic relations. Is the abstract worker white as well as male? Are white-male-dominated organizations also built on underlying assumptions about the proper place of people with different skin colors? Are racial differences produced by organizational practices as gender differences are? CONCLUSION Feminists wanting to theorize about organizations face a difficult task because of the deeply embedded gendering of both organizational processes and theory. Common sense notions, such as jobs and positions, which constitute the units managers use in making organizations and some theorists use in making theory, are posited upon the prior exclusion of women. This underlying construction of a way of thinking is not simply an error, but part of processes of organization. This exclusion in turn creates fundamental inadequacies in theorizing about gender-neutral systems of positions to be filled. Creating more adequate theory may come only as organizations are transformed in ways that dissolve the concept of the abstract job and restore the absent female body. Such a transformation would be radical in practice because it would probably require the end of organizations as they exist today, along with a redefinition of work and work relations. The rhythm and timing of work would be adapted to the rhythms of life outside of work. Caring work would be just as important and well rewarded as any other; having a baby or taking care of a sick mother would be as valued as making an automobile or designing computer software. Hierarchy would be abolished, and workers would run things themselves. Of course, women and men would share equally in different kinds of work. Perhaps there would be some communal or collective form of organization where work and intimate relations are closely related, children learn in places close to working adults, and workmates, lovers, and friends are all part of the same group. Utopian writers and experimenters have left us many possible models (Hacker 1989). But this brief listing begs many questions, perhaps the most important of which I show, given the present organization of economy and technology and the pervasive and powerful, impersonal, textually mediated relations of ruling (Smith 1988), so radical a change could come about. Feminist research and theorizing, by continuing to puzzle out how gender provides the subtext for arrangements of subordination, can make some contributions to a future in which collective action to do what needs doing producing goods, caring for people, disposing of the garbage - is organized so that dominance, control, and subordination, particularly the subordination of women, are eradicated, or at least minimized, in our organization life. REFERENCES Acker, Joan. 1982. Introduction to women, work and economic democracy. Economic and Industrial Democracy 3(4):i-viii. (). 1987. Sex bias in job evaluation: A comparable worth issue. In Ingredients for women's employment policy, edited by Christine Bose and Glenna Spitze. Albany: SUNY Press. (). 1988. Class, gender and the relations of distribution. Signs 13:473-97. (). 1989. Doing comparable worth: Gender, class and pay equity. Philadelphia: Temple University Press. (). 1990. The Oregon case. In State experience and comparable worth, edited by Ronnie Steinberg. Philadelphia: Temple University Press. Acker, Joan, and Donald Van Houten. 1974. Differential recruitment and control: The sex structuring of organizations. Administrative Science Quarterly 19:152-63. Bielby, William T., and James N. Baron. 1987. Undoing discrimination: Job integrations and comparable worth. In Ingredients for women 's employment policy, edited by Christine Bose and Glenna Spitze. Albany: SUNY Press. Braverman, Harty. 1974. Labor and monopoly capital. New York: Monthly Review Press. Brown, Wendy. 1984. Challenging bureaucracy. Women's Review of Books 2 (November):14-17. Burrell, Gibson. 1984. Sex and organizational analysis. Organization Studies 5:97-118. (). 1987. No accounting for sexuality. Accounting, Organizations and Society 12:89-101. Calas, Mana B., and Linda Smircich. 1989. Voicing seduction to silence leadership. Paper presented at the Fourth International Conference on Organizational Symbolism and Corporate Culture in Fountainbleau, France. Clegg, Stewan, and David Dunkerley. 1980. Organization, class and control. London: Routledge & Kegan Paul. Cockburn, Cynthia. 1983. Brothers: Male dominance and technological change. London: Pluto Press. (). 1985. Machinery of dominance. London: Pluto Press. Cohn, Samuel. 1985. The process of occupational sex-typing. Philadelphia: Temple University Press. Collinson, David L. 1988. Engineering humour: Masculinity, joking and conflict in shop-floor relations. Organization Studies 9: 181-99. Connell, R. W. 1987. Gender and power. Stanford, CA: Stanford University Press. Crompton, Rosemary, and Gareth Jones. 1984. White-collar proletariat: deskilling and gender in clerical work. Philadelphia: Temple University Press. Dexter, Carolyn R. 1985. Women and the exercise of power in organizations: From ascribed to achieved status. In Women and work: An annual review, Vol. I, edited by Laurie Larwood, Ann H. Stromberg, and Barbara A. Gutek. Beverly Hills, CA: Sage. Edwards, Richard. 1979. Contested terrain. New York: Basic Books. Eisenstein, Zillah R. 1981. The radical future of liberal feminism. New York: Longman. Feldberg, Roslyn, and Evelyn Nakano Glenn. 1979. Male and female: Job versus gender models in the sociology of work. Social Problems 26:524-38. Ferguson, Kathy E. 1984. The feminist case against bureaucracy. Philadelphia: Temple University Press. Foucault, Michel. 1979. The history of sexuality, Vol. I. London: Allen Lane. Freeman, Jo. 1975. The politics of women's liberation. New York: Longman. Fry, John. 1986. Toward a democratic rationality. Aldershot: Gower. Game, Ann and Rosemary Pringle. 1984. Gender at work. London: Pluto Press. Gould, Meredith. 1979. When women create an organization: The ideological imperatives of feminism. In The international yearbook of women's studies 1979, edited by D. Dunkerley and G. Salaman. London: Routledge & Kegan Paul. Hacker, Barton C., and Sally Haeckel 1987. Military institutions and the labor process: Noneconomic sources of technological change, women's subordination, and the organization of work. Technology and Culture 28:743-75. Hacker, Sally. 1981. The culture of engineering women: Women, workplace and machine. Women's Studies International Quarterly 4:341-54. (). 1987. Women workers in the Mondragon system of industrial cooperatives. Gender & Society I :358-79. (). 1989. Pleasure, power and technology. Boston: Unwin Hyman. Harding, Sandra. 1986. The science question in feminism. Ithaca, NY: Cornell University Press. Hartmann, Heidi. 1976. Capitalism, patriarchy and job segregation by sex. Signs 1:137-70. Heam, Jeff, and P. Wendy Parkin. 1983. Gender and organizations: A selective review and critique of a neglected area. Organization Studies 4:219- 42. (). 1987. Sex at work. Brighton: Wheatsheaf. Hochschild, Arlie R. 1983. The managed heart: Commercialization of human feeling. Berkeley: University of California Press. International Labour Office. 1986. Job evaluation Geneva: ILO. Izraeli, Dafna N. 1983. Sex effects or structural effects? An empirical test of Kanter's theory of proportions. Social Forces 61:153-65. Kanter, Rosabeth Moss. 1975. Women and the structure of organizations: Explorations in theory and behavior. In Another voice, edited by Rosabeth Kaoter and Marcia Millman. New York Doubleday. (). 1977. Men and women of the corporation. New York: Basic Books. Knights, David, and Hugh Willmott. 1985. Gender and the labour process. Aldershot: Gower. Lipman-Blumen, Jean. 1980. Female leadership informal organizations: Must the female leader go formal? In Readings in managerial psychology, edited by Harold J. Leavitt, Louis R. Pondy, and David M. Boje. Chicago: University of Chicago Press. Lorber, Judith. 1984. Trust, loyalty, and the place of women in the organization of work. In Women: A feminist perspective, 3d ed., edited by Jo Freeman. Palo Alto, CA: Mayfield. MacKinnon, Catharine A. 1979. Sexual harassment of working women. New Haven, CT: Yale University Press. (). 1982. Feminism, Marxism, method and the state: An agenda for theory, Signs 7:515-44. Manin, Patricia Yancey. 1985. Group sex composition in work organizations: A structural-normative view. In Research in the sociology of organizations, edited by S. A. Bacharach and R. Mitchell. Greenwich, CT: JAI. (). 1987. A commentary on The feminist case against bureaucracy. Women's Studies International Forum 10:543-48. (). 1990. Rethinking feminist organizations. Gender & Society 4:182-206. Mills, Alben J. 1988. Organization, gender and culture. Organization Studies 9(3):351-69. Mitchell, Diane. 1986. Review of Ferguson, The feminist case against bureaucracy. Unpublished manuscript. Morgan, Gareth. 1986. Images of organization. Beverly Hills, CA: Sage. Newman, Katherine. 1980. Incipient bureaucracy: The development of hierarchies in egalitarian organizations. In Hierarchy and society, edited by Gerald R. Britan and Ronald Cohen. Philadelphia: Institute for the Study of Human Issues. Pateman, Carole. 1970. Participation and democratic theory. Cambridge: Cambridge University Press. (). 1983a. Feminist critiques of the public private dichotomy. In Public and private in social life, edited by S. I. Benn and G. F. Gaus. Beckenham, Kent: Cronm Helm. (). 1983b. Feminism and democracy. In Democratic theory and practice, edited by Graeme Duncan. Cambridge: Cambridge University Press. (). 1986. Introduction: The theoretical subversiveness of feminism, In Feminist challenges, edited by Carole Pateman and Elizabeth Gross. Winchester, MA: Allen & Unwin. (). 1988. The sexual contract. Cambridge, MA: Polity. Phillips, Anne, and Barbara Taylor. 1986. Sex and skill. In Waged work edited by Feminist Review. London: Virago. Reskin, Barbara F., and Patricia Roos. 1987. Status hierarchies and sex segregation. In Ingredients for women's employment policy, edited by Christine Bose and Glenna Spitze. Albany: SUNY Press. Ressner, Ulla. 1986. Review of K. Ferguson, The feminist case against bureaucracy. Economic and Industrial Democracy 7:130-34. (). 1987. The hidden hierarchy. Aldershot: Gower. Rothman, Barbara Katz. 1989. Recreating motherhood: Ideology and technology in a patriarchal society. New York: Norton. Rothschild, Joyce. 1987. Do collectivist-democratic forms of organization presuppose feminism? Cooperative work structures and women's values. Paper presented at Annual Meetings, American Sociological Association, Chicago. Rothschild-Whitt, Joyce. 1979. The collectivist organization. American Sociological Review 44:509-27. Scott, Joan. 1986. Gender: A useful category of historical analysis. American Historical Review 91:1053-75. Smith, Dorothy E. I 979. A sociology for women. In The prism of sex: Essays in the sociology of knowledge, edited by Julia A. Sherman and Evelyn Torten Beck. Madison: University of Wisconsin Press. (). 1988. The everyday world as problematic. Boston: Northeastern University Press. Sorenson, Bjorg Aase. 1984. The organizational woman and the Trojan horse effect. In Patriarchy in a welfare society, edited by Harriet Holter. Oslo: Universitetsforlaget. Steinberg, Ronnie, and Lois Haignere. 1987. Equitable compensation: Methodological criteria for comparable worth. In Ingredients for women 's employment policy, edited by Christine Bose and Glenna Spitze. Albany: SUNY Press. Strober, Myra H., and Carolyn L. Arnold. 1987. Integrated circuits/segregated labor: Women in computer-related occupations and high-tech industries. In Computer chips and paper clips: Technology and women's employment, edited by H. Hartmann. Washington, DC: National Academy Press. Wallace, Phyllis A., ed. 1982. Women in the workplace. Boston: Auburn House. West, Candace, and Don H. Zimmerman. 1983. Small insults: A study of interruptions in conversations between unacquainted persons. In Language, gender and society, edited by B. Thorne, C. Kramerae, and N. Henley. Rowley, MA: Newbury House. (). 1987. Doing gender. Gender & Society 1:125-51. Young, Iris. 1981. Beyond the unhappy marriage: A critique of dual systems theory. In Women and revolution, edited by L. Sargent. Boston: South End Press. Joan Acker wrote this article while she was a visiting professor at The Swedish Center for Working Life. She has also recently published Doing Comparable Worth: Gender, Class, and Pay Equity (Philadelphia: Temple University Press, 1989), "Class, Gender, and the Relations of Distribution," Signs 13 (1988), and "Making Gender Visible," in Feminism and Sociological Theory, edited by Ruth A. Wallace (Newbury Park CA: Sage, 1989).