Men and Women of the Corporation

By Kanter, Rosabeth Moss.

men + Women cover

Chapter 8


Numbers: Minorities and Majorities

The token woman stands in the Square of the Immaculate Excep­tion blessing pigeons from a blue pedestal. . The token woman is placed like a scarecrow in the long haired corn: her muscles are wooden. Why does she ride into battle on a clothes horse? -Marge Piercy, “Living in the Open” [1]

Up the ranks in Industrial Supply Corporation (Indesco), one of the most consequential conditions of work for women was also among the simplest to identify: there were so few of them. On the professional and managerial levels, Industrial Supply Corporation was nearly a single-sex organization. Women held less than 10 percent of the exempt (salaried) jobs starting at the bottom grades-a 50 percent rise from a few years earlier-and there were no women at the level reporting to officers. When Indesco was asked to participate in a meeting on women in business by bringing their women executives to a civic luncheon, the corporate personnel committee had no difficulty selecting them. There were only five sufficiently senior women in the organization.

The numerical distributions of men and women at the upper reaches created a strikingly different interaction context for women than for men. At local and regional meetings, training programs, task forces, casual out-of-the office lunches with colleagues, and career review or planning sessions with managers, the men were overwhelmingly likely to find themselves with a pre­dominance of people of their own type -other men. For men in units with no exempt women, there would be, at most, occasional events in which a handful of women would be present alongside many men. Quite apart from the con­tent of particular jobs and their location in the hierarchy, the culture of cor­porate administration and the experiences of men in it were influenced by this fact of numerical dominance, by the fact that men were the many.

Women, on the other hand, often found themselves alone among male peers. The twenty women in a three hundred-person sales force were scat­tered over fourteen offices. Their peers, managers, and customers were nearly all men. Never more than two women at a time were found in twelve-person personnel training groups. There was a cluster of professional women on the floor at corporate headquarters housing employee administration and training, but all except three were part of different groups where they worked most closely with men.

The life of women in the corporation was influenced by the proportions in which they found themselves. Those women who were few in number among male peers and often had “only woman” status became tokens: symbols of how-women-can-do, stand-ins for all women. Sometimes they had the advan­tages of those who are “different” and thus were highly visible in a system where success is tied to becoming known. Sometimes they faced the loneli­ness of the outsider, of the stranger who intrudes upon an alien culture and may become self-estranged in the process of assimilation. In any case, their turnover and “failure rate” were known to be much higher than those of men in entry and early grade positions; in the sales function, women’s turnover was twice that of men. What happened around Indesco women resembled other reports of the experiences of women in politics, law, medicine, or manage­ment who have been the few among many men.

At the same time, they also echoed the experiences of people of any kind who are rare and scarce: the lone black among whites, the lone man among women, the few foreigners among natives. Any situation where proportions of significant types of people are highly skewed can produce similar themes and processes. It was rarity and scarcity, rather than femaleness per se, that shaped the environment for women in the parts of Indesco mostly populated by men.

The situations of Industrial Supply Corporation men and women, then, point to the significance of numerical distributions for behavior in organiza­tions: how many of one social type are found with how many of another. [1] As proportions begin to shift, so do social experiences.





Georg Simmers classic analysis of the significance of numbers for social life argued persuasively that numerical shifts transform social interaction, as in the differences between two-person and three-person situations or between small and large groups. [2] But Simmel, and then later investigations in this tradition, dealt almost exclusively with the impact of absolute numbers, with group size as a determinant of form and process. We have no vocabulary for dealing with the effects of relative numbers, of proportional representation: the difference for individuals and groups that stem from particular numerical distributions of categories of people.

Yet questions of how many and how few confound any statements about the organizational behavior of special kinds of people. For example, certain popular conclusions and research findings about male-female relations or role potentials may turn critically on the issue of proportions. One study of mock jury deliberations found that men played proactive, task-oriented leadership roles, whereas women in the same groups tended to take reactive, emotional, and nurturant postures-supposed proof that traditional stereotypes reflect behavior realities. But, strikingly, men far outnumbered women in all of the groups studied. Perhaps it was the women’s scarcity that pushed them into classical positions and the men’s numerical superiority that encouraged them to assert task superiority. Similarly, the early kibbutzim, collective villages in Israel that theoretically espoused equality of the, sexes but were unable to fully implement it, could push women into traditional service positions because there were more than twice as many men as women. Again, relative numbers interfered with a fair test of what men or women can “naturally” do, as it did in the case of the relatively few women in the upper levels of Indesco. Indeed, recently Marcia Guttentag has found sex ratios in the population in general to be so important that they predict a large number of behavioral phenomena, from the degree of power women and men feel to the ways they cope with the economic and sexual aspects of their lives. [3]

To understand the dramas of the many and the few in the organization requires a theory and a vocabulary. Four group types can be identified on the basis of different proportional representations of kinds of people, as Figure 8-1 shows. Uniform groups have only one kind of person, one significant social type. The group may develop its own differentiations, of course, but groups called uniform can be considered homogeneous with respect to salient exter­nal master statuses such as sex, race, or ethnicity. Uniform groups have a typological ratio of 100:0. Skewed groups are those in which there is a large preponderance of one type over another, up to a ratio of perhaps 85:15. The numerically dominant types also control the group and its culture in enough ways to be labeled “dominants.” The few of another type in a skewed group can appropriately be called “tokens,” for, like the Indesco exempt women, they are often treated as representatives of their category, as symbols rather than individuals. If the absolute size of the skewed group is small, tokens can also be solos, the only one of their kind present; but even if there are two tokens in a skewed group, it is difficult for them to generate an alliance that can become powerful in the group, as we shall see later. Next, tilted groups begin to move toward less extreme distributions and less exaggerated effects. In this situation, with ratios of perhaps 65:35, dominants are just a “majority” and tokens become a “minority.” Minority members have potential allies among each other, can form coalitions, and can affect the culture of the group. They begin to become individuals differentiated from each other as well as a type differentiated from the majority. Finally, at about 60:40 and down to 50:50, the group becomes balanced. Culture and interaction reflect this balance. Majority and minority turn into potential subgroups that mayor may not generate actual type-based identifications. Outcomes for individuals in such a balanced peer group, regardless of type, will depend more on other structural and personal factors, including formation of subgroups or differen­tiated roles and abilities.


It is the characteristics of the second type, the skewed group, that un­derlay the behavior and treatment of professional and managerial women ob­served at Indesco. If the ratio of women to men in various parts of the organiza­tion begins to shift, as affirmative action and new hiring and promotion policies promised, forms of relationships and peer culture should also change. But as of the mid-1970S, the dynamics of tokenism predominated in Indesco’s exempt ranks, and women and men were in the positions of token and domi­nant. Tokenism, like low opportunity and low power, set in motion self-perpetuating cycles that served to reinforce the low numbers of women and, in the absence of external intervention, to keep women in the position of token.




The proportional rarity of tokens is associated with three perceptual ten­dencies: visibility, contrast, and assimilation. These are all derived simply from the ways any set of objects are perceived. If one sees nine X’s and one 0:

x X x x X X O X x X


The O will stand out. The O may also be overlooked, but if it is seen at all, it will get more notice than any X. Further, the X’s may seem more alike than dif­ferent because of their contrast with the O. And it will be easier to assimilate the O to generalizations about all O’ s than to do the same with the X’s, which offer more examples and thus, perhaps, more variety and individuation. The same perceptual factors operate in social situations, and they generate special pressures for token women.

First, tokens get attention. One by one, they have higher visibility than dominants looked at alone; they capture a larger awareness share. A group member’s awareness share, averaged over other individuals of the same social type, declines as the proportion of total membership occupied by the category increases, because each individual becomes less and less surprising, unique, or noteworthy. In Gestalt psychology terms, those who get to be common more easily become “ground” rather than “figure”; as the group moves from skewed to tilted, tokens turn into a less individually noticed minority. But for tokens, there is a “law of increasing returns”: as individuals of their type repre­sent a smaller numerical proportion of the overall group, they each potentially capture a larger share of the awareness given to that group.

Contrast- or polarization and exaggeration of differences-is the second perceptual tendency. In uniform groups, members and observers may never become self-conscious about the common culture and type, which remain taken for granted and implicit. But the presence of a person or two bearing a different set of social characteristics increases the self-consciousness of the numerically dominant population and the consciousness of observers about what makes the dominants a class. They become more aware both of their commonalities and their difference from the token, and to preserve their commonality, they try to keep the token slightly outside, to offer a boundary for the dominants. There is a tendency to exaggerate the extent of the differences between tokens and dominants, because as we see next, tokens are, by defini­tion, too few in numbers to defeat any attempts at generalization. It is thus easier for the commonalities of dominants to be defined in contrast to the token than in tilted or balanced groups. One person can be perceptually isolated and seen as cut off from the core of the group more than many, who begin to represent too great a share of what is called the group.

Assimilation, the third perceptual tendency, involves the use of stereo­types, or familiar generalizations about a person’s social type. The character­istics of a token tend to be distorted to fit the generalization. Tokens are more easily stereotyped than people found in greater proportion. If there were enough people of the token’s type to let discrepant examples occur, it is even­tually possible that the generalization would change to accommodate the ac­cumulated cases. But in skewed groups, it is easier to retain the generalization and distort the perception of the token. It is also easier for tokens to find an in­stant identity by conforming to the preexisting stereotypes. So tokens are, ironically, both highly visible as people who are different and yet not permit­ted the individuality of their own unique, non-stereotypical characteristics.

All of these phenomena occurred around the proportionally scarce women in Indesco, but there was, of course, no way to compare these same women’s behavior and treatment when they were not in the token position. However, a clever and suggestive laboratory experiment showed that the same person may be perceived differently depending on whether he or she is a token in a skewed group or one of many in a balanced group. (Because the categories used in the experiment were black-white rather than male-female, it also demonstrated the generality of such perceptual tendencies beyond token women.) Shelley Taylor and Susan Fiske played a tape of a group discus­sion to subjects while showing them pictures of the “group,” and then asked them for their impressions of group members on a number of dimensions. The tape was the same for all subjects, but the purported composition of the group varied. The pictures illustrated either an otherwise all-white male group with one black man (the “token” condition) or a mixed black-white male group. In the token condition, disproportionate attention was paid to the token, his prominence in the group was overemphasized, and his attributes were exag­gerated. Similarly, the token was perceived as playing out special roles in the group, often highly stereotypical ones. By contrast, in “integrated” groups, subjects recalled no more about blacks than whites, and their attributes were evaluated about the same. [4]

Visibility, contrast, and assimilation are each associated with particular forces and dynamics that, in turn, generate typical token responses. These dy­namics are, again, similar regardless of the category from which the tokens come, although the specific kinds of people and their history of relationships with dominants provide cultural content for specific communications. Visibil­ity tends to create performance pressures on t4e token. Contrast leads to heightening of dominant culture boundaries, including isolation of the token. And assimilation results in the token’s role encapsulation. The experiences of exempt women at Industrial Supply Corporation took  their shape from these processes.



Indesco’s upper-level women, especially those in sales, were highly visi­ble, much more so than their male peers. Even those who reported they felt ignored and overlooked were known in their immediate divisions and spotted when they did something unusual. But the ones who felt ignored also seemed to be those in jobs not enmeshed in the interpersonal structure of the com­pany: for example, a woman in public relations who had only a clerical assis­tant reporting to her and whose job did not occupy a space in the competitive race to the top.

In the sales force, where peer culture and informal relations were most strongly entrenched, everyone knew about the women. They were the subject of conversation, questioning, gossip, and careful scrutiny. Their placements were known and observed through the division, whereas those of most men typically were not. Their names came up at meetings, and they would easily be used as examples. Travelers to locations with women in it would bring back news of the latest about the women, along with other gossip. In other func­tions, too, the women developed well-known names, and their characteristics would often be broadcast through the system in anticipation of their arrival in another office to do a piece of work. A woman swore in an elevator in an Atlanta hotel while going to have drinks with colleagues, and it was known all over Chicago a few days later that she was a “radical.” And some women were even told by their managers that they were watched more closely than the men. Sometimes the manager was intending to be helpful, to let the woman know that he would be right there behind her. But the net effect was the same as all of the visibility phenomena. Tokens typically performed their jobs under public and symbolic conditions different from those of dominants.

The Two-Edged Sword of Publicity

The upper-level women became public creatures. It was difficult for them to do anything in training programs, on their jobs, or even at informal social affairs that would not attract public notice. This provided the advantage of an attention-getting edge at the same time that it made privacy and ano­nymity impossible. A saleswoman reported: ”I’ve been at sales meetings where all the trainees were going up to the managers-‘Hi, Mr. So-and-So’ ­trying to make that impression, wearing a strawberry tie, whatever, some­thing that they could be remembered by. Whereas there were three of us [women] in a group of fifty, and all we had to do was walk in and everyone recognized us. [5]

But their mistakes or their intimate relationships were known as readily as other information. Many felt their freedom of action was restricted, and they would have preferred to be less noticeable, as these typical comments in­dicated: “If it seems good to be noticed, wait until you make your first major mistake.” “It’s a burden for the manager who gets asked about a woman and has to answer behind-the-back stuff about her. It doesn’t reach the woman unless he tells her. The manager gets it and has to deal with it.” “I don’t have as much freedom of behavior as men do; I can’t be as independent.”

On some occasions, tokens were deliberately thrust into the limelight and displayed as showpieces, paraded before the corporation’s public but in ways that sometimes violated the women’s sense of personal dignity. One of Indesco’s most senior women, a staff manager finally given two assistants (and thus managerial responsibilities) after twenty-six years with the company, was among the five women celebrated at the civic lunch for outstanding women in business. A series of calls from high-level officers indicated that the chairman of the board of the corporation wanted her to attend a lunch at a large hotel that day, although she was given no information about the nature of the event. When she threatened not to go unless she was given more information, she was reminded that the invitation had come down from the chairman himself, and of course she would go. On the day of the luncheon, a corsage arrived and, later, a vice-president to escort her. So she went, and found she was there to represent the corporation’s “prize women,” symbolizing the strides made by women in business. The program for the affair listed the women executives from participating companies, except in the case of Indesco, where the male vice-presidential escorts were listed instead. Pictures were taken for the em­ployee newsletter and, a few days later, she received an inscribed paper­weight as a memento. She told the story a few weeks after the event with visi­ble embarrassment about being “taken on a date. It was more like a senior prom than a business event.” And she expressed resentment at being singled out in such a fashion, “just for being a woman at Indesco, not for any real achievement.” Similar sentiments were expressed by a woman personnel manager who wanted a pay increase as a sign of the company’s appreciation, not her picture in a newspaper, which “gave the company brownie points but cost nothing.”

Yet the senior woman had to go, the personnel manager had to have her picture taken, and they had to be gracious and grateful. The reaction of tokens to their notice was also noticed. Many of the tokens seemed to have developed a capacity often observed among marginal or subordinate peoples: to project a public persona that hid inner feelings. Although some junior management men at Indesco, including several fast trackers, were quite open about their lack of commitment to the company and dissatisfaction with aspects of its style, the women felt they could not afford to voice any negative sentiments. They played by a different set of rules, one that maintained the split between public persona and private self. One woman commented, “I know the company’s a rumor factory. You must be careful how you conduct yourself and what you say to whom. I saw how one woman in the office was discussed endlessly, and I decided it would be better to keep my personal life and personal affairs sepa­rate.” She refused to bring dates to office parties when she was single, and she did not tell anyone at work that she got married until several months later­ this was an office where the involvement of wives was routine. Because the glare of publicity meant that no private information could be kept cir­cumscribed or routine, tokens were forced into the position of keeping secrets and carefully contriving a public performance. They could not afford to stumble.

Symbolic Consequences

The women were visible as category members, because of their social type. This loaded all of their acts with extra symbolic consequences and gave them the burden of representing their category, not just themselves. Some women were told outright that their performances could affect the prospects of other women in the company. In the men’s informal conversations, women were often measured by two yardsticks: how as women they carried out the sales or management role; and how as managers they lived up to Images of womanhood. In short, every act tended to be evaluated beyond its meaning for the organization and taken as a sign of “how women perform.” This meant that there was a tendency for problematic situations to be blamed on the woman–on her category membership–rather than on the situation, a phe­nomenon noted in other reports of few women among many men in high-rank­ing corporate jobs. In one case of victim-blaming, a woman in sales went to her manager to discuss the handling of a customer who was behaving seductively. The manager jumped to the assumption that the woman had led him on. The result was an angry confrontation between woman and manager in which she thought he was incapable of seeing her apart from his stereotypes, and he said later he felt misunderstood.

Women were treated as symbols or representatives on those occasions when, regardless of their expertise or interest, they would be asked to provide the meeting with “the woman’s point of view” or to explain to a manager why he was having certain problems with his women. They were often expected to be speaking for women, not just for themselves, and felt, even in my inter­views, that they must preface personal statements with a disclaimer that they were speaking for themselves rather than for women generally. Such individ­uality was difficult to find when among dominants. But this was not always generated by dominants. Some women seized this chance to be a symbol as an opportunity to get included in particular gatherings or task forces, where they could come to represent all women at Indesco. “Even if you don’t want me per­sonally,” they seemed to be saying to dominants, “you can want me as a sym­bol.” Yet, if they did this, they would always be left with uncertainty about the grounds for their inclusion; they were failing to distinguish themselves as individuals.

Women also added symbolic consequences to each other’s affairs. Upper-­level women were scrutinized by those on a lower level, who discussed the merits of things done by the higher-ranking women and considered them to have implications for their own careers. One woman manager who was passed over for a promotion in her department was the subject of considerable discus­sion by other women, who felt she should have pushed to get the opening and complained when she did not.

The extension of consequences for those in token statuses may increase their self-consciousness about their self-presentation and about their deci­sions, and can change the nature of the decisions that get made. Decisions about what to wear and who to sit with at lunch are not casual. One executive woman knew that her clothing and leisure choices would have impact. She de­liberately wore pants one day as she walked through an office-not her own–of female clerks supervised by a man who wanted them to wear dresses, and she noted that a few women cautiously began to wear pants occasionally. She decided to let it be known that she was leaving at four p.m. for ballet les­sons once a week, arguing that the men at her level did the same thing to play golf, but also knowing that ballet was going to have a very different meaning from golf. Her act was a gesture performed with an audience in mind as much as an expression of preference. The meaning of “natural” in such situations is problematic, for in doing what they might find natural as private beings, tokens as public personae are also sending messages to the organization.

Business as well as personal decisions were handled by tokens with an awareness of their extended symbolic consequences. One woman manager was faced with the dilemma of deciding what to do about a woman assistant who wanted to go back to the secretarial ranks from which she had recently been promoted. The manager felt she jeopardized her own claims for mobility and the need to open the system to more women if she let her assistant return and had to admit that a woman who was given opportunity had failed. She spent much more time on the issue than a mere change of assistants would have warranted, going privately to a few men she trusted at the officer level to discuss the situation. She also kept the assistant on much longer than she felt was wise, but she thought herself trapped.

Sometimes the thought of the symbolic as well as personal consequences of acts led token women to outright distortions. One was an active feminist in a training staff job who, according to her own reports, “separated what I say for the cause from what I want for myself.” Her secret ambition was to leave the corporation within a year or two to increase her own professional skills and become an external consultant. But when discussing her aspirations with her own manager in career reviews or with peers on informal occasions, she always smiled and said, “Chairman of the board of Industrial Supply Corporation.” Every time a job at the grade level above her became vacant, she would inquire about it and appear to be very interested, making sure that there was some reason at the last minute she could not take it. “They are watching me,” she explained, “to see if women are really motivated or if they will be content to stay in low-level jobs. They are expecting me to prove something one way or the other.”

The Tokenism Eclipse

The token’s visibility stemmed from characteristics-attributes of a mas­ter status-that threatened to blot out other aspects of a token’s performance. Although the token captured attention, it was often for her discrepant charac­teristics, for the auxiliary traits that gave her token status. The token does not have to work hard to have her presence noticed, but she does have to work hard to have her achievements noticed. In the sales force, the women found that their technical abilities were likely to be eclipsed by their physical appear­ances, and thus, an additional performance pressure was created. The women had to put in extra effort to make their technical skills known, and said they worked twice as hard to prove their competence.

Both male peers and customers could tend to forget information women provided about their experiences and credentials while noticing and remem­bering such secondary attributes as style of dress. For example, there was this report from a salesman: “Some of our competition, like ourselves, have women sales people in the field. It’s interesting that when you go in to see purchasing agent, what he has to say about the woman sales person. It is always what kind of a body she had or how good-looking she is or “Boy, are you in trouble on this account now.” They don’t tell you how good-looking your competitors are if they’re males, but I’ve never heard about a woman’s technical competence or what kind of a sales person she was– only what her body was like.” And a saleswoman complained in an angry outburst, “There are times when I would rather say to a man, ‘Hey, listen, you can have our bodies and look like a female and have the advantage of walking in the room and being noticed. ‘ But the noticeability also has attached to it that surprise on the part of men that you can talk and talk intelligently. Recognition works against you as well as for you.” And another; “Some of the attention is nice, but some of it is demeaning to a professional. When a man gets a job, they don’t tell him he’s better looking than the man who was here before-but they say that to me.” The focus on appearance and other non-ability traits was an al­most direct consequence of the presence of very few women.

Fear of Retaliation

The women were also aware of another performance pressure; not to make the dominants look bad. Tokenism sets up a dynamic that can make tokens afraid of being too outstanding in performance on group events and tasks. When a token does well enough to “show up” a dominant, it cannot be kept a secret, since all eyes are upon the token, and therefore, it is more dif­ficult to avoid the public humiliation of a dominant. Thus, paradoxically, while the token women felt they had to do better than anyone else in order to be seen as competent and allowed to continue, they also felt, in some cases, that their successes would not be rewarded and should be kept to themselves. They needed to toe the fine line between doing just well enough and too well. One woman had trouble understanding this and complained of her treatment by managers. They had fired another woman for not being aggressive enough, she reported; yet she, who succeeded in doing all they asked and brought in the largest amount of new business during the past year, was criticized for being “too aggressive, too much of a hustler.”

The fears had some grounding in reality. In a corporate bureaucracy like Indesco, where “peer acceptance” held part of the key to success in securing promotions and prized jobs it was known how people were received by colleagues as well as by higher management. Indeed, men down the ranks resented the tendency for some top executives to make snap judgments about people after five minutes’ worth of conversation and then try to influence their career reviews and create instant stars. So the em­phasis on peer acceptance in performance evaluations, a concept known to junior managers, was one way people lower down the managerial hierarchy re­tained some control over the climbing process, ensured themselves a voice, and maintained a system they felt was equitable, in which people of whom they approved had a greater chance for success. Getting along well with peers was thus not just something that could make daily life in the company more pleasant; it was also fed into the formal review system.

At a meeting of ten middle managers, two women who differed in peer acceptance were contrasted. One was well liked by her peers even though she had an outstanding record because she did not flaunt her successes and mod­estly waited her turn to be promoted. She did not trade on her visibility. Her long previous experience in technical work served to certify her and elicit col­league respect, and her pleasant but plain appearance and quiet dress mini­mized disruptive sexual attributes. The other was seen very differently. The mention of her name as a “star performer” was accompanied by laughter and these comments: “She’s infamous all over the country. Many dislike her who have never met her. Everyone’s heard of her whether or not they know her, and they already have opinions. There seems to be no problem with direct peer acceptance from people who see her day-to-day, but the publicity she has received for her successes has created a negative climate around her.” Some thought she was in need of a lesson for her cockiness and presumption. She was said to be aspiring too high, too soon, and refusing to play the promotion game by the same rules the men had to use: waiting for one’s turn, the requi­site years’ experience and training. Some men at her level found her overrated and were concerned that their opinions be heard before she was automatically pushed ahead. A common prediction was that she would fail in her next assign­ment and be cut down to size. The managers, in general, agreed that there was backlash if women seemed to advance too fast.

And a number of men were concerned that women would jump ahead of them. They made their resentments known. One unwittingly revealed a cen­tral principle for the success of tokens in competition with dominants: to always stay one step behind, never exceed or excel. “It’s okay for women to have these jobs,” he said, “as long as they don’t go zooming by me.”

One form peer retaliation against success took was to abandon a success­ful woman the first time she encountered problems. A dramatic instance in­volved a confrontation between a very dignified woman manager, the only woman in a management position in her unit, who supervised a large group of both male and female workers, and an aggressive but objectively low-perform­ing woman subordinate, who had been hired by one of the other managers and was unofficially “sponsored” by him. The woman manager had given low rat­ings to the subordinate on her last performance appraisal, and another review was coming up; the manager had already indicated that the rating would still be low, despite strong protests of unfairness from the worker. One day after work, the manager walked through a public lounge area where several work­ers were standing around, and the subordinate began to hurl invectives at her, accusing her of being a “bitch, a stuck-up snob,” and other unpleasant labels. The manager stood quietly, maintaining her dignity, then left the room, fear­ing physical violence. Her feelings ranged from hurt to embarrassment at the public character of the scene and the talk it would cause. The response over the next few days from her male peers ranged from silence to comments like, “The catharsis was good for X. She needed to get that off her chest. You know, you never were responsive to her.” A male friend told the manager that he heard two young men who were passed over for the job she was eventually given commenting on the event; “So Miss High-and-Mighty finally got hers!” The humiliation and the thought that colleagues supported the worker rather than her was enough to make this otherwise-successful woman consider leav­ing the corporation.

 Tokens’ Responses to Performance Pressures

A manager posed the issue for scarce women this way: “Can they survive the organizational scrutiny?” The choices for those in the token position were either to over-achieve and carefully construct a public performance that mini­mized organizational and peer concerns, to try to turn the notoriety of public­ity to advantage, or to find ways to become socially invisible. The first course means that the tokens involved are already outstanding and exceptional, able to perform well under close observation where others are ready to notice first and to attribute any problems to the characteristics that set them apart-but also able to develop skills in impressions management that permit them to re­tain control over the extra consequences loaded onto their acts. This choice in­volved creating a delicate balance between always doing well and not generat­ing peer resentment. Such dexterity requires both job-related competence and political sensitivity that could take years to acquire. For this reason, young women just out of college had the greatest difficulty in entering male domains like the Indesco sales force and were responsible for much of the high turnover among women in sales. Women were successful, on the other hand, who were slightly older than their male peers, had strong technical backgrounds, and had already had previous experiences as token women among male peers. The success of such women was most likely to increase the prospects for hiring more women in the future; they worked for themselves and as symbols.

The second strategy, accepting notoriety and trading on it, seemed least likely to succeed in a corporate environment because of the power of peers. A few women at Indesco flaunted themselves in the public arena in which they operated and made a point out of demonstrating their” difference,” as in refusing to go to certain programs, parading their high-level connections, or by­passing the routine authority structure. Such boldness was usually accom­panied by top management sponsorship. But this strategy was made risky by shifting power alliances at the top; the need to secure peer cooperation in cer­tain jobs where negotiation, bargaining, and the power of others to generate advantage or disadvantage through their use of the rules were important; and the likelihood that some current peers would eventually reach the top. Furth­ermore, those women who sought publicity and were getting it in part for their rarity developed a stake in not sharing the spotlight. They enjoyed their only-women status, since it gave them an advantage, and they seemed less consciously aware than the other women of the attendant dangers, pressures, psychic costs, and disadvantages. In a few instances, they operated so as to keep other women out by excessive criticism of possible new-hires or by sub­tly undercutting a possible woman peer (who eventually left the company), something that, we shall see later, was also pushed for by the male dominants. Thus, this second strategy eventually kept the numbers of women down both because the token herself was in danger of not succeeding and because she might keep other women out. This second strategy, then, serves to reinforce the dynamics of tokenism by ensuring that, in the absence of external pres­sures like affirmative action, the group remains skewed.

The third choice was more often accepted by the older generation of cor­porate women, who predated the women’s movement and had years ago ac­commodated to token status. It involved attempts to limit visibility, to become “socially invisible.” This strategy characterizes women who try to minimize their sexual attributes so as to blend unnoticeably into the predominant male culture, perhaps by adopting “mannish dress,” as in reports by other inves­tigators. Or it can include avoidance of public events and occasions for perfor­mance-staying away from meetings, working at home rather than in the of­fice, keeping silent at meetings. Several of the saleswomen deliberately took such a “low profile,” unlike male peers who tended to seize every opportunity to make themselves noticed. They avoided conflict, risks, or controversial situ­ations. They were relieved or happy to step into assistant or technical staff jobs such as personnel administration or advertising, where they could quietly play background roles that kept men in the visible forefront-or they at least did not object when the corporation put them into low-visibility jobs, since for many years the company had a stake in keeping its “unusual” people hidden.

Those women preferring or accepting social invisibility also made little at­tempt to make their achievements publicly known or to get credit for their own contributions to problem-solving or other organizational tasks, just like other women reported in the research literature who have let men assume vis­ible leadership or take credit for accomplishments that the women really produced-the upper corporate equivalent of the achieving secretary. In one remarkable laboratory experiment, women with high needs for dominance, paired with a man in a situation where they had to choose a leader, exercised their dominance by appointing him the leader. [5]


Women making this choice, then, did blend into the background and control their performance pressures, but at the cost of limited recognition of their competence. This choice, too, in­volved a psychic splitting, for rewards for such people often came with secret knowledge-knowing what they had contributed almost anonymously to an ef­fort that made someone else look good. In general, this strategy, like the last, also reinforces the existence of tokenism and keeps the numbers of women down, because it leads the organization to conclude that women are not very effective: low risk-takers who cannot stand on their own.

The performance pressures on people in token positions generate a set of attitudes and behaviors that appear sex-linked, in the case of women, but can be understood better as situational responses, true of any person in a token role. Perhaps what has been called in the popular literature “fear of success in women,” for example, is really the token woman’s fear of visibility. The original research that identified the fear of success concept created a hypothetical situation in which a woman was at the top of her class in medical school-a token woman in a male peer group. Such a situation is the kind that exacts extra psychic costs and creates pressures for some women to make themselves and their achievements invisible-to deny success. Replication of this research using examples of settings in which women were not so clearly proportionately scarce produced very different results and failed to confirm the sex¬-linked nature of this construct. Seymour Sarason also pointed out that minorities of any kind, trying to succeed in a culturally alien environment, may fear visibility because of retaliation costs and, for this reason, may try to play down any recognition of their presence, as did Jews at Yale for many years. [6] Fear of visibility, then, is one response to performance pressures in a token’s situation. The token must often choose between trying to limit visibility-and being overlooked-or taking advantage of the publicity-and being labeled a “troublemaker.”





Contrast, or exaggeration of the token’s differences from dominants, sets a second set of dynamics in motion. The presence of a token or two makes dominants more aware of what they have in common at the same time that it threatens that commonality. Indeed, it is often at those moments when a collectivity is threatened with change that its culture and bonds become exposed to itself; only when an obvious “outsider” appears do group members suddenly realize aspects of their common bond as insiders. The “threat” a token poses is twofold. First, the token represents the danger of challenge to the dominants’ premises, either through explicit confrontation by the token or by a disaffected dominant who, through increased awareness, sees the culture for what it is and sees the possibility of alternatives. Second, the self-consciousness created by the token’s presence is uncomfortable for people who prefer to operate in casual, superficial, and easygoing ways, without much psychological self-awareness and without the strain of reviewing habitual modes of action-a characteristic stance in the corporate environment.

Furthermore, as Everett Hughes pointed out, part of the hostility peer groups show to new kinds of people stems from uncertainty about their behav­ior when non-structured, non-routine events occur. Tokens cannot be as­sumed to share the same unspoken understandings that the rest of the members, share because of their common membership in a social category, one basis for closing ranks against those who are different. For smooth interaction, groups require both discretion (the ability to put statements in their proper perspective) and a shared vocabulary of attitudes (the ability to take feelings and sentiments for granted) so that they can avoid the time-consuming process of translation. At best, then, members of the dominant category are likely to be uncomfortable and uncertain in the pres­ence of a member of a different category. Other analysts have also shown that people with “incongruent statuses,” like women in male jobs, strain group in­teraction by generating ambiguity and lack of social certitude.[7] It is not only the first of a kind that arouses discomfort. People who are usually not found in that setting and come from a category with a history of special forms of interac­tion with the numerical dominants, as rare women among men, are also po­tentially disruptive of peer interaction.

The token’s contrast effect, then, can lead dominants to exaggerate both their commonality and the token’s “difference.” They move to heighten boundaries of which, previously, they might even have been aware. They erect new boundaries that at some times exclude the token or at others let her in only if she proves her loyalty.

Exaggeration of Dominants’ Culture

Indesco men asserted group solidarity and reaffirmed shared in-group un­derstandings in the presence of token women, first, by emphasizing and exag­gerating those cultural elements they shared in contrast to the token. The token became both occasion and audience for the highlighting and dramatizing of those themes that differentiated her as the outsider. Ironically, tokens, unlike people of their type represented in greater proportion, are thus in­struments for underlining rather than undermining majority culture. At Indesco, this phenomenon was most clearly in operation on occasions that brought together people from many parts of the organization who did not nec­essarily know each other well, as in training programs and at dinners and cock­tail parties during meetings. Here the camaraderie of men, as in other work and social settings, [8] was based in part on tales of sexual adventures, ability with respect to “hunting” and capturing women, and off-color jokes. Other themes involved work prowess and sports, especially golf and fishing. The ca­pacity for and enjoyment of drinking provided the context for displays of these themes. They were dramatized and acted out more fervently in the presence of token women than when only men were present. [9] When the men were alone, they introduced these themes in much milder form and were just as likely to share company gossip or talk of domestic matters such as a house being built. This was also in contrast to more equally mixed male-female groups in which there were a sufficient number of women to influence and change group culture and introduce a new hybrid of conversational themes based on shared male-female concerns. [10]

Around token women, then, men sometimes exaggerated displays of aggression and potency: instances of sexual innuendos, aggressive sexual teas­ing, and prowess-oriented “war stories.” When a woman or two were present, the men’s behavior involved” showing off,” telling stories in which” masculine prowess” accounted for personal, sexual, or business success. They high­lighted what they could do, as men, in contrast to the women. In a set of train­ing situations for relatively junior salespeople, these themes were even acted out overtly in role plays in which participants were asked to prepare and per­form demonstrations of sales situations. In every case involving a woman, the men played the primary, effective roles, and the women were objects of sexual attention. Sexual innuendos were heightened and more obvious and exagger­ated than in all-male role plays, as in these two examples:

[I] Two men and a woman simulated a call on a buyer; the woman was introduced as the president of the company, but the sales manager and his assistant did all the talking. The company was in the business of selling robots. The sales manager brought in a male “robot” to demonstrate the product. The sales manager leered at him, saying, “Want a little company?” He then revealed that the woman in­troduced as the president was actually one of the female robots.

[II]  The two-man, one-woman team was selling wigs; the woman was the wig stylist. The buyer on whom they were calling adopted an exaggerated homosexual carica­ture, which broadened considerably during the” sales call.” Toward the end of the role play, one of the men, trying to wrap up the sale, said, “We have other special services along with wigs. Other women who work with our stylist will come to your store to work for you.” The buyer’s response made it clear that he would be interested in those women sexually (though he was simulating homosexuality). Said the seller, ‘They’ll be on your payroll; you can use them any way you want.” Said the buyer, leering, “Any way I want?” The seller answered, “We might offer other services like a massage along with the wig.” Said the buyer, “That sounds interesting. Can I have one right now?” After these role plays, the group atmosphere seemed quite tense, and the women especially appeared highly uncomfortable.

The women themselves reported other examples of “testing” to see how they would respond to the “male” culture. They said that many sexual innu­endos or displays of locker-room humor were put on for their benefit, espe­cially by the younger men. (The older men tended to parade their business successes,) One woman was a team leader at a workshop (and the only woman), when her team decided to use as its slogan, “The [obscenity] of the week,” looking at her for a reaction. By raising the issue and forcing the woman to choose not to participate, the men in the group created an occasion for uniting against the outsider and asserting dominant group solidarity. Such events, it must be pointed out, were relatively rare and occurred only at those informal occasions outside of the business routine in which people were un­winding, letting themselves go, or, as in the training role plays, deliberately creating unreal situations, Most behavior at Indesco was more businesslike in tone. But the fact that such interaction ever occurred, even infrequently, around women served to isolate them and make them uncomfortable at those very moments when, ironically, people were supposed to be relaxing and having fun.

A sales meeting at Indesco provided an interesting example of how the dominant culture could simultaneously acknowledge the presence of tokens and retain its own themes and flavor. It was traditional for salesmen to tell traveling salesman/farmer’s daughter jokes at informal gatherings. On this oc­casion, four years after women first entered the sales force, a raunchy traveling saleswoman/farmer’s son joke was told, a story currently going around the company. The form was the same, but the content reflected the presence of women. Tokens functions as audience for dominant cultural expressions also played a part in the next set of processes,

Interruptions as Reminders of “Difference”

On more formal occasions, as in meetings, members of the numerically dominant category underscored and reinforced differences between tokens and dominants, ensuring that tokens recognized their outsider status, by making the token the occasion for “interruptions” in the flow of group events, Dominants prefaced acts with apologies or questions about appropriateness directed at the token; they then invariably went ahead with the act, having placed the token in the position of interrupter or interloper, of someone who took up the group’s time. This happened often in the presence of the saleswomen. Men’s questions or apologies represented a way of asking whether the old or ex­pected cultural rules were still operative-the words and expressions permit­ted, the pleasures and forms of release indulged in. (Can we still swear? Toss a football? Use technical jargon? Go drinking? Tell “in” jokes?) [11] Sometimes the questions seemed motivated by a sincere desire to put the women at ease and treat them appropriately, but the net effect was the same regardless of dominants’ intentions. By posing these questions overtly, dominants made the culture clear to tokens, stated the terms under which tokens enter the rela­tionship, and reminded them that they were special people. It is a dilemma of all cross-cultural interaction that the very act of attempting to learn what to do in the presence of the different kind of person so as to integrate him can rein­force differentiation.

The answers about conduct almost invariably affirmed the under­standings of the dominants. The power of sheer numbers means that an indi­vidual rarely feels comfortable preventing a larger number of peers from en­gaging in an activity they consider normal. Most women did not want to make a fuss, especially about issues they considered trivial and irrelevant to their job status, like saying “goddamn” or how to open doors. Their interest in not being signaled out for special treatment made them quickly agree that things should proceed as they would if women were not present, and to feel embar­rassment about stopping the flow of conversation. None wanted to be a “wet blanket”! As one said, “They make obscene suggestions for slogans when kidding around, looking to me for a reaction. Then they jump on me for not liking it.”

Secondly, the tokens have been put on notice that interaction will not be “natural,” that dominants will be “holding back,” unless they agree to ac­knowledge and permit (and even encourage) majority cultural expressions in their presence. (It is important that this be stated, of course, for one never knows that another is holding back unless the other lets a piece of the sup­pressed material slip out.) At the same time, tokens have also been given the implicit message that majority members do not expect those forms of expres­sion to be “natural” to the tokens’ home culture; otherwise, majority members would not need to raise the question. (This is a function of what Judith Long Laws called the “double deviance” of tokens: deviant first because they are women in a man’s world and second because they inappropriately aspire to the privileges of the dominants.) [12] Thus, the saleswomen were often in the odd position of reassuring peers and customers that they could go ahead and do something in the women’s presence, like swearing, that the women them­selves would not be permitted to do. They listened to dirty jokes, for example, but reported that they would not dare tell one themselves. In fact, whether or not to go drinking or tell jokes was a major question for women: “You can’t tell dirty jokes. Clean jokes would go over like a lead balloon. So I sit there like a dummy and don’t tell jokes.”

Via difference-reminding interruptions, then, dominants both affirm their own shared understandings and draw the cultural boundary between themselves and tokens. The tokens learned that they caused interruptions in “normal” communication, and that their appropriate position was more like that of audience than that of full participant. But the women also found the au­dience position frustrating or wearying, as these statements indicated: “I felt like one of the guys for a while. Then I got tired of it. They had crude mouths and were very immature. I began to dread the next week because I was tired of their company. Finally, when we were all out drinking, I admitted to myself, this is not me; I don’t want to play their game.” And: “I was at a dinner where the men were telling dirty jokes. It was fun for a while; then it got to me. I moved and tried to have a real conversation with a guy at the other end of the table. The dinner started out as a comrade thing, but it loses its flavor, especially if you’re the only woman. I didn’t want them to stop on my account, but I wish I had had an alternative conversation.”

Overt Inhibition: Informal Isolation

In some cases, dominants did not wish to have tokens around all the time; they had secrets to preserve or simply did not know how far they could trust the women, especially those who didn’t seem to play by all the rules. They thus moved the locus of some activities and expressions from public settings to which tokens had access to more private settings from which they could be excluded. When information potentially embarrassing or damaging to domi­nants is being exchanged, an outsider-audience is not desirable, because dom­inants do not know how far they can trust the tokens. As Hughes and Chapter 3 pointed out, colleagues who rely on unspoken understandings may feel un­comfortable in the presence of “odd kinds of fellows” who cannot be trusted to interpret information in just the same way or to engage in the same rela­tionships of trust and reciprocity. [13] There was a sense that it was not possible to level with a woman or be real with her, as one could with other men.

The result was sometimes “quarantine”-keeping tokens away from some occasions. Informal pre-meeting meetings were sometimes held. Some topics of discussion seemed rarely raised by men in the presence of many of their women peers, even though they discussed them among themselves: admis­sions of low commitment to the company or concerns about job performance, ways of getting around formal rules, political plotting for mutual advantage, strategies for impressing certain corporate executives. Many of the women did not tend to be included in the networks by which informal socialization oc­curred and politics behind the formal system were exposed, as researchers have found in other settings. One major project found that people with in­congruent statuses, like the Indesco exempt women, were likely to become isolates in peer groups and to have less frequent interaction with the group than other members, outside of formally structured occasions. [14] Toward the upper levels of the corporation, any tendency for peer groups to quarantine women was reinforced by men-only social establishments; a senior personnel administrator committed to placing more women in top executive jobs was concerned about whether they could overcome the limitation on their busi­ness effectiveness placed by exclusion from informal exchanges at male clubs.

In a few cases, overt inhibition worked directly against women in their jobs. They missed out on important informal training by peers.1S There were instances in which women trainees did not get direct criticism in time to improve their performance and did not know they were the subjects of criti­cism in the company until told to find jobs in other divisions. They were not part of the buddy network that uncovered such information quickly, and their managers were reluctant to criticize a woman out of uncertainty about how she would receive the information. (One man put quite simply how he felt about giving negative feedback to a woman: ”I’m chicken.”) Here feelings that it was impossible to level with a different kind of person stood in the way.

Loyalty Tests

At the same time that tokens may be kept on the periphery of colleague in­teraction, they may also be expected to demonstrate loyalty to their dominant peers. Failure to do so could result in further isolation; signs of loyalty, on the other hand, permitted the token to come closer to being included in more of the dominants’ activities. Through loyalty tests, the group sought reassurance that the tokens would not turn against the dominants or use any of the infor­mation gained through their viewing of the dominants’ world to do harm to the group. In the normal course of peer interactions, people learn all sorts of things about each other that could be turned against the other. Indeed, many colleague relationships are often solidified by the reciprocal knowledge of po­tentially damaging bits of information and the understanding that they both have an interest in preserving confidentiality. Tokens, however, pose a dif­ferent problem and raise uncertainties, for their membership in a different social category could produce loyalties outside the peer cadre.

This was a quite rational concern on occasion. With government pres­sures and public interest mounting, Indesco women were often asked to speak to classes or women’s groups or to testify before investigating committees. One woman was called in by her manager before her testimony at hearings on discrimination against women in business; he wanted to hear her testi­mony in advance and have censorship rights. She refused, but then made only very general and bland statements at the hearing anyway.

Peers seek reassurance about embarrassing as well as damaging disclo­sures. There is always the possibility that tokens will find some of what the dominants naturally do silly or ridiculous and will insult them where they feel vulnerable. Dominants also want to know that tokens will not use their inside information to make the dominants look bad or turn them into figures of fun to members of the token’s category outside with whom they must interact. The joking remarks men made when seeing women colleagues occasionally eating with the secretaries (e.g. “What do you ‘girls’ find so interesting to talk about?”) revealed some of their concerns.

Assurance could be gained by asking tokens to join with or identify with the dominants against those who represented competing loyalties; in short, dominants pressured tokens to turn against members of their own category, just as occurred in other situations where women were dominants and men tokens. [16] If tokens colluded, they made themselves psychological hostages to the majority group. For token women, the price of being “one of the boys” was a willingness to occasionally turn against “the girls.”

There were three ways token women at Indesco could demonstrate loyalty and qualify for a closer relationship with dominants. First, they could let slide (or even participate in) statements prejudicial to other members of their cate­gory. They could allow themselves to be viewed as “exceptions” to the “gen­eral rule” that others of their category have a variety of undesirable or unsuita­ble characteristics; Hughes recognized this as one of the “deals” token blacks might make for membership in white groups, [17] Women who did well were sometimes told they were “exceptions” and exceptional, not like a “typical woman.” It is an irony of the token situation that women could be treated as both representatives of their type and exceptions to it, sometimes by the same people.

At meetings and training sessions, women were occasionally the subjects of ridicule or joking remarks about their incompetence. Some of the women who were insulted by such innuendos found it easier to appear to agree than to start an argument. A few accepted the dominants’ view fully. One of the first saleswomen denied in interviews having any special problems because she was a woman, calling herself “skilled at coping with a man’s world,” and said the company was right not to hire more women. Women, she said, were unreliable and likely to quit; furthermore, young women might marry men who would not allow them to work. (She herself quit a few years later.) In this case, a token woman was taking over “gatekeeping” functions for dominants, letting them appear free of prejudice while a woman acted to exclude other women. [18]

Tokens could also demonstrate loyalty by allowing themselves and their category to provide a source of humor for the group. Laughing with others, as Rose Coser indicated, is a sign of a common definition of the situation; to allow oneself or one’s kind to be the object of laughter signals a further willingness to accept the others’ culture on their terms. [19] Just as Hughes found that the ini­tiation of blacks into white groups might involve accepting the role of comic in­ferior, [20] Indesco women faced constant pressures to allow jokes at the expense of women, to accept “kidding” from the men around them. When a woman objected, the men denied any hostility or unfriendly intention, instead accus­ing the woman, by inference, of “lacking a sense of humor.” In order to cope, one woman reported, she “learned to laugh when they try to insult you with jokes, to let it roll off your back.” Tokens could thus find themselves colluding with dominants through shared laughter.

Thirdly, tokens could demonstrate their gratitude for being included by not criticizing their situation or pressing for any more advantage. One major taboo area involved complaints about the job or requests for promotion. The women were supposed to be grateful for getting as far as they had (when other women clearly had not) and thus expected to bury dissatisfaction or aspirations.

Responses of Tokens to Boundary Heightening

The dilemma posed here for tokens was how to reconcile their awareness of difference generated by informal interaction with dominants with the need, in order to belong, to suppress dominants’ concerns about the difference. As with performance pressures, peer group interaction around the tokens in­creased the effort required for a satisfactory public appearance, sometimes ac­companied by distortions of private inclinations.

Of course, ‘not all men participated in the dynamics noted. And some tokens managed to adapt very well. They used the same kind of language and expressed the same kinds of interests as many of the men. One woman loved fishing, she said, so when she came on as a manager and her office was con­cerned that that would end fishing trips, she could show them they had noth­ing to fear. Another had a boat on which she could take customers (along with her husband and their wives). A professional woman joined the men on “woman hunts,” taking part in conversations in which the pro’s and con’s of particular targets were discussed. There were women known to be able to “drink the men under the table.” It was never clear what the psychic toll of such accommodation was-whether, for example, such people would have made different choices in a balanced context-for they were also unlikely to talk about having any problems at all in their situation; they assumed they were full members.

Numerical skewing and polarized perceptions left tokens with little choice about accepting the culture of dominants. There were too few other people of the token’s kind to generate a “counterculture” or to develop a shared intergroup culture. Tokens had to approach the group as single individ­uals. They thus had two general response possibilities. They could accept isolation, remaining only an audience for certain expressive acts of dominants. This strategy sometimes resulted in friendly but distant peer relations, with the risk of exclusion from occasions on which informal socialization and politi­cal activity took place. Or they could try to become insiders, proving their loy­alty by defining themselves as exceptions and turning against their own social category.

The occurrence of the second response on the part of tokens suggests a reexamination of the popularized “women-prejudiced-against-women” hy­pothesis, also called the “Queen Bee syndrome,” for possible structural (nu­merical) rather than sexual origins. Not only has this hypothesis not been con­firmed in a variety of settings, [21] but the analysis offered here of the social psychological pressures on tokens to side with the majority provides a more compelling explanation for the kinds of situations most likely to produce this effect. To turn against others of one’s kind (and thus risk latent self-hatred) can by a psychic cost of membership in a group dominated by another culture.



Tokens can never really be seen as they are, and they are always fighting stereotypes, because of a third tendency. The characteristics of tokens as indi­viduals are often distorted to fit preexisting generalizations about their cate­gory as a group–what I call “assimilation.” Such stereotypical assumptions about what tokens “must” be like, such mistaken attributions and biased judg­ments, tend to force tokens into playing limited and caricatured roles. This constrains the tokens but is useful for dominant group members.

Whatever ambiguity there might be around a strange person is reduced by providing a stereotyped and thus familiar place for tokens in the group, allowing domi­nants to make use of already-learned expectations and modes of action, like the traditional ways men expect to treat women. Familiar roles and assump­tions, further, can serve to keep tokens in a bounded place and out of the mainstream of interaction where the uncertainties they arouse might be more difficult to handle. In short, tokens become encapsulated in limited roles that give them the security of a “place” but constrain their areas of permissible or rewarded action.

Status Leveling

Tokens were often initially misperceived as a result of their numerical rar­ity. That is, an unusual woman would be treated as though she resembled women on the average-a function of what has been called “statistical discrim­ination” rather than outright prejudice. [22] Since people make judgments about the role being played by others on the basis of probabilistic reasoning about what a particular- kind of person will be doing in a particular situation, such misperceptions are the result of statistical mistakes. Thus, women exempts at Indesco, like other tokens, encountered many instances of “mistaken iden­tity”-first impressions that they were occupying a usual female position rather than their unusual (for a woman) job. In the office, they were often taken for secretaries; on sales trips on the road, especially when they traveled with a male colleague, they were often taken for wives or mistresses; with cus­tomers, they were first assumed to be temporarily substituting for a man who was the “real” salesperson; with a male peer at meetings, they were seen as the assistant; when entertaining customers, they were assumed to be the wife or date. (One woman sales trainee accompanied a senior salesman to call on a customer, whose initial reaction was laughter: “What won’t you guys think up next? A woman!” She had the last laugh, however, for that company’s chief en­gineer happened to be a woman with whom she had instant rapport.)

Mistaken first impressions can be corrected, although they give tokens an extra burden of spending more time untangling awkward exchanges and es­tablishing accurate and appropriate role relations. But meanwhile, status lev­eling occurs. Status leveling involves making adjustments in perception of a token’s professional role to fit with the expected position of the token’s category-that is, bringing situational status in line with what has been called “master status,” the token’s social type. Even when others knew that the women were not secretaries, for example, there was still a tendency to treat them like secretaries or to make secretary-like demands on them. In one bla­tant case, one woman was a sales trainee along with three men, all four of whom were to be given positions as summer replacements. The men were all assigned to replace salesmen; the woman was asked to replace a secretary ­and only after a long and heated discussion with the manager was she given a more professional assignment. Similarly, when having professional contacts with customers and managers, the women felt themselves to be treated in more wife-like or date-like ways than a man would treat another man, even though the occasion was clearly professional.

A professional woman at Indesco asked for a promotion and talked about looking for a better job; her manager’s first assumption was that she did not feel “loved” and it was his fault for failing to give love to a woman. In all these instances, it was easier for others to fit the token woman to their preexisting generalizations about women than to change the category; numerical rarity provided too few examples to contra­dict the generalization. (Instances of status leveling have also been noted around other kinds of tokens, such as male nurses; [23] in the case of tokens whose master status is higher than their situational status, leveling can work to their advantage, as when male nurses are called “Dr.”)

The Woman’s Slot

There was also a tendency to encapsulate women and to maintain gener­alizations by defining special roles for women, even on the managerial and professional levels, that put them slightly apart as colleagues. Again, it was easy to do this with a small number and would have been much harder with many more women spilling over the bounds of such slots. A woman could en­sure her membership by accepting a special place but then find herself con­fined by it. Once women began to occupy certain jobs, those jobs sometimes gradually came to be defined as “women’s slots.” One personnel woman at Indesco pointed this out. In her last career review, she had asked to be moved, feeling that, in another six months, she would have done and learned all she could in her present position and was ready to be upgraded. “They [the managers] told me to be patient; if I waited a year or two longer, they had just the right job for me, three grades up. I knew what they had in mind. Linda Martin [a senior woman] would be retiring by then from a benefits administra­tion job, and they wanted to give it to me because it was considered a place to put a woman. But it had no real responsibilities despite its status; it was all routine work.”

Affirmative action and equal employment opportunity jobs were also seen as “women’s jobs.” Many women, who would otherwise be interested in the growth and challenge they offered, said that they would not touch such a posi­tion: “The label makes it a dead end. It’s a way of putting us out to pasture.” There was no way to test the reality of such fears, given the short time the jobs had been in existence, but it could be observed that women who worked on women’s personnel or training issues were finding it hard to move out into other areas. These women also found it hard to interest some other, secretly sympathetic managerial women in active advocacy of upward mobility for women because of the latter’s own fears of getting too identified with a single issue. (Others, though, seized on it as a way to express their values or to get visibility.)

Committees, task forces, and other ad hoc events had a tendency, too, to develop a woman’s slot for those women selected to participate. Sometimes it would take the form of giving the women areas of responsibility that were stereotypically “female” concerns, or, as mentioned earlier, giving them the role in the group of” expert on women.” This posed major dilemmas for the women seriously interested in being women’s advocates but who were also aware of how the role encapsulation process could undercut their effectiveness and limit their organizational mobility. They had to carefully balance the time spent as woman-symbols with other activities and with attention to the tech­nical/professional aspects of their jobs.

Stereotyped Informal Roles

Dominants can incorporate tokens and still preserve their generalizations by inducting tokens into stereotypical roles that preserve familiar forms of interaction between the kinds of people represented by the token and the domi­nants. In the case of token women in colleague groups at Indesco, four informal role traps were observed, all of which encapsulated the tokens in a category the men could respond to and understand. Each was formed around one be­havioral tendency of the token, building this into an image of the token’s place in the group and forcing her to continue to live up to the image; each defined for dominants a single response to the token’s sexuality. Two of the roles are classics in Freudian theory: the “mother” and the “seductress.” Freud wrote of the need for men to handle women’s sexuality by envisioning them either as “madonnas” or “whores”-as either asexual mothers or overly sexual, debased seductresses, perhaps as a function of Victorian family patterns, which en­couraged separation of idealistic adoration toward the mother and animalistic eroticism.[24] The others, termed the “pet” and the “iron maiden,” also have family counterparts in the kid sister and the virgin aunt.


A token woman sometimes found that she became a “mother” to men in the group. One by one, they brought her their private troubles, and she was expected to comfort them. The assumption that women are sympathetic, good listeners and easy to talk to about one’s problems was common, even though, ironically, men also said it was hard to level with women over task-related is­sues. One saleswoman was constantly approached by her all-male peers to lis­ten to their problems with their families. In a variety of residential training groups, token women were observed acting out other parts of the traditional nurturant-maternal role: doing laundry, sewing on buttons for men.

The mother role was comparatively safe. A mother is not necessarily vul­nerable to sexual pursuit (for Freud it was the very idealization of the madonna that was in part responsible for men’s ambivalence toward women), nor do men need to compete for her favors, since they are available to everyone. However, the typecasting of women as nurturers can have three negative con­sequences for the woman’s task performance: (1) The mother is rewarded primarily for service and not for independent action. (2) The dominant, power­ful aspects of the maternal image may be feared, and thus the mother is ex­pected to keep her place as a non-critical, accepting, “good mother” or lose her rewards. Since the ability to differentiate and be critical is often an indica­tor of competence in work groups, the mother is prohibited from exhibiting this skill. (3) The mother becomes an emotional specialist. This provides her with a place in the life of the group and its members. Yet, at the same time one of the stereotypically “feminine” characteristics men in positions of authority in industry most often criticize in women is excess “emotionality.” Although the mother herself might not ever cry or engage in emotional outbursts in the group, she remains identified with emotional matters. As long as she is in the scarce position of token, however, it is un­likely that nurturance, support, and expressivity will be valued or that a mother can demonstrate and be rewarded for critical, independent, task­-oriented behaviors.


The role of seductress or sexual object is fraught with more tension than the maternal role, for it introduces an element of sexual competition and jeal­ousy. The mother can have many sons; it is more difficult for the sexually at­tractive to have many swains. Should the woman cast as sex object (that is, seen as sexually desirable and potentially available–seductress is a percep­tion, and the woman herself may not be consciously behaving seductively) share her attention widely, she risks the debasement of the whore. Yet, should she form a close alliance with any man in particular, she arouses resentment, particularly so because she represents a scarce resource; there are just not enough women to go around.

In several situations I observed, a high status male allied himself with the seductress and acted as her “protector,” partly because of his promise of res­cue from sex-charged overtures of the rest of the men as well as because of his high status per se. The powerful male (staff member, manager, sponsor, etc.) could easily become the “protector” of the still “virgin” seductress, gaining through masking his own sexual interest what the other men could not gain by declaring theirs. However, this removal of the seductress from the sexual marketplace contained its own problems. The other men could resent the high-status male for winning the prize and resent the woman for her ability to get an “in” with the high-status male that they could not obtain as men. Al­though the seductress was rewarded for her femaleness and insured attention from the group, then, she was also the source of considerable tension; and needless to say, her perceived sexuality blotted out all other characteristics.

Men could adopt the role of protector toward an attractive woman, re­gardless of her collusion, and by implication cast her as sex object, reminding her and the rest of the group of her sexual status. In the guise of “helping” her, self-designated protectors may actually put up further barriers to the solitary woman’s full acceptance by inserting themselves, figuratively speaking, be­tween the woman and the rest of the group. A male management trainer typically offered token women in management assessment groups extra help and sympathetic attention to the problems their male peers might cause, tak­ing them out alone for drinks at the end of daily sessions. But this kind of “help” also preserved the sex object role.


The “pet” was adopted by the male group as a cute, amusing little thing and symbolically taken along on group events as mascot-a cheerleader for shows of prowess. Humor was often a characteristic of the pet. She was ex­pected to admire the male displays but not to enter into them; she cheered from the sidelines. Shows of competence on her part were treated as special and complimented just because they were unexpected (and the compliments themselves can be seen as reminders of the expected rarity of such behavior). One woman reported that when she was alone in a group of men and spoke at length on an issue, comments to her by men after the meeting often referred to her speech-making ability rather than the content of what she said (e.g., “You talk so fluently”), whereas comments the men made to one another were almost invariably content- or issue-oriented. Competent acts that are taken for granted when performed by males were often unduly “fussed over” when per­formed by exempt women, considered precocious or precious-a kind of look­ what-she-did-and-she’s-only-a-woman attitude. Such attitudes on the part of men encouraged self-effacing, girlish responses on the part of solitary women (who, after all, may be genuinely relieved to be included and petted) and prevented them from realizing or demonstrating their own power and competence.

Iron Maiden

The “iron maiden” is a contemporary variation of the stereotypical roles into which strong women are placed. Women who failed to fall into any of the first three roles and, in fact, resisted overtures that would trap them in a role (such as flirtation) might consequently be responded to as “tough” or danger­ous. (One woman manager developed such a reputation in company branches throughout the country.) If a token insisted on full rights in the group, if she displayed competence in a forthright manner, or if she cut off sexual innu­endos, she could be asked, “You’re not one of those women’s libbers, are you?” Regardless of the answer, she was henceforth regarded with suspicion, undue and exaggerated shows of politeness (by inserting references to women into conversations, by elaborate rituals of not opening doors), and with dis­tance, for she was demanding treatment as an equal in a setting in which no person of her kind had previously been an equal. Women inducted into the “iron maiden” role were stereotyped as tougher than they are (hence the name) and trapped in a more militant stance than they might otherwise take. Whereas seductresses and pets, especially, incurred protective responses, iron maidens faced abandonment. They were left to flounder on their own and often could not find peers sympathetic to them when they had problems.

Responses of Tokens to Role Encapsulation

The dynamics of role entrapment tended to lead tokens to a variety of conservative and low-risk responses. The time and awkwardness involved in correcting mistaken impressions led some tokens to a preference for already-­established relationships, for minimizing change and stranger-contact in the work situation. It was also often easier to accept stereotyped roles than to fight them, even if their acceptance meant limiting the tokens’ range of expressions or demonstrations of task competence, because they offered a comfortable and certain position. The personal consequence for tokens, of course, was a mea­sure of self-distortion. John Athanassiades, though not taking into account the effects of numerical representation, found that women in organizations tended to distort upward communication more than men, especially those with low risk-taking propensity, and argued that many observed work behaviors of women may be the result of such distortion and acceptance of organizational images. Submissiveness, frivolity, or other attributes may be feigned by peo­ple who feel they are prescribed by the dominant organizational culture. [25] This suggests that accurate conclusions about work attitudes and behavior can­not be reached by studying people in the token position, since there may always be an element of compensation or distortion involved.

The analysis also suggests another way in which tokenism can be self-per­petuating: acceptance of role encapsulation and attendant limitations on dem­onstration of competence may work to keep down the numbers of women in the upper ranks of the organization, thus continuing to put people in token positions. Role encapsulation confirms dominants’ stereotypes and proves to them how right they were all along. On the other hand, some women try to stay away from the role traps by bending over backwards not to exhibit any characteristics that would reinforce stereotypes. This strategy, too, is an un­easy one, for it takes continual watchful effort, and it may involve unnatural self-distortion. Finally, token women must steer a course between protec­tiveness and abandonment. Either they allow other people to take over and fight their battles for them, staying out of the main action in stereotypical ways, or they stand much too alone. They may be unable by virtue of scarcity even to establish effective support systems of their own.



The examination of numerical effects leads to the additional question of tipping points: How many of a category are enough to change a person’s status from token to full group member? When does a group move from skewed to tipped to balanced? What is the impact for a woman of the presence of another?

In the exempt ranks of Indesco, there were a number of instances of situa­tions in which two rather than one woman were found among male peers, but still constituted less than 20 percent of the group. Despite Solomon Asch’s classic laboratory finding that one potential ally can be enough to reduce the power of the majority of secure conformity, [26] in the two-token situation in or­ganizations, dominants several times behaved in ways that defeated an alliance between the two women. This was done through setting up invidious compari­sons. One woman was characteristically set up as superior, and the other as in­ferior–exaggerating traits in both cases. One was identified as the success, the other as the failure. The one given the success label felt relieved to be in­cluded and praised, recognizing that alliance with the identified failure would jeopardize her acceptance. The consequence, in one office, was that the iden­tified success stayed away from the other woman and did not give her any help with her performance, withholding criticism she had heard that might have been useful, and the second woman soon left. In another case, a layer of the hierarchy was inserted between two women who were at the same level: one was. made the boss of the other, causing great strain between them. Dominants also could defeat alliances, paradoxically, by trying to promote them. Two women in a training group of twelve were treated as though they were an automatic pair, and other group members felt that they were relieved of re­sponsibility for interacting with or supporting the women. The women reacted to this forced pairing by trying to create difference and distance between them and becoming extremely competitive.

Thus, structural circumstances and pressures from the majority could further produce what appeared to be intrin­sically prejudicial responses of women to each other. There were also in­stances in which two women developed a close alliance and refused to be turned against each other. Strong identification with the feminist cause or with other women was behind such alliances. Allied, two tokens could reduce some of the pressures and avoid some of the traps in their position. They could share the burden of representing womankind, and they could each be active on some pieces of “the woman’s slot” while leaving time free to demonstrate other abilities and interests. Two women personnel trainers, for example, on a six-person staff, could share responsibility for programs on women without ei­ther of them becoming over-identified with it.

A mere shift in absolute numbers, then, as from one to two tokens, could potentially reduce stresses in a token’s situation even while relative numbers of women remained low. But two were also few enough to be rather easily divided and kept apart. It would appear that larger numbers are necessary for supportive alliances to develop in the token context.

But these issues were part of a transitional status out of which new-hires soon passed, and, in any event, men did not so routinely or dramatically encounter them. Similarly, age and experience helped women make a satisfactory accom­modation, and over time, many women settled into comfortable and less token-like patterns. Some said that there was no problem they could not handle with time and that the manifestations of discrimination in their jobs were trivial. But still, the issues stemming from rarity and scarcity arose for women in every new situation, with new peers, and at career transitions. Even successful women who reported little or no discrimination said that they felt they had to “work twice as hard” and expend more energy than the average man to succeed. It is also clear that not all women in the token situa­tion behave alike or engender the same responses in others. There was variety in the individual choices, and there were alternative strategies for managing the situation. But a system characteristic–the numerical proportion in women and men were found-set limits on behavioral possibilities and de­fined the context for peer interaction.

The token position contains a number of dilemmas and contradictions: [27]

-Tokens are simultaneously representatives and exceptions. They serve as symbols of their category, especially when they fumble, yet they also are seen as unusual examples of their kind, especially when they succeed. -They are made aware of their differences from the numerical dominants, but then must often pretend that the differences do not exist, or have no implications.

-Tokens are among the most visible and dramatized of performers, no­ticeably on stage, yet they are often kept away from the organizational back­stage where the dramas are cast.

-Tokens are the quintessential “individuals” in the organization, since they stand apart from the mass of peer group members; yet they lose their in­dividuality behind stereotyped roles and carefully constructed public personae that can distort their sense of self.

-Those situations in which organizational peers are supposedly “relax­ing” (after-work drinks, celebratory dinners, sports events) are often the most stressful for tokens, for on such occasions the protection of defined positions and structured interaction disappears. So tokens,






The point is not that all of these things happen to token women, or that they happen only to people who are tokens. Some young men at Indesco complained that as new-hires they, too, felt performance pressures, uncertain­ties about their acceptance, and either over-protected or paradoxically, may be most relaxed and feel the most “natural” during the official parts of the business day when other people are the most constrained by formal roles.

-Tokens suffer from their loneness, yet the dynamics of interaction around them create a pressure for them to seek advantage by dissociating themselves from others of their category and, hence, to remain alone.

-As long as numbers are low, disruptions of interaction around tokens (and their personal problems) are seen by the organization as a huge deflection from its central purposes, a drain of energy, leading to the conclusion that it is not worth having people like the tokens around. Yet the disruptions are pri­marily a function of the numbers being low and could be remedied by propor­tional increases.

In short, organizational, social, and personal ambivalence surrounds peo­ple in token situations. It is likely that the burdens carried by tokens in the management of social relations take their toll in psychological stress, even if the tokens succeed in work performance. Research on people with inconsis­tent or poorly crystallized statuses has identified a number of psycho-social difficulties, including unsatisfactory social relationships, unstable self-images, frustration from dealing with contradictory demands (from others as well as the self), and insecurity. More serious physical and mental stress has also been found to be associated with status incongruities and from role pressures at work. [28]

Even the best coping strategy is likely to have some internal repercus­sions, ranging from inhibition of self-expression to feelings of inadequacy and, perhaps, self-hatred. Sidney Jourard hypothesized that one of the “lethal” aspects of the male role (that literally kills men off at an early age) is the inhibi­tion of self-disclosure. [29] Self-repression and refraining from certain kinds of expressiveness are, as we have seen, part of the culture in large organizations. But tokens of any kind are especially in a position where true disclosure to peers is not possible, and tokens may not even easily join work peers in their characteristic modes of tension release. Finally, to the extent that tokens ac­cept their exceptional status, dissociate themselves from others of their cate­gory, and turn against them, tokens may be denying parts of themselves and engaging in self-hatred. This can produce inner tension.

There is a small positive psychological side to tokenism: the self-esteem that comes from mastering a difficult situation and from getting into places that traditionally exclude others of one’s kind. If the token can segregate conflicting expectations and has strong outside support groups with which to relax, then perhaps a potentially stress-producing situation can be turned into an opportu­nity for ego enhancement. Indeed, one study showed that people whose racial ethnic statuses were “lower”, than their Occupational ranks on social prestige scales (the upwardly mobile) did not report the physical symptoms otherwise associated with inconsistent statuses. [30] But, on balance, token situations seem more stressful than beneficial.




What have been identified as the major issues in the situation of the numerically few at Industrial Supply Corporation are also characteristic of the token position in general. The same pressures and processes can occur around people of any social category who find themselves few of their kind among others of a different social type, as a few examples indicate. Bernard Segal studied the male token situation in a hospital in which 22 out of 101 nurses were men. He found that male nurses were isolates in the hospital social struc­ture–not because the men dissociated themselves from their women peers but because the women felt the men were out of place and should not be nurses. The male and female nurses had the same objective rank, but people of both sexes felt that the men’s subjective status was lower. The women placed the men in stereotypical positions, expecting them to do the jobs the women found distasteful or considered “men’s work.” [31] One male nursing student whom 1 interviewed reported that he thought he would enjoy being the only man in a group of women. Then he found that he engendered a great deal of hostility and that he was teased every time he failed to live up to a manly image–e.g., if he was vague or subjective in speech. The content of interaction when men are tokens may appear to give them an elevated position, but the process is still one of role encapsulation and treating tokens as sym­bols. Deference can be a patronizing reminder of difference, too.

Similarly, a blind man indicated that when he was the only blind person among sighted people, he often felt conspicuous and attended to more than he would like, creating pressure for him to work harder to prove himself. In the solo situation, he was never sure that he was getting the same treatment as other members of the group (first fellow students, later fellow members of an academic situation), and he suspected that people tended to protect him. When he was a token, compared with balanced situations in which other blind people were present, sighted people felt free to grab his arm and pull him along and were more likely to apologize for references to visual matters, rein­forcing his sense of being different and being cast in the role of someone who is more helpless than he, in tact, perceived himself to be.

People’s treatment, then, is not automatically fixed by inflexible charac­teristics but depends on their numbers in a particular situation. Change in the behavior and treatment of women in token positions is strongly tied to shifting proportions. But to argue for the importance of numbers smacks of advocacy of quotas, and many Americans object to quotas. Quantitative limits on expan­sion (which is how quotas are seen: not as more jobs for women, but fewer for men) have always seemed objectionable to some in the United States, espe­cially when “individual rights” for the advantaged are involved, as shown in concerns over gasoline rationing or income ceilings. Yet, it seems clear that numbers, especially relative numbers, can strongly affect a person’s fate in an organization. This is a system rather than an individual construct-located not in characteristics of the person but in how many people, like that person in sig­nificant ways, are also present. System phenomena require system-level inter­vention to make change.

In the absence of external pressures for change, tokenism is a self-per­petuating system. Tokens of Type O who are successful in their professional roles face pressures and inducements to dissociate themselves from other O’s, and thus they may fail to promote, or even actively block, the entry of more O’s. At the same time, tokens who are less than successful and appear less than fully  competent confirm the organization’s decision not to recruit more O’s, unless they are extraordinarily competent and not like most O’s. And since just a few O’s can make the X-majority people feel uncomfortable, X’s will certainly not go out of their way to include more O’s. In short, outside intervention is required to break the cycles created by the social composition of groups.

Most arguments made in favor of numerical guidelines in hiring and job placement limit their own effectiveness by making only part of the case. They say that equality of opportunity is the goal, but that this goal is hard to mea­sure without proof of outcome. Therefore, numbers hired serve as a shorthand for, a measure of, non-discrimination in selection. However, there is also a strong case that can be made for number-balancing as a worthwhile goal in it­self, because, inside the organization, relative numbers can playa large part in further outcomes-from work effectiveness and promotion prospects to psy­chic distress.


Notes for Chapter 8

  1. There are two questions that must be answered, of course, in any analysis of the effects of proportional representation: first, the circumstances under which a given characteristic is “salient” for differentiating group members; and second, the nature of the boundaries defining the extent and limits of the “group” within which proportional representation is measured. Here I am defin­ing ascribed characteristics with a physical manifestation (such as sex) as salient, and the “group” as work peers in equivalent statuses.
  2. Kurt H. Wolff, The Sociology of Georg Simmel (Glencoe, Illinois: Free Press, 1950).
  3. On juries: Fred L. Strodtbeck and Richard D. Mann, “Sex Role Differentiation in Jury Deliberations,” Sociometry, ‘9 (March 1956): pp. 3-11; and Fred L. Strodtbeck, Rita M. James, and Charles Hawkins, “Social Status in Jury Deliberations:’ American Sociological Redew, 22 (1957): pp. 7’3-‘9′ On the kibbutz: Lionel Tiger and Joseph Shepher, Women in the KibbutzNew York: Harcourt, Brace Jovanovich, 1975), and my critique in terms of sex ratios, Rosabeth Moss Kanter, “Interpreting the Results of a Social Experiment:’ Science, ’92 (14 May 1976): pp. 662-63. “Behavioral demography” as the study of population ratios: Marcia Guttentag, Too Many Women (New York: Basic Books, in press).
  1. Shelley E. Taylor and Susan Fiske, “The Token in the Small Group: Research Findings and Theoretical Implications,” in Psychology and Politics, J. Sweeney, ed. (New Haven: Yale Uni­versity Press, ‘976).
  2. Margaret Hennig, Career Development for Women Executives, Unpublished Doctoral Dissertation, Harvard Business School, ‘970, p. vi–2l. The experimental study of high­ dominance women was Edwin 1. Megaree, “Influence of Sex Role.’ on the Manifestation of Lead­ership,” Journal of Applied Psychology, 53 (196g): pp. 377–82. Cynthia Epstein’s book is Woman’s Place (Berkeley: University of California Press, ‘970). See also Edith M. Lynch, The Executive Suite: Feminine Style (New York: AMACOM, 1973); Margaret Cussler, The Woman Executive (N ew York: Harcourt, Brace, 1958).
  3. See Adeline Levine and Janice Crumrine, “Women and the Fear of Success: A Problem in Replication,” American journal of Sociology, 80 (January 1975): pp. 967-7+ Seymour Sarason’s argument is in his “Jewishness, Blackness, and the Nature-Nurture Controversy,” American Psy­chologist, 28 (November ‘973): pp. 962-71.
  4. The Hughes citation is to Everett Hughes, Men and Their Work (Glencoe, Illinois: Free Press, 1958), p. 109. See also Hughes, “Dilemmas and Contradictions of Status,” American Jour­nal of Sociology, 50 (March 1944): pp. 353-59. Judith Lorber came to some similar conclusions in “Trust, Loyalty, and the Place of Women in the Informal Organization of Work,” presented at the ‘975 Meetings of the American Sociological Association. On “status incongruence” as an explana­tory variable in group member behavior, used in many of the same ways I use “tokenism” but without the explicit numerical connotations, see A. Zaleznik, C. R. (Christensen, and F. J. Roeth­lisberger, The Motivation, Productivity, and Satisfaction of Workers: A Prediction Study (Boston: Harvard Business School Division of Research, 1958), especially pp. 56-68.
  1. Jeane Kirkpatrick, Political Woman (New York: Basic Books, ‘974), p. “3; Lionel Tiger, Men in Groups (New York: Random House, 1969).
  2. Clearly I was limited in first-hand observations of how the men acted when alone, since, by definition, if I, as a female researcher, were present, they would not have been alone. For my data here I relied on tape recordings of several meetings in which the tape was kept running even during breaks, and on informants’ reports immediately after informal social events and about meetings.
  3. For supportive laboratory evidence, see Elizabeth Aries, Interaction Patterns and Themes of Male, Female, and Mixed Groups, Unpublished Doctoral Dissertation, Harvard Uni­versity, ‘973. Her work is discussed further in Appendix II.
  4. For examples in another context, see Marcia Greenbaum, “Adding ‘Kenntnis’ to ‘Kirch, Kuche, und Kinder: ” Issues in Industrial Society, 2 (1971): pp. 61-68.
  5. Judith Long Laws, “The Psychology of Tokenism: An Analysis,” Se, Roles, 1 (1975): pp.51-67.
  6. Hughes, “Men and Their Work” and “Dilemmas and Contradictions of Status”: Lorber, “Trust, Loyalty, and the Place of Women.
  7. 14- Zaleznik et aI., Motivation, Productivity, and Satisfaction of Workers; Hennig, Career Development for Women Executives; Epstein, Woman’s Place; Brigid O’Farrell, “Affirmative Ac­tion and Skilled Craft Work,” Unpublished Report, Cambridge, Mass., 1973 (available from Center for Research on Women, Wellesley College); Carol Wolman and Hal Frank, “The Solo Woman in a Professional Peer Group,” American Journal of Orthopsychiatry, 45 January 1975): pp. 164-71.
  1. On secrecy and loyalty to peers in training, see Donald Roy, “Quota Restriction and Goldbricking in a Machine Shop,” American Journal of Sociology, 57 (March 1952): pp. 427-42; Howard S. Becker and Anselm L. Strauss, “Careers, Personality, and Adult Socialization,” Ameri­can Journal of Sociology, 62 (November 1956): pp. 253-D3·
  2. Bernard E. Segal, “Male Nurses: A Study in Status Contradiction and Prestige Loss,”Social Forces, 41 (October 1962): pp. 31-38. Personal interviews were used as a supplement.
  1. Hughes, “Dilemmas and Contradictions of Status.”
  2. Laws, “Psychology of Tokenism.”
  3. Rose Laub Coser, “Laughter Among Colleagues: A Study of the Social Functions of Humor Among the Staff of a Mental Hospital,” Psychiatry, 23 (1960): pp. 81-95·
  4. Everett C. Hughes, “Race Relations in Industry,” in Industry and Society, W. F. Whyte, ed. (New York: McGraw-Hill, 1946), p. “5.
  1. Marianne Abeles Ferber and Joan Althaus Huber, “Sex of Student and Instructor: A Study of Student Bias,” American Journal of Sociology, 80 (January 1975): pp. 94g–B3.
  2. Annual Report of the Council of Economic Advisers (Washington, D.C.: U.S. Govern­ment Printing Office, 1973), p. 106.
  3. Segal, “Male Nurses,” supplemented by my personal interviews.
  4. Philip Rieff, ed., Freud: Sexuality and the Psychology of Love (New York: Collier Books, 1963); Bryan Strong, “Toward a History of the Experiential Family: Sex and Incest in the Nine­teenth Century Family,” Journal of Marriage and the Family, 35 (August 1973): pp. 457-D6·
  5. John C. Athanassiades, “An Investigation of Some Communication Patterns of Female Subordinates in Hierarchical Organizations,” Human Relations, 27 (1974): pp. 195-209.
  1. Solomon E. Asch, “Effects of Group Pressure on the Modification and Distortion of Judg­ment,” in Group Dynamics, second edition, D. Cartwright and A. Zander, eds. (Evanston, Illi­nois: Row Peterson, 1960), pp. 189-200.
  2. Hughes used this phrase in a slightly different way in “Dilemmas and Contradictions of Status.”
  3. Gerhard Lenski, “Status Crystallization: A Non-Vertical Dimension of Status,” American Sociological Review, 19 (August 1954): pp. 405-13; Lenski, “Social Participation and the Crys­tallization of Status,” American Sociological Review, 21 (August 1956): pp. 458-64; G. H. Fenchel, J. H. Monderer, and E. L. Hartley, “Subjective Status and the Equilibrium Hypothesis, “Journal of Abnormal and Social Psychology, 45 (October 1951): pp. 476-79; Elton F. Jackson, “Status Con­sistency and Symptoms of Stress,” American Sociological Review, 27 (August 1962): pp. 4~0. On role overload and coronary disease see Stephen M. Sales, “Organizational Roles as a Risk Fac­tor in Coronary Heart Disease,” Administrative Science Quarterly, 14 (196g): pp. 235-336. For general reviews, see James S. House, “The Effects of Occupational Stress on Physical Health,” in Work and the Quality of Life, James O’Toole, ed. (Cambridge, Mass.: MIT Press, 1974), pp. 145–’70; and Stanislav V. Kasl, “Work and Mental Health,” in Work and the Quality of Life, James O’Toole, ed. (Cambridge, Mass.: MIT Press, 1974), pp. 171-g6.
  4. Sidney M. Jourard, The Transparent Self: Self-Disclosure and Well-Being (Princeton: D, Van Nostrand, 1964).
  1. Jackson, “Status Consistency and Symptoms of Stress.”
  2. Segal, “Male Nurses.”

Library of Congress Cataloging in Publication Data


Copyright ©  1976 by Marge Piercy. Reprinted by permission of Alfred A. Knopf, Inc.

Men and women of the corporation.

Bibliography: p. 326 Includes index.

  1. Organizational behavior. 2. Industrial organization. 3. White collar workers. 4. Women in business. I. Title.

HD58.7.K36    301.18’32        76-43464

ISBN: 0-465-04452-2 (cloth) ISBN: 0-465-04453-0 (paper)


Copyright © 1977 by Rosabeth Moss Kanter Printed in the United States of America


Designed by Vincent Torre




pete padilla 


Leave a Comment

Your email address will not be published. Required fields are marked *