GENDER

10.1177/0891243204265349
Risman / GENDER
& SOCIETY
AS A/ August
SOCIAL 2004
STRUCTURE

Sociologists for Women in Society Feminist Lecture

GENDER AS A SOCIAL STRUCTURE

Theory Wrestling with Activism

BARBARA J. RISMAN
North Carolina State University

In this article, the author argues that we need to conceptualize gender as a social structure, and by doing
so, we can better analyze the ways in which gender is embedded in the individual, interactional, and
institutional dimensions of our society. To conceptualize gender as a structure situates gender at the
same level of general social significance as the economy and the polity. The author also argues that while
concern with intersectionality must continue to be paramount, different structures of inequality have dif-
ferent constructions and perhaps different influential causal mechanisms at any given historical
moment. We need to follow a both/and strategy to understand gender structure, race structure, and other
structures of inequality as they currently operate while also systematically paying attention to how these
axes of domination intersect. Finally, the author suggests we pay more attention to doing research and
writing theory with explicit attention to how our work can indeed help transform as well as inform
society.

Keywords: feminist theory; gender theory; social structure; social change

Gender has become a growth industry in the academy. In the years between my
own college education and today, we have moved from not enough having been
published in 1972 to justify my writing a literature review for an undergraduate
course paper to more sociologists’ studying and teaching about gender than any
other single substantive area in American society. In 1998, I published Gender Ver-
tigo: American Families in Transition (Risman 1998), which offered both a histori-
cal narrative about how the field of gender had developed and an integrative

AUTHOR’S NOTE: There are too many scholars who have read this work and helped to improve it to
thank each and every one. I do owe a great deal to the feminist intellectual community of Sociologists for
Women in Society. Special thanks are due to Shannon Davis, Patricia Yancey Martin, Michael Schwalbe,
Donald Tomaskovic-Devey, and the students in my 2003 and 2004 graduate seminars in sociology of the
family, sociology of gender, and feminist thought. All of them have helped improve my argument, but of
course I alone remain responsible for the content, flaws and all.
REPRINT REQUESTS: Barbara J. Risman, North Carolina State University, Department of Sociology,
Raleigh, NC 27695-8107.

GENDER & SOCIETY, Vol. 18 No. 4, August 2004 429-450

DOI: 10.1177/0891243204265349
© 2004 Sociologists for Women in Society
429
430 GENDER & SOCIETY / August 2004

theoretical explanation for the tenacity of gender stratification in families. In this

article, I briefly summarize my earlier argument that gender should be conceptual-
ized as a social structure (Risman 1998) and extend it with an attempt to classify the
mechanisms that help produce gendered outcomes within each dimension of the
social structure. I then provide evidence from my own and others’ research to sup-
port the usefulness of this theoretical schema. Finally, using gender structure as a
starting point, I engage in conversation with ideas currently emerging about
intersectionality and wrestle with how we might use theory in the service of social
change.

GENDER AS SOCIAL STRUCTURE

With this theory of gender as a social structure, I offer a conceptual framework,

a scheme to organize the confusing, almost limitless, ways in which gender has
come to be defined in contemporary social science. Four distinct social scientific
theoretical traditions have developed to explain gender. The first tradition focuses
on how individual sex differences originate, whether biological (Udry 2000) or
social in origin (Bem 1993). The second tradition, perhaps portrayed best in
Epstein’s (1988) Deceptive Distinctions, emerged as a reaction to the first and
focuses on how the social structure (as opposed to biology or individual learning)
creates gendered behavior. The third tradition, also a reaction to the individualist
thinking of the first, emphasizes social interaction and accountability to others’
expectations, with a focus on how “doing gender” creates and reproduces inequal-
ity (West and Zimmerman 1987). The sex-differences literature, the doing gender
interactional analyses, and the structural perspectives have been portrayed as
incompatible in my own early writings as well as in that of others (Fuchs Epstein
1988; Kanter 1977; Ferree 1990; Risman 1987; Risman and Schwartz 1989). Eng-
land and Browne (1992) argued persuasively that this incompatibility is an illusion:
All structural theories must make assumptions about individuals, and individualist
theories must make presumptions about external social control. While we do gen-
der in every social interaction, it seems naive to ignore the gendered selves and cog-
nitive schemas that children develop as they become cultural natives in a patriarchal
world (Bem 1993). The more recent integrative approaches (Connell 2002; Lorber
1994; Ferree, Lorber, and Hess 1999; Risman 1998) treat gender as a socially con-
structed stratification system. This article fits squarely in the current integrative tra-
dition.
Lorber (1994) argued that gender is an institution that is embedded in all the
social processes of everyday life and social organizations. She further argued that
gender difference is primarily a means to justify sexual stratification. Gender is so
endemic because unless we see difference, we cannot justify inequality. Lorber
provided much cross-cultural, literary, and scientific evidence to show that gender
difference is socially constructed and yet is universally used to justify stratification.
She wrote that “the continuing purpose of gender as a modern social institution is to
 Risman / GENDER AS A SOCIAL STRUCTURE 431

construct women as a group to be subordinate to men as a group” (p. 33). I share this
presumption that the creation of difference is the very foundation on which
inequality rests.
Martin (forthcoming) extended Lorber’s (1994) use of the term “institution” in
her argument that gender should be conceptualized as such. She identified the crite-
ria for a social institution as follows: (1) Characteristic of groups; (2) persists over
time and space; (3) includes distinct social practices; (4) constrains and facilitates
behavior/action; (5) includes expectations, rule/norms; (6) is constituted and
reconstituted by embodied agents; (7) is internalized as identities and selves; (8)
includes a legitimating ideology; (9) is contradictory, rife with conflict; (10)
changes continuously; (11) is organized by and permeated with power; and (12) is
mutually constituted at different levels of analysis. I build on this notion of gender
as an institution but find the institutional language distracting. The word “institu-
tion” is too commonly used to refer to particular aspects of society, for example, the
family as an institution or corporations as institutions. My notion of gender struc-
ture meets the criteria offered by Martin (forthcoming) as well. While the language
we use may differ, our goals are complementary, as we seek to situate gender as
embedded not only in individuals but throughout social life (Patricia Martin,
personal communication).
I prefer to define gender as a social structure because this brings gender to the
same analytic plane as politics and economics, where the focus has long been on
political and economic structures. While the language of structure suits my pur-
poses, it is not ideal because despite ubiquitous usage in sociological discourse, no
definition of the term “structure” is widely shared. Smelser (1988) suggested that
all structuralists share the presumption that social structures exist outside individ-
ual desires or motives and that social structures at least partially explain human
action. Beyond that, consensus dissipates. Blau (1977) focused solely on the con-
straint collective life imposes on the individual. In their influential work, Blau and
his colleagues (e.g., Blau 1977; Rytina et al. 1988) argued that the concept of struc-
ture is trivialized if it is located inside an individual’s head in the form of internal-
ized norms and values. Blau focused solely on the constraint collective life imposes
on the individual; structure must be conceptualized, in his view, as a force opposing
individual motivation. Structural concepts must be observable, external to the indi-
vidual, and independent of individual motivation. This definition of “structure”
imposes a clear dualism between structure and action, with structure as constraint
and action as choice.
Constraint is, of course, an important function of structure, but to focus only on
structure as constraint minimizes its importance. Not only are women and men
coerced into differential social roles; they often choose their gendered paths. A
social structural analysis must help us understand how and why actors choose one
alternative over another. A structural theory of action (e.g., Burt 1982) suggests that
actors compare themselves and their options to those in structurally similar posi-
tions. From this viewpoint, actors are purposive, rationally seeking to maximize
their self-perceived well-being under social-structural constraints. As Burt (1982)
432 GENDER & SOCIETY / August 2004

suggested, one can assume that actors choose the best alternatives without presum-
ing they have either enough information to do it well or the options available to
make choices that effectively serve their own interests. For example, married
women may choose to do considerably more than their equitable share of child care
rather than have their children do without whatever “good enough” parenting
means to them if they see no likely alternative that the children’s father will pick up
the slack.
While actions are a function of interests, the ability to choose is patterned by the
social structure. Burt (1982) suggested that norms develop when actors occupy
similar network positions in the social structure and evaluate their own options vis-
à-vis the alternatives of similarly situated others. From such comparisons, both
norms and feelings of relative deprivation or advantage evolve. The social structure
as the context of daily life creates action indirectly by shaping actors’perceptions of
their interests and directly by constraining choice. Notice the phrase “similarly sit-
uated others” above. As long as women and men see themselves as different kinds
of people, then women will be unlikely to compare their life options to those of
men. Therein lies the power of gender. In a world where sexual anatomy is used to
dichotomize human beings into types, the differentiation itself diffuses both claims
to and expectations for gender equality. The social structure is not experienced as
oppressive if men and women do not see themselves as similarly situated.
While structural perspectives have been applied to gender in the past (Epstein
1988; Kanter 1977), there has been a fundamental flaw in these applications.
Generic structural theories applied to gender presume that if women and men were
to experience identical structural conditions and role expectations, empirically
observable gender differences would disappear. But this ignores not only internal-
ized gender at the individual level (which indeed purely structural theorists deny
exists) but the cultural interactional expectations that remain attached to women
and men because of their gender category. A structural perspective on gender is
accurate only if we realize that gender itself is a structure deeply embedded in
society.
Giddens’s (1984) structuration theory adds considerably more depth to this
analysis of gender as a social structure with his emphasis on the recursive relation-
ship between social structure and individuals. That is, social structures shape indi-
viduals, but simultaneously, individuals shape the social structure. Giddens
embraced the transformative power of human action. He insisted that any structural
theory must be concerned with reflexivity and actors’ interpretations of their own
lives. Social structures not only act on people; people act on social structures.
Indeed, social structures are created not by mysterious forces but by human action.
When people act on structure, they do so for their own reasons. We must, therefore,
be concerned with why actors choose their acts. Giddens insisted that concern with
meaning must go beyond the verbal justification easily available from actors
because so much of social life is routine and so taken for granted that actors will not
articulate, or even consider, why they act.
 Risman / GENDER AS A SOCIAL STRUCTURE 433

This nonreflexive habituated action is what I refer to as the cultural component

of the social structure: The taken for granted or cognitive image rules that belong to
the situational context (not only or necessarily to the actor’s personality). The cul-
tural component of the social structure includes the interactional expectations that
each of us meet in every social encounter. My aims are to bring women and men
back into a structural theory where gender is the structure under analysis and to
identify when behavior is habit (an enactment of taken for granted gendered cul-
tural norms) and when we do gender consciously, with intent, rebellion, or even
with irony. When are we doing gender and re-creating inequality without intent?
And what happens to interactional dynamics and male-dominated institutions
when we rebel? Can we refuse to do gender or is rebellion simply doing gender dif-
ferently, forging alternative masculinities and femininities?
Connell (1987) applied Giddens’s (1984) concern with social structure as both
constraint and created by action in his treatise on gender and power (see particu-
larly chapter 5). In his analysis, structure constrains action, yet “since human action
involves free invention . . . and is reflexive, practice can be turned against what con-
strains it; so structure can deliberately be the object of practice” (Connell 1987, 95).
Action may turn against structure but can never escape it. We must pay attention
both to how structure shapes individual choice and social interaction and to how
human agency creates, sustains, and modifies current structure. Action itself may
change the immediate or future context.
A theory of gender as a social structure must integrate this notion of causality as
recursive with attention to gender consequences at multiple levels of analysis. Gen-
der is deeply embedded as a basis for stratification not just in our personalities, our
cultural rules, or institutions but in all these, and in complicated ways. The gender
structure differentiates opportunities and constraints based on sex category and
thus has consequences on three dimensions: (1) At the individual level, for the
development of gendered selves; (2) during interaction as men and women face dif-
ferent cultural expectations even when they fill the identical structural positions;
and (3) in institutional domains where explicit regulations regarding resource
distribution and material goods are gender specific.

Advantages to Gender Structure Theory

This schema advances our understanding of gender in several ways. First, this
theoretical model imposes some order on the encyclopedic research findings that
have developed to explain gender inequality. Thinking of each research question as
one piece of a jigsaw puzzle, being able to identify how one set of findings coordi-
nates with others even when the dependent variables or contexts of interest are dis-
tinct, furthers our ability to build a cumulative science. Gender as a social structure
is enormously complex. Full attention to the web of interconnection between
gendered selves, the cultural expectations that help explain interactional patterns,
and institutional regulations allows each research tradition to explore the growth of
their own trees while remaining cognizant of the forest.
434 GENDER & SOCIETY / August 2004

A second contribution of this approach is that it leaves behind the modernist

warfare version of science, wherein theories are pitted against one another, with a
winner and a loser in every contest. In the past, much energy (including my early
work; Risman 1987) was devoted to testing which theory best explained gender
inequality and by implication to discounting every alternative possibility.1 While
this is perhaps an effective technique for building academic careers, as a model for
explaining complex social phenomena, it leaves much to be desired. Theory build-
ing that depends on theory slaying presumes parsimony is always desirable, as if
this complicated world of ours were best described with simplistic monocausal
explanations. While parsimony and theory testing were the model for the twentieth-
century science, a more postmodern science should attempt to find complicated
and integrative theories (Collins 1998). The conceptualization of gender as a social
structure is my contribution to complicating, but hopefully enriching, social theory
about gender.
A third benefit to this multidimensional structural model is that it allows us to
seriously investigate the direction and strength of causal relationships between
gendered phenomena on each dimension. We can try to identify the site where
change occurs and at which level of analysis the ability of agentic women and men
seem able, at this historical moment, to effectively reject habitualized gender rou-
tines. For example, we can empirically investigate the relationship between gend-
ered selves and doing gender without accepting simplistic unidirectional argu-
ments for inequality presumed to be either about identities or cultural ideology. It is
quite possible, indeed likely, that socialized femininity does help explain why we
do gender, but doing gender to meet others’ expectations, surely, over time, helps
construct our gendered selves. Furthermore, gendered institutions depend on our
willingness to do gender, and when we rebel, we can sometimes change the institu-
tions themselves. I have used the language of dimensions interchangeably with the
language of levels because when we think of gender as a social structure, we must
move away from privileging any particular dimension as higher than another. How
social change occurs is an empirical question, not an a priori theoretical
assumption. It may be that individuals struggling to change their own identities
(as in consciousness-raising groups of the early second-wave women’s movement)
eventually bring their new selves to social interaction and create new cultural
expectations. For example, as women come to see themselves (or are socialized to
see themselves) as sexual actors, the expectations that men must work to provide
orgasms for their female partners becomes part of the cultural norm. But this is
surely not the only way social change can happen. When social movement activists
name as inequality what has heretofore been considered natural (e.g., women’s seg-
regation into low-paying jobs), they can create organizational changes such as
career ladders between women’s quasi-administrative jobs and actual manage-
ment, opening up opportunities that otherwise would have remained closed, thus
creating change on the institutional dimension. Girls raised in the next generation,
who know opportunities exist in these workplaces, may have an altered sense of
 Risman / GENDER AS A SOCIAL STRUCTURE 435

possibilities and therefore of themselves. We need, however, to also study change

and equality when it occurs rather than only documenting inequality.
Perhaps the most important feature of this conceptual schema is its dynamism.
No one dimension determines the other. Change is fluid and reverberates through-
out the structure dynamically. Changes in individual identities and moral account-
ability may change interactional expectations, but the opposite is possible as well.
Change cultural expectations, and individual identities are shaped differently. Insti-
tutional changes must result from individuals or group action, yet such change is
difficult, as institutions exist across time and space. Once institutional changes
occur, they reverberate at the level of cultural expectations and perhaps even on
identities. And the cycle of change continues. No mechanistic predictions are pos-
sible because human beings sometimes reject the structure itself and, by doing so,
change it. Much time and energy can be wasted trying to validate which dimension
is more central to inequality or social change. Instead, the feminist project is better
served by finding empirical answers to particular questions and by identifying how
particular processes explain outcomes in need of change. If our goal is to do schol-
arship that contributes to transforming society, the identification of the processes
that explain particular outcomes is the first step in effectively changing those
processes and subsequently the outcomes themselves.

Social Processes Located by Dimension in the Gender Structure

When we conceptualize gender as a social structure, we can begin to identify

under what conditions and how gender inequality is being produced within each
dimension. The “how” is important because without knowing the mechanisms, we
cannot intervene. If indeed gender inequality in the division of household labor at
this historical moment were primarily explained (and I do not suggest that it is) by
gendered selves, then we would do well to consider the most effective socialization
mechanisms to create fewer gender-schematic children and resocialization for
adults. If, however, the gendered division of household labor is primarily con-
strained today by cultural expectations and moral accountability, it is those cultural
images we must work to alter. But then again, if the reason many men do not equita-
bly do their share of family labor is that men’s jobs are organized so they cannot
succeed at work and do their share at home, it is the contemporary American work-
place that must change (Williams 2000). We may never find a universal theoretical
explanation for the gendered division of household labor because universal social
laws may be an illusion of twentieth-century empiricism. But in any given moment
for any particular setting, the causal processes should be identifiable empirically.
Gender complexity goes beyond historical specificity, as the particular causal pro-
cesses that constrain men and women to do gender may be strong in one institu-
tional setting (e.g., at home) and weaker in another (e.g., at work).
The forces that create gender traditionalism for men and women may vary
across space as well as time. Conceptualizing gender as a social structure contrib-
436 GENDER & SOCIETY / August 2004

utes to a more postmodern, contextually specific social science. We can use this
schema to begin to organize thinking about the causal processes that are most likely
to be effective on each dimension. When we are concerned with the means by
which individuals come to have a preference to do gender, we should focus on how
identities are constructed through early childhood development, explicit socializa-
tion, modeling, and adult experiences, paying close attention to the internalization
of social mores. To the extent that women and men choose to do gender-typical
behavior cross-situationally and over time, we must focus on such individual expla-
nations. Indeed, much attention has already been given to gender socialization and
the individualist presumptions for gender. The earliest and perhaps most com-
monly referred to explanations in popular culture depend on sex-role training,
teaching boys and girls their culturally appropriate roles. But when trying to under-
stand gender on the interactional/cultural dimension, the means by which status
differences shape expectations and the ways in which in-group and out-group
membership influence behavior need to be at the center of attention. Too little atten-
tion has been paid to how inequality is shaped by such cultural expectations during
interaction. I return to this in the section below. On the institutional dimension, we
look to law, organizational practices, and formal regulations that distinguish by sex
category. Much progress has been made in the post–civil rights era with rewriting
formal laws and organizational practices to ensure gender neutrality. Unfortu-
nately, we have often found that despite changes in gender socialization and gender
neutrality on the institutional dimension, gender stratification remains.
What I have attempted to do here is to offer a conceptual organizing scheme for
the study of gender that can help us to understand gender in all its complexity and
try to isolate the social processes that create gender in each dimension. This is nec-
essary before we can begin to imagine how to change these processes and thus to
change the way we socially construct gender. Table 1 provides a schematic outline
of this argument.2

Cultural Expectations during Interaction and the Stalled Revolution

In Gender Vertigo (Risman 1998), I suggested that at this moment in history,

gender inequality between partners in American heterosexual couples could be
attributed particularly to the interactional expectations at the cultural level: the dif-
ferential expectations attached to being a mother and father, a husband and wife.
Here, I extend this argument in two ways. First, I propose that the stalled gender
revolution in other settings can similarly be traced to the interactional/cultural
dimension of the social structure. Even when women and men with feminist identi-
ties work in organizations with formally gender-neutral rules, gender inequality is
reproduced during everyday interaction. The cultural expectations attached to our
sex category, simply being identified as a woman or man, has remained relatively
impervious to the feminist forces that have problematized sexist socialization prac-
tices and legal discrimination. I discuss some of those processes that can help
 Risman / GENDER AS A SOCIAL STRUCTURE 437

TABLE 1: Dimensions of Gender Structure, by Illustrative Social Processes

Dimensions of the Gender Structure

Individual Interactional
Level Cultural Expectations Institutional Domain

Social Socialization Status expectations Organizational practices

Processesa Internalization Cognitive bias Legal regulations
Identity work Othering Distribution of resources
Construction Trading power Ideology
of selves for patronage
Altercasting

a. These are examples of social processes that may help explain the gender structure on each
dimension. They are meant to be illustrative and not a complete list of all possible social pro-
cesses or causal mechanisms.

explain why social interaction continues to reproduce inequality, even in settings

that seem ripe for social change.
Contemporary social psychological writings offer us a glimpse of possibilities
for understanding how inequality is reconstituted in daily interaction. Ridgeway
and her colleagues (Ridgeway 1991, 1997, 2001; Ridgeway and Correll 2000;
Ridgeway and Smith-Lovin 1999) showed that the status expectations attached to
gender and race categories are cross-situational. These expectations can be thought
of as one of the engines that re-create inequality even in new settings where there is
no other reason to expect male privilege to otherwise emerge. In a sexist and racist
society, women and all persons of color are expected to have less to contribute to
task performances than are white men, unless they have some other externally vali-
dated source of prestige. Status expectations create a cognitive bias toward privi-
leging those of already high status. What produces status distinction, however, is
culturally and historically variable. Thus, cognitive bias is one of the causal mecha-
nisms that help to explain the reproduction of gender and race inequality in every-
day life. It may also be an important explanation for the reproduction of class and
heterosexist inequality in everyday life as well, but that is an empirical question.
Schwalbe and his colleagues (2000, 419) suggested that there are other “generic
interactive processes through which inequalities are created and reproduced in
everyday life.” Some of these processes include othering, subordinate adaptation,
boundary maintenance, and emotion management. Schwalbe and his colleagues
suggested that subordinates’ adaptation plays an essential role in their own disad-
vantage. Subordinate adaptation helps to explain women’s strategy to adapt to the
gender structure. Perhaps the most common adaptation of women to subordination
is “trading power for patronage” (Schwalbe et al. 2000, 426). Women, as wives and
daughters, often derive significant compensatory benefits from relationships with
the men in their families. Stombler and Martin (1994) similarly showed how little
sisters in a fraternity trade affiliation for secondary status. In yet another setting,
438 GENDER & SOCIETY / August 2004

elite country clubs, Sherwood (2004) showed how women accept subordinate sta-
tus as “B” members of clubs, in exchange for men’s approval, and how when a few
wives challenge men’s privilege, they are threatened with social ostracism, as are
their husbands. Women often gain the economic benefits of patronage for them-
selves and their children in exchange for their subordinate status.
One can hardly analyze the cultural expectations and interactional processes
that construct gender inequality without attention to the actions of members of the
dominant group. We must pay close attention to what men do to preserve their
power and privilege. Schwalbe et al. (2000) suggested that one process involved is
when superordinate groups effectively “other” those who they want to define as
subordinate, creating devalued statuses and expectations for them. Men effectively
do this in subversive ways through “politeness” norms, which construct women as
“others” in need of special favors, such as protection. By opening doors and walk-
ing closer to the dirty street, men construct women as an “other” category, different
and less than independent autonomous men. The cultural significance attached to
male bodies signifies the capacity to dominate, to control, and to elicit deference,
and such expectations are perhaps at the core of what it means for men to do gender
(Michael Schwalbe, personal communication).
These are only some of the processes that might be identified for understanding
how we create gender inequality based on embodied cultural expectations. None
are determinative causal predictors, but instead, these are possible leads to reason-
able and testable hypotheses about the production of gender. I offer them as part of a
conceptual scheme to help us think about how different kinds of processes are
implicated at each dimension of the gender structure. Martin’s (2003) research on
men and women workers in a corporate setting can help illustrate how such a con-
ceptual scheme might work. She wrote about a male vice-president’s asking his
female counterpart to pick up a phone call, which she does unreflectively, but she
soon thereafter identifies this request as problematic. Martin presented this as an
example of how interactional status expectations attached to sex category create
inequality within professional relationships. This empirical example supports the
thesis that shared but routine cultural expectations re-create inequality even with-
out the conscious intent of the actors. Gender structure theory does not presume
that this man and woman do not bring gendered selves to the office to accept Mar-
tin’s analysis. In fact, one might suggest that a vice-president who had more thor-
oughly internalized traditional femininity norms would not have noticed the ineq-
uity at all. Nor does one need to have a company that has purged all discriminatory
practices from its policies to see the import of the cultural expectations that Martin
identified. A meta-analysis that looks at the effects of gender inequality in the
workplace should integrate findings about social processes at the level of individual
identities, cultural expectations, and organizational practices. In the next section of
this article, I provide empirical illustrations of this conceptual scheme of gender as
a social structure.
 Risman / GENDER AS A SOCIAL STRUCTURE 439

Empirical Illustrations

I begin with an example from my own work of how conceptualizing gender as a

social structure helps to organize the findings and even push forward an under-
standing of the resistance toward an egalitarian division of family work among con-
temporary American heterosexual couples. This is an area of research that incorpo-
rates a concern with nurturing children, housework, and emotional labor. My own
question, from as early as graduate school, was whether men could mother well
enough that those who care about children’s well-being would want them to do so.
Trained in the warfare model of science, my dissertation was a test of structural ver-
sus individualist theories (Kanter 1977) of men’s mothering. As someone who con-
sidered herself a structuralist of some generic sort, I hypothesized (Risman 1983)
that when men were forced into the social role of primary parent, they could
become just like mothers: The parenting role (e.g., a measure of family structure)
would wipe out the effects of individual gendered selves in my models. What I
found was, alas, more complicated. At the time, I concluded that men could
“mother” but did not do so in ways identical to women (Risman 1983). After having
been influenced by studies showing that tokenism worked differently when men
were the tokens (Williams 1992; Zimmer 1988) and that money could not buy
power in marriage for women quite as it seemed to for men (Brines 1994; Ferree
1990), I came to the realization that gender itself was a structure and would not dis-
appear when men and women were distributed across the variety of structural
positions that organize our social world.
To ask the question, Can men mother, presuming that gender itself is a social
structure leads us to look at all the ways that gender constrains men’s mothering and
under what conditions those change. Indeed, one of my most surprising, and unan-
ticipated, findings was that single fathers who were primary caretakers came to
describe themselves more often than other men with adjectives such as “nurturant,”
“warm,” and “child oriented,” those adjectives we social scientists use to measure
femininity. Single fathers’identities changed based on their experiences as primary
parents. In my research, men whose wives worked full-time did not, apparently, do
enough mothering to have such experiences influence their own sense of selves.
Most married fathers hoard the opportunity for leisure that frees them from the
responsibilities of parenting that might create such identity change. My questions
became more complicated but more useful when I conceptualized gender as a
social structure. When and under what conditions do gendered selves matter?
When do interactional expectations have the power to overcome previous internal-
ized predispositions? What must change at the institutional level allow for expecta-
tions to change at the interactional level? Does enough change on the interactional
dimension shift the moral accountability that then leads to collective action in
social organizations? Could feminist parents organize and create a social move-
ment that forces workplaces to presume that valuable workers also have family
responsibilities?
440 GENDER & SOCIETY / August 2004

These questions led me to try to identify the conditions that enable women and
men to actually succeed in creating egalitarian relationships. My next research pro-
ject was an in-depth interview and qualitative study of heterosexual couples raising
children who equally shared the work of earning a living and the family labor of
child care, homemaking, and emotion work. The first interesting piece of data was
how hard it was to find such people in the end of the twentieth century, even when
recruiting at daycare centers, parent-teacher associations, university venues, and
feminist newsletters (all in the southeastern United States). Three out of four volun-
teer couples failed the quite generous criteria for inclusion: Working approximately
the same number of hours in the labor force (within five hours per week), sharing
the household labor and child care tasks within a 60/40 split, and both partners’
describing the relationship as equitable. There are clearly fewer couples who live
equal lives than those who wish fervently that they did so.
What I did find from intensive interviews and home observations with 20 such
couples was that the conditions that enabled their success spread across each
dimension of the gender structure. Although I would have predicted otherwise
(having once been committed to a purely structural theory of human behavior),
selves and personalities matter. The women in my sample were strong, directive
women married to relatively laidback men. Given the overwhelming gendered
expectations for men’s privilege in heterosexual marriage, this should have been
expected, but to someone with my theoretical background, it was not. Less surpris-
ing to me, the women in these couples also had at least the income and career status
of their partners and often bettered them. But this is not usually enough to dent
men’s privilege, or we would have far more egalitarian marriages by now. In addi-
tion, these couples were ideologically committed to equality and to sharing. They
often tried explicitly to create social relationships with others who held similar val-
ues, for example, by joining liberal churches to meet like-minded others. Atypical
gendered selves and shared feminist-inspired cultural expectations were important
conditions for equality, but they were not enough. Men’s workplace flexibility mat-
tered as well. Nearly every father in this sample was employed in a job with flexible
working hours. Many women worked in jobs with flexibility as well, but not as uni-
formly as their male partners. These were privileged, educated workers for whom
workplace flexibility was sometimes simply luck (e.g., a father who lost a corporate
job and decided to sell real estate) but more often was a conscious choice (e.g., clin-
ical psychologists choosing to teach at a small college to have more control over
working hours despite decreased earning power). Thus, these couples experienced
enabling contexts at the level of their individual selves, feminist ideology to help
shape the cultural expectations in their most immediate environments (within the
dyad and among at least some friends), and the privilege within the economy to
have or find flexible jobs. By attending to each dimension of the gender structure, I
amassed a more effective explanation for their ability to negotiate fair relationships
than I could have without attention to selves, couple interaction, and their work-
places. The implications for feminist social change are direct: We cannot simply
attend to socializing children differently, nor creating moral accountability for men
 Risman / GENDER AS A SOCIAL STRUCTURE 441

to share family work, nor fighting for flexible, family-friendly workplaces. We

must attend to all simultaneously.
The research on gender in occupational settings (Williams 1992; Zimmer 1988)
and quantitative studies of household division of labor (Brines 1994; Greenstein
2000) also provide good examples of how using gender structure as a conceptual
framework can help organize meta-analytic reviews of the literature to create
cumulative knowledge. Kanter’s (1977) early structural hypotheses presumed that
tokenism per se was an important mechanism that explained women’s and men of
color’s continued subordination in the labor force. But as research testing this
tokenism hypothesis expanded to include men in women’s jobs, it became clear that
the theory was not indeed only about numbers. Tokenism did not work the same
way for white men. Men tokens rode glass escalators while women and racial
minorities hit glass ceilings (Reskin 1998; Williams 1992; Yoder 1991). Gender
and race remained important; the cultural interactional expectations remained dif-
ferent even in integrated work settings. Status expectations (Ridgeway 1991;
Ridgeway et al. 1998) favored men and devalued women, whatever their numbers.
We can conceptualize this as the interactional cultural level impeding further
changes that realignments on the institutional dimension would predict.
Similarly, quantitative research findings about the household division of labor
have made it quite clear that even when women work outside the home full-time,
they shoulder the majority of household and child care. Over time, researchers have
tested a variety of theories for why, sometimes presuming that as time pressures and
resources equalized between husbands and wives, so too would the burden of
household labor (Bianchi et al. 2000; Coverman 1985; Pleck 1985; Presser 1994;
Shelton 1992). Not so. The data are unequivocal. Even in dual worker families,
women do considerably more work and retain the majority of responsibility, even if
they do share (or perhaps delegate) some of the family work to husbands and chil-
dren. Sociology has provided solid evidence (Fenstermaker Berk 1985; Greenstein
1996, 2000; Robinson and Milkie 1998; Twiggs, McQuillan, and Ferree 1999) that
domestic work, whether cleaning toilets or changing diapers, is as much about the
production and display of gender as it is about clean toilets and dry bottoms. But
such information only gets us so far analytically. We can integrate such research by
asking questions about when and how the different effects of the gender structure
remain resistant to change and when some progressive feminist change has
occurred. Do young women in the twenty-first century, raised by feminists, suc-
cessfully negotiate fair families? Or does the moral accountability to do gender as
mothers and wives combined with devalued status in the workplace still defeat even
women socialized for equality? Does workplace flexibility for men allow feminist
women more success in their negotiations at the family level? The conceptualiza-
tion of gender as a structure, and attention to the mechanisms at work in each
dimension of the gender structure, helps to frame the kind of research that might
answer such queries.
Gender structure theory allows us to try to disentangle the “how” questions
without presuming that there is one right answer, for all places, times, and contexts.
442 GENDER & SOCIETY / August 2004

It is easy to illustrate that a combination of gender wage gap and the organization of
careers requiring inflexible hours and full-time commitment pushes married moth-
ers outside the labor force and creates stressful lives for mothers who remain within
it, married or not. But we must still ask why this is true for women but not men. Per-
haps, under some conditions, women socialized for emphasized femininity do
indeed hold themselves accountable for being personally responsible for more than
good enough mothering and sparkling households. Research should identify under
what conditions and to what extent gendered selves help to account for objective
inequalities (e.g., women working more hours a day than their partners) and when
other factors are more significant. My own hypothesis is that feminist women are
often defeated in their attempt at egalitarian heterosexual relationships by cultural
gendered interactional expectations. Within the past year, memoirs have been writ-
ten by young feminists, academics, and daughters of famous women’s movement
leaders (Fox 2003; Hanauer 2002) bemoaning the impossible expectations facing
career women who choose motherhood as well. Similarly, a recent feminist
cyberspace conversation on the Listserve of Sociologists for Women in Society
described the struggle to combine motherhood and career in the academy in nearly
as despairing a tone as did Arlie Hochschild (1975) in her classic article first pub-
lished three decades ago. I have yet to see recent memoirs, or hear of painful
listserver conversations, among twenty-first-century fathers. Little cultural change
has occurred around fathering. Most men are still not morally responsible for the
quality of family life, and women have yet to discover how to avoid being held
accountable.
Gender structures are even more complicated than my discussion suggests thus
far because how gender identities are constructed on the individual and cultural
dimensions vary tremendously over time and space. Even within contemporary
American society, gender structures vary by community, social class, ethnicity, and
race.

GENDER STRUCTURE AND INTERSECTIONALITY

Perhaps the most important development in feminist thought in the last part of
the twentieth century was the increasing concern with intersectionality (Andersen
and Collins 1994; Baca Zinn and Thornton Dill 1994; Collins 1990). Women of
color had been writing about intersectionality from nearly the start of the second
wave of feminist scholarship. It was, however, not until several decades into the
women’s movement when they were heard and moved from margin closer to center
(Myers et al. 1998). There is now considerable consensus growing that one must
always take into consideration multiple axes of oppression; to do otherwise pre-
sumes the whiteness of women, the maleness of people of color, and the
heterosexuality of everyone.
I concur with this consensus that gender must be understood within the context
of the intersecting domains of inequality. The balkanization of research and theory
 Risman / GENDER AS A SOCIAL STRUCTURE 443

into specializations of race or ethnicity or gender or stratification has undermined a

sophisticated analysis of inequality (but see Reskin 2002; Schwalbe et al. 2000;
Tilly 1999). I do not agree, however, with an operational strategy for scholarship
that suggests the appropriate analytic solution is to only work within an inter-
sectionality framework. While various axes of domination are always intersecting,
the systems of inequality are not necessarily produced or re-created with identical
social processes. The historical and current mechanisms that support gender
inequality may or may not be those that are most significant for other kinds of
oppression; whether this is the case is an empirical question. Gender research and
theory can never again ignore how women’s subordination differs within racial and
ethnic communities or is constructed within class dynamics. Yet we should not
therefore only study gender, race, and class simultaneously. There is a difference
between an analysis of psychological, historical, or sociological mechanisms that
construct inequality and the subjective experience of the outcomes of such mecha-
nisms. There may be similarity of outcomes (e.g., experiences of oppression) along
axes of oppression that arise from different causal mechanisms, but that is an
empirical question, not a logical necessity. To focus all investigations into the com-
plexity or subjective experience of interlocking oppressions would have us lose
access to how the mechanisms for different kinds of inequality are produced. Femi-
nist scholarship needs a both/and strategy (Collins 1998). We cannot study gender
in isolation from other inequalities, nor can we only study inequalities’ intersection
and ignore the historical and contextual specificity that distinguishes the
mechanisms that produce inequality by different categorical divisions, whether
gender, race, ethnicity, nationality, sexuality, or class.
Calhoun (2000) exemplifies this both/and strategy in her argument that
heterosexism cannot simply be understood as gender oppression and merged into
feminist theory. She argued that we must study heterosexism as a separate system
of oppression. While it is clearly the case that gender subordination and
heterosexism support one another, and a gendered analysis of homophobia is criti-
cal, the two oppressions should not be conflated. It is often presumed that Rich’s
(1980) argument about “compulsory heterosexuality and lesbian existence” sug-
gests that heterosexism is primarily a product of men’s dominance, an attempt to
ensure sexual access to women by men, and that this is a primary explanation for
lesbian oppression. While this is surely an important component of heterosexism,
Calhoun argued that it is a mistake to presume that it is the whole of it. She sug-
gested instead that challenging men’s dominance is a necessary condition of ending
the subordination of lesbians and gay men but not a sufficient condition to end such
oppression. It is important for analytic clarity, and therefore to the scholarly contri-
bution to social change, to identify causal mechanisms for heterosexism and gender
oppression distinctly.
Other examples also illustrate the analytic usefulness of paying attention to the
distinct properties of different axes of oppression. Gendered images support racial
domination, but racial domination can hardly be attributed to gender inequality. For
example, Black men’s inferiority gets promoted through constructions of hyper-
444 GENDER & SOCIETY / August 2004

sexuality (Collins 2004), and Black women’s inferiority gets promoted through
sexualized images such as Jezebel or welfare queen (Collins 2000). Similarly,
Asian American men’s autonomy and even citizenship rights were abrogated by
constructions of effeminacy (Espiritu 1997). Yet it is implausible to argue that
racial domination is nothing but a product of gender oppression. While we must
pay attention to how axes of oppression affect one another and how the experience
of their oppressions are simultaneous, we must continue to study and work to trans-
form each one independently as well as in conjunction with one another.
Each categorical inequality (Tilly 1999) that is deeply embedded in society can
be conceptualized as a social structure. Bonilla-Silva (1997) has made this argu-
ment persuasively for conceptualizing race as a social structure. He argued that race
is a social structure that influences identities and attitudes but is also incorporated
into how opportunities and constraints work throughout every societal institution.
According to Bonilla-Silva, to conceptualize race as a social structure forces us to
move beyond seeing racial inequality as constructed simply by racist attitudes and
to understand the ways in which our society embeds white privilege at every level
of analysis. I hardly need to argue that class inequality should be conceptualized as
a structure as the economic structure of society has long been a primary concern of
social scientists. Similarly, political structures have long been studied both at the
national and comparative level because here too, politics are routinely considered a
basic component of human society. My argument is that race, gender, and sexuality
are as equally fundamental to human societies as the economy and the polity. Those
inequalities that are fundamentally embedded throughout social life, at the level of
individual identities, cultural expectations embedded into interaction, and institu-
tional opportunities and constraints are best conceptualized as structures: The gen-
der structure, the race structure, the class structure, and the sexuality structure. This
does not imply that the social forces that produced, nor the causal mechanisms at
work in the daily reproduction of inequality within each structure, are of similar
strength or type at any given historical moment. For example, gender and race
structures extend considerably further into everyday life in the contemporary
American context, at home and at work, than does the political structure.3 I propose
this structural language as a tool to help disentangle the means by which inequali-
ties are constructed, recreated, and—it is hoped—transformed or deconstructed.
The model for how gender structure works, with consequences for individuals,
interactions/cultural expectations, and institutions, can be generalized to the study
of other equally embedded inequalities such as race and sexuality. Each structure of
inequality exists on its own yet coexists with every other structure of inequality.
The subjective experience of actual human beings is always of intersecting inequal-
ities, but the historical construction and contemporary reproduction of inequality
on each axis may be distinct. Oppressions can be loosely or tightly coupled, can
have both common and distinct generative mechanisms.
 Risman / GENDER AS A SOCIAL STRUCTURE 445

THEORY WRESTLING WITH ACTIVISM

Within any structure of inequality, perhaps the most important question a critical
scholar must ask is, What mechanisms are currently constructing inequality, and
how can these be transformed to create a more just world? If as critical scholars, we
forget to keep our eye on social transformation, we may slip without intention into
the implicitly value-free role of social scientists who study gender merely to satisfy
intellectual curiosity (Risman 2003). The central questions for feminists must
include a focus on social transformation, reducing inequality, and improving the
status of women. A concern with social change brings us to the thorny and as yet too
little explored issue of agency. When do subordinate groups collectively organize
to challenge their oppression? When do superordinate groups mobilize to resist?
How do we know agency when we see it, and how can we support feminist versions
of it?
Feminist scholarship must seek to understand how and why gender gets done,
consciously or not, to help those who hope to stop doing it. I end by focusing our
attention on what I see as the next frontier for feminist change agents: A focus on
the processes that might spur change at the interactional or cultural dimension of
the gender structure. We have begun to socialize our children differently, and while
identities are hardly postgender, the sexism inherent in gender socialization is now
widely recognized. Similarly, the organizational rules and institutional laws have
by now often been rewritten to be gender neutral, at least in some nations. While
gender-neutral laws in a gender-stratified society may have short-term negative
consequences (e.g., displaced homemakers who never imagined having to support
themselves after marriage), we can hardly retreat from equity in the law or organi-
zations. It is the interactional and cultural dimension of gender that have yet to be
tackled with a social change agenda.
Cognitive bias is one of the mechanisms by which inequality is re-created in
everyday life. There are, however, documented mechanisms for decreasing the
salience of such bias (Bielby 2000; Reskin 2000; Ridgeway and Correll 2000).
When we consciously manipulate the status expectations attached to those in sub-
ordinate groups, by highlighting their legitimate expertise beyond the others in the
immediate social setting, we can begin to challenge the nonconscious hierarchy
that often goes unnoticed. Similarly, although many subordinates adapt to their sit-
uation by trading power for patronage, when they refuse to do so, interaction no
longer flows smoothly, and change may result. Surely, when wives refuse to trade
power for patronage, they can rock the boat as well as the cradle.
These are only a few examples of interactive processes that can help to explain
the reproduction of inequality and to envision strategies for disrupting inequality.
We need to understand when and how inequality is constructed and reproduced to
deconstruct it. I have argued before (Risman 1998) that because the gender struc-
446 GENDER & SOCIETY / August 2004

ture so defines the category woman as subordinate, the deconstruction of the cate-
gory itself is the best, indeed the only sure way, to end gender subordination. There
is no reason, except the transitional vertigo that will accompany the process to dis-
mantle it, that a utopian vision of a just world involves any gender structure at all.
Why should we need to elaborate on the biological distinction between the sexes?
We must accommodate reproductive differences for the process of biological
replacement, but there is no a priori reason we should accept any other role differ-
entiation simply based on biological sex category. Before accepting any gender
elaboration around biological sex category, we ought to search suspiciously for the
possibly subtle ways such differentiation supports men’s privilege. Once two
salient groups exist, the process of in-group and out-group distinctions and in-
group opportunity hoarding become possible. While it may be that for some com-
petitive sports, single-sex teams are necessary, beyond that, it seems unlikely that
any differentiation or cultural elaboration around sex category has a purpose
beyond differentiation in support of stratification.
Feminist scholarship always wrestles with the questions of how one can use the
knowledge we create in the interest of social transformation. As feminist scholars,
we must talk beyond our own borders. This kind of theoretical work becomes
meaningful if we can eventually take it public. Feminist sociology must be public
sociology (Burawoy forthcoming). We must eventually take what we have learned
from our theories and research beyond professional journals to our students and to
those activists who seek to disrupt and so transform gender relations. We must con-
sider how the knowledge we create can help those who desire a more egalitarian
social world to refuse to do gender at all, or to do it with rebellious reflexiveness to
help transform the world around them. For those without a sociological perspec-
tive, social change through socialization and through legislation are the easiest to
envision. We need to shine a spotlight on the dimension of cultural interactional
expectations as it is here that work needs to begin.
We must remember, however, that much doing gender at the individual and
interactional levels gives pleasure as well as reproduces inequality, and until we
find other socially acceptable means to replace that opportunity for pleasure, we
can hardly advocate for its cessation. The question of how gender elaboration has
been woven culturally into the fabric of sexual desire deserves more attention.
Many of our allies believe that “viva la difference” is required for sexual passion,
and few would find a postgender society much of a feminist utopia if it came at the
cost of sexual play. No one wants to be part of a revolution where she or he cannot
dirty dance.
In conclusion, I have made the argument that we need to conceptualize gender as
a social structure, and by doing so, we can analyze the ways in which gender is
embedded at the individual, interactional, and institutional dimensions of our soci-
ety. This situates gender at the same level of significance as the economy and the
polity. In addition, this framework helps us to disentangle the relative strength of a
variety of causal mechanisms for explaining any given outcome without dismissing
the possible relevance of other processes that are situated at different dimensions of
 Risman / GENDER AS A SOCIAL STRUCTURE 447

analysis. Once we have a conceptual tool to organize the encyclopedic research on

gender, we can systematically build on our knowledge and progress to understand-
ing the strength and direction of causal processes within a complicated multidi-
mensional recursive theory. I have also argued that our concern with intersec-
tionality must continue to be paramount but that different structures of inequality
have different infrastructure and perhaps different influential causal mechanisms at
any given historical moment. Therefore, we need to follow a both/and strategy, to
understand gender structure, race structure, and other structures of inequality as
they currently operate, while also systematically paying attention to how these axes
of domination intersect. Finally, I have suggested that we pay more attention to
doing research and writing theory with explicit attention to how our work can come
to be “fighting words” (Collins 1998) to help transform as well as inform society. If
we can identify the mechanisms that create gender, perhaps we can offer
alternatives to them and so use our scholarly work to contribute to envisioning a
feminist utopia.

NOTES

1. See Scott (1997) for a critique of feminists who adopt a strategy where theories have to be simpli-
fied, compared, and defeated. She too suggested a model where feminists build on the complexity of each
others’ ideas.
2. I thank my colleague Donald Tomaskovic-Devey for suggesting the visual representation of these
ideas as well as his usual advice on my ideas as they develop.
3. One can certainly imagine a case where political structures extend far into everyday life, a nation in
the midst of civil war or in the grips of a fascist state. One can also envision a case when race retreats to
the personal dimension, as when the Irish became white in twentieth-century America.

REFERENCES

Andersen, Margaret, and Patricia Hill Collins. 1994. Race, class, and gender: An anthology. Belmont,
CA: Wadsworth.
Baca Zinn, Maxine, and Bonnie Thornton Dill. 1994. Women of color in U.S. society. Philadelphia: Tem-
ple University Press.
Bem, Sandra. 1993. The lenses of gender. New Haven, CT: Yale University Press.
Bianchi, Suzanne M., Melissa A. Milkie, Liana C. Sayer, and John P. Robinson. 2000. Is anyone doing
the housework? Trends in the gender division of household labor. Social Forces 79 (1): 191-228.
Bielby, William T. 2000. Minimizing workplace gender and racial bias. Contemporary Sociology 29 (1):
120-29.
Blau, Peter. 1977. Inequality and heterogeneity. New York: Free Press.
Bonilla-Silva, Eduardo. 1997. Rethinking racism: Toward a structural interpretation. American Socio-
logical Review 62 (3): 465-80.
Brines, Julie. 1994. Economic dependency, gender, and the division of labor at home. American Journal
of Sociology 100 (3): 652-88.
Burawoy, Michael. Forthcoming. Public sociologies contradictions, dilemmas and possibilities. Social
Forces.
448 GENDER & SOCIETY / August 2004

Burt, Ronald S. 1982. Toward a structural theory of action. New York: Academic Press.
Calhoun, Cheshire. 2000. Feminism, the family, and the politics of the closet: Lesbian and gay displace-
ment. New York: Oxford University Press.
Collins, Patricia Hill. 1990. Black feminist thought: Knowledge, consciousness, and the politics of
empowerment. New York: Routledge.
. 1998. Fighting words: Black women and the search for justice. Minneapolis: University of Min-
nesota Press.
. 2004. Black sexual politics: African Americans, gender, and the new racism. New York:
Routledge.
Connell, R. W. 1987. Gender and power: Society, the person, and sexual politics. Stanford, CA: Stanford
University Press.
. 2002. Gender: Short introductions. Malden, MA: Blackwell.
Coverman, Shelley. 1985. Explaining husbands’ participation in domestic labor. Sociological Quarterly
26 (1): 81-97.
England, Paula, and Irene Browne. 1992. Internalization and constraint in women’s subordination. Cur-
rent Perspectives in Social Theory 12:97-123.
Espiritu, Yen Le. 1997. Asian American women and men: Labor, laws, and love. Thousand Oaks, CA:
Sage.
Fenstermaker Berk, Sarah. 1985. The gender factory: The apportionment of work in American house-
holds. New York: Plenum.
Ferree, Myra Marx. 1990. Beyond separate spheres: Feminism and family research. Journal of Marriage
and the Family 53 (4): 866-84.
Ferree, Myra Marx, Judith Lorber, and Beth Hess. 1999. Revisioning gender. Thousand Oaks, CA: Sage.
Fox, Faulkner. 2003. Dispatches from a not-so-perfect Life: On how I learned to love the house, the man,
the child. New York: Harmony Books.
Fuchs Epstein, Cynthia. 1988. Deceptive distinctions: Sex, gender, and the social order. New Haven, CT:
Yale University Press.
Giddens, Anthony. 1984. The constitution of society: Outline of the theory of structuration. Berkeley:
University of California Press.
Greenstein, Theodore N. 1996. Husbands’ participation in domestic labor: Interactive effects of wives’
and husbands’ gender ideologies. Journal of Marriage and the Family 58:585-95.
. 2000. Economic dependence, gender, and the division of labor in the home: A replication and
extension. Journal of Marriage and the Family 62 (2): 322-35.
Hanauer, Cathi. 2002. The bitch in the house: 26 women tell the truth about sex, solitude, work, mother-
hood, and marriage. New York: William Morrow.
Hochschild, Arlie. 1975. Inside the clockwork of male careers. In Women and the power to change,
edited by Florence Howe. New York: McGraw Hill. Repr. in The commercialization of intimate life.
Berkeley: University of California Press, 2003.
Kanter, Rosabeth. 1977. Men and women of the corporation. New York: Basic Books.
Lorber, Judith. 1994. Paradoxes of gender. New Haven, CT: Yale University Press.
Martin, Patricia. 2003. “Said and done” versus “saying and doing”: Gendering practices, practicing gen-
der at work. Gender & Society 17:342-66.
. Forthcoming. Gender as a social institution. Social Forces.
Myers, Kristen A., Cynthia D. Anderson, and Barbara J. Risman, eds. 1998. Feminist foundations:
Toward transforming society. Thousand Oaks, CA: Sage.
Pleck, Joseph H. 1985. Working wives/working husbands. Beverly Hills, CA: Sage.
Presser, Harriet B. 1994. Employment schedules among dual-earner spouses and the division of house-
hold labor by gender. American Sociological Review 59 (3): 348-64.
Reskin, Barbara. 1998. The realities of affirmative action in employment. Washington, DC: ASA.
. 2000. The proximate causes of employment discrimination. Contemporary Sociology 29 (2):
319-28.
 Risman / GENDER AS A SOCIAL STRUCTURE 449

. 2002. How did the poison get in Mr. Bartlett’s stomach? Motives and mechanisms in explaining
inequality. Presidential address given at the 97th annual meetings of the American Sociological
Association, Chicago, August.
Rich, Adrienne. 1980. Compulsory heterosexuality and lesbian existence. Signs: Journal of Women in
Culture and Society 5 (4): 631-60.
Ridgeway, Cecilia L. 1991. The social construction of status value: Gender and other nominal character-
istics. Social Forces 70 (2): 367-86.
. 1997. Interaction and the conservation of gender inequality: Considering employment. Ameri-
can Sociological Review 62 (2): 218-35.
. 2001. Gender, status, and leadership. Journal of Social Issues 57 (4): 637-55.
Ridgeway, Cecilia L., and Shelley J. Correll. 2000. Limiting inequality through interaction: The end(s)
of gender. Contemporary Sociology 29:110-20.
Ridgeway, Cecilia L., Kathy J. Kuipers, Elizabeth Heger Boyle, and Dawn T. Robinson. 1998. How do
status beliefs develop? The role of resources and interactional experience. American Sociological
Review 63:331-50.
Ridgeway, Cecilia L., and Lynn Smith-Lovin. 1999. The gender system and interaction. Annual Review
of Sociology 25:191-216.
Risman, Barbara J. 1983. Necessity and the invention of mothering. Ph.D. diss, University of Washington.
. 1987. Intimate relationships from a microstructural perspective: Mothering men. Gender &
Society 1:6-32.
. 1998. Gender vertigo: American families in transition. New Haven, CT: Yale University Press.
. 2003. Valuing all flavors of feminist sociology. Gender & Society 17:659-63.
Risman, Barbara J., and Pepper Schwartz. 1989. Gender in intimate relationships. Belmont, CA:
Wadsworth.
Robinson, John P., and Melissa A. Milkie. 1998. Back to the basics: Trends in and role determinants of
women’s attitudes toward housework. Journal of Marriage and the Family 60 (1): 205-18.
Rytina, Steve, Peter Blau, Jenny Blum, and Joseph Schwartz. 1988. Inequality and intermarriage: Para-
dox of motive and constraint. Social Forces 66:645-75.
Schwalbe, Michael, Sandra Godwin, Daphne Holden, Douglas Schrock, Shealy Thompson, and
Michele Wolkomir. 2000. Generic processes in the reproduction of inequality: An interactionist
analysis. Social Forces 79 (2): 419-52.
Scott, Joan Wallach. 1997. Comment on Hawkesworth’s “Confounding Gender.” Signs: Journal of
Women in Culture and Society 22 (3): 697-702.
Shelton, Beth Anne. 1992. Women, men and time: Gender differences in paid work, housework and lei-
sure. Westport, CT: Greenwood.
Sherwood, Jessica. 2004. Talk about country clubs: Ideology and the reproduction of privilege. Ph.D.
diss., North Carolina State University.
Smelser, Neil J. 1988. Social structure. In Handbook of sociology, edited by Neil J. Smelser. Beverly
Hills, CA: Sage.
Staples, Robert. 1990. Social inequality and Black sexual pathology: The essential relationship. Black
Scholar 21 (3): 29-37.
Stombler, Mindy, and Patricia Yancey Martin. 1994. Bring women in, keeping women down: Fraternity
“little sister” organizations. Journal of Contemporary Ethnography 23:150-84.
Tilly, Charles. 1999. Durable inequality. Berkeley: University of California Press.
Twiggs, Joan E., Julia McQuillan, and Myra Marx Ferree. 1999. Meaning and measurement: Recon-
ceptualizing measures of the division of household labor. Journal of Marriage and the Family 61 (3):
712-24.
Udry, J. Richard. 2000. Biological limits of gender construction. American Sociological Review 65:
443-57.
West, Candace, and Don Zimmerman. 1987. Doing gender. Gender & Society 1:125-51.
450 GENDER & SOCIETY / August 2004

Williams, Christine. 1992. The glass escalator: Hidden advantages for men in the “female” professions.
Social Problems 39:253-67.
Williams, Joan. 2000. Unbending gender: Why family and work conflict and what to do about it. New
York: Oxford University Press.
Yoder, Janice. 1991. Rethinking tokenism. Social Problems 5:178-92.
Zimmer, Lynn. 1988. Tokenism and women in the workplace: The limits of gender-neutral theory. Social
Problems 35:64-77.

Barbara J. Risman is Alumni Distinguished Research Professor of Sociology at North Carolina

State University. She studies gender in intimate and family relationships. She is the author of
Gender Vertigo: American Families in Transition (1998, Yale University Press). She also edits
The Gender Lens book series with Judith Howard and Joey Sprague, designed to transform the
discipline of sociology by mainstreaming a gender perspective throughout the curriculum. She is
past president of Sociologists for Women in Society and is currently co-chair of the Council on
Contemporary Families. Her current research focuses on gender and sexual ideology among teen-
agers.