Rethinking Stratification from a Feminist Perspective: Gender, Race, and Class in Mainstream Textbooks

by Myra Marx Ferree, Elaine J. Hall
Citation
Title:
Rethinking Stratification from a Feminist Perspective: Gender, Race, and Class in Mainstream Textbooks
Author:
Myra Marx Ferree, Elaine J. Hall
Year: 
1996
Publication: 
American Sociological Review
Volume: 
61
Issue: 
6
Start Page: 
929
End Page: 
950
Publisher: 
Language: 
English
URL: 
Select license: 
Select License
DOI: 
PMID: 
ISSN: 
Abstract:

RETHINKING STRATIFICATION
FROM A FEMINIST PERSPECTIVE:
GENDER, RACE, AND CLASS IN MAINSTREAM TEXTBOOKS*
Myra Marx Ferree Elaine J. Hall

University of Connecticut Kent State University

Economic stratification and social class occupy a central position in socio- logical discourse as the core organizing features of modern societies. Yet such economically centered models of stratification often disregard factors like physical violence and the intra-household distribution of resources that shape power and autonomy for all groups. Using a sample of textbooks from 1983 through 1988, we examine "mainstream" sociolog.~, that is, the sociol- ogy that teachers, students, and textbook publishers have treated as nonproblematic. We show how stratification anal.ysis is applied to class, race, and gender in profoundly unequal ways. Rather than integrating macro, meso, and micro levels of social structure as interactive and mutually deter- minative in their discussions of race, class, and gender, introductory sociol- ogy textbooks segregate stratification processes. They discuss class at the societal (or macro) level of anal.ysis, gender at the individual (or micro) level, and race at a group (or meso) level. We anal.yze the quantitative and qualitative elements of the coverage of class, race, and gender in indexes, texts, pictures, and captions, and suggest that attention to feminist theories of gender would produce a more integrated, multilevel, and interactive view of stratification.

In their landmark evaluation of the disci- ". . . neither have been integrated into, nor pline of sociology, Stacey and Thorne have they transformed the old, 'general' (1985) conclude that, unlike history and an- theories of society" (p. 65). Although femi- thropology, sociology has not undergone the nist work challenges theories based on a nor- conceptual transformation necessary for a mative male subject (Pateman 1988; Wallace "feminist revolution." Similarly, Acker 1989), mainstream research often does little (1989) argues that the extensive knowledge more than "add women and stir" into exist- and criti~ues of recent feminist scholarshiu ing frameworks (McIntosh 1984; Grant, ward, and Rong 1987).

* Direct Marx we address three aspects of this feminist Ferree, Department of Sociology, U-68, Univer-

to First, we define whatsity of Connecticut, Storrs, CT 06269-2068 we See as the mainstream view of

(ferree@uconnvm.uconn.edu). The order of au-

stratification, review feminist critiques of its

thorship is alphabetical and reflectsdifferent but equal contributions by both authors. We are in- limitations, and outline new perspectives on debted to colleagues for their constructive criti- gender stratification. Second, we examine cisms, including Mark Abrahamson, Maxine the mainstream analysis of class, race, and

Baca Zinn, Christine Bose, Nancy Breen, Joe gender found in introductory sociology text-

Feagin, Davita Silfen Glasberg, Beth Hess, Judith books and indicate the consequences of its Lorber, Patricia Yancey Martin, Gail McGuire, unexamined assumptions about gender, race, Barbara Risman, Silke Roth, Gaye Tuchman, and class stratification. Finally, we suggest Carole Turbin, Wayne Villemez, Ruth Wallace how feminist insights can help realize the

and the ASR Editor and anonymous reviewers,

discipline's promise of an analysis of the so-

~ acknowledged by the authors i ~ ~ ~

[ ~ ~are ~ ~ Linda M. Grant, Harold Kerbo, and Barbara cial order that combines both Structure and Laslett. -ED.] agency at the micro, meso, and macro levels.

929

American Sociological Review, 1996, Vol. 61 (December:929-950)
WADING INTO THE MAINSTREAM

Sociology is a diverse discipline, and its theo- retical positions are contested. To avoid be- coming mired in the details of sociology's myriad perspectives and theories, we define four central concepts: a structural perspec- tive, the paradigm of sociology, the sociologi- cal mainstream, and economic stratification.
The Structural Perspective

One premise of sociology is the claim to look at society structurally, that is, in terms of per- sistent patterns of social relations. Embrac- ing a structural perspective does not mean that the discipline can or should ignore hu- man agency. Even if the conditions are not of their own choosing, people do make his- tory. Sociologists use the term agency to re- fer to such self-motivated actions, and they recognize that actors' individual histories shape their behavior. However, sociological explanations resist reduction to arguments of agency alone; doing sociology demands at- tention to forces that arise from and impact on collectivities as such. Such structural forces exist at all levels of society, from the micro (individuals and their interactions), through the meso (groups and their interac- tions), to the macro (whole societies or insti- tutional sectors and their interactions). These levels are not equivalent in scope or scale, and each has its distinctive forms of social organization; yet all three levels can be un- derstood, at least in part, in terms of their structures. Most sociologists, we believe, en- dorse this principle, and in this sense it is a core feature of the sociological paradigm.
The Sociological Paradigm

By sociological "paradigm" we do not mean any of the particular theories that sociologists advance. Nor do we refer to the schools into which theories are usually organized. Rather we mean the shared assumptions of sociologi- cal discourse that define certain problems as significant, identify relevant evidence, and produce agreed-upon social facts (solutions) as well as troubling anomalies (Kuhn 1970). In comparison to the natural sciences, sociology's paradigm is relatively weak and unconstraining, but some elements are
AMERICAN SOCIOLOGICAL REVIEW

broadly shared within the discipline (Westhus 1976; Barnes 1982). Such shared assump- tions are essentially pre-theoretical in that they are rarely part of theoretical debates.'
Mainstream Sociology

We use the term "mainstream sociology" to refer to this weak paradigm as evidenced in sociology departments in the United States today. The term excludes forms of sociology practiced outside the United States and in- cludes elements of American culture that shape this particular form. On the one hand, mainstream sociology represents an ideal type toward which publishers push American textbook authors: A noncontroversial amal- gam of generally acceptable assertions orga- nized to reflect shared assumptions of what goes with what and what belongs where. This common-denominator version of sociology is driven in part by publishers9 empirically- based appraisals of what U.S. sociologists find appealing and useful. Research sociolo- gists who do not teach may be outside this mainstream, although they are surely famil- iar with it. On the other hand, the "main- stream" is only a central tendency in a di- verse community of teaching sociologists, and while the mean defines the market, it probably represents relatively few individu- als' conceptions of sociology precisely. Be- cause of market pressures, textbooks are more mainstream and less idiosyncratic or innovative than even their own authors would be in another setting. This is an advantage when using textbooks to explore the nature of the mainstream-they can show what sort of analysis counts as noncontroversial for most teaching sociologists. Over time, the mainstream responds to changes in the aver-

] The invisibility of "pre-theoretical" assump- tions distinguishes them from the specific pre- mises of particular schools that are explicitly theorized and debated (e.g., structural-functional- ist, Marxist, symbolic interactionist, etc.). We consider such pre-theoretical assumptions no less controversial and debatable (and hence also de- fensible in principle by those who disagree with our analysis), but they are rarely theorized explic- itly or defended within the discipline because they constitute a core of premises shared so widely (outside of feminist circles) that they ap- pear self-evident.

age preferences of the population of sociolo- gists. One evidence of this is the addition of gender as a significant topic in introductory textbooks in the past 20 years.

Consequently, any study of the "main- stream," especially one that uses introductory textbooks to capture this weak paradigm, is chasing a moving target. Textbooks are time- lagged measures of the state of the disci- pline. Research on gender over the past de- cade is only now being incorporated in spe- cialized textbooks, and the lag before new theories and findings reach mainstream soci- ologists (most of whom are long out of graduate school) is even more substantial. Because textbooks are revised frequently (of- ten on two-year cycles) while research moves slowly, our analysis does not fully reflect the current state of even this textbook knowl- edge. The general trend in the mainstream is toward increased inclusion of race and gen- der, and the perspectives of research sociolo- gists in these specific subareas continue to influence what instructors, authors, and pub- lishers perceive as important. Thus we use these textbooks only to illustrate tendencies in the mainstream, not to blame the books or their authors for past limitations or to sug- gest that they do not or will not change. Given these important caveats, we examine introductory textbooks from 1983 through

1988 to make visible mainstream under- standings of stratification shared by their producers and adopters.
Economic Stratification

By "economic stratification" we mean mod- els of inequality that focus on differences in economic resources (income or wealth) and on factors conceptualized in terms of their relation to such economic resources (occupa- tion, education, prestige). Economic stratifi- cation and the associated topics of mobility and poverty have been central to sociology since its inception (Bernard 1929; Odum 1951; Westhus 1976; Connell 1990). Eco- nomic stratification is so prominent in main- stream sociology that it is often equated with social stratification in general. To disentangle the two concepts, we define social stratifica- tion as processes regulating "(1) access to re- sources; (2) autonomy, that is, the freedom to make life choices and the freedom of move- ment; and (3) power, that is, participation in the making of decisions concerning the so- cial group" (Agassi 1989: 168). As Acker (1973) noted two decades ago, mainstream studies of stratification are often limited to the first of these dimensions (taking occupa- tion and income as the paradigmatic mea- sures). Differential access to resources, which we call economic stratification, is often treated in textbooks under the rubric of "class." We also use the term "class" for eco- nomic stratification, but distinguish our us- age from specific theoretical positions about economic stratification, whether they be Marxist, functionalist, or other.

Because sociologists viewed economic stratification as self-evidently primary (Haug 1977:52), introductory textbooks before 1975 typically provided several chapters on economic stratification as the only aspect of inequality. Over the past two decades, text- book authors added chapters on race and gender stratification, but many still label only economic stratification as "social" strat- ification pertaining to society as a whole. Race and gender are often presented as fac- tors contributing to an individual's position in purely economic stratification systems, while the noneconomic aspects of stratifica- tion (i.e., autonomy, power) remain invisible. As feminist critics have noted (Ciancanelli and Berch 1987; Glazer 1993), explaining who has the autonomy to enter the labor mar- ket in the first place or the power to set the terms under which certain skills are recog- nized or rewarded is not an integral part of the mainstream model of stratification. Eco- nomic stratification Drocesses are often con- ceptualized as abstract structures, while race and gender appear as either "illusions" (reducible to class) or "essences" (of individu- als) (Wharton 1991; Omi and Winant 1994).

Although the economic relations studied are nominally abstract, they are constructed analytically by examining only those posi- tions and processes that affect status differ- ences among male heads of households. That is, the development of a market system pro- duced a division between paid and unpaid forms of labor that sociology failed to theo- rize as a stratification process. Because it was assumed that males were and should be "heads" of households. the market's estab- lishment of new dimensions of power and privilege on preexisting lines of gender dif- ferentiation was treated as "natural" rather than as needing explanation. This invisibility persists. For example, research on social mo- bility and status-attainment concerns the changing placement of individuals within the "empty slots" of a paid labor system (Smel- ser 1990), without asking how a gender divi- sion of labor shapes the existence and nature of these available slots (Papanek 1973; Wharton 1991). This is a ure-theoretical assumption, not simply an oversight. As Acker (1990) argues, the worker posited by theory is abstract and disembodied-an ideal type that most closely fits the life experience of male heterosexuals, since from their social location sexuality, emotionality, and repro- ductive labor can be identified with the pres- ence of women and divided from the work- place into a "separate sphere" of the family. This division then appears "natural" rather than as part of a social stratification process. Thus only the paid labor market becomes the focus of "stratification" research.

This approach marginalizes race and gen- der as attributes of non-White, non-male in- dividuals. Such attributes are invoked to ex- plain the disadvantaged status of White women and people of color in a system orga- nized around positioning "abstract" individu- als in a market-based hierarchy (Acker 1990). Feminist researchers have instead suggested that race and gender are institu- tionalized systems that organize labor mar- kets and construct hierarchies among and within institutions (Lorber 1994). These cri- tiques direct attention to the missing dimen- sions of stratification.
FEMINIST CRITIQUES OF THE ECONOMIC STRATIFICATION MODEL

Because the pre-theoretical assumption of the centrality of economic position marginalizes sociological concern with non- economic hierarchies and subordination pro- cesses, it is difficult, if not impossible, to de- termine any unitary status of women cross- nationally. The multiple dimensions and pro- cesses that generate women's social standing are rarely addressed (Sudarkasa 1973; Blumberg 1984; Tiano 1987). Women may have autonomy but little power, economic resources but limited autonomy, or power as
AMERICAN SOCIOLOGICAL REVIEW

members of kin groups but no access to eco- nomic resources independent of kin ties. In the absence of awareness of such multidi- mensionality, inequalities seem reducible to natural "differences." For example, studying the relative economic standing of household heads without understanding autonomy and power as stratification processes that estab- lish men as the "normal" heads of families leads to attributing the "disadvantages" of women household heads only to their indi- vidual characteristics (e.g., work experience, "being female").

To be sure, women are now more visible in analyses that assume an "abstract" worker in an economic stratification system than they were when these models were formu- lated-in part because more White women and people of color around the globe are now found in the formal positions that such theo- ries recognize and analyze. However, this in- clusion has not been accompanied by a fun- damental rethinking of classic stratification models, despite the anomalies it has gener- ated (e.g., in prestige measures). Although feminist critics (Hess and Ferree 1987; Smith 1987; Wallace 1989; Lorber and Farrell 1991; Wharton 1991; England 1993) raise many issues, we emphasize just four recur- rent themes.

First, the role of violence and physical domination is omitted from the mainstream account of stratification, yet domestic and public violence contribute significantly to the subordination of women (Gordon and Riger 1989; Bart and Moran 1993). When issues of power and autonomy are raised, it seems clear that substantial limits are imposed on women by male ~iolence.~

If the experiences of women were theorized, threats of violence and the use of brute force would not have disappeared from accounts of the stratifica- tion process in industrial societies. Legiti- mate control over the means of violence is also a gender stratification issue. For example, gender is a factor in military experi- ence, which in turn becomes a qualification

Racial harassment, discrimination in housing and public accommodations, and the differential legitimacy of violence by Whites against Blacks and violence by Blacks against Whites also re- strict the autonomy and power of people of color in the United States, regardless of their incomes (Feagin and Sikes 1994).

for political office and access to other posi- tions of power (Stiehm 1982; Enloe 1987).

Second, the apparently "objective" and abstract nature of money as a source and measure of economic power has been ques- tioned by theoretical accounts of the dis- counting of women's economic resources (Blumberg 1984) and by empirical analysis of the "special" meaning of women's earnings (Zelizer 1989).~ The discounting of women's earnings often takes the form of defining them as "not essential" and thus as not creating the same social entitlements (at home or in social policy) as men's earnings. When money does not mean the same thing for women as for men, or does not translate into power and autonomy at the same rate, location in the economic system alone will not explain social position. Gender norms also define the extent to which women (and men) control their incomes or transmit ad- vantages to their children. The structural po- sitions and processes generated by class models are not gender-neutral and only ap- pear so by defining male outcomes as the "general" case (e.g., taking intergeneration- a1 mobility to mean father-son links [Smith

1987; Acker 19891).

Third, class models assuming that male heads of households are "natural" fail to inte- grate the effects of stratification processes within households with those among pseudo- independent household heads (Curtis 1986; Folbre 1988; Ferree 1990). Most such discus- sions, such as the debate between Goldthorpe (1983) and his critics (Stanworth 1984; Wright 1989), are mired in theoretical anomalies that arise from having to choose either the individual or the household as the sole proper unit of social stratification for empirical research, rather than considering them interactively: Individuals are indeed stratified in households that are themselves stratified, and both hierarchies affect each other. Eitherlor models treat normative fam- ily structures as "natural," so that the eco-

Race matters here as well. For example, Afri- can Americans pay more for comparable housing. Race-based discounting of income, wealth, or skills (discrimination) is not integrated into a gen- eral model of stratification that would explain when and how the White "race" was constructed as a social group entitled to more rewards (Roe- diger 1991; Allen 1994; Omi and Winant 1994).

nomic position of non-White and/or female- headed U.S. families is treated as a product of their "deviant" household types, while the advantages produced for men by the conven- tional family are ignored (Baca Zinn 1991). In fact, so-called "individual" economic suc- cess within the current system depends on unpaid family labor, and the gendered divi- sidn of household labor hasdemonstrable economic benefits for men (Ferree 1990; Bellas 1992). The historical process of recon- ceptualizing a male wageworker as "indepen- dent" and entitled to a "family wage" was steered by the organized gender interests of those men who had the power to define "skills" (Kessler-Harris 1990; Fraser and Gordon 1993). Entitlement to a family wage is not merely a matter of gender and race con- tributing to one's position in an economic hi- erarchy, but is a factor contributing to power in a gendered stratification system. This is re- flected in the Moynihan report and in other discussions of African American men as "emasculated" by limited economic opportu-

nities (U.S. Department of Labor 1965).

Finally, modern political rights were his- torically constructed from analogies of the king to the patriarch and of the state to the household and developed as parts of gender- ed systems in which "the individual" was imagined as an adult male head of house- hold (Pateman 1988). Despite the extension of the franchise to women, political systems remain notably male-dominated structures of power and control. Although the state's role in maintaining class stratification is considered theoretically important (and much debated), the role of the state in ex- pressing and maintaining gender stratifica- tion is typically ignored (Fraser 1989; Gor- don 1990). Yet gender is part of the political process: State-practices maintain certain family types as the norm, reallocate resources between men and women based on gendered concepts of "deservingness" and "dependency," and respond to organized ex- pre'ssions of gender interests (Skocpol 1992; Fraser and Gordon 1993). But these activi- ties are hard to see when class is accorded a priori the central position in models of the stratification system.4

-

The social construction of races is also a po- litical process. States play an active role by de

In sum, feminist critics point out that the mainstream perspective on stratification ob- scures many processes that generate male privilege: the legitimate use of violence, the discounting of women's incomes, the inter- action between households and other eco- nomic structures, and the political construc- tion of groups differentially entitled to rights and resources. Defining "stratification" on the basis of economic resources alone mar- ginalizes stratification processes in which race and gender figure significantly and minimizes the advantages these processes convey to White men, who really are not "ab- stract" or "ungendered" either.

As a result, in the mainstream model race and ethnicity are often viewed as something that non-Whites "have" that "explains their lower status" (Andersen 1987; Rothenberg 1992; Omi and Winant 1994) and gender is interpreted to mean something that women "have" that makes them "different kinds of people" than men (Hess 1990; Lorber 1994).~This perspective in regard to race has been widely and effectively criticized as

fining what sorts of family relationships produce "biological groups," for example, by defining a person with any African A'merican ancestor as "Black" or by legally constructing an Asian or Hispanic "race" (Gilkes 1992; Glenn 1992) and then using these categories to define political or social rights. As stratification systems, the con- struction of race and gender makes the social per- ceptions of "difference" held by more powerful groups into legal facts (Omi and Winant 1994). This process is also obscured by a pre-theoreti- cal assumption that race and gender are self-evi- dent individual attributes relevant to stratification only as they affect placement in an economic hi- erarchy.

Anderson (1987) labels this analytic strategy the "victims-and-variables approach" and argues that it allows sociologists to bring in women and minorities without including an analysis of race and gender as ranking systems in their own right. Of course, in quantitative analyses the variables "race" and "gender" necessarily include males and Whites, but the issue is to determine the questions for which such variables are deemed necessary: When does valid analysis or interpretation call for the variable, and when differences are found, which value is treated as the "baseline"? Whites and males are often treated empirically and theo- retically as the normative cases whose outcomes do not need explanation in terms of institutional- ized race or gender stratification systems.

"blaming the victim" (Ryan 1971). In gen- eral, we see socialization models as distorted into victim-blaming when they (1) put a dis- proportionate emphasis on characteristics of disadvantaged individuals (features of per- sonality or behavior that differentiate them from a White male norm) and (2) use indi- vidual differences to explain differential out- comes (ignoring the effects of systems that organize difference and its meanings). Be- cause treating race and gender as sources of individual-level deficiencies leads analysts to ignore their effects in constructing and main- taining systematic privilege, some feminist theorists now define race and gender as stratification systems that organize interests and institutionalize privilege and oppression for Whites and non-Whites, males and non- males (Scott 1986; Connell 1987; Chafetz 1990; Wallace 1991; Lorber 1994).

FEMINIST ALTERNATIVES

This emerging view of gender as a stratifica- tion system differs from the two earlier per- spectives on gender-the sex-differences ap- proach and the sex-roles approach (Ferree and Hess 1987). The sex-differences model ex~lains behavioral and attitudinal differ- ences by invoking the biological categories of "male" and "female" as a taken-forgranted reality and attempting to identify how much "real" difference remains when socialization is "removed" (Maccoby and Jacklin 1974; Benbow and Stanley 1983). The social processes translating such differ- ences into inequalities are ignored (Ward and Grant 1985). The sex-roles model uses socialization to explain gender differences, fo- cusing on socialization processes in which people take on masculine and feminine iden- tities and learn quasi-permanent "gender roles" (Weitzman 1984; Basow 1992).

Although used by some feminists, the sex- role explanation is similar to a sex-differ- ences account of inequality in emphasizing individual differences as causes. Many soci- ologists studying gender are critical because "sex roles" do not fit a sociological under- standing of other social roles in which there

u

is a relationship between self and other and a process of ongoing interaction to be accounted for. Instead, "sex roles" are typically treated as two polarized and internally con- sistent sets of predispositions acquired in early childhood socialization and exercised regardless of context, leaving the social structure to which people are socialized un- explained (Lopata and Thorne 1978; Connell 1987). Because what boys and girls learn comes from the institutionalized patterns of adult gender inequality, these patterns need explanation at a meso and macro level, which "sex-role theory" fails to provide.6 Instead, "role" is misused to refer to everything from personality traits to concrete behaviors to classes of occupations, while processes of re- sistance, conflict, and change are ignored. This produces an "oversocialized" view of individuals (Gerson and Peiss 1985; Ferree and Hess 1987).

Because the sex-role socialization approach focuses so heavily on the constraints operating in childhood, it makes the rest of the gender regime look voluntary, even agentic. Its explanations rest on how indi- viduals (mostly women) have formed their identities, preferences, and habits rather than addressing the ongoing, multilevel processes of social expectation, control, and struggle that sustain and subvert gender systems. Used to explain gender hierarchy, this sex-role ver- sion of "socialization" encourages a victim- blaming distortion of socialization theory. Like the frequently critiqued "culture-of-pov- erty" model, the sex-role model places the

When looking at mothers socializing daugh- ters to do housework, for example, sex-role mod- els rarely address the real processes of mothers (specific people) teaching daughters (specific people) to do (certain historically and culturally specific tasks of) housework, which are the mi- cro level interactions. The phrase is invoked to implicitly include, but not to analyze, the general social arrangements for housework that organize it as a gender-specific responsibility (and hence the changing palette of tasks that women do in different societies all get called "housework") and to define the tasks called housework as what women will be held socially responsible for do- ing (or failing to do). This is what makes brides struggle to learn what their mothers didn't teach them, and these are really macro-level institu- tional arrangements. When sociologists talk and think about housework as if it were all micro- level socialization, the result is either blaming mothers or feeling helpless when one can't indi- vidually change one's daughters (and hence blaming their biology).

onus for inequality on the supposed charac- teristics of the subordinate group. The un- equal social locations of men and women are individualized into personal identities and "preferences," the distinction between in- equality and difference is obscured, and women are seen as sharing certain behaviors and traits responsible for their position (Risman 1987; Kimmel 1991). Although a gender perspective should include socializa- tion as a means of maintaining a gender stratification system, the system itself cannot be reduced to its socialization processes.

Although theorists of gender (themselves a subset of feminist social theorists) do not agree on any one comprehensive theory of stratification, their critiques suggest some parameters for evaluating the models evident in textbooks. First, they point to processes that establish inequality at micro (interper- sonal), meso (intergroup, intrasocietal), and macro (institutional, cross-societal) levels, rather than resorting to a context-free view of socialization (Smith 1987; Acker 1990; Chafetz 1990; Huber 1991). We expect that the micro-level dominates mainstream ap- proaches to gender because the sex-differ- ences and sex-roles models frame gender as a characteristic of individuals.

Second, some feminist analyses have de- veloped gender parallels to the critiques of models of race that fail to address inequality as a function of something other than "dif- ference" (Collins 1990; Baca Zinn 1991; Rothenberg 1992). We expect to find that the way mainstream textbooks present socializa- tion encourages victim-blaming by a focus on women-as-different and on socialization as producing such differences (as if social- ization alone generated and explained in- equality) to the neglect of meso-level issues like group conflict and macro-level issues like the structure of the economy or political system in various countries.

Third, many gender theorists argue that the effects of stratification processes are interactivt, not additive. ~hu;, for example, within the social framework constructed by race, gender is organized differently for Whites and Blacks; in turn, gender shapes race in different ways for women and men, as was graphically illustrated in the Hill-Thomas hearings (Morrison 1992; Willie 1993). Class is organized by gender, and vice versa, for example, as management was historically separated from clerical work on gender lines (Kanter 1977) and female sex-typed service work still moves between paid and unpaid forms (Glazer 1993).7 Rather than replacing class as the primary dimension, gender can be seen as one of a number of intersecting, interacting forces that stratify access to re- sources, autonomy, and power in society (Andersen and Collins 1992; Rothenberg 1992; West and Fenstermaker 1995). We an- ticipate, however, that gender, race, and class typically will appear as separate dimensions in mainstream sociology.

Attention to such interactive processes does not imply that race, class, and gender are equal in their effects, but argues that they need to be analyzed as theoretically codeter- minative in order to assess their relative sig- nificance for particular individuals, groups, and societies. From this perspective, the mainstream assertion that economic stratifi- cation is primary is regarded as not proven because the claim has not yet been subjected to empirical tests that are not derived from theories that assume this very primacy. We expect that distinctive groups will be pigeon- holed into particular discussions at each level of analysis and for each separate dimension, with social class situated as a structural fea- ture of societies, race as characteristic of groups, and gender as part of individuals.

METHODOLOGY

We sampled introductory sociology text- books published between 1982 and 1988'that appeared in the 1988 edition of Books in Print under the headings of "sociology" or "society." Citations to individual textbooks are referred to by their number on the list of the full sample (see Appendix A). We select quotes from these textbooks to illustrate pat- terns found in our qualitative analysis, but the quotes should not be used to infer which books are "better" or "worse," because the

There are, of course, three-way interactions as well, but these are poorly theorized and rare. For an important exception that looks at the struc- turing of domestic and reproductive labor as paid and unpaid work differently for women and men of different raciallethnic groups in the United States, see Glenn (1992).

AMERICAN SOCIOLOGICAL REVIEW

data are not organized or analyzed in that way. Our focus is on the mainstream sociol- ogy to which these textbooks collectively are oriented, not on the merits or limitations of individual books or their impact on students (Hall 1988). We also compared the eight "best-sellers" to the other textbooks, but we found no important differences on dimen- sions of interest here, even though the best- sellers were more inclusive on some quanti- tative measures (e.g., they were less likely to have only fragments of chapters on gender or race). This suggests that the trend toward more inclusion of gender and race will con- tinue.

Because we incorporate and extend previ- ous research with this data set (Hall 1988; Ferree and Hall 1990), the size of the sample varies. We provide a quantitative analysis-of the full sample of 35 textbooks listed in Ap- pendix A, but select a subsample of 24 books that had distinct chapters or chapter frag- ments on all three topics for more intensive analysis of how race, class, and gender are conceptualized in these chapters.8 We exam- ine 79 such chapters or chapter fragments in these 24 books (numbered 1 through 24 in Appendix A) and 729 pictures in th&e chap- ters. By focusing on those books that include chapters or chapter fragments on gender and race as well as on social class, we bias our sample toward books with a more inclusive perspective and toward best-sellers. Gener- ally, the textbooks placed race, class, and gender chapters or chapter fragments (both of which we refer to as chapters) together

Thus 1 1 textbooks in the initial sample are not analyzed here. Of these, 8 lacked even a chapter fragment on gender or sex as a sociological phe- nomenon, and 3 of these also lacked a chapter on race; 2 additional books were dropped because data on gender and race were too mixed to code as chapters (Kornblum; Walton) and 1 (Landis) was no longer available for additional data col- lection (4 are missing data on cross-national cita- tions'for this reason). Three of these 11 books were judged to be among the best-sellers. How- ever, to maintain maximum generalizability to the universe of 1988 textbooks, the full sample of 35 was used for a survey of overall index citations, and all but the 4 with missing data were included in the analysis of cross-societal citations. The re- sults of these analyses are consistent with the study of the 24 selected books.

under a heading of "inequality" or "social in- equality."

We use index citations, illustrations, and the written text of chapters as the basis of our various measures. Index citations are a con- venient, but possibly biased, measure of in- clusiveness to cover text chapters outside of the race, class, and gender chapters, because we could not scrutinize all this material in detail. Because some authors and/or publish- ers hire indexers and some publishers and au- thors invest more in having a high quality in- dex, we expect variation among textbooks in the reliability of the index, but we do not ex- pect differences in reliability by subject mat- ter. We use the index data primarily to estab- lish a context for the more focused analysis. In the books with sections on class, race, and gender, we examined the actual pages and il- lustrations in these chapters rather than rely- ing on their inde~es.~

We treat coverage of socialization as an in- dicator of micro-level analysis; we also look for attention to gender socialization that is disproportionately extensive relative to other explanations of inequality and for discus- sions of attributes or behaviors of women and/or non-Whites that would indicate vic- tim-blaming. To operationalize dispropor- tionate coverage we count index headings by both topical content and chapter location in the book as a whole; we also compare the lo- cation and nature of pictures and captions about socialization in chapters that focus on race, class, and gender. To operationalize the victim-blaming focus we examine the num- ber, location, and content of the discussions of socialization, whether indexed or not.

We take the coverage of cross-societal data as an indicator of macro-level analysis. We use the term cross-societal to refer to com- parisons both of nation-states (industrialized

We rely on the index for counts of countries mentioned in the three targeted chapters because we consider the names of countries (capitalized nouns) to be relatively unambiguous and to be an item included in even weak indexes. Our own count of such citations would be no less error- prone. Among the 24 books examined in detail, comparing index citations with the actual text they cite in the books' class, race, and gender sec- tions indicated variability among books in the quality of their indexes, but we found no system- atic biases among the three sections.

or not, Western or not) and of cultures, that is, tribal or ethnic groups within or across nation-states; we refer to the former as "cross-national" and the latter as "cross-cul- tural." Cross-societal analysis is undoubtedly an imperfect measure of the presence of macro-level analysis because it ignores any such analysis done completely in the United States. While producing some degree of undercount, we think cross-national data pro- duce a reasonable, conservative measure of macro-analysis, because a primary reason for introducing cross-national data in introduc- tory sociology is to illustrate differences in how whole societies function. If gender is understood as including such macro-level stratification processes, an important way to make this clear is to show variation among nations-differences, for example, in the proportion of women in high government po- sitions, in the size of the wage gap between women and men, and in the extent to which women can move freely without physical constraint or fear of assault.

To operationalize the extent of cross-soci- eta1 comparison, we count the index citations that identify a nation or culture outside the United States;'' we compare the location and content of these citations in sections on class, race, and gender; we code the location and subject matter of illustrations set outside the United States found in these three sections; and we examine written text in these sections for the nature of inequalities described and the explanations offered for inequality in non-U.S. societies.

FINDINGS

To set the context for our comparison of the sections on class, race, and gender, we ex- amine where these sections fit in the book as a whole. Table 1 presents an overview of the

'" Each mention of page numbers was counted as a citation (e.g., a discussion of India's caste system included in the index three times was counted as three citations). Location refers to where the index sent readers to find the discus- sion of India's caste system-chapters on class, race, gender, or elsewhere in the textbook. Con- tent was coded two ways: on the basis of the topic mentioned in the index heading (for non-race/ classlgender sections) and by direct examination of the written material (for these three sections).

Table 1. Frequency Distribution of Coverage of Class, Race, and Gender in Introduc- tory Sociology Textbooks, 1988

To~ic
Measure of Coverage     Class     Race     Gender
Total Sample (N = 35)             
More than one chapter     9     2     0
One chapter     26     25     13
Chapter fragment     0     3     14
Omits topic     0     5     8
"Inclusive" Sample (N = 24)         
More than one chapter     6     1     0
One chapter     18     21     13
Chapter fragment     0     2     11
Total number of pages     839     751     600
Mean number of pages     35.0     31.3     25.0
Standard deviation     11.9     8.2     9.0

differences in coverage. Overall, the text- books give much more attention to economic stratification than to either race or gender. In the full sample of 35 books, all contain at least one chapter on class and 9 contain more than one full chapter. However, 5 books lack even a chapter fragment on race and 8 have no specific section on gender. In the 24 "in- clusive" textbooks, 11 provide less than a full chapter on gender and 2 have less than a chapter on race; but all have at least one chapter on class, and 6 have more than one chapter. Even in "inclusive" books, class re- ceives an average of 35.0 pages of coverage compared to 31.3 for race and 25.0 for gen- der.'' The chapter fragments for gender typi- cally combine gender with age (7 of 10 books), sometimes suggesting that gender is seen as a biological feature of persons (e.g., "Social Differentiation: Sex, Age and Handi- cap"), but sometimes in ways that indicate more attention to social structural issues (e.g., "Gender Roles and Ageism").

Gender as a Trait: Socialization

Disproportionate attention. The tie between gender and socialization is strong. Table 2 indicates that two-thirds of all index citations

I' The pages are somewhat more evenly distrib- uted among the eight best-sellers, although gen- der still lags: 33.3 pages for class, 32.5 for race, and 25.9 for gender.

AMERICAN SOCIOLOGICAL REVIEW

that associate socialization with any form of inequality (class, race, or gender) explicitly link gender to socialization. Three-fourths of these gender socialization citations (48 of 64) send the reader to the chapter on gender inequality, while all but 2 of the 26 class so- cialization citations refer the reader to dis- cussions of basic social processes in the so- cialization chapter. When class socialization is discussed at all, it is not used to explain what class is as a stratificat~on system, but rather is presented as part of what produces the generalized individual.

This same pattern is found with regard to pictures. In the chapters on class, race, and gender, 14 of 18 illustrations captioned as showing socialization or learning are in the gender chapter. The 4 exceptions are all in the race chapter, and all 3 non-Black pictures highlight learning a positive identification with a minority culture. The fourth depicts a young Black woman standing in a poor rural house and is captioned: "The effects of pov- erty and discrimination can continue to pro- duce powerfully debilitating socializing ex- periences" (9:260).

The pictures of gender socialization dis- proportionately focus on children. Twelve of the 14 gender socialization pictures include children. Two pictures that include only little boys show them actively hammering a piece of wood or playing with a toy motorcycle and suggest that such active play is the means of acquiring their "sex role" and is important for later life outcomes. When girls are shown alone they are often depicted as learning to display sexuality. A little girl with fake breasts is captioned as "imitating an adult woman's figure" (18:253), and pre- schoolers in a beauty contest are described as learning to "display femininity" (24:357). A picture of teenage girls learning to use makeup in beauty school tells readers that "it takes a great deal of learning to become the men and women society requires. These young women are making the 'extra effort' to become women" (9:200). Boys are never shown learning to ogle women or learning the masculine version of heterosexuality (at fraternity parties, for example). When paired in the same picture, girls are shown as pas- sive compared to boys: A characteristic ex- ample shows a little girl watching a little boy play with a toy car (15:236).

Table 2. Frequency Distribution of Coverage of Socialization in Chapters on Class, Race, and Gen- der in 24 Introductory Sociology Textbooks, 1988

Chapter
Measure of Coverage     Total     Class     Race     Gender
Index citations associating class, race, or gender with socialization
Refer reader to inequality section
Refer reader to other topicsa
    96    26     6     64
Pictures labeled as about socialization in class, race, or gender chaptersb     18     0     4     14
Books with text devoted to socialization in class, race, or gender chapters One page or more Less than one page No text     ---    0 0 24     0 3 2 1     18 2 4

" Of the 42 referrals to other topics, 33 referred the reader to socialization, 1 to work, 6 to education, and 2 to social change.

Pictures not labeled as being about socialization, especially in the gender chapters, also carry learning

themes.

The text in the class, race, and gender chapters also puts a disproportionate empha- sis on gender socialization. Although the gender chapters are shorter overall, they pro- vide more pages on socialization: 18 books offer more than one page (a mean of 4.3, with a range from 1.5 to lo), 2 provide a page or less, and only 4 provide no direct dis- cussion. In contrast, none of the books offers more than a page of discussion of either race or class socialization in sections on stratifi- cation. Only 3 books provide a page or less on race socialization, and none explicitly cover class socialization at all in the chapter on class. Four books have some discussion in the class chapter that could be interpreted as about socialization, but they never use the term. We thus conclude that the sheer volume of information indicates that sociology un- derstands socialization to be a salient part of what gender is but views socialization as not comparably important for class or race.

Text in the gender chapters frequently pre- sents socialization as an unavoidable and nonproblematic development of masculine and feminine personalities. Readers are told that children "internalize" their parents' con- ceptions of appropriate behavior (6:13 I), that a "variety of personality characteristics are promoted in men and discouraged in women" (20:249), and that gender identity occurs because "social agents . . . transmit social scripts outlining gender-appropriate behavior from one generation to another" (1 1:139). The resulting image is one of oversocialized people; external social control and sanctions are ignored. In pictures and text, gender operates at a personality level, with adult behavior reflecting differences created and reified in childhood. These di- chotomized "roles" are clear, coherent, and uniform rather than full of conflict, variabil- ity, and resistance. Gender "roles" then ex- plain all the ways women "must" act differ- ently than men, and these differences in turn explain inequality.

The deficiency focus. Many textbooks ex- plain patterns of inequality, such as occupa- tional distributions or disparities in office holding, by women's lack of the personality traits needed for achievement. For example, one textbook states that "for girls, computers seem to fit into the traditional math anxiety syndrome, so they shy away from using them. . . . [Tlhis is likely to be a source of continued disadvantage to women" (2:273). Another claims that the "three personality traits associated with adult sex differences are gender identity, self-esteem, and achieve- ment motivation" (23:337). Efforts that women and girls make to claim nonconventional rights and occupational roles are presented in terms of changing socialization. For example, the caption for a picture of a woman lawyer standing in front of shelves of law books says that "gender-role social- ization traditionally de-emphasized aggres- siveness and competitiveness for women. A profession, such as law, which requires these traits, has seen a large influx of women in recent years as gender roles have been reex- amined" (23:337). Here the emphasis is on individual level "traits" rather than on changes in the structure of the profession or elimination of barriers previously erected by law schools and law firms. The invisibility of men in the picture and caption suggests that socialization is literally "for women" and "of women." Although this is just one example, we found many instances of women's personalities being highlighted as the source of greater equality (and so as the primary cause of inequality). Another .text claims, for instance, that "women in success- ful careers have often received socialization that has made them feel particularly secure and certain about being successful" (6:267). The implicit message is that most women have not had this advantage, which is pre- sumably why they are typists rather than managers. Although the statement is prob- ably equally true of successful men, no par- allel sentences appear in the section on class.

The difference between class and gender in whether socialization "counts" as an expla- nation of inequality can also be seen in dis- cussions of class socialization. These few references typically report the development of different attitudes about work and educa- tion and often cite Kohn's (1976) research on working-class and middle-class parents' val- ues for children. They uniformly say that so- cialization explains how individuals react to their places in the system, rather than imput- ing it a role in the origin of inequality: For example, "An individual socialized into one class may never feel completely at home in a higher class or be fully accepted by his or her new peers" (20: 1.86). By contrast, gender so- cialization explains why women are not in conventionally male jobs rather than why men may not fully accept them. The place- ment of most of these discussions outside the chapters on class also reinforces the idea that class socialization does not explain class in- equality.

Using socialization to explain gender in- equality is sometimes explicit. Four of the 18

AMERICAN SOCIOLOGICAL REVIEW

books with more than one page on gender socialization place it in a section that claims to explain inequality, variously entitled "Gender Differences and Sexual Stratifi- cation," "Explanations of Gender Stratifica- tion," "Origins of Sex Differences," or even "What Produces Gender Inequality?" Two others contrast gender socialization with bio- logical origins of sex differences and ignore the social structural link between differences and inequality. One of these books begins a chapter on "Sex Roles" with a six-page dis- cussion of "how different are the sexes?" It includes biological, cultural, historical, and psychological evidence, and concludes that women are different but that socialization rather than biology makes them so. Indeed, only 3 of the 24 books explicitly critique a socialization explanation at all, although 22 present and 18 critique a biological explana- tion. The uncritical use of socialization to explain "sex differences" as women's "defi- ciencies in achievement" stands in marked contrast to the presentation of class and race socialization. Not only are such discussions rare, but they take a different form.

Meso-Analysis: Race as a Group-Level Concept

The chapters on race use socialization con- cepts in a way that links them to meso-level group processes that are ignored for gender and class. The race chapter uses the "social definition" of race to show the arbitrariness of race categories, rather than citing the so- cial construction of differences in "traits." In Omi and Winant's (1994) terms, race is more often seen as "an illusion" than as "an essence." The "culture of poverty" is the most common trait-type argument offered to ex- plain socioeconomic outcomes as a result of race and/or class socialization, but it is rarely offered in the race chapter (1 of 9 books) and is never used without a discussion of its limi- tations. Culture-of-poverty arguments are not labeled "socialization" (either in the index or in the text), nor are they used to illustrate supposedly typical patterns of learning "roles" (e.g., a "poor role").12 Instead, a criti-

"This is why we do not consider their omis- sion from the index to be an error or evidence of bias. They are obviously not intended to be teach-

cal presentation undercuts students' tendency to individualize explanations of economic in- equality; this is lacking from discussions of gender inequality.

Race socialization is depicted as a source of a positive identity, unlike gender or class socialization. Pictures of Indian children dancing in traditional outfits or of a Jewish man teaching the Sabbath rituals emphasize the acquisition of positive ethnic and racial identities; captions comment that "teaching children their cultural heritage is one way of preserving ethnic identity" (1:312) and that "cultural and religious celebrations serve as binding and solidifying events for the Jewish people and other ethnic minorities"(8:243).13 The text elaborates extensively on such posi- tive identity themes. In contrast, class iden- tity or consciousness is seldom explicitly mentioned (only 4 books), and with one ex- ception, the brief, non-indexed references to class-based learning all present class identity as a deficiency, not as a positive resource for working-class people: For example, "Poor people are not the only ones whose psyches are scarred. Most working-class people also endure shame and a sense of failure" (10:239). Gender socialization is also not shown as creating positively valued solidar- ity among women or among men. We were unable to find a single sentence in the sec- tions on inequality that suggested that social- ization builds a positive gender identity.

The race chapters are also the only ones to use socialization models to explain the development of prejudice. For example, one textbook states that patterns of racial preju- dice and discrimination "become incorpo- rated into the values and norms of a group, . . . legitimated, transmitted to new members through the socialization process, and fre- quently internalized" (21:207). In contrast, discussions of gender socialization do not say that men learn prejudice against women or male chauvinism (overvaluing maleness). Few books even raise the issue of men and

ing students what sociologists think about social- ization, learning, or acquiring role proficiency.

l3 African Americans, who are otherwise strongly represented in the chapters on race, do not appear in any of the textbooks' pictures of forming a positive ethnic identity, but an African American woman in a race chapter illustrates so- cialization as producing deficiencies.

boys devaluing and denigrating women, and those that do fail to place it in the context of socialization. ~ender chapters focus so strongly on women as the ones being social- ized that in all 24 books we found only one sentence identifying men or boys as learn- ing prejudice or anti-woman attitudes: "Some feminists have argued that rape and other forms of violence are the end result of norms of aggressiveness that men learn to display to women" (8:281). When prejudice, misogyny, and sexism are mentioned at all, these mentions stress women's learning of sexism (e.g., "sexism is found among women as well as men" [22:246]), women's developing prejudice against other women (often citing Goldberg [1968]), and women's acquiring low self-esteem (e.g., "a damaged sense of self-worth, self-hatred, and dislike of their gender identity" [20: 2411). Cultural messages are interpreted ex- clusively in terms of their impact on women and girls (e.g., "sexist language teaches girls to use 'he' to mean both male and female sex" [10:355] and "sexism appears in the behavior patterns women havetraditionally learned" [6:268]).

When boys and men are presented as so- cialized, the emphasis is on a pseudo-paral- lelism of "sex-role learning" in which male dominance is presented as offering the "bur- den of being male" (22:260) rather than its rewards (e.g., "sex role stereotypes and gen- der roles are also damaging to men. . . . First, men are expected to be strong, aggressive, confident and in control of all situations at all times" [19:354]; "Males are strongly pressured to be assertive, competitive and in control-which is surely a burden to many" [12:298]). We found no examples in which men's "learning to expect to be in control" was advanced as an explanation for wife-bat- tering or other negative outcomes for women. When rape and violence against women are mentioned, textbooks neither dis- cuss male socialization nor present violence as part of a process of gender stratification from which men may derive benefits. Simi- larly, class prejudice, class biases, and class conflict are also neglected themes. Racial and ethnic socialization processes are best represented at the meso-level of forming a group identity and perpetuating intergroup conflict.

Table 3. Frequency Distribution of Cross-Society Comparisons in Chapters on Class, Race, and Gen- der in 31 Introductory Sociology Textbooks, 1988

Measure

Index Citations

Total Cross-Societal Citations

Number of books with citations

Number of citations

Mean per book

Standard deviation

Cross-National Citations

Number of books with citations

Number of citations

Mean per book

Standard deviation

Cross-Cultural Citations

Number of books with citations

Number of citations

Mean per book

Standard deviation

Percentage of all cross-societal
citations that are cross-national

Percentage of cross-national citations accounted for by the three most frequently cited nations

Number of non-U.S. pictures"

Percentage non-U.S. pictures of all pictures in class, race, and gender chapters"

"n the 24 books of the "inclusive" sample.
Macro-Level Comparison: Class Structures

To see what these textbook chapters say about the workings of societies as wholes, we turn to their use of cross-societal data. We distinguish two types of such data: cross-cul- tural data, which use non-state societies (tribes or indigenous groups), and cross-na- tional data, which compare characteristics of state societies. Although cross-cultural data evoke alternative ways of living, their focus on small tribal societies limits their utility for understanding macro-level issues in modern societies. Cross-national data, in contrast, place contemporary U.S. society in a frame- work of alternative social structures. All three indicators (index citations, pictures, and content of text) show consistent differ- ences in the extent and content of compari- sons used for class, race, and gender.

Comparative data are more often used to discuss class than to describe race or gender

Chapter

Total Class Race Gender

-41 44 64

74 33 19 22

-15 7 11

inequality (Table 3). There were an average of 3.39 cross-societal citations for discus- sions of class stratification, 3.19 for race, and

2.13 for gender. This pattern is even more pronounced for the cross-national citations

(3.12 for class, 2.61 for race, and 1.06 for gender) because nearly all cross-societal ref- erences for class (92 percent of 105) are cross-national, while only one-half of the smaller number of cross-societal references in the gender chapters are cross-national (50 percent of 66). The data on pictures support the conclusion that class is more likely to be seen comparatively. Of 74 pictures of non-

U.S. locales, 45 percent occur in discussions of class, at least half again as many as ap- pear in discussions of race (26 percent) or gender (30 percent).

Although only 22 percent (59 of 270) of cross-societal index citations refer to cross- cultural groups, gender in particular is treated this way: More than one-half (56 per- cent) of all 59 non-state comparisons relate to gender, but less than a third (1 8, or 3 1 percent) to race and only 8 (or 14 percent) to class. Cross-cultural references to gender convey specific messages and are drawn from comparatively few cultures. Of 26 dif- ferent non-state cultures mentioned in at least one textbook, only 9 are used for gen- der. Thus, although race and class chapters have fewer cross-cultural comparisons over- all, the examples used are drawn from nearly twice as many different cultural groups. Three-fourths of all cross-cultural references to gender are to the Arapesh, Tchambuli, and Mundugamor cultures of New Guinea stud- led by Margaret Mead. In these discussions, gender appears as culturally acquired person- ality differences rather than as an organizing basis of macro-level social structures. Varia- tion seems to be found only in preindustrial cultures, not contemporary industrial society.

Conversely, most (78 percent) cross-soci- ety comparisons use nation-states and nearly one-half (46 percent) of these comparisons involve class. Of 21 1 cross-national compari- sons, three times as many concern class (97) as concern gender (33), the reverse of the pat- tern found for cross-cultural comparisons. Different nation-states are typically chosen to illustrate each topic, and a few particular countries are ideal types for examining class or race or gender. Of cross-national class comparisons, 41 percent (40 of 97) use the three countries of the Soviet Union, India, and Britain. The top three nations-South Africa, Switzerland, and Brazil-account for nearly one-half (44 percent or 36 of 81) of the comparisons concerning race, and the Soviet Union, Sweden, and China account for two-thirds (64 percent or 21 out of 33) of the comparisons on gender. Thus, far fewer cross-national comparisons are made for gen- der, and these are concentrated among fewer nations, providing a more limited range of evidence for gender than for class or race.

When a cross-national perspective on gen- der is presented, it is often used to suggest the difficulty of change, for example, empha- sizing that policies in the former Soviet Union (N = 9), Sweden (N = 4), or China (N = 3) failed to produce gender equality. Al- though some texts note that women's status improved to some degree as a result of such structural changes, their theme often remains the immutability of women's "different" in- terests or traits. Socialist societies are pre- sented as having made "scarcely any greater progress" (17:303) than the majority of in- dustrial capitalist nations in reducing the gender division of labor. Swedish women are described as "pressured to enter the work force" although they want to stay home with their children (13:212). Another book claims that the division into men's and women's work persists in Sweden because "traditional socialization" teaches women to want to be nurturing and men to be aggressive (24:353). Chinese women are portrayed as "having to work" even though they carry a heavy "double shift," made worse by the absence of household time-saving technology (3:224); no sociological explanation for men "having to work" or for the absence of men from housework is offered. The gender division of labor remains unanalyzed, and social struc- ture appears helpless in the face of women's "preferences" to be nonemployed wives and mothers, suggesting that government pro- grams to reduce gender inequality are of little value. Instead, throughout the gender chapters, resocialization (also termed "edu- cation," "questioning women's traditional roles," or "a new wave of enlightenment") typically appears as the primary source of social change.

In contrast, the persistence of class in- equality in socialist nations is not traced to "class socialization" or "preferences," but is said to reflect structural features of these so- cieties. Although one could argue that citi- zens' "questioning economic hierarchies" (on the model of "questioning women's tra- ditional roles") has indeed fueled social change in class systems, mainstream text- books do not present this as an engine for fu- ture progress toward equality. Instead, eco- nomic hierarchies are presented as stable and fundamental elements of the social order, so that expectations of change seem naive-ironically so, given the massive transforma- tions in Eastern Europe that have occurred since these textbooks were published.

Only one country, the Soviet Union, is dis- cussed in terms of both class and gender (5 textbooks use this example for class only, 5 for gender only, and 4 for both). The conclu- sion for class and gender is the same-gen- der and class inequalities persist despite of- ficial ideology and government policy-but the arguments are markedly different. The class chapter describes opportunity struc- tures and readers are told that the Soviet Union is stratified along political lines, that privileged elites control economic surpluses, and that segments of the population have dif- ferent incomes, occupations, and privileges. Discussions in the gender chapters, however, use the Soviet Union to suggest that oppor- tunity structures matter little. For example, all four gender pictures about the Soviet Union show women in unconventional occu- pations (as doctors, a surveyor, and a laborer using a shovel), yet the texts associated with them emphasize domestic obligations. One caption claims, for example, "In the Soviet Union, 76 percent of all physicians are women. But a female doctor earns substan- tially less money than her male counterparts and must continue to fulfill traditional role expectations at home" (24:354). No connec- tion is made between the demand for unpaid housework and her status in the labor market (why "must" she do domestic labor?). No mention is made of gender as an issue in po- litical control over evaluations of an occupation's worth. Absent any model of what matters, students can only conclude that structural changes are irrelevant. Pictures and text about women's employment in Swe- den and China are used in a similar fashion.

Cross-national comparisons about race are often cast at a meso level of.intergroup rela- tions among segments within a society by contrasting personal prejudice with more in- stitutionalized forms of discrimination. Some race chapters also present macro-level per- spectives on stratifying structures and sys- tems, such as segregation. Although every measure of U.S. labor force segregation sug- gests that gender has a larger effect than race in producing a dual labor market, twice as many textbooks (N = 7) discuss this concept in the race chapter than in the chapter on gender (3 of these 7 books and one other). Jim Crow and apartheid are also often used to illustrate segregation. Unlike gender, most cross-national comparisons (N = 19) of race identify systematic patterns of "intergroup relations" between countries. In sections en- titled "Patterns of Ethnic Relations," "Cross- Cultural Variations in Intergroup Relations," or "Patterns of Minority and Majority Inter-

AMERICAN SOCIOLOGICAL REVIEW

action," several types of group relations are defined, each with its own illustrative nation. Brazil is used as an exam~le of cultural and racial assimilation or even amalgamation, Switzerland for pluralism, and South Africa depicts segregation and subjugation. We found no gender distinctions made at this level (e.g., between rape-prone or relatively rape-free societies, or between countries ac- cording to the degree of segregation between men and women in various institutional realms)

~einforcin~

the message of the importance of structures (not just people and their pref- erences) in group relations, all pictures that illustrate race in South Africa depict signs segregating Blacks to different-beaches, restrooms, buses, or stairways. In contrast, pictures in the gender chapters rarely lack indentifiable individuals (only 11 of 206 to- tal), and none show physical evidences of segregation such as separate entrances for boys and girls in a school.

Any tendency to see race as an attribute of individuals is also actively contested in other ways. Contrasting the race categories of the United States with those of Brazil or South Africa highlights the social arbitrari- ness of definitions of race. When defining a minority group sociologically, textbooks contrast it to a numerical minority by refer- ring to South Africa and emphasize the is- sue of power by referring to the absence of any minority group in Switzerland. Al- though 4 textbooks acknowledge the signifi- cance of power for gender relations (by casting women as "the 52 percent minority" or explicitly developing an analogy with race), most textbooks fail to mention power differences.

Cross-national comparisons in the text ex- plicitly frame class as a structure of society by naming social processes (e.g., "the degree of social mobility in different societies var- ies" [1:276]) and discussing types of stratifi- cation systems. For example, a section (usu- ally entitled "Systems of Stratification") will compare the caste system of India, the estate system of Great Britain, and the class system of the United States, or "open" and "closed" mobility patterns. This is not true for gender: In only 4 books are gender systems named or described, and even these do not challenge the focus on individual differences. For ex-

ample, one book defines its focus as "macro- sociology" and contrasts hunting-gathering, horticultural, agrarian, and industrial societ- ies in their patterned gender division of la- bor. But after presenting such patterns, their significance is discounted in favor of the uni- versality of male dominance by claiming: "Evidence from observations of infrahuman primate behavior, cross-cultural studies of sex differences, and studies of sex differ- ences in human development all point to the conclusion that human biology is a signifi- cant component in differential behavior of the sexes. . . . [Mlen's greater strength is cer- tainly one reason they monopolize the more strenuous and demanding tasks in all societ- ies" (17:307). The three other books that use any cross-national evidence to indicate social structural variation in gender relations none- theless share the dominant emphasis on resocialization (e.g., "Even in Sweden, where the situation is geared for change in gender role definition, time must pass before deep-rooted traditions and attitudes are re- placed" [24:353]). Of course, gender is not shown as invariant; the books make a point of using the Tchambuli, Mundugamor, and Arapesh to show that culture can produce gender-typed people different from those in the United States. But macro-level analysis of the social structures that might underlie this variation is lacking.

CONCLUSION

In sum, gender socialization not only receives far more emphasis than race or class socialization processes, but the nature of what is said about socialization is very dif- ferent for gender than for race or class. "Sex- role socialization" is typically described as the acquisition of traits that impede achieve- ment rather than critiqued as a popular view that ignores macro or meso levels of social structure, as is the "culture-of-poverty" argu- ment it resembles. Women are the ones who are seen as socialized, but not to a positive gender identity that could be a resource in social organization or a source of self-es- teem. Men are never presented as acquiring prejudice or even group solidarity. Gender socialization becomes an explanation of in- equality, offered in a vacuum in which meso- level (group conflict) and macro-level (divi- sion of labor, power, and social control) so- cial processes are virtually invisible.

The introductory textbooks on the market in 1988 show that mainstream sociology in- cluded material on gender and race without reevaluating the concept of stratification guiding the choice of materials to be in- cluded. The differences between the text- books' portrayals of gender and their treat- ments of race and class suggest what was be- ing assumed. Overall, gender is treated most individualistically of all three forms of in- equality in both the extent and nature of cov- erage. The emphasis on socialization as pro- ducing deficiency and the lack of cross-na- tional data interpreted as patterned variation combine to define gender in terms of indi- vidual differences rather than in terms of a system of stratification.

Race and class also have presumed char- acteristics. Race, in particular, is conceptual- ized as intergroup relations, and to some ex- tent both individual-level socialization and national-level patterns of opportunity are also considered relevant. For race, issues of identity, solidarity, and prejudice at micro or meso levels are analyzed as expressing power relations and showing patterns of con- flict and resistance. Class receives predomi- nantly macro-sociological treatment, often at the cost of attention to individual and group levels of analysis, and with a concomitant neglect of themes of class identity and class conflict. The virtually exclusive focus on so- cialization in presenting gender leaves it dis- connected from the sociological modes of analysis that students are otherwise encour- aged to adopt to understand inequality-to see variation in structural arrangements as having consequences for how individuals act and to place the causes of such patterns in something other than "nature" or personal preferences. While the micro-sociological bias in conceptualizing gender is troubling, the macro-sociological bias in discussions of class is no less a distortion. Although in the case of gender, childhood socialization pro- duces women whose preferences as adults

(agency) then dominate structural forces, in

the case of class, both socialization and indi-

vidual agency virtually disappear and social

structures seem more permanent and less

questionable than they in fact are (as the case

of Eastern Europe shows).

Perhaps surprisingly, race and gender are not treated similarly as additive characteris- tics of individuals or as deficiencies. The cul- ture-of-poverty argument is contested, clas- sification of persons into racial groups is analyzed as a social practice, and discussions of race include patterns of cross-national variation. Race is intermediate, not only in the sense of receiving less cross-national analysis than class, and less socialization emphasis than gender, but also in the unique attention paid to the interaction of macro and micro levels (e.g., between power relations and learning prejudice) and the significance given to group processes at the meso level (e.g., solidarity and conflict). This interme- diate, interactive level of analysis might fruitfully be employed for class and gender as well, but is not.

In dealing with class, the lack of group process (meso analysis) and macro-micro links makes the mainstream picture exces- sively static, disembodied, and conflict-free. Themes such as class conflict, class con- sciousness, and changing class identities are precisely the sort of issues that the single- minded focus on the macro level neglects and that an integration of levels would help U.S. sociology to address (Wharton 1991). In dealing with gender, the meso level view would attend to the dynamics of power and autonomy that pattern life chances for White males as well as for other groups. Recent re- search has paid more attention to bringing men into gender analyses (Reskin 1988; Collinson and Hearn 1994) as active defend- ers of their privileges, as enjoying rewards from rape (Scully 1990), or as learning to feel entitled to having housework done for them (Hochschild 1989).

Sociology itself will become better able to explain social dynamics if it avoids reducing gender to personality traits and reifying class into impermeable structures existing inde- pendently of people, because these hidden as- sumptions are pre-theoretical obstacles to adoption of effective theories of structure and agency. Rethinking basic assumptions about stratification, as theories of gender require, suggests new insights into race, class, and gender at all three levels.I4 Such an integrated

l4 To appreciate some of these insights, con- sider what textbooks might look like if they went

sociology would look at gender at the macro level, bringing in more social structure with- out discarding agency (e.g., Dunn, Almquist, and Chafetz 1993), as well as at the meso and micro levels of class systems in which race and gender visibly interact with class forma- tion processes (e.g., the gender and race com- position of corporate networks [Useem 19791, working-class high schools [Bowles and Gintis 19761, and factories [Halle 19841). Attention to the gender and race of White

-

men and to the processes by which they se- cure gender and race advantages at all class levels is central to such a reworking.

We do not expect such rethinking to be re- flected in textbooks until it is more widely

through such a feminist transformation. The choices of pictures and their captions for them suggest some possible directions. Like the racial montages (pictures of people from around the world) that now exist in some textbooks, a mon- tage of women of different ages, ethnic types, and physical appearances could be included with a caption reading, "Despite their obvious differ- ences, all of these people are grouped together as 'women' and their life chances will vary accord- ingly." Or a picture of an Arab woman walking down the street in a chador (a picture some text- books now include) might be captioned, "A chador is a signal in Arab cultures that this is a 'respectable' woman who should not be assaulted by men. In U.S. society, women who walk alone at night are often viewed as 'not respectable' and may be treated as legitimate targets of assault." Or the standard picture of an overwhelmingly White male group of legislators might be cap- tioned, "Welfare policy is typically made by people who are at little risk of ever being recipi- ents of such aid."

Interaction effects of gender, race, and class would also be incorporated. For example, paired pictures of Black and White mothers-and-babies might be captioned, "The work mothers do at home is not valued when their children are not valued. White mothers are often blamed for their children's problems if they don't give up their paid jobs to care for them; Black mothers are blamed if they do." A picture of a White male professional at his desk could be captioned, "Without the invisible labor of women at home, men would not have the ability to work 40 to 60 hours a week to gain success in a professional ca- reer," and an impersonal picture of a store with a Korean sign could be captioned, "The so-called 'self-made man' is often the result of an entire family laboring long hours so that one or more children can get ahead."

practiced in the discipline than it is today, nor do we fault textbooks or their authors for reflecting the mainstream thinking of their times. Yet we hope that as awareness of re- search on gender grows, the anomalies of ex- isting stratification theories will be more strongly felt and debates over how to incor- porate gender and race will intensify. If re- cent gender theories enter the mainstream, we expect that both the demand for inte- grated understandings of class, race, and gender and the supply of research that can be used in textbooks to illustrate this ap- proach will increase. Piecemeal inclusion of such studies in textbooks, however, would still fall short of the deep transformation of stratification theory that a true integration of gender, race and class analysis could provoke and toward which feminist sociologists are beginning to struggle (England 1993). Such a sociological perspective on stratification seems a promising direction for further de- velopment of interdisciplinary theories of gender, as well as a valuable contribution to sociology itself. A feminist transformation of sociology, while perhaps tardy relative to an- thropology or other fields, may ultimately be more far-reaching.

Myra Marx Ferree is Professor of Sociology and Women's Studies at the University of Connecti- cut. Among her recent publications are Controversy and Coalition: The New Feminist Move- ment (coauthored with Beth Hess, rev. ed., Twayne, 1994), Feminist Organizations: Harvest of the New Women's Movement (co-edited with Patricia Yancey Martin, Temple University Press, 1995),and various articles on gender, work, and family, social movements, and feminism in the United States and Germany. Her current research is on abortion discourse in U.S. and German newspapers over the past 25 years and on global

feminist support for the emergence of a women's movement in Russia.

Elaine J. Hall is Assistant Professor of Sociol- ogy at Kent State Universitjl. In addition to being interested in the intersection of class, race, and gender, her areas of interest include the develop- ment of a structural conceptualization of gender, the way work organizations create gendered jobs, and the role of reproductive rights in the mainte- nance of gender inequality. Her current work on the development of a new scale to measure "mod- ern" sexism is the first step in an agenda on the multiple forms of modern or symbolic prejudice- racism, sexism, and classism.

Appendix A. Introductory Textbooks Used in the Analysis

Numbered books have at least a partial chapter on class, on race, and on gender.

(1)
    Bassis, Michael, Richard Gelles, and Ann Le- vine. 1988. Sociology: An Introduction. New York: Random House.

(2)
    Brinkerhoff, David and' Lynn White. 1985. Sociology. St. Paul, MN: West. Charon, Joel. 1980. The Meaning of Sociology. New York: Harper and Row.

(3)
    Conklin, John. 1984. Sociology: An Introduc- tion. New York: Macmillan.
(4)
    Coser, Lewis, Buford Rhea, Patricia Steffan, and Steven Nock. 1983. Introduction to So- ciology. New York: Harcourt.
(5)
    DeFleur, Melvin, William D'Antonio, and Lou- is DeFleur. 1984. Sociology: Human Soci- ety. New York: Random House.
(6)
    Doob, Christopher. 1985. Sociology: An Intro- duction. New York: Holt, Rinehart, and Winston.
(7)
    Eitzen, D. Stanley. 1988. In Conflict and Or- der: Understanding Society. Boston. MA: Allyn and Bacon.
(8)
    Eshleman, J. Ross and Barbara Cashion. 1985. Sociology: An Introduction. Boston, MA: Little, Brown.
(9)
    Hess, Beth, Elizabeth Markson, and Peter Stein.

1988. Sociology. New York: Macmillan. Karp, David and William Yoels. 1986. Sociol

ogy and Everyday Life. Itasca, IL: Peacock.

Kornblum, William. 1988. Sociology in a Changing World. New York: Holt, Rine- hart, and Winston.

Landis, Judson. 1986. Sociology: Concepts and Characteristics.Belmont, CA: Wadsworth.

Lenski, Gerhard and Jean Lenski. 1987. Human Societies: An Introduction to Macrosociol- ogy. New York: McGraw-Hill.

Levin, William. 1984. Sociological Ideas, Con- cepts and Applications. Belmont, CA: Wad- worth.

(10)
    Liazos, Alexander. 1985. Sociology: A Liber- ating Perspective. Boston, MA: Allyn and Bacon.
(11)
    Light, Donald and Suzanne Keller. 1985. Sociology. New York: Knopf.
(12)
    Macionis, John. 1987. Sociology. Englewood Cliffs, NJ: Prentice-Hall.

Orenstein, David. 1985. The Sociological Quest: Principles of Sociology. Saint Paul, MN: West.

(13) Perry, John and Erna Perry. 1988. The Social Web: An Introduction to Sociology. New York: Harper and Row.

(14) Persell, Caroline. 1987. Understanding Society, An Introduction to Sociology. New York:

Harper and Row.

(15)
    Ritzer, George, Kenneth Kammeyer, and Nor- man Yetman. 1987. Sociology: Experienc- ing a Changing Society. Boston, MA: Allyn and Bacon.
(16)
    Robertson, Ian. 1987. Sociology. New York: Worth.
(17)
    Sanderson, Stephan. 1988. Macrosociology, An Introduction to Human Societies. New York: Harper and Row.
(18)
    Schaefer, Richard. 1986. Sociology. New York: McGraw-Hill.
(19)
    Shepard, John. 1984. Sociology. Saint Paul, MN: West. Stark, Rodney. 1985. Sociology. Belmont, CA: Wadsworth.

Stewart, Elbert and James Glynn. 1985. Introduction to Sociology. New York: McGraw-Hill.

REFERENCES

Acker, Joan. 1973. "Women and Social Stratifi- cation: A Case of Intellectual Sexism." American Journal of Soc~ology 78:936-45. , 1989. "Making Gender Visible." Pp. 35- 81 in Feminism and Sociological Theory, edited by R. Wallace. Newbury Park, CA: Sage. . 1990. "Hierarchies, Jobs, Bodies: A Theory of Gendered Organizations." Gender and Society 4: 139-58.

Agassi, Judith Buber. 1989. "Theories of Gender Equality: Lessons from the Israeli Kibbutz." Gender and Society 3(2): 160-86.

Allen, Theodore. 1994. The Invention of the White Race. New York: Verso. Andersen, Margaret. 1987. "Denying Differences: The Continuing Bases for Exclusion in Curriculum." Working Paper Curriculum Se- ries No.4, Memphis State University Center for Research on Women, Memphis, TN. Andersen, Margaret and Patricia Hill Collins, eds. 1992. Race, Class and Gender: An Anthology. Belmont, CA: Wadsworth. Baca Zinn, Maxine. 1991. "Family, Feminism, and Race in America." Pp. 119-33 in The So- cial Construction of Gender, edited by J. Lorb- er and S. Farrell. Newbury Park, CA: Sage. Barnes, Barry. 1982. T.S. Kuhn ancl Social Sci- ence. New York: Columbia University Press. Bart, Pauline and Eileen Moran, eds. 1993. Violence against Women: The Bloody Footprints. Newbury Park, CA: Sage. Basow, Susan. 1992. Gender Stereotypes and Roles. Pacific Grove, CA: BrooksICole. Bellas, Marcia. 1992. "The Effect of Marital Sta- tus and Wives' Employment on the Salaries of Faculty Men: The (House)wife Bonus." Gender and Society 6(4):609-21. Benbow, Camilla and Julian Stanley. 1983. "Sex

(20)
    Stokes, Randall. 1984. Introduction to Sociolo- gy. Dubuque, IA: Brown.
(21)
    Sullivan, Thomas and Kenrick Thompson. 1986. Sociology: Concepts, Issues, and Ap- plications. New York: Macmillan.
(22)
    Thio, Alex. 1986. Sociology, an Introduction. New York: Harper and Row.
(23)
    Tischler, Henry, Phillip Whitten, and David Hunter. 1986. Introduction to Sociology. New York: Holt, Rinehart, and Winston.
(24)
    Wallace, Richard and Wendy Wallace. 1985. Sociology. Boston, MA: Allyn and Bacon.

Walton, John. 1986. Sociology and Critical In- quiry: The Work, Tradition, and Purpose.

Chicago, IL: Dorsey. Zeitlin, Irving. 1983. The Social Conditions qf Humanity: An Introduction to Sociology.

New York: Oxford University.

Differences in Mathematical Reasoning: More Facts." Science 222: 1029-3 1.

Bernard, Jessie. 1929. "The History and Prospects of Sociology in the United States." Pp. 1-71 in Trends in American Sociology, edited by G. Lundberg, R. Bains, and N. Anderson. New York: Harper.

Blumberg, Rae L. 1984. "A General Theory of Gender Stratification." Sociological Theory 2: 23-101. Bowles, Samuel and Herbert Gintis. 1976. Schooling in Capitalist America. New York: Basic. Chafetz, Janet. 1990. Gender Equity: An Inte- grated Theory of Stability and Change. New-bury Park, CA: Sage. Ciancanelli, Penelope and Bettina Berch. 1987. "Gender and the GNP." Pp. 244-66 in Analyzing Gender, edited by B. Hess and M. M. Ferree. Newbury Park, CA: Sage. Collins, Patricia Hill. 1990. Black Feminist Thou- ght: Knowledge, Corzsciousness, and the Poli- tics of Empowerment. Boston, MA: Unwin Hyman. Collinson, David and Jeff Hearn. 1994. "Naming Men as Men." Gender, Work and Organiza- tions 1:1-24. Connell, R. W. 1987. Gender and Power. Stanford, CA: Stanford University. , 1990. "Notes on American Sociology and American Power." Pp. 265-71 in Sociology irz America: A~nerlcan Sociological Association Presidential Series, edited by H. Gans. New- bury Park, CA: Sage. Curtis, Richard. 1986. "Household and Family in Theory on Inequality." American Sociological Review 51:168-83. Dunn, Dana, Elizabeth M. Almquist, and Janet Saltzman Chafetz. 1993. "Macrostructural Per- spectives on Gender Inequality." Pp. 69-90 in

Theory on Gencler/Feminism on Theory, edited by P. England. New York: Aldine de Gruyter.

England, Paula, ed. 1993. Theory on Gender/ Feminism on Theory. New York: Aldine de Gruyter.

Enloe, Cynthia. 1987. "Feminists Thinking about War, Militarism and Peace." Pp. 526-48 in Analyzing Gender: A Handbook of Social Sci- ence Research, edited by B. Hess and M. M. Ferree. Newbury Park, CA: Sage.

Feagin, Joe and Melvin Sikes. 1994. Living with Racism: The Black Midclle-Class Experience. Boston, MA: Beacon.

Ferree, Myra Marx. 1990. "Beyond Separate Spheres: Feminism and Family Studies." Journal of Marriage and the Family 52:866-84.

Ferree, Myra Marx and Elaine J. Hall. 1990. "Vi- sual Images of American Society: Gender and Race in Introductory Sociology Textbooks." Gender and Society 4:500-33.

Ferree, Myra Marx and Beth Hess. 1987. ."Intro- duction." Pp. 9-30 in Analyzing Gender: A Handbook of Social Science Research, edited by B. Hess and M. M. Ferree. Newbury Park, CA: Sage.

Folbre, Nancy. 1988. "The Black Four of Hearts: Toward a New Paradigm of Household Eco- nomics." Pp. 248-64 in A Home Divided: Wotnen arzd Income in the Thircl Worlcl, edited by J. Dwyer and D. Bruce. Stanford, CA: Stanford University Press.

Fraser, Nancy. 1989. "Women, Welfare and the Politics of Need Interpretation." Pp. 144-60 in Unruly Practices: Power Discourse and Gen- der in Contemporary Social Theory, edited by

N. Fraser. Minneapolis, MN: University of Minnesota Press.

Fraser, Nancy and Linda Gordon. 1993. "A Ge- nealogy of Dependency: Tracing a Keyword of the U.S. Welfare State." Signs 19(2):309-36.

Gerson, Judith and Kathy Peiss. 1985. "Bound- aries, Negotiations and Consciousness: Recon- ceptualizing Gender Relations." Social Prob- lems 32: 3 17-31. Gilkes, Cheryl Townsend. 1992. "Disentangling 'Race' in the United States: History, Law, Con- sciousness and the Crises of Diversity." Pre- sented at Eastern Sociological Society, April 4, Arlington, VA. Glazer, Nona. 1993. Women's Paid and Unpaid Labor. Philadelphia, PA: Temple University Press. Glenn, Evelyn Nakano. 1992. "From Servitude to Service Work: Historical Continuities in the Racial Division of Paid Reproductive Labor." Signs 18(1): 1-43.

Goldberg, Philip. 1968. "Are Women Prejudiced against Women?" Transaction 5(5):28-30. Goldthorpe, John. 1983. "Women and Class Analysis: In Defense of the Conventional

View." Sociology 17:465-88.

Gordon, Linda. 1990. Women, the State and Wel- fare. Madison, WI: University of Wisconsin Press.

Gordon, Margaret and Stephanie Riger. 1989. The Female Fear: The Social Cost of Rape. New York: The Free Press.

Grant, Linda, Katherine Ward, and Xue Lan Rong. 1987. "Is There an Association between Gender and Methods in Sociological Research?" American Sociological Review 52(6): 856-62.

Hall, Elaine J. 1988. "One Week for Women? The Structure of Inclusion of Gender Issues in Introductory Textbooks." Teaching Sociology 16143 1-42.

Halle, David. 1984. America's Working Man. Chicago, IL: University of Chicago Press.

Haug, Marie. 1977. "Measurement in Social Stratification." Pp. 51-77 in Annual Review of Sociology, edited by A. Inkeles, J. Coleman, and N. Smelser. Palo Alto, CA: Annual Re- views. Hess, Beth B. 1990. "Beyond Dichotomy: Draw- ing Distinctions and Embracing Differences." Sociological Forum 5:75-93. Hess, Beth B. and Myra Marx Ferree. eds. 1987. Analyzing Gencler: A Handbook of Social Sci- ence Research. Newbury Park, CA: Sage. Hochschild, Arlie. 1989. The Second Shift: Work- ing Parents ancl the Revolution at Home. New York: Viking. Huber, Joan. 1991. "Macro-Micro Links in Gen- der Stratification." Pp. 11-25 in Macro-Micro Linkages in Sociology, edited by J. Huber. Newbury Park, CA: Sage. Kanter, Rosabeth Moss. 1977. Men and Women of the Corporation. New York: Basic. Kessler-Harris, Alice. 1990. A Woman's Wage: Historical Meanings and Social Consequences. Lexington, KY: University of Kentucky Press. Kimmel, Michael. 1991. "From Separate Spheres to Sexual Equality: Men's Responses to Femi- nism at the Turn of the Century." Pp. 301-22 in The Sociology of Gender, edited by L. Kram- er. New York: St. Martin's. Kohn, Melvin L. 1976. "Social Class and Paren- tal Values: Another Confirmation of the Rela- tionship." American Sociological Review 41: 538-45. Kuhn, Thomas. 1970. The Structure of Scientific Revolutions. Chicago, IL: University of Chi- cago Press. Lorber, Judith. 1994. Paradoxes of Gerzcler. New Haven, CT: Yale University Press. Lorber, Judith and Susan Farrell, eds. 1991. The Social Construction of Gender. Newbury Park, CA: Sage. Lopata, Helena Z. and Barrie Thorne. 1978. "On the Term 'Sex Roles'." Signs 3:718-21.

McIntosh, Peggy. 1984. "Interactive Phases of Curriculum Re-Vision: A Feminist Perspec- tive." Working Paper No. 124, Wellesley Col- lege Center for Research on Women, Well- esley, MA.

Maccoby, Eleanor and Carol Jacklin. 1974. The Psychology of Sex Differences. Stanford, CA: Stanford University Press.

Morrison, Toni. 1992. Race-ing Justice, En- gendering Power. New York: Pantheon.

Odum, Howard. 1951. American Sociology: The Story of Sociology in the United States through 1950. New York: Longmans, Green.

Omi, Michael and Howard Winant. 1994. Racial Formation in the United States: From the 1960s to the 1980s. New York: Routledge.

Papanek, Hannah. 1973. "Men, Women, and Work: Reflections on the Two-Person Career." Pp. 90-1 10 in Changing Women in a Changing Society, edited by J. Huber. Chicago, IL: Uni- versity of Chicago Press.

Pateman, Carole. 1988. The Sexual Contract. Stanford, CA: Stanford University Press.

Reskin, Barbara. 1988. "Bringing the Men Back In: Sex Differentiation and the Devaluation of Women's Work." Gender and Society 2(1):58

8 1.

Reskin, Barbara and Patricia Roos. 1990. Job Queues, Gender Queues: Explaining Women's Entry into Male-Dominated Occupations. Philadelphia, PA: Temple University Press.

Risman, Barbara. 1987. "Intimate Relationships from a Microstructural Perspective: Men Who Mother." Gender and Society 1(1):6-32.

Roediger, David. 1991. The Wages of Whiteness: Race and the Making of the American Working Class. New York: Verso.

Rothenberg, Paula, ed. 1992. Race, Class and Gender in the United States: An Integrated Framework. New York: St. Martin's.

Ryan, William. 1971. Blaming the Victim. New York: Pantheon. Scott, Joan. 1986. "Gender: A Useful Category of Historical Analysis." American Historical Re- view 91:1053-75. Scully, Diana. 1990. Understanding Sexual Vio- lence: A Study of Convicted Rapists. New York: Routledge. Skocpol, Theda. 1992. Protecting Soldiers and Mothers: The Political Origins of Social Policy in the United States. Cambridge, MA: Harvard University Press. Smelser, Neil. 1990. "External Influences on So- ciology." Pp. 49-60 in Sociology in America: American Sociological Association Presiden- tial Series, edited by H. Gans. Newbury Park, CA: Sage.

Smith, Dorothy. 1987. The Everyday World as Problematic. Boston, MA: Northeastern Uni-

versity Press.

Stacey, Judith and Barrie Thorne. 1985. "The Missing Feminist Revolution in Sociology." Social Problems 32:301-16.

Stanworth, Michelle. 1984. "Women and Class Analysis: A Reply to Goldthorpe." Sociology 18:159-69.

Stiehm, Judith Hicks. 1982. "Women, Men and Military Service: Is Protection Necessarily a Racket?" Pp. 282-93 in Women, Power and Policy, 1st ed., edited by E. Boneparth. New York: Pergamon.

Sudarkasa, Niara. 1973. Where Women Work: A Study of Yoruba Women in the Marketplace and at Home. Ann Arbor, MI: University of Michigan Press.

Tiano, Susan. 1987. "Gender, Work and World Capitalism: Third World Women's Role in De- velopment." Pp. 216-43 in Analyzing Gender: A Handbook of Social Science Research, edited by B. Hess and M. M. Ferree. Newbury Park, CA: Sage.

U.S. Department of Labor. 1965. The Negro Farn- ily: The Case for National Action. Washington DC: Government Printing Office.

Useem, Michael. 1979. "The Social Organization of the American Business Elite: Participation of Corporation Directors in the Governance of American Institutions." American Soc~ological Review 44:553-72.

Wallace, Ruth, ed. 1991. Feminism and Socio- logical Theory. Newbury Park, CA: Sage.

Ward, Kathryn and Linda Grant. 1985. "The Feminist Critique and a Decade of Published Research in Sociology Journals." Sociological Quarterly 26: 139-57.

Weitzman, Lenore. 1984. "Sex-Role Socializa- tion: A Focus on Women." Pp. 157-237 in Women: A Feminist Perspective, edited by J. Freeman. Palo Alto, CA: Mayfield. West, Candace and Sarah Fenstermaker. 1995. "Doing Difference." Gender and Society 9(1): 8-37. Westhus, Kenneth. 1976. "Class and Organiza- tion as Paradigms in Social Science." American Sociologist 1 1 :38-49. Wharton, Amy. 1991. "Structure and Agency in Socialist-Feminist Theory." Gender and Soci- ety 5(3):373-89. Willie, Sarah. 1993. "What Have Sociologists Learned from the Anita Hill-Clarence Thomas Hearings?" Presented at the Eastern Sociologi- cal Society annual meetings, March 28, Bos- ton, MA. Wright, Eric 0. 1989. "Women in the Class Structure." Politics and Society 17:35-6.

Zelizer, Viviana. 1989. "The Social Meaning of Money: 'Special Monies'." American Journal of Sociology 95:342-77.

Comments
  • Recommend Us