CHANGING GENDER PRACTICES WITHIN THE HOUSEHOLD
A Theoretical Perspective
ORIEL SULLIVAN
Ben-Gurion University
While recent emphasis has been placed on transformations of gender in the public sphere, changes in gender relations between heterosexual couples in the domestic sphere have been less fully developed in the theoretical literature. The author presents evidence for change at various levels, from the discursive to the quantitative. She outlines a theoretical framework for the analysis of such change based on the "doing gender" and gender consciousness perspectives, readdressed in the light of the new emphasis on discourses of reflexivity and intimacy. She argues for a conception of change that is slow and uneven, in which daily practices and interactions are linked to attitudes and discourse, perhaps over generations.
Keywords:    gender relations; household division of labor; time use; social change
A growing body of literature documents changes in gender practices among heterosexual couples within the domestic sphere. However, there is still controversy over whether these changes should be interpreted merely as adjustments occurring in response to changes in the public sphere or as more meaningful indicators of change in gender ideologies and relations in the domestic sphere. In this article, I attempt to do three things: First, I provide a critique of a theoretical context (or contexts) that, I argue, led to a research agenda that mainly emphasized the understanding of stasis in domestic gender relations, rather than the theorization of processes of change. Second, I present a slice of the multilayered evidence for changing gender relations at home. From the wider discursive level, I discuss changes in attitudes to gender equality, images of masculinity/fatherhood, and intimacy in personal relationships. I then present cross-national quantitative evidence based on time-use diaries suggesting long-term changes in gender practices around the domestic division of labor. Finally, I outline a theoretical framework that makes connections
AUTHOR'S NOTE: Sincere thanks are due to Orly Benjamin and Jonathan Gershuny whose insights and collaboration have been invaluable in the development of these ideas. Many thanks go also to the editors and reviewers o/Gender & Society/or their helpful comments and criticisms on this manuscript.
REPRINT REQUESTS: Oriel Sullivan, Department of Behavioral Sciences, Ben-Gurion University; e-mail: sullivan @ bgumail.bgu.ac.il.
GENDER & SOCIETY, Vol. 18 No. 2, April 2004 207-222
DOI: 10.1177/0891243203261571
© 2004 Sociologists for Women in Society
207
208     GENDER & SOCIETY / April 2004
between the wider discursive environment and the micro-level interactions and negotiations that individuals engage in on a day-to-day basis, with the focus on gender practices in the domestic sphere. It is of course the combination of all these different levels that represents the most challenging theoretical task.
In the section of the article that follows, I give a preliminary outline of this theoretical framework. In subsequent sections, I discuss some of the wider theoretical inhibitions to the recognition and analysis of meaningful change in the domestic sphere and present some of the multilevel evidence for such change. In the final section of the article, I discuss examples of research that has explicitly addressed processes of change at the micro level and present a theoretical framework for understanding such change.
THEORIZING CHANGE
As a starting point, I return to the idea of "doing gender" (West and Fenster-maker 1993; West and Zimmerman 1987) as an active (re)construction of gender, located in daily interaction. This is because I do not conceive of the changes I describe as an inevitable process, disconnected from the actions of acting subjects. They have been struggled for, fought over, and hard won over decades not only in the public and political arena but also in innumerable daily contestations and negotiations.
Many presentations of the doing gender approach have addressed its potential for examining alternative and changing constructions of gender (Ferree 1990). However, while the guidelines that regulate appropriate gender behavior are contingent, and vary from time to time and from place to place, the main research emphasis has tended to be on the way in which "doing gender" serves to reproduce existing normative constructions of gender (DeVault 1991; Fenstermaker Berk 1985). It may therefore be helpful to combine the emphasis on interaction and negotiation of the doing gender approach with a theoretical frame in which the conditions for the accomplishment of change in interaction are explicitly addressed. One such frame is provided by the concept of "gender consciousness" (Gerson and Peiss 1985), which according to Thompson (1993), constitutes a central component of our understanding of women's attempts at change. This concept is described as a continuum, at one end of which a generalized awareness of gender issues may be succeeded by a full consciousness of the rights associated with specific gender locations. The conditions under which this consciousness develops depends partly on information from the wider society. For example, the rise of feminism provided new conditions for an awareness of rights and thus for the development of gender consciousness. Gender consciousness may therefore be utilized as a means of describing the potential for change in the domestic sphere in light of new emphases on discourses of intimacy and equality in personal relationships, because these discourses provide new conditions and information for the development of a consciousness of rights.
Sullivan / CHANGING HOUSEHOLD GENDER PRACTICES     209
But day-to-day social interaction also has an influence on the development of gender consciousness. Therefore, the linking together of the analysis of daily interaction with a concept of changing gender consciousness involves the development of a multilevel theoretical frame that can link changes in the wider discursive sphere to processes of change as they occur in interaction between men and women in the domestic sphere. On one hand, it is critical to identify changes at the level of the ideologies and discourses that structure gendered interaction. On the other, the key lies in the detailed analysis of daily interactive processes of change as described by the actors themselves. Women's and men's day-to-day negotiations and struggles around the domestic division of labor can be conceived within such a framework as part of a wider social process that involves slow changes in both consciousness and practices. In this way, it becomes possible to incorporate into the theoretical frame a recognition of the simultaneously constituted and constitutive nature of day-to-day interaction.
Risman (1998) has identified four strands of research on gender, one of which involves the combination of the gender structure and doing gender approaches. I would locate my approach, as she locates hers, in this category. We differ in that I would put more emphasis on the potentially transformative aspect of doing gender. I therefore do not agree that it is necessary to undergo a period of "gender vertigo" to arrive at a transformation of the gender structure. I argue that incremental change can be slow and piecemeal and still in the end effect a radical transformation if we can take the longer perspective.
THEORETICAL CONTEXTS
The literature on late modernity contains a peculiarly apocalyptic vision of change. Large-scale upheavals and dramatic changes are presented as characteristic: Witness the references to "juggernauts," "volcanoes," and the "runway society" (Beck et al. 1994). Such metaphors emphasize size, speed, and drama. The perception of the nature of change that they support and reflect is that only large and dramatic change really constitutes change. This perspective is, I argue, in keeping with the masculinist tradition of classical history, in which large and dramatic moments of change (wars, revolutions, coups, elections, market busts and booms, and other upheavals) are regarded as the material of history. However, these moments of dramatic change may be contrasted with other less dramatic but equally meaningful sorts of change that go largely unrecorded in the pages of classical history but that have informed the content of feminist history. I refer to the small social and economic changes affecting the real-life circumstances of individuals on a day-to-day basis and that, accumulating slowly, practiced and contested in daily interaction, amount in the end to real, substantial, and substantive change.
So in contrast to metaphors of rapid, dramatic change, I want to point to the significance of change based on a different metaphor, of a slow dripping of change that is perhaps unnoticeable from year to year but that in the end is persistent enough to
210     GENDER & SOCIETY / April 2004
lead to the slow dissolution of previously existing structures (see also the section in Connell 2000 titled "The Historical Moment"). In these sorts of changes, daily practices and interactions both reflect and are constitutive of attitudes and discourse, in processes that stretch perhaps over generations. These changes are important, but we should not expect too much from them in a short period of time. However, it is the idea of revolutionary change, the rapid, large-scale upheaval of structures, that has had the most profound influence on modernist thinking on the nature of change. As we might expect, the idea of revolutionary change has had a significant influence on the development of feminist thinking, reaching its peak perhaps in the period following the eruption of the women's liberation movement in the 1960s.
In the early burgeoning of second-wave feminist writing in the late 1970s and early 1980s, emphasis was put on detailing patterns of employment segregation according to gender, on the identification of housework as labor, and on the unequal division of such labor. This literature provided a counterargument to the idea of the "symmetrical family" (Smart and Neale 1999) and made a significant contribution both to the definition of household labor as work (Oakley 1974) and to the construction of theories of structured inequality in gender relations (Lupri 1983). While there has been an increasing recognition of change in the public sphere, there has been among many feminist authors a certain reluctance to accept the significance of changing gender relations within the domestic sphere. A common argument has been that although we might perceive some changes around the edges of the performance of domestic labor, the overwhelming evidence points to a persistence in existing gender relations (Hochschild 1989, 1996; Lorber 1994; Morris 1990; Thompson and Walker 1989).
However, for some time now in both Europe and North America, there has been a growing body of empirical findings documenting change in the relative contribution of women and men to household work. Partly in response to the initial documentation of such changes, more research attention has been focused on issues such as the management and responsibility for household work, as opposed to its performance. Likewise, the focus has shifted to understandings of equity and fairness in domestic labor, posing the important question of how a myth of "fairness" can be maintained in the face of obvious ongoing inequality (Bittman and Lovejoy 1993; Hochschild 1989; Pyke and Coltrane 1996; Sanchez and Kane 1996; Thompson 1991). North American literature, in particular, has examined the socio-psychological mediators and consequences (for example, marital quality) of inequity in household labor allocations (Lennon and Rosenfield 1994; Voydanoff and Donnelly 1999). The careful and critical documentation of housework as labor; the teasing out of the relationships between task performance, task management, and feelings of entitlement and gratitude around domestic labor; and the analysis of how inequity affects life quality has made a huge contribution to our understanding of these issues. Nevertheless, I would argue that some of the theoretical underpinnings of the early research agenda have acted to inhibit the development of a full feminist theorization focused on processes of change.
Sullivan / CHANGING HOUSEHOLD GENDER PRACTICES     211
So what have been these theoretical underpinnings? Willinger (1993) has distinguished two major theoretical starting points used for addressing changes in gender relations: Sex stratification theory and cultural lag theory. In the first, no meaningful change in gender relations can be conceived of without a fundamental upheaval in the structured gender inequality associated with patriarchal social relations. The second carries an implication of evolutionary progressive change occurring across social groups. These theoretical starting points parallel the more general distinction I made above about ways of conceptualizing change. In the early stages of the development of gender theory, there was a strong and politically necessary (Lorber 1994) ideological commitment to the identification and description of patriarchal relations identified with the sex stratification position. The argument for no meaningful change therefore developed within a set of political and theoretical contingencies that made it the right position for the time.
On the other hand, exchange theory and the middle-level theories deriving from it, such as relative resources and time availability (Brines 1994), have appeared regularly in the empirical literature from North America, together with gender role socialization, to explain domestic labor allocations (Coltrane 2000). Where household labor practices are seen as changing directly in response to the movement toward greater equality for women in the public sphere, such models can carry an underlying implication of progressive change. The major feminist criticism of such models has been that they do not sufficiently address existing structural inequalities in gender power and thus questions of constraint. In addition, if an increase in husbands' housework occurs as an ad hoc response to practical contingencies (such as a wife moving into full-time employment), while it may certainly be significant for the women who experience it, it may not necessarily indicate deeper change at the ideological or structural level. So, while being skeptical of the no-change arguments associated with an uncompromising sex stratification theoretical position, it is also necessary to avoid simplistic progressive/evolutionary connotations of movement toward equality.
During the past decade or so, there has been a shift in emphasis on the subject of change in the domestic sphere. Following the development of the gender perspective in the North American literature, with its emphasis on situated interaction, and the move toward a sociopsychological approach including an emphasis on perceptions of inequity and fairness (Kroska 1997; Sanchez 1994; Sanchez and Kane 1996), there is now a more grounded body of situated, processual theory to support the conclusion that we are in the throes of a significant change in gender ideologies and structures in the domestic sphere. Arguments explicitly addressing the extent, nature, and meaning of such change have recently appeared in the U.K. literature (Benjamin and Sullivan 1996,1999; Sullivan and Gershuny 2001). The difficulties are how to generate pressure for it and how to recognize its preconfigurations. In accordance with a general shift within the gender literature toward an emphasis on issues of plurality, differences, and shifting constructions of identity, we need to develop theoretical frameworks of change that can incorporate the complexity of the intersection between ideology, attitudes, and practice in situated, located
212     GENDER & SOCIETY / April 2004
contexts. Such perspectives need to be able to address the tension between the growing cultural emphasis on intimacy in personal relationships and the existing structural bolstering of gender inequality within the domestic sphere.
Some examples of analytical frameworks addressing these elements have appeared recently. I am thinking in particular of Risman's (1998) analysis of the gender structure and Connell's (2000) concept of "configurations of gender practice." But before building on these ideas, I present some of the main elements of the evidence for change. The range of such evidence is large and multilayered. In the following sections, I present a slice of this evidence, ranging from changes in the wider discursive environment to changes in gender practices in the domestic sphere.
EVIDENCE FOR CHANGE I: THE WIDER DISCURSIVE ENVIRONMENT
Attitudes
The association between attitudes to gender equality and the division of domestic labor is by now well established using large-scale data (Goldscheider and Waite 1991), although the direction of causality is not always clear. Lending strength to this association, though, middle-level variables such as self-reported participation in domestic tasks also suggest a relationship with attitudes. In general, those men and women whose attitudes to gender equality are more liberal or progressive tend to share domestic work more equally. However, among the large volume of research on the attitudes of men and women toward their work and family, there have been relatively few large-scale studies that attempt to assess changes in attitudes and values using consistent and comparable measures over time. It is on these studies that I will concentrate. To date, the majority of this work has been conducted in the United States and Britain, but some studies have also appeared that make cross-national comparisons of change. Among the earliest of these was a review of changes in values during the 1980s based on the World Values Survey (Inglehart 1997), but more recently Scott et al. (1996) have provided cross-national comparisons of attitudes featuring several countries of Europe based on large-scale surveys for the 1990s.
Most of the research on attitudes and values conducted in the 1980s found that there had been a movement toward a rejection of traditionally defined gender roles. This mainly took the form of a greater acceptance of nonfamilial roles for women, particularly among younger women with higher levels of education, and a rather less clear movement toward acceptance by men of a more familial role (Willinger 1993). One of the aims of Scott et al. (1996) was to investigate whether changes of this kind had continued into the 1990s. Using cross-national comparative data, they found different patterns of change emerging across different countries. For example, in a three-way comparison of the United States, Britain, and Germany, it was
Sullivan / CHANGING HOUSEHOLD GENDER PRACTICES     213
suggested that the particularly slow rate of change in attitudes among women in Britain is related to structural factors such as the high rate of part-time employment, involving j uggling home and paid work responsibilities. These sorts of associations are supported by various recent cross-national studies (Crompton and Harris 1999; Hofstede and Associates 1998).
Scott et al. (1996,489) concluded that despite intercountry and cross-time variations, "the ideology surrounding traditional gender roles is increasingly rejected, though there is evidence that the pace of change has slowed in the 1990s." They also noted that although overall, women have been much more prepared than men to reject traditional gender role attitudes, within-cohort change in Britain has been more rapid recently among men. This type of change implies a more rapid process of change than between-cohort change. In Inglehart's (1997) study, for example, the emphasis is on intergenerational shifts in values, of which the move toward gender equality is one. However, Scott el al. found that a significant proportion of change (40 to 60 percent) in Britain, the United States, and Germany during the period from the 1970s has been due to within-cohort changes.
Intimacy
The late 1980s and early 1990s saw the publication of a number of relatively optimistic accounts of changes in the nature of personal relationships, in the direction of greater intimacy. Cancian (1987) was among the first to provide an account of changing intimacy that is sensitive to issues of gender. She described the emergence of anew model of "interdependency" in intimate relationships. In this model, "self-development" (which in the traditional model is the preserve only of men) and love are mutually reinforcing. The causal connection is made not to individualism per se but to a more androgynous image of love emerging in the 1970s, together with an increased emphasis on self-development for both men and women. Other dimensions were introduced in the literature on late modernity (Beck and Beck-Gernsheim 1995; Giddens 1992). In this latter literature, a theoretical link was made between changes associated with the late modern period and the development of more intimate and equal relationships, leading to a progressive undermining of gender inequality. These changes were seen as related to the growth of individualism and its emphasis on self-development in an age when kinship and community ties are weakening. These arguments can be linked to gender consciousness through the connection, identified by Skolnick (1992, 41), between the growth of individualism in the late modern period and the "rights revolution," by which is meant the claiming of rights by disadvantaged groups.
Images of Masculinities and Fatherhood
The recent worldwide growth of research on masculinities has been documented by Connell (2000). Writers on masculinity have been at the forefront of arguments
214     GENDER & SOCIETY / April 2004
both for the significance of changing symbolic representations of masculinity and fatherhood—in particular the "new father" who is, according to Knijn (1995), becoming part of male gender identification—and for the existence of change in masculine caring behavior (Coltrane 1996; Hearn and Morgan 1990). More recently, the argument has developed in line with postmodernist thinking on issues of diversity and shifting meanings. For example, the "nurturant new father" referred to by Hochschild (1995), is "born in a context of multiplying ideals and images of a good father," in contradistinction to the simplistic media image of "the new father." Likewise, the changing images associated with "masculinity" emerge from a context of plurality, in which, according to Connell (2000), there exists a hierarchy of masculinities, including a hegemonic masculinity and other forms that contain elements usually associated with feminine caring. The question is, To what extent can the emergence of new and shifting images and ideals of fatherhood and masculinity be linked to empirical changes in practice?
Research from the 1980s documented some changes in paternal behavior, particularly the increase in paternal involvement in child care (Lamb 1986), but at the time, there was little hard evidence in support of the claim for change in behavior. At the turn of the twenty-first century, there is more evidence and more agreement in support of Coltrane's (1998, 106) claim that "the move is towards uncoupling gender from caring." For instance, recent (and not so recent) studies show changes in masculine behavior associated with the taking on of routine responsibility for child care (Coltrane 1996). A growing body of research focuses on "involved fathers" or even "equal caretakers," fathers who participate to greater degrees in caring for children, as opposed to the traditional breadwinner role (Gerson 2001). A number of authors have directly addressed the theoretical basis of the reasons for such changes in the meaning and practice of fatherhood (Hearn and Morgan 1990; LaRossa 1988). While some authors have combined arguments about increasing individualization with the growing significance of the enduring parent-child bond (Beck and Beck-Gernsheim 1995), others have placed more emphasis on the issues of personal identity, arguing that increases in involved fatherhood are "in line with the growing awareness of, or belief in, personal identity as a reflexive identity" (Knijn 1995).
Some of the more optimistic prognoses of future change should probably be regarded with caution. On the other hand, it may be the case that new ideals of fatherhood, currently manifested mostly among the middle classes, will be more likely to spread in the future. There is indeed some evidence that class differences are changing in this respect. Gerson's (2001) in-depth study of involved fathers showed no differences between classes, although the sample was small. There is, in addition, some large-scale quantitative evidence that suggests a narrowing of the class difference in male partners' contribution to domestic labor (see next section).
Sullivan / CHANGING HOUSEHOLD GENDER PRACTICES     215
EVIDENCE FOR CHANGE II: THE USE OF TIME
There are several sources of quantitative evidence for long-term changes over time in the allocation of household labor. The most reliable are based on successive surveys with comparable measures, or on longitudinal data, such as the National Survey of Families and Households panel (Gupta 1999). Time-use diary studies are by now perhaps the main source of such evidence, and they have already confirmed that within specific countries, there have been changes in the amount of time that men and women spend in housework in the direction of greater equity (see, for example, Gershuny 2000 for Britain and Robinson and Godbey 1999 for the United States; for a discussion and defense of the methodology, see Gershuny and Sullivan 1998). Some of the most up-to-date comparative research is based on the multinational time-use study archive held at the University of Essex, England. The large-scale, nationally representative surveys of the archive cover the period from the 1960s to 2000 for more than 20 countries of North America and Europe (see Gershuny 2000, appendix 2, for an introduction to the data set). Analysis of these data for 6 countries of Europe and North America during the period from 1960 to 1997 has shown that while the time per day that women devote to core domestic labor (cooking and cleaning) has decreased by slightly less than an hour, there has been an increase of 20 minutes per day in the time men devote to these activities (Gershuny 2000; Sullivan and Gershuny 2001). These changes, which control for changes over time in both employment status and family status as well as for individual country effects, may not seem that impressive if we calculate the overall decade changes in men's contribution (20 minutes over 30 years). Nor, as is well known, does men's increased contribution compensate for the larger decrease in the amount of time that women devote to these tasks. However, the trends are statistically significant and in the same direction across different countries. They may therefore be described as both robust and internally consistent. In addition, the fact that trends for men and women move in opposing directions is an important feature in the adjudication of a hypothesis for slow change toward a situation of greater equality.
These findings provide us with a cross-national perspective on long-term trends in the amount of time allocated by men and women to domestic work. However, recent research, particularly that influenced by poststructuralist feminist analysis, has pointed to the importance not just of documenting change at the average level but of focusing on plurality and differences as a critical element of understanding. Indeed, Ferree (1991) suggested that aprocessual approach that can incorporate the idea of change can only really arise in the context of a focus on plurality and difference among households. However, most of this research has focused on cross-sectional differences and not on the issue of changing differences over time (for example, Coltrane's 2000 review of household labor research contains a full
216     GENDER & SOCIETY / April 2004
discussion of findings on differences but nothing directly on change). The next example therefore focuses on changing differences over time in partners' contribution to domestic labor using British data from the multinational time-use study collected in 1975, 1987, and 1997.
Controlling for household structure and employment status, the British data show that there was a noticeable growth during the study period in the proportion of couples in which the woman contributed less than 60 percent of domestic labor time, especially among those couples in which both the man and the woman were employed full-time. In 35 percent of such couples in 1975, the woman contributed less than 60 percent of domestic labor time, which increased to 58 percent in 1997. The equivalent figures for those couples wherein the man was employed full-time and the woman part-time were 9 percent in 1975 and 24 percent in 1997. Taken together, the conclusion is that there are now more (i.e., a higher proportion of) equal or more nearly equal couples, particularly among couples wherein both partners are full-time employed.
Cross-sectional research on difference also tells us that husbands from higher socioeconomic statuses are more likely to be contributing more domestic work time. From the British couples data, it is also possible to say something about the relative rate of change among couples in different social classes. While the relative decline in women's contribution applies equally to major social class groups, there is evidence for a larger increase in men's relative contribution (from a slightly lower base) among the manual and clerical category (see Sullivan 2000). Indeed, there may be a suggestion that the manual/clerical group in which both members of the couple are employed have eliminated, if not reversed, the class gap in the division of domestic labor during the study period. This development provides us with a clear example of a changing difference, and it indicates greater change in the direction of equality among a group that had previously been more unequal.
Again, it should be noted that these changes do not challenge the fact that women still perform the bulk of household domestic work or the existence of the dual burden for employed women, particularly those with young children. However, the macro-level changes outlined above control for changing employment statuses and household structures. By factoring out these structural changes, we can compare changes at other levels of behavior, which cannot easily be explained with reference to piecemeal adaptive change to changing structural circumstances.
MODELS FOR CHANGE
As described earlier, the dialectic between interaction and gender consciousness forms the pivot of the approach I am proposing. When individuals do gender within a couple relationship, they do it as part of a dialectic process, which involves both an interpretation of the other partner's gender consciousness and an interaction with their respective doing of gender. This process is also affected by the material and relational resources of both partners and is embedded within a wider discursive
Sullivan / CHANGING HOUSEHOLD GENDER PRACTICES     217
and ideological frame. The attempt to address the interface between these different levels of analysis is grounded in the idea that an understanding of possibilities of change in the sphere of gender relations necessarily involves both interactional and institutional dimensions: "The gender perspective simultaneously emphasizes the symbolic and the structural, ... the interactional and the institutional levels of analysis" (Ferree 1990, 868).
However, studies of the interactional dimension of relationships with the emphasis on understanding processes of change in household labor are relatively rare. Researchers have tended to apply themselves more to examining women's consent and adaptation to their position as the caregivers in their families rather than their attempts to alter their domestic arrangements (DeVault 1991; Thompson 1993). Even in studies of "sharing" couples (Schwartz 1994), the participating couples usually present themselves as intending to share from the starting point of their relationships. Relatively few studies have focused on the detail of processes of change at the level of interaction within couples. Below, I present some examples of studies that have addressed such processes at this level of analysis.
In a study described in more detail elsewhere (Benjamin and Sullivan 1996, 1999), we make a connection between the "relational resources" acquired through occupational training among white English women in different professions and the influence they perceive this to have had on their attempts to negotiate the boundaries regulating the domestic division of labor. We argue that the development of relational resources is connected to a changing gender consciousness and plays a role in both the desire and the ability to accomplish change. Women who have been exposed (through professional development activities, personal counseling, etc.) to influences at the institutional level that promote the development of reflexivity and self-awareness in intimate relationships are more likely to have acquired the relational resources necessary to achieve the successful challenging of normative boundaries regulating both marital "talk" (Zvonkovic et al. 1996) and the material measure of the division of domestic labor. In in-depth interviews with different groups of professional women, some of those with a high level of training and exposure to therapeutic discourses in their professional lives (marriage guidance counselors and social workers) felt themselves to have utilized relational resources developed through this exposure to introduce change into their relationships.
The use of occupation as a key independent variable in the study of the domestic division of labor is a relatively new development, but its significance is that it acts as a mediating connection between the worlds of the public and the private, having a strong influence on both the relational and the material resources that women command within the household. Crompton and Harris (1999), in a cross-national analysis of survey data on attitudes and equality in the home among female doctors and bankers, found that the domestic division of labor was more equal among bankers than among doctors, despite the similarity in terms of material circumstances. They attributed this finding to the managerial training of bankers, by implication including negotiation and personnel skills, as opposed to the professional training of doctors. The results of other studies are also interesting in this respect. Schwartz (2001)
218     GENDER & SOCIETY / April 2004
admitted, for example, that she expected that most of the couples in "peer marriages" that she interviewed would be "yuppie" or "post-yuppie" professional couples. What she found is that most of them were in fact from more modest middle-class occupations such as small business owners, social workers, teachers, and health professionals (but not doctors!). What was missing from her sample were high-earning, high-status professionals such as litigators, investment bankers, and high-status doctors (Schwartz 2001, 189). Her surmise is that a career fast track with associated material rewards is inimical to the development of egalitarian relationships.
In in-depth interviews about micro-level processes of change around the domestic division of labor with Chicana women, Pesquera (1997) also found a relationship between success in efforts to alter the distribution of household labor according to occupational class, reflecting differentials in material and relational resources. One of her main conclusions is that professional and blue-collar workers were able to alter the distribution of household labor to a greater degree than were clerical workers. They did so by using a combination of "gender strategies" including both an "underground approach" (retraining, coaching, praising) and more directly confrontational techniques. She wrote, "A strong relationship emerged between women's willingness to engage in 'political struggle' and the level of male involvement in household labor" (Pesquera 1997, 218). Significantly, the importance of struggle in achieving change was mediated by the differing gender ideologies of the women, for professional and blue-collar workers started out with higher expectations of equality (Pesquera 1997). Where women do not perceive their situation as unequal, or where they have fewer expectations of equality, the stimulus for direct conflict is considerably reduced (Lennon and Rosenfield 1994; Thompson 1991). This is where the link to gender consciousness comes in, since, as described above, such consciousness is strongly linked to differences in expectations and perceptions of rights.
Most of the studies referred to have relied on in-depth interviewing techniques to explore processes of change occurring at the level of daily interaction. Despite the information they provide on process, they are subject to criticism based on the known problems of interview methodology, including recall difficulties and the ex post facto rationalization of events and feelings. Outstanding among the work of authors who have been concerned with the actual observation and description of day-to-day interactions in the household is that of Hochschild (1989, 1996). Her method of family ethnography, although potentially the most effective in that it permits observation of processes of interaction both in situ and in real time, is on the other hand extremely time consuming. In relation to the length of time over which couples often describe any significant change as occurring (which may be over years), it is clearly problematic as a methodology in this area. Undertaken over relatively short periods of time, it will have the tendency to reinforce ideas of stability rather than change.
Doucet (1996) has introduced a different methodological tool with the "household portrait," which is explicitly designed as a participatory technique for facilitat-
Sullivan / CHANGING HOUSEHOLD GENDER PRACTICES     219
ing couples' discussion of how they feel their domestic division of labor has changed over the years. Based on visual cues representing different household tasks, areas, and responsibilities, this technique allowed her to metaphorically follow couples' experiences "through the years of their lives and into the rooms of each household" (Doucet 1996,161). These visual cues provided a reference point for the discussion of household life: "It encouraged discussion, analysis, debate, agreement and disagreement over how each household's particular division of labor had changed over the years" (Doucet 1996,169). Although problems of recall may still be involved, the use of visual cues and the debate aspect of the technique appears to stimulate couples to remember and tell the detailed story of the processes of conflict and change that they have experienced. While interviewing couples together may tend to underrepresent the extent of conflict, it is efficient in getting at both the amount of change and the detailed processes involved. Indeed, one of Doucet's central substantive points is that the picture that we are accustomed to receive of "outstanding stability" of gender divisions of household labor is only a partial one because it fails to take into account the complex explanations that were used by the women and men she interviewed to describe the "daily, weekly, monthly, and yearly layers of difference which may move and change" (Doucet 1996, 166).
All of the above studies involve detailed description at the level of daily interaction of the accomplishment of (or the failure to accomplish) change. To make the theoretical linkage to gender consciousness, and thus to the wider environment, among the issues that need to be focused on are the "boundary points" over which couples experience conflict (Gerson and Peiss 1985) and women's conscious strategies for the stimulation of change, for example, through the deliberate use oranger (Benjamin 1998).
CONCLUSION
One goal of feminist research has been the empowerment of subjects. By focusing on daily interaction as a potentially transformative process, it is possible to conceive of women's everyday struggles around the domestic division of labor as a constitutive part of a wider societal process, involving slow changes both in consciousness and in practice. In the absence of determined policy changes in the area of the family, we can expect such changes to continue to be slow. But an important question must be how we can reintegrate change as a political project not just in the public domain but also at the individual level. I think this is a message that feminism has partially lost. We should acknowledge that, while part of that everyday process involves a struggle to justify or disguise existing inequalities, another part of it also involves a struggle for change, which sometimes in the end bears fruit.
The problem has been that the processes of change that have been described here are slow; they are not easy to identify or analyze; they do not look like changes of the upheaval kind that we are accustomed to identify with late modernity or like a
220     GENDER & SOCIETY / April 2004
transformation of the kind that has occurred in the public arena. I believe that the key to a theoretical framework for understanding these changes lies in the integration of different levels of analysis. Analyses of middle-level variables such as attitudes have indicated substantial shifts across and within generations. But how these attitudes translate into processes of interaction is less well researched. The burgeoning of media on change both in representations of intimacy and in masculinity and fatherhood is also suggestive of wider change. Empirical observations of changes in practice within the home are by now well documented. But again, far less is known about the interactive processes that have led to these changes. At the level of individuals, the key lies in the detailed analysis of processes of change as they occur in day-to-day intimate interaction. Important here are not just observations of changes but also of the processes, negotiations, and struggles that have led to changes, as described by the actors themselves. Combining these levels is the most difficult theoretical task of all. What I have attempted here is to offer some suggestions for a framework for thinking about this combination, in which changes in gender consciousness can be linked in a recursive system both to the wider discursive environment (moving up the level of analysis) and to changes in daily interaction and struggle.
REFERENCES
Beck, Ulrich, and E. Beck-Gernsheim. 1995. The normal chaos of love. Cambridge, UK: Polity.
Beck, Ulrich, A. Giddens, and S. Lasch. 1994. Reflexive modernization. Cambridge, UK: Polity.
Benjamin, Orly. 1998. Therapeutic discourse, power and change: Emotion and negotiation in marital conversations. Sociology 32 (4): 771-93.
Benjamin, Orly, and O. Sullivan. 1996. The importance of difference: Conceptualising increased flexibility in gender relations at home. Sociological Review 44 (2): 225-51.
---------. 1999. Relational resources, gender consciousness and possibilities of change in marital
relationships. Sociological Review 47 (4): 794-820.
Bittman, Michael, and F. Lovejoy. 1993. Domestic power: Negotiating an unequal division of labour within a framework of equality. Australian and New Zealand Journal of Sociology 29 (3): 302-21.
Brines, Julie. 1994. Economic dependency, gender, and the division of labor at home. American Journal of Sociology 100:652-688.
Cancian, Frances M. 1987. Love in America: Gender and self-development. New York: Cambridge University Press.
Coltrane, Scott. 1996. Family man. Oxford, UK: Oxford University Press.
---------. 1998. Gender and families. Thousand Oaks, CA: Pine Forge Press.
---------. 2000. Research on household labor: Modeling and measuring the social embeddedness of routine family work. Journal of Marriage and the Family 62 (4): 1208-33.
Connell, Robert W. 2000. The men and the boys. Cambridge, UK: Polity.
Crompton, Rosemary, and F. Harris. 1999. Attitudes, women's employment and the changing domestic division of labor. In Restructuring gender relations and employment, edited by R. Crompton. Oxford, UK: Oxford University Press.
De Vault, Marjorie. 1991. Feeding the family: The social construction of caring as gendered work. Chicago: University of Chicago Press.
Sullivan / CHANGING HOUSEHOLD GENDER PRACTICES      221
Doucet, Andrea. 1996. Encouraging voices: Towards more creative methods for collecting data on gender and household labor. In Gender relations in public and private, edited by Lydia Morris and E. S. Lyon. London: Macmillan.
Fenstermaker Berk, Sarah. 1985. The gender factory: The apportionment of work in American households. New York: Plenum.
Ferree, Myra Marx. 1990. Beyond separate spheres: Feminism and family research. Journal of Marriage and the Family 52:866-84.
---------. 1991. The gender division of labor in two-earner marriages: Dimensions of variability and
change. Journal of Family Issues 12:158-80.
Gershuny, Jonathan. 2000. Changing times. Oxford, UK: Oxford University Press.
Gershuny, Jonathan, and O. Sullivan. 1998. Sociological uses of time-use diary data. European Sociological Review 14 (1): 69-85.
Gerson, JudithM., andK. Peiss. 1985. Boundaries, negotiation, consciousness: Reconceptualizing gender relations. Social Problems 32:317-31.
Gerson, Kathleen. 2001. Dilemmas of involved fatherhood. In Shifting the center: Understanding contemporary families, edited by S. J. Ferguson (2nd Edition). Mountain View, CA: Mayfield.
Giddens, Anthony. 1992. The transformation of intimacy. Cambridge, UK: Polity.
Goldscheider, Frances K., and L. J. Waite. 1991. New families, no families: The transformation of the American home. Berkeley: University of California Press.
Gupta, Sanjiv. 1999. The effects of transitions in marital status on men's performance of housework. Journal of Marriage and the Family 61 (3): 700-711.
Hearn, Jeffrey, and D. H. J. Morgan, eds. 1990. Men, masculinities and social theory. London: Unwin Hyman.
Hochschild, Arlie R. 1989. The second shift. Berkeley: University of California Press.
---------. 1995. Understanding the future of fatherhood. In Changing fatherhood: An interdisciplinary
perspective, edited by M. C. van Dougen, G. Frinking, and M. Jacobs. Amsterdam: Thesis.
---------. 1996. The time bind. New York: Henry Holt.
Hofstede, Geert, and Associates. 1998. Masculinity andfemininity: The taboo dimension ofnational cultures. Thousand Oaks, CA: Sage.
Inglehart, Ronald. 1997. Modernization and postmodernization: Cultural, economic and political change in 43 societies. New York: Princeton University Press.
Knijn, Trudie. 1995. Towards post-paternalism? In Changing fatherhood: An interdisciplinary perspective, edited by M. C. van Dougen, G. Frinking, and M. Jacobs. Amsterdam: Thesis.
Kroska, Amy. 1997. The division of labor in the home: A review and reconceptualization. Social Psychology Quarterly 60:304-22.
Lamb, Michael E., ed. 1986. The father's role: Applied perspectives. London: Wiley.
LaRossa, Ralph. 1988. Fatherhood and social change. Family Relations 37:451-57.
Lennon, Mary C, and S. Rosenfield. 1994. Relative fairness and the division of housework: The importance of options. American Journal of Sociology 100:506-31.
Lorber, Judith. 1994. Paradoxes of gender. New Haven, CT: Yale University Press.
Lupri, Eugen, ed. 1983. The changing position of women in family society. Leiden, the Netherlands: E. J. Brill.
Morris, Lydia. 1990. The workings of the household. Cambridge, UK: Polity.
Oakley, Ann. 1974. The sociology of housework. London: Martin Robinson.
Pesquera, Beatriz. 1997. In the beginning he wouldn't even lift a spoon. In Situated lives: Gender and culture in everyday life, edited by L. Lamphere, H. Ragone, and R Zavella. London: Routledge.
Pyke, Karen, and S. Coltrane. 1996. Entitlement, obligation and gratitude in family work. Journal of Family Issues 17 (1): 60-82.
Risman, Barbara. 1998. Gender vertigo. New Haven, CT: Yale University Press.
Robinson, John P., and G. Godbey. 1999. Time for life: The surprising ways Americans use their time (2nd Edition). Philadelphia: Penn State University Press.
222     GENDER & SOCIETY / April 2004
Sanchez, Laura. 1994. Gender, labor allocations and the psychology of entitlement with the home. Social Forces 73:533-53.
Sanchez, Laura, andE. Kane. 1996. Women's and men's constructions of perceptions of housework fairness. Journal of Family Issues 17:358-87.
Schwartz, Pepper. 1994. Peer marriage: How love between equals really works. New York: Free Press.
---------. 2001. Peer marriage. In Shifting the center: Understanding contemporary families, edited by S.
J. Ferguson (2nd Edition). Mountain View, CA: Mayfield.
Scott, Jacquie, D. F. Alwin, and M. Brown. 1996. Changing sex-role attitudes. Sociology'iO (3): 427-45.
Skolnick, Arlene. 1992. The intimate environment: Exploring marriage and the family. New York: HarperCollins.
Smart, Carol, and B. Neale. 1999. Family fragments. Cambridge, UK: Polity.
Sullivan, Oriel. 2000. The division of domestic labor: 20 years of change? Sociology 34 (3): 437-56.
Sullivan, Oriel, and J. Gershuny. 2001. Cross-national changes in time-use: Some sociological (histories re-examined. British Journal of Sociology 52:331-347.
Thompson, Linda. 1991. Family work: Women's sense of fairness. Journal of Family Issues 12 (2): 181-96.
---------. 1993. Conceputalising gender in marriage: The case of marital care. Journal of Marriage and
the Family 55:557-69.
Thompson, Linda, and A. J. Walker. 1989. Gender in families: Women and men in marriage, work, and parenthood. Journal of Marriage and the Family 51:845-871.
Voydanoff, Patricia, and B. W. Donnelly. 1999. The intersection of time in activities and perceived unfairness in relation to psychological distress and marital quality. Journal of Marriage and the Family 61:739-51.
West, Candace, and S. B. Fenstermaker. 1993. Power, inequality and the accomplishment of gender: An ethnomethodological view. In Theory on gender/feminism on theory, edited by P. England. New York: Aldyne de Gruyter.
West, Candace, and D. H. Zimmerman. 1987. Doing gender Gender & Society 1:125-51.
Willinger, Beth. 1993. Resistance and change. In Men, work and family, edited by J. C. Hood. London: Sage.
Zvonkovic, Anisa, K. Greaves, C. Schmiege, and L. Hall. 1996. The marital construction of gender through work and family decisions: A qualitative analysis. Journal of Marriage and the Family 58:91-100.
Oriel Sullivan is a senior lecturer in the Department of Behavioral Sciences at Ben-Gurion University, previously at the Department of Sociology, University of Essex, England. She has published widely in the areas of the domestic division of labor and the analysis of time-use diaries.