As Sapiro (Reference Sapiro1981) pointed out many years ago, recognizing that women's interests are interesting is the vital first step in establishing the significance of women's political representation (or the lack thereof)—both in the “real” world and in the scholarly world. Indeed, the assumption that women's interests exist, that women have political interests that can be defined and measured, is central to much of the subsequent research and discussion of women in politics. It is central to our own research on the relationship between women's descriptive and substantive representation (e.g., Reingold Reference Reingold2000; Swers Reference Swers2002), and it is central to this symposium. Yet we come together in this symposium not simply because we share this assumption, but more tellingly because we all grapple with this assumption. Defining and measuring women's political interests pose a number of very difficult questions or dilemmas, which we elaborate in the following. We highlight these challenges not to dismiss such endeavors as futile or necessarily misguided. Rather, we argue that the very uncertainty surrounding women's interests is what makes them so interesting.

Gender politics scholars can and should embrace the uncertainties, critiques, and debates that surround women's interests by taking what we call an “endogenous” approach. Instead of thinking about women's interests as something we must define and operationalize a priori, we might think about women's interests and issues as political phenomena—or variables—worth studying in and of themselves. If we begin with the assumption that women's interests are socially constructed, politically contested, and empirically contingent, then we can further explore how and why the meaning and significance of women's interests vary across time, space, institutions, groups, and individuals. Such an endogenous approach to the study of women's interests raises a wealth of interesting research questions and suggests any number of innovative measurement strategies. In the second half of our essay, we explore some of these possibilities by outlining a research agenda focused on the motivations and perspectives of elected representatives in legislative institutions.

The Difficulty of Defining Women's Interests

Our own research on the policy impact of electing women to office has largely centered on the question of whether increasing the descriptive representation of women leads to more or better substantive representation of women's interests. To examine this question, we (and many other scholars) typically identify a subset of legislation that constitutes “women's issues” and analyze gender differences in legislative activity on these issues.Footnote ¹ In doing so, we inevitably confront a number of questions and criticisms (Jónasdóttir Reference Jónasdóttir, Jones and Jónasdóttir1988; Reingold Reference Reingold2000, 36–46). Can or should researchers define women's interests and issues objectively or subjectively, that is, according to our knowledge of women's social locations and how they differ from men's, or according to women's own conceptions of what their interests and issues are? What do we do when the objective and subjective conflict? Can we or should we even try to identify interests that are inclusive of most women (and exclusive of most men) across racial, class, and ideological lines? Are any interests so widely shared among women and so widely rejected, ignored, or discounted among men? On the other hand, are there any policy debates or political issues in which women do not have a gendered interest? Are all issues women's issues, in this sense? How do we account for the saliency of issues to different groups of women and the intensity of preferences among and across groups of women and men? In short, how can we avoid essentializing or objectifying women but still recognize the political, social, and economic power of gender?

No doubt, many of our colleagues also struggle with these dilemmas. And judging from the variety of definitions and measures of women's interests/issues available in the literature, we have devised numerous responses. Some look to public opinion polls to identify women's interests in terms of gender gap issues—those issues on which women tend to take more liberal positions than do men (e.g., Burrell Reference Burrell1994; Reingold Reference Reingold2000; Welch Reference Welch1985). Some, not surprisingly, look to women's interest groups (or groups claiming to represent women's interests) to identify women's interests (e.g., Burrell Reference Burrell1994; Thomas Reference Thomas1989). Many define women's issues more “objectively” as those they believe are particularly salient to women—either because they primarily, most directly, or disproportionately concern or affect women in particular or because they reflect the more “traditional” concerns (or interests) that women presumably have about others, especially children, families, and those generally in need (e.g., Carroll Reference Carroll1994; Dodson and Carroll Reference Dodson and J. Carroll1991; Reingold Reference Reingold2000; Swers Reference Swers2002). Some distinguish and include both feminist initiatives that promote women's rights or equality and more general or “traditional” social welfare issues (e.g., Reingold Reference Reingold2000; Saint-Germain Reference Saint-Germain1989; Swers Reference Swers2002). Others restrict designated women's issues/interests to only those that are feminist, or at the very least not blatantly antifeminist (e.g., Bratton Reference Bratton2005; Dodson and Carroll Reference Dodson and J. Carroll1991).

Despite these variations, all of these attempts to define and measure women's interests and issues have one thing in common: none relies on the subjects of inquiry (elected officials, in this case) to define (actively or passively, directly or indirectly) or discuss their own conceptions of women's interests/issues. Instead, each of us has defined our terms exogenously, assuming perhaps that our primary task or challenge as researchers is to create a priori a valid, defensible, and appropriate definition of women's interests/issues. Yet no matter how careful, conscientious, or inclusive we may be, we still risk some degree of oversimplification or overgeneralization. By predefining the terms ourselves, we fail to recognize or fully appreciate the degree to which the meaning and significance of women's interests and issues vary and the degree to which they are themselves political. In short, we bypass the opportunity to study and understand women's interests as political variables that are interesting and significant in and of themselves.Footnote ²

An Endogenous Approach to the Study of Women's Interests

Adopting an alternative, endogenous approach to women's interests opens several important avenues for research. To explore some of these possibilities, we turn again to the study of women's representation in legislative institutions. Much fruitful research has illuminated the connection between descriptive and substantive representation by examining the impact of female legislators on policy outcomes, including bill sponsorship, cosponsorship, committee and floor products, and voting (e.g., Dodson Reference Dodson2006; Reingold Reference Reingold2000; Swers Reference Swers2002). An endogenous approach, we suggest, requires scholars to shift their attention to the more deliberative aspects of representation that shape those connections between individual lawmakers and policy outcomes (Mansbridge Reference Mansbridge1999; Phillips Reference Phillips1995). Frequently repeated assertions that “women bring a different perspective to policy deliberations” and that “all issues are women's issues” assume that women draw on their gendered experience as they consider policy dilemmas and develop solutions. These dynamics can only be examined by unpacking women's participation in, and impact on, the deliberative process.

Focusing on deliberation within parties, committees, and among legislators will give us greater leverage over the question of how specific definitions of women's interests and issues emerge in the political process (Celis et al. Reference Celis, Childs, Kantola and Krook2008; Childs, Webb, and Marthaler Reference Childs, Webb and Marthaler2010). By examining the political contest to define women's issues, we will gain an understanding of how the ideologies of liberal and conservative women influence their view of the depth and dimensions of women's interests and the range of acceptable policy solutions (Klatch Reference Klatch1987; Schreiber Reference Schreiber2008; Swers and Larson Reference Swers, Larson, Thomas and Wilcox2005). The increased polarization of the political parties and the expanding number of conservative Republican women at the national and state level require us to examine how gendered life experiences and political ideologies shape legislators’ competing definitions of women's interests/issues. Sarah Palin and her explicit strategy of endorsing female candidates, whom she referred to as “mama grizzlies,” has drawn new attention to the rising activism of conservative women. The relationship of these women to their parties and their views on women's issues, however, remain poorly understood. Furthermore, clarifying how gender and ideology interact to shape legislators’ policy deliberations will allow us to move beyond traditional feminist or social welfare categorizations of women's issues to highlight how gender influences legislator behavior on a range of topics that are not typically characterized as women's issues, such as defense and foreign policy.

Paying close attention to the endogenous process by which women's issues emerge will also lead to a greater understanding of the power relations among and within different groups of legislators. Numerous scholars note that legislative institutions are raced and gendered (Hawkesworth Reference Hawkesworth2003; Kenny Reference Kenney1996). Focusing on the deliberative process will allow us to explain not only how the definition of women's interests differs for a representative from a low-income minority district versus a legislator who represents a district composed largely of white, upper-middle-class professionals, but also how these definitions and their advocates compete for legislative recognition. Processes of marginalization that allow some conceptions of women's interests to gain traction while others are dismissed or ignored can, in turn, reveal much about “raced-gendered” (Hawkesworth Reference Hawkesworth2003) power relations among legislators (Cohen Reference Cohen1999; Strolovitch Reference Strolovitch2007).

Finally, conceptualizing the fight to define women's interests as a political process requires us to focus on the strategic behavior of political parties and the motivations of legislators as they seek to advance their multiple political goals. Political parties work to highlight definitions of women's issues/interests that favor their own interests and mobilize groups of targeted voters (Sanbonmatsu Reference Sanbonmatsu2002; Wolbrecht Reference Wolbrecht2000). Individual legislators work within this partisan context to advance their multiple goals of gaining reelection, serving constituent interests, making good public policy, and advancing their own power within the party and institution (Fenno Reference Fenno1978). As we seek to understand how legislators define women's interests, we must also examine their individual and collective motivations for advancing specific policies. We need to explicate, for example, the normative and empirical implications of a party's decision to select a woman to sponsor a bill because of the symbolic import of having a woman sponsor a particular policy proposal, or a legislator's desire to champion legislation in order to win the women's vote in his/her district. These decisions reflect the multiple motivations of parties and legislators and have important effects on the political contest to define the range and substance of women's interests.

Defining and measuring women's interests are not simply matters of academic debate or methodological hand-wringing. Rather, contestations about the meaning of women's interests are also profoundly political. Endogenous approaches like the ones outlined here prompt us to think about women's interests/issues as deeply embedded in, and integral to, the political processes and conflicts that animate political actors, groups, organizations, and institutions. They direct our attention to describing and explaining how some issues/interests come to be associated with women while others do not; how some are associated with some women and others with other women; and how some “women's” interests/issues are widely accepted as such while others are not. And they compel us to think about the empirical effects and normative implications of such political choices, debates, and outcomes. This is what makes women's interests interesting.