When are “the people”—with all its different groups—represented? It is commonly accepted that democratic representation implies that no significant parts of the population are excluded from the right to vote or to stand for election and, similarly, that parliaments and even governments should, to a certain extent, mirror the represented and governed population. If authorization and accountability indicate the democratic quality of the formal dimension of representation (i.e., it is democratic when it is accountable), then representativeness allows for evaluating descriptive representation (i.e., it is democratic when it is representative of society) (see also Celis Reference Celis2009). But what are our standards for judging the democratic quality of substantive representation? According to Hanna F. Pitkin, that normative standard is the representative's responsiveness: substantive representation is “acting in the interest of the represented, in a manner responsive to them.” Responsiveness turns what representatives do into substantive representation of the demos. It is a metacriterion for democratic representation in the sense that accountability and descriptive representativeness need also to ensure responsiveness. But, again, questions arise: How should we understand responsiveness? How do representatives establish it? Where and when does responsiveness need to be established?
How substantive representation is achieved and the conditions for its realization are, in my view, the key questions for future representation studies.Footnote 1 I side with Pitkin in rejecting too strong a focus on descriptive representation because that diverts attention from substantive representation, which she believes is a far more important dimension of representation. This was challenged by feminist research on political representation, which attaches great importance to the descriptive representation of women. Feminist scholars have done so for intrinsic reasons—democratic values such as fairness, justice, inclusion, and legitimacy—but also to further the substantive representation of women. In earlier work on the substantive representation of women, I argued that our understanding of women's representation would benefit from shifting the attention from the relationship between descriptive and substantive representation toward substantive representation as such: how it is done, who is implied, where it happens, and for what reasons (Celis et al. Reference Celis, Sarah Childs and Krook2008).
These matters, however, pose important challenges, not the least at the conceptual and normative levels. When do women representatives make a difference and come to realize the women's movement priorities? Under which conditions does the substantive representation of women occur? Shifting the focus from the first question to the second requires a clear understanding of what constitutes substantive representation—and the criteria we could use to assess the quality of substantive representation. When can we say that women or ethnic minorities are substantively represented, especially when, fully appreciating the existence of intersectional identities, we recognize that “women” and “ethnic minorities” are highly diversified groups? This is a major challenge for representation studies. In this contribution, I will sketch how we might conceptualize substantive representation and responsiveness in such a way that they can, in the best feminist tradition, be assessed from a normative point of view (Childs and Lovenduski Reference Childs, Lovenduski, Georgina, Karen, Johanna and Laurel2013).
I would like to suggest that plural and possibly conflictual or contradictory interests and perspectives are the basic ingredients of democratic substantive representation. Since women are not a homogeneous group and are expected to have diverging views (e.g., the opinions of leftists versus conservatives/neoliberals, whites versus people of color, high versus lower classes, for instance, on the burqa, quotas, or maternity leave), responsiveness increases when divergent and even contradictory views are included in the representation process. Political deliberation in the course of which different views are argued for and are substantiated contributes to a greater extent to responsive representation in diverse societies than pretending that the conflicting interests do not exist.
I do not, however, exclude the possibility that, with regard to some issues, universal consensus could exist and that, in these cases, representatives should bridge ideological and partisan differences and join forces to translate these particular issues into decisions and policies.Footnote 2 Nevertheless, with regard to most issues, diverging interests do exist, and in these cases, voicing differences in the course of the representative process is key. This approach to the substantive representation of women thus differs fundamentally from investigating whether female representatives or women's policy agencies act in women's interests or realize women's movement interests in several respects. Most importantly, it places political debate and conflict about what is in the interest of women at center stage.
When is substantive representation of women truly responsive? This quality assessment consequently focuses on the diversity of claims and the extent to which these establish a substantive link with the various and varying groups and subgroups in society (see also Celis and Childs Reference Celis and Childs2012a and Reference Celis and Childsb). The outcomes of the representative process—laws and policy—are the temporary result of the struggle over what is in the interest of women; for instance, equality policy should be regarded as “an arena for feminist struggles about what feminism is about and what it can become at a certain time and location” (Verloo Reference Verloo2011, 6). These outcomes should, of course, be responsive to women in society or to women's movement demands. Whereas the possibility that not all women agree with the final outcomes does not affect its democratic quality, I contend that the exclusion of specific women's interests that are supported by specific groups of women in society in the process leading to a law or policy on women decreases its democratic quality and legitimacy. Responsiveness during the process of representation should thus be broad and include a plurality of women, women's organizations, and women's interests.
As a consequence, the level of responsiveness of the process and the outcome of substantive representation greatly depends on the reflexivity of the political and policymaking system (i.e., its capacity to draw out and engage with different groups and perspectives). Reflexivity is a system capacity criterion that measures the extent to which the system furthers input, contestation, and objection from the represented citizens (Disch Reference Disch2011). Legitimate representation is in practice generated by “mobilization democracy,” enabling the interaction of claims and empowering the represented to form their preferences and to make judgments. Reflexivity can be understood as the political system (for instance, parties, MPs, women's policy agencies, and administrative actors) actively searching for divergent, conflicting interests related to women's issues on the political agenda. It is a strategy to capture diverging and conflicting interests and, hence, a prerequisite to establishing responsiveness with as many groups in society as possible.
One might, however, not think of reflexivity as merely “bringing in” preexisting interests and perspectives. Pitkin's conceptualization of substantive representation has recently been criticized by several scholars, especially Saward (Reference Saward2010), on the grounds that it assumes such a static notion of interests as entities that exist “out there,” ready to be brought into the representational process. Newer theories on representation stress the importance of “creative” acts by the representative whereby the constituency in whose interests the representative claims to speak is constructed within the framework of the “representative claim.” Furthermore, substantive representation should, according to Saward, be considered “a market of claims,” an interactive process of interest articulation during which a multitude of interests and perspectives can be formulated by many actors and during which the representative and the represented respond to one another in an iterative fashion.
This dynamic, deliberative, and process-oriented conception of representation makes sense in today's societies marked by diversity and in which meaningful social or political collectives are evaporating and representative democracies face high levels of party dealignment and electoral volatility. It is also better fit to deal with complex and intersecting identities. Nevertheless, responsiveness, the key component of Pitkin's “old” definition, remains a sine qua non: substantive representation fails without firm linkage between what politicians do and what citizens want and need. Responsiveness should, however, also be understood dynamically, as it can be established by elected representatives' claims echoing preexisting claims (classic responsiveness) or by the represented's acceptance of their elected representatives' claims (a posteriori responsiveness) (Severs Reference Severs2010).
Increasing descriptive representativeness as Phillips (Reference Phillips1995) asserts in her politics of presence theory, establishing structural ties with civil society, hearings, women's policy agencies, the participative-democratic model of gender mainstreaming: these are some of the ways in which the political system could operationalize reflexivity. But in order to tap into the diversity of women, it is important to have women present in all political parties and their parliamentary groups and to establish relationships and linkages with different kinds of women's movements (feminist and nonfeminist ones, minority women's organizations, LGBT groups, and organizations anchored in different age and class groups). This is because we need for the substantive representation of women to be responsive to a diversity of women, and to do so, it must be able to reflect diverse, competing, and conflicting claims about women's issues.
If the representative process lacks diversity, competition, and contradiction in terms of the claims that are included, then the chances of the representative process being responsive are low, except when the political system's attempts to capture diverse and competing claims have proven that there are none (although that seems quite unlikely). The absence of a diversity of claims is most probably a sign of the representative system's lacking or slacking efforts to engage with society, of its low mobilizing powers, and thus of incomplete substantive representation. Diverse, complementing, and conflicting claims about what is in the interest of women—a true political debate about what is good and advantageous for women—however, point to the elected assembly's capacity to take into account societal demands through reflexivity leading to classic responsiveness and to produce representations that resonate with specific groups in society through reflexivity leading to a posteriori responsiveness.
