INTRODUCTION
In 2010, the Senegalese president Abdoulaye Wade signed the Law on Parity. This gender quota law obliges all political parties to nominate an equal number of women and men on party lists and as constituency candidates in an alternating matter in legislative and local elections. If parties fail to achieve gender parity on candidate lists, they are not allowed to run for election (JORS 2010). This law proved to be highly effective in the 2012 legislative election when the number of women representatives to the National Assembly increased from 22.7% to 42.7% (IPU 2016), placing Senegal among the top countries in Africa – and the leading Muslim country in the world – in terms of women's political representation.
A central argument among women's activists for implementing parity was that exposure to more women in decision-making roles would challenge people's perception of politics as a male domain and increase belief in women's ability to govern. This constitutes what scholars refer to as the symbolic representation of women. This article seeks to explore whether such expectations have been fulfilled in Senegal through a case study of the Law on Parity. Nationally representative surveys show that the adoption of parity has led to a dramatic and gendered polarisation of attitudes towards women as political leaders: Whereas women's belief in other women's ability to govern has increased substantially, men have become much more negative towards female political leaders. The same goes for a minority of women (Afrobarometer 2018). This article uses these findings as a point of departure and studies the attitudinal effects of the Law on Parity on a different audience: political elites in the National Assembly. My purpose is to examine whether a polarisation of opinions is found among political elites, who are the ones most directly affected by the sudden increase in women parliamentarians. This is important because perceptions of women's ability to govern are likely to influence what they can achieve in terms of policy and career goals. One can also imagine this to have consequences for the recruitment of new female political candidates. Another contribution of this study is that it provides new insight into women's symbolic representation in Muslim Africa.
This study provides a set of indicators that can serve as an analytical framework for future studies on symbolic representation in terms of attitudes towards women as political leaders that has previously been lacking in the literature. Through examining assumptions about women parliamentarians’ competence, gender relations in parliament, delegation of posts of responsibility, perspectives on representation and the legitimisation of gender quotas, I trace whether and how attitudinal changes have taken place in the Senegalese National Assembly since the adoption of parity. My data will be largely based on interviews with current and former parliamentarians of both genders conducted in Dakar, Senegal, in November 2015. I study opinions and experiences before and after the implementation of the parity law and the subsequent influx of women to the parliament. My findings show that the strong polarisation in public opinion is not reflected in the parliament. On at least some indicators, there seems to have been a moderate yet positive development in the belief in women's ability to govern among political elites of both genders.
I will start by explaining the concept of symbolic representation and how this relates to gender quotas. Next, I will present my analytical framework for measuring belief in women's ability to govern, followed by a brief presentation on how the Law on Parity was adopted. I will then analyse my findings based around the framework, and finish with a discussion of the findings.
GENDER QUOTAS AND THE SYMBOLIC REPRESENTATION OF WOMEN
Scholars of women's political representation often build upon Hannah Pitkin's (Reference Pitkin1967) concepts of descriptive, substantive and symbolic representation when studying the impact of gender quotas. This study focuses on the relationship between descriptive (or numerical) and symbolic representation by investigating how an influx of women into parliament through gender quotas may affect people's perceptions of women as political leaders. Feminist scholars have considered increasing the number of women in legislative bodies as valuable if it leads to changes at a broader level. Most research has focused on whether more women in national legislatures leads to more women-friendly policies, which constitutes substantive representation (Thomas Reference Thomas1994; Bratton Reference Bratton2005; Bauer Reference Bauer2012; Muriaas & Wang Reference Muriaas and Wang2012). However, few have studied attitudinal, or symbolic, effects. This may be attributed to the difficulties of operationalising and measuring such an intangible phenomenon.
When it comes to symbolic representation, the importance lies in how people react emotionally to the representative. The feelings and attitudes evoked are not based on any ‘rationally justifiable criteria’ (Pitkin Reference Pitkin1967: 100). What a representative is currently taken to stand for can change through training and habit. The symbolic representation of women in particular is thus related to the social construction of gender roles: After years of women being first excluded, then underrepresented in politics, people have learnt to associate the image of a political leader with that of a man (Kymlicka Reference Kymlicka2002). At the same time, the lack of exposure to women who rule has given people the impression that women are not fit to rule (Mansbridge Reference Mansbridge1999). This can create a bias towards female political candidates competing in elections against male candidates (Dahlerup Reference Dahlerup2006). In theory, this process can be reversed through exposure to more women as political leaders. Even in cases where the inclusion of those previously excluded has no consequences for the resulting policy, it is argued that due to the strong signals it sends, women's mere presence matters (Phillips Reference Phillips1995; Mansbridge Reference Mansbridge1999). Exposure to more women in politics can challenge socially constructed gender roles and public beliefs about politics as a male domain (Mansbridge Reference Mansbridge1999). Also, visible female representatives can act as role models for other women, encouraging women to be more politically involved (Clayton Reference Clayton2014). This does not mean that all women automatically accept women as political leaders; internalised gender roles can be as rigid among women as men (Kymlicka Reference Kymlicka2002: 393). Scholars of symbolic representation have also focused on how the presence of women in politics may affect attitudes towards political institutions. Based on our modern inclusive definition of democracy, the number of women in political institutions may affect the legitimacy of these institutions and democracy in general. This article will follow the first strand by focusing on perceptions and attitudes towards women as political leaders.
Today, gender quotas are found in more than 130 countries worldwide (Krook & Zetterberg Reference Krook and Zetterberg2017), and expectations of positive change in attitudes towards women as political leaders have been used as an argument for adopting such measures. The quota literature is vast, but only recently have scholars started studying its symbolic effects. The results from these studies are however inconsistent. In Rwanda, the quota-driven influx of women to the legislature has led to improved attitudes toward women as political leaders, and an increase in women's status beyond the scope of parliament (Burnet Reference Burnet, Franceschet, Krook and Piscopo2012). In Tanzania, some of the quota-assigned women have produced ‘good governance’ which has lessened previous scepticism towards women's ability to govern, thereby connecting descriptive and symbolic representation through substantive representation (Yoon Reference Yoon2011). In Uganda (Bauer Reference Bauer2012), more women in local legislatures have inspired other women to take part in community activities, while simultaneously affecting men's attitudes toward women as political leaders positively (Johnson et al. Reference Johnson, Kabuchu and Kayonga2003; Bauer Reference Bauer2012). Addressing women's presence in legislatures around the world, Alexander (Reference Alexander2012) shows that more female parliamentarians have positive effects on other women's belief in their own gender's ability to govern. However, among men, there are no significant positive effects, suggesting that women and men react differently to women in political office. In Latin America, a higher number of women legislators, and substantive representation through the passage of laws related to women's issues, tend to have positive effects on attitudes towards government, although this relationship is not systematically linked to quotas (Schwindt-Bayer Reference Schwindt-Bayer2010).
Several scholars also find that gender quotas have no, and sometimes even negative, attitudinal effects. In Lesotho, citizens show less political engagement in constituencies with quota-assigned female leaders (Clayton Reference Clayton2014). In her study of political elites in Belgium, Meier (Reference Meier, Franceschet, Krook and Piscopo2012) finds no support for the claim that gender quotas may lead to changes in attitudes among men, exemplified by the lack of internalisation of the underlying equality norms used for legitimising gender quotas. In Morocco, women elected through quotas are seen as tokens without real decision-making power and are therefore not taken seriously (Lloren Reference Lloren2014). Zetterberg (Reference Zetterberg, Franceschet, Krook and Piscopo2012) finds no positive effects of gender quotas in the form of political engagement among women in Mexico. Finally, women in the Ugandan parliament elected to reserved seats through quotas are less recognised in plenary debates over time than their male and female colleagues in open seats, which signals a lack of respect towards quota-assigned female deputies (Clayton et al. Reference Clayton, Josefsson and Wang2014).
Factors such as quota type, culture, initiators, the media, electoral system and regime type are all presented as possible explanations for the variation in findings illustrated above. This points to the need for a deeper understanding of contextual factors in diverse cases. It is also suggested that the variation is caused by differences in the substantive contribution by women representatives (Franceschet et al. Reference Franceschet, Krook and Piscopo2012: 18). Scholars do not seem to agree on how much time is needed for attitudes to change. Differences in the operationalisation of attitudinal change further complicate matters, which reflect the novelty of the subject in the quota literature. Symbolic representation entails complex phenomena such as attitudes, beliefs, feelings and assumptions. This can be expressed in various and sometimes unexpected ways, which makes operationalisation challenging. Whereas Schwindt-Bayer (Reference Schwindt-Bayer2010) provides a fruitful framework for measuring symbolic representation in terms of attitudes towards government, I here present a framework for measuring attitudes towards women representatives (see Table I). This framework is based on an open research strategy where I combine indicators grounded in existing theory with new discoveries made in the field. The result is a set of indicators that can help measure different aspects of symbolic representation. In this emerging field of research, there is a need for a standard that allows us to compare across cases and thus develop more systematic knowledge. This framework identifies mechanisms and how they play out, which helps in understanding how the parity law relates to the belief in women's ability to govern in the National Assembly. We can thus detect whether changes are manifested in some areas, and not in others.
Table I. Measuring symbolic representation.
Assumptions about competence
Theoretically, gender quotas can generate positive attitudes towards women's competency (Burnet Reference Burnet, Franceschet, Krook and Piscopo2012), which can be traced in assumptions made about the women representatives’ competence. By having parliamentarians describe the characteristics and competence of the women representatives, one can reveal attitudes towards female political leadership that would possibly not have been made apparent through more direct questions. This indicator may show whether more is expected of women representatives than their male colleagues, which would suggest that the image of the political leader as a man remains.
Gender relations in parliament
Because of the historical exclusion of women, parliaments contain a highly masculine culture. When women enter parliament, they may be regarded as outsiders by their male colleagues, which might trigger negative reactions. Sometimes men develop tactics aimed at humiliating their female colleagues with demeaning comments about looks or remarks based on stereotypes (Lovenduski Reference Lovenduski2002). Another theory suggests that exposure to more women in parliament may transform the pervasive masculine culture into being more conducive to women (Wang Reference Wang2014). Reports of hostility or mutual respect between parliamentarians of opposite genders can say something about attitudes towards women as political leaders. Cooperation and dynamics in open debates and subsequent reactions are examples of activities where such attitudes are expressed (Clayton et al. Reference Clayton, Josefsson and Wang2014).
Delegation of posts
The political field is a hierarchical one, which is reflected in the different posts of responsibility within representative bodies. Some posts are considered more important and prestigious than others, but these are rarely taken by women (Miguel Reference Miguel, Franceschet, Krook and Piscopo2012). The hierarchy inside the National Assembly can say something about attitudinal developments. If women are not given posts of responsibility after an increase in their presence, it can suggest that they are considered less able to hold powerful positions than men.
Perspectives on representation
Gender quotas can contribute to positive changes in attitudes toward women as political leaders through widening the perspective both of who the representative can be, and of who should be represented (Meier Reference Meier, Franceschet, Krook and Piscopo2012), thereby transforming the image of the political actor. Who women deputies are claimed to represent, and the importance attributed to their function, can say something about attitudes towards their role as political representatives.
Legitimisation of quotas
The legitimacy of gender quotas rest on the assumption that the public space is not gender-neutral, and that affirmative action is needed to lower the structural barrier for women's access to politics. Support for gender quotas among respondents may therefore involve an internalisation of its inherent gender equality norms ‘by challenging attitudes about the maleness of the public space and by underlining the need for the equality of men and women as citizens and politicians’ (Meier Reference Meier, Franceschet, Krook and Piscopo2012: 162). Thus, the degree to which gender quotas have been legitimised can say something about attitudes towards women's political leadership.
THE ROAD TO PARITY
Theory suggests that to understand symbolic effects of gender quotas, one needs to understand how gender quotas are adopted. The Senegalese women's movement is among the strongest in Africa. Still, the road to parity has been a long and complex process and attempts at promoting women's rights were uncoordinated and weak until preparations for the 1995 United Nation's Fourth World Conference in Beijing started. During this period, the Senegalese Council of Women, COSEF, saw the light of day. COSEF drew its members from all over the political spectrum and were present in all 10 regions of Senegal. Its main focus was women's political marginalisation (Diop Reference Diop2011). COSEF established broad alliances with civil society organisations such as the Senegalese Association of Women Jurists (AFJ) who helped develop proposals to the electoral code to achieve parity. COSEF and AFJ claimed that women's contribution to politics was mostly as political marionettes mobilising votes and that a ‘true female citizenship’ was lacking. COSEF stressed the need to challenge people's perceptions of what a political actor could be through a greater presence of women in real decision-making positions (Diaw Reference Diaw1999: 31). Hence, expectations of symbolic effects were an important motivation behind the parity process.
COSEF campaigned vigorously through the years, and arranged marches, held meetings in rural districts, distributed leaflets, made T-shirts and theatre plays and more. During these campaigns, the word quota was avoided while parity was highlighted. Activists lobbied for parity in the National Assembly and focused on familiarising the public with the parity concept through cooperation with media associations. This may have been a wise strategy. Parity is a newer concept and is stressed by its proponents to entail something philosophically different than gender quotas (Krook Reference Krook2013), thereby avoiding some of the controversy surrounding quotas.
COSEF found an ally when Abdoulaye Wade was elected as president in 2000. Initially, Wade had the image of a modern reformer, and campaigned under the promise of change. He adopted a new constitution, which guaranteed women's rights and access to decision-making instances. This women-friendly behaviour fuelled COSEFs mobilisation for parity, and its jurist members started working on a parity model law. However, President Wade's popularity started waning after years of concentrating power within the executive, infrastructure problems and increasing costs of living. Wade himself was also aware of the fact that parity was in line with international norms; the promotion of gender quotas worldwide had been on the international agenda since the Beijing Conference in 1995. It is not uncommon for political leaders to use gender quotas to increase support from the female electorate (Krook Reference Krook2006). They sometimes also adopt quotas with the purpose of presenting themselves and their countries as ‘modern’ or ‘democratic’ to the rest of the world. Countries like Senegal, which are heavily indebted, largely dependent on foreign aid and/or tourism, are more likely to adopt gender quotas than more financially independent countries (Bush Reference Bush2011; Hughes et al. Reference Hughes, Krook and Paxton2015). Although there were some opposition to parity in the parliament, the president holds great powers and parliament rarely votes down his proposals. Furthermore, political pragmatism often defeats personal convictions when highly media-covered issues such as parity are on the agenda (Murray et al. Reference Murray, Krook and Opello2011). Thus, in 2010 President Wade signed the Law on Parity, which first came into effect in the 2012 legislative election. Two years later, local elections with parity lists were held for the first time. In both elections, women's representation almost doubled.
In the 2014 local elections, the power of religious leaders came into play when the holy city of Touba presented an all-male list of political candidates, ignoring the obligations of the Law on Parity completely. Despite criticism from the Electoral Commission (CENA), the Khalif refused to succumb, declaring that the only valid law in Touba was the Sharia (Le Dakarois 2016). While non-compliance to parity has led CENA to reject party lists elsewhere, Touba has faced no such sanctions. In fact, in 2014 CENA declared that Touba constituted a special case, and that the ‘sociological realities of Senegal’ should be considered when implementing new laws (Faye Reference Faye2014). This illustrates that although Senegal is a secular republic, religion remains paramount: Sufi Islam is omnipresent in Senegalese society, where a majority belong to a Sufi brotherhood (The Pew Forum 2012). The powerful Sufi leaders, the marabouts, have exercised considerable economic and political power since before colonial times. The French increased this power through a cooperative approach towards the marabouts to maintain control over parts of the territory where their direct authority did not reach (Creevey Reference Creevey1996: 275). Today Senegalese society is marked by a ‘lack of uniformity’: On one side there is the importance of Islam and its leaders which has long marked the link between politicians and the people. On the other side, politics are dominated by the often Western-educated urban elites concerned with the modernisation of Senegal (Creevey Reference Creevey, Purkayastha and Subramaniam2004: 62). One can expect this context to have consequences for the belief in women's ability to govern.
CASE SELECTION: THE SENEGALESE LAW ON PARITY
The Senegalese parity law provides an interesting case for examining the attitudinal effects of gender quotas for several reasons. Firstly, while quotas are spreading to other parts of Africa, most research still focuses on the pioneer countries in the southern and eastern part of the continent, while West Africa has been mainly overlooked. Whereas Kang (Reference Kang2015) provides a thorough study of gender quotas in Niger, it does not address the issue of attitudinal effects. Apart from Lloren's (Reference Lloren2014) study of Marocco, little research on symbolic representation has focused on Muslim countries. Some scholars link Islamic heritage to patriarchal values and lack of gender equality (Alexander & Welzel Reference Alexander and Welzel2011; Inglehart & Norris Reference Inglehart and Norris2003a, Reference Inglehart and Norris2003b). However, patterns are rapidly changing in many African Muslim states with regards to women's legislative representation. Tanzania, Mauritania, Eritrea, Sudan, Niger and Somalia all have significant Muslim populations and gender quotas (Tripp Reference Tripp2013). There is thus a need for research to keep up with these developments.
Secondly, many of the African countries that have adopted gender quotas have done so in a post-conflict context. In the institutional vacuum that appears after a destructive civil war, women's movements often use the opportunity to claim their rightful place in the new society and create women-friendly laws and institutions such as gender quotas (Tripp Reference Tripp2015). Senegal, however, is one of the most stable countries in West Africa and has never experienced a coup d’état or military rule. The positive symbolic effects found in Uganda (Johnson et al. Reference Johnson, Kabuchu and Kayonga2003; Bauer Reference Bauer2012) and Rwanda (Burnet Reference Burnet, Franceschet, Krook and Piscopo2012) may be attributed as much to the upheaval of traditional gender roles caused by violent conflict as to the effect of gender quotas. One can thus argue that the study of quotas in stable Senegal holds important explanatory power with regards to symbolic effects.
Another reason the Law on Parity makes an interesting case is that the strong influx of women to parliament during the 2012 election facilitates the tracing of attitudinal effects, and the proximity in time allows capturing people's attitudes before and after the quota implementation. Consequently, this study addresses the short-term symbolic effects of gender quotas.
Finally, survey data from Senegal shows some puzzling results that calls for closer scrutiny of the dynamics of attitudinal change (see Table II). A surprisingly strong and gendered polarisation of public attitudes has taken place in the aftermath of the implementation of parity. People's strong support of women's rights to be elected have increased by 11% from 2005/2006 to 2014/2015. Simultaneously, those who agree very strongly with men being better political leaders than women, have increased more, by 14%. Those who just agree with either statement, has decreased. It thus appears that people have had their original perceptions reinforced.
Table II. Women vs. men as political leaders in Senegal.
Statement 1: ‘Women should have the same chance of being elected to political office as men.’
Statement 2: ‘Men make better political leaders than women, and should be elected rather than women.’
Source: Afrobarometer (2018).
The findings become even more interesting when separating between genders. There is a slight positive change in men's strong support for the statements that women should have the same chance at being elected to political office as men. However, among women, strong support has increased by 20%. Furthermore, strong beliefs about men being more fit to govern than women have risen dramatically among men during the same period by 23%. Fewer women share this opinion, but there is still an almost doubling in strong support for men being better political leaders.
The explanatory power of the Law on Parity on these dramatic changes increases when comparing public attitudinal change in Senegal with countries that has not adopted such radical quota legislation. Neither sub-Saharan Africa nor West Africa has seen a sudden upturn in women's strong support for female political leadership (see Appendix). Simultaneously, the strong support for men as political leaders has increased more in Senegal than in the rest of Africa. A paired comparison of Senegal and Niger, which are both secular republics with a predominantly Muslim population, and similar in terms of colonial past, geographic location, development- and democracy level, further strengthens the explanatory power of gender quotas with regards to attitudinal change. Niger has a much more moderate form of gender quotas than Senegal, and women made up just 13% of the Nigerien parliament at the time of this study (IPU 2016). The gendered polarisation of opinions found in Senegal is not reflected in Niger. In fact, Nigerien men and women seem to agree more today than before. There are thus strong reasons to believe that the Law on Parity has played an important role in attitudinal changes in Senegal. Are these changes reflected in the Senegalese parliament?
Methodological design
To capture the impact of gender quotas on women's symbolic representation, I study opinions and experiences before and after the implementation of the parity law and the subsequent influx of women to the Senegalese National Assembly. I rely on semi-structured interviews with 20 interviewees conducted in Dakar, Senegal, in November 2015. To capture attitudinal change, I divided the sample in two: three female and two male former parliamentarians who sat in the National Assembly prior to the parity law who did not get re-elected, and four male and four female parliamentarians elected after the implementation of parity. By asking these two groups a similar set of questions regarding their challenges, accomplishments, main issues, perceived roles and contributions, as well as views on female parliamentarians and the consequences of the law on parity, I was able to compare whether attitudes toward female political leadership had changed. One criticism of this set-up could be that those who did not get re-elected could be more negative towards their experiences in parliament than those who recently got elected. However, the reverse dynamic might also apply; people tend to forget negative experiences that happened in the past, whereas the present is judged more harshly.
I used snowball sampling in order to access participants and was invited to the parliament through a gatekeeper who worked there as a civil servant. He helped me connect with current deputies, who then introduced me to former deputies. To avoid interviewing only people who knew each other – and possibly shared the same views – I also approached deputies wandering around the corridors of the parliament and asked them for interviews.Footnote 1 For contextual insight, I relied on additional information from a total of seven academics, civil servants and representatives from civil society.
BELIEF IN WOMEN'S ABILITY TO GOVERN AMONG POLITICAL ELITES
Assumptions about competence
Theoretically, more women in legislatures can generate positive attitudes about women's competences. However, after the implementation of the Law on Parity, an alleged lack of competence among the newly elected ‘quota women’ grew to be an issue. Because of the new requirements of a 50/50 gender balance, some parties reportedly struggled to find ‘educated’ or ‘experienced’ women for their candidate lists in the following elections, and there were reported cases of ‘just filling up lists’ (Wade, Diallo & Sy 2015 Int.). The claim that gender quotas promote unqualified women is not new and is often voiced in quota-adopting countries (Jamieson Reference Jamieson1995). This study finds no support for such claims. According to the parliamentarians’ biographies on the parliament's webpage, both the men and women deputies are equally highly educated and have relevant experience (L'Assemblée Nationale de la République du Sénégal 2018). However, around 50% of the representatives lack biographies, thus one cannot assume that this level of competence applies to everyone. Still, the number of deputies’ lacking biographies is equally distributed among men and women, implying that one cannot assume that there are more unqualified women than men either. These findings are supported by several studies that show that quota-assigned women are not less competent than other members of parliament (O'Brien Reference O'Brien, Franceschet, Krook and Piscopo2012; Murray Reference Murray, Franceschet, Krook and Piscopo2012; Franceschet & Piscopo Reference Franceschet, Piscopo, Franceschet, Krook and Piscopo2012; Josefsson Reference Josefsson2014; Nugent & Krook Reference Nugent and Krook2016). Some ‘quota women’ in fact have more political experience than their male colleagues, and a high performance in parliament. Rather than promoting unqualified women to parliament, gender quotas contribute to lowering the barriers that keep qualified women out (Nugent & Krook Reference Nugent and Krook2016).
This very discussion of competences can reveal much about attitudes towards women as political leaders. Reportedly, there have been many uneducated men in the National Assembly without their competences ever being an issue both before and after 2012 (Sané 2015 Int.). A female former deputy stated that: ‘We always see illiterate men get elected to the municipalities as well as to the National Assembly. So why say now that women lack education and cannot be there?’ (M. Kane 2015 Int.). This suggests that there are different requirements for female and male politicians. Even where the competence level is equal, men appears to be advantaged by the virtue of being men, regardless of competences or any other rationally justifiable criteria. There thus seems to be little evidence for positive developments in deputies’ belief in women's ability to govern based on the discussion of competence alone.
Gender relations in parliament
As for gender relations in the National Assembly, more positive developments seem to have taken place. Members of the post-2012 parliament reported that gender relations were good and not hostile. As stated by a female parliamentarian: ‘I think that here, we respect each other mutually, whether you're a man or a woman. We consider each other colleagues’ (Dionne 2015 Int.). This was reflected in the men's statements: ‘To be honest, I don't think about whether I am with a woman or a man. I think that I am with someone, a colleague with whom I defend the same ideal. Together we discuss different positions in a democratic manner’ (Sy 2015 Int.). All the women I interviewed in the current parliament had spoken before the assembly without experiencing negative reactions. According to the female ex-parliamentarians, however, things used to be less collegial between the men and women. One claimed that her male colleagues were not always pleased with women's presence in political activities: ‘They were obliged to work with us because they had to, but once “on the ground” relations were not the best’ (N. M. Kane 2015 Int.). Another female deputy claimed that the men used other women to annoy her and described the environment as hostile (Fatou Touré 2015 Int.). However, as more women entered parliament, the mentality of some of the male parliamentarians gradually changed towards more acceptance, as illustrated by one of them: ‘At the time, people said that a man is superior to a woman, but today when I am with my deputy colleagues, we have equal dignity. There are even women who are superior to me, because they are vice presidents in the National Assembly hierarchy. One no longer has this “power complex” when working with women as a collaborator or partner’ (Diallo 2015 Int.).
These findings signify that relations between men and women have improved since the introduction of the Law on Parity. The increased presence of women may have helped reformulate the role of the political actor, manifested in more acceptance and respect from male colleagues. These seemingly positive findings become more complicated when compared to another tendency that reportedly has taken place simultaneously: the decline in women's activism. Women in parliaments with a large male majority can either ‘blend in’ or promote themselves and their achievements. In skewed parliaments, the few women present may be particularly active, reflected in how they voluntarily enter such a competitive political arena (Bratton Reference Bratton2005). There seems to have been much activism among the women in the pre-2012 Senegalese parliament, exemplified by the successful adoption of the Law on Parity itself: ‘Just before the law was to be voted, we were a group of women who rebelled. We blocked the election until very late in the night … They [the men] saw us as troublemakers’ (N. M. Kane 2015 Int.). Some of these activist women also accused the women in today's National Assembly of ‘sleeping’ and not being good role models (Female ex-parliamentarian 2015 Int.). Another stated that ‘[today] even with 64 female deputies, they cannot do what the 22 of us did in terms of being heard and exercising leadership’ (Female ex-parliamentarian 2015 Int.).
Interestingly, this apparent decline in activism was reflected in the men's statements but presented as something positive. As stated by one, ‘the consequence [of the law on parity] that one can see is that women have calmed down. First, they fought for their rights, and now they have calmed down because at least [the law] permits social peace and cohesion’ (Male parliamentarian 2015 Int.). Another man claimed that thanks to the law, the ‘cleavage of feminism’ has diminished in parliament: ‘You don't see feminists, but representatives of the people in general. Feminism needs for women to be on an unequal footing. But today the women are sufficiently represented. They don't have to be supported by feminism or women's activists. Equality is reality’ (Male parliamentarian 2015 Int.).
These statements clearly show two different conceptions of the same phenomenon. That women have ‘calmed down’ or are ‘sleeping’ can be seen as a consequence of having their most urgent requirements satisfied, but also as a consequence of more women – perhaps with less experience from the competitiveness of the legislative arena and thus less clout – occupying parliament. Most women will probably disagree with the claim that feminism loses its purpose because women have the same access to parliament as men. If this decline in activism – and thus potentially in conflict level – is the reason behind the reported harmony in today's parliament, one cannot draw the conclusion that the presence of more women has transformed a pervasive culture of masculinity into a parliament more conducive to women.
Perspectives on representation
In the Senegalese National Assembly, gender quotas appear to have contributed to a broadening of the perspective on who should represent and who should be represented, thereby contributing positively to women's symbolic representation. Most of my participants claimed that one of the main reasons for including more women into politics was because women contributed something fundamentally different from men, which was of value to Senegalese society. Women were described as more in contact with the population and their problems than men, and as representing their sisters who remained back home in the village. A recent translation service was believed to facilitate rural female parliamentarians’ participation in parliament, allowing them to focus on representing the population better (Dionne 2015 Int.). Many parliamentarians saw the increased diversity as to who should be represented as a direct consequence of the Law on Parity: ‘No matter their background, women are now entering decision-making instances. Because of parity, the National Assembly now represents the population better’ said one current female parliamentarian (Dionne 2015 Int.). One man claimed that the parliament needs both the rural women who can mobilise and be ‘close to the population’, as well as the educated and more politically experienced women. From this perspective, one can thus argue that parity contributes to attitudinal change ‘in that they broaden the perspective of whom representatives, and politicians more broadly speaking, should represent and to whom they should be responsive’ (Meier Reference Meier, Franceschet, Krook and Piscopo2012: 161). Based on my sample, I was not able to conclude that this was a direct consequence of parity, yet many of my interviewees described this process as one of the main advantages of the new law.
Apart from one male current deputy who applauded his female colleagues for ‘talking like men’, there was a consensus that women contributed by being women, not by imitating men. Research from other parts of the world show that women are perceived to hold specific ‘communal’ qualities connected to their gender, such as the welfare of others, while men are seen as more assertive, controlling and confident (Nugent & Krook Reference Nugent and Krook2016: 132). From a western feminist point of view, this gendered partition of qualities might seem contradictory to the principles of parity, namely that gender should not determine roles in politics. However, Tripp (Reference Tripp2003: 250) argues that the women's movements in Africa use different resources to draw from when mobilising for political change. For instance, the concept of ‘motherhood’ has been used actively to legitimise women's inclusion into politics. By introducing more ‘female’ values such as nurturing, justice and caring to the political scene, these are meant to replace some of the more negative ‘masculine’ traits, like corruption. Further, women entering politics sometimes draw on their domestic experiences to create a new type of political imagery that challenges the paternal one that evolved with the colonial state and has remained in the post-colonial context. Drawing on motherhood is not considered controversial or problematic in the African tradition, which does not draw a sharp divide between the public/political and domestic sphere as is common among Western feminists. This shows the importance of considering cultural factors not only when measuring symbolic effects, but also when deciding what a symbolic effect actually is. Hence, the fact that women are still believed to hold distinct moral attributes does not mean that no advancement has been made with regards to their attitudinal change.
It is important to note that this gendered division of political issues is not entirely unproblematic. So-called ‘soft politics’ associated with women are often seen as less prestigious than issues associated with men, which typically provide better opportunities for ascension in political life. This essentialisation of politics may work against gender equality within the National Assembly. It is however possible that a bigger presence of women in parliament may gradually draw people's attention towards women's issues and increase its prestige (Miguel Reference Miguel, Franceschet, Krook and Piscopo2012). In sum, the widened perspective on who should represent and who should be represented may point to positive developments in attitudes towards women as political leaders.
Delegation of posts
In the Senegalese National Assembly, men are reportedly still prioritised to the detriment of women when choosing leaders of commissions and other posts of responsibility, even when competences are equal (M. Kane 2015 Int.). This suggests that the perception of political decision-making as a male domain lingers. The gendered division of political issues discussed above can help explain the uneven distribution of posts. Furthermore, men tend to join politics 5 to 10 years earlier than women, which is connected to unbalanced child-rearing responsibilities (Rosenbluth et al. Reference Rosenbluth, Kalla and Teele2015). One study also suggests that women are involved in politics for a shorter period before getting nominated as candidates (Magnussen & Svarstad Reference Magnussen and Svarstad2013). Seniority may therefore be part of the reason why men are more often given posts of responsibility. One should however note the political will in today's parliament to even out this unbalance, as parity has been expanded to include elective posts within the parliament as well (M. Kane & Balde 2015 Int.).
Legitimisation of gender quotas
Among Senegalese political elites, support for gender quotas was mixed. According to male and female ex-parliamentarians, the quota law was reported to correct social injustice, recognise women's legitimate rights and symbolise social progress. Among the current male parliamentarians, some criticism was made as to what the substantial consequences of the law would be (D. Wade 2015 Int.). One claimed that the law was introduced too early and too fast for the Senegalese people to accept it (Diallo 2015 Int.). It was also reported that some of the representatives from religious families in the National Assembly criticised the Law on Parity vigorously, arguing that it violated the rule of equality before the law, and promoted favouritism and unfairness.
Although most of the current male parliamentarians claimed to support the Law on Parity, some did not think gender discrimination existed in Senegal in the first place. This stands in stark contrast to established knowledge about Senegal and its patriarchal culture. It also undermines the legitimisation of the Law of Parity itself. These findings are not surprising since discrimination is more often acknowledged by those facing it than those who do not (Crosby et al. Reference Crosby, Iyer and Sincharoen2006). This lack of acknowledging discriminatory practices suggests that the gender equality norms underlying the parity law have not been internalised among the male deputies. The intense campaigning leading up to the adoption of parity can help explain why gender equality norms seemed more internalised among the male ex-parliamentarians.
In sum, the findings from these five indicators show the complexity of measuring symbolic representation. Changes are detected in some indicators, but not in others. Possible improvements in gender relations in parliament and a broader perspective on representation indicates positive symbolic effects of gender quotas. The discussion of competences, the unbalanced assignment of posts, and a lack of internalisation of the norms underlying gender quotas suggest few positive changes.
DISCUSSION
Has the Law on Parity had a strong polarising effect on political elites in Senegal, as was witnessed among the population at large? This study suggests no; there appear to be modest, yet positive developments in attitudes towards female political leaders among deputies of both genders. In this section I will discuss what might explain these findings.
First, it is possible that people react to the affirmative action itself, instead of or in addition to the increase in women's numerical representation. This study focuses on short-term effects, but attitudes change slowly. The strong reactions witnessed among Senegalese citizens might therefore be a reaction towards the Law on Parity rather than an effect of being exposed to more women in parliament. People react differently to gender quotas depending on who benefits from them. For those previously excluded, the women, parity can be seen as a fundamental right and quotas as a necessary mean to counter patriarchal political culture. Senegalese women strongly supported female political leaders after the adoption of the parity law, whereas men's support remained relatively stable. According to Alexander (Reference Alexander2012), women's attitudes towards gender relations in the political sphere are less rigid than men's. Since women are the victims of discrimination they are more likely to acknowledge its existence and thus support gender quotas. For them, gender quotas symbolise inclusion. On the other hand, privileged men who have never experienced discrimination or marginalisation, often favour the status quo (Crosby et al. Reference Crosby, Iyer and Sincharoen2006). For people who do not see the need for quotas, such measures are regarded as giving preferential treatment to women at men's expense, or as an illegitimate tool promoting ‘undeserving’ women who are not elected based on merit alone. Instead of inclusion, quotas thus symbolise exclusion. The gendered polarisation of public opinion can thus be understood in light of support for quota policies.
How parity was adopted can also explain women's attitudes. In Senegal, quota adoption was driven by a broad, locally based coalition of women's organisations focusing on raising awareness about the importance of parity for Senegalese women and for the country's development, involving grassroots women in the campaigning. Such coalitions are seen as vital for the successful adoption of gender quotas (Kang & Tripp Reference Kang and Tripp2018). The positive attitudinal change witnessed among female Senegalese citizens suggests that it also matters for women's symbolic representation. In cases where quotas are adopted in a top-down matter without including the grassroots, as in Lesotho (Clayton Reference Clayton2014), such positive effects are not a given.
However, the survey data show that also among some women, scepticism towards women as political leaders increased. Not all women consider themselves discriminated against, and among these, quota policies lack a basis of legitimacy. In addition, both men and women may react negatively towards quotas when they are seen as violating appropriate gender roles which can be deeply rooted in parts of society (Clayton Reference Clayton2014). According to social psychology research, both genders ‘punish’ women who behave counter-stereotypically by aiming at leadership positions (Nugent & Krook Reference Nugent and Krook2016: 132). This is perhaps of extra importance in deeply patriarchal societies like Senegal where religion regulates everyday life. In some parts of rural Senegal, parité has become a negative word in the local language Wolof, meaning ‘I no longer accept my husband's authority’ (Hirsch Reference Hirsch2012). In other words, while the framing of quotas as parity might have helped its implementation among the pragmatic political elites, this has not necessarily contributed to parity endorsement among the rural population. As stated by one of the bureaucrats in the National Assembly:
There is a big cleavage between the political elite vis-à-vis the population. When you ask a Senegalese person on the street … what he thinks of the quotas, he will tell you ‘No, man and woman will never be equal’. In the mentality of the population, [parity] is not entirely accepted, but this is not a problem on elite level. (Ka 2015 Int.)
This reflects the complexity of the Senegalese society, and might help us understand why the deeply polarising effect of the parity law on public opinion towards female political leaders is not found among political elites in the National Assembly. In Senegal, politics is dominated by urban, educated elites, influenced by French and Western thinking. In addition, these politicians operate in an international climate that favours ‘modern’ states focusing on women's rights. Although my findings among political elites showed limited attitudinal change, few parliamentarians – except for some belonging to religious families – displayed direct opposition toward the parity law. They acknowledged its challenges but expected that the situation would improve with time. Theory suggests that once elected to a legislative body, elite opinion often converges with ‘the rules of the political game’ such as for instance tolerance for minority groups and support for democratic values (McAllister Reference McAllister1991: 259). One can argue that the Law on Parity represents similar values as a guarantee for marginalised women's right to access decision-making positions. In addition, international pro-quota norms are likely to have influenced elite's opinions. On the other hand, there is the rest of society, where many people are more in touch with their marabout than their local representative. One can expect attitudinal changes to develop more slowly among these segments of society. These findings can also be seen in relation to class. Although research from Africa suggests that citizens’ and deputies’ views often converge, class differences may be an explanation when they don't converge. Poverty and gender inequality are interconnected, and elites may not have experienced the same level of gender inequality as the poorer segments of society (Clayton et al. Reference Clayton, Josefsson, Mattes and Mozaffar2019). It is not unlikely that this class difference is translated into a stronger polarisation of attitudes toward women as political leaders among ordinary citizens than political elites.
It is beyond the scope of this study to say for sure whether the polarisation of public opinion is a reaction towards the quota law itself, the increase of women parliamentarians, or a combination of both. This study of short-term effects captures the instant reactions to a policy. One can expect attitudes to continue developing as people become more used to seeing women in untraditional roles as political leaders. How these women perform in parliament may matter for this development. As suggested by political elites in the National Assembly, and by some scholars (Yoon Reference Yoon2011; Beaman et al. Reference Beaman, Pande, Cirone, Franceschet, Krook and Piscopo2012), there is a possible link between the three types of representation: descriptive representation may matter for symbolic representation through substantive representation. In other words, women's actions in parliament might matter more than their mere presence. Some of the newly elected women are new to legislative politics, and non-Francophone deputies have previously not been able to fully participate in legislative politics. However, a newly installed translation service shows that there is a political will present in parliament to facilitate the substantive contribution of the newly entered women (Kane 2015 Int.). Exactly what such a contribution from female deputies entails, is a different question, but several of my interviewees highlighted the importance of women deputies representing other women through their actions.
CONCLUSION
After the implementation of the 2012 Law on Parity in Senegal, we saw a strong and gendered polarisation of public opinion towards women versus men as political leaders. Generally, men became more convinced than before that other men made better political leaders, while most women had their belief in other women's ability to govern strengthened. This study has used this as a point of departure for exploring whether similar attitudinal changes have taken place within the Senegalese National Assembly. My findings show that some moderate changes have occurred, but the strong polarisation of opinions found among the public is not reflected within the parliament. Reports of better gender relations between male and female parliamentarians and a broader perspective on who should be represented indicate that positive attitudinal change has taken place. However, assumptions of women's competence, unbalanced delegation of posts and a lack of internalisation of gender equality norms do not suggest such attitudinal change.
I suggest that the strong reaction among the population has as much to do with attitudes towards the affirmative action itself as towards the increased presence of women deputies, which might again be a short-term symptom. How people react to quota policies depends on whether they see it as a legitimate tool to counter discriminatory practices or not. There seems to be a need for COSEF and other parity supporters to continue raising awareness and promoting parity among the Senegalese population; even though the term parity has been embraced by the political elite, it has not yet been successfully transmitted to the whole population. It remains for future research to follow the evolution of this process.
One of the main contributions of this study is the creation of a framework for measuring symbolic effects of gender quotas in parliament. Parts of this framework can also be used for measuring women's symbolic representation in other institutions such as businesses, courts, executive bodies or local legislatures. The fact that we see effects in some aspects of women's symbolic representation but not in others, points to the complicated endeavour of measuring such an intangible concept. Future research may build upon and refine this framework and apply it to different audiences and contexts.
Appendix
Table AI. Women vs. men as political leaders in Niger.
Table AII. Women vs. men as political leaders in sub-Saharan Africa.
Table AIII. Women vs. men as political leaders in West Africa.
