Why civilians participate in genocide.

FROM VIOLENCE TO MOBILIZATION: WOMEN, WAR, AND THREAT IN RWANDA*

Marie E. Berry†

Theories of social movement emergence posit “threat” as an important concept in ex- planations of mobilization. This article uses the case of the 1994 Rwandan genocide to investigate whether threats that stem from mass violence can also have a mobilizing effect. Drawing from interviews with 152 women in Rwanda, I reveal how threatening conditions created by the genocide and civil war initiated a grassroots mobilization process among women. This mobilization featured women founding and joining community organizations, engaging in new forms of claims making toward state institutions, and eventually running for political office. Two mechanisms facilitated this process: the social appropriation of feminine values for the reconceptualization of women as legitimate political actors, and the brokerage of connections between individual women, organizations, and government institutions by foreign actors. I conclude by suggesting that this mobilization served as a necessary, but not sufficient, condition for the meteoric rise of women in Rwanda’s politics.

Social movement scholarship has long investigated the interplay between opportunity and threat in connection with mobilization. While most of the literature looks at how political opportunities facilitate collective action (McAdam 1982; Meyer and Minkoff 2004; Tarrow 1998; Tilly 1978), we are reminded that threats can drive collective action as well (Alimi 2009; Almeida 2003; Goldstone and Tilly 2001; Khawaja 1993; Loveman 1998; Tilly 1978; Van Dyke and Soule 2002). Threats are generally understood as emanating from a repressive state, as a government or its agents target certain groups via mass arrests, massacres, or other means of suppression. However, several studies suggest that threats can also originate from sources beyond the political arena, such as during periods of rapid economic changes (McVeigh 1999; Van Dyke and Soule 2002). Extending this line of research, in this article I investigate whether mass violence may also generate threats that have mobilizing effects.

To do so, I examine a puzzle: In 1994, Rwanda experienced one of the most horrific genocides in history, which left between 500,000 and one million people dead, an estimated 250,000 women raped, and over 100,000 children orphaned (Des Forges 1999; Prunier 1995).1 The genocide and corresponding civil war destroyed Rwanda’s infrastructure, displaced up to one-half of the population, and damaged the agricultural foundations of the economy (Prunier 1995). Yet shortly after the violence ended, Rwandan women took on new roles in their households, joined nascent community organizations, and soon began to emerge as leaders in their communities. Ten years later, thousands of women’s organizations had emerged as vital and robust social institutions, and women were elected to the world’s highest percentage of seats in Parliament. These surprising developments complicate the expectations of social _______________________________

* I would like to thank the three anonymous reviewers and Rory McVeigh for their detailed and thoughtful comments on earlier drafts of this article, as well as Andreas Wimmer, Ed Walker, Michael Mann, Abigail Saguy, Doug McAdam, Sid Tarrow, Laura Mann, Susan Watkins, Hollie Nyseth Brehm, Pamela Prickett, Zeynep Ozgen, Matt Baltz, Molly Jacobs, and Josh Sadlier for their advice and guidance on this article and broader project as it developed. This research was supported by grants from the UCLA International Institute and the UCLA Center for the Study of Women. † Marie E. Berry is a Ph.D. candidate in the Department of Sociology at the University of California, Los Angeles. Please direct all correspondence to the author at [email protected].

© 2015 Mobilization: An International Quarterly 20(2): 135-156 DOI 10.171813.1086-671X-20-2-135

Mobilization 136

movement scholars who have long argued that social capital, networks, and resources are critical preconditions for mobilization. Instead, this case suggests that although mass violence is destructive, it can also upset traditionally male-dominated social spaces, transform social relationships, and perhaps even open opportunities for women to participate in new social and political roles. In this article I thus ask: How does mass violence precipitate the mobilization of women? Under what conditions and through which social mechanisms can it do so?

Drawing on interviews with 152 women at all levels of Rwandan society, I use a social movement framework to illustrate how the violence in Rwanda triggered a grassroots mobil- ization process among women. This case is of theoretical and empirical importance to the study of social movements for three primary reasons. First, as stated above, this case allows us to extend the dominant understanding of threat beyond state repression. To my knowledge, it is the first attempt to explicitly investigate the mobilizing effects of threats caused by mass violence. As such, this article aims to deepen our understanding of the process by which vio- lence can lead to mobilization.

Second, looking at threats that originate outside of the political arena allows me to in- vestigate how such threats might be gendered. This advances previous scholarship on threat that has largely ignored how men and women experience threats differently. Moreover, illustrating the agency women exhibit when responding to such threats complicates normative assumptions of women’s victimhood and suffering after mass violence. This article thus con- tributes to feminist literature that challenges such simplistic, victim-centered interpretations of women’s experience during and after violence.

Finally, this article answers the call for social movement research that looks beyond highly formalized social movements in Western democratic settings. Unlike more contentious forms of mobilization frequently referenced in the literature, here I understand mobilization as “the process by which a group goes from being a passive collection of individuals to an active participant in public life” (Tilly 1978: 69). Investigating this kind of mobilization is essential for understanding network formation and civic activism more generally, which scholars argue is a precondition of social movement emergence. Given the lack of networks and resources in a post-genocide context, it seems unfathomable that mobilization would occur in such a setting. Yet as I describe below, in Rwanda this mobilization took the form of women founding and joining grassroots self-help organizations, engaging in new forms of dialogue with established political institutions, and eventually running for political office. This more inclusive under- standing of mobilization aligns with feminist scholarship on gender and social movements, which shows how women’s collective action often happens outside of the formal political realm and in more fluid and diffuse forms of organization (Ferree 1992; Taylor 1999; L. Tilly 1981). Indeed, women’s mobilization may not draw headlines, but it can dramatically change the political dynamics of society (Tripp 2000: 17).

I find that two key “threatening” conditions created by the violence are at the origin of women’s mobilization in Rwanda. These include (1) the demographic imbalance between men and women, and (2) the urgent need for basic goods like food, water, medical care, and shelter. Combined, these two conditions were immediate and potentially lethal for many Rwandans—and particularly for women. After describing how these threatening conditions initiated the process of women’s mobilization, I identify two mechanisms that worked to facilitate this mobilization process: (a) the social appropriation of feminine values for the re- conceptualization of women as legitimate political actors, and (b) the brokerage of con- nections between individual women, grassroots organizations, and government institutions by foreign actors. Identifying these mechanisms allows me to illustrate the various components of the mobilization process (figure 1). My findings suggest that scholars of social movements should pay more attention to the unexpected mobilizing effects of large-scale atrocities, as they may actually open spaces for the increased participation of certain groups in public life. In what follows, I explain how mass violence can be a liminal historical moment that has both transformative and destructive potential.

From Violence to Mobilization

137

Figure 1. Relationship Between Mass Violence and Women's Mobilization

MOBILIZATION:

Mass$Violence$

Demographic, vacuum,of,

men,

Urgent, material,needs,

Social, appropria7on,, Brokerage,,

THREAT:

Women,take,on, new,roles,

outside,their, homes,

Women,begin,to, form,small,

informal,self= help,groups,

Women,engage, in,new,forms,of, dialogue,with,

gov.,ins7tusions,

Women,take,on, leadership,

posi7ons/run, for,poli7cal,

office,

MOBILIZING EFFECTS OF THREAT

In the social movement literature, threat denotes the likelihood that harms will be inflicted on a group if members of the group fail to act collectively (Alimi 2009; Almeida 2003; Goldstone 2002; Khawaja 1993; Loveman 1998; Tilly 1978; Van Dyke and Soule 2002). In most previous studies, threats are understood as stemming from a repressive state. For ex- ample, Kurzman (1996: 161) describes how the Iranian state’s brutal treatment of civilians during the 1970s helped catalyze the opposition movement to the Shah, while Loveman (1998) details how human rights organizations in Chile, Uruguay, and Argentina formed in the face of life-threatening state repression. In some of the definitive studies of revolutions, state repression is often depicted as a threat that can inspire revolutionary movements to form (see Goldstone and Tilly 2001; Goodwin 2001; Zhao 2001). Almeida (2003) breaks the concept of threat into three more specific dimensions (e.g., the erosion of rights), which can all be traced back to the state. Each of these studies finds that repressive conditions can lead to increased protest mobilization and action.

Some scholarship extends the analysis of threat beyond the traditional focus on state repression. Drawing from elements of strain theory, scholars like Tarrow (1996) and Moore (1966) have shown how European peasants staged uprisings in response to economic restructuring caused by changes in the commercialization of agriculture. Van Dyke and Soule (2002) suggest that economic, political, or demographic shifts may produce threats that in- spire reactive mobilization among certain types of militia groups. Macro-level phenomena, like rising unemployment and demographic shifts, have also been shown to lead to the mobil- ization of right-wing groups (Beck 2000; McVeigh 1999).

Extending the idea of threat even further, natural disasters and other major calamities may produce the threat of death, injury, disease, or exposure, and thus may also catalyze col- lective action. For instance, Walsh (1981) describes the rapid expansion of citizen protest after the nuclear disaster on Three Mile Island, and Kaplan (1997) chronicles how home- owners in the Love Canal neighborhood of Niagara Falls organized to protest the toxic waste dump causing lethal health problems in the community. Einwohner and Maher’s (2011) study of resistance in ghettos and camps during the Holocaust argues that threats can come from the government directly (e.g., the threat of being killed by an SS guard) or indirectly (e.g., the threat of dying from starvation or exposure). This latter point is key, as it suggests that poten- tially lethal environmental conditions—such as those that exist after mass violence—may also motivate collective action.

Mobilization 138

The idea that environmental conditions can constitute a threat extends research on social movements in non-Western contexts, where collective action is often motivated by a “crisis of consumption”—such as the lack of food, health care, or other basic resources (Molyneux 1985; Noonan 1995; Safa 1990). Critically, where mobilization is not in response to state re- pression but is rather a response to environmental conditions or a lack of access to basic resources, it is likely to be highly gendered. For example, suffocating air pollution in Czecho- slovakia led women to mobilize against the authoritarian regime to defend their children’s health (Shriver, Adams, and Einwohner 2014). Women’s traditional caregiving roles in countries across the world mean that they often are the most responsible for providing basic supplies to their families, and thus may be first to mobilize when their ability to provide for their families is threatened. Many studies have illustrated women’s mobilization around “practical gender interests” (Molyneux 1985: 233) like food or fuel scarcity, and have looked particularly at the role of “ordinary” women in these move-ments (see Jacquette 1994; Martin 2011; Molyneux 1998; Safa 1990). These studies note that women’s organizing may consist of public protests, but may also involve less contentious forms of mobilization such as expressive rituals of reconciliation (Taylor and Rupp 2002) or the organization of women’s self-help organizations (Safa 1990). While these studies do not conceptualize women’s organizing as a response to threat per say, they reinforce the idea that mobilization can be spurred by threatening conditions that stem from beyond the political sphere.

Analyzing threats from beyond the state echoes work by feminist scholars that calls for a broadening of the focus in many social science fields from the state and political institutions to one that encompasses other institutional arenas (Blee 1991; Ferree 1992; Nagel 1998; Taylor 1999; L. Tilly 1981). These scholars join others in criticizing the overt focus in the social movement literature on highly formalized social movement organizations, which ob- scures more diffuse and less political spaces of collective action. When cut out of politics or denied the ability to directly lobby political leaders, women often organize in the civil sector (Bop 2001; Gal and Kligman 2000; Tamale 1999). Thus, research on threat—like research on social movements more generally—has much to gain from investigating the emergence of threats from outside the political arena, particularly as such threats may be gendered.

Building on these ideas—that immediate and potentially lethal threats can come from beyond the state, and that those that do may be particularly salient for women—I investigate whether the threat of starvation, disease or destitution following a period of mass violence is sufficient to precipitate collective action. I also investigate whether this threat—and the mobilization it may inspire—may be gendered, and show in detail how this process of mobilization unfolded in Rwanda.

HISTORICAL BACKGROUND

Rwanda has been a patriarchal society since before the colonial era. Although valued as ad- visors and as mothers, women were considered subordinate to men; they were confined to social realms within the home, with the children, and on the land, as the gendered division of labor structured daily life (Codere 1973; Des Forges 1999; Jefremovas 1991; Newbury 1988). Little had changed by 1994, when the genocide began. A few women had entered national politics—including Agathe Uwilingiyimana, the first woman prime minister—but most were able to do so only because of well-connected husbands or fathers, and wielded little practical power once in these roles. No women held political offices at the provincial or district levels (Nowrojee 1996: 2; Women for Women 2004). According to one member of Parliament, “Politics . . . was a male domain. It was like their personal bedroom where no one else could go in” (interview, Deputy A, July 17, 2009). Women were also deprived of basic economic rights, including the right to inherit property, belong to a profit-making organization, and, until 1992, to open a personal bank account (Polavarapu 2011). Community organizations

From Violence to Mobilization 139

were very rare in Rwanda prior to the early 1990s, and the few that existed were generally appendages of the ruling political party with membership confined to a small group of women with close ties to the regime (Women for Women 2004).

Before the genocide and civil war, the population of Rwanda was divided into three ethnic groups (Hutu at 84 percent, Tutsi at 14 percent, and Twa at 1 percent), and political conflict between these groups has characterized Rwandan politics for the last century.2 During the German and Belgian colonial era, Rwandan politics was primarily defined by an ethnic- class system. Tutsis were favored by the colonial powers as they controlled the monarchy and comprised the cattle-raising class. Hutus, primarily sedentary farmers, were politically and economically disadvantaged (Codere 1973; Newbury 1988; Pottier 2002). The 1959 Hutu revolution secured the country’s independence from Belgium and installed a Hutu regime for the first time. It also unleashed a period of violence that sent tens of thousands of Tutsi and political dissidents into exile in neighboring countries. In the decades that followed, the Kayibanda and subsequent Habyarimana regimes favored Hutus and imposed quotas on the number of Tutsi who could enroll in school or work in the public sector (Prunier 1995). In 1990, amidst a climate of regional political and economic instability, a group of mostly Tutsi exiles launched a military attack against the Hutu regime in Kigali from Uganda. Seen as an attempt to bring the state back under Tutsi domination, the Rwandan Patriotic Front (RPF) invasion posed a threat to the existing government, and a low-level civil war began in the north of the country.

The story of the Rwandan genocide itself has been thoroughly explored by social scientists (see Des Forges 1999; Fujii 2009; Mamdani 2001; Prunier 1995; Straus 2006). On April 6, 1994, while peace negotiations between the RPF and the Habyarimana regime were ongoing, the plane carrying Rwandan President Juvenal Habyarimana and the president of Burundi was shot down as it approached the airport in Kigali. Within hours the genocide began, as hard-liners within the government and members of Habyarimana’s inner circle blamed the RPF for the death of the president and amplified their claim that all Tutsis were “the enemy.” Over the next several months, Tutsis and moderate Hutus were hacked, beaten, and shot to death by Hutu extremists and government-affiliated militias, as well as by many ordinary Rwandans. Throughout the bloodshed, the RPF continued its battle for control of the state, meaning that the genocide effectively occurred within the context of a civil war (Des Forges 1999; Straus 2006). All told, between 500,000 and one million people were killed in the violence. By the time the RPF secured control of Kigali on July 4, Rwanda was in ruins.

DATA AND METHODS

The analysis in this article relies primarily on interviews I conducted with 152 Rwandan women during six months of fieldwork between 2009 and 2013. The analysis is also informed by dozens of informal interviews I conducted during two prior summers spent in Rwanda in 2007 and 2008. Alongside the interviews are myriad corroborating sources from government documents, organizational reports, secondary literature, and archival collections of survivor testimonies, which allow me to situate the experiences of individual women within the broader historical context.

Interviewees were selected using a stratified purposive sampling design, which aimed to select women respondents from three categories: (1) high-level government officials and executives in internationally linked NGOs or government-affiliated institutions (“elite” women); (2) members/employees of less formal domestic organizations; and (3) poor urban and rural women (“ordinary” women).3 This sampling strategy was designed in order to see how the mass violence affected women’s lives from across the class, ethnic, and regional spectrum. In order to capture variation in income, urbanization, and experience of violence during the genocide within these categories, respondents were selected from three focus

Mobilization

140

regions: Kigali City (the capital), Musanze District in Northern Province, and Bugasera District in Eastern Province.

I conducted forty semistructured biographical interviews with respondents from the first category of political elites, with the goal of mapping patterns in the ways these women ascended to political office. I conducted twenty-five of the forty interviews with current mem- bers of Rwanda’s Parliament. After securing permission from Rwandan authorities to conduct research, I requested interviews from all high-ranking women in government (e.g., members of Parliament) and interviewed those who agreed to meet. Within this group of elite respon- dents, 35 percent (14) of the women were widows or survivors of genocide (likely Tutsi), 38 percent (15) were returning refugees (likely Tutsi), and 28 percent (11) made no mention of their victimization during the genocide (likely Hutu). Given that ethnicity is illegal to discuss in Rwanda, I was unable to ask questions about it directly and therefore relied on each respon- dent’s life story to situate them within these social/ethnic categories.

I also conducted individual and small-group interviews with an additional 112 women who fell into either the second or third category, in order to compare the impact of the violence on women in lower social classes and of different backgrounds.4 These women were selected by local informants in the focus regions in order to capture respondents from a range of different careers, income levels, ages, and ethnicities. The majority of these women were not mobilized as political actors, and thus provide an important comparison group for the political elites in the first category. Ultimately, the analysis in this article draws most heavily on the interviews conducted with women in the first category, as their personal stories of mobilization reveal patterns in the way mass violence led to women’s increased participation in politics and civic life.

All respondents were interviewed once at a mutually agreed upon location, and were given the option of conducting their interview in one of the three official languages of Rwanda: Kinyarwanda, English, or French. Most chose Kinyarwanda, and while I speak basic Kinyarwanda, most interviews were conducted with the help of a Rwandan translator. Inter- views ranged from approximately thirty minutes to more than three hours, and were almost always audio recorded. Upon their completion, interviews from all groups were transcribed and coded for patterns in Dedoose, a qualitative coding software. Codes for this article were broadly organized into six subject themes: (1) shifts in career or household roles before/after genocide; (2) childhood career aspirations; (3) experience during genocide (e.g., survivor, returning refugee); (4) involvement with grassroots organizations; (5) involvement with politics; and (6) qualities of women. These interviews serve as a rich source of data, as they allow me to trace the way the violence impacted broader social mobilization through patterns in individual women’s stories.

The interviews revealed patterns in the new activities women engaged in after the violence. These activities generally fell into three general categories—working together in groups; redefining their presence in the public arena; and interacting with government institutions and NGOs for the first time. Upon a closer examination of my data, it became clear that the latter two were mechanisms facilitating the broader process of mobilization, which is captured by the former. By identifying these mechanisms and tracing them throughout the mobilization process, I follow the sociological tradition of using mechanisms to explain the relationships between actors, institutions, and the political context (Elster 1989; Hedström and Swedberg 1998; McAdam, Tarrow, and Tilly 2001). Doing so allows me to illustrate how the violence in Rwanda facilitated the mobilization of women, as it enables me to break up the processes of mobilization into specific components that can be identified and independently studied (McAdam et. al. 2001).

The interview excerpts selected for this article represent themes that were consistent in multiple interviews. To preserve the anonymity of my interview subjects who still hold political offices, I omit their names and instead assign an identifying letter.

From Violence to Mobilization

141

THREAT: CONDITIONS FOR MOBILIZATION Threat: Demographic Imbalance

Even if your husband is dead, even if your husband isn’t there, you can’t stop living. You have to keep living, you have to take his responsibilities and you have to keep moving (interview, Deputy D, July 15, 2009). By the time the Rwandan Patriotic Front halted the genocide, approximately one out of

every ten Rwandans was dead and two million had fled over the border to neighboring countries.5 In the following months, approximately 120,000 Rwandans—mostly men—were rounded up and imprisoned (Des Forges 1999; Prunier 1995). An estimated two million of the eight million people in Rwanda were displaced internally, squatting in existing households that were often left empty after the slaughter or displacement of their owners, or in makeshift Internally Displaced Person (IDP) camps (Newbury and Baldwin 2000; Prunier 1995). Fifty- six percent of the dead were men (Republic of Rwanda 2004). Since Tutsis were targeted during the genocide, they comprised a majority of the dead. However, a sizable minority were Hutus who were either killed because of their “moderate” political position or affiliation with Tutsis, or because of pervasive uncertainty during the ongoing civil war with the RPF (Des Forges 1999; Fujii 2009). The disproportionate participation of men in the ongoing civil war is a likely cause of the gender disparity in the death toll. In addition, “mercy” was sometimes shown to women—often at the price of rape rather than death (Nowrojee 1996). As a result of this unequal death, displacement, and imprisonment of men, immediately after the violence an estimated 70 percent of Rwanda’s nonincarcerated population was female (Powley 2004; Republic of Rwanda 2007). Across the country, women headed approximately 34 percent of all households—up from 21 percent prior to the violence (Newbury and Baldwin 2000; Republic of Rwanda 1999;). These numbers reflect a demographic imbalance that was the first threatening condition caused by the violence that precipitated the mobilization of Rwandan women.6 Here I understand this demographic imbalance as a “threat” to women’s lives—rather than as an opportunity—because widespread poverty in Rwanda in 1994 positioned women as more vulnerable to hunger, physical insecurity, and economic destitu- tion in the absence of such a substantial percentage of the male population.

Many women’s roles shifted as a result of this demographic imbalance at the household and community level. According to a mayor of a large urban district, there were three dif- ferent groups of women who were the most vulnerable:

There were women who were widows of the genocide who lost their husbands to the genocide. There were widows—or not necessarily widows—but women who found themselves alone because their husbands had participated in the genocide and found themselves in jail. Then there were women who lost their husbands in the liberation struggle [the RPF invasion]…. All of these women, they faced different social circumstances, but the point is that they found themselves in the same situation of having to fulfill responsibilities that they never thought that they would have. (interview, June 26, 2009)

As this quote suggests, at the household level, women who lost their husbands and families to mass violence, exile, or prison found themselves shouldering massive new responsibilities. Tutsi survivors of the genocide were particularly vulnerable as their homes were often looted or destroyed during the violence, leaving them without basic shelter or belongings. One mem- ber of Parliament described,

I was married to a successful man, [and thought] he will give me anything, do everything for me. But after the genocide it was different because my husband was dead. And I was the head of the home. So I had to do everything the same for the kids as when their dad was around. (interview, Deputy D, July 15, 2009)

Mobilization

142

Regardless of ethnicity or experience during the genocide, if their husbands were absent, women now had to replant land that had turned fallow, make bricks to rebuild their houses, and sell anything they could to generate an income—all activities that had only been done by men before the genocide (Newbury and Baldwin 2000; Uwineza and Pearson 2009). As another member of Parliament put it,

In my region . . . there were some activities that were done by men, and others done by women. Like cutting down a bush; women were not supposed to do that. Or hard stone farming; the men had to go do that. But for a widow, since she had no husband, she would have to go to that for herself. (interview, Deputy K, July 19, 2012)

If women had not taken on these new roles, many would have risked their lives and well- being, as well as the lives and wellbeing of children in their care.

The scarcity of men was also apparent at the community level, as businesses, organi- zations, and local government institutions suddenly lacked staff, suppliers, and patrons. Patterns in my interviews revealed how pervasive the lack of men was after the genocide. According to one parliamentarian, “At the end of the day you would have entire umudugudus [small villages] that had only women in [them].” This parliamentarian described this phenom- enon as a catalyst for changes in women’s roles, since from those villages “inevitably had to come out leaders, had to come out some sort of organization, had to come out some sort of way of living together” (interview, July 6, 2009).7 The absence of men compounded the devastation caused by the genocide and civil war as it further unraveled the social structure of society and left many women vulnerable to physical insecurity and hunger. As such, it rep- resents the first threatening condition created by the violence that facilitated women’s mobilization.

Threat: Urgent Material Needs

A second threatening condition is related, as widows, survivors, and other vulnerable

Rwandans faced an urgent threat of hunger or exposure during and after the violence. Before the genocide, Rwanda was one of the poorest countries in the world in 1994, with an average per capita GDP of $330 (World Bank Data [1993] 2012). During the violence, its agriculture- based economy was devastated and its GDP per capita decreased by nearly half, as crops were spoiled, land went uncultivated, and infrastructure collapsed. In 1995 and 1996 the govern- ment estimated that 10 percent of households were at extreme risk of severe malnutrition or starvation, while the rest of the population found it increasingly challenging to find enough food to feed themselves and their families (Newbury and Baldwin 2000). One study estimated that, in the aggregate, the daily available calories were only a little more than one-third of what was needed to meet the minimum caloric requirements of the population (UNICEF 2007, cited in Newbury and Baldwin 2000: 12). As many as 250,000 women were victims of rape during the genocide, of which perhaps 20,000 to 50,000 survived the violence (Cohen, d’Adesky and Anastos 2005; Nowrojee 1996). Many of these survivors were infected with HIV/AIDS and other sexually transmitted diseases, and needed urgent medical treatment (Newbury and Baldwin 2000). Moreover, without any state-sponsored system of care, tens of thousands of orphans of the genocide and war became the wards of anyone who would take them in (Prunier 1995). Most orphans were informally adopted by households led exclusively by women, which averaged between six to seven young dependents each (Republic of Rwanda 1999). One member of Parliament, who was widowed during the genocide, described how women in her situation “found themselves alone after the genocide. It wasn’t a choice, it was an obligation. Either you do it, or you die. Either you provide for yourself, your children, or others, or you die” (interview, Deputy C, July 8, 2009).

Land and shelter insecurity compounded the crisis. Widespread arson and looting destroyed an estimated 150,000 Tutsi homes during the violence (Prunier 1995). Those without homes

From Violence to Mobilization

143

gathered in IDP camps or resettlement areas where their food situation remained uncertain, and social networks or other community support systems that might have existed before the genocide were gone. The arrival of hundreds of thousands of returning “old caseload”8 refu- gees served to further destabilize the housing situation and drew clear distinctions between different categories of women based on their ethnicity, class, and experience during the violence.

Most critically, under both customary law and the existing legal code, all Rwandan women were ineligible to inherit land from their deceased husbands or fathers. If a widow or de facto widow had no male children, her rights to her land were particularly tenuous and largely dependent on the benevolence of her husband’s family (Ali, Deininger, and Goldstein 2011; Newbury and Baldwin 2000). Widows were repeatedly denied access to the land they lived on or farmed prior to the genocide (UNICEF 1997, cited in Newbury and Baldwin 2000). Since over one-third of all households had no adult male present, the land insecurity crisis in Rwanda was enormous.

The absence of men and urgent survival needs combined to present a threat to many Rwandan women that was both immediate and potentially lethal, rendering it a “total threat” that required collective action to resist (Maher 2010: 253). Women recognized that the costs of inaction (i.e., death or suffering) were higher than the costs of action (i.e., overcoming traditional gender roles and trauma to forge a new path forward). This threat pushed many women to join together with neighbors, friends, and strangers to secure food and shelter and retain control of their land—in other words, it compelled women to mobilize.

MOBILIZATION BEGINS

Soon, small informal self-help organizations began to emerge around the country to meet the challenges of finding food, cultivating fields, rebuilding homes, or locating other basic supplies (interviews with NGO directors, members; Gervais 2003: 544). In what follows, I refer to these informal organizations as grassroots organizations, distinguishing them from NGOs, which I understand as more formalized and often linked to international sources of funding. The earliest of these grassroots organizations were generally initiated by genocide survivors (predominantly Tutsi women), who were among the most economically, emotion- ally, and personally vulnerable (Cohen et. al. 2005; Zraly 2008). These women came together for mutual emotional support. Agnes, a member of a widow’s association in Eastern Province described how she first met other widows in her area and came up with the idea to form a regular group:

At first we usually met in the places we got food from. That is how we would see who survived, who was there. We would get together, talk to each other . . . and then people would start crying and we could try to console each other. . . . And it became a regular thing. (inter- view, July 24, 2012) Agnes’s story was similar to the stories of dozens of my other respondents from across all

interview categories, who described how groups of five to twenty women who were neighbors, friends, or simply new acquaintances decided on places to meet and invited others to come along. These grassroots organizations were “entirely Rwandan driven” (interview, mayor, July 26, 2009; Burnet 2008: 375), and most, if not all, were feminine, not feminist in nature; in other words, they did not explicitly aim to challenge traditional patriarchal society and instead often essentialized women’s roles as “mothers” or as “peaceful” (Kaplan 1982; Stephen 1997). These “self-help” groups often cultivated a form of “emotion culture” that prioritized open displays of grief and empathy, confining them to an essentially feminine logic (Taylor 1999: 20; Taylor and Rupp 2002). The rapid formation of these organizations

Mobilization 144

reflected newly forming social networks, as women joined with other women for the first time in public spaces and, in the process, upset gender roles that had long existed in the country.

Thousands of women were de facto widows as their husbands were in jail or in exile, and these women faced similar challenges of finding food, cultivating their fields, and caring for children who were now their dependents. Of course, each woman’s status in the community was often determined by her experience during the genocide. “Survivors” (i.e., Tutsis) had the support of the new RPF regime wherever it was present, while Hutus faced the burden of having to prove that they did not participate in the atrocities. This meant that in many cases it was easiest for groups of Tutsi widows to join together in an organization, as they were con- sidered “pure” victims and were thus supported by local government and foreign relief organizations.

In most cases, the founders of these grassroots organizations were individuals with some sort of training, higher education, or resources (Zraly, Rubin-Smith, and Betancourt 2011). For example, one group of relatively well-educated widows in Kigali formed an organization called Avega-Agahozo.9 The former director of Avega described how women initially came together to cry. Their early meetings were informal, without structure, leadership, or an agenda. She described:

The first thing that we wanted to take care of and “repair” was ourselves, because everything had been destroyed. We had lost our families. We ourselves were torn apart, so we needed to take care of each other, so that was the first priority, to be able to take care of each other. The first need was emergency relief; then it was taking care of people’s rights, especially the rights of women with inheritance. (interview, director of Avega, July 13, 2009)

Over the next few years, as this director suggested, the mission expanded. The founding widows decided to help other women and children in their community obtain basic neces- sities, including access to health care, trauma counseling, and eventually legal resources to secure their property. As the group secured funding from donor organizations, it grew rapidly. By 1999, over 1,500 women had turned to Avega for emotional and social support, and today Avega is one of the largest organizations for genocide widows and survivors in the country.

Within five years of the genocide, women’s collective efforts to obtain legal rights to land and basic social support created a network of organizations that interacted directly with the government. Soon it became easier to get aid if you were a member of a grassroots organ- ization. As such, Rwandans were motivated to form their own organizations that adhered to a basic organizational model: each organization needed leadership, rules and regulations, and a system to communicate with its members. This led to a process of institutional isomorphism, as these nascent organizations reproduced the formal structure of the organizations from which they sought funding—including international NGOs (DiMaggio and Powell 1983). The women (and men) who came to lead these organizations became well known in their com- munities; one study found that 50 percent of the members of two large widows associations held leadership positions in their communities (Zraly and Nyirazinyoye 2010). Women’s participation in these organizations thus disrupted traditional understandings of what the gen- dering of public spaces should look like.

Mechanisms of Mobilization: Social Appropriation

As women increasingly participated in grassroots organizations and in public spaces, they were gradually certified as legitimate actors within the political sphere. This reflected a profound shift in cultural attitudes from before the genocide, when politics and public leadership roles were almost exclusively a male domain. The shift in the cultural acceptance of women as political actors was possible because of the social appropriation of culturally meaningful values by women and the new government of Rwanda in the post-violence

From Violence to Mobilization

145

political climate. Women were heralded by the RPF leadership—and, more critically, heralded themselves—as peacemakers, as mothers, and as the “heart of the home.” Across the board, elite and nonelite women interviewed for this study agreed that it was important to have women in politics because they were more peaceful than men. One woman senator put it simply, “Women are peace-actors; they are the ones who carry out peace” (interview, July 17, 2009).

These types of “affirmative essentializations” can further entrench women in the dom- estic sphere (Helms 2002). In a post-violence context, however, they also celebrated a new type of political actor that was understood as less prone to violence than men. Indeed, the very identification of women as potential political actors reflected what Kaplan (1982) referred to as an evolving “female consciousness.” This consciousness embraced the gendered hier- archies in society, in which women were charged with bearing children and preserving life, in the process of stimulating women’s increasing participation in local and national politics. Groups of women such as Las Madres de la Plaza de Mayo in Argentina, the Mothers’ Fronts in Sri Lanka, and the Mothers of Srebrenica in Bosnia-Herzegovina have used similar argu- ments to justify their engagement outside of the home. By defining themselves first as women and as mothers, Rwandan women coalesced around a shared vision of their role in society. Emerging women leaders crafted a narrative about women’s strength during the genocide, and emphasized that all women—regardless of ethnicity—shared the experience of suffering and victimhood. Such narratives held a profound moral weight, strategically justifying women’s engagement in the predominantly male political realm. The prominence of this narrative was apparent in the responses of nearly all of the female politicians to my questions about why it was important to include women in politics in Rwanda. It was also reflected in government documents. Women were described this way in one of those documents:

[Women are] bearers of life [who] can offer a special perspective and experience which will help to overcome prevailing life-destroying methods of dealing with human problems and conflicts. Since military conflicts and diplomacy, which have traditionally been exclusively orchestrated by men, have failed to be a reliable system to safeguard peace, the inclusion of women in all stages of the peace process becomes imperative. (Republic of Rwanda, 2005) As this excerpt makes clear, this emerging female consciousness and resulting partici-

pation of women in public life was largely uncontested for one reason: men had over- whelmingly perpetrated the genocide. While some studies have highlighted exceptions to this notion (African Rights 1995; Sharlach 1999), in general very few women participated in killing, and many were actively involved in ending the atrocities. Some women did provide passive support and material assistance to the perpetrators of the genocide. But as a whole, women participated in the killing in much smaller numbers than men did, perhaps most clearly evidenced by the fact that less than 9 percent of those tried for any sort of genocide crimes have been women (Nyseth Brehm, Uggen, and Gasanabo forthcoming). Verwimp’s (2005) study suggests that the majority of the crimes women committed were theft related, as merely 1.4 percent of women in his sample were involved with killing.

Instead, as the violence came to an end, some women were recognized as being res- ponsible for the recovery efforts. They buried the dead, found homes for orphans, and began to rebuild their homes (Powley 2003). The new RPF regime and the international community often viewed men—particularly Hutu men—as collectively complicit in the genocide, whereas women—even Hutu women—were more easily regarded as victims. As such, they were more easily conceived of as being leaders palatable to Tutsi and Hutu alike. Deputy E noted,

The shift [in women’s roles] happened after the genocide, because one, women were the ones that were bearing most of the burden of the genocide. Everybody suffered but women suffered the most…. And they all got together and said, you know, now there are new doors open to us. [Men] are now recognizing that we are capable. We need to take advantage of this, we need to

Mobilization

146

be determined, and we need to have the will and the strength to make this happen for us. (interview, July 10, 2009).

This quote reflects a pattern that emerged in my data, which makes clear that the approp- riation of cultural values was at times explicitly tactical: emerging women leaders recognized the benefits of juxtaposing an essentialized idea of women as peaceful with the idea that men were responsible for the genocide in the first place.

Countless anecdotes emerged in interviews and government documents in which women were called upon to settle disputes, discipline children, or serve as judges on gacaca courts,10 simply because of their status as “peacemakers” or as “mothers” in the community. In one popular story, women threatened to turn in their husbands and sons to the authorities unless they stopped waging periodic attacks across the boarder from Congo in an effort to “finish the job” of exterminating Tutsis (Newbury 2011: 230; Prunier 2009: 73). According to Deputy C,

[Women in Ruhengeri said] “Please come out of the bush. If this is not the case, we shall never shelter you anymore. When you come out we shall put you out and never give you food.” And women did it that way—and those women started reporting even their own sons. . . . They’d say, “This one is a rebel!” Soon, many fighters deserted the militias and entered rehabilitation centers, where they were disarmed and demobilized. Women took credit for this accomplishment, saying they had stopped the war without using the bullet and the gun! (interview, July 20, 2009) The historical accuracy of this anecdote is less important than the impact this story had

on the reconceptualization of women in Rwanda. As stories that referred to women’s tradi- tional influence in preventing men from waging war became more common, women pointed to them in public speeches to legitimize their presence in politics at all levels (Uwineza and Pearson 2009). This positive reference to the idea of women as more peaceful than men led to the popular claim that the genocide would not have happened had women been in charge of the country (field notes; Uwineza and Pearson 2009). Indeed, many of my respondents em- phasized that “the contribution of women is the end of the genocide” (interview with NGO director, July 9, 2009).

During this period, emerging women leaders and the RPF leadership also reintroduced into political discourse the image of the Queen Mother, the historically powerful mother of the Tutsi King in pre-revolutionary Rwanda. The Queen Mother served as proof that the political engagement of women in Rwanda was not a foreign imposition or a threat to the cultural values of the society, but was rather “authentically Rwandan” (Republic of Rwanda 2005). References to the Queen Mother as the advisor and partner of her son were frequent in my interviews and in government reports in the post-genocide period. Today, the Queen Mother has been reincorporated into national discourse as a symbol of women’s strength. According to one member of Parliament,

Today, we are coming back to what has ever been there. It is like history repeats itself. So the Rwandan ideology today in governance somehow goes back far to revisit Rwandan history at the time of the kingdomship and the governance of the king, because the Queen Mother had a voice. She had a say. (interview, Deputy F, July 13 2009).

Rwandan women appropriated culturally salient values of peacefulness and motherhood

in order to legitimize their increased presence in public leadership positions. By linking their current role to a unique narrative deeply embedded in Rwandan history, women charted a course of social mobilization that was palatable to the broader Rwandan society in the post- violence political context. Shared sensibilities about their strengths and potential were mani- fested in women’s increased participation in grassroots organizations, and soon in trainings and conferences convened by international NGOs. Women’s value as legitimate political

From Violence to Mobilization 147

actors emerged from the nexus between immediate social needs—such as the need to care for children and feed their families—and a pressing national need for a new, less-violent type of political actor. As this case study shows, after mass violence the social appropriation of “affirmative essentializations” of women can be critical for mobilization, as it helps justify women’s presence in new social and political spaces.

Mechanisms of Mobilization: Brokerage

The genocide in Rwanda occurred one year before the 1995 Beijing World Conference on Women, and during an unprecedented historical moment when gender equality was vigorously promoted in international discourse. The timing of the genocide thus aided the mobilization of women, because international actors and NGOs rapidly arrived in Rwanda after the genocide and served as brokers of connections between previously unconnected actors—such as a small women’s association and an international funding organization like UNIFEM. The impact of international actors is difficult to measure; it is easy to assume they diffused international norms during the post-violence period, or that they initiated the mobil- ization of women in Rwanda. My data, however, suggest that women’s mobilization was underway long before international actors were able to extend their influence across the country. Further, my data also suggest that the most significant impact of international NGOs—beyond the tremendous amount of aid they distributed—was their ability to facilitate the flow of knowledge, goods, and opportunities between ordinary women, grassroots organizations, and government institutions. For example, Odette, a genocide widow and current leader of a national widows’ organization, was living in Eastern Province after the genocide. She “saw so much need . . . widows were really suffering; they needed houses, food, health care.” So in 1995, she organized a group of several other widows from her area who had lost their homes during the genocide. Together they approached a representative of United Nations Development Programme (UNDP) and requested iron sheets that would allow them to rebuild their homes. The representatives at UNDP listened and eventually provided them with the requested materials. Encouraged by their success, Odette and the group of women decided to form their own organization and continue applying for funding from international sources (interview, July 9, 2012).

The involvement of these foreign actors bolstered the emerging local voices calling for women’s increased presence in public spaces and in leadership roles. They did this by delib- erately soliciting proposals from organizations that promoted women and advocated for their involvement in recovery efforts. In particular, many prioritized housing, health care, and income-generating projects for widows and orphans (Gervais 2003; Newbury and Baldwin 2000). They also organized trainings, paying locals to participate. Soon a group of “interstitial elites” emerged (Swidler and Watkins 2009). This group was comprised of young Rwandans with some education and foreign language competency who were employed by international agencies to implement programs in local communities. They thus served as brokers of connections between “ordinary Rwandans” and foreign and domestic elites. In some cases, these “interstitial elites” and other brokers were returning refugees—the “best-educated and most highly motivated diaspora ever to return in such numbers to an African country” (Kinzer 2008: 312)—and they brought with them their connections to people, organizations, and financial networks across the globe. These new connections then facilitated the exchange of information and ideas through interpersonal interactions and networks, and connected many rural Rwandan women with centralized government bureaucracy. Moreover, many inter- national actors interacted with both the Rwandan government and ordinary people, and thus served as a conduit for further interaction between them.

While ordinary women were forming grassroots organizations and beginning to make claims on local government institutions, larger and more-connected domestic NGOs deepened

Mobilization

148

their reach into the rural sectors of the country. For example, Pro Femmes Twese Hamwe, an umbrella organization for an array of women’s groups, expanded rapidly after the genocide. Pro Femmes was partially funded by international agencies, and it frequently hosted con- sultants from the UN High Commissioner for Refugees and other international organizations that provided inspiration and technical trainings to the organization’s staff (Burnet 2008; Republic of Rwanda 1999; interview, Pro Femme director, July 9, 2009). This allowed Pro Femmes to serve as a broker between large international institutions and rural women’s associations that were in its network. Through the coordinated efforts of local women’s associations and well-connected NGOs like Pro Femmes, indigenously initiated campaigns to increase literacy, develop leadership skills, and practice family planning techniques were implemented across the country, all of which were shaped by the priorities of the international community.

In 2003, Rwanda passed a new constitution that included a provision requiring that 30 percent of all decision-making positions be held by women. The most notable effect of this policy was that it guaranteed at least 30 percent of the Parliament be comprised of women—a number that was surpassed by 18 percent during the same elections. This so-called “gender quota” epitomized the impact of international actors’ involvement, and has been used by scholars since to explain the high level of women in the Rwandan Parliament (Burnet 2008; Longman 2006; Powley 2003). In Rwanda, many international actors encouraged the government to incorporate the platform established at the Beijing Conference, and in doing so legitimized the basic tenets of gender equality espoused within that platform. But these ideas about the importance of women were not solely a foreign idea: many women I interviewed described the policies advanced in the Beijing Platform of Action as “tools” they could draw on to realize their preexisting political ambitions. The implementation of the gender quota thus became a way to guarantee that the progress made by women at the grassroots would result in a minimum level of women in the national legislature—a level that is more than doubled today.

Thus far, I have argued that the mass violence in Rwanda created demographic and economic “threats” that pushed women to mobilize. I then described how this mobilization began, and offered two mechanisms that facilitated this process. In the following section, I describe some of the manifestations of women’s mobilization in Rwanda, including the formal- ization of these nascent organizations and the eventual rise of women in Rwanda’s govern- ment. Manifestations of Mobilization in Civil Society

By 1997, an astonishing 15,400 new women’s associations had formed across Rwanda—

a dramatic increase from the mere handful of civil society organizations that existed before the genocide (Newbury and Baldwin 2000; Powley 2003).11 While the earliest organizations that formed were informal and provided solidarity and support to anyone affected by the genocide, a latter type of initiative advocated for women’s issues at the local, regional, or national level (Gervais 2003). These groups directed their efforts toward local land-rights disputes, health care initiatives, and other critical issues facing their communities, which led to coordinated claims against the state. While no comprehensive data exists that can shed light on the types of women who founded and led these organizations, my data suggest that most were Tutsi with some secondary or university education. At the membership level, however, this type of organization included women of a wide range of backgrounds.

In particular, many of these organizations formed to address land-rights issues. Without legal rights to land or a formalized system of land certification, disputes about ownership were pervasive, and these disputes escalated as hundreds of thousands of old and new case- load refugees began to return from neighboring countries in the mid-1990s (Ali et. al. 2011; Prunier 1995). With over 90 percent of Rwandans dependent on agriculture for their liveli-

From Violence to Mobilization

149

hood, access to land was of vital importance (Ansoms 2008; Gervais 2003). This compelled many returning refugees and landless women—both of whom were predominantly Tutsi—to join together in groups to petition local authorities for access to communal land, land aban- doned by its owner, or less-desirable marshland that was still considered part of the commons (Gervais 2003; field notes).

Many of these collective efforts at making claims to land succeeded, and lobbying local officials through the institutional structure of an organization became one of the only ways women could gain formal access to land (Gervais 2003; multiple interviews). Centralized “umbrella” organizations like Pro Femmes Twese Hamwe helped coordinate many of these efforts, and led a coalition of women’s groups in a campaign to lobby legislators to change the law to allow women to inherit property from deceased husbands or family members (Burnet 2008). The mayor mentioned above described how her personal situation led to her involve- ment with a variety of organizations that eventually lobbied for the inheritance law to be changed. She stated,

I was faced with raising 14 orphans. . . . Other women were faced with raising the same thing— family from their husband’s side and family from their own. So faced with those respon- sibilities, women decided to get together . . . organizing around the rights of women; so traditional organizations fought for the 1999 law for women to be able to inherit land from both their family and their husband’s family. (interview June 26, 2009) As the mayor mentions, women’s collective efforts to obtain land rights eventually

manifested in a campaign, run by various women’s groups in conjunction with women in government, to amend the legal code to allow women to inherit land from deceased family members. In 1999, the law was passed, granting women the right to inherit land and property for the first time. While the law had several notable shortcomings (Pottier 2006), it was nevertheless a major accomplishment of women’s mobilization in Rwanda. It also set the stage for subsequent legislation that established children’s rights and protected against gender-based violence (Powley 2006; Powley and Pearson 2007).

Manifestations of Mobilization in Politics

While explaining the bottom-up process of women’s mobilization is this article’s primary

goal, in the final section I develop a preliminary argument about one of the key manifestations of women’s mobilization in Rwanda—namely, the high percentage of women in Rwanda’s national government. Today women in Rwanda hold 64 percent of the seats in Parliament, 38 percent of ministerial positions, 40 percent of the positions in local government, and top positions in the judiciary, police, and other government institutions (Rwanda Women Parliamentary Forum 2007, 2011; Inter-Parliamentary Union 2014). In what follows, I argue that the process of grassroots mobilization previously discussed was a necessary—but not sufficient—condition for the high levels of women’s political representation in Rwanda today.

As shown in table 1, the career trajectories of the forty women in government12 interviewed for this project suggest that women reached positions in national politics after

Table 1. Biographical Information on Sample of Political Elite Respondents (n=40)

Org member INGO Local gov. Total Widow/survivor of genocide (likely Tutsi) 93% (13) 79% (11) 57% (8) 14 Returning refugee (likely Tutsi) 80% (12) 80% (12) 53% (8) 15 No mention of victimization (likely Hutu) 54% (6) 36% (4) 45% (5) 11 Total 31 27 21 40

Note: Individual respondents often participated in more than one caterogy

Mobilization

150

having worked with a grassroots organization or with an internationally linked NGO, or after having served in local government. Many women currently in government had served in all three capacities prior to their current position. Of the women currently in high-ranking government positions who were widowed or lost family during the genocide, 93 percent became involved with a grassroots women’s association at some point after the genocide. Interestingly, for those members of government whose ethnicity it was difficult to ascertain from the interview (i.e., likely Hutu women), merely 54 percent had been members of a grassroots organization. This suggests that organizations for widows and survivors were perhaps the most important organizations for linking members to positions in national govern- ment.

Of course, the mobilization of women does not fully explain the high level of women in Rwanda’s government. Instead, I find that women’s national political successes depended on two additional conditions: comprehensive regime change and gender sensitive leadership. First, after the genocide there was a complete overhaul of the former political regime. This meant that an entirely new governing elite came to power after the violence, as most of the old regime was dead, on the run, or in jail. This led to an enormous demand for new political office holders at all levels of government. In effect, this was a political opportunity that women involved in community life were well positioned to take advantage of.13 The demand for new office holders was so great, in fact, that anthropologist Jennie Burnet (2008: 378) laments, “the most vibrant leaders of women’s civil society organizations left to take positions in the government.”

Second, there was an overt promotion of women by leaders of the new regime, and particularly by current President Paul Kagame (Republic of Rwanda 2000).14 The reasons for this are debated, but a few are particularly plausible. Most critically, despite the RPF’s suc- cess at bringing the genocide to an end, in 1994 it was still considered the enemy and distrusted by the majority of Rwandans. This is because the RPF was predominantly com- prised of Tutsis, and was rumored to have committed revenge killings and war crimes while stopping the genocide (Prunier 1995; Reyntjens 2004). As the RPF attempted to secure and legitimize its control, women emerged as a large, nonethnic political constituency that could safely be championed by the RPF. It was ostensibly in the context of this political tension that Kagame and the RPF began to vocally encourage women to join politics, and they targeted the women who were emerging as leaders of local grassroots organizations. This may have also inspired support for the 30 percent gender quota implemented in 2003.15 Regime change and gender sensitive leadership allowed the grassroots mobilization of women in Rwanda to culminate in formal political representation (see figure 2). The high level of women in government is thus a result of both bottom-up and top-down forces, and could not be explained without looking at the critical new roles women took on in their households and communities in the aftermath of the violence. This mobilization process resulted in a pool of available women who were well trained in local community politics, and who aspired to participate in government at a higher level.

Figure 2. Theorized Relationship between Mobilized Women After Violence and Women’s Parliamentary Representation

From Violence to Mobilization

151

Today, the demographic balance in society has been largely restored, as women comprise 53 percent of the population. Moreover, foreign actors and the new regime have radically transformed the country’s economic landscape, and the percentage of the population in ex- treme poverty has decreased (Republic of Rwanda 2012). Yet, while women have managed to continue their involvement in politics and civil society, this certainly does not mean that allwomen in Rwanda are “empowered.” The authoritarian nature of the state means that few beyond the executive have any meaningful political control (Reyntjens 2011). Moreover, past research has shown that women in politics often advance legislation that is distinctly non- feminist, or that reinforces the power of the ruling regime (Bauer and Britton 2006; Goetz and Hassim 2003; Tamale 1999). This is partly the case in Rwanda; for example, the women- dominated legislature passed a law in 2009 that shortened maternity leave from twelve to six weeks (Debusscher and Ansoms 2013).

Moreover, as I have detailed elsewhere (Berry forthcoming), inequality in Rwanda is in- creasing, and the vast majority of Rwandan women remain impoverished and have yet to reap the full benefits of the legal rights they have on paper (Ansoms 2009; Burnet 2012; Debusscher and Ansoms 2013). This is partly a result of the intrusion of the state into the private lives of its citizens, through strict regulatory measures that seek to manage and control all areas of Rwandan life. Moreover, customary beliefs about the second-class status of women prevail across the country, perpetuating the gendered power structures that have existed for generations. Perhaps most importantly, as my data reflect, the women that have reached high levels of political power are disproportionately Tutsi women and returning refugees, suggesting that Hutu women face additional impediments to their incorporation in politics. These factors should cause us to pause before championing Rwanda as a model of women’s empowerment, regardless of the impact mass violence had on the transformation of women’s roles from private to public space.

CONCLUSION

This article illustrates the relationship between mass violence and women’s mobilization in Rwanda. It aims to show how mass violence can create threatening conditions that mobilize women to form grassroots organizations, engage in new forms of claims making with govern- ment institutions, and eventually run for political office. The social appropriation of values that are traditionally feminine like “motherhood” and the involvement of international actors as brokers facilitate this mobilization process. By illustrating this case study, I aim to make three contributions to the literature on social movements. First, I follow social movement scholars by suggesting that “threat” is a critical concept in studies of mobilization. Building on studies that look at threats that emanate from beyond the state, I suggest that threats emerging from mass violence can have similar mobilizing effects. This should encourage future researchers to look at how women, and perhaps other marginalized groups, may have unforeseen opportunities to transform their position in society in the immediate aftermath of mass violence. Second, the argument I have developed above aims to advance social movement research by illustrating how nonstate threats are particularly gendered. As women hold the majority of care responsibilities in countries around the world, the lack of access to basic supplies can, as scholars have shown, inspire mobilization among women. Describing this mobilization contributes to feminist scholarship that challenges the normative assump- tions about female victimhood during mass violence. Finally, I answer the call for research on less formal movements in non-Western settings. The conceptualization of mobilization in this article includes more diffuse and informal forms of organization, which better capture the arenas in which women’s mobilization is likely to occur. In doing so, I aim to illustrate the gendered mobilizing effects of large-scale atrocities.

Mobilization

152

These findings provide empirical support for the claim that mass violence precipitated women’s mobilization in Rwanda, but how well might this argument hold up in other cases of mass violence? Future research should investigate whether threats created by mass violence can have mobilizing effects for women in other contexts. Based on the findings presented here, we would expect mass violence to be mobilizing for women if there is both a demographic imbalance between men and women after the violence has ended, and an urgent material need for basic goods and supplies. This might mean that mass violence in contexts where women and men were killed, imprisoned, or displaced at the same rate would not result in similar rates of women’s mobilization after violence. Moreover, mass violence in more developed countries might also fail to have mobilizing effects, as the economic needs created by the violence would likely pose a less significant threat to individual lives. It is also possible that peace and security are preconditions for such mobilization to occur.

While the empirical focus of this article is on Rwanda, the broader theoretical goal of this project is to complicate our dominant understanding of mass violence as only destructive, and to look for social arenas in which it may also be transformative. Mass violence is a liminal historical event in which gendered social hierarchies can be dissolved, institutions can be reconstructed, and society as we know it can be reimagined. The Rwandan genocide was overwhelmingly devastating, and yet it also precipitated a dramatic transformation of women’s roles in society.

NOTES

1 Rwanda’s genocide unfolded in the context of a broader civil war, which began in 1990. This article treats both processes together under the concept of “mass violence” and often refer to the violence simply as “violence” or “genocide.” 2 These ethnic categories have come to be understood as dynamic identities that have held different meanings in different time periods. In general, Tutsis were of higher economic and social status than Hutus, who were generally sedentary farmers. Members of these groups traditionally spoke the same language (Kinyarwanda), lived in the same areas, practiced the same religion, and belonged to the same clans (Newbury 1988; Mamdani 2001). 3 Currently the rate of NGO or organization membership in Rwanda is very high due to the popularity of income- saving and loan cooperatives; therefore, at times it became difficult to delineate between categories 2 and 3, as nearly all of my interviewees claimed membership in some form of civil society organization. Thus, for the purpose of this paper, they are lumped together into a single category. 4 I conducted seven small group interviews with groups of 3-5 participants within this group of “ordinary” women, which were usually several hours in length. This allowed me to ask individualized questions of participants, while facilitating a more open and supportive interview environment. 5 Including Zaire (Democratic Republic of the Congo), Tanzania, and Burundi. 6 Of course while I refer to “women” as a category often throughout the analysis that follows, it is worth noting that differences in ethnicity, genocide experience, education, and class all profoundly shaped individual women’s mobilizing experiences. 7 As this member of Parliament was male, he was not counted in the total interview sample. 8 “Old caseload” refugees refers to predominantly Tutsi refugees who fled Rwanda during waves of violence beginning during the 1959 Revolution. The majority of these refugees settled in the Democratic Republic of Congo, Uganda, Burundi, and Tanzania, although others made their way to other parts of Africa, Europe, and even to North America. “New caseload” refugees, who were predominantly Hutu, refers to those Rwandans who fled Rwanda during the genocide and its aftermath, and settled mostly in the Democratic Republic of Congo. 9 AVEGA is the French acronym for “Widows of the April Genocide,” and Agahozo is a Kinyarwanda word that means “wipe away your tears.” 10 The gacaca court system was a traditional community justice institution that was repurposed after the genocide to face the daunting task of trying over one million suspected genocide perpetrators. 11 A report by Réseau des Femmes lists the number of women’s organizations in 1999 even higher, at nearly 100,000 (USAID 2001; Zraly 2008). 12 Or women in government-affiliated institution or high-ranking NGO positions. 13 There was also a complete overhaul of the Rwandan constitution, penal code, and legal system, requiring new laws and policies to be rewritten with sensitivity to the post-genocide context. Since women had taken on new roles in their communities and were being reconceptualized as legitimate political actors, their new status was reflected in many of these new laws and documents. 14 Pasteur Bizimungu, a Hutu member of the RPF, held the title of President from 1994-2000, although Kagame, as the successful head of the military that toppled the old regime and stopped the genocide, was firmly in control behind the scenes.

From Violence to Mobilization

153

15 In addition to this, there are potentially two other factors that help to explain the RPF’s sustained promotion of gender equality. First, the RPF’s early leadership was oriented toward a revolutionary ideology that is reflected in its strict code of military discipline, governing strategy, and willingness to treat women as equals (Kinzer 2008). Second, because of this ideological orientation and the dire situation Rwandan refugees found themselves in while in Uganda, the RPF depended heavily on women within the movement during the years it was growing from a fledgling philan- thropic organization into a highly trained military operation.

REFERENCES

African Rights. 1995. Rwanda: Not So Innocent–When Women Become Killers. London: African Rights. Ali, Daniel Ayalew, Klaus Deininger, and Markus Goldstein. 2011. “Environmental and Gender Impacts

of Land Tenure Regularization in Africa: Pilot Evidence from Rwanda.” Journal of Development Economics 110(1): 262-75.

Alimi, Eitan Y. 2009. “Mobilizing Under the Gun: Theorizing Political Opportunity Structure in a Highly Repressive Setting.” Mobilization 14(2): 219-37.

Almeida, Paul D. 2003. “Opportunity Organizations and Threat-Induced Contention: Protest Waves in Authoritarian Settings.” American Journal of Sociology 109(2): 345-400.

Ansoms, An. 2008. “Striving for Growth, Bypassing the Poor?: A Critical Review of Rwanda’s Rural Sector Policies.” Journal of Modern African Studies 46(1): 1-32.

———. 2009. “Re-Engineering Rural Society: The Visions and Ambitions of the Rwandan Elite.” African Affairs 108(431): 289-309.

Bauer, Gretchen, and Hannah Britton. 2006. Women in African Parliaments. Boulder, CO: Lynne Rienner Publishers.

Beck, Elwood M. 2000. “Guess Who’s Coming to Town: White Supremacy, Ethnic Competition, and Social Change.” Sociological Focus 33(2): 153-174.

Berry, Marie E. Forthcoming. “When ‘Bright Futures’ Fade: Paradoxes of Women’s Empowerment in Rwanda.” Signs: Journal of Women in Culture and Society.

Blee, Kathleen M. 1991. Women of the Klan: Racism and Gender in the 1920s. Berkeley, CA: University of California Press.

Bop, Codou. 2001. “Women in Conflict: Their Gains and Their Losses.” Pp. 19-35 in The Aftermath: Women in Post-Conflict Transformation, edited by Sheila Meintjes, Meredeth Turshen, and Anu Pillay. London: Zed Books.

Burnet, Jennie E. 2008. “Gender Balance and the Meanings of Women in Governance in Post-Genocide Rwanda.” African Affairs 107(428): 361-86.

———. 2012. Genocide Lives in Us: Women, Memory, and Silence in Rwanda. Madison: University of Wisconsin Press.

Codere, Helen. 1973. The Biography of an African Society: Rwanda 1900-1960. Tervuren, BE: Musée royal de l'Afrique centrale.

Cohen, Mardge H., Anne-Christine d'Adesky, and Kathryn Anastos. 2005. “Women in Rwanda: Another World Is Possible.” Journal of the American Medical Association 294(5): 613-15.

Debusscher, Petra, and An Ansoms. 2013. “Gender Equality Policies in Rwanda: Public Relations or Real Transformations?” Development and Change 44(5): 1111-1134.

Des Forges, Alison Liebhafsky. 1999. “Leave None to Tell the Story”: Genocide in Rwanda. New York: Human Rights Watch.

DiMaggio, Paul, and Walter Powell. 1983. “The Iron Cage Revisited: Institutional Isomorphism and Collective Rationality in Organizational Fields.” American Sociological Review 48(2): 147-160.

Einwohner, Rachel L., and Thomas V. Maher. 2011. “Threat Assessment and Collective-Action Emergence: Death-Camp and Ghetto Resistance During the Holocaust.” Mobilization 16(2): 127-146.

Elster, Jon. 1989. Nuts and Bolts for the Social Sciences. Cambridge, UK: Cambridge University Press. Ferree, Myra Marx. 1992. “The Political Context of Rationality: Rational Choice Theory and Resource

Mobilization.” Pp. 29-52 in Frontiers in Social Movement Theory, edited by Aldon Morris and Carol McClurg Mueller. New Haven, CT: Yale University Press.

Fujii, Lee Ann. 2009. Killing Neighbors: Webs of Violence in Rwanda. Ithaca, NY: Cornell University Press. Gal, Susan, and Gail Kligman. 2000. The Politics of Gender After Socialism: A Comparative-Historical

Essay. Princeton, NJ: Princeton University Press. Gervais, Myriam. 2003. “Human Security and Reconstruction Efforts in Rwanda: Impact on the Lives of

Women.” Development in Practice 13(5): 542-51.

Mobilization

154

Goetz, Anne-Marie, and Shireen Hassim. 2003. No Shortcuts to Power: African Women in Politics and Policy Making. London: Zed Books.

Goldstone, Jack A. 2002. Revolutions: Theoretical, Comparative, and Historical Studies. Belmont, CA: Wadsworth/Thomson Learning.

Goldstone, Jack A., and Charles Tilly. 2001. “Threat (and Opportunity): Popular Action and State Response in the Dynamics of Contentious Action.” Pp. 179-194 in Silence and Voice in the Study of Contentious Politics, edited by Ronald Aminzade, Jack Goldstone, Doug McAdam, Elizabeth Perry, William Sewell, Jr., Sidney Tarrow, and Charles Tilly. New York: Cambridge University Press.

Goodwin, Jeff. 2001. No Other Way Out: States and Revolutionary Movements, 1945-1991. Cambridge, MA: Cambridge University Press.

Hedström, Peter, and Richard Swedberg. 1998. Social Mechanisms: An Analytical Approach to Social Theory. Cambridge, UK: Cambridge University Press.

Helms, Elissa. 2002. “Women as Agents of Ethnic Reconciliation? Women’s NGOs and International Intervention in Post-war Bosnia-Herzegovina.” Women’s Studies International Forum 26(1): 15-33.

Inter-Parliamentary Union. 2014. “Women in Parliaments: World and Regional Averages, Online Database.” PARLINE. (http://www.ipu.org/parline/:Inter-ParliamentaryUnion). Retrieved Novem- ber 12, 2014

Jaquette, Jane S. 1994. The Women’s Movement in Latin America: Participation and Democracy. Boulder, CO: Westview Press

Jefremovas, Villia. 1991. “Loose Women, Virtuous Wives, and Timid Virgins: Gender and the Control of Resources in Rwanda.” Canadian Journal of African Studies 25(3): 378-95.

Kaplan, Temma. 1982. “Female Consciousness and Collective Action: The Case of Barcelona, 1910- 1918.” Signs 7(3): 545-66.

———. 1997. Crazy for Democracy: Women in Grassroots Movements. New York: Routledge. Khawaja, Marwan. 1993. “Repression and Popular Collective Action: Evidence from the West Bank.”

Sociological Forum 8(1): 47-71. Kinzer, Stephen. 2008. A Thousand Hills: Rwanda’s Rebirth and the Man Who Dreamed It. Hoboken,

NJ: John Wiley & Sons. Kurzman, Charles. 1996. “Structural Opportunity and Perceived Opportunity in Social Movement

Theory: The Iranian Revolution of 1979.” American Sociological Review 61(1): 153-170. Longman, Timothy. 2006. “Rwanda: Achieving Equality or Serving an Authoritarian State?” Pp. 133-

150 in Women in African Parliaments, edited by Gretchen Bauer and Hannah Britton. Boulder, CO: Lynne Rienner Publishers.

Loveman, Mara. 1998. “High Risk Collective Action: Defending Human Rights in Chile, Uruguay, and Argentina.” American Journal of Sociology 104(2): 477-525. Maher, Thomas V. 2010. “Threat, Resistance, and Collective Action.” American Sociological Review

75(2): 252-72. Mamdani, Mahmood. 2001. When Victims Become Killers: Colonialism, Nativism, and the Genocide in

Rwanda. Princeton, NJ: Princeton University Press. Martin, Vanessa. 2011. “Women and Popular Protest: Women’s Demonstrations in 19th Century Iran.”

Pp. 50-66 in Subalterns and Social Protest: History From Below in the Middle East and North Africa, edited by Stephanie Cronin. New York: Routledge.

McAdam, Doug. 1982. Political Process and the Development of Black Insurgency, 1930-1970. Chicago: University of Chicago Press.

McAdam, Doug, Sidney G. Tarrow, and Charles Tilly. 2001. Dynamics of Contention. Cambridge, UK: Cambridge University Press.

McVeigh, Rory. 1999. “Structural Incentives for Conservative Mobilization: Power Devaluation and the Rise of the Ku Klux Klan, 1915-1925.” Social Forces 77(4): 1461-96.

Meyer, David S., and Debra Minkoff. 2004. “Conceptualizing Political Opportunity.” Social Forces 82(4): 1457-92.

Molyneux, Maxine. 1985. “Mobilization Without Emancipation? Women’s Interests, the State, and Revolution in Nicaragua.” Feminist Studies 11(2): 227-54.

———. 1998. “Analysing Women’s Movements.” Development and Change 29(2): 219-45. Moore, Barrington. 1966. Social Origins of Dictatorship and Democracy: Lord and Peasant in the

Making of the Modern World. Boston, MA: Beacon Press. Nagel, Joane. 1998. “Masculinity and Nationalism: Gender and Sexuality in the Making of Nations.”

Ethnic and Racial Studies 21(2): 242-69.

From Violence to Mobilization

155

Newbury, Catharine. 1988. The Cohesion of Oppression: Clientship and Ethnicity in Rwanda, 1860- 1960. New York: Columbia University Press.

———. 2011. “High Modernism at the Ground Level: The Imidugudu Policy in Rwanda.” Pp. 223-39 in Remaking Rwanda, edited by Scott Straus and Lars Waldorf. Madison, WI: University of Wisconsin Press.

Newbury, Catherine, and Hannah Baldwin. 2000. “Aftermath: Women in Post-genocide Rwanda.” Washington, DC: USAID Center for Development Information and Evaluation.

Noonan, Rita. 1995. “Women Against the State: Political Opportunities and Collective Action Frames in Chile’s Transition to Democracy.” Sociological Forum 10(1): 81-111.

Nowrojee, Binaifer. 1996. “Shattered Lives: Sexual Violence during the Rwandan Genocide and its Aftermath.” Pp. 1-103 in Human Rights Watch Africa, Women’s Rights Project. New York: Human Rights Watch.

Nyseth Brehm, Hollie, Christopher and Uggen, and Jean-Damascene Gasanabo. 2013. “Age, Sex, and the Crime of Genocide.” Unpublished Paper (shared by the authors).

Polavarapu, Aparna. 2011. “Procuring Meaningful Land Rights for the Women of Rwanda.” Yale Human Rights and Development Law Journal Vol. XIV: 101-150.

Pottier, Johan. 2002. Re-Imagining Rwanda: Conflict, Survival, and Disinformation in the Late Twentieth Century. Cambridge, UK: Cambridge University Press.

———. 2006. “Land Reform for Peace? Rwanda’s 2005 Land Law in Context.” Journal of Agrarian Change 6(4): 509-37.

Powley, Elizabeth. 2003. “Rwanda: Women Hold Up Half the Parliament.” Pp. 154-163 in Women in Parliament: Beyond Numbers, Revised ed., edited by Julie Ballington and Azza Karam. Stockholm, SE: International Institute for Democracy and Electoral Assistance.

———. 2004. “Strengthening Governance: The Role of Women in Rwanda’s Transition.” Pp. 1-37 in Women Waging Peace, edited by Sanam Naraghi Anderlini. Washington, DC,: Hunt Alternatives Fund.

———. 2006. “Rwanda: The Impact of Women Legislators on Policy Outcomes Affecting Children and Families.” Pp. 1-18 in State of the World’s Children Background Paper, edited by UNICEF.

Powley, Elizabeth, and Elizabeth Pearson. 2007. “‘Gender Is Society’: Inclusive Lawmaking in Rwanda's Parliament.” Critical Half (Women for Women International): Winter: Pp.16-21.

Prunier, Gérard. 1995. The Rwanda Crisis, 1959-1994: History of a Genocide. London: Hurst. ———. 2009. Africa’s World War: Congo, the Rwandan Genocide, and the Making of a Continental

Catastrophe. Oxford, UK: Oxford University Press. Reyntjens, Filip. 2004. “Rwanda, Ten Years On: From Genocide to Dictatorship.” African Affairs

103(411): 177-210. ———. 2011. “Constructing the Truth, Dealing with Dissent, Domesticating the World: Governance in

Post-Genocide Rwanda.” African Affairs 110(438): 1-34. Rwanda, Republic of. 1999. “Country Progress Report on Implementation of Women’s World Regional

and National Action Platformes.” Kigali, RW: Ministry of Gender & Women Development. ———. 2000. “Rwanda Vision 2020.” Kigali, RW: Ministry of Finance and Economic Planning. ———. 2004. “Survey of Genocide Deaths.” Kigali, RW: Ministry of Local Administration and

Community Development. ———. 2005. “The Role of Women in Reconciliation and Peace Building in Rwanda: 10 Years After

Genocide 1994-2004.” Kigali, RW: National Unity and Reconciliation Commission. ———. 2007. “Single Report Equal to the 4th, 5th, and 6th Reports CEDAW: 1994-2005.” Kigali, RW:

Ministry of Gender and Family Promotion. ———. 2012. “EICV3 Thematic Report: Economic Activity.” Kigali, RW: National Institute of Statistics. Rwanda Women Parliamentary Forum (FFRP). 2007. “Rwanda’s Progress Towards a Gender Equitable

Society.” Pp. 1-52. Kigali, RW: Government of Rwanda. ———. 2011. “Unity, Dignity for a Better Future.” Pp. 1-24. Kigali, RW: Government of Rwanda. Safa, Helen Icken. 1990. “Women’s Social Movements in Latin America.” Gender & Society 4(3): 354-69. Sharlach, Lisa. 1999. “Gender and Genocide in Rwanda: Women as Agents and Objects of Genocide.”

Journal of Genocide Research 1(3): 387-99. Shriver, Thomas, Alison Adams, and Rachel Einwohner. 2013. “Motherhood and Opportunities for

Activism Before and After the Czech Velvet Revolution.” Mobilization 18(3): 267-88. Stephen, Lynn. 1997. Women and Social Movements in Latin America: Power From Below. Austin, TX:

University of Texas Press. Straus, Scott. 2006. The Order of Genocide: Race, Power, and War in Rwanda. Ithaca, NY: Cornell

University Press.

Mobilization 156

Swidler, Ann, and Susan Cotts Watkins. 2009. “‘Teach a Man to Fish’: The Sustainability Doctrine and its Social Consequences.” World Development 37(7): 1182-1196.

Tamale, Sylvia. 1999. When Hens Begin to Crow: Gender and Parliamentary Politics in Uganda. Boulder, CO: Westview Press.

Tarrow, Sidney. “States and Opportunities: The Political Structuring of Social Movements.” Pp. 41-61 in Comparative Perspectives on Social Movements: Political Opportunities, Mobilizing Structures, and Cultural Framings, edited by Doug McAdam, John D. McCarthy and Mayer N. Zald. New York: Cambridge University Press, 1996.

———. 1998. Power in Movement: Social Movements, Collective Action, and Politics. New York Cambridge University Press.

Taylor, Verta. 1999. “Gender and Social Movements: Gender Processes in Women’s Self-Help Movements.” Gender & Society 13(1): 8-33.

Taylor, Verta, and Leila Rupp. 2002. “Loving Internationalism: The Emotion Culture of Transnational Women’s Organizations, 1888-1945.” Mobilization 7(2): 141-158.

Tilly, Charles. 1978. From Mobilization to Revolution. Reading, MA: Addison-Wesley. Tilly, Louise A. 1981. “Paths of Proletarianization: Organization of Production, Sexual Division of

Labor, and Women’s Collective Action.” Signs 7(2): 400-17. Tripp, Aili Mari. 2000. Women and Politics in Uganda. Madison, WI: University of Wisconsin Press. USAID Rwanda. 2001. “Civil Society in Rwanda: Assessment and Options.” Retrieved March 1, 2011

(http://pdf.usaid.gov/pdf_docs/ PNACM181.pdf). Uwineza, Peace, and Elizabeth Pearson. 2009. “Sustaining Women’s Gains in Rwanda: The Influence of

Indigenous Culture and Post-Genocide Politics.” Pp. 1-32 in The Rwanda Project. Washington, DC: The Hunt Alternative Fund.

Van Dyke, Nella, and Sarah A. Soule. 2002. “Structural Social Change and the Mobilizing Effect of Threat: Explaining Levels of Patriot and Militia Organizing in the United States.” Social Problems 49(4): 497-520.

Verwimp, Philip. 2005. “An Economic Profile of Peasant Perpetrators of Genocide: Micro-level Evidence from Rwanda.” Journal of Development Economics 77(2): 297-323.

Walsh, Edward J. 1981. “Resource Mobilization and Citizen Protest in Communities Around Three Mile Island.” Social Problems 29(1): 1-21.

World Bank Data. [1993]2012. “World Development Indicators: Rwanda.” The World Bank. Retrieved November 18, 2013 (http://www.worldbank.org/en/country/rwanda).

Zhao, Dingxin. 2001. The Power of Tiananmen: State-Society Relations and the 1989 Beijing Student Movement. Chicago, IL: University of Chicago Press.

Zraly, Maggie. 2008. “Bearing: Resilience Among Genocide-Rape Survivors in Rwanda.” Ph.D. dissertation, Department of Anthropology, Case Western Reserve University, Cleveland, OH.

Zraly, Maggie, and Laetitia Nyirazinyoye. 2010. “Don’t Let the Suffering Make You Fade Away: An Ethnographic Study of Resilience Among Survivors of Genocide-Rape in Southern Rwanda.” Social Science & Medicine 70(10): 1656-64.

Zraly, Maggie, Julia Rubin-Smith, and Therese Betancourt. 2011. “Primary Mental Health Care for Survivors of Collective Sexual Violence in Rwanda.” Global Public Health 6(3): 257-70.