Sisters Are Doing It for Themselves: How Female Combatants Help Generate Gender-Inclusive Peace Agreements in Civil Wars

Abstract

This article examines the effect rebel women have on the shape of civil war peace agreements, paying particular attention to the specific gender-inclusive provisions female rebels advocate for. I argue that, through conflict experiences and socialization, rebel women develop group identities that foster collective demands. Their identities as fighters and women from marginalized groups encourage rebel women to lobby for provisions that address the grievances of women from these societal groups. Using data on women’s participation in conflict and the terms written into contemporary peace agreements, I find support for this contention. Greater participation of female combatants is associated with an increased likelihood of observing gender-inclusive agreement provisions calling for the inclusion of women from marginalized groups and addressing the specific post-conflict needs of female ex-combatants. This study is one of the first to show that women’s participation in rebellion matters for the shape of post-conflict peace.

INTRODUCTION

In 2016, the Revolutionary Armed Forces of Colombia (FARC) signed a transformative, yet controversial peace agreement with the Colombian government containing over two hundred mentions of women and more than 50 mentions of gender. Numerous gender-inclusive provisions were imbedded in the agreement, including a commitment to maintain a gendered approach to its truth and reconciliation commission, gender integration in its transitional justice and rehabilitation and reconstruction programs, guarantees for women’s political participation, and resolutions to address gender-based violence in the country. Victoria Sandino Palmera, a member of the FARC’s peace delegation, proffered the agreement aimed to “advance equal opportunities between men and women…meet the needs and proposals of women living in vulnerable conditions, such as women heads of household, victims of conflict, rural, Indigenous, Afro-descendant, and Black women, among others” (Sandino Palmera 2016). FARC women aimed to recognize the cross-cutting inequities or “multiple discriminations” that result when gender interacts with Colombian women’s other identities. Why did this conflict yield such an inclusive peace agreement?

According to Bouvier (2016, 19), “at the table, around the table, behind the table, and at side tables, women [were] having their say and shaping the path to peace” in Colombia. Women in civil society organizations, whose efforts were facilitated by transnational advocacy groups and the international community, have been widely recognized for their roles in pushing women’s inclusion in the peace process. However, little attention has been paid to the impact female combatants have had on the terms of the agreement. This is surprising since FARC guerrilleras served in many formal capacities during the peace talks and figured prominently in the organization more broadly. During talks held between 2012 and 2016, Farianas (FARC women) comprised 20% of the group’s negotiators and nearly half of its peace delegation (Simanca Herrera 2018). The FARC was first to appoint a woman, Sandino, to its negotiation team and female FARC members served on the unprecedented Gender Subcommittee, which gathered members of the government and guerrilla delegations together to ensure the final agreement took concerns of women seriously.

Women were also well-represented within the FARC’s rank and file; as much as 40% of it membership was female. Though few Farianas were in the room during negotiations, their presence within the organization impacted its approach to the peace process and made it possible for FARC negotiators like Sandino to represent the gender perspective during the talks. According to Sandino, her position at the negotiating table created “a window so women could have more visibility” in the process (Ruiz-Navarro 2019).

I examine whether women’s integration into rebel forces influences the adoption of specific gender-inclusive peace terms. Peace agreements can provide political opportunities for disadvantaged groups, including women (Anderson 2015). Since peace agreements constitute legal obligations that can lead to substantial changes in domestic legislation (Bell 2008; Dancy 2018), the incorporation of gender-inclusive terms into accords can yield important alterations to the political and social landscape for women. Thus, women push for provisions that ensure a post-conflict society inclusive of their interests. I argue that female fighters in rebel organizations can be important drivers of this process.

I emphasize the idea that rebel women develop collective identities around their specific lived experiences, which induces female combatants to push for provisions that address their particular needs. Rebel women will lobby for provisions that generate equality between men and women and ensure equity for historically marginalized women. They are also likely to push for provisions that cater to former female combatants’ post-conflict needs.

This project departs from existing work in three important ways. First, previous studies examining the determinants of gender-inclusive agreements focus solely on the role of women in civil society (Anderson 2015).¹ While they have been the group to secure access to peace negotiations most often (Martín de Almagro 2018), attention should be given to the roles of rebel women as well. Even where they are locked out of formal negotiation processes, rebel women can indirectly influence the demands articulated during peace talks, especially where they constitute a sizable lobby within rebel organizations. I assess female rebels’ influence on armed groups by examining the extent to which women participate in combat, given the centrality of these roles to many rebel organizations. Second, I focus on rebel women’s indirect impact on the peace process. Some previous studies focus on women’s formal participation as invited guests at negotiating tables, overlooking the ways that they may impact the terms of peace from outside. The potential for backchannel advocacy is important since women’s formal participation in negotiation processes is rare. Thus, an agreement’s focus on gender is unlikely to result from women’s presence at the peace table alone. Some variation may be explained by women’s advocacy outside of formal channels.

Third, this article deviates from previous work in its focus on the specific provisions enshrined in agreements since terms vary significantly across documents. Focusing only on whether agreements provide any references to women impedes our understanding of the purposeful design of agreements and makes it difficult to understand why and how specific terms became adopted. I argue that female combatants’ intersectional identities as women from marginalized political groups and rebel groups, will lead them to advocate for specific provisions that directly benefit female combatants. This approach is in line with existing work arguing the quality of the specific women in the room matters more for the outcome of peace processes than the quantity (Karam 2000).

Using data on over 1,200 peace agreements signed between 1990 and 2019, I show the proportion of female combatants in a rebel organization bears a strong influence on the presence of certain gender-inclusive agreement provisions, specifically those affecting female combatants and the populations from which they tend to be recruited. Greater female participation in rebel combat forces also increases the probability that women-inclusive demobilization, disarmament, and reintegration (DDR), rehabilitation, and equality provisions are incorporated.

WOMEN’S POST-CONFLICT TRANSFORMATION

Despite efforts by the international community to push the women, peace, and security (WPS) agenda, women’s formal participation in conflict resolution processes remains surprisingly low. For instance, in peace processes between 1992 and 2019, women have participated as formal negotiators in only 13% of cases and signatories and mediators in 6% of the processes, respectively. According to Tamaru and O’Reilly (2018), women have been the largest group left out of political processes historically. Despite their minimal representation in formal talks and their larger scale marginalization in conflict resolution efforts, women have managed to find ways to seize on opportunities to gain representation in peace processes anyway (Tripp et al. 2008).

War can be transformative for women and gender relations (Karam 2000; Tripp 2015). During conflicts, women experience substantial changes in their status and opportunities (Tripp 2015), sometimes taking on societal roles that were inaccessible prior to the outbreak of war (Hughes and Tripp 2015; Mazurana, Krystalli, and Baaré 2018). These experiences offer them a glimpse into a drastically altered society—one in which women have expanded rights and freedom. These liberties do not always stick, however (Molyneux 1985), and some women lament the return to the pre-war gendered status quo (Mazurana, Krystalli, and Baaré 2018; Veale 2003).

Across the globe, women have organized to prevent these reversions (Tripp 2015; Tripp et al. 2008). Through their participation in the conflict resolution process, women attempt to usher in peace on their own terms. Participation in the crafting of negotiated settlements presents women with an opportunity to cement wartime transformations since compromise settlements, by nature, assume significant deviations from existing conditions. The United Nations Development Fund for Women (UNIFEM) asserts that “where women have been involved, even as nonspeaking observers, in peace processes, they have been able to ensure that matters of importance to them are included in peace accords” (Goetz 2009, 3).

Existing research finds that more inclusive solutions are proposed when women participate in the drafting of peace agreements and post-conflict constitutions (Bell 2004; Tamaru and O’Reilly 2018). Agreements are also likelier to contain political, social, and economic reforms (Krause, Krause, and Bränfors 2018) and provide for the input and participation of a diverse set of societal actors, yielding more equitable and inclusive documents (Tamaru and O’Reilly 2018). When civil society women have significant input in a peace process, agreements are more apt to include provisions on women’s rights (Anderson 2015), which ultimately leads to improvements in women’s post-conflict rights (Bakken and Buhaug 2021; Reid 2021). Peace is also more durable when women have formal roles at the bargaining table (Krause, Krause, and Bränfors 2018; Stone 2014). Together, these studies offer evidence that female participation can shore up peace processes and help craft more inclusive post-conflict states. Yet, these studies prioritize only women’s formal access to decision-making channels during the peace process, which is surprising given the infrequency with which women are granted explicit entry into these processes. Women also have the ability to influence these processes indirectly (Dayal and Christien 2020). If women make up less than 15% of peace delegations, yet 40% of conflicts that end in an agreement are women-inclusive (Anderson 2015), there is the potential that some of the heavy lifting is being done by women not in the room.

The literature has also largely focused on the role of civil society women in generating substantive reforms on gender issues (Anderson 2015) to the exclusion of other domestic groups of women, including those operating within violent parties. However, previous research emphasizes that practitioners prioritize the inclusion of combatants at the peace table, often to the exclusion of third parties (Paffenholz 2014). This may suggest civil society actors have a harder time gaining access to the peace process than women operating within armed groups. Consequently, rebel women may have significant means by which to influence the peace process since they constitute primary parties to peace talks. According to Dr. Anne Itto, a member of the Sudan People’s Liberation Movement’s (SPLM/A) peace delegation, “Women were never simply guests at the negotiating table. The roles they play as combatants, supporters of fighting forces and peacemakers qualify them to sit at the negotiating table and to assume an active role in implementation” (Itto 2006, 56). This potential impact for female combatants is examined here.

A significant body of research shows women are represented in rebellions across the globe, sometimes in large concentrations (Darden, Henshaw, and Szekely 2019; Henshaw 2016; Thomas and Bond 2015; Thomas and Wood 2018; Wood 2019; Wood and Thomas 2017). Globally, women have participated as fighters in 40% of rebel organizations (Wood and Thomas 2017) and as leaders in 28% of rebel groups (Darden, Henshaw, and Szekely 2019). Therefore, female rebels represent a nontrivial class of women already organized within preexisting institutions. These women also experience some of the most drastic alterations of traditional gender roles during conflict, which may offer them powerful incentives to seek inclusion in the peace process.

Given strong reasons to prevent the return of pre-conflict gender roles (Mazurana, Krystalli, and Baaré 2018), large cohorts of women within rebel groups are likely to push for gender reforms to be taken up during the peace process. Particularly, I expect rebel women to pursue negotiated settlements laden with gender-inclusive provisions to ensure that the commitments offered to attract their participation are upheld in post-conflict transitions. Below, I develop a novel argument that explains how female rebels’ conflict experiences and socialization can lead to the development of a gendered group consciousness that cultivates a set of shared goals. These goals then translate into collective demands rebel women advocate for during peace processes.

FEMALE COMBATANTS’ EFFECT ON GENDER-INCLUSIVE PEACE AGREEMENT PROVISIONS

Women are often motivated to participate in militant organizations by politics. Although personal motivations (e.g., revenge and retribution) and material inducements spur some women’s participation (Alison 2009), others are enticed by the idea that their involvement in violent politics can produce broad social and political changes in society (Kampwirth 2014; Viterna 2013). Rebellion has also been viewed as an avenue for women’s empowerment (Kampwirth 2002; 2014; Molyneux 1985; Viterna 2013); some women join explicitly to experience the emancipation promised by violent organizations (Mazurana, Krystalli, and Baaré 2018; Thomas and Bond 2015; Wood and Thomas 2017).

For example, although many women joined the CPN-M to fight against class oppression, a subset joined expressly to combat gender oppression (Parvati 2005). A survey of its female membership showed that nearly two-thirds of female Maoists identified class repression as the primary driver of their participation, while 16% claimed rebellion against gender discrimination as their main impetus (Parvati 2005). Moreover, recognition of women’s unique plight only grew as they participated, which led women to seek redress through the peace process; 22% identified equality between CPN-M men and women as one of the most attractive aspects of their participation in the People’s War, while 15% believed that sensitivity toward “women specific problem[s]” was required to maintain women’s continued support and participation within the movement (Parvati 2005, 5233, 5236). Participating in a rebel organization with other women can illuminate women’s concerns about their relative status as a group and inspire them to add these concerns to the list of grievances they are fighting for.

I argue that mobilization for war can help women identify shared interests through routine interaction and offers them the opportunity to mobilize in support of those aims. Even apolitical women become sensitized to gender disparities as they operate within conflicts (Enloe 2014) due to both experience and socialization. First, revolutionary organizations go to great lengths to raise the consciousness of their recruits (Enloe 2014; Kampwirth 2014). While political indoctrination processes increase women’s political literacy and awareness of political, social, and economic disparities in society, they can also lead to the awareness of political exclusion and discrimination on the basis of gender (Enloe 2014; Ortega 2015). Tamil women, for instance, gained both political and gender consciousness through their participation in the nationalist struggle (Enloe 2014). After recognizing the differential treatment of men and women, women began to mobilize to better understand the interrelationship between their oppression as Tamils and their marginalization as women (Enloe 2014).

Through regularized interaction in armed groups with other women, female rebels develop a group consciousness and begin to adopt concerns common to individuals within their networks (Gurin 1985). Awareness of group deprivation grows as women interact intimately and routinely with other female in-group members and recognize they possess shared experiences, values, and interests (Gurin 1985). This results in a shared-identity and sense of collectivity that facilitates women’s mobilization for change (Gurin 1985). According to Kelly and Breinlinger (1995), women’s collective mobilization is predicated on both the existence of a group identification and the belief that change will occur through collective action.

The experience of rebellion is rife with inequality. First, labor within most rebellions is divided along gender lines (Henshaw 2020; Thomas and Wood 2018). Even in egalitarian organizations such as El Salvador’s Farabundo Martí National Liberation Front (FMLN), women are less likely to be assigned high-prestige roles than their male colleagues, which has implications for women’s ability to advance within an organization’s ranks (Viterna 2013). Individual women are likely to observe these disparities, and collectively, women are likely to attempt to challenge these practices. Second, though not exclusive to women, sexual- and gender-based violence is often an inescapable part of a woman’s experience in rebel groups. In Namibia, El Salvador, and Colombia, female combatants were exposed to sexual torture, assault, and forced abortions by both rebels and governments (Herrera 2010; Olsson 2005). Relatedly, challenges associated with participation in resistance while bearing children affects women disproportionately (Herrera 2010). The lack of institutional support for women’s unique experiences gives rise to shared discontent that they are likely to attempt to address through collective action. Women’s willingness to organize for change during a peace process requires, however, that women not only recognize individual-level discrimination but view it as collective discrimination (Ortega 2015).

Former FMLN combatant, Morena Herrera admits that when she was first recruited, “being a women seemed of little relevance.” However, she realized during her decades-long participation that the experiences of male and female combatants were “marked by gender” (Herrera 2010, 292). Gender disparities were evidenced by the ways men’s and women’s roles and duties were assigned, how justice within the organization was applied and the recognition that combatants were afforded (Herrera 2010). Female participation in the armed struggle not only helped militant women become conscious of what they were denied, but enabled them to see what they were entitled to. Consequently, the disparities that women faced during conflict facilitated FMLN’s female combatants and broader female constituency’s mobilization under the Asociación de Mujeres por la Dignidad y la Vida (Las Dignas). The organization formed to draw attention to women’s specific needs and rights, which they continued to do during the peace process (Herrera 2010; Ortega 2015).

Ortega (2015) suggests the timing of female combatants’ collective organization can explain the degree to which women rebels are interested in and emboldened to push for their unique demands. Gaining access to a peace process early can affect how successful women are in pushing their demands forward (Tamaru and O’Reilly 2018). Colombian female combatants in Mouvement 19 de Abril’s (M-19) Mujeres de April were discouraged from focusing on gender equality during the armed struggle and were only mobilized as a coherent group around the opening of peace negotiations. Given their lack of previous organization, and failure to achieve group consciousness, female combatants did not advocate for a women’s agenda or attempt to push through agreement provisions that were specifically tailored to women’s interests (Ortega 2015). By contrast, where women organized and gained consciousness before the opening of negotiations, they had greater opportunity to advocate for provisions that benefited women during the peace process (Ortega 2015). FMLN outgrowths Las Dignas and Las Melidas were both formed years prior to El Salvador’s peace negotiations and worked to facilitate women’s collective identity formation during the peace process (Herrera 2010; Ortega 2015). Both pushed women’s interests during negotiations with the state, despite resistance from the FMLN. Moreover, Ortega (2015) argues that access to insurgent infrastructure may have aided in opportunities to effectively channel their demands. Where female combatants lacked the cover of rebel groups, they face more difficulties gaining legitimacy for their demands.

Perceiving an opportunity to affect change is as important as identifying women’s collective relative deprivation (Kelly and Breinlinger 1995). Groups are likely to be more successful when they can exploit points of access (Ellerby 2016; Htun and Ossa 2013). The peace process, where rebel demands are being considered, can provide a forum for female rebels’ demands. According to Yami (2010), “The peace process gave the opportunity [for] CPN (M) to institutionalize the issues they had been raising, and practicing during [the people’s war],” especially as it related to gender equity.

How Do Female Combatants’ Influence Peace Processes

According to Ariño (2008) and Bell (2004), the communication mechanisms between those taking part in conversations and those outside of negotiations can be more important than women’s token presence in peace talks. Essentially, women’s access is necessary to shape peace, but access can take on direct and indirect forms. I argue that when women constitute a coherent and vocal force within rebel groups, they may generate sufficient pressure to secure the inclusion of gendered provisions in peace accords, even if indirectly. This form of indirect representation has been used to broaden women’s participation in other political processes, including post-conflict constitution drafting (Tamaru and O’Reilly 2018). There are several means by which women may influence rebel demands during peace talks.

First, women become more central to an organization as they constitute a critical mass. When women comprise a sizable proportion of an armed organization, their ability to influence group politics grows and their voices become harder to sideline. I concentrate on the potential influence of female combatants since previous studies point to combat participation as a signal of women’s integration into a rebel organization and a measure of their ability to attract prestige and status (Thomas and Wood 2018, 3). Duties within rebel organizations are typically delegated along gender lines, with women taking on support work and men primarily engaging in combat duties. Therefore, when women also engage in combat, they are afforded equal opportunities to participate as men (Thomas and Wood 2018, 7). These roles also tend to be more respected. According to Kampwirth (2014, 9), in Chiapas, Mexico, “any woman who served in combat automatically enjoyed some prestige, given the glorification of violence that played a not so insignificant role in guerrilla culture.” As in other Central American rebellions (Kampwirth 2002), front-line combat duties were viewed as high-prestige assignments while support tasks were viewed as less prestigious in the FMLN (Viterna 2013). Rebel women in Sierra Leone, Liberia, Colombia, and Sri Lanka also sought out the esteem of participating as soldiers within rebel organizations (Coulter 2008; Herrera and Porch 2008; Jordan and Denov 2007).

Access is likely to accompany prestige and offer rebel women the ability to influence organizations’ substantive agendas during negotiations. Itto (2006) asserts that the Government of Sudan refused to accept a gender quota put on the table by SPLM/A’s female negotiators because they believed that women had not actively participated in fighting, suggesting women’s status as combatants can better position them to make demands during negotiations.² Women’s competence in combat roles can also help afford female rebels recognition; engagement in fighting helped women gain legitimacy and ultimately convinced the Liberation Front of Mozambique’s (FRELIMO) male leadership to integrate women’s concerns into the group’s platform (Katto 2019).

Second, rebels remain conscious of their levels of support and hesitate to engage in actions that marginalize sub-factions to the point of fractionalization (Mampilly 2011). Thus, militant organizations may decide to yield on women’s issues to mollify aggrieved factions within their ranks in order to maintain cohesion.³ Reluctant rebel groups will be more likely to accede to women’s demands, when women’s representation is tied to other forms of power and influence that groups rely on. Some armed groups intentionally recruit and organize women to further their own goals, including attracting international support and building legitimacy (Manekin and Wood 2020; Shekhawat 2015; Stack 2009; Wood 2019). Destacamento Feminino (DF), FRELIMO’s women’s wing, was formed to better mobilize the peasant population in support of the armed movement (Arnfred 2011). Female fighters in the EPLF and FMLN were also recruited and organized to broaden domestic support for the rebellion (Herrera 2010). Female visibility in the FMLN was intended to help the armed group garner legitimacy and international financial support (Ortega 2015).

While female visibility can be helpful for an organization’s viability, it can also tie its hands making it difficult to sideline women’s issues during negotiations. Once attention is brought to women’s concerns, the desire to maintain support from the civilian constituency, international community and civil society actors can compel armed groups to take women’s issues seriously. Noncombatant women can use female participation within rebellion as a justification for attention to women’s status and rights. Lohani-Chase (2014, 33–4) asserts that political and social changes occurred in Nepal because women’s representation within the conflict made it “no longer possible to ignore women as an electorate or as a constituent power within and outside the political parties.” Similarly, women in the Philippines’ Moro Islamic Liberation Front (MILF) eventually secured seats in negotiations because of pressure exerted by civil society (Ellerby 2016). Once seated at the table, MILF women used their opening to advocate for women’s rights (Ellerby 2016).

Women have also been able to gain traction on their interests by building broad coalitions across disparate groups of women (Tamaru and O’Reilly 2018). According to Htun and Ossa (2013), women must form coalitions to be politically effective. In many conflicts (e.g., Eritrea, Somalia, Israel/Palestine, Bosnia, and Northern Ireland) women have demonstrated their willingness to work across the aisle to push for greater attention to women’s interests, needs, and representation during negotiations (Cockburn 1998; Handrahan 2004; Karam 2000). Female combatants in countries like Sudan and Colombia have been able to push their demands further when they have collaborated with women outside of their organizations to draw attention to gender inequity. Women in the CPN-M engaged the group’s extensive network, particularly through the All Nepal Women’s Association, to push for change to Nepal’s inheritance laws, to create a Women’s Commission and to prohibit trafficking. Their many activities, including lobbying members of parliament and conducting sit-ins and marches, along with other forms of dissent, collectively “built up political pressure on the state” and informed women of their own rights (Tamang 2009, 68).

Finally, rebel women will find it easier to accomplish their objectives when their aims are compatible with the movement’s broader goals. Gaining support for provisions on gender equality is likely less complicated when women are mobilized within organizations already advocating for egalitarianism. In such cases, feminist efforts may facilitate the movement’s goals. According to Kampwirth (2014), the relationship between guerrillas and feminists in Chiapas, Mexico was symbiotic; women’s organizations influenced the Zapatista movement while the Zapatistas influenced the work of the feminists. However, a leftist agenda is not necessary, nor is it sufficient to guarantee that women’s rights will be taken up by rebel groups during the peace process. Enloe (2014) argues that leftist revolutionaries, such as the Vietnamese Indochinese Communist Party, often encourage women’s activism only to squelch their legitimate concerns about gender relations later.

Rebel Women’s Cross-Cutting Identities

In the preceding section, I argue that participation in militancy can help female combatants generate a group consciousness that both incentivizes and enables them to lobby for women’s political interests during peace processes. Below, I theorize about the specific types of demands rebel women make during the peace process. I argue the origins of female combatant mobilization and organization—first as rebels, then as a group of women—influence the types of concerns they voice. Female rebels articulate interests in line with their identities as members of politically marginalized groups and rebel groups. Therefore, peace agreement provisions will reflect these interests when female rebels are prevalent and influential.

Although scholars and practitioners often allude to women’s interests, conceiving of women’s interests around conflict in such broad terms presents a false homogeneity and neglects the complexity of their concerns (Molyneux 1985). Feminist scholars have long argued that women’s experiences are varied, diverse, and intersectional; and their interests and politics are dependent on their unique positions in society (Crenshaw 1990; Smooth 2011; Stephen 2007; Weldon 2011). Race, ethnicity, class, and caste are all salient identities that determine how women experience oppression, including gender discrimination. In this context, the shape of a women’s agenda may depend on which women are involved in the conflict and attendant peace process (Bell 2004).

Bell (2004, 99) argues that especially in divided societies, women’s identities, affiliations, and experiences shape their perspectives on conflict resolution. That is, what women view as viable solutions to conflicts will be predicated on their backgrounds and where they are located within society and war. Thus, agreements that make reference to all women will be inadequate for addressing the diverse needs of specific groups of women and will be unable to address inequalities between groups (Ní Aoláin and Rooney 2007, 341). Likewise, incorporating a few women into peace processes risks treating female negotiators like token representatives for all women and fails to appreciate women’s “multi-layered and diverse experiences” in conflict (Céspedes-Báez and Jaramillo Ruiz 2018, 96). It also takes for granted that any women could or would advocate for all women (Vastapuu 2021). On this point, Henshaw (2020, 66) asserts that gendered peace-building interventions often fail to acknowledge “the divides running between combatant (defined broadly) and civilian women and among diverse categories of race/ethnicity, class/caste, gender, religion, sexuality, and ability.” Treating women as a cohesive identity group belies the fact that women-centered accommodations seldom benefit all individual women or groups of women equally (Berry and Lake 2021; Giri 2021; Hughes 2011; Martín de Almagro 2018; Ní Aoláin and Rooney 2007; Weldon 2011). Thus, the specific constellation of pro-women provisions that are included in an agreement will depend on the quality of the women involved not the quantity (Karam 2000).

Female Rebels’ Intersectional Identities

John Garang, the leader of the Sudan People’s Liberation Movement/Army (SPLM/A), asserted that women were “the marginalized of the marginalized” (Itto 2006, 58). Not only were women within the group’s constituency marginalized because of their political and social affiliations, they were marginalized because of their gender. Likewise, Maoist women believed they were fighting a “two-line struggle” against class exploitation and patriarchy, thus experiencing “double oppression” (Parvati 2003). In Colombia, issues of gender were salient but so too were issues of age, class, ethnicity, and religion (Bouvier 2016; Céspedes-Báez and Jaramillo Ruiz 2018). In each of these cases, rebel women’s fight for gender equality was inseparable from their struggles for equality on other dimensions (e.g., race, caste, and class).

Gender is considered a cross-cutting cleavage that is often believed to span across political divides (Htun 2004). An intersectional perspective, however, suggests one’s gender does not override other salient identities, nor do women’s concerns about gender equity transcend partisan politics. Especially when women mobilize to advance specific political aims first and mobilize for their gendered interests second, women’s political identities are likely to shape the ways in which women’s demands are conceived and articulated. Since rebel organizations tend to recruit from disempowered groups in society, rebel women often hold marginalized identities. Consequently, I argue militant women’s advocacy efforts will concentrate on the incorporation of provisions that redress historical oppression of women from excluded groups in society.

Political and economic exclusion are potent tools for rebel recruitment (Cederman, Weidmann, and Gleditsch 2011; Gurr 1993; Ives and Breslawski 2022; Østby, Nordås, and Rød 2009). Civil conflict becomes likelier as a greater share of the civilian population is excluded from state power on the basis of ethnicity (Wimmer, Cederman, and Min 2009). Horizontal inequalities are associated with an increased likelihood of rebellion, as communities that are politically and economically marginalized rebel against dominant groups to remedy their exclusion (Cederman, Weidmann, and Gleditsch 2011). Like men (Humphreys and Weinstein 2008), women are pushed into rebellion by their ethnic, economic, and religious grievances (Henshaw 2016).

If political marginalization motivates the decision to rebel, female rebels are likely to hold identities as members of oppressed groups. Moreover, some rebel organizations go to lengths to educate potential recruits on prevailing political inequities, appealing specifically to groups that have been repressed by traditional political elites (Eck 2009). Such political education heightens recruits’ sense of relative deprivation, raises recruits’ consciousness about their own marginalization and publicizes the group’s demands and strategies for achieving change (Eck 2009; Weinstein 2006).

Given their personal experiences and political education, female rebels will be motivated to fight for the concerns of marginalized communities, but are likely to advance the interests of marginalized women in particular. These interests are likely to differ from those advanced by women that do not hail from politically, socially, or economically marginalized backgrounds. Women at the cross-section of two or more marginalized identities often experience these oppressions simultaneously and recognize them as inseparable, causing them to pitch their concerns more broadly than those facing discrimination on one dimension (Crenshaw 1990).

Jones (2020, 34) argues that Black women, facing “enslavement, sexual violence, economic exploitation, and cultural denigration” in the nineteenth and twentieth centuries, sought to improve their own circumstances by fighting to “realize the equality and dignity of all persons.” The intersection of their identities as Black people and women, and the attendant oppressions led them to frame their demands around universal human rights. Neither Black male abolitionists nor white female suffragettes cast their concerns in such broad terms.

Further Yuval-Davis (2006, 195) argues that racial oppression is always “intermeshed in other social divisions” such as “gender, social class, disability status, sexuality, age, nationality, immigration status, and geography,” and individuals that are disenfranchised on one dimension often experience oppression on another. During civil conflict, rebel women can experience marginalization on multiple dimensions simultaneously, causing them to advocate for broad notions of equality that account for their numerous, intersecting identities.⁴

The FARC recruited heavily from rural, Afro-Colombian, and Indigenous communities (Herrera and Porch 2008) and used their influence to address the concerns of people from these groups (Bigio, Vogelstein, and Connell 2017). As such, Farianas pushed for land restitution and an egalitarian distribution of property for rural women. They also used their platform to advocate for terms that would remedy the “oppressions that women, Black and Indigenous women, experience” and acknowledged the inherent human rights of marginalized, traditionally underrepresented, and vulnerable communities in Colombia (Simanca Herrera 2018). Similarly, women in the CPN-M sought to “address Dalit, regional, and ethnic oppression because they are interrelated to women’s oppression” (Yami 2010). Cadres from these groupings were believed to constitute up to half of the organization’s female recruits.

Tamaru and O’Reilly (2018, 2) find that when included in drafting post-conflict constitutions, “women repeatedly advocated for issues relating to gender equality and the rights of marginalized groups.” When women adopt identities as the largest group historically excluded from politics (Tamaru and O’Reilly 2018), it can broaden their identification with others that have been excluded from power on the basis of identity as well. The propensity for women to advocate for minority groups more often than men, may also be facilitated by women’s interest in creating coalitions with women from across identity groups, including those from perceived opposition groups. Female combatants have been willing to liaise with women from civil society, government and those unaligned, which may reflect the benefits of their perceived in-group status. Even when in coalitions, rebel women will possess a unique perspective, influencing the quality of resulting agreements; accords will better represent the interests of historically marginalized or excluded women. As a result, I expect the following hypothesis to obtain:

Hypothesis 1: The probability that a peace agreement will include a provision calling for equality for women increases with the proportion of female fighters participating in a conflict.

Female Rebels’ Militant Identities

Over time, female guerrillas internalize their identities as combatants (Tarnaala 2016) and attempt to pursue special protections for their class. Female combatants’ identities become salient through their participation in high-risk collective action. Armed groups often undertake significant efforts to socialize their recruits to generate a coherent fighting force (Cohen 2013). These attempts to bond and cohere a disparate group of recruits, often through group participation in violence, can serve to create a shared identity around female participation in militancy.

Viterna (2013) argues that guerrilla recruits often develop a coherent identity through sustained interaction with other group members; this new identity is forged through and reinforced by network connections. Individuals within rebel organizations develop a participation identity when their activism within a movement harmonizes with or advances their preexisting, salient identities (Viterna 2013). Moreover, when more recruits share similar identities, the guerrilla identity becomes inseparable from these other identities (Viterna 2013). With respect to Salvadoran women in the FMLN, Viterna anticipates that if a woman’s salient identities as a Christian campesina (peasant) “come to be aligned with the identity of a guerrilla, she may feel compelled to act in ways that sustain not only the identities of campesina and Christian, but also the identity of guerrilla” (Viterna 2013, 53). In this regard, female combatants not only develop strong identities as rebels they may also coalesce around a shared identity as rebel women. Rebel women, therefore, are likely to act in ways that afford this group status and push for special protections within peace agreements.

Molyneux (1985) proposes that rebel women tend to focus on their practical gender interests or immediate perceived needs. During a peace process, these practical gender interests are likely to revolve around post-conflict reintegration or rehabilitation. Rebel women often have post-conflict needs that differ from both civilian womens’ (Tarnaala 2016) and male rebels’ (Mazurana, Krystalli, and Baaré 2018). These needs can be marginalized by both. The men drafting demobilization plans rarely consider women’s needs, and the elite women occasionally invited to participate in talks have little incentive to advocate for benefits that do not pertain to them (Vastapuu 2021). Yet, researchers and practitioners have been vocal about the unique trials rebel women face when attempting to assimilate back into society when conflicts end (Coulter, Persson, and Utas 2008; Farr 2001; 2003; Henshaw 2020; Mazurana, Krystalli, and Baaré 2018).

Women and girls often struggle to return to their communities after having transgressed traditional gender roles by joining rebellions, participating in violence, and enduring trauma within these groups (Mazurana, Krystalli, and Baaré 2018). Women are stigmatized by presumptions about their relationships, behavior, and treatment during war (Coulter, Persson, and Utas 2008; Farr 2001). Rebel women often lack the resources and support to facilitate their transitions back into society (Mazurana, Krystalli, and Baaré 2018). Women often experience significant physical and mental health challenges that can impede their return to normalcy (Coulter, Persson, and Utas 2008; Henshaw 2020). Moreover, some women actively resist returning to their pre-war lives. Participation in rebellion can allow women to experience power and agency that they are not ready to relinquish when wars end (Mazurana, Krystalli, and Baaré 2018; Veale 2003). While male rebels may be integrated into government armies and are reborn as politicians, rebel women rarely enjoy these same options (Mazurana, Krystalli, and Baaré 2018). Their post-conflict roles seldom make use of the extensive skill-sets they developed during war, even though these credentials could be useful to society (Coulter, Persson, and Utas 2008).

Many of these concerns could be ameliorated by female combatants’ participation in DDR or rehabilitation programs. Many DDR programs include job training and education programs that can help former female combatants re-tool or locate jobs that utilize their skills (Jennings 2009). However, women are often excluded from the construction of these processes and are inhibited from participating in them (Mazurana and Carlson 2004); they are less likely to be rehabilitated or reintegrated in post-conflict societies than their male counterparts (MacKenzie 2009; Mazurana et al. 2002). Their exclusion from this process is sometimes encouraged by male commanders and institutionalized by the international community (Mazurana, Krystalli, and Baaré 2018).

While many female rebels are excluded, others select out of demobilizing. Scholars argue, however, that women are more likely to participate in demobilization processes when their interests are considered and taken seriously (Coulter, Persson, and Utas 2008). Women’s influence in devising post-conflict re-entry programs could ensure female combatants’ concerns are addressed (Farr 2003). Since the terms and conditions of DDR programs are decided during negotiations (Coulter, Persson, and Utas 2008; Mazurana, Krystalli, and Baaré 2018), female rebels may attempt to influence the uptake of gender-inclusive DDR programs when they perceive opportunities to do so.

According to Laura Cardozo, Gender Advisor for the FARC, Farianas were apprised of the historical treatment of demobilized women during a Norwegian government-sponsored meeting with other female ex-combatants from around the world. Cardozo stated, “The Salvadorans told us after the peace process they planned a return to civil life in mixed and impartial terms, but as they became a political party, the ex-guerrilla women were left behind. They began to take on the traditional gender roles while the men were the candidates for political office” (Ruiz-Navarro 2019). With awareness of the harsh reality of demobilized women from other contexts, it should be unsurprising that FARC women sought to protect their own post-conflict futures by pushing for a gender-inclusive demobilization and integration program and a separate reconstruction project that integrated the perspectives of militant women. Relatedly, SPLM/A women warned Darfurian women negotiating the Abuja agreement that their own interests had been sidelined during negotiations for Sudan’s 2005 Comprehensive Peace Agreement. Expectations of their marginalization incentivized women in Darfur to push harder to ensure a women’s agenda was reflected in the subsequent agreement (Itto 2006). I expect this behavior to constitute a broader pattern:

Hypothesis 2: As more female fighters participate in a conflict, the probability of a gender-inclusive peace agreement provision addressing women’s participation in DDR increases.

RESEARCH DESIGN

To assess the effect female rebels on the incidence of gender-inclusive provisions in peace agreements, I utilize data from the Political Settlements Research Programme’s Peace Agreements Database (PA-X) (Bell and Badanjak 2019), which codes agreements that resulted from over 150 peace processes globally. These data contain a larger set of agreements and more information on the terms than any other dataset. Its focus on gender-inclusive language is exemplary. The present sample includes all agreements associated with an intrastate conflict coded in the UCDP Armed Conflict Dataset between 1990 and 2019 (Pettersson and Eck 2018).⁵ The unit of analysis is the agreement, of which there are over 1,200 in these data.

Dependent Variable

In this article, I ask how does female rebel’s combat participation influence the adoption of particular gender-inclusive peace agreement provisions? To assess this question, I examine whether an agreement includes gender-inclusive provisions, as coded in the Women and Peace Agreements (PA-X Women, V2) database (Bell and Badanjak 2019).

The main hypotheses suggest that rebel women advocate for provisions that correspond with their identities as female fighters and women from marginalized society groups. To test the first hypothesis that female rebels push for provisions that provide for women’s equality, I include a binary measure, Equality, recording whether the agreement references social or general equality between between men and women or whether terms directly address the exclusion of women from traditionally marginalized societal groups. Such provisions include references to women’s inherent rights to equal protection under the law, general prohibitions on social, political, and economic discrimination and references to women from specific societal groups that face exclusion. The latter category includes references to Indigenous women, rural women, Afro-descendent women, and Dalit women, among others. Examples of such provisions can be found in the Supplementary Material.

Few peace agreements (14%) include provisions for equality; yet only a small subset of those agreements—fewer than one-quarter—make explicit claims about the importance of women’s equality. By contrast, more than half of the agreements that discuss societal equality make allusions to the inclusion or nondiscrimination of ethnic or racial groups. Therefore, even agreements that provide for equality between social groups and pinpoint specific groups that should be accommodated, do not necessarily consider women.

To test the second hypothesis that female combatants will be particularly concerned with ensuring they are included in post-conflict programs, I examine whether gender inclusive DDR programs are written into agreements. DDR provisions include terms that provide demobilization benefits to former female combatants and those that establish women’s rights to participate in restructured security forces. I also assess whether agreements include gender-inclusive plans for rehabilitation programs. Gender-inclusive rehabilitation provisions establish plans to rehabilitate or provide for social/economic reconstruction for women in society. These provisions include plans to restructure the state in a way that takes into account marginalized women’s needs as well as provisions establishing aid and support for women from war-affected communities. Such provisions should, in theory, be extended to women who are demobilized from fighting forces but not exclusively. In fact, of the 50 times where DDR or rehabilitation provisions are present in an agreement, they co-occur in only six cases (12%), suggesting they are substantively different. Both types of provisions are rare; only 2.5% and 2.2% of all agreements include provisions for DDR and rehabilitation, respectively. Using data on rehabilitation programs ensures the results do not merely reflect a correlation between female combatants and women-inclusive DDR because there exist women fighters to be demobilized. In other words, DDR programs may be gender-inclusive because women are engaged in conflict, not because of their influence and advocacy. Rehabilitation provisions, on the other hand, do not require women’s participation in rebellion.

I combine provisions for gender-inclusive DDR and rehabilitation programs into a single binary variable, Reintegration. Detailed examples of these categories can be found in the Supplementary Material.

Independent Variable

I proxy female combatants’ representation within rebel organizations by examining the extent to which women serve in violent roles. Although women’s broader participation is important, not all women have the capacity to influence an organization’s substantive agenda. Not only must women make up a critical mass, they must also be influential. Participation in combat roles suggests women occupy central roles in a fighting force, while the scope of their participation reflects whether women constitute a meaningful contingent of the group, making them indispensable to a war effort. The primary independent variable, Female Combatant Prevalence, estimates the proportion of women in a rebel organization’s fighting force (Thomas and Wood 2018). This ordinal measure is coded “0” when women do not participate in combat, “1” when female participation in fighting forces is low, “2” when female combat participation is moderate, and “3” when there is a high proportion of women fighters in a conflict.

Since the unit of analysis is the agreement, which can be signed by a number of groups involved in a peace process, I aggregate the female combatant measure to the conflict-level and code the highest level of female participation within any group involved in the conflict. Twenty-one percent of agreements are signed in the context of a conflict with no female combat participation. Forty-two percent of agreements are penned where there is a low prevalence of women soldiers. Eighteen and nineteen percent of agreements are signed in contexts where there is high and moderate proportion of female fighters, respectively.

Control Variables

Existing research finds that women’s civil society and women’s transnational advocacy groups are largely responsible for instigating gender inclusive provisions (Anderson 2015). Therefore, I include measures that account for the strength of domestic and international women’s organizations. I account for the strength of civil society women’s movements in two ways. First, I employ Murdie and Peksen’s (2015) data on the Number of Anti-state Women’s Protests, which records the number of protest events where women are identified as primary actors engaged in acts of nonviolent resistance against the state. Second, I include, V-Dem’s Women’s Civil Society Participation index (v2x_gencs), which records the extent to which women are able to form and participate in civil society organizations and express political ideas freely. This measure ranges from 0 to 100, where “0” corresponds with low participation, and “100” denotes the converse. To assess the activity of international women’s groups, I use data on women’s transnational advocacy organizations from Hughes et al. (2018). These data measure country-level membership for women’s international nongovernmental organizations.⁶

To ensure the female combatant variable does not simply proxy women’s participation at the negotiating table, I include Women Signatories, which records whether any women from rebel groups, government or civil society held formal roles such as agreement signatories in a process. Based on Krause, Krause, and Bränfors (2018), I code women’s participation as signatories for all agreements signed in the Democratic Republic of Congo in 2003, El Salvador in 1992, Guatemala in 1996, Liberia in 2003, Papua New Guinea in 2001, and Northern Ireland in 1998.

The international community also plays an important role in pressuring or helping belligerents adopt gender-inclusive language in agreements. UN women was established in 2010 to assist states in implementing UNSC 1325, which created legal obligations for conflict actors to take gender and women into account during conflict resolution processes. The institution provides guidance and information, building capacity and increasing opportunities for women to participate in peace processes. To assess if agreements are more gender-inclusive after the availability of these resources, UN Women is coded 1 for all agreements signed in or after 2010; any agreements signed before this date are coded 0. I also include binary variables measuring whether the United Nations or individual Major Powers are formally involved in the peace process. Such involvement was coded only when recorded in the text or preamble of the peace agreement.

I also include variables recording the ideology of participating rebel organizations, as existing research finds leftist organizations recruit female soldiers more often, whereas religious organizations do so less frequently (Wood and Thomas 2017). Including a measure for leftist ideology helps alleviate concerns of omitted variable bias since an egalitarian philosophy may confound the relationship between female combatants and gender-inclusive peace agreement terms.

Additionally, longer agreements may be more likely to include any type of provision, whereas substantive agreements may cover any type of topic, including those on gender. Therefore, I include variables measuring the page length of an agreement (Agreement Length) and a binary measure of whether the agreement could be considered a Substantive agreement as opposed to a ceasefire agreement, renewal or renegotiation, among other types of agreements. Equality Provision and DDR provision control for the inclusion of equality, and DDR provisions that make no clear reference to women. Information on all of these agreement features comes from the PA-X dataset.

Finally, to ensure that gender inclusion in peace processes is not just an extension of women’s inclusion in or exclusion from society, I include a measure of women’s civil liberties from V-Dem, which measures the extent to which women have the right to domestic movement and private property, are free from forced labor, and whether they have equal access to justice (Coppedge et al. 2021).

STATISTICAL RESULTS

Since the two response variables are binary, I use logistic regression to evaluate the hypotheses. Robust standard errors are clustered on the country to address potential dependencies between agreements. In Table 1, Models 1 and 3 show the relationship between female combatant prevalence and gender-inclusive peace agreement provisions without controls, whereas Models 2 and 4 include relevant control variables.

Table 1. Logistic Regression Analyses Examining the Effect of Female Combatant Prevalence on Gender-Inclusive Peace Agreement Provisions

Note: Standard errors clustered on country in parentheses; * $ p<0.10 $ , ** $ p<0.05 $ , *** $ p<0.01 $ .

Models 1 and 2 tests the first hypothesis, which suggests that as the volume of female combatants within a conflict increases, the probability that an agreement will include a provision advancing women’s right to equality increases. The findings offer strong support for this hypothesis. In particular, the results show that female combatants exert a positive influence on provisions for general gender equality. As the proportion of female combatants increases, peace agreements are more likely to include terms for women’s equality.

Models 3 and 4 examine the relationship between female combatant prevalence and the incorporation of gender-inclusive DDR and rehabilitation programs. The intuition is that the development of strong identities as female combatants push rebel women to prioritize the unique post-conflict needs of women that served in fighting forces. This is likely to manifest as clear programs designed to rehabilitate and reintegrate rebel women, which can smoothen their transition from conflict to post-conflict peace. The results lend support to Hypothesis 2. When rebel forces comprise a greater proportion of women, peace agreements are more likely to include terms providing for gender-inclusive demobilization and reintegration programs and separate rehabilitation programs.

Figure 1 plots the substantive effects from Models 2 and 4, respectively, with 95% confidence intervals.⁷ Figure 1a shows that when female combatants are missing from an organization’s fighting force, there is a slightly greater than 5% chance of a gender-inclusive equality provision. The probability of a gender inclusive equality provision more than doubles when the prevalence of female combatants is high. While the incidence of these gender-inclusive provisions appears relatively low, the probability of observing provisions that do not explicitly reference gender is also low. For perspective, the likelihood of observing any provision for equality is 15%.

Figure 1. Predicted Probability of a Gender-Inclusive Peace Agreement Provision across Values of Female Combatant Prevalence

Note: 95% confidence intervals shown.

Figure 1b displays the predicted probability of observing either a gender-inclusive DDR or rehabilitation provision at different values of female combatant prevalence. When female combatants are absent from a conflict, the probability of a provision calling for women’s reintegration into society is 0.01. At low levels of women’s participation in rebel fighting forces, the probability of a rehabilitation provision accounts for women’s interests is 0.02. The probability of observing a women-inclusive rehabilitation provision is 0.04 and 0.08 for high and moderate female combatant prevalence, respectively. Overall, there is a 30% chance of observing a DDR provision that does not make explicit reference to women.

This article argues rebel women have interests that inform their demands during the peace process. In particular, while female combatants are expected to care about their narrow self-interests as rebels, they also make broader demands given their identities as women from marginalized backgrounds. The results in Table 1 support this contention by examining whether agreements mention broad notions of equality when female combatants are more prevalent. One concern may be that the variables in the main analyses are overly inclusive and too broad to provide a precise test of this argument. To address this concern, I recode the dependent variable in a number of ways.

First, I consider only provisions that make reference to women from specific identity groups in society. I then code whether this subset of provisions clearly references terms such as discrimination, exclusion, marginalization/marginalized, minority, or inclusion. Thus, these provisions examine discrimination or exclusion of specific groups of women. Second, I reconceptualize the measure to include only provisions that reference discrimination and women’s inclusion simultaneously. Finally, I consider the additional implications of female combatants advocating for marginalized women. I examine whether terms propose development support or resources for women from marginalized or underdeveloped regions, recognizing their past political and economic exclusion. These results, which can be found in Section 2.1 of the Supplementary Material, demonstrate that a one-unit change in female combatant prevalence increases the probability that any of these provision types are incorporated into agreements. Interestingly, female combatants exert the greatest influence on provisions for women-inclusive development. These analyses offer additional support for the argument that female combatants address grievances of marginalized groups of women through their advocacy.⁸ Additionally, when Equality is restricted to include only calls for equity and inclusion for particular groups of women, the results remain consistent.

Control Variables

There are a number of control variables that influence the degree to which peace agreements are inclusive. First, the results show that other domestic and international women’s groups influence the incorporation of gender-inclusive peace terms. The results in Table 1, Model 2 confirm that women’s transnational advocacy networks and women’s civil society activity, proxied by women’s nonviolent movement events, both have a hand in producing provisions that call for equality for women. When domestic civil society strength is measured by V-Dem’s civil society participation index, there is a negative effect of civil society on gender-inclusive equality terms. These conflicting results suggest these two variables may measure different things. While women’s participation in peaceful activism captures women’s participation in practice, V-Dem’s civil society participation measures perceptions of women’s de jure rights.

Surprisingly, whether women are among the signatories of peace accords has no impact on the presence of a gender-inclusive equality provision. International women’s transnational advocacy networks increase the probability of observing a gender-inclusive equality provision. Other types of international community involvement are less influential. In particular, equality provisions are no more likely when major powers or the UN are formally involved in the negotiation process. Agreements that were signed after the United Nations established UN Women in 2010 are no more likely to include gender-inclusive terms calling for women’s equality. Additional analyses also show that agreements signed after the Beijing Declaration in 1995 and UN 1325 in 2000 are no more likely to include these specific terms either. This is consistent with previous studies that suggest the international community is more likely to support an apolitical women’s agenda during negotiations (Anderson 2015; Martín de Almagro 2018).

Rebel group ideology also matters. Agreements signed by leftist rebels are more likely to call for equality for women, while a religious ideology does the converse. Importantly, this result shows that, on average, leftist organizations are not merely using the rhetoric of equality to attract female recruits; they also tend to advocate for equality in practice. However, after the effect of ideology is considered, female combatant prevalence still has a statistically significant effect on the inclusion of provisions calling for equality for marginalized women, suggesting female combatants’ efforts may have a separate impact from that of their organizations.

Many other features of agreements are salient. Long peace agreements and those with a substantive focus are more likely to include provisions paving the way for equality for women, as are agreement provisions that call for equality without respect to gender. The existing level of women’s civil liberties does not affect the chance of seeing terms for women’s inclusion.

The effect of the control variables differ between Models 2 and 4, which underscores the importance of considering the inclusion of specific terms of agreements, rather than focusing on the inclusion of any-gender inclusive language. In contrast to Model 2, both measures of women’s civil society have a negative effect on the addition of gender-inclusive DDR terms. Only women’s peaceful activism is significant, however. This result supports the argument that civil society women are less likely to advocate for the interests of female combatants (Vastapuu 2021). Women’s participation as agreement signatories also has a negative but statistically negligible effect. There are varying effects for international community involvement. In both models, transnational advocacy networks increase the probability of gender-inclusive reintegration or rehabilitation provisions. Major powers are more likely to push for women-specific programs devoted to DDR in peace agreements but the explicit involvement of the UN does not appear to influence the uptake of these programs. Whether the UN was a formal party to the agreement or whether the treaty was signed after the establishment of UN Women has no bearing on these terms.

Since the demobilization of female fighters was an explicit aim of UNSC 1325 and subsequent resolutions, this finding is perplexing. However, when changing international norms are measured by whether the agreement was penned in the wake of UNSC 1325, there appears to be a stronger effect of international pressure. In contrast to equality provisions, agreements signed with leftist groups are no more likely to provide for female combatants’ reintegration and rehabilitation than those without such groups. Agreements signed with religious groups are still less likely to include such provisions. Consistent with the findings in Model 2, longer texts and substantive agreements are more likely to include women-inclusive reintegration and rehabilitation provisions. Agreements including DDR terms that are not gender-specific are also likely to include gender-inclusive ones. Again, women’s civil liberties do not affect the probability of observing terms that make the way for women’s inclusion in DDR or rehabilitation.

Robustness Checks

I conduct a number of robustness tests to ensure the main results hold under different specifications. All additional analyses mentioned in this article can be found in the Supplementary Material.

First, I disaggregate each of the two outcome variables into its component parts. I display results disaggregating provisions that address historical exclusion of marginalized women from those that advocate for general equality for women. I also separate gender-inclusive DDR provisions from those for rehabilitation programs. I recode the main explanatory variables, female combatant prevalence into a dichotomous measure and consider the effect of female rebel leaders. I also show that changes to the sample do not impact the interpretation of the results. I subset the sample to include only substantive agreements, those with ethnic dimensions and agreements in the UCDP Peace Agreement dataset (Högbladh 2019). I also conduct a number of tests to address dependencies between the agreements including alternative clustering and controls for the number of previous agreements in a process. I also assess the significance of several measures of gender equality measured both at the start of the conflict and just prior to the agreement’s signing. Importantly, measures of civil liberties, civil society participation and women’s empowerment fail to explain the scope of women’s combat participation and do not explain the relationship between female combatants and gender-inclusive provisions. This is consistent with Paffenholz et al.’s (2016) findings that female participation was more impactful in countries with lower scores on gender equality.

Finally, I consider alternative measures of international pressure, including the amount of official development aid (ODA) a country received from the OECD’s Development Assistance Committee (DAC) as well as the total amount of DAC’s “aid activities targeting gender equality and women empowerment” to capture whether conflict actors are being pressured into gender reforms. The main results are robust to these specification changes.

CONCLUSION

This article examines female combatants’ effect on the incorporation of gender-inclusive peace terms. I argue female rebels’ interests and experiences incentivize them to push for provisions in line with their identities as women from marginalized groups and fighting forces. The findings offer empirical support for this contention; female combatant prevalence is associated with peace terms advocating for the equality of marginalized groups of women and gender-inclusive DDR and rehabilitation programs. They appear to push for terms that are sensitive to the needs of violent women and advocate for provisions that smoothen their transition from conflict to peace. Rebel women also lobby for the inclusion of provisions supporting equality for marginalized groups of women, given their recruitment from marginalized communities.

Consequentially, these findings support existing arguments emphasizing the importance of considering the unique interests of different groups of women; these disparate preferences can lead to distinct agreement outcomes. The results show, for instance, that there are few advocates for gender-inclusive DDR programs aside from female combatants. Neither participation by women’s civil society, women signatories, nor UN involvement are associated with provisions for DDR programs that include women. These terms are also rare. If demobilization, reintegration and disarmament is an international imperative, as suggested by UNSC 1325, including rebel women’s perspectives in peace processes is important, as their inclusion may be one of the few means by which these provisions become endemic. In 2020, UN Women declared that “understanding intersecting forms of vulnerability and discrimination” is essential to the realization of the WPS, Sustainable Development Goals, and Leave No One Behind agendas.⁹ Therefore, better understanding the genesis of provisions that redress the grievances of marginalized women can add value to academic and policy discussions.

While the findings suggest rebel women are capable of influencing the outcome of peace processes when they constitute a larger contingent, the degree to which they are able to influence the terms of peace is likely contingent upon female combatants’ level of organization as well as their avenues for participation and influence afforded by rebel organizations. Female rebels may have more influence when they ascend to higher rungs on the political ladder in large numbers, though preliminary analyses show that the existence of female rebel leaders has no positive impact on the incorporation of the gender-inclusive provisions examined here. This is consistent with research on female state leaders (Caprioli 2000) showing that token women do not always change the norms of militaries. Other aspects of rebel organization and structure may also facilitate rebel women’s effectiveness in pushing for gender-inclusive changes. Participatory governance structures may provide increased access for female combatants (Israelsen 2018), as they can provide clear mechanisms for input. Rebel women’s cohesion is also likely to be greater when organizations include formal channels for them to mobilize. Rebel women will more readily establish a collective identity when they have a measure of autonomy and institutionalization. Therefore, examining how women’s organization into wings within rebel organizations influences their ability to push for gender inclusive peace agreements may be worthwhile.

This study advances our understanding of the way in which rebel women’s intersectional identities influence their advocacy efforts in peace processes. However, scholars have established that individual women within rebel groups have different lived experiences, which inform their interests (Berry and Lake 2021; Giri 2021). Therefore, generalizing women’s interests even in sub-groups likely glosses over important nuance. Future work should examine the ways heterogeneity in rebel recruits may influence female rebels’ ability to develop group consciousness or organize effectively.

Additionally, it is important to appreciate the limitations of rebel women’s activism even where they are able to get gender-inclusive terms on paper. This study focuses solely on whether agreements include gender-inclusive peace terms. However, there are foreseeable hurdles that can affect the ultimate realization of these concessions. When women are demobilized, it is not obvious that they will possess resources to ensure the implementation of agreement terms. Although female ex-combatants may continue to mobilize for their rights post-conflict (e.g., Las Dignas) their impact on implementation remains unclear. Second, although female rebels’ participation increases the odds that these important gender-inclusive provisions are integrated into agreements, the overall probability of these terms remains low, suggesting the difficulty of pushing these terms through. Even if rebel groups are amenable, other stakeholders, including governments and mediators, may veto their inclusion. Rebel groups may also fail to advocate for women’s interests, even when women comprise a large contingent of the organization. Future research should investigate the determinants of the implementation of gender-inclusive terms, paying careful attention to whether they are instituted at the same rate as other provisions and the roles former female combatants play in this process. Research should also consider potential barriers erected by other veto players in both the establishment and implementation of these provisions.