Peacekeeping Prevention: Strengthening Efforts to Preempt Conflict-related Sexual Violence

Introduction: Enhancing Dialogue and Research

Louise Olsson and Angela Muvumba Sellström

Peace Research Institute Oslo (PRIO) and Uppsala University

Since the adoption of UNSCR 1820 in 2008, United Nations peacekeeping operations have come under increased pressure to prevent conflict-related sexual violence (CRSV) conducted by local security actors. Still, the outcomes from peacekeeping preventive actions are reported to often fall far short of public expectations.¹ To address that problem, there is an ongoing debate on how to enhance peacekeeping operations’ effectiveness. For example, the UN Security Council held a Debate on CRSV on 23 April 2019 with the Nobel Peace Prize laureates Ms. Nadia Murad and Dr. Denis Mukwege speaking before the Council. The resolution adopted, UNSCR 2467, stated that the Security Council ‘[r]ecognizes the need to integrate the prevention, response and elimination of sexual violence in conflict and post-conflict situations […] including in relevant authorizations and renewals of the mandates of peace missions through the inclusion of operational provisions’. In preparations for the anniversary of UNSCR 1325 in 2020, the first resolution to recognize that violence targeting women is relevant for international peace and security, further calls on the UN to improve its track record of preventing CRSV is a critical theme. At the practitioner level, military forums are discussing how to clarify military responsibilities and improve their practical contributions to prevention.² As noted by Lotze in this Forum: ‘Consolidating the gains made to date, and building on these for the future, will be key if the Security Council’s call to bring a total halt to CRSV is to be realized’ (see page 27).

Engaging with this debate on how to strengthen peacekeeping prevention of CRSV, this Forum explores the role of uniformed, primarily military, peacekeeping contingents and their interactions vis-à-vis local state- and non-state military organizations. As noted in the entries, state militaries tend to be more commonly represented among the perpetrators. This can constitute a specific challenge given the relationship between the host state and the peacekeeping operation.³ All entries in this Forum do, however, consider CRSV as preventable, here defined as systematic, proactive and/or reactive effort(s) to limit the perpetration of conflict-related sexual violence.⁴ While it is a frequent feature of many wars, research clearly demonstrates that CRSV is not an inevitable consequence. Many warring parties choose not to use it as a strategy or policy and simultaneously seek to prevent their members from engaging in its practice. This understanding of prevention thereby means that we discuss what research says are risk factors for military organizations’ carrying out, or allowing CRSV (see entries by Moncrief and Wood, Hoover Green, and Muvumba Sellström in this Forum) and how traits and practices of military peacekeeping contingents relate to such factors (see Johansson, Ruffa, Sjoberg and Lotze).⁵ We further take as a starting point that when severe and persistent CRSV does occur, there is variation in form, targeting, and frequencies between and within warring parties and conflicts. Such variations and nuances mean that peacekeeping policy can learn from the warring actors’ successful and unsuccessful efforts to control and limit this violence.

Enhancing Dialogues on CRSV and Peacekeeping Operations

To support research and practical learning, this Forum brings together leading researchers on CRSV and on peacekeeping operations in dialogue. This is important, as few research projects on peacekeeping operations thus far have directly studied CRSV.⁶ In engaging with the policy debate, the Forum seeks to deepen the dialogue between researchers and key decision-makers, including contributions comprising direct experiences of working to prevent CRSV. This is critical as there appears to be a widening gap between a quickly developing systematic research front and the dominant policy narrative in two respects.

First, we are currently seeing a rise in the production of high-quality evidence and results relevant for peacekeeping operations and CRSV (see Moncrief and Wood and Johansson in this Forum for an overview). These results can be used to enhance political will and for formulating targeted mandate objectives which can contribute to successfully preventing violence. That said, recent results have additionally entailed that research now takes issue with some of the more fundamental policy statements on CRSV. Most notably, research remains hesitant to the persistent labelling of CRSV as a ‘weapon of war’ and with the claim that CRSV is primarily driven by societal gender inequality. As demonstrated in this Forum, empirical research often finds the opposite; while CRSV occasionally can be ordered as a weapon, it is generally more likely to be tolerated or to be perceived as impossible to stop by a weak military leadership.⁷ This means that peacekeeping operations need to increasingly focus on understanding the features of the groups responsible and to be able to differentiate between CRSV which is the result of leadership orders (a strategy/policy), and CRSV which is the result of the leadership’s lack of control (a practice).⁸ Importantly, as the Forum will show, the fact that sexual violence can occur in different forms and varies across different types of actors as well as within armed organizations in the same geographical area makes the explanation of societal gender inequality unlikely.⁹ We agree that support to gender equality is imperative and can serve to provide vital support to the voices of victims of violence; central for strengthening processes where women survivors mobilize for rights and resources.¹⁰ But research does clearly suggest that only focusing on gender equality might not be the most effective peacekeeping strategy for preventing further CRSV by a warring party.

Second, concurrently, peacekeeping prevention policy has been shaped by fairly extreme instances of CRSV as interventions have had to prioritize responses to atrocity.¹¹ This has entailed gathering empirical information about the violence committed in detention camps by Bosnian Serbs, the Interhamwe during the Rwandan genocide, the myriad militia and state actors in the DRC, and, more recently, the Islamic State of Iraq and al-Sham Islamic State (IS). These are very serious situations. However, in addition to these, research clearly shows that only a minority of armed political and security organizations commit widespread or pernicious sexual violence in conflict. While there are always difficulties to document and count CRSV,¹² Cohen and Nordås, covering 129 active conflicts involving 625 armed actors for the period 1989–2009, find that 43% of the individual conflicts had no reports of this form of violence.¹³ Muvumba Sellström’s events-based dataset between 1989 and 2011 of 23 armed actors in sub-Saharan Africa,¹⁴ shows that only eight actors were reported as responsible for 68% of abuses and assaults. While policies to anticipate violence through peacekeeping mandates and intervention practices have thus originated in models of pathology, it is important to now continue to move toward including lessons on institutions, norms, and preferences that exist where there are robust measures against sexual violence. This move toward nuance further underlines the need for continuously improved collaboration between researchers, policy-makers and practitioners in order to collect and analyse better data.

This Forum engages in dialogue with these two contentions in several important respects. The first three contributions set the frame by outlining key findings from systematic empiricist research on CRSV (Moncrief and Wood) and on peacekeeping operations (Johansson), and provide insights from ongoing operations (Lotze). Johansson notes that it has been nearly twenty years since Skjelsbaek observed that CRSV poses a particular challenge to peacekeeping operations. Johansson suggests that (a) a more systematic approach is needed in the response to CRSV; and that (b) peacekeeping operations are not as effective in preventing CRSV as they are in stopping the killing of civilians. To address this, Moncrief and Wood argue that there is a need to better understand the military organizations responsible for CRSV – their culture, structures, ideology, leadership and social norms in relation to training – before effective preventive measures can be developed (a point backed up by Ruffa, Hoover Green and Muvumba Sellström). Moreover, an analysis in preparation for an operation cannot assume civilian women victims. We need to take seriously that men and boys, and sexual minorities, are also targeted and we need to consider the risk for intra-force victims (see also Sjöberg and Lotze).¹⁵ Lotze, drawing on research and his own peacekeeping experiences, thereafter provides key insights on these dynamics from the internal viewpoint of UN peacekeeping operations in the Democratic Republic of the Congo (DRC) working in cooperation with the DRC Government to prevent CRSV. Lotze outlines a two-tracked approach: (1) by conditioning that elements of the state armed forces which committed violations against the civilian population would not be eligible for UN support; and (2) by providing assistance to the government’s military justice system in the fight against impunity, including a focus on holding higher ranking officers responsible.

Two entries then scrutinize what key traits the military peacekeeping contingents bring with them when deployed – military culture and understanding of professional responsibilities (Ruffa) and command and control and discipline (Hoover Green) – mean for prevention of CRSV, finding common concerns relevant for also addressing sexual exploitation and abuse (SEA). Ruffa’s commentary underlines that preventing CRSV has never been seen as a core task of most state military organizations. This means that the understanding of the problem, the sense of responsibility, as well as the practical capacity are likely to be low among many peacekeeping contingents when they are deployed. But while military contingents bring additionally potential problematic traits with them to the field, it can be difficult to realize prevention of CRSV through other channels. Hoover Green argues that CRSV more frequently is the result of a failure of military organizations to uphold command and control rather than it being a strategy. This means, she argues, that core military traits affecting the military personnel’s expectations about detection and discipline can drive the perpetration of unordered sexual violence. As suggested by both Ruffa and Hoover Green, formal and informal learning between military organizations – peacekeeping contingents and the host state’s military organization – could constitute a venue for prevention but only under the condition that it is used in an aware and conscious manner by the peacekeeping contingents.

The Forum further nuances this discussion by empirically illustrating prevention dynamics in local non-state military organizations. Muvumba Sellström offers insights from two different rebel groups from Burundi’s civil war (1994–2008), the CNDD-FDD and Palipehutu-FNL, which despite their many similarities, developed different approaches to CRSV. This resulted in one creating a more permissive environment and the other establishing a preventive culture. The result was different levels of CRSV conducted by these groups. Drawing on her own experiences from working with armed non-state actors to create such preventive regimes, Sjöberg suggests that there often is a need to improve the responsible organization’s understanding of international humanitarian law and support the development of codes of conduct and rules of engagement in order to be able to prevent CRSV. Such military documents, she claims, have to be owned by the group itself. In creating space for such policies, broader discussion on both gender equality and sexual violence have proven important for progress in spite of expectations of the opposite.

Finally, a common underlying theme in the Forum is the discussion of how efforts for peacekeeping prevention of CRSV should be understood in relation to the prevention of other forms of violence, such as, SEA (Ruffa and Hoover Green), and sexual violence more broadly (Sjöberg). Kishi makes a central contribution to this theme by arguing that it is important for a peacekeeping operation to better understand trends in many forms of violence targeting women and how these relate to women’s political agency. This allows for (a) more contextually appropriate and comprehensive protective measures which enable continued participation, and (b) ensures that preventive peacekeeping measures avoid contributing to a stereotyping of women as only victims of CRSV.¹⁶ For example, in addition to CRSV, abductions, forced disappearances, and mob violence, disproportionately affects women. The gender aware conflict analysis called for by the UN Secretary General in his 2019 report on Women, Peace and Security (S/2019/800) is consequently central.

Future Research: Moving Forward on Peacekeeping Prevention

By engaging in dialogue between research fields and with policy, the authors in this Forum collectively provide additional essential insights for developing research on peacekeeping prevention, that is, systematic, proactive and/or reactive effort(s) to limit the perpetration of conflict-related sexual violence, beyond this Forum. The shared insights and exchanges contribute in two respects: (a) they underline the need to move toward understanding prevention as a multi-level, multi-actor interactive process which is more comprehensive than fighting impunity; and (b) we need to develop indicators which offer insights into ongoing internal processes of military organizations in order to estimate risk and to understand how to effectively prevent CRSV.

Let us look closer at these two aspects. First, impunity is much in focus of current efforts. We agree that enacting legal recourse and justice are critical. However, the Forum demonstrates that prevention can be thought of as a multi-level, multi-actor process involving the peacekeeping operation interacting with the host government and local security forces. We claim that prevention involves removing the root causes through these interactions by altering the conditions in the organizations that precipitate CRSV. Practically, in the short term, peacekeeping prevention means promoting the exchange of information between military organizations in order for them to develop the capacity to effectively interact on CRSV prevention, create and guard humanitarian safe spaces, and other efforts which can reduce opportunity. Such efforts need to enhance the understanding of diverse victim groups, including men and boys. As this Forum has demonstrated, there are additional long-term measures needed. These should clarify and integrate the responsibility for preventing CRSV in the military profession – in countries contributing to peacekeeping operations and in those hosting operations – and involve discussions on military culture and training. That state military organizations deploying troops to a peacekeeping operation practice at home what they preach in theatre has to be a cornerstone. This is imperative for credibility and for overcoming state-to-state sensitivities.

Second, as all multi-actor processes, implementing prevention is contingent upon the nature and position of the involved organizations and their role within a conflict. To contribute to this continuous adaption to context, this Forum suggests that it is difficult to discuss prevention solely based on reported incidents of violence. Such reports are difficult to interpret when analysed separately from other indicators. For example, reports of violence might indicate either a rampant problem or an established accountability mechanism. Thus, this Forum proposes that we should capture prevention by combining formal and informal indicators. Formal indicators might include adopted codes of conduct; political education messaging; documentation to name and shame, such as listing credibly suspected perpetrators; sanctions; and penalties and judgments. These demonstrate active efforts by an organization to stop or deter violence. Informal indicators seek to capture social norms which can involve what can be construed as sexual misconduct and include sexual symbolism that disparages sexual coercion; or socialization among peers that promotes sexual cultures of consent and stigmatizes sexual predators. A combination of formal and informal indicators, we argue, would make it possible to create typologies of particular regimes of prevention and then to systematically examine their respective potential impact on the level of CRSV. Thereby, research and policy can jointly contribute with knowledge relevant for meeting the Security Council’s objective of eliminating CRSV.

Understanding Conflict-Related Sexual Violence: What Peacekeepers Should Know

Stephen Moncrief and Elisabeth Jean Wood

Yale University

Conflict-related sexual violence (CRSV) poses a difficult challenge for peacekeeping operations in the aftermath of armed conflict. Sexual violence by combatants, or former combatants, may continue after conflict’s supposed end. It often leaves profound legacies of displaced families, transformed communities, and ruptured social networks in its wake.

That rape and other forms of sexual violence by combatants during and after war now loom large on the peacekeeping agenda is an achievement of the international women’s movement, one largely driven by the narrative that, when frequent, rape during war is a strategy.¹⁷ Peacekeeping operations are increasingly tasked with preventing and addressing CRSV; to do so effectively requires understanding its patterns and drivers. Some armed actors do purposefully adopt rape as a strategy (that is, in pursuit of military objectives); others adopt rape as a policy for some other purpose (often, to manage the sexual and reproductive lives of combatants). However, members of other organizations engage in high levels of rape not because it is a policy, but because it is driven by social dynamics among combatants and tolerated by commanders. In such cases, it is a ‘practice’ of war.

In this commentary, we analyse the implications of recent social science research for efforts by peacekeeping operations to prevent, or at least mitigate, CRSV. We first document the sharp variation across armed actors (state forces, rebel organizations, or militias) in patterns of CRSV, emphasizing that some engage in little sexual violence of any form. Rape is thus not inevitable during war, which suggests that peacekeeping policy can make a difference if properly attuned to why it occurs. We distinguish between CRSV as a policy of the armed organization (whether adopted as a military strategy or to manage the sexual and reproductive lives of combatants), and its occurrence as a practice. We then analyse challenges to prevention posed by the complex patterns of CRSV. Finally, we discuss the implications for efforts by peacekeeping operations to address CRSV, and suggest how peacekeepers might most effectively combat sexual violence.

Variation in Patterns of Conflict-Related Sexual Violence

Perhaps the most important social science finding is that CRSV varies sharply in form (rape, sexual torture, forced abortion, etc.), targeting (against what social group), and frequency across armed organizations.¹⁸ Some organizations target only women and girls, while others target males as well; still others also target sexual and gender minorities.¹⁹ And some organizations target specific social groups with particular forms of sexual violence, as in the case of the Islamic State, which sexually enslaved Yazidi but not Sunni Muslim girls and women.

Crucial for peacekeeping operations is the particular finding that rape by combatants can be effectively prohibited. For example, the Farabundo Martí National Liberation Front engaged in little CRSV during El Salvador’s civil war (1980–1992) – even as state forces committed widespread rape during massacres and sexually tortured political prisoners.²⁰ There is of course severe under-reporting of rape and other forms of sexual violence, but the documented differences are too sharp to only reflect differences in reporting.

Because CRSV can be effectively prohibited, it is not inevitable in war or its aftermath. Peacekeeping policy toward sexual violence therefore matters. If state forces are susceptible to international sanction, or interact frequently with peacekeeping personnel in the context of reform programs, there is an opportunity to advance CRSV prevention efforts, especially because the fraction of state forces that engage in moderate to high levels of CRSV is higher than the fraction of rebel organizations that do so.²¹

To be sure, some armed organizations do engage in high levels of CRSV. Some do so as a military strategy purposefully adopted as organizational policy during ethnic cleansing, genocide, or sexual torture (in some settings). Examples include state forces during Guatemala’s civil war, Bosnian Serb militias in the former Yugoslavia, Janjaweed militias in Sudan, Hutu forces in Rwanda, and the Myanmar military.²²

However, some armed actors adopt some form of sexual violence as organizational policy not as a military strategy, but as a way to manage and regulate the sexual and reproductive lives of combatants.²³ Examples include the sexual enslavement of Yazidi women and girls by the Islamic State and the forced prostitution of women held in brothels by the Japanese military during World War II.

When some form of CRSV against some social group(s) is adopted as an organizational policy (whether for strategic or other reasons), it may be authorized by commanders, but not ordered. Neither Japanese soldiers nor Islamic State combatants were ordered to sexually abuse civilians, but were authorized to do so under specific conditions.

A fundamental challenge to the effective prevention of CRSV is that it may be committed frequently by combatants without having been ordered or authorized as an organizational policy. Rape by American soldiers in the Vietnam War was frequent because it was tolerated by US commanders and driven by peer social dynamics – rape was a practice. During formal and civic investigations, US soldiers claimed that they had been ordered or authorized to kill civilians but did not make the same claim for rape.²⁴ Several, however, mentioned social pressure to participate; very few were prosecuted.

Patterns of rape in other wartime contexts are also well characterized as a practice. Gang rape builds cohesion among the recruits of those insurgencies and state forces that rely on abduction and press-ganging (respectively).²⁵ Rape in these cases is not purposefully adopted by commanders as policy; rather, members of small units participate and insist that all recruits – including women – participate.

When entrenched as a practice, rape (or other forms of sexual violence) may be difficult to eradicate. For example, a pattern of sexual assault of both men and women within the ranks of the US military has persisted despite two decades of claimed but not realized ‘zero tolerance’.²⁶ Each year, almost 5% of active duty servicewomen and 1% of men experience sexual assault by a colleague(s). The frequency of gang rape and retaliation by peers as well as the perpetrator against the survivor for reporting suggest that sexual assault is largely driven by social dynamics.

Rape, or some other form of CRSV, is likely to be frequent as a practice when both combatants and commanders act on gendered norms and beliefs that license sexual violence against certain populations as acceptable despite its formal prohibition. Combatants may believe in such norms and hold such beliefs if they are recruited from a society that itself permits sexual violence against a particular population. Or they may come to hold new norms and beliefs through a process of peer socialization, which may itself be violent, that transforms recruits’ norms, proclivities, and beliefs toward participation in CRSV, often including exceptionally brutal forms of rape, including gang rape (very frequent in war). Commanders may tolerate CRSV for various reasons. The commander may think its effective prohibition would be too costly because it would require disciplining otherwise effective subordinates; might divert scarce resources to an issue he sees as unimportant; or might undermine vertical cohesion, perhaps because it would lessen the respect of subordinates for him. Or he may tolerate rape because he is little troubled by the suffering of those targeted, because it is too much trouble, because he himself engages in rape, or is too weak to effectively enforce the organization’s norms and rules. The commander acts on an understanding of the benefits and costs of toleration that reflect his own gendered norms and beliefs.

Implications for Peacekeeping Operations

Peacekeeping operations are increasingly mandated to respond to, and prevent, conflict-related sexual violence.²⁷ However, as Johannson (this issue) notes, research on peacekeeping operations and CRSV is at a relatively early stage. Johannson and Hultman argue that peacekeeping operations may not be efficient tools for responding to CRSV, and may be most effective when (1) the peacekeeping operation has a protection mandate and (2) armed organizations have strong internal control.²⁸ Kirshner and Miller’s results are more optimistic, and they note the role civilian personnel may play in promoting institutional responses to CRSV.²⁹ Peacekeeping personnel can reduce the power of combatants to victimize civilians, and the presence of international observers who document and punish CRSV may discourage potential perpetrators.³⁰ Further, some peacekeeping operations now attempt to reform security institutions and strengthen formal justice mechanisms in conflict-affected areas, which can complement efforts to mitigate CRSV and prevent sexual violence in the future.³¹ In this section, we highlight nine implications of recent academic research on CRSV, and discuss these implications in the context of peacekeeping.

First, whether an armed group commits high or low levels of CRSV, its institutions and organizational culture must be understood. CRSV – its form, targeting and frequency – varies across armed organizations, sometimes within the same conflict.³² While an organization’s sexual practices may reflect the broader society’s gender hierarchies, it is imperative to understand an armed group’s internal institutions and organizational culture. These features can help explain whether (and why) a particular armed organization practices restraint, or engages in frequent CRSV. This is crucial information for peacekeeping operations that seek to understand and address CRSV. For example, Johannson and Hultman (see note 28) suggest that the nature of armed group institutions, particularly as they relate to internal control, might condition the effectiveness of peacekeeping operations seeking to reduce CRSV.

Second, CRSV does not only affect women. Although commonly framed as a ‘women’s issue’, we emphasize that there is a wider range of CRSV survivors (and perpetrators) than is often assumed, and sensitivity to this diversity can help strengthen peacekeeping operations’ responses to CRSV. Many of the UN’s early warning indicators justifiably emphasize threats to women and girls, and Women’s Protection Advisors play a key role in monitoring and analysing CRSV during peacekeeping operations. However, we note that men and boys, as well as sexual and gender minorities, may also be targeted. This has implications for how CRSV should be addressed after it occurs. The needs of diverse survivors may differ because of social and legal contexts, especially in places where homosexual acts (even if by violence) are considered crimes, or are deeply stigmatized.³³ Specialized services may also be necessary if diverse survivors were targeted for different types of CRSV (e.g. women for forced pregnancy or abortion, but men for sterilization or sexual mutilation). Further, female combatants may also be perpetrators, particularly among groups that recruit forcibly and practice gang rape. As evidence from Sierra Leone suggests, this is a more common phenomenon than prevailing assumptions about CRSV would predict.³⁴

Third, CRSV does not only affect civilians. While many peacekeeping imperatives justifiably emphasize threats to civilians, intra-force sexual assault may also be prevalent.³⁵ When peacekeeping operations are implementing or facilitating DDR programs, some demobilized combatants may require additional, specialized services if they were targeted for intra-force sexual assault.

Fourth, peacekeeping personnel documenting CRSV should look for evidence of its emergence not just as a policy, but also as a practice. Rhetoric suggesting that sexual violence has been authorized, rather than directly ordered, may constitute evidence of sexual violence as an organizational policy.³⁶ However, peacekeeping personnel should also look for evidence that commanders do not punish CRSV, despite its formal prohibition. This can advance the cause of post-conflict justice and accountability, and advance prevention efforts.

Fifth, whether a policy or practice, commanders are responsible for restraining, promoting, or tolerating combatants’ use of sexual violence. Toleration by commanders is a necessary condition for the emergence of CRSV as a practice by members of an organization. When an armed organization perpetrates CRSV, commanders as well as individual perpetrators should be held legally accountable. If peacekeeping personnel are able to gather evidence of CRSV as a practice, it may strengthen justice and accountability efforts. Seelinger and Wood (forthcoming) suggest that sexual violence, even when it occurs as a practice rather than a policy, can be charged and prosecuted as a core international crime.³⁷

Sixth, the emergence of CRSV as a practice entails a difficult challenge for peacekeeping operations. Because CRSV may emerge as a practice and become embedded within an armed organization’s culture, peacekeeping personnel may face a particular challenge in addressing this type of violence, especially if it is a long-running feature of life in the armed organization.³⁸ However, there are avenues for addressing CRSV as a practice. Changing the gendered norms that facilitate both the violence itself, and commanders’ toleration of it, will be crucial to ending CRSV as a practice, and will require cooperation among civil society groups, government entities, and other international actors, not just peacekeepers. Peacekeeping personnel may be able to promote long-term changes in attitudes toward sexual violence in society by supporting domestic justice institutions, as well as awareness-raising campaigns and support for survivor services.³⁹ Ruffa (this issue) also highlights how military peacekeeping contingents might transfer norms of CRSV prevention. However, the short-term protection efforts of peacekeeping operations will remain indispensable.

Seventh, groups that continue to commit CRSV frequently should be identified, and pressured to end sexual violence. Because engaging in CRSV constitutes a choice, rather than an unavoidable wartime phenomenon, the UN should pressure commanders to declare and enforce prohibitions on sexual violence by their subordinates. This may be especially effective when an armed organization has strong internal institutions that allow commanders to punish combatants. Recent research suggests that organizations with high levels of internal control may be more responsive to the presence of peacekeeping personnel, and therefore more likely to show restraint.⁴⁰

Eighth, organizational change should be encouraged. Both state and non-state organizations should be encouraged to declare and enforce prohibitions on sexual violence. Peacekeeping personnel interact with members of state security institutions, especially in the context of security sector reform, and several police and military institutions have worked with the Special Representative on CRSV ‘in order to develop specific, time-bound commitments and action plans to address violations’.⁴¹ However, the UN should continue to pressure non-state forces to prohibit and punish sexual violence as well. Geneva Call’s Deed of Commitment for the Prohibition of Sexual Violence in Situations of Armed Conflict and Towards the Elimination of Gender Discrimination may provide an example for how such commitments from armed non-state organizations could be structured.

Ninth, local efforts to address sexual violence through formal institutions should be supported. In the long term, the international community should support local efforts to build domestic institutions that address sexual violence. This may involve support to domestic justice institutions, as well as awareness raising and services for survivors. Recent research has also suggested that UN peacekeeping operations can help to rebuild the rule of law after a civil war ends,⁴² and formal justice institutions that prosecute sexual violence may deter its occurrence in the future. Peacekeeping personnel may also provide specialized training for members of the security sector that is specific to preventing sexual violence. However, we caution that while gender balancing efforts in post-conflict armed forces may be beneficial, they may not be sufficient to prevent CRSV. A recent study of the Liberian National Police does not find evidence that adding women to police teams increases a team’s sensitivity to gender-based or sexual violence.⁴³

Conclusion

Conflict-related sexual violence constitutes a major challenge for peacekeeping operations. We have summarized recent research on CRSV and derived several implications for peacekeeping operations seeking to address this type of violence. We have described variation in form, targeting, and frequency, while also distinguishing between CRSV as a policy and CRSV as a practice. In particular, we highlight the importance of tailoring policy to whether the organization has adopted CRSV as a policy, or is tolerating CRSV as a practice. We emphasize holding both perpetrators and their commanders accountable for sexual violence, and we note the value of encouraging organizational change. Ultimately, both short and long-term efforts will be required to eradicate CRSV, and attention to the gender norms and internal institutions of armed organizations will be necessary for this endeavour.

Peacekeeping and Protection from Conflict-Related Sexual Violence

Karin Johansson

Uppsala University

Nearly twenty years have passed since Skjelsbaek⁴⁴ concluded that conflict-related sexual violence (CRSV) poses a particular challenge to United Nations (UN) peacekeeping operations. Since then, we have witnessed an increased attention to CRSV and a number of adjustments have been made within the UN Department of Peace Operations to adequately address CRSV in mandate implementation. For reasons related to data availability, it has only recently been possible to begin assessing the impact of these efforts statistically, over time, across peacekeeping operations and warring parties. Findings are not yet conclusive, leaving significant questions unanswered. To progress, I argue that we need to further strengthen partnerships between researchers, policy-makers, and practitioners. This would allow for more fine-grained analyses with higher external validity.

In this commentary, I share insights from empirical research focusing on peacekeeping operations effectiveness in mitigating CRSV perpetrated by warring parties. While bearing in mind the delimitations of our current data and the – only partial – conclusions we should draw from them, I highlight four findings: (1) the Security Council does respond to information about CRSV although not in a consistent way; (2) current peacekeeping operations practices succeed in curbing CRSV under certain circumstances, but the average success-rate in protecting civilians from CRSV is unclear; (3) peacekeeping operations, and their police contingents in particular, have an important role to play in securing that survivors’ voices are heard; and (4) despite the expectation that female peacekeeping personnel would be particularly well suited to combat CRSV, this remains largely unverified. To put the findings in perspective, I regularly conclude by clarifying what we still do not know and what is needed in order to improve our understanding of how peacekeeping operations adequately could prevent CRSV.

Evidence suggests that policy-makers increasingly pay attention to CRSV: The heightened number of country-specific resolutions⁴⁵; the increased probability of both peacekeeping deployment⁴⁶; and gender-mainstreamed mandates⁴⁷ indicate that the Security Council pays attention to CRSV. The amount of attention and responsiveness however varies. The link between CRSV and peacekeeping deployment seems to depend on the origin of the CRSV reports.⁴⁸ Furthermore, the language in resolutions generally⁴⁹ and in peacekeeping mandates specifically⁵⁰ is inconsistent in the extent to which it reflects the overarching Women Peace and Security agenda. Equally concerning as the inconsistent response, is the limited knowledge we have about the adequacy of today’s peacekeeping practices as tools to address CRSV. This is the issue to which I turn next.

While a robust body of research demonstrates the life-saving impact of peacekeeping operations,⁵¹ the same cannot be said about its effect on CRSV. Theoretically, we could expect some of the mechanisms that prevent killings to protect against CRSV too. But there are also reasons to expect that CRSV requires specific considerations. Unless staged for a wider audience, CRSV often takes place in the private sphere far from military confrontations. Relatedly, military peacekeeping personnel – which constitutes the lion’s share of the entire operation – generally has very limited experience of managing and preventing CRSV.⁵² It is also worth considering that CRSV historically has been excluded from ceasefire and peace agreement stipulations,⁵³ suggesting lower priority for peacekeeping operations and hence lower costs associated with such violations from the perspective of the warring parties. Notably, this has changed in recent years.⁵⁴ Other aspects may hinder CRSV even being reported: the reputation, and sometimes practice, of peacekeeping personnel to commit sexual offences themselves⁵⁵; the tradition in certain communities to shame survivors of sexual violence rather than perpetrators⁵⁶; the risk of reprisal following reporting,⁵⁷ all reduce the likelihood that CRSV becomes recorded. The hesitation or inability to report (which varies across contexts) represents a fundamental challenge to peacekeeping operations and practitioners in the field. It complicates our understanding of its prevalence, in particular across cases and over time, which in turn puts boundaries on possible research and reachable conclusions. The scarcity of data is notably also a consequence of the previously low interest in CRSV, and the assumption that CRSV is an inevitable side effect of any war.

Until now, cross-national research on CRSV has relied on one data collection covering CRSV worldwide 1989–2015: Sexual Violence in Armed Conflict (SVAC).⁵⁸ Drawing on reports from the US State Department, Human Rights Watch and Amnesty International, SVAC holds a collection of yearly prevalence estimates for every warring party during conflict and five years thereafter.

Based on these data, researchers have begun to investigate the impact of peacekeeping operations on CRSV. Judging by the outcome of all UN peace operations deployed 1989–2009, peacekeepers do not mitigate CRSV very well. Using a global sample, Johansson and Hultman⁵⁹ (hereafter referred to as the ‘global’ sample) find on average no reduction in state-perpetrated CRSV the year after peacekeeping operations deployment. Importantly, this non-finding does not hold true across all cases of CRSV and all types of peacekeeping operations.⁶⁰ In a subset of conflicts in Africa during the same time period, Kirschner and Miller (hereafter referred to as the ‘African’ sample) do find a dampening effect of peacekeeping operations on CRSV perpetrated by state forces as well as rebel groups.⁶¹ In what follows, I review the findings of both these studies in relation to peacekeeping operations characteristics and armed groups’ command structures. While the latter is beyond the control of peacekeeping operations, it still gives valuable insights into why the challenge to mitigate CRSV is difficult and when extra and/or different resources need to be invested.

Beginning with peacekeeping operations characteristics, the global sample finds that the number of personnel is particularly important in relation to CRSV as it sometimes takes place in remote and more private spaces than violence intended for a wider audience. Also crucial is a mandate focused on civilian protection. In a conflict where CRSV is frequent, the risk of continuously high levels of sexual violence by rebel groups decreases substantially the year after a sizable police contingent mandated to protect civilians has been deployed. Without protection-mandate, the risk of rebel-perpetrated CRSV instead increases.⁶² In the study of African conflicts, which does not differentiate between protection mandates and other mandates, the authors instead find a reducing effect of police contingents on CRSV by state forces.⁶³ The findings also diverge in relation to military contingents. In the global sample, there is no discernible impact of military peacekeeping personnel, suggesting that prevention of CRSV to a large degree hinges upon UN police.⁶⁴ The African sample leads to a different conclusion as the authors find a mitigating effect of military peacekeeping personnel on CRSV by both state forces and rebel groups.⁶⁵ The same study argues for the pacifying effect of civilian contingents. While the knowledge about this impact is limited, it remains an important avenue for future research. Overall, more research is needed to teas out why these studies differ in outcome. Possibly, they reflect heterogeneity across geographical regions. In addition, they highlight a sensitivity of the analyses to different modelling choices including whether or not to single out protection mandates, what time lags to use as well as how to deal with the non-random deployment of peacekeeping personnel. In addition, they point to the crudity of current measurements in quantitative research. To better understand when peacekeeping operations succeed in improving the situation on the ground, we would need more detailed information on both the specific undertakings of various peacekeeping contingents at different points in time, across different territories, and periodical reports of CRSV with more precise information about location and timing than conflict-year. To strengthen the external validity of our measurements and models, stronger partnerships between researchers (qualitative and quantitative), policy-makers and practitioners is needed.

The success of current peacekeeping operations is not only a product of composition and mandate. It is also a matter of the degree to which an armed actor exercises central command over its forces. While the international spotlight that follows from a large international operation expectedly has the power to impact warring parties’ political will, their actual capacity to act upon it lay more in how the armed group functions and is organized. In an organization such as the Integrated Armed Forces in the Democratic Republic of Congo (FARDC) where the relationship between the leadership and the rank-and-file soldiers has been characterized by disdain and broken promises,⁶⁶ we cannot expect adjustments in the incentive structure at the top leadership to leave much of a footprint at lower levels of the echelons, at least not in the short term.

Based on rough measures of control, there is statistical evidence in support of this notion: While military contingents on average do not leave a measurable impact on CRSV in the global sample, they do indeed have a substantive, violence-reducing effect on those warring parties that exercise control over their forces.⁶⁷ While centralized groups also commit CRSV,⁶⁸ there is a substantial proportion of perpetrating forces lacking such command. Indeed, recent research⁶⁹ shows that rebel group fragmentation often correlates with CRSV which further speaks to the importance of developing alternative strategies to make peacekeeping operations better equipped to gain control over CRSV by decentralized or fragmenting groups. From a research perspective, more precise measures of control as well as in-depth case studies are needed to better nail down the processes through which the effects of peacekeeping operations resonate (or not) throughout the line of command. It will also be important to examine how other types of group characteristics, such as within-group prohibitions, influence peacekeeping operation effectiveness in curbing CRSV.⁷⁰

Until now I have discussed the impact of peacekeeping in terms of increases or decreases in reported prevalence levels of CRSV, but peacekeeping has additional implications for CRSV through other pathways. Recent studies have found that CRSV makes women more likely to collectively mobilize⁷¹ and pushes warring parties to the negotiation table.⁷² The importance of women’s ability to mobilize is demonstrated in yet another study, which finds that CRSV by rebels makes negotiated settlements more likely only in societies where women’s voices are acknowledged in the public sphere.⁷³ Thus, to increase the probability that negotiation efforts are successful, UN police contingents play an important role to help create a safe environment for nonviolent mobilization where survivors’ voices can be heard.⁷⁴ Needless to say, conflict settlements are key to long-term elimination of CRSV.

Lastly, I turn to the policy expectation that peacekeeping operations with female peacekeeping personnel are better equipped to deal with CRSV.⁷⁵ According to the UN, women are critical in all types of positions (civilian as well as uniformed).⁷⁶ Indeed, women expectedly widen the overall skillset, are more approachable to local women and ‘act as role models’⁷⁷ in any capacity. While research has corroborated the positive impact of diversity in terms of cultural background of peacekeeping personnel,⁷⁸ there is no published cross-national study testing the impact of gender diversity specifically.⁷⁹ There are at least two reasons for why such a study is difficult to conduct: (1) female peacekeeping personnel is less likely to be deployed to areas suffering from CRSV than other areas⁸⁰ (which induces selection bias); (2) a study of Liberia describes how female personnel deployed to high-risk conflicts rarely leave the compound and actually meet civilians⁸¹ (requiring research to use fine-grained information seldom accessible to comparative scholars). These obstacles aside, the aspiration rests on a biased yet common understanding of CRSV only affecting women.⁸² Even if female personnel were successful in reaching local women, this could never be more than a partial solution considering that CRSV against men would remain unaddressed.

Conclusions

There are strong theoretical reasons to believe that CRSV requires specific considerations when designing peacekeeping operations. Initial studies of the average effect of peacekeeping operations on CRSV are inconclusive, suggesting that the impact differ depending on various factors such as peacekeeping mandate and command structure of armed groups. Some empirical evidence suggests that police contingents may have a particularly important role to play. Similarly, there are theoretical reasons to expect that diverse operations (in terms of gender and cultural background of personnel) would be more effective but this is not yet supported by statistical evidence. Partnerships where unique insights from policy, practitioners and researchers are used to extend and improve our current measurements and data collections would have the capacity to further teas out how peacekeeping operations most effectively could prevent CRSV.

The Evolving United Nations Approach to Preventing and Addressing Conflict-Related Sexual Violence: Experiences from the Democratic Republic of the Congo

Walter Lotze

Mason Fellow for the Mid-Career Masters in Public Administration Programme with the Kennedy School of Government, Harvard University.

For more than a decade, the United Nations (UN) has been increasingly concerned by acts of sexual violence when used as a tactic of war or terrorism, and when systematically used to target civilians, both because this targeting of the civilian population exacerbates conflict, and because it hinders the restoration of international peace and security (see Johansson this Forum). The Security Council has therefore increasingly mandated UN presences in conflict and post-conflict areas to work both to prevent and respond to the use of conflict-related sexual violence (CRSV). The efforts of UN peace operations have become central to these efforts, and the work which has been undertaken in the Democratic Republic of the Congo (DRC), first through MONUC⁸³ and later MONUSCO,⁸⁴ has perhaps been most formative for the organization in this regard. This commentary builds on research and personal views, in part arising from my experience as Head of Office of the Office of the UN’s Special Representative of the Secretary General in the DRC (January 2017 – June 2019) and Deputy Chief of Staff of MONUSCO (August 2016–January 2017).

The role of UN peace operations in Addressing CRSV

Over the years, the UN has worked to prevent and address CRSV in conflict and post-conflict settings across the globe. The bulk of these efforts have taken place in zones of conflict or instability, including Afghanistan, the Central African Republic, Colombia, the Democratic Republic of the Congo, Iraq, Libya, Mali, Myanmar, Somalia, South Sudan, Sudan, and Yemen. Yet the organization has also worked to address sexual violence crimes in post-conflict settings, such as in Bosnia and Herzegovina, Côte d’Ivoire, Nepal and Sri Lanka, and monitors situations of concern, such as in Burundi and Nigeria.⁸⁵The work of peace operations has been given particular attention in recent years. By the end of 2018, five UN peace operations (in the Central African Republic, Darfur, the DRC, Mali, and South Sudan) and two UN special political missions (in Afghanistan and Somalia) had been mandated to address CRSV. When issuing mandates, the Security Council has specifically tasked these operations to establish the required Monitoring, Analysis, and Reporting Arrangements (MARA), to engage with parties to secure time-bound commitments to end CRSV, to support parties to implement these commitments, and to support Security Sector Reform (SSR) efforts to build capacity to address this form of violence.⁸⁶ By 2018, all peacekeeping operations with protection of civilians mandates had also established monitoring arrangements and incorporated early warning indicators for CRSV into their protection structures.⁸⁷

The role of the Women Protection Advisors (WPAs) has proven important. Tasked with ensuring that the CRSV aspects of mandates are implemented, and are regularly reported on, by 2019, 21 WPAs were working in peace operations.⁸⁸ Whereas such individuals previously worked alongside the human rights components, in recent years these functions have been integrated, with WPAs now reporting directly to the heads of human rights components. Efforts that were previously viewed as complementary were thus better recognized as being integral to one another. Today, WPAs work with the senior mission leadership to mainstream conflict-related sexual violence considerations into various aspects of mandate implementation; coordinate the monitoring and reporting arrangements; ensure regular reporting to the Security Council; engage with parties to the conflict to secure commitments to end violations, and support parties to implement these commitments; and work to strengthen capacities of national actors and all stakeholders to prevent and address conflict-related sexual violence.⁸⁹ It is in the DRC that the work of UN peace operations has perhaps been most formative to how UN peace operations work to prevent and address conflict-related sexual violence.

Partnering in Support of National Efforts

In late 2008, the United Nations Organization Mission in the Democratic Republic of the Congo (MONUC) was mandated to support the Forces Armeés de la RDC (FARDC) in their operations against armed groups, as a way of enhancing the protection of the civilian population. The challenge was that Congolese armed forces were not fully adhering to international humanitarian law and human rights law when conducting these operations, and violations of the civilian population, including the use of sexual violence, were being recorded. In response, MONUC developed a two-tracked approach. First, a conditionality policy was developed, which stipulated that elements of the armed forces which committed violations against the civilian population would not be eligible for UN support. This would later serve as the foundation for the United Nations Human Rights Due Diligence Policy, applied globally wherever the UN provides support to state security actors or other parties. Second, the mission started to provide assistance to the military justice system, supporting the Congolese government in the fight against impunity. Here, priority would be placed on high-ranking military officers, to demonstrate that where abuses were committed, justice would be served, no matter the rank. The approach was new, as until that time the UN had been hesitant to support national military justice systems.⁹⁰ Key to this evolving thinking was that survivors of CRSV often risk being violated twice; once through the violent acts which are committed against them, and again by the reactions, or the lack thereof, by the state and society.

The Government of the DRC and the UN thus started working hand in hand in the fight against armed groups, the protection of civilians, the fight against impunity, and the rule of law. Combatting CRSV and violations against children featured strongly in this two-tracked approach. Three key areas of engagement were identified, namely the fight against impunity, judicial reform, and establishing functioning justice institutions and prisons in areas affected by armed conflict. The initial prioritization of emblematic cases of war crimes and crimes against humanity in three provinces most affected by conflict helped to establish precedents.⁹¹ Some high-level cases have attracted most of the attention at global level in recent years, yet it is also important to note that on an almost annual basis FARDC officers have been tried for, and convicted of war crimes or crimes against humanity for rape and sexual violence. Indeed, it has been remarked that as efforts have progressed, military justice officers have often displayed greater rigour in investigations, in upholding fair trial principles in proceedings, and in implementing measures to protect victims and witnesses. Some Congolese authorities have also indicated that the threat of prosecution is the only credible deterrent to the commission of crimes within the FARDC and Police National Congolaise (PNC) ranks.⁹² The first-ever UN Comprehensive Strategy against Sexual Violence in Conflict was then developed in the DRC, which the Government endorsed through its zero-tolerance policy and included in its own gender-based violence National Strategy.

With time, these efforts started to show real results. Whereas before, rape was being used by state security forces as a weapon of war, in recent years this slowly stopped being the case. Although security forces did still commit acts of sexual violence, this was increasingly done in an opportunistic manner, rather than being organized at command level. In 2012, still around half of the recorded violations were attributed to the FARDC and the PNC.⁹³ By 2014, a significant change had been recorded, and for the first time since reporting began, less than half of the reported cases of sexual violence were attributed to Congolese security forces.⁹⁴ These efforts were reinforced and given executive level attention when President Joseph Kabila established an advisory role in his office on sexual violence and child recruitment in July 2014. In her new role, the Presidential Advisor placed the fight against impunity and providing better support to victims at the very centre of her efforts. That same year, the FARDC launched its first Action Plan against sexual violence, and a National Commission was established to oversee its implementation.

Overall, the number of recorded violations on the part of state actors has continued to decrease over the years, with most violations committed by non-state armed actors since 2014. Between 2017 and 2019 some worrying trends did emerge, when a rise in recorded violations on the part of the FARDC and the PNC was observed, as the country experienced political tensions. Of concern was that violations were being committed in new areas by armed groups (the Kamuina Nsapu and Bana Mura groups in the Kasai region), but also by the FARDC deploying in the same areas, and a significant rise in sexual assault by the PNC of persons in police custody was observed.

Cooperation continued nevertheless, and by early 2018, joint efforts had resulted in the prosecution of over 1200 accused, of whom 963 had been convicted, including senior elements of the FARDC, armed groups and a provincial parliamentary deputy.⁹⁵ Importantly, between 2016 and 2018 316 military field commanders signed undertakings to prevent and address sexual violence, and 570 commanders were trained on their legal obligations.⁹⁶ In this period, the UN also documented 159 convictions of members of the state security forces for sexual violence crimes. Verdicts were also handed down to members of numerous armed groups during this period, highlighting that justice was served to both state and non-state actors who committed violations.

Partnership has remained key to progress throughout. The Ministry of Justice and Human Rights and military judicial authorities, supported by the UN Team of Experts, the United Nations Development Programme, MONUSCO and the International Centre for Transitional Justice, have continued to work to prioritize the most grave cases for prosecution. The United Nations Children’s Fund has also worked to support victims, through the provision of medical, psychosocial, legal and socio-economic support to survivors of rape by combatants. Challenges remain, and reaching all victims and being able to provide the same level of support to those who may need it cannot always be assured. This is particularly so when it comes to providing access to post-exposure prophylaxis, treatment for sexually transmitted diseases, and the provision of mental health support. To address some of these and other challenges, the Government in 2016 developed a three-year roadmap (2017–2019) of national priorities to further advance progress in this regard, which is supported by the UN.⁹⁷

In early 2018, with the support of MONUSCO, the national police developed an action plan against sexual violence.⁹⁸ Later that same year, the DRC launched its second National Action Plan on Women, Peace and Security (2019–2022), which contains core elements of work to eliminate impunity for sexual violence.

Having higher number of women peacekeepers has also played a key role. By early 2019, MONUSCO had over 3% of its military component and over 12% of its police component staffed by women. While these numbers may seem low, their impact has been profound. Women officers take part in patrols, visit detention facilities, form part of all-Female Engagement Teams (FETs), and undertake special missions. Women serve both to strengthen engagement by the mission with communities, and with state security actors. Women Protection Advisors also continue to engage with newly deployed FARDC and PNC personnel, provide training, sensitize state security actors and non-state armed actors alike, and work to reinforce local early warning and protection mechanisms (see also Johansson, in this Forum).

Consolidating Efforts: Implications for UN Peacekeeping

The UN approach to preventing and addressing CRSV in the DRC has thus evolved significantly over the years. Similarly, the experiences and lessons of other contexts is shaping the direction which the Security Council sets, and how the UN will implement these efforts going forward. Two key recent developments are indicative of this.

First, since 2017 the Secretary-General has taken numerous measures to improve the way the organization responds to sexual exploitation and abuse (SEA) by UN personnel and those deployed under its banner. This involves placing victims’ rights at the centre of response mechanisms, re-doubling efforts to end impunity, and working with Member States to better detect, control and prevent incidents of SEA within the context of a zero-tolerance framework. Leading by example is essential, if the UN is seen to be a credible partner in the fight against sexual violence (see Ruffa and Hoover Green in this Forum for a discussion).

Second, drawing on the formative experiences in the DRC and from across its work in other countries, the UN in 2018 started work to develop its first comprehensive set of guidance on preventing and responding to CRSV through UN peace operations.

The policy would do three key things. First, it outlines the six guiding principles which should govern all efforts in relation to preventing and responding to CRSV – do no harm; confidentiality of victims and families; informed consent by all those cooperating with the UN; the use of a gender-sensitivity; maintaining a victim-centred approach; and protecting and promoting the best interests of the child where applicable. Second, it outlines the five priority objectives of UN peace operations in relation to addressing conflict-related sexual violence – the prevention of such acts and the protection of populations vulnerable to them; ending impunity; raising awareness of such acts, and condemning them where they occur; strengthening the capacity of national actors to address conflict-related sexual violence; and empowering victims through political process and referring them for support where possible.⁹⁹ Third, the policy offers specific guidance on roles and responsibilities in field operations, and how the conflict-related sexual violence mandates are to be implemented in a comprehensive manner with a range of stakeholders.¹⁰⁰

Conclusions

While progress is being made in terms internal policy consolidation, and measures are being undertaken to reinforce delivery on the ground, attention will also need to be paid to a few dynamics going forward. First, the importance of partnerships with national security actors should be further emphasized. The UN is most effective when it is working in support of national efforts, and reinforcing these. Second, the provision of support to justice systems, both military and civilian, has been a key to success, and is the best means of ensuring that the gains made can be sustained in the long term. Investing in justice for victims is also an investment in sustainable peace. Yet investments in this area have been varied. Third, the risks of not properly applying the Human Rights Due Diligence Policy are high, and this should not be compromised. However, the risks of not engaging at all with actors who commit violations are even higher. The policy should be applied to manage the risks of violations in the best manner possible, but it should not become a reason to not engage at all. Fourth, reparations for victims matter, both symbolically and materially where lives and where livelihoods have been impacted. In some cases courts award damages, but this is not always the case. Working to ensure uniformity of approach would go a long way towards the equitable administration of justice. Fifth, while reporting has improved in recent years following the establishment of dedicated reporting mechanisms, the deployment of the WPAs and the mainstreaming of reporting on CRSV, barriers to reporting remain. Peace operations generally report lower numbers of violations than other UN actors (like the United Nations Population Fund – UNFPA, or UNICEF). It is therefore often difficult to know the true extent of the violations, and where the most support may be needed. And sixth, a deeper understanding of vulnerability may be required. Although it is recognized that CRSV is not solely a women’s issue as men and boys are both victims of and deeply affected by such violations, for male survivors, sexual violence remains shrouded in cultural taboos. In 2018, in over 60 countries no provisions for male victims existed within the scope of national sexual violence legislation, leaving male and boy victims particularly vulnerable to accusations of homosexuality, in particular in countries where this is criminalized. Lesbian, gay, bisexual, transgender and intersex individuals who are victims of violations are often vulnerable to stigmatization, exclusion or criminal persecution if they come forward.¹⁰¹

The UN has made significant progress on preventing and addressing CRSV in a very short span of time. In the course of just over a decade, the issue has been placed at the centre of Security Council deliberations, an organizational architecture has been developed, peace operations have been provided with increasingly strong mandates to focus on this, and Women Protection Advisors have been deployed to all major peace operations to both drive and consolidate these efforts. And significant progress has been attained, such as in the DRC. Yet as conflicts continue to evolve, so too the complexity of the challenge has evolved, and in turn the UN’s approach must continue to be built and shaped. Consolidating the gains made to date, and building on these for the future, will be key for the Security Council’s call to bring a total halt to CRSV to be fully realized.

Peacekeeping Military Contingents in Preventing CRSV by Local State Forces

Chiara Ruffa

Uppsala University/Swedish Defence University

Research findings on whether or not military peacekeeping contingents can prevent conflict-related sexual violence (CRSV) perpetrated by local state forces are contradictory and inconclusive (see Johansson, this Forum).¹⁰² This lack of knowledge is highly problematic as local state forces constitute the largest group of perpetrators. When addressing this knowledge gap, I argue that we can improve prevention if we explore the role of two core traits that military peacekeeping contingents bring along when they deploy to an operation – military culture and self-understanding of professionalism – and how these are transferred to the local state security forces. By borrowing analytical tools from the field of security studies, this commentary contributes to this enhanced understanding and thereby also outlines how military peacekeeping contingents can further develop the capacity to become credible interlocutors when supporting the prevention of CRSV.

While a wealth of literature has explored sexual exploitation and abuse (SEA) perpetrated by peacekeeping personnel (see Hoover Green in this Forum), no study has so far explored the effect of culture and self-understanding of professionalism that military peacekeeping contingents may have on the perpetration of sexual violence by local actors.¹⁰³ My argument for why these are central for prevention is presented in three steps. First, I outline how the military culture and the self-understanding of professionalism in military peacekeeping contingents may harbour ideas that affect the prioritization of CRSV and which include a set of gendered conditions that can tolerate sexual violence and abuse. If the peacekeeping contingent has issues of this kind, it is less probable to prevent CRSV and may even contribute to harm. Second, once peacekeeping contingents deploy, their culture and ideas of professionalism are likely to be imitated by local state forces. We need to understand how these play into prevention. Third, and relatedly, we need to understand that the norm of prevention is more likely to transfer from peacekeeping contingents to local state forces via informal learning rather than formal programs.¹⁰⁴ In discussing these arguments, I further relate them to understandings of military responsibilities and capacity. After discussing the arguments, I outline opportunities of addressing the problems and transferring more preventing norms of CRSV to local security forces given the internal dynamics of the military peacekeeping contingents themselves. I conclude by outlining ways ahead for future research and policy.

Military Contingents’ Traits and the Prevention of CRSV

I argue that we need to consider the state military organizations which are deployed to peacekeeping operations as organizations with special traits. Fundamental to this consideration is the fact that preventing CRSV has never been a core concern nor a core task of state military training and norms. At present, across the spectrum, military organizations still hold combat as their clear core task and conflate ‘masculinity with idealized traits of the warrior’.¹⁰⁵ When they deploy, they tend to fall back into those core operational tasks. This entails that we need to start from the expectation that the understanding of CRSV, the sense of responsibility for preventing it, as well as the practical capacity to have an impact, are all likely to be low among many military peacekeeping contingents when they are deployed. To make matters worse, I argue, some military organizations implicitly seem to tolerate sexual violence which can make a military peacekeeping contingent a potential problem in and by itself. Before a military peacekeeping contingent can become effective at preventing CRSV, they therefore need to consider the roles that their own military culture and professional understanding of responsibility play, and the adherent capacity issues which follows from these.

Let us start by looking closer at these traits on culture and profession. Often depicted as ‘total’ or ‘greedy’ institutions, the military profession requires high levels of commitment and embracement of a collective identity.¹⁰⁶ Key in this respect is military culture, ‘the set of beliefs, attitudes, norms and values that become deeply embedded and profoundly ingrained within a military unit and that guide the way the unit manages its internal and external life, the way it interprets its tactical and operational objectives and the way it learns and adapts to external influences’.¹⁰⁷ Research finds that state militaries and their culture are also ‘inherently gendered’ and male-dominated, rooted in the idea that war should be fought by men.¹⁰⁸ Few militaries in the world allow women to serve in any role, and many perform notably poorly at retaining women. When women deploy in peacekeeping operations, Karim and Beardsley showed that female peacekeeping personnel are often relegated to safe spaces¹⁰⁹ and King demonstrated that in the large boots-on-the-ground operations in Afghanistan and Iraq, female personnel has taken great risks but has not ben awarded as much as male personnel in terms of carrier progression. To make it worse, they have often boxed women in ancillary care roles.¹¹⁰

While military organizations have changed substantially over the past thirty years, they retain some of the above characteristics. Notably, most state military organizations struggle to prevent sexual violence and assault even within their own ranks. In the US, ‘between 9.5% and 33% of women report experiencing an attempted or complete rape while serving in the military’ and ‘male victims are reported between 1% and 12%’.¹¹¹ If an organization with such a problem deploy contingents to a peacekeeping operation, this may logically hamper a military’s ability to transfer a norm on preventing CRSV in its interaction with local state forces, worsened by the fact that such behaviour may continue to take place during deployment. Some literature even suggests that the experience of deployment, the long period out-of-area may make problems of sexual violence and assault even worse.¹¹² In any case, the presence or persistence of sexual violence, assault and abuse would profoundly delegitimize the contingent but also open a space of reflection of unsolved problems.

These issues thereby have potential practical implications when a contingent is deployed. Research suggests that peacekeeping personnel act as role models for the local population, including members of security sector institutions, just by routine interactions. Arguably, these dynamics should be enhanced if peacekeeping personnel takes on the roles of monitors and trainers of local security forces and/or undertake joint operations. If peacekeeping contingents then fall back into those core operational tasks of combat where preventing sexual violence is not considered as a priority, this implies signalling to the local forces that CRSV is not important. This signal is even stronger if tasks in the peacekeeping mandate on CRSV are not implemented at all. While in recent years, military organizations have started to adapt to more complex peacekeeping tasks – ranging from the delivery of humanitarian aid to reconstruction – internal promotion logics and dynamics in military organizations persist.¹¹³ This unfortunate structural problem hence have additional perverse consequences for the ability of soldiers to implement important components of peacekeeping mandates such as the prevention of CRSV.

That said, the issues with organizational culture and ideas of professional responsibilities discussed above are neither a necessity – some military organization may prioritize prevention – nor a justification. Recent trends within the field of war studies suggests that military culture varies profoundly across military organizations and that it may shape norms, behaviour, and indirectly misconduct, in different ways. While some organizations cannot prioritize restraint, others have been more successful in doing so.¹¹⁴ Importantly, this means that military culture is not an inherently toxic driver of behaviour, it is changeable and should be deprived of its gendered and toxic masculinity nature to help prevent sexual violence. This suggests that while preventing CRSV might not have been a priority thus far, there is potential for unpacking important variations across state militaries in terms of propensity to support local security actors in preventing sexual violence. The first condition, however, for norm transfer to happen is to have a well-established norm punishing sexual abuse, assault and violence within the contingent that deploys to the peacekeeping mission.

How Peacekeeping Contingents Could Transfer a Prevention Norm

While previous research has identified the UN as an international norm entrepreneur,¹¹⁵ I argue that one of its most important components – the military contingents deployed – could and do act as a ‘norm transfer entrepreneur’. In this section, I outline two main channels that could help military peacekeeping personnel address existing internal problems and transfer a set of norms that could contribute to the prevention of CRSV to local security actors.¹¹⁶

First, the great advantage of military organizations is that they are likely to imitate each other. This is particularly so when it comes to learning from military organizations perceived to be more professional as the case is in peacekeeping.¹¹⁷ Local security actors may here learn from peacekeeping personnel either directly via training about sexual violence or indirectly by learning and adopting other patterns and practices from the military peacekeeping personnel deployed. An example of when direct formal learning can take place is during training on the prevention of sexual violence in Security Sector Reform (SSR) programs.¹¹⁸ Within those programs, the UN states that ‘rather than imposing strict, prefabricated standards, the international community seeks to suggest agenda items, or rather: the agenda, for reforms’.¹¹⁹ Given this approach, quite some margins are left for those who implement the mandate on the ground which enables the military peacekeeping personnel to support norm transfer. We should here note that this will only take place under the condition that CRSV is understood and prioritized by the peacekeeping contingent. In addition, disarmament, demobilization, and reintegration (DDR) programs can be an arena for norm transfer, in particular, when rebel-military integration into state forces is envisioned. Through including CRSV when supporting the establishment of appropriate vetting and recruitment processes, as well as pre-integration training, sexual violence may be minimized in future security forces.¹²⁰ Informal learning between peacekeeping contingents and local state forces, on the other hand, remains understudied. ‘The most important mode of knowledge formation and transfer behind these similar attitudes is the informal interaction with experienced interveners and local actors, not official trainings or information’, claims Carey.¹²¹ Training is not enough, we need socialization together with informal encounters for creating trust and enhancing norm transfer. Of equal importance is having norm enforcement procedures in cases when peacekeeping personnel witness incidents of misconduct by local security forces.¹²² Imperatively, this also means that if a peacekeeping contingent has not dealt with a pre-existing negative culture or problems regarding harassment, these could also be transferred under such conditions.

Second, the content and context of the norm transfer matter. While the literature on the effect of training on preventing sexual violence is at its infancy, broader findings suggest that militaries learn about norms of restraint in general, including sexual violence prevention.¹²³ Peacekeeping operations involved in training should therefore better integrate the prevention of sexual violence in training on neighbouring norms, work ethics, international humanitarian law, protection of civilians, and other related topics. Naturally, learning about preventing sexual violence should here be given a clear centre stage. Relatedly, in SSR and DDR programs, a lot of emphasis is placed on the transfer of the vague notion of military professionalism, which is usually understood as training and equipment for making the military more militarily effective. Yet, we know from research that military professionalism also refers to the notion of ethics and restraint as much as weapons and technology.¹²⁴ These aspects need to become more explicit in training. Moreover, integrating the norm of CRSV prevention as part of training on military professionalism could grant it higher status and priority. Vital here is also to note that one cannot transfer the norm of CRSV prevention without addressing the importance of control within the armed forces (see Hoover Green in this Forum). It is crucial to have a clear and functioning command and control structure across the military echelon and to form responsible commanders across levels. In this context, it needs to be made absolutely clear that sexual violence undermines military effectiveness and does not boost cohesion (see also Sjoberg, this Forum, regarding non-state security actors).

Conclusions and Future Research

In this commentary, I started by underlining the urgency of tackling the fact that most CRSV perpetrators are state security forces. Peacekeeping operations’ ability to contribute to CRSV prevention in their interactions with this group is hence imperative. However, this is mitigated by the fact that military peacekeeping contingents themselves can have issues regarding military culture, professional understandings, and sexual assault and violence in their ranks. In part, this is due to gendered views and hypermasculine military ideas. I argue that it is only military peacekeeping contingents that address more openly, long-term and effectively these issues and prevent sexual assault and violence within their own organizations that can become credible interlocutors when addressing CRSV by local state security actors.

Importantly, future research should now explore how a military peacekeeping contingent’s ‘track-record’ in sexual violence may signal and undermine their ability to spread the norm of preventing CRSV. Changing the culture is difficult but not impossible – it is taking place in many military organizations. As a short-term solution, much effort should be placed on the practices and effects of norms transfer during already ongoing operations. The main avenue for research and policy in this respect is to move beyond the presence of training alone and explore ways to enhance isomorphism among virtuous military peacekeeping contingents and local state forces. This requires virtuous contingents to emphasize CRSV prevention. Also, protection norms adjacent to CRSV – such as protection of civilians – could help creating synergies and help spread norms of prevention of CRSV among local security actors. Norm entrepreneurs serving in key positions in the military organizations, such as Force Commanders or senior military trainers, are crucial in this regard. Finally, using the norm of military professionalism as a ‘Trojan horse’ to make the prevention of sexual violence norm heard seems a very promising, yet underexplored, path to follow. All local state forces want to become more professional. But in order to be professional, they have to actively prevent CRSV.

Command and (Maybe) Control: Military Institutions and Sexual Violence During Peacekeeping Operations

Amelia Hoover Green

Drexel University

Liberia’s civil war (1989–2003¹²⁵) featured epidemic levels of conflict-related sexual violence (CRSV).¹²⁶ At the close of the conflict, the United Nations Security Council mandated a significant peacekeeping operation with a substantial military contingent of ‘up to 15,000’ armed personnel, plus civilian support staff to prevent violence.¹²⁷ Yet sexual violence did not end. According to data from the Sexual Violence in Armed Conflict dataset,¹²⁸ at least two parties to the conflict (the government armed forces and government-sponsored militias) continued to perpetrate CRSV even after the deployment of the peacekeeping operation. Moreover – despite the fact that the peacekeeping operation was deployed in part to quell epidemic levels of CRSV by conflict actors – the arrival of UN peacekeeping personnel itself created a wave of exploitative transactional sex, i.e. Sexual Exploitation and Abuse (SEA).¹²⁹ More than half of women in greater Monrovia, where the peacekeeping force was concentrated, reported having engaged in transactional sex during their lifetimes, with approximately 46% of women reporting transactional sex with UN personnel.¹³⁰ Of course, sexual violence – including both CRSV and SEA – affects peacekeeping operations far beyond Liberia. Of the 14 post-zero-tolerance UN peacekeeping operations averaging at least 1000 peacekeepers per year – comprising 65 mission-years in total – just one appears to have prevented SEA altogether.

Why do peacekeeping operations, in Liberia and elsewhere, fail to prevent both sexual violence and sexual exploitation? In this Commentary, I show that soldiers’ training and socialization experiences play an important role in the perpetration (or prevention) of unordered sexual abuse. I argue that these findings hold important implications for the prevention of sexual violence and abuse by peacekeeping personnel and conflict actors. However, I also argue that several factors limit the UN’s ability to prevent sexual violence and abuse in post-conflict settings. These include, most notably, the lack of control over home-country training and socialization, inability to effectively sanction sexual violence by other armed actors, language barriers, and the lack of robust mandates. In what follows, I specify ‘CRSV’ or ‘SEA’ only when the research I cite does so. This is for the simple reason that there is considerable overlap between the two concepts. For example, a significant amount of CRSV by conflict actors takes the form of nominally consensual relationships between civilians with very limited resources and soldiers with, relatively speaking, considerable power; this is precisely the dynamic that the (rather euphemistic) term ‘SEA’ is intended to reflect. Similarly, some ‘SEA’ encompasses stereotypical ‘CRSV’ behaviours, such as forcible rape.

Command, Control, and CRSV by Military Organizations

Early scholarly research on CRSV focused on cases in which sexual violence was so massive and systematic as to be genocidal.¹³¹ Systematic, strategic rape, however, remains the exception rather than the rule. More importantly in the context of this Commentary, strategic explanations for SEA by peacekeeping personnel are inexplicable as ‘strategy’: peacekeeping operations’ overarching strategic goal is reducing violence, including sexual violence.¹³² More recent work has focused on explaining variation, emphasizing that rape is not an inevitable by-product of war.¹³³ In particular, sexual violence frequently results from the failure of armed group (whether state, rebel, or peacekeeping) institutions, rather than from the successful implementation of a strategy.¹³⁴ By ‘institution’, in this context, I mean any relatively durable formal/written or informal/understood set of rules and practices.¹³⁵ In the armed group context, several key institutions may affect how combatants behave towards civilians, including whether they engage in CRSV or SEA. Institutions common to most armed groups include vetting, recruitment, training, formal and informal socialization, deployment, discipline, and promotion/advancement processes. Hoover Green’s The Commander’s Dilemma highlights a tension between these processes: to succeed, armed group leaders must recruit and train fighters who use violence unhesitatingly.¹³⁶ However, the normalization of violence frequently results in ‘spillover’¹³⁷ to acts of violence that armed group leaders do not prefer. Thus, the second prong of the commander’s dilemma: because maximal violence is usually not strategically beneficial, (most) commanders must also seek to limit violence to appropriate levels, times, types, and targets.¹³⁸

How do commanders thread this needle, in practice? They may attempt to recruit only fighters who appear unlikely to commit unordered violence, and/or recruit only voluntarily.¹³⁹ They may educate recruits about the laws of war, or indoctrinate recruits in an ideology that valorizes restraint.¹⁴⁰ They may (attempt to) impose strict discipline, such as a clear system of rules, rewards, and punishments regarding conduct toward civilians.¹⁴¹ They may ban or attempt to ban certain practices of socialization, such as violent hazing.¹⁴² They may offer career advancement only to fighters and officers who deploy violence in strategically useful ways. Of these institutional solutions, the most intuitive is probably discipline; after all, discipline is what we are speaking of when we speak of ‘ending impunity’, a common refrain with respect to CRSV/SEA.¹⁴³ And indeed, empirical results suggest that consistent discipline is associated with lower levels of unordered violence.¹⁴⁴ However, consistent discipline is not always possible (consider, for example, the difficulty of constant behavioural observation of an entire peacekeeping operation). Nor is it always effective. George Patton famously informed Moroccan leaders in 1942 that, while rapists would be promptly hanged, there would nevertheless ‘unquestionably be some raping’ during the American incursion.¹⁴⁵ Similarly, interviews with former combatants in El Salvador suggest that even Communist rebels, whose rules forbade sexual violence and called for severe punishment when it occurred, initially failed to prevent it.¹⁴⁶

Discipline fails for many reasons, all common in conflict and post-conflict settings. First, fear and other stressors shorten individuals’ time horizons, meaning that the risk of future punishment is unlikely to outweigh the immediate benefit (whether personal or social) of violence today. Second, in many contexts (including most non-conventional warfare and the disorder of post-conflict recovery), the probability of detection is low, even if the probability of punishment in case of detection is very high. Finally, with respect to sexual violence specifically, most fighters in most fighting forces perform, or even genuinely endorse, ‘militarized masculinity’, a set of beliefs and practices that emphasize physical toughness, denigration of the feminine, (hetero)sexual competency, homophobia, and male dominance.¹⁴⁷

Both in-depth qualitative and quantitative research suggest that armed groups wishing to limit CRSV/SEA must go beyond discipline, investing in institutions for ongoing training and socialization that valorize restraint and connect restraint to specific mission goals. For example, the institutional change that appears finally to have ended opportunistic sexual violence by rebel combatants in El Salvador was the implementation of formal political education across all units in 1982–83. Despite overall violence reaching a crescendo, rebel sexual violence, never widespread, dwindled to almost nothing during this period.

To the extent that the Commander’s Dilemma framework applies to peacekeeping operations, we would expect the highest levels of sexual abuse to be associated with specific units whose training, discipline, and socialization – in home militaries or in mission – failed to discipline sexual violence, de-emphasized restraint as an element of the mission, and/or strongly reinforced militarized masculinity. In the following section, I review what is known about the relationship(s) between peacekeeping, CRSV, and SEA, attending particularly to findings that support or disconfirm the analytical framework outlined above.

Peacekeeping Operations, CRSV and SEA

Research on peacekeeping operations and sexual abuse (including both CRSV and SEA) is limited, relative to the somewhat larger literature addressing wartime CRSV. However, what is known seems to support the view that institutions can help explain variation in sexual violence, at least to some extent. Examining data from the Sexual Violence in Armed Conflict dataset,¹⁴⁸ Johansson and Hultman find that, relative to a ‘baseline’ mission with neither a protection of civilians mandate nor a sizeable police contingent, missions with both a protection mandate and a police force were associated with significantly less sexual violence by armed actors.¹⁴⁹ Moreover, such missions are even more successful at preventing sexual violence when (formerly) warring parties have relatively strong, centralized command structures.¹⁵⁰ That is, peacekeeping operations can strengthen armed groups’ incentives to restrain sexual violence, but only groups with effective command and control fully succeed in acting on those incentives. Depending on the specific mechanisms of ‘control’ (not specified by Johansson and Hultman), this appears consistent with the institutional theory outlined above. Hultman and Johansson, in a separate piece, have also shown that peacekeeping onset is associated with the overall incidence of sexual violence in a given war, as well as with rebel sexual violence in particular, and with rebel strength.¹⁵¹ This finding suggests a caveat to the finding that peacekeeping operations can produce lower levels of CRSV: if more sexually violent armed actors are more likely to receive peacekeeping operations, then it is unclear whether these operations are causing the downturn in sexual violence.

However, other examinations of sexual abuse and peacekeeping operations seem also to support the idea that institutions can propel or restrain sexual violence and abuse. Given that in-mission military socialization is limited, Moncrief argues, the effect of home-country military socialization and training is conditioned primarily by in-mission discipline.¹⁵² These findings complement those of Horne, Robinson, and Lloyd, who find that ‘the behaviour of PKO [peacekeeping operation] contributor states toward their own populations strongly and consistently predicts the behaviour of these states’ military forces in UN PKOs’.¹⁵³ That is, despite dramatic differences between the objectives of traditional military operations and peacekeeping operations, the behaviours of peacekeeping personnel toward civilians match their home-country training and socialization, rather than the strategic imperatives of the new mission context.

In addition, following UNSCR 1325,¹⁵⁴ both scholars and advocates have considered the likely effect(s) of incorporating more women in peacekeeping operations.¹⁵⁵ A traditional view suggests that the presence of women (and particularly women’s substantive involvement) in peacekeeping operations should limit SEA, either because, in effect, women are less likely to perpetrate it, or because women have a restraining effect on male colleagues. Karim and Beardsley find some support for this notion, although the estimated effect is curvilinear: the marginal effect of adding women peacekeeping personnel diminishes as the proportion of women increases.¹⁵⁶ However, the causal relationship here is unclear. Is there a per se effect of women in the peacekeeping setting, or are women more likely to join militaries that are, independently, less likely to perpetrate SEA? Regardless of the precise causal relationship, the association suggests that improving women’s participation in peacekeeping operations should be a continued priority.

Discussion

If insights from institutional approaches are correct, a major goal of UN peacekeeping operations should be strengthening institutions, both in troop-contributing countries’ militaries and (perhaps counterintuitively) among parties to conflict. ‘Strengthening institutions' might mean several things from the perspective of the UN. First, with respect to CRSV perpetrated by other parties, peacekeeping operations cannot directly alter armed group institutions. Rather, peacekeeping operations must demonstrate that they can and will sanction sexual violence, thereby increasing the parties’ incentives to gain effective control over their combatants. Second, peacekeeping operations may wish to appeal to some groups’ desire to ‘win the peace’. However, as Johansson and Hultman show, even these indirect objectives are significantly more difficult without a civilian protection mandate and a police force, and consequently the mandate is of particular importance.

Third, the UN must invest in ideological strategies within its own forces, in particular de-normalizing sexual violence, exploitation, and abuse. Doing so requires that troop contributors create recruitment, training, and disciplinary institutions that reinforce UN norms of restraint, particularly around sexual violence, exploitation, and abuse. On the UN side, this work begins with vetting troop contributing armed forces for consistent discipline around civilian protection; it continues, crucially, by harmonizing home-country military training and socialization with UN expectations and values. Of course, the utility of these ideas in controlling SEA/CRSV is conditional on rebel group goals and structure, on troop contributing countries’ existing institutions and practices, and on any given mission’s mandate.

Finally, both scholars and practitioners need a stronger understanding of the relationship between SEA and CRSV.¹⁵⁷ The two appear somewhat correlated, but the relationship may not be causal. Most post-2003 missions report some SEA, and CRSV is associated with peacekeeping onset. There are important reasons to suspect a causal relationship, however. For example, SEA may weaken armed actors’ incentives to refrain from CRSV by sending a message to civilians that sexual abuses are not viewed as sufficiently serious to merit a response from peacekeeping operations. Or, SEA may undermine confidence in peacekeeping operations more generally, allowing erstwhile conflict actors to return to fighting or gain de facto control over ostensibly peaceful areas. Most insidiously, SEA by peacekeeping personnel may normalize sexual violence and abuse among civilians, as well as fighters.

‘Inside’ Armed Actor Processes and the Conflict ‘Outside’: Research on the Prevention of CRSV

Angela Muvumba Sellström

Uppsala University.

A feature of the United Nations (UN) Security Council’s Women, Peace and Security (WPS) agenda has been its orientation toward protecting populations from conflict-related sexual violence (CRSV).¹⁵⁸ While peacekeeping personnel may be tempted to understand their responsibilities as bulwarks of female vulnerability, the complexity of CRSV goes beyond assumptions about women’s defencelessness and to the heart of how armed groups inculcate sexual norms. Indeed, military organizations from the same social and cultural backgrounds do not all exhibit CRSV.¹⁵⁹ For some, sexual violence is not a preferred strategy. For others, these behaviours will vary over the many years of a conflict. How can peacekeeping operations deployed to dynamic situations anticipate and effectively prevent CRSV?

New trends in research demonstrate that some rebel groups employ political education and training of soldiers, shaping perceptions among their fighters that sexual violence is negative and consequential (see also for example Moncrieff and Wood; and Hoover Green in this Forum). I argue that peacekeeping operations’ efforts to prevent CRSV should start with the premise that understanding an armed actor’s own institutional culture is fundamental to prevention. Actors can and do tighten or loosen their control over their force members, even shifting expectations of proper sexual conduct. This is demonstrated in fluctuations in degree, intensity or type of violence over time; changes which if accurately observed and assessed, could provide entry-points and opportunities for peacekeeping prevention of CRSV.¹⁶⁰

In order to contribute to our understanding of these prospects for peacekeeping prevention, this commentary presents fresh insight from qualitative research on armed groups in Burundi’s civil war (1994–2008), CNDD-FDD (National Council for the Defence of Democracy-Forces for the Defence of Democracy) and Palipehutu-FNL (Palipehutu-Forces for National Liberation),¹⁶¹ discussed in terms of whether or not sexual violence permissiveness or prevention was constant over time. The material demonstrates that during strategic moments, the groups altered their institutional cultures (in part) as a response to needs for civilian support and based on their respective internal political and military aims. These results are relevant beyond Burundi. These armed organizations embody gradations in military and political structures and organizational approaches that are similar to many other rebel groups.

Burundi Civil War 1994–2008

Burundi carries the historical burden of political repression, organized massacres (in 1965, 1972, 1988, 1991 and 1993) and widespread displacement and exile. The civil war of 1994–2008 is a continuation of this legacy. The CNDD-FDD launched its armed rebellion against the Tutsi-led regime in 1994, as a response to the 1993 assassination of Burundi’s first democratically-elected Hutu president. The insurgency came to be composed of 8000 to 12,000 members and it instated stratified, formal political and military structures.¹⁶² Following several years of internal fragmentation, Pierre Nkurunziza emerged as CNDD-FDD’s chief leader. The Palipehutu-FNL also produced a dominant commander, Agathon Rwasa. Unlike CNDD-FDD, Palipehutu-FNL was small, just 2000 to 3000 fighters. It was less likely to receive weaponry or financing from regional leaders and relied mainly on support from local villagers in and around Bujumbura Rural. While Palipehutu-FNL was organized also along political and military structures, these were suffused with religious dogma from evangelical Christianity.

During the civil war, observers from civil society and international agencies reported high levels of sexual violence. In 2006, human rights advocates reported 1930 cases of sexual violence, approximately 5 victims every day¹⁶³ and that the armed actors in the war relied upon sexual violence to frighten the local population.

Permitting or Preventing CRSV

Of the two rebel groups, CNDD-FDD was more likely to be associated with, and permissive of sexual violence.¹⁶⁴ There is no evidence that FNL used or allowed CRSV in its assaults on civilians.¹⁶⁵ There is no evidence that levels of sexual violence coincided with patterns of other types of violence against civilians, for indeed CNDD-FDD committed less attacks on innocents than Palipehutu. Indeed, FNL was known widely for carrying out attacks on civilians, including an incident at Gatumba which has been described as particularly egregious but was still not associated with sexual violence.¹⁶⁶ Moreover, a subsequent version of the FNL later operated in Uvira, South Kivu, DRC, and its fighters were not reported to commit sexual violence as part of their operations (Van Acker, personal communication, November 18, 2014).¹⁶⁷ I argue that this outcome can be traced to distinctive internal institutional cultures, and to their respective positions and responses to the external conflict and peace processes. Strikingly, while CNDD-FDD and Palipehutu-FNL both put in place rules against sexual misconduct, including prohibiting rape, the nature and quality of these codes were different and changed over time.

Permissive to Preventive: CNDD-FDD

The CNDD-FDD’s long-term political ambition was to govern Burundi. The military trainees who founded CNDD-FDD presumed that sexual violence should be penalized. However, rule-making, enforcement, and dissemination was unclear, diffuse and of marginal institutional importance. Former CNDD-FDD members I have interviewed had no recollection of being trained or briefed about any negative repercussions for sexual violence. They could not recall seeing or discussing a written code of conduct. The rank-and-file could not anticipate punishment for committing rape. Some ex-combatants understood that the penalty was detention; others presumed it was corporal punishment; and still others believed perpetrators were executed. Commanders did not deliver negative messages against sexual violence, but generated positive or mixed signals about abuses. The CNDD-FDD created a permissive institutional culture.

Former members that I have interviewed highlighted that the opportunities to force sex from a girl varied considerably. Some members of CNDD-FDD could commit such acts openly, while others concealed this behaviour, particularly during operations. A commander who had status or prestige within the armed organization could have a reputation for coercing sexual partners without losing status within the group. Indeed, if their commander was also engaging in sexual violence, foot soldiers might feel assured that punishment would not be severe. Meanwhile, cohorts in other units might not have the same opportunity or similar leniency. Finally, only those perpetrators who violated children or relatives of higher-ranking officers seemed likely to face a penalty. A within-group double standard was normalized. The institutional culture made sexual violence inconsequential.

Despite its permissive institutional culture, CNDD-FDD’s prohibition changed over time. During the conflict, the rebel group showed little interest in prevention of CRSV. However, toward the end of the conflict, leaders such as Nkurunziza changed their approach. The group declared that it would investigate and punish perpetrators.¹⁶⁸ The majority of fighters I interviewed noted that the rule against sexual violence became more defined and severe after the conclusion of the 2003 ceasefire between CNDD-FDD and the transitional government. The period that followed entailed demobilization at assembly points and integration into the new national army, the Burundi National Defense Force (NDF).¹⁶⁹ At this point in time, the rebel army began to see itself as a government-in-waiting. It planned to participate in elections in 2005. Consequently, the CNDD-FDD sought also to have better relations with civilians, and needed to control its members.

What can we make of this change? One explanation is that the leadership sought to deter sexual violence for the purposes of credibility with the local population and the international community. Indeed, the UN made entreaties to Nkurunziza about the level of wartime rape in early 2004, prompting public attention to CRSV behaviour in combatant assembly areas.¹⁷⁰ Another source, a top-ranking CNDD-FDD officer,¹⁷¹ explained to me that CNDD-FDD took a long time to evolve and professionalize. During the war, the group relied on support from other governments, and could control the population with its guns and projection of power. Coercion of the population and a sense of entitlement on the part of CNDD-FDD generals had to change, ahead of elections planned for 2005. International pressure paralleled the CNDD-FDD’s increasing efforts to gain support from the population.¹⁷² Its leaders attempted to put in place a less permissive institutional culture.

Preventive to Permissive: Paliphutu-FNL

Throughout the civil war, Palipehutu-FNL actively invested time and energy in an instrument to prohibit sexual violence.¹⁷³ The mechanism was a code of conduct, shared through repetitive and recurrent training and indoctrination and maintained in written form. Former combatants from the other armed actors in the Burundian civil war, civil society experts, UN officials stationed in the country and local civilians have corroborated this with me. They emphasize that FNL’s punishment for rape was execution (even reports of incidences of rape make mention of the code).¹⁷⁴ FNL’s within-group messaging prioritized temperance. Members were barred also from consensual sex during the war. Drinking alcohol and smoking were forbidden. However, differences between consensual and coercive sex were clear. Offenders expected maximum and swift punishment for rape. Interviewees recall the execution of offenders. The predictability of punishment was so constant, individuals suspected of sexual violence preferred to defect. Ex-FNL members uniformly recount training and prayer sessions that inculcated a principle against CRSV. This institutional culture made sexual violence both status-reducing and shameful, and as such was preventive.

The Paliphutu-FNL code applied to foot soldiers as well as squad-level commanders, extending up the chain of command. While some commanders were demoted and spared execution, even the highest-ranking leaders of the group were not ‘free to commit rape’. Demotion in the midst of a peasant rebellion can be costly, resulting in fewer avenues for survival, access to information, material rewards and power. The armed group’s institutional culture was that sexual violence was consequential.

It is striking that Palipehutu-FNL sought to prevent sexual violence. It shared many characteristics with CNDD-FDD. Burundi’s pre-war context seemed permissive with regard to sexual violence, and the population from which both rebel armies recruited were familiar with forced marriage. The asymmetry in power and status between women and men and the general chaos of war could have led to weak or flexible prohibitions, as with CNDD-FDD. Why would FNL invest in rule-making, dissemination and enforcement to this degree?

Palipehutu adhered to the narrative of struggle for ethnic emancipation from Tutsi oppression.¹⁷⁵ At the same time, the group was also religious. Its members were expected to become born-again Christians. They referred to themselves as ‘God’s Army’. While this alone is not uncommon – consider the Lord’s Resistance Army (LRA) in northern Uganda, which had a so-called ten commandments but had a record of abduction and forced marriage¹⁷⁶ – FNL used its religious orientation to inculcate sexual discipline. Ex-FNL members uniformly recount that prayer sessions valorized sexual discipline and were platforms for calling-out offenders. Sexual misconduct was exposed through an imposition of ‘self’-confession. Actually, peers would put pressure on one another to confess transgressions.

Nevertheless, religious orientation does not explain why Palipehutu-FNL invested in preventing CRSV. A former CNDD-FDD combatant now journalist believes that the FNL code was not driven by genuine religious beliefs on the part of its members, but rather, used as a rhetorical tool and socialization mechanism in order to maintain cohesion and capability.¹⁷⁷ The FNL discipline was necessary for achieving a common goal. Rules were also a way of normalizing the notion that good fighters could (and had to) have self-control.¹⁷⁸ Palipehutu-FNL’s prohibition contributed to accountability for personal conduct and projection of the individual as part of a collective, directed toward a shared cause. The group’s attitude also seemed attuned to the welfare of the local population – the Hutu majority for whom FNL relied upon for food, shelter, information and material resources. Its prohibition allowed members to coexist fraternally, while avoiding negative entanglements with the civilians they depended upon.

Startlingly, Palipehutu-FNL’s institutional culture of prevention shifted in the last stages of the civil war. As the pressure to negotiate a settlement mounted, Rwasa acquiesced to demands from other leaders in Palipehutu to soften punishment practices. Some commanders married or cohabited with girlfriends while the group awaited the conclusion of settlement talks during assembly for disarmament and demobilization. The FNL leadership formally reduced the punishment for sexual violence to demotion or physical hardship. The death penalty no longer applied. By 2008, it allowed its armed fighters to engage in consensual sexual relations. This strategy also seemed to be geared toward increasing the rebellion’s numbers as FNL was encouraging indiscriminate enrolment of new recruits. The leadership wanted to swell its numbers to secure more positions in the state’s political apparatus and military. Rwasa had to respond to internal organizational politics, particularly since his efficacy in negotiations was contingent upon the support of all the rebel group’s factions.¹⁷⁹ Human Rights Watch noted that ‘in the past, the FNL has exercised tight control from the centre. However, this control decreased in late 2007 and early 2008, when increased numbers of rapes, cases of assault, and robberies were also attributed to FNL members, many of them [by] recent recruits’.¹⁸⁰

Conclusions: Peacekeeping Entry-Points and Opportunities for CRSV Prevention

These dynamics offer entry-points and opportunities for peacekeeping operations. First, as the material has shown, the institutional cultures of rebel groups can change. Information about the internal orientation of rebel groups should be targeted for collection and analysis by peacekeeping operations. Insights about armed actors’ institutional CRSV cultures could be integrated more often in operational planning, implementation, monitoring and reporting to the Security Council. Second, changes in permissiveness or prevention occurred at particular moments in conflict and peace processes. When peacekeeping operations understand how events and timing correspond to CRSV behaviour they will be better placed to implement their mandated responsibilities.

Peacekeeping operations may see these as opportunities in two ways. First, they can better anticipate when an armed group may be more likely to be permissive of CRSV, and if civilians are threatened. Operational staff should know well the leadership of armed actors; their political histories and leanings; the actor’s training and disciplinary practices; who provides them with arms, information, shelter and food and supplies; and if there is active civilian cooperation and support, or not. This information is already collected in other aspects of peace operations, but should be integrated with WPS (see Lotze in this Forum) as well as used to encourage better CRSV policies by security actors. Second, peacekeeping operations can mobilize international and regional bystanders to put in place special measures to prevent CRSV during turning points in the conflict. Peacekeeping prevention should be particularly mindful of periods when armed groups are not fighting and combatants are more likely to interact with civilians. Armed groups are dynamic actors. They may become more laissez-faire and CRSV incidents could increase. Counterintuitively, periods of protracted negotiations or assembly for demobilization and disarmament may see weakening of internal disciplinary codes by commanders. It is during these periods that proactive advocacy and engagement with armed actors can be impactful. Finally, peacekeeping operations should avoid perceptions of partiality and address the potential threat of CRSV by all armed actors, not only any so-called usual suspects. Effective prevention means sending consistent and coherent messages that CRSV cannot be brushed aside.

Engagement and Positive Role-Modelling in the Prevention of Conflict-Related Sexual Violence

Ann-Kristin Sjöberg

Fight for Humanity, Geneva

It is human to focus on the negative, as our ‘fight or flight’ response has taught us. Nevertheless, it is also human to find solutions to problems, to learn from those we identify with and adopt new behaviours. Building on my own experiences and research, the commentary focuses on presenting lessons from working with armed non-state actors (ANSAs) in order to promote the prevention of conflict-related sexual violence (CRSV). I have two fundamental starting points. First, most acts of violence during armed conflict will be a crime under national law, including CRSV. Yet, during armed conflict the ‘normal’ legal system is by default not functioning, as the state has lost the monopoly on the use of armed force. In these circumstances the law of armed conflict or International Humanitarian Law (IHL) applies. Rape and other types of CRSV has been prohibited by international law for over a century, in both international and non-international armed conflicts, for all conflict actors.¹⁸¹ Differently from other war crimes, CRSV is always a war crime, also when committed against combatants. ¹⁸² Moreover, not just the direct perpetrators, but also persons that allow, encourage or influence the commission of CRSV can be held responsible under international law.¹⁸³ A complication is that while the 70-year old Geneva Conventions are amongst the most universally accepted treaties, they lack comprehensive accountability mechanisms. Hence, efforts to improve the respect for IHL largely rely on voluntary measures taken by the conflict parties themselves, State and non-State.

Second, in spite of being organizers of violence, most ANSAs care about their reputation and establish self-images according to their own norms. Indeed, there are ANSAs for which sexual violence is the worst crime that their members can commit. Under such circumstances, the perpetrators, be they members or civilians in the areas ANSAs control or influence, receive severe punishments. Personally, I have met commanders of ANSAs that have ordered executions for such acts. In another case, sexual violence had been stigmatized to the point that one ANSA member was horrified to hear that their recruitment of children was considered as equivalent to an act of rape. On other occasions, ANSAs commanders have angrily asked for evidence that they were ‘using sexual violence as a weapon of war’.

Based on this, I discuss practical lessons from engaging ANSAs on IHL and CRSV. This involves how to sell the topic, how it is positioned in relation to sexual violence more broadly, and the role of accurate information. The essay further explores lessons from research on informal socialization and positive role-modelling and finally conclude with some suggestions for possible action.

Lessons from Engaging ANSAs on IHL and to Prevent CRSV

While some ANSAs make reference to IHL, quite a few generally or explicitly prohibiting CRSV,¹⁸⁴ in their internal rules and regulations, there is a global debate on how to improve compliance. At a fundamental level, this means that the ANSAs and their fighters must have knowledge of these rules. Moreover, ANSA superiors should also be aware that they can be held accountable when their subordinates commit acts which violate IHL. This means that it should be in the interest of ANSA leaders to promote IHL. Hence, this is an obvious starting point for engaging these actors in order to prevent CRSV. Pragmatically, there are some key steps to be taken with this approach: obtaining the greenlight to conduct dissemination sessions with ANSA members, supporting the modification of rules, and their translation into military doctrine and rules of engagement, etc. However, the road to getting an agreement and having it implemented, is often bumpy. For ANSAs, there are some additional difficulties in this process as compared to states, particularly linked to security, access, and existing capacities. Still, most of the time, if you have done your homework by having studied the organization and its behaviour, and made a good target selection, the leaders will be interested in listening to what you have to say.

The Need to ‘Sell’ the Topic

My experience is that the decision to go along in a dialogue and dissemination process will in many cases depend on how you present the issue. With my former organization, the NGO Geneva Call, we created and promoted the ‘Deed of Commitment under Geneva Call for the Prohibition of Sexual Violence in Situations of Armed Conflict and Towards the Elimination of Gender Discrimination’.¹⁸⁵ In the beginning of this process, it was debated if including a broader discussion on gender discrimination would not deter more conservative ANSAs from signing the Deed. The fear was that they would not be in favour of gender equality and this, in turn, would negatively affect their commitment to preventing CRSV. Surprisingly, it turned out many ANSAs were interested in discussing gender equality and more easily recognized facing challenges here rather than relating to CRSV. The focus on gender equality hence allowed for gender – and sexual violence, conflict-related or not – to be put be put on the agenda and also allowed us to access female members of the ANSA, a group which we otherwise often were restricted from approaching. Moreover, it helped mitigate ANSAs’ concerns that the fact that they were addressing the CRSV issue would look like as if before they were deliberately practicing CRSV.

The Need to Focus on Sexual Violence More Broadly

Another lesson learned concerns the limitations of the CRSV concept itself. Human rights are indivisible and belong to everyone. As such, the legal rights to protection from sexual violence are extended to everyone in an armed conflict, including fighters. In relation to the dialogue with ANSAs, it’s my experience that if the framing of ‘conflict-related’ sexual violence was modified to addressing also other aspects of sexual violence, interlocutors are more open to address the issue. First, this term appeared to be simply confusing to them. Being placed within a broader frame of armed conflict, struggle or armed violence, it is difficult for ANSAs to draw the border between what is related or not to the conflict. Moreover, only focusing on ‘public’ violence is too limiting to some ANSAs, as a main concern is related to domestic violence by their members or by civilians in areas they control or influence. Secondly, many have amongst their ranks and constituencies victims of sexual violence and torture, victimized by other conflict actors, by civilians or by other ANSA members. Yet, there is a perception and response gap to the fact that sexual violence affects ANSA and their members in many different ways.

The Need for Accurate Information

In an ideal scenario, ANSAs have a well-functioning internal monitoring mechanism. In reality, they often have limitations in their internal communications, and CRSV expertise. Under such conditions, external sources of information are important. In fact, even informal mechanisms can help give results. For example, in one context I was involved in, relating to IHL, a report from a well-established human rights organization had been released just days before a high-level workshop we were organizing to raise support for an IHL training process with an armed actor. Although our expectation was that the ANSA would not show up, all the high-level leaders attended the workshop, very upset with the report and the fact that they were being criticized. Later that day they committed to the training process and, later on, to develop an IHL compliant Code of Conduct.

Another example specifically related to CRSV, concerns a research process I was involved in to find data on the types of violence committed against women and girls by ANSAs in a specific subregion, in order to inform the advocacy work toward specific ANSAs. The study, conducted by external researchers, found extensive use of violence against women and girls (notably threats and killings), but limited evidence that the ANSAs were using CRSV strategically. There few cases of CRSV found appeared to be perpetrated by individual members rather than as a policy. The study became a main door opener for working on gender with the two target ANSAs examined, as it showed our interest in understanding the actual situation.

Next Steps: Learning from the Research

Informal Socialization Processes

One of the main findings of the 2018 ICRC ‘Roots of Restraint’ study was that we need to emphasize values as well as the law.

How informal socialization mechanisms work within an ANSA can explain, for example, the gap between the protective standards in internal rules and regulations and lack of implementation in reality. If a Code of Conduct sounds ‘too good’ and is framed in standard language, most likely it doesn’t reflect the internal culture of an organization. ‘Borrowed’ rules or rules adopted to please an external audience will not make a significant difference: even the best rules would be disrespected if the organization or its subunits follow different rules. Simply supporting the drafting of a Code of Conduct or conducting training for an ANSA doesn’t do it. We need to understand what are the values that they display. Specifically related to CRSV, and, sexual violence more generally, we need to understand the gender relations of the organization and the society it forms part of. What does a prohibition of sexual violence actually mean to them: what do they understand by rape, by sexual violence? What are the estimated consequences (organizational and individual) of sexual violence? Are prohibitions explicit enough (not just ‘you can’t go around and abuse people’)? Are there certain categories whom they don’t see as covered (such as men and boys,¹⁸⁶ fellow combatants, enemy fighters or certain categories of civilians)? Are these understandings influenced by the ‘dehumanization’ processes that take place in war,¹⁸⁷ and elsewhere?¹⁸⁸

Positive Role-Modelling and Incentives

From my experience, ANSA are also open to role-modelling/positive peer-pressure, notably in relations to other ANSAs for which they may have admiration or respect. I have on numerous occasions taken examples from an ANSA in one continent and shared them with ANSAs in another context. Yet, so far this has been easier to do on gender equality and women’s participation rather that CRSV for two reasons. First, CRSV is stigmatized and sensitive. Second, there are simply more good practices and positive examples available to share and mimic on gender equality than for CRSV.

The value of positive role-modelling has also been evidenced in different studies.¹⁸⁹ Yet, when it comes to prevention beyond negative examples (deterrence) we’re often limited in our toolbox. How do we generate positive dynamics and prime the ‘non-rape’ and the ‘non-abuse’?

An advisor on sexual and reproductive health once proposed promoting ‘consensual sex’ and ‘consent’ also towards ANSA members. This could work in situations where violence is perpetrated because of lack of knowledge of the law and of the concept of ‘consent’ or because of perceived ‘gains’ for the perpetrators, and not when the goal is humiliation and pain for the victim and/or a community. Approaching ANSAs and offering education on sexually transmitted diseases while raising the issue of sexual violence is also possible, as well as having them nominate ‘gender ambassadors' or by conducting targeted campaigns.

Yet, if the aims and strategies (outspoken or de facto) of the ANSA by nature violate IHL, the impact of a direct dialogue may be slow or counterproductive. In fact, given limited resources, in some contexts we may do better in concentrating our efforts to the ANSAs where CRSV is simply tolerated as a practice.¹⁹⁰ Or, if our assessment is that the ANSA may listen to certain community members or high-level influencers, such an indirect approach may be more productive.

Moreover, research in neuroscience¹⁹¹ has shown that showcasing violations may have a counterproductive effect – contributing to more, not fewer violations – by creating a link in the minds of people between certain groups of people (for example an ethnic group) and certain categories (people to be ‘victimized’). Repeated exposure to images and accounts of violent acts can also normalize such acts to our brains and simultaneously strengthen the perpetrator group’s identity. But doesn’t ‘naming and shaming still work’? My experience is that it can help push armed actors towards cooperation. Yet, this usually requires other favouring factors or incentives, such as: (1) there is also a solution being offered, such as a training process and/or other support; and (2) leverage is exercised by allies or other influential actors which provide an incentive to take positive and proactive action. But reports on violations should not only document abuses: many armed actors are operating in difficult contexts where they may be coping with humanitarian crisis at the same time as they are fighting a war, with various degrees of support – or not – from international actors. Not recognizing the positive aspects and the potential gaps in their capacities and resources doesn’t help build their willingness to take action.

Conclusion and Importance for Peacekeeping

Based on a discussion on some lessons from engaging with ANSAs to commit to IHL and to prevent CRSV, and on research on informal socialization and positive role-modelling, a number of conclusions can be drawn. These are to various degrees relevant for peacekeeping operations on the ground. Notably:

• Working on gender issues can be a door opener with ANSAs, as discussions on CRSV are sensitive and provides high risks and limited ‘carrots’ for them.

• Focusing on sexual violence and not just CRSV could help create and maintain the interest of ANSA interlocutors.

• Context-specific and targeted data on violations provides a basis for advocacy work towards ANSAs and allows them to find solutions that fit the actual problem. If well done and acknowledging efforts undertaken it can also help build confidence in the dialogue process.

• Positive role-modelling is a as way of helping ANSAs accept a change in behaviour. Some ideas for how this can be done have been presented, notably linked to the topics of gender equality, consensual sex and sexually transmitted diseases, but this would need further development.

Clearly, efforts towards ANSAs would benefit immensely from cooperation and support from peacekeeping forces. Importantly, peacekeeping policies and concrete on-the-ground measures that create the space for humanitarian/human rights/peace actors to engage in dialogue with ANSAs in prevention, would be pivotal, particularly given existing security or legal hurdles.

Placing Peacekeeping Prevention of Conflict-Related Sexual Violence in the Context of Political Violence Targeting Women

Roudabeh Kishi

The Armed Conflict Location & Event Data Project [ACLED]

Research and data collection initiatives have found that conflict-related sexual violence (CRSV) is prevalent.¹⁹² In this light, it is an important atrocity that peacekeeping operations increasingly are mandated to address and prevent. While agreeing with the urgency of this mandate, this commentary argues that other forms of political violence – such as abductions and forced disappearances, as well as mob violence – also disproportionately target women.¹⁹³ We must therefore avoid a ‘reliance on dominant narratives’ in which sexual violence alone overshadows the other concurrent threats that women especially face.¹⁹⁴ To that end, this commentary argues that it is important that peacekeeping operations provide security equality to the population. This means that different groups should be equally protected by considering which form of threats affect each groups’ security and then adapt protection accordingly.¹⁹⁵ In this vein, peacekeeping operations must recognize that some threats that women face within conflict and post-conflict settings are different than those threats which target the population at large. For example, men are more likely to be participants in political violence, driven in part by ‘honour ideology’¹⁹⁶; women are more likely to be targeted through sexual violence, abductions/forced disappearances, and mob violence.¹⁹⁷ In this light, it is important for peacekeeping operations to adapt to the various contexts in which they are deployed to account for such variance. Based on new data from the Armed Conflict Location & Event Data Project (ACLED), this commentary consequently asks: how can peacekeeping operations protect women from the spectrum of violence which they face if the discussion regarding violence targeting women is limited to only one aspect of that full spectrum?

Introducing New Data from ACLED¹⁹⁸

Until recently, data did not exist to capture the spectrum of political violence targeting women (PVTW). New data from ACLED seek to address this gap.¹⁹⁹ These data record the various forms of political violence targeting women, and the perpetrators of this violence, both during and outside of conflict contexts. Data are updated weekly, allowing for analysis to be based on the most current information. Researchers compiling these data rely on information from not only traditional media, but also from select ‘new media’, reports from international organizations, and, importantly, from partnerships with local organizations and conflict observatories gathering information and data.²⁰⁰ The latter, in addition to prioritization of local media, is integral.²⁰¹

Violence Targeting Women: What Peacekeeping Operations Should Know

In the ACLED data, PVTW is understood as ‘the use of [physical] force by a group with a political purpose or motivation’ in the public sphere in which women and girls are the only, the majority, or the primary targets.²⁰² The data focus on five different types of political violence: (1) sexual violence, (2) attacks that are non-sexual in nature, (3) abductions and forced disappearances, (4) mob violence, and (5) explosions and other forms of remote violence. The difference in the prevalence of these forms across different regions is depicted in Figure 1.²⁰³

Figure 1. Political violence targeting women, by sub-event type.

Sexual violence disproportionately affects women, and such events are quite common during conflict contexts. This violence makes up well over one-third of PVTW in both Africa as well as Southeast Asia. Numerous incidents of rape have been reported during the South Sudanese Civil War at the hands of both the state military as well as the Sudan People's Liberation Movement-in-Opposition (SPLM/IO) (the primary rebel group). Efforts by the United Nations Mission in South Sudan (UNMISS) to proactively patrol and to escort women leaving protection sites has helped to deter and prevent some of these continued attacks.²⁰⁴

Despite the conversation around gender-based violence focusing almost exclusively on sexual violence, attacks of a non-sexual nature are the most prevalent form of PVTW. This type makes up over two-thirds of PVTW in the Middle East, South-eastern and Eastern Europe and the Balkans, the Caucasus and Central Asia, as well as in Latin America and the Caribbean. In Iraq, especially at the height of the war in 2016, women faced a very serious threat at the hands of ISIS, with reports of women facing mass executions in Mosul, as well as elsewhere in the country. In Ukraine, there are reports of women targeted and killed in Crimea and the Donbas region – the site of ongoing fighting between separatists and state forces – by militants as well as Ukrainian Armed Forces.

In addition to sexual violence, abductions and forced disappearances disproportionately target women. This violence is most commonly reported across Africa. Women in conflict zones are especially susceptible to this type of violence at the hands of a myriad of actors. The Lord’s Resistance Army (LRA) is responsible for a number of abductions of women and girls in recent years in both the Congo as well as in the Central African Republic (CAR). Both states are home to ongoing conflicts, though in both cases the LRA is not one of the primary conflict parties. The group has been able to take advantage of the security vacuum that is created as a result of the larger conflict, making the threat to local women even more pervasive.

This is similar to the case with mob violence, in which spontaneously organized mobs engage in violence targeting women. This violence too disproportionately impacts women, and in conflict settings can be exacerbated. While mobs, given their informal nature, are not parties to large-scale conflicts themselves, they can become more active in the vacuum that such contexts create. This can be seen in CAR. While levels of violence decreased significantly between the Séléka Rebel Coalition / Séléka Miltia and Anti-Balaka in 2017/2018 relative to the height of the war in late 2013/early 2014, civilians continued to be targets of atrocities by both sides, with reports of summary executions and mutilated bodies left exposed to terrorize populations.²⁰⁵ In this context, vigilante mobs began targeting women blamed for ‘witchcraft’, which they believed was the only explanation for ‘‘the occurrence of otherwise inexplicable misfortune’,²⁰⁶ which may stem from large-scale conflict.²⁰⁷ This type of violence continued in parts of the country even despite the deployment of United Nations Multidimensional Integrated Stabilization Mission in the Central African Republic (MINUSCA) peacekeepers contributing to improvement in the security situation.

Lastly, explosions and other forms of remote violence can also be used to target women. This type of violence is rarer given that the remote nature of these events, by definition, makes using them in targeted attacks where gender is a salient identity more difficult. Despite this fact, such violence can be a fixture in certain conflicts and regions (see Figure 1). The blowing up of girls’ schools has occurred on more than one occasion in 2019 alone in Afghanistan by armed militants (assumed to be the Taliban), despite attempts by the United Nations Assistance Mission in Afghanistan (UNAMA) to improve girls’ access to education in the country.

Recognizing the Diverse Group of Perpetrators

The primary perpetrators of PVTW vary across regions, with patterns largely mirroring those of political violence strategies at large across those regions. The data focus on seven different types of perpetrators of this violence: (1) unidentified or anonymous armed groups, (2) state forces, (3) political militias, (4) external forces, (5) mobs, (6) identity militias, and (7) rebel groups. The difference in the prevalence of these perpetrators across different regions is depicted in Figure 2.

Figure 2. Political violence targeting women, by perpetrator.

Anonymous or unidentified armed groups are the primary perpetrators of PVTW; over one-third of all reported PVTW has been perpetrated by such agents. These agents can be unknown due to insufficiently detailed reporting, or as a result of strategic anonymity – doing the bidding of others who have ‘outsourced’ violence to them. Given the role they play in conflict settings, where they may act on behalf of conflict parties, including them in assessments of threats to women is integral.

State forces, such as military or police forces, are the primary perpetrators of PVTW in the Middle East, Southeast Asia, and the Caucasus and Central Asia. While these are not limited solely to conflict settings – for example, Israeli state forces active in the West Bank are one of the primary perpetrators of this violence – they are especially prevalent in such contexts. During the Yemen Civil War, women have been targeted as the victims of snipers by both pro-Houthi and pro-Hadi soldiers.²⁰⁸

Political militias are armed, organized political gangs, who act on behalf of political elites; sometimes these political elites may be government affiliates, making them into pro-government militias. Such groups are active in a variety of settings, including in conflict settings where they carry out violence on behalf of primary conflict parties. In Sudan, there are numerous reports of attacks and rapes by the pro-government Janjaweed, especially targeting those in IDP camps. Mass killings by the Janjaweed were what initially spurred the deployment of the United Nations – African Union Mission in Darfur (UNAMID) to Darfur in 2007. In Syria, there have been numerous reports of pro-regime militias shooting, killing, and raping women, especially at the height of the war in 2017.

External or other forces refer to actors such as state forces active outside of their home country and multilateral coalitions, as well as peacekeeping forces. These agents make up about one-tenth of PVTW in the Middle East. In Yemen, there are numerous reports of coalition forces, part of Operation Restoring Hope, targeting women and girls; these include the bombing of girls’ schools by Saudi airstrikes as well as rapes of women by Sudanese military forces. While those who are parties to internationalized conflicts can be engaged in PVTW, more damning is when peacekeeping forces themselves engage in such violence; a number of rapes by MINUSCA agents reported in CAR in 2014 and 2015 are an example.

Mobs, who can have links to political parties or religious groups, or can be vigilante groups taking what they perceive to be justice into their own hands, may be the primary perpetrator of PVTW in South Asia, but they are active outside of that region as well. In Ethiopia in 2009, a UN national staff member was injured when her vehicle was attacked by a mob throwing stones at her – signalling how PVTW can impact aid workers and peacekeepers as well as the local population.

Identity militias are armed groups that are organized around a collective, common feature including community, ethnicity, region, religion or, in exceptional cases, livelihood; these violent groups often act locally, in the pursuance of local goals, resources, power, security, and retribution.²⁰⁹ Such groups are active in conflict contexts where they can act as local security providers for their communities or can fight alongside conflict parties. In Somalia, where Al Shabaab continues to battle successive Somali governments as well as the allied AMISOM forces, local clan militias are especially active. There are numerous reports of these clans targeting other clans – killing, abducting, and/or raping women and girls – as a part of clan-based fighting over livelihoods.

Lastly, rebel groups are armed, organized groups with a stated political agenda for national power, either through regime replacement or separatism.²¹⁰ Such groups make up significant proportions of PVTW in both the Middle East as well as Southeast Asia. Myanmar is home to a number of such rebel groups who contribute to a complex conflict environment. Conflict has been ongoing between the government and a number of ethnic armed organizations (EAOs) for decades – with women often caught in the crosshairs. There are reports of EAOs, such as the Kachin Independence Army, abducting girls and holding them hostage as punishment for family members refusing recruitment, or the Ta'ang National Liberation Army detaining women accused of being military informants.

Conclusions and Implications for Peacekeeping Operations

The trends and examples in the ACLED data underline the importance of peacekeeping operations adopting a comprehensive and substantiated understanding of political violence targeting women and the role CRSV can play in this dynamic. Understanding context specific trends in violence enables the continued development of effective preventive measures for conflict and post-conflict settings. This was visible in the examples of UNAMID efforts to handle Janjaweed and the AMISOM forces in relation to Al Shabaab. Furthermore, the analysis underpinning peacekeeping operation mandates and operation plans must recognize that the threats that women face vary from those which target the entire population at large.²¹¹ The ACLED data thereby underline the importance of the recommendation in the UN Secretary General’s 2019 report on Women, Peace and Security of conducting a gender aware conflict analysis. In this analysis, the data strongly suggest that we must recognize that the threats that women face vary from those which target the entire population at large and that analysis needs to extend beyond the narrative of CRSV. Moreover, as the primary perpetrator groups vary across regions and conflicts, this means that peacekeeping engagement with the concerned conflict parties must, by definition, vary (a point also underlined by the other contributions to this Forum). What this means for peacekeeping operations is that not only is security equality important to recognize (i.e. women face unique threats), but also that this security equality can vary across contexts (i.e. the unique threats that women face are not the same across all conflicts and regions). This means that it is important to have a full mandate of protection of women as civilians and that approaches must be tailored to each context to be appropriate. Gender aware conflict analysis recognizing women’s agency will also ensure that preventive peacekeeping operation policies and approaches avoid stereotyping women only as victims.

Concluding, the ACLED data suggest that the need for a nuanced understanding of violence targeting women extends beyond peacekeeping contexts and into the future: efforts must ‘take into account the widespread and insidious political violence that undermines the ability of women to participate meaningfully in all political processes, including peace negotiations, post-conflict reconstruction efforts, and civil society activism’.²¹²

Acknowledgements

This Forum originated in discussions at the workshop The Prevention of Sexual Violence by Armed Political and Security Actors held on 13–14 of December 2018 in Uppsala, Sweden. Presenters included UN Special Representative on Sexual Violence in Conflict Pramila Patten and representatives from the ICGLR Regional Training Facility on Prevention of Sexual and Gender Based Violence, South African National Defence Force (retired personnel) and the Nordic Center for Gender in Military Operations. We would like to thank all participants for their insightful contributions and in particular the authors who contributed to this Forum. We are very grateful for our constructive collaboration with International Peacekeeping and would like to express a warm thank you to Govinda Clayton, Nina Wilén, Robert U Nagel, Alistair James Knott and the anonymous reviewers. We have also greatly benefited from Christiana Lang's assistance and wish to thank Chiara Tulp, in particular, for her contributions to arrangements for the workshop and in editing the Forum.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

About the Authors

Louise Olsson is Senior Researcher at the Peace Research Institute Oslo (PRIO) and the former Senior Advisor on Women, Peace, and Security at Folke Bernadotte Academy, Sweden. Her research focuses on gender aware peace agreement implementation, state strategies to gender equality and mainstreaming in the military, and the role of gender inequality for conflict risk. For example, she is the editor of ‘Gender, Peace and Security: Implementing UN Resolution 1325’ (Routledge 2015) and ‘Women and International Peacekeeping’ (Frank Cass 2001) and the author of articles appearing in Global Governance and Journal of Global Security Studies.

Angela Muvumba Sellström is an assistant professor at Uppsala University’s Department of Peace and Conflict Research, Sweden. She leads research projects on wartime sexual violence and contributes to research collaborations on international interventions, urban violence and civil wars and the non-permanent members of the UN Security Council. She has over 20 years of policy and research experience and has worked at the Folke Bernadotte Academy, the Nordic Africa Institute, the African Union Commission and think-tanks in New York, Cape Town and Durban.

Stephen Moncrief holds a PhD in Political Science from Yale University. His research focuses on peacekeeping, statebuilding, and international intervention. Starting in July 2020, he will be a Postdoctoral Scholar in the Center for Strategic Studies at the Fletcher School of Law and Diplomacy.

Elisabeth Jean Wood is the Crosby Professor of the Human Environment and Professor of Political Science, International and Area Studies at Yale University. Her work focuses on political violence, particularly conflict-related sexual violence, and its legacies.

Karin Johansson is a PhD candidate in the Department of Peace and Conflict Research at Uppsala University, Sweden. Her research focuses on conflict-related sexual violence in relation to peacekeeping and other types of external interventions. Together with co-authors, she has previously published in International Conflict Resolution, Conflict Management and Peace Science as well as in International Peacekeeping’’.

Walter Lotze is a Mason Fellow Mid-Career Masters in Public Administration Candidate at the Harvard Kennedy School of Government. He has worked with non-governmental organizations, research centres, the African Union and the United Nations on peace and security issues in Africa. He holds an MA in International Relations from the University of Stellenbosch, and a DPhil in International Relations from the University of St Andrews.

Chiara Ruffa is an academy fellow at the Department of Peace and Conflict Research at Uppsala University and associate professor in War Studies at the Swedish Defense University. Her research is about civil-military relations, peacekeeping operations, state militaries, norms, and cultures. She is the author of ‘Military Cultures in Peace and Stability Operations’ (University of Pennsylvania Press, 2018) and of ‘Composing Peace. Mission Composition in UN Peacekeeping’ (with Vincenzo Bove and Andrea Ruggeri, Oxford University Press, 2020).

Amelia Hoover Green is Associate Professor in the Department of Politics at Drexel University, and a consultant with the Human Rights Data Analysis Group (San Francisco, California, USA). Her research interests include armed group institutions, violence against civilians, and quantitative methods in political violence. In addition to her 2018 volume The Commander’s Dilemma: Violence and Restraint in Wartime (Cornell University Press), her work has been published in the Journal of Peace Research, Demographic Research, and the Journal of Global Security Studies.

Ann-Kristin ‘Anki’ Sjöberg is Co-Director and co-founder of the NGO Fight for Humanity. She holds a PhD in International Relations focused on Political Science from the Graduate Institute of International and Development Studies, Geneva, Switzerland. Her research interests are armed non-state actors and their restraint and use of violence, and human rights and gender in conflict and situations of violence.

Roudabeh Kishi is Research Director of the Armed Conflict Location & Event Data Project (ACLED) — a disaggregated data collection, analysis, and crisis mapping project. She holds a PhD from the University of Maryland at College Park in Government and Politics, with a focus on quantitative methodology. Her work has appeared in the Journal of Conflict Resolution, Terrorism and Political Violence, and Research & Politics, amongst others.

Additional information

Funding

The Forum is funded through the project Disciplining Fighters: Understanding Armed Political Actors’ Control of Sexual Violence supported by the Swedish Research Council [grant number VR 2015-03094]. Chiara Ruffa gratefully acknowledges support from the Swedish Research Council [grant number 2017-02139] and the Royal Swedish Academy on Letter, History and Antiquities.

Notes

1 see, for example, Oladipo, UN Peacekeeping Nightmare.

2 For example, Nordic Center for Gender in Military Operations organized an expert meeting in November 2019 with a specific focus on Conflict-Related Sexual and Gender-Based Violence. See NCGM, “Policybrief.”

3 While important discussions, this means that the more encompassing aspects conflict prevention and CRSV as well as the broader effects on CRSV on political and social developments will be outside of the scope of this Forum. For a discussion on these, see, for example, Davis and True, “Handbook” and Kreft, “Gendered Conflicts.”

4 Muvumba-Sellström, “Toward Indicators of Prevention?”

5 Ergo, we do not make a causal claim but seek to explore the potential connection and what that means for research and practice.

6 Hultman and Muvumba Sellström, “WPS Protection of Civilians.”

7 Cohen, “Ties that Bind”; Hoover Green, Repertoires of Violence, “Commander’s Dilemma”; Wood, “Typology of Political Violence.”

8 Wood, “Variation in Sexual Violence”, “When is Wartime Rape Rare?”, “Typology of Political Violence”; Hoover Green, Commander’s Dilemma, “Commander’s Dilemma”; Muvumba Sellström, “Stronger than Justice.”

9 Alison, “Wartime Sexual Violence”; Eriksson Baaz and Stern, Complexity of Violence, Weapon of War?; Marks, “Sexual Violence in Sierra Leone”; Stern and Nystrand, Gender and Armed Conflict.

10 For a discussion, see Kreft, “Responding to Sexual Violence.”

11 Wood, “Variation in Sexual Violence”, “When is Wartime Rape Rare?”; Cohen and Nordås, “SVAC Dataset”; Muvumba-Sellström, “Stronger than Justice.”

12 See, for example, Davies and True, “The Politics of Counting and Reporting Conflict-related Sexual and Gender-based Violence.”

13 Cohen and Nordås, “SVAC Dataset.”

14 “Stronger than Justice.”

15 For further discussion, see also Touquet and Gorris, “Out of the Shadows” and Schultz, “Displacement from Gendered Personhood.”

16 Speaking also to the protection logic, see Muvumba-Sellström in this Forum and Nagel, “Talking to the Shameless” for an overview.

17 Crawford, Wartime Sexual Violence.

18 On variation in pattern, see Brownmiller, Against Our Will; Wood, “Variation in Sexual Violence,” 307–42; Cohen, “Explaining Rape During Civil War,” 461–77; Cohen and Nordås, “SVAC Dataset,” 418–28; Cohen, Rape During Civil War. On sexual violence adopted as a policy versus a practice, see Wood, “Policy Implications of Recent Research,” 457–78; Wood, “Typology of Political Violence,” 513–37; Seelinger and Wood, Practice of War.

19 Touquet and Gorris, “Out of the Shadows?”; Myrttinen and Daigle, “When Merely Existing is a Risk.”

20 Wood, “When is Wartime Rape Rare,” 131–61; Hoover Green, “Creating and Controlling,” 619–32; Hoover Green, Commander’s Dilemma.

21 Cohen, “Explaining Rape during Civil War”; Cohen and Nordås, “Sexual Violence in Armed Conflict.”

22 Wood and Bleckner, “Conflict-Related Sexual Violence,” 25–38; Meger, Rape Loot Pillage.

23 Wood, “Policy Implications of Recent Research”, “Typology of Political Violence” (see note 18 above).

24 Weaver, Ideologies of Forgetting; Richardot, “You Know What to Do with Them,” 83–90; Wood, “Typology of Political Violence” (see note 18 above).

25 Cohen, “Explaining Rape During Civil War” and Rape During Civil War (see note 18 above).

26 Wood and Toppelberg, “Persistence of Sexual Assault,” 620–33.

27 See, for instance, UNSC Resolution 1888.

28 Johansson and Hultman, “UN Peacekeeping and Protection,” 1656–81.

29 Kirschner and Miller, “Does Peacekeeping Bring Peace,” 2043–70.

30 Ibid.

31 Ibid.

32 See note 18, above.

33 Touquet and Gorris, “Out of the Shadows?”; Goodley, “Ignoring Male Victims.”

34 Cohen, “Female Combatants,” 383–415.

35 Wood and Toppelberg, “The Persistence of Sexual Assault,” 620–33.

36 Seelinger and Wood, Practice of War.

37 Ibid.

38 Further, CRSV may influence conflict dynamics in ways that create challenges for peacekeeping operations. For example, Nagel and Doctor find that rebel groups that perpetrate CRSV are more likely to fragment, potentially increasing the number of armed actors in a conflict. See Nagel and Doctor, “Conflict-related Sexual Violence.”

39 Kirschner and Miller (see note 29, above) find that civilian personnel are associated with lower levels of CRSV, implying the possibility of social and institutional change with the presence of civilian personnel over time.

40 See note 28, above.

41 United Nations, S/2019/280.

42 Blair, “International Intervention,” 365–98.

43 Karim et al., “International Gender Balancing Reform,” 618–31.

44 “Sexual Violence in Times of War.”

45 Benson and Gizelis, “A Gendered Imperative.”

46 Hultman and Johansson, “Responding to Wartime Sexual Violence.”

47 Kreft, “The Gender Mainstreaming Gap.”

48 Hultman and Johansson, “Responding to Wartime Sexual Violence.”

49 Kirby and Shepherd, “The Futures Past of the Women, Peace and Security Agenda,” 378.

50 Kreft, “Gender Mainstreaming Gap.”

51 e.g. Doyle and Sambanis, “International Peacebuilding”; Fortna, “Does Peacekeeping Keep Peace?”; Hultman, Kathman, and Shannon, “United Nations Peacekeeping and Civilian Protection in Civil War”; Hultman, Kathman, and Shannon, “Beyond Keeping Peace”; Beardsley and Gleditsch, “Peacekeeping as Conflict Containment”; Ruggeri, Dorussen, and Gizelis, “Winning the Peace Locally”; Hegre, Hultman, and Nygård, “Evaluating the Conflict-Reducing Effect of UN Peacekeeping Operations.”

52 UN Women, “Addressing Conflict-Related Sexual Violence,” 9. Read also contribution by Ruffa in this Forum.

53 UN Women, “Women’s Participation in Peace Negotiations.”

54 See contribution by Lotze in this Forum.

55 Nordås and Rustad, “Sexual Exploitation and Abuse by Peacekeepers”; read also Ruffa’s and Hoover Green’s contributions to this Forum.

56 e.g. Skjelsbæk, “Sexual Violence in Times of War.”

57 Davies and True, “The Politics of Counting and Reporting Conflict-Related Sexual and Gender-Based Violence.”

58 Nordås and Nagel, “Continued Failure to End Wartime Sexual Violence”; Cohen and Nordås, “Sexual Violence in Armed Conflict.” Cohen’s data collection is essentially a subset of SVAC as it includes reports of wartime rape, compared to SVAC which includes other types of sexual violations too (Cohen, “Explaining Rape during Civil War”).

59 “UN Peacekeeping and Protection From Sexual Violence.”

60 Ibid.

61 Kirschner and Miller, “Does Peacekeeping Really Bring Peace?”

62 Johansson and Hultman, “UN Peacekeeping and Protection from Sexual Violence.”

63 Kirschner and Miller, “Does Peacekeeping Really Bring Peace?”

64 Johansson and Hultman, “UN Peacekeeping and Protection from Sexual Violence.”

65 Kirschner and Miller, “Does Peacekeeping Really Bring Peace?”

66 Eriksson Baaz et al., “Making Sense of Violence,” 76–80.

67 Johansson and Hultman, “UN Peacekeeping and Protection from Sexual Violence.”

68 Ibid.

69 Nagel and Doctor, “Conflict-Related Sexual Violence and Rebel Group Fragmentation.”

70 Muvumba Sellström, “Stronger than Justice” Read also Muvumba Sellström in this Forum.

71 Kreft, “Responding to Sexual Violence.”

72 Nagel, “Talking to the Shameless?” (note that Nagel’s finding refer to CRSV by rebel groups only).

73 Chu and Braithwaite, “The Effect of Sexual Violence on Negotiated Outcomes in Civil Conflicts,” 242.

74 Belgioioso, Di Salvatori, and Pinckney, “Tangled up in Blue.”

75 UN Women, “Addressing Conflict-Related Sexual Violence”; United Nations Peacekeeping, “Women in Peacekeeping.”

76 United Nations Peacekeeping, “Women in Peacekeeping.”

77 Ibid.

78 Bove and Ruggeri, “Kinds of Blue” Notably, this study measures PKO effectiveness in terms of lethal violence, not CRSV.

79 Unpublished work by Sarwari “UN Responsiveness to Wartime Sexual Violence.” suggests a decline in CRSV as the proportion of female peacekeeping personnel increases. Read also Karlsrud and Solhjell, “Gender-Sensitive Protection and the Responsibility to Prevent.”

80 Karim and Beardsley, “Female Peacekeepers and Gender Balancing.”

81 Karim, “Reevaluating Peacekeeping Effectiveness.”

82 see e.g. Touquet and Gorris, “Out of the Shadows?”

83 United Nations Organization Mission in the Democratic Republic of the Congo.

84 United Nations Organization Stabilization Mission in the Democratic Republic of the Congo.

85 United Nations, Reports of the Secretary-General, 2017, 2018, 2019.

86 United Nations. Draft Policy, 4.

87 United Nations, Report of the Secretary-General, 2019, 2.

88 Ibid., 1.

89 United Nations, Draft Policy, 9.

90 Zerrougui, “Strengthening the Rule of Law.”

91 Ibid.

92 Ibid.

93 United Nations, Report of the Secretary-General, 2012, 10.

94 Ibid., 8.

95 Zerrougui, “Strengthening the Rule of Law.”

96 United Nations, Reports of the Secretary-General, 2017, 2018, 2019.

97 Ibid., 11–12.

98 Ibid., 11.

99 United Nations, Draft Policy, 5–8.

100 Ibid., 15–16.

101 United Nations, Report of the Secretary-General on Conflict-Related Sexual Violence, 2018, 6.

102 Kirschner and Miller, “Does Peacekeeping bring Peace?”; Hultman, “UN Peacekeeping and Protection.”

103 Beber et al., “Peacekeeping, Compliance with International Norms,” 1–30; Ferstman, Criminalizing Sexual Exploitation and Abuse; Aoi et al., Unintended Consequences of Peacekeeping Operations; Simić and O’Brien, “Peacekeeper Babies,” 345–63.

104 Finnemore and Sikkink, “International Norm Dynamics and Political Change,” 887–917.

105 Karim and Beardsley, Equal Opportunity Peacekeeping, 31; Janowitz, The Professional Soldier.

106 Wechsler Segal, “The Military and the Family as Greedy Institutions,” 9–38; Goffman, Asylums.

107 Ruffa, Military Cultures.

108 Eichler, Militarizing Men; Dana M. Britton, “The Epistemology of the Gendered Organization,” 420.

109 Karim and Beardsley, Equal Opportunity Peacekeeping, 37.

110 Anthony King, “The Female Combat Soldier,” 122–43.

111 Turchik and Wilson, “Sexual Assault in the U.S. Military,” 267–77.

112 Ibid., 267–77.

113 Howard, “Peacekeeping is Not Counterinsurgency,” 1–4; Ruffa, Military Cultures.

114 Bell, “Military Culture and Restraint,” 488–518; ICRC, Roots of Restraint in War.

115 United Nations, Security Sector Reform.

116 Björkdahl, “Promoting Norms through Peacekeeping.”

117 Grissom, “The Future of Military Innovation Studies,” 905–34; Fox, Learning to Fight; Farrell, “Improving in War,” 567–94.

118 United Nations, Security Sector Reform.

119 Ibid.

120 Coomaraswamy et al., Preventing Conflict, Transforming Justice; Bacon, “Liberia’s Gender-Sensitive Police Reform,” 372–97.

121 Westendorf and Searle, “Sexual Exploitation and Abuse in Peace Operations,” 365–87.

122 Carey, “Women and Peace and Security,” 49–68.

123 ICRC, Roots of Restraint in War; Bell, “Military Culture and Restraint.”

124 Wilén, “Security Sector Reform and Peacekeeping Troop Contribution”; Böhmelt, Escribà-Folch, and Pilster, “Pitfalls of Professionalism?”; Wilén, Birantamije, and David Ambrosetti, “The Burundian Army’s Trajectory,” 120–35.

125 Or alternatively 1989–1996 and 1997–2003.

126 See discussion in Cohen and Hoover Green, “Dueling incentives.”

127 UN Security Council Resolution 1509, 3. Note that 15,000 peacekeeping personnel was approximately 0.48% of Liberia’s 2003 population. Given the concentration of the force in the capital city of Monrovia (population approximately 600,000 in 2003), the concentration of UN personnel in Monrovia was likely even higher.

128 Cohen and Nordås, “SVAC Dataset.”

129 It is important to note that transactional sex is not the same phenomenon as rape, and that sex workers frequently choose transactional sex as a livelihood, or a source of extra income, because it represents their perceived best option. But, given the power and resource differentials between UN personnel and (most) Liberian citizens, especially at the close of civil war, it also cannot be viewed in this context as freely chosen, or as fully consensual.

130 Beber et al., “Peacekeeping, Compliance with International Norms, and Transactional Sex,” 10.

131 See review in Cohen, Hoover Green, and Wood, “Wartime Sexual Violence.”

132 Strategic explanations also fail in the context of intra-military sexual violence; see Wood and Toppelberg, “The Persistence of Sexual Assault Within the US Military.” Note, however, that strategy could explain sexual violence by conflict actors in the post-war period.

133 Ibid.

134 See, e.g. Winslow, “Misplaced Loyalties”; Wood, “When is Wartime Rape Rare?” Hoover Green, The Commander’s Dilemma.

135 This definition accords with, for example, March and Olsen, “Elaborating the New Institutionalism.”

136 Hoover Green, The Commander’s Dilemma, especially ch. 2.

137 Meltzer, “Gender, Work, and Intimate Violence;” Anderson and Lo, “Intimate Partner Violence.”

138 The argument is premised on the assumption that there exists a strategic or other incentive to avoid maximal violence.

139 Weinstein, Inside Rebellion; Cohen, “Explaining Rape during Civil War.”

140 Warner et al., “Battlefield Ethics Training;” Hoover Green, “The Commander’s Dilemma;” Muvumba Sellström, this issue.

141 Humphreys and Weinstein, “Handling and Manhandling Civilians.”

142 Matthews, Hall, and Lim, “A Commander’s Guide.”

143 See discussion in Kirby, “Ending Sexual Violence.”

144 Moncrief, “Military Culture.”

145 Patton, War as I Knew It, 23; and see Brownmiller, Against our Will, 31.

146 Hoover Green, The Commander’s Dilemma, 26, 103.

147 See discussion in Baaz and Stern, “Why Do Soldiers Rape?” 498–99.

148 Cohen and Nordås, “SVAC Dataset.”

149 Johansson and Hultman, “UN Peacekeeping and Protection,” 1669.

150 Ibid., 1671–72.

151 Hultman and Johansson, “Responding to Wartime Sexual Violence,” 140.

152 Moncrief, “Military Socialization,” 718–25.

153 Horne, Robinson, and Lloyd, “The Relationship Between Contributors.”

154 United Nations, Security Council Resolution 1325.

155 Karim and Beardsley, “Female Peacekeepers;” Bridges and Horsfall, “Increasing Operational Effectiveness.” And see more generally Britton, “Epistemology of the Gendered Organization.”

156 Karim and Beardsley, “Explaining Sexual Exploitation.”

157 For a broader view of the necessity of comparative gender research, including in international institutions, see Chappell, “Comparative Gender and Institutions.”

158 Hultman and Muvumba Sellström, “WPS and Protection of Civilians.”

159 Gottschall, “Explaining Wartime Rape.” See Muvumba Sellström, “Burundi’s Rebel Groups” and “Stronger than Justice”; Hoover Green, “Repertoires of Violence” and The Commander’s Dilemma; ICRC, Roots of Restraint; Wood, “Armed Groups and Sexual Violence”, “Policy Implications of Recent Research”, and “Rape as a Practice of War,” 513–37.

160 Muvumba Sellström, “Burundi’s Rebel Groups” and “Stronger than Justice.”

161 Ibid. More specifically, it builds on interviews and focus groups with ex-combatants of CNDD-FDD and Palipehutu-FNL; expert and civil society input; and secondary literature such as ethnographies, historical studies and non-governmental reports.

162 Nindorera. The CNDD-FDD in Burundi.

163 ACAT and WOAT, Violence Against Women in Burundi.

164 Muvumba Sellström, “Burundi’s Rebel Groups” and “Stronger than Justice.”

165 HRW, Burundi: The Gatumba Massacre.

166 Ibid.

167 Van Acker, personal communication.

168 Muvumba Sellström, “Stronger than Justice.”

169 Wilén, Birantamije, and Ambrosetti, “The Burundian Army’s Trajectory.”

170 Ibid.

171 Ibid.

172 Ibid.

173 Muvumba Sellström, “Burundi’s Rebel Groups” and “Stronger than Justice.”

174 See for example HRW, Everyday Victims.

175 Palipehutu, Memorandum to the President.

176 See for example Baines, “Forced Marriage” and Donnelly, “Gender and Strategy.”.

177 Muvumba Sellström, “Stronger than Justice.”

178 Ibid.

179 Ibid.

180 HRW, Pursuit of Power, 48–49.

181 According to international customary IHL Rule 93. Rape and other forms of sexual violence are prohibited. Common Article 3 of the 1949 Geneva Conventions is here the main tool, but other Geneva Conventions, such as Third and Fourth Geneva Conventions, as well as Additional Protocols I and II also prohibits sexual violence and adds some more explicit language. “Customary IHL, Rule 93.”

182 Module 4, Individual Criminal Responsibility. “Tool for SGBV-Investigation/Documentation Trainers.”

183 Ibid.

184 For ANSAs some internal regulations prohibiting sexual violence can be found under the Geneva Call ‘Their Words’ database: https://theirwords.org.

185 Geneva Call.

186 As noted for example in the Rohingya crisis, where humanitarians had initally missed that men and boys had been massively exposed to sexual violence, which was widely known in the community, as there was not a common understanding of what sexual violence was. The New Humanitarian “Male rape survivors.”

187 See for example Herbert Kelman, quoted in Frésard, “The Roots of Behaviour of War,” 94.

188 Ligouri, “Brain Research.”

189 Ibid.

190 Wood, “Typology of Political Violence,” 1–25.

191 Ligouri, “Brain Research.”

192 Cohen and Nordås, “SVAC Dataset”; Bastick et al., Sexual Violence in Armed Conflict; Leatherman, Sexual Violence in Armed Conflict.

193 Kishi et al., Terribly and Terrifyingly Normal.

194 Autesserre, Peaceland.

195 Olsson, “Same Peace, Different Quality?”

196 Bjarnegård et al., “Honor and Political Violence.”

197 Kishi et al., Terribly and Terrifyingly Normal.

198 Data can be accessed at www.acleddata.com. For more on what the data do and do not measure, how data are categorized as such, etc., see ACLED FAQs: What Can and Cannot Be Done with ACLED’s New Gendered Associate Actors.

199 The first tranche of this data was launched in May 2019 in which historical data collected by ACLED was recoded to account for the targeting of women; this was collated in partnership with the Robert Strauss Center for International Security and Law at The University of Texas at Austin.

200 For more on ACLED sourcing, see ACLED FAQs: ACLED Sourcing Methodology.

201 For example, see Autesserre, Peaceland.

202 Raleigh et al., “Introducing ACLED”; Kishi et al., Terribly and Terrifyingly Normal.

203 The regions depicted here are limited to the regions that ACLED covers at the time of this writing; for more on ACLED spatial coverage, see the ACLED website.

204 This makes allegations of child rape in South Sudan by UN peacekeepers from Nepal, however, particularly damning (see Agence France Presse, “UN Peacekeepers Accused of Child Rape”).

205 Doctors Without Borders, Central African Republic.

206 La Fontaine, Witchcraft Belief.

207 Kishi, Witchcraft and Vigilante Justice.

208 In the Yemeni context, both pro-Houthi and pro-Hadi forces are coded as state forces. For more, see ACLED’s Yemen coding methodology document.

209 ACLED Codebook.

210 Ibid.

211 Kishi and Olsson, How Does Political Violence Target Women.

212 Kishi and Turkington, How to Thwart Political Violence Targeting Women to Stabilize Societies.