Later this week, ministers from more than 140 countries, along with an estimated 1,500 invited delegates, are gathering in London for the Global Summit to End Sexual Violence in Conflict. The summit — the largest gathering of its type — is co-chaired by British Foreign Secretary William Hague and the actress Angelina Jolie, in her capacity as the special envoy for the U.N. High Commissioner for Refugees.
As scholars who study conflict-related sexual violence, we are heartened by the increased attention to this terrible problem. For far too long, policymakers overlooked wartime rape and other forms of sexual violence, and the current wave of political will to create solutions to mitigate the horrors of these forms of violence is a most welcome change.
For policymakers to develop effective policies, they will require a clear understanding of the ways that rape and other forms of sexual violence have varied across time, space and perpetrator group. As social scientists, we are committed to tracking these forms of variation, which are essential to understanding where — and helping us test theories of why — rape has occurred in recent conflicts. Efforts to mitigate sexual violence in future conflicts may be strengthened by holding accountable perpetrators of past crimes. Accomplishing this goal necessitates careful documentation and analysis of patterns of sexual violence, particularly because direct proof of command responsibility is often elusive.
To that end, we have recently released the Sexual Violence in Armed Conflict dataset, which tracks reports of seven forms of sexual violence (rape, sexual slavery, forced prostitution, forced pregnancy, forced sterilization/abortion, sexual mutilation and sexual torture) against noncombatant victims (both women and men) by all major armed groups involved in wars over a recent 20-year period. (The SVAC data include reports of sexual violence perpetrated by armed groups only, not interpersonal violence or violence by groups not related to the conflict, such as abuses by local police). For more details, please see this article, or the FAQ section of the SVAC dataset Web site.)
Findings from these data, along with results from other work of ours, challenge some of the conventional wisdom on wartime sexual violence and may offer guidance to summit-goers as they consider solutions later this week.
Among the important, and policy-relevant, patterns we find:
1. Rape and other forms of sexual violence are reported in many conflicts and by many armed groups — but not all.
During the study period, 14 percent of the conflicts were characterized by reports of massive sexual violence, whereas 43 percent of conflicts had no reports of sexual violence. Turning from variation across conflicts to variation across groups within conflicts, we find that 42 percent of state armed groups were reported as perpetrators of sexual violence at some point during the study period, compared with only 24 percent of rebel groups and 17 percent of pro-government militias. Scholars have argued that this variation in and of itself is very hopeful news; the fact that some armed groups exercise restraint implies that sexual violence is not inevitable during wartime and that it can be effectively prevented. A key future area of research will be to better understand the features of those armed groups that perpetrate widespread rape versus those that are successful at preventing it — whether it is due to socialization, education, training or something else altogether.
2. State militaries are more likely to be reported as perpetrators of sexual violence than either rebel groups or pro-government militias.
Indeed, 42 percent of state militaries engaged in conflict were reported as perpetrators of sexual violence at some point during the study period, while only 24 percent of rebel groups and 17 percent of pro-government militias were. As we have argued elsewhere, this presents an opportunity for policymakers: Research suggests that states can be effectively shamed by international campaigns, and it is likely that states are more susceptible to such campaigns than are rebel groups.
3. Rape need not be directly ordered to occur on a massive scale.
The members of some armed groups engage in high levels of rape even though they were not ordered (even implicitly) to do so. To be sure, commanders ordered their combatants to rape civilians in some well-documented cases; common forms of strategic rape include cases of rape carried out during some (but not all) campaigns of ethnic cleansing and rape as sexual torture of prisoners. Widespread rape that is not ordered but is tolerated by commanders is best understood as a “practice.” Such commanders tolerate rape by their combatants because prohibition would mean disciplining otherwise-capable troops, as well as expending scarce resources on an issue that the commander does not see as serious, withholding what combatants see as “booty” of war — or because they themselves engage in rape. Groups that forcibly abduct new members often forge social cohesion through participation in gang rape. Of course, where commanders have effective control over troops, even if they did not order their combatants to rape, they are legally responsible if they knew or should have known of its occurrence.
4. Sexual violence by armed groups can continue to be reported into the “post-conflict” period, sometimes at very high levels.
We collected data on reports of sexual violations by armed groups for the first five years that those groups were considered “inactive” (using the standard definition of no longer killing at least 25 people per year). We find that 21 percent of state actors, 13 percent of rebel groups and 5 percent of pro-government militias were reported to perpetrate acts of sexual violence after the groups were considered no longer active. Of these, 3 percent of the reports of sexual violence perpetrated by states, 6 percent of that perpetrated by rebels and 3 percent of that perpetrated by pro-government militias were at high or severe levels. This suggests that “conflict-related” sexual violence can continue even after the lethal violence of the conflict has ended, confirming patterns noted by others. Policymakers would be wise to recognize that lethal violence and sexual violence may not be closely correlated, and should remain attuned to reports of sexual violence well into the post-war period.
As the systematic study of rape and other forms of sexual violence continues to progress, it is our hope that our research, together with that by other scholars, will aid policymakers and activists in combating this terrible scourge of war.
Dara Kay Cohen is an assistant professor of political science at the Harvard Kennedy School; Ragnhild Nordås is a senior researcher at the Peace Research Institute Oslo; Elisabeth Jean Wood is a professor of political science at Yale University.