Reports of sexual violence during the ongoing unrest in Libya have captured headlines across the world. Libyan leader Muammar al-Qaddafi's forces, some have alleged, were given Viagra to facilitate their rape of hundreds, if not thousands, of victims. Recently, Luis Moreno-Ocampo, the chief prosecutor at the International Criminal Court, and U.S. Secretary of State Hillary Clinton both expressed outrage at what was apparently a purposeful campaign. Yet recent reports by the U.N. and by advocacy groups shed doubt on these claims. Amnesty International, for example, has been unable to locate a single rape victim, or even anyone who knows a victim.
As the veracity of stories about sexual violence in Libya came into question, the American Journal of Public Health published a study estimating that the prevalence of rape in the Democratic Republic of the Congo was far worse than previously documented. The article estimated that between 2006 and 2007 more than 400,000 women between the ages of 15 and 49 were raped during the war there -- 26 times the U.N.'s official count.
Read more at at Foreign Affairs' Special Report: Global Public Health.
So what are we to make of these two cases -- a possible exaggeration of rape in Libya and a gross underestimation of it in the DRC? Wartime sexual violence has rightly been called a hidden epidemic; in truth, we know very little about its actual magnitude and impact. Reports of rape are increasingly common in countries wracked by conflict, such as Colombia, the DRC, East Timor, Côte d'Ivoire, Libya, and Sudan, but no one knows what the relationship is between increased reports and increased rape. Even in peacetime, sexual violence is severely and unevenly underreported. Beyond prevalence, patterns of where, when, and by whom rape is committed -- not to mention why it is committed -- are even less clear. War exponentially worsens these problems. As a result, estimates of rape in prominent conflicts are often unreliable.
The lack of clarity about wartime sexual violence is hardly surprising. Wartime rape is violent and brutal. Few of us want to think about it, talk about it, or read about it. Even when it does make headlines, it is the sensational details, not meaningful analysis of patterns and causes, that are often reported. And researchers are as apt as anyone to look away. The study of wartime rape has been marginalized as a "feminist" or "gender" issue. For many years, research on sexual violence was considered too unseemly a topic for top academic publications. Social scientists and statisticians rarely collect systematic data on wartime rape; instead, they frequently rely on victim narratives recorded by human rights advocates, aid workers, and journalists; as well as law enforcement reports; and hospital records.
Although often profoundly moving and rich with detail, victim narratives cannot replace rigorous large-scale analyses. And all three sources can misrepresent wartime rape in a number of important ways, primarily by undercounting it. Because of stigma, shame, and the perceived impunity of the perpetrators, very few victims report their assaults. Likewise, only a small proportion of victims -- those with the most severe injuries or with the best access to facilities -- seek clinical care. During war, moreover, murder can follow rape; in such cases, the victim's story is buried with her or, less frequently, him. Uniform undercounting, moreover, is not the only problem. Victims who have been publicly gang raped may have less to lose by reporting the crime. But for the thousands raped in private, reporting it -- even anonymously -- may seem not worth the risk. In such cases, some types of victims may be better counted than others, distorting researchers' understanding of the crime.
Perhaps surprisingly, overcounting also is a concern. Mats Utas, a cultural anthropologist, has noted that "every single" female Sierra Leonean refugee he interviewed as part of his research in Liberia reported being raped. In Utas' view, the refugees sensed that reporting rape was an efficient way to procure material assistance from aid agencies. Some experts on sexual violence in the DRC have suggested that a similar dynamic is at work there. According to Maria Eriksson Baaz and Maria Stern, the growing focus of the media, aid organizations, and the international community on helping victims of sexual violence has come at the expense of other programs. They argue that a lack of interest in maternal health care, women's economic empowerment, and political participation incentivizes women to present themselves as rape victims in order to access basic health care services.
So the majority of rape estimates are really nothing more than highly educated guesswork. Yet some of the world's most trusted international news sources have cited as fact statistics that are implausible on their face. A 2002 BBC article, citing an advocacy organization, stated that "a girl born in South Africa will have a greater chance of being raped than learning how to read." It is true that South Africa has one of the highest rates of sexual violence in the world, but it is almost certainly not true that the rape rate surpasses 97.5 percent, which is the U.N. estimate for literacy among females aged 15 to 24 in South Africa.
Similarly, prominent advocacy organizations and media outlets have reported that 75 percent of Liberian women suffered wartime rape. The best available survey data indicates the number is closer to ten to twenty percent -- a staggering problem, but nothing close to what has become the conventional wisdom, which seems to have arisen from a misinterpretation of a small-scale study.
Presenting extrapolations as fact hurts efforts to prevent and address wartime sexual violence. Falsely specific figures may attract the attention of an otherwise apathetic public, but they may also endanger future victims of rape. Will anyone care if "only" five percent of women get raped in the next civil war?
Better prevention efforts require better analysis, which requires better evidence. Major players have already started to take that need seriously. In December 2010, the U.N. Security Council adopted Resolution 1960, which calls for a commitment to "enhance data collection and analysis of incidents, trends, and patterns of rape and other forms of sexual violence" to inform and improve policy initiatives. The U.N. has committed considerable resources to this goal, including creating an internal umbrella group, the U.N. Action Against Sexual Violence in Conflict, which aims to improve coordination between other U.N. agencies, amplify programming and advocacy, and support national efforts to prevent sexual violence.
But further efforts are needed. Researchers must also approach the problem of sexual violence more broadly, expanding their studies beyond the rape of women to include forced marriage by abduction, sexual mutilation and torture, sex trafficking, forced prostitution, forced abortion, forced sterilization, and sexual violence against men. Meanwhile, they must stop relying on "emergency data" -- numbers based on rapid assessments or careless extrapolations -- and instead invest in less immediate, but more accurate, studies on the prevalence and causes of sexual violence. Of course, no single analysis can provide the "final answer," but more comprehensive studies are a good place to start.