Acts of War

Acts of War

On the sidelines of the civil strife that has long embroiled the people of South Sudan, women and girls are in the crosshairs.

May 27, 2015

The world’s newest nation is also one of its most volatile. South Sudan emerged in 2011, rising from nearly a half century of civil wars. But by the end of 2013, ethnic conflicts had erupted again in warring that has been marked by gender-based violence. “The bodies of women and children are the battleground of this conflict,” the United Nations’ special representative on sexual violence in conflict, Zainab Hawa Bangura, said last fall after visiting one of the war’s hot spots.

GW Global Women’s Institute Director Mary Ellsberg and research director Manuel Contreras traveled to the fledgling country last August. They were preparing for potentially groundbreaking research that aims to estimate the scope of gender-based violence in South Sudan and to better understand its causes and consequences, while also attempting to create a standard for measuring this kind of violence in conflict zones the world over. The study, proposed during South Sudan’s window of peace, is now taking on new dimensions of urgency, danger and complexity, as the researchers try to measure the violence of war in real time.

After the trip, Dr. Ellsberg spoke with Danny Freedman about the project and its relevance.

What do we know already about violence against women and girls in conflict zones?

We’ve known for a long time that rape exists in war. Now there’s a recognition that, in some cases, it’s on a massive scale and used as a tactic of war. We saw it in Bosnia and Herzegovina, we saw it in Rwanda, we’re seeing it in Democratic Republic of Congo and in South Sudan. In many places, rape is not just an opportunistic crime, it’s actually part of fighters’ tactics to demoralize villagers and to terrify them. And it’s done in horrendous ways that are meant to completely devastate families and whole communities. Domestic violence also increases in times of conflict, and when women are getting displaced, for example, this gets piled on to their already extreme vulnerabilities to child sexual abuse and to sexual assault in communities. 

So that’s what we do know; there’s a lot that we don’t know. 

We have pretty good methods now for doing this kind of research in non-conflict circumstances. We’ve also done some research in postconflict settings. But how to apply those methods to an area that’s in complete upheaval—I think there’s been some skepticism about whether it can be done. But the fact is: If you don’t have good data, it’s much harder to convince people to invest in solutions. So what we’re doing is figuring out how to adapt and apply some of these tools, this rigorous scientific method, to an area that’s so fluid.


What’s the goal of the study?

We’re trying to understand the prevalence, the triggers and the characteristics of gender-based violence in both the most recent conflict and earlier conflicts in South Sudan. That’s going to be going on for a couple years. Then we’re going to help design and evaluate an intervention for violence prevention.

The research is part of a broader five-year program led by the International Rescue Committee, along with us at GWI and CARE International, called “What Works to Prevent Violence Against Women and Girls in Conflict and Humanitarian Emergencies,” which is funded by the United Kingdom’s Department for International Development. 


There are no measures for this kind of violence in war zones?

Only a few studies have been done. And here’s another problem: Everybody’s using different questionnaires, their methods are different, so the results aren’t comparable. That’s the problem we had with studying domestic violence years ago. As a result, there was the World Health Organization’s Multi-country Study on Women’s Health and Domestic Violence against Women in 2005, which I was a part of, that was really the first effort to have a standard methodology that would allow us to compare the levels of violence across different settings. 

For the South Sudan study, to ensure that we’re building on best practices for this kind of research, we convened a meeting in February with researchers and other stakeholders from all over the world. They shared their experiences with interviewing methods, like face-to-face with iPads or on paper; ways to make it safe and comfortable for people to disclose violence; safety issues; how we measure and how we define violence, those kinds of things—so we can try to develop the gold-standard instrument. That’s our goal here.


Why South Sudan?

In July 2013, we proposed a set of different studies that we thought would contribute to the gaps in the evidence, and we wanted a place where we could conduct both qualitative and quantitative research on the prevalence and drivers of violence. We thought of South Sudan because it was newly independent, and we were hearing really positive stories. There was a feeling that the people there had been through so much, and here they are finally getting a chance to move forward as an independent country and live in peace. It was a real time of optimism. 

There also were no data on gender-based violence, the government was interested in learning more, and it’s a place where they could actually use the data for programs. 

That’s how we proposed it. In November we got word that we received the contract, and in December everything blew up and civil war was back on. In some ways the research is just as important, or even more important, but it’s definitely more challenging to do the research than we initially expected.

When we were there, the hardest thing was to choose where we would be able to do an ethnically diverse, representative study in six to eight months, because nobody knows where anybody’s going to be then, or what places it will be possible to enter. It’s just so fluid.


Hence calling this a feasibility visit. What did you find when you got there?

We were going to find out whether and where there were conditions on the ground to be able to do this research. With the conflict ongoing we had real doubts, we thought that it would be impossible to do. But talking to people on the ground, they wanted us to do the study. People felt it was important to document what was happening and to do it in real time, not wait till years later.

The first week we were in Juba, the capital, meeting with the rest of the local and international research team, government officials—we’re working closely with the Ministry of Gender, Child and Social Welfare and the National Bureau of Statistics—United Nations agencies, health care workers and women’s groups. We also went to possible study sites, including a Protection of Civilians camp in the surrounding county. 


What was that like?

Everything is tense. People are living in very small tents, basically, each made out of one big sheet of canvas. It’s hot, dusty; there’s not a lot of food. Women have to leave the camps to look for food and firewood, and that’s a big risk for getting raped. People are just hugely frustrated over the situation. It’s a tinderbox, and that breeds more domestic violence.

In the second week, Manuel Contreras, our institute’s director of research, traveled on to Rumbek in the central state of Lakes, to continue the assessment. Things are very tense there, too. 

It was a packed trip. While we were there, and in our ongoing efforts with the technical advisory team I mentioned earlier, we’re trying to get everybody’s buy-in that what we’re doing is useful and appropriate, both scientifically and culturally. At every step we’re trying to do a lot of consultation to make sure that we’re getting it right.

One of the lines we have to walk carefully on is: We’re not doing this as human rights activists. The IRC is a humanitarian actor, and they have to be very careful to not appear to be taking sides. And we are public health researchers; we’re trying to document the problem and, based on that, to raise the flag of awareness. 

We also have to collect the data in a way that is completely impeccable, in terms of its transparency, its rigor, its objectivity, or it won’t have credibility.


In the end, how are you going to be conducting the research?

Choosing the sites was important, because we had already decided that it would not be possible to carry out a nationally representative survey. We needed to find places that represent at least the three main ethnic groups: the Dinka, Nuer and Murle. 

We ultimately settled on doing a door-to-door field survey in Juba County and the state of Lakes. Juba County is ethnically diverse and includes urban and rural areas, including two Protection of Civilians camps, while Lakes is largely rural and most of the people are Dinka. Lakes hasn’t seen the kind of fighting that’s going on in other areas, but there are conflicts between communities related to cattle rustling and bride prices that also increase women’s risk of violence. 

Those two sites will have both the quantitative study and qualitative work—interviews with survivors, humanitarian and health care workers, and others, maybe even perpetrators of violence. We’d planned on a third site, Unity state in the north, but that’s one of the areas most affected by the war now. Going door-to-door would be too dangerous. So in order to try to still represent that state, which is mostly Nuer, we’ll conduct only the qualitative piece there.

For the population survey, which will likely begin this fall, we’ll be interviewing 2,000 women and 1,000 men. A few studies have reported sexual violence against men during conflicts, and though most of the people we talked with in South Sudan hadn’t heard of incidents like that, it may be that it goes unreported because of stigma. We’d like to gather data and see, one way or another.

So we’ll be asking women and men about their experiences with domestic violence, sexual abuse in childhood and other violence they’ve witnessed or experienced during and before the current war.


The survey is door-to-door?

You get maps of each village and neighborhood—and the last census was in 2008, before South Sudan was even a country, so we’ll be updating the maps as we go—and you randomly select, say, every sixth house systematically through the neighborhood. You find out who lives there. We’re looking for women and men age 15 or older. If they have more than one, then we randomly select one and interview him or her.

The trick, then, is getting them alone. Sometimes you take them outside, you go sit under a tree, you go to the river where they’re washing. That’s a key piece of this, so they can feel comfortable talking about violence. It’s such a stigmatized, sensitive issue that people often won’t talk about it. But that secrecy is also important so the person doesn’t face reprisals afterward. 

We have a whole set of safety measures, developed by the World Health Organization, and that involves having dummy questionnaires, or if somebody walks into the room you change the subject and start talking about breast-feeding or immunizations, and these are all things that you discuss with the woman beforehand. 

We don’t tell the family that we’re talking about violence. We tell them it’s something about women’s health and then when we get her alone we tell her what we’re going to be talking about. We’ll also have basic psychological and social services available, directly or by referral, both for the women in the study and for the interviewers. It’s hugely stressful for interviewers to be listening to these stories, and a lot of times they’ve suffered abuse, too, so they need counseling as well during the process.

For everyone’s safety and for the comfort of the people in the study, this piece will be done by trained local interviewers, matched with respondents by gender and ethnicity.


Is there anything that works to prevent this kind of violence?

We just published a review paper in a special issue of [British medical journal] The Lancet that looks at the effectiveness of different programs to prevent all forms of violence, including female genital mutilation, child marriage, domestic violence and sexual assault. And what we found was pretty interesting. 

We did a review of all the rigorous studies we could find that aimed to reduce violence against women and girls. Most of the studies in the U.S. and Canada involved screening women for domestic violence or treatment for batterers through the justice system. We found that there is almost no evidence that these programs actually prevent future violence. They may have other benefits but, for the most part, relying on screening programs and batterers’ treatment does not work to reduce violence overall.

We also now know a lot more about the drivers of violence. A big reason why there’s more violence in one country than another, or even one community to another, has to do with social norms. When you’re in a setting where 50 percent of the men are beating their wives, it’s not because 50 percent of the male population are monsters. It’s because it’s socially acceptable, and if you don’t beat your wife you’re not considered to be a man. So deterrents through stronger laws—“you’ll go to jail if you do this”—or better services are not going to change that. You have to work on making it not OK to beat your wife and talking about gender equality. 

That sounds kind of touchy-feely. But now there have been a few rigorous, randomized, controlled trials that have shown that it is possible to prevent violence against women by changing social norms. One community mobilization program carried out in Uganda, called SASA!, reduced domestic violence by 50 percent in the study communities in two years. They just saturated the communities with messages about how harmful domestic violence was, not only to women but to families and communities. While many programs have focused on women, this one also included everyone from men and children to religious leaders, police, teachers and health workers. It has been hugely successful and is being adapted and scaled up in many countries. These are the types of interventions we think have the potential to make a lasting impact on violence worldwide.

The evidence showing that violence can be prevented sends a very powerful message. We’re hoping to show that programs like this can be successful in postconflict settings, too. 

In conflict settings, like South Sudan, our emphasis is on documenting the magnitude and types of violence women and girls are facing, and the kinds of basic measures that can make a difference in the acute phase, such as ensuring safe access to firewood and water. Often, women in refugee camps are raped when they have to leave the camps to look for firewood or food or to carry water. 

Humanitarian groups are also paying a lot more attention to issues like sanitation. If the toilets are too far away, or if the women’s toilets are too close to the men’s toilets, or if people have to go at night and there’s no lighting, that’s when women get raped.

So even pretty simple measures around infrastructure and lighting—like access to charcoal,
so women don’t have to get firewood—can actually make a big difference in terms of women’s safety in the camps.