What makes people commit violence against women? World Bank report tries to find out

Last week, an important resource on violence against women was placed in the public domain without fanfare. "Violence against Women and Girls: Lessons from South Asia" is the kind of study NGOs and academics in the region wish they could undertake. But it takes the resources of the World Bank to commission and publish a work of this scope (almost encyclopaedic) and quality.

In its almost-300 pages, report authors Jennifer Solotaroff and Rohini Pande tackle at length and in depth three important aspects of violence against women: the prevalence of violence, causes of violence, and the efficacy of interventions against violence against women. For each of these, it offers a dense review of the literature from across South Asia; so much so, it becomes an excellent point of departure for anyone just starting to teach themselves about the subject.

What predisposes people to commit violence against women? Predictably, coming from a development organisation, the report turns in its discussion of factors to how education, employment and gender violence are related. I highlight this because the middle class in India likes to believe education and economic independence secure a woman from the experience of violence, and much awareness work involves disabusing people of this easy correlation. 

This report finds (via other studies) that when only one or two women are educated and educated just a little, they may actually be at greater risk of violence. However, as the numbers of educated women increase and their levels of education rise, then the risk might diminish (page 96). The more women study, the more they are likely to disclose their experience of violence. 

Schooling, via scholarships, is also a chosen method in the campaign against child marriage; when girls are in school, they are less likely to be married off. This report finds no conclusive evidence about the correlation between working outside the home and violence. On the one hand, challenging men's roles as primary bread-winners could provoke violence, and on the other hand, bringing in supplemental income could cancel out the risk of violence (page 97).

Evidence also seems mixed on the impact of a husband's education on his propensity to resort to violence. Some studies suggest that more educated men have a less rigid idea of masculinity, are less inclined to seek to control their wives, and are less prone to violence. Others suggest more educated men command larger dowries which come with greater risks of violence (page 98). Stable employment is thought to diminish the likelihood of violence (page 99). But what is clear in all this confusion is that neither cause nor solution can be that simple; the work of eliminating violence is far more complex than an easy formula like "educate the women/men".

The authors repeatedly remark upon the lack of data on violence against women, something NGOs and academics have repeatedly pointed to. There are at least four reasons why we do not have those numbers. First, violence against women continues to be under-reported. 

Second, official counting (in India) is tagged to filed police reports, and those are tagged to existing laws. Where a complaint could be classified under more than one law, such as those relating to cruelty and domestic violence, we cannot be sure what the difference in numbers means. Where there is no law, such as 'honour killing', there is no tally. 

Third, service providers who could track cases of sexual assault, intimate partner violence or domestic violence do not have the wherewithal to really contribute to a collective documentation effort. Resources-money, ergo staffing-do not permit reliable record-keeping (or evaluation, one of this report's big complaints). 

Finally, quality data collection and analysis have to be well-paid, full-time careers. The result is it falls to organisations like the World Bank to come up with reports this rich, and even then, inevitably, what they can see and who they can reach is limited. This report illustrates what is possible when serious resources are allocated for research on serious issues.

Two very useful contributions of this report are to be found in the boxed text that describes local manifestations of gender-based violence or successful interventions, and in the appendices that list evaluated interventions categorised by the kind of violence they address. For all of us that work in the field, this is the kind of intervention we have to struggle to get even in an age where we think "everyone" has a website or that "everyone" is on social media. Accessible, summary information is still hard to come by.

The report is not just concerned about the availability of data but also monitoring. It reviews 101 interventions on different aspects of VAW, and the filter appears to have been whether monitoring and evaluation were undertaken. In fact, one of the strong recommendations of this report is that donors should emphasise, NGOs should undertake and researchers should help with monitoring and evaluation. I read this with discomfort, for two reasons.

Granted, every piece of research needs some practical filters, but having said that, this necessarily filters out small organisations. The report's description of various interventions across the region is interesting and informative. For small organisations, which are the majority in civil society, surely there is more to learn from the resource-strapped efforts of other small organisations. I would love to see a well-funded, systematic platform for such sharing, which also looks at low-cost, easy-to-do monitoring and evaluation methods. Perhaps that too is something the World Bank can undertake.

The other thing that makes me uncomfortable when I read prescriptions for NGOs to undertake this or that sort of evaluation is that these things cost money. To get a competent person or team to do a proper evaluation, you need the kind of grant money that is impossible for small organisations to access. Large organisations check off the boxes more easily, enabling them to get the funds to check off more boxes, and to have a voice. Small NGOs lie completely outside this loop, and sometimes with the best of intentions, cannot meet the criteria that are set in capital city conference rooms.

But should the World Bank even be writing about violence against women? The Executive Summary lists reasons why. Violence limits the human and citizenship potential of women. It affects their health across a lifetime, and also spirals across generations. It undercuts progress along the Millennium Development Goals. 

I would add to these: someone needs to invest serious time, money and human resources to creating and compiling knowledge in this area, and this report suggests the World Bank could be well-placed to do that-in terms of both resources and sensibility. I would hope this is the first of many definitive resources, and that others down the line will also adopt a ground-level perspective, learning from local service providers and human rights defenders at the village and district level.

Swarna Rajagopalan is a political scientist and the founder of Prajnya, a non-profit centre working on peace, justice and security issues, including women's rights and violence against women.