Domestic violence in Africa: What’s gender got to do with it?

by Hilde Jakobsen

Jakobsen_image3“Gender-based violence”: A familiar term in international health and development circles. This is how researchers, policymakers and practitioners refer to violence against women in the Global South. Who can even remember when they first heard it?

I can. July 2001. Cambridge. I’d just completed my criminology degree, and was applying for jobs that could get me back to Tanzania, where I’d completed high school. Anyone who’s struggled with that first real job search can imagine how ecstatic I was when an opportunity finally appeared in my inbox. I called my sister in the US to share my good news: I was going to Western Tanzania to evaluate the “Gender-Based Violence” program there. “What does Gender-Based Violence mean?” she asked. “No clue,“ I admitted : “a polite way of saying rape?” It was a euphemism, we decided, invented by one of these well-meaning Europeans and Americans, with their PC notions like gender, who knew nothing about Africa.

The program I evaluated turned out to be the first of what was to be hundreds of GBV programmes in refugee camps throughout Africa and emergencies across the world. Soon ‘everyone’ from Guinea to Chad to Geneva was dealing with “GBV”. I’d had no reason to worry about being sent on some crusade for Western-made ‘gender’ notions in Africa. We treated the violence as a health problem caused by armed conflict, poverty, neo-liberalism, disasters or displacement. It wasn’t ordinary social practices that the violence reflected, but the disruption of them. This approach was in perfect harmony with my pre-feminist stance in that phone conversation with my sister. I was happy to work with ‘gender-based violence’, but didn’t see what was ‘gendered’ about the violence, and I knew that ‘gender’ was a Western invention ill-suited to Africa. It’s a stance you could summarize to a Tina Turner tune : “What’s gender got to do (got to do) with it? What’s gender but a second-hand notion?”


This way of thinking, I have come to see, has led to some problems. If violence is gendered only because it happens to women, then any violence that also happens to men cannot be gendered. In North America and Europe, this conceptual framework has enabled a powerful academic lobby to argue that gender has nothing to do with partner violence, since women beat men too. In many African contexts, the “women beat men” claim serves as an objection to interventions against wife-beating. These are some real-world consequences of just calling violence against ‘gender-based’ without a clear idea of what this means. What exactly is gendered about gender based violence?

I try to answer this question in my Gender and Society article, by showing how people discuss wife-beating in two Tanzanian districts. I asked groups of men and women to discuss what was right and what was wrong in hypothetical cases of wife-beating. I left the room during the discussions so people would speak to one another and not to me. Which arguments people choose to make in a discussion reflects their assessment of how the others are thinking. This was what I wanted data on.

I examined the transcripts of 27 discussions to see how the speakers gauged the group opinion as they spoke. I found a shared idea of a “good beating” that a husband could expect others in the community to approve of. Drunken beatings, beating in a fit of temper, or beatings that resulted in serious injury were bad. A beating was good and necessary, however, if it made a wife perform her duties and behave as a woman should, or made it clear who was the man and therefore the boss.

I then compared what I found in my data with theoretical literature describing gender and how it plays out in social relations. Wherever I found a description that matched a finding in the data, I checked to see how representative that finding was for all my data. Was it something only women said? Or only Muslims, or only people from one district or one ethnic group? I only included findings that cut across these differences.


In doing this, I found that the “good” beating that participants expected society to approve was all about gender as described in the ‘doing gender’ tradition. It ensured the “accountability to sex category membership” that West and Zimmerman say is key to doing gender. It was considered “good” because it made sure that relations between spouses were organized according to biological sex. In this sense, it was gender as a cultural frame that was enforced, and its intended effect was concrete material inequality. The violence was both supported by gender norms, and enforced those same gender norms. In this way, it was cyclically intertwined with gender.

My point is this: When I actually looked at how ordinary people understand socially approved wife-beating, I could see that it was all about gender. From this I argue that gender is far from alien to Tanzania. Yes, maybe the most systematic theories around gender have been developed in other countries, but in my research, men and women, Muslims and Christians, rich and poor, of different ethnic groups and in two very disparate districts, all discuss wife-beating in terms of practices organizing spousal relations of inequality based on a two-sex difference. In other words, although they do not use the precise word ‘gender’, the concept that ‘gender’ refers to permeates their discussions of wife-beating.

Hilde Jakobsen, University of Bergen, Norway. Her article, “What’s Gendered About Gender-Based Violence? An Empirically Grounded Theoretical Exploration from Tanzania,” is published in the August 2014 issue of Gender & Society. To read the article, click here.  


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s