Do Unto Others focuses on the Middle East, (nonviolent) social movements, and how I make sense of my place in the world. I'm currently based in Cairo, Egypt doing peacebuilding and community development.

Monday, July 30, 2012

Where rape, murder, and gross income inequality are classified as "peace"

Over the last week, a common theme has randomly emerged out of lectures I have attended and blog posts I’ve read: scholars and practitioners of conflict (resolution), international affairs, peacebuilding, and transitional justice are fixated on political violence, often at the expense of other, more discreet, but equally as menacing forms of violence.

At a seminar hosted by the Institute for Security Studies in Cape Town, Australian academic, John Braithwaite spoke about “cascades of violence” that tend to move and morph in location and in type.  Violence in the eastern provinces of the Democratic Republic of the Congo seeps across the border to Rwanda.  Policymakers often fail to see the impacts of their political decisions and military movements. One acute example of a violence cascade is the Iraq sectarian war that was fomented (or mobilized) by the US invasion and occupation of that country. Not only did the sectarian divide contribute significantly to the more than 100,000 Iraqis that are now dead as a result of the war, but sectarianism is increasingly problematic throughout the region.  Iran and Saudi Arabia’s feud is starting to resemble US-USSR relations during the Cold War period, albeit on a much smaller scale. Iran funds Shia, and non-Sunni, movements and regimes while Saudi Arabia generally supports Sunnis. 

Not only does political violence cascade from one place to another, but violence can shift and cascade from one type to another.  In South Africa, political violence has decreased in the last 18 years, since the political transition to democracy in 1994.  However, other types of violence in South Africa have remained static or have even increased. Violent crime in South Africa remains quite high.  According to 2010 data, the southern Africa region has the highest rates of intentional homicide in the world, followed closely by Central America.  South Africa, as a country, ranks in the top 10 for intentional homicides per 100,000 people.  As of the early 2000s, South Africa had the highest rape rate per capita in the world.


In addition to high rates of violent crime, South Africa remains one of the most economically unequal societies on the planet.  South Africa has an incredibly high GINI coefficient, a World Bank statistic that measures the level of economic inequality in a given country (however, comparing GINI coefficients between countries is difficult because of the different ways that data is gathered and because the type of numbers that go into the production of a GINI coefficient aren’t uniform from country to country). South Africa also ranks among the top 10 most unequal societies in the world when the average income of the top 10% is compared with the average income of the bottom 10%, the same holds when the top 20% is compared with the bottom 20%.


Many community organizations working in townships and formal/informal settlements in South Africa will also attest to the disparity of quality service delivery in townships as compared with wealthier (and usually whiter) areas.  Crime, income inequality, and high rates of HIV (among many other health and education indicators) attest to acute forms of structural violence in South Africa.  Many donors and scholars have turned their attention away from South Africa, yet continue to refer to it as a prototypical success story, despite the ongoing economic and social maladies that plague the country. 

In his new, highly-touted book The Better Angels of Our Nature: Why Violence has Declined, Steven Pinker suggests that violence has decreased over time.  However, Christian Davenport recently pointed out that Pinker doesn’t look at the full scope of violence.
OK, so maybe the magnitude of violence has been diminished (which Pinker says) but the form may have shifted (which he does not).     
Davenport goes on to suggest that governments aren’t more angelic towards one another, nor to their own citizenry; instead, the forms of violence have simply changed.  States have also become quite effective at silencing voices of dissent and weakening movements that attempt to change the status quo and dethrone the throne-sitters. 

Will H. Moore also gets in on the conversation and suggests that the fixation of violent political conflict fails to account for structural violence and the connection between justice and peace.  Peace is too often (mis)understood as the absence of political violence.  Within this (misunderstood) understanding, peace can be at hand when wealth is disproportionately in the hands of whites while black South Africans don’t have access to electricity or proper toilets and peace can be manifest while women have a better chance of being raped than getting an education.     

Therein lies South Africa: high rates of violent crime, income inequality, and a plethora of social structures and social institutions that prevent the vast majority of South Africans (and a disproportionate number of non-whites) from meeting their basic needs; yet, there is a relative absence of overt political violence.

Has political violence in South Africa cascaded into more severe forms of structural violence? Have social and economic agreements made during the political transition only entrenched the (unequal) economic and social status quo while the political landscape changed without benefit for South Africa’s poor (black) majority?          

No comments: