show me the data
A couple of months ago, a certain grad student/atrocity humor blogger who shall remain nameless emailed with the following question: "Could you point me towards anyone who's done research on the linkage / lack thereof between the mineral trade and sexual violence?" It seems that in her graduate school endeavors, solid research requires actual evidence to support the "cell phones/minerals cause rape" thesis that's become quite popular due to efforts of various activist groups, most notably the Enough Project.
It just so happened that this particular email arrived just a few days after I gave a talk on the subject of minerals and violence in the Congo, so I had already been searching for such evidence.
Long story short: there isn't any. As far as I can tell, there has as yet been no published report that systematically demonstrates a rigorous causal relationship between the mineral trade and the epidemic of sexual violence in the eastern Congo.
But, wait, you might say. There are lots of reports claiming that the mineral trade causes sexual and other forms of violence.
This is true. I certainly do not want to argue that there's no connection whatsoever. Much of the gender-based violence in the Congo is perpetrated by armed groups that are involved in the mineral trade. We know that just about every armed group in the eastern Congo has engaged in violence, looting, and rape at one time or another, and that many (but not all) of those armed groups also benefit from the trade in minerals. There's a definite correlation between some of the violence and the fact that armed groups profit from the mineral trade. And we know beyond any doubt that armed groups terrorize populations who live near their respective mines.
But the question we need to be asking is whether the majority of gender-based and other forms of violence in the eastern DRC are actually caused by the mineral trade. As long-time readers of this blog know, I am not yet convinced that the mineral trade causes the bulk of violence in the eastern Congo, or that getting the mineral supply chains under control would end the war against Congolese women and girls.
Why don't I believe this? Because there's no data showing that the mineral trade is the primary cause of violence in the eastern Congo. It's just not there. There are anecdotal accounts and reports on the mineral supply chains and reports on the horrific conditions in the mines. There are somewhat bizarre journalistic accounts like last week's 60 Minutes piece, which misled viewers into thinking that a gold pit in the largely peaceful Ituri district is at the epicenter of the current fighting. I sometimes hear that there might be data out there, or that an activist group is about to come out with a report showing a clear causal relationship, but as of yet, as far as I've been able to tell, there's nothing that shows anything that:
- that violence happens more frequently or with greater intensity near the mines than it does in non-mining areas of the region,
- that violence happens more frequently or with greater intensity along the primary mineral supply routes,
- that armed groups engaged in the mineral trade proportionally commit more acts of violence than those not engaged in the mineral trade,
- or that those groups that control more mines or make more money from mining engage in proportionally higher levels of violence than those who control fewer mines and make less money.
The sad fact is that violence happens everywhere in the Kivu provinces. Armed groups that benefit from the mineral trade buy weapons and rape women. But so do armed groups that don't benefit from the mineral trade. And a significant number of rapes are committed by civilians who see that they will have total impunity for their crimes and take advantage of the situation.
Earlier this week, @alanna_shaikh linked to this map. The creator meshed data from Ushahidi's crisis reporting system onto a map of "coltan mining areas" to make the point that there is a "conspicuous correlation" between the two variables.
While I certainly appreciate the creator's intentions to draw attention to the Congo conflict, this map is horribly misleading. First of all, there aren't coltan mines throughout North Kivu. A more useful mashup would overlap the violence incidents with the actual locations of specific mines. This matters because when you look at the Ushahidi map itself, it's clear that, while violence is high in mining areas, it's also high elsewhere. (There aren't any coltan mines, for example, between Goma and Sake. What's there? IDP camps and a road.) The eastern Congo is an incredibly violent place. Then of course, there's the problem that Ushahidi's data, while a wonderful resource, is far from complete due to the difficulties of reporting and collecting this sort of data. Sexual violence cases are still underreported in the DRC due to victims' concerns about shame and being shunned by their families and communities. We also know that there are mines - even in the Kivus - that are not militarized, despite the fact that there is gender-based violence in those areas as well.
More importantly, however, this map doesn't show us what else is in these regions. For one thing, there are large tracts of extremely fertile, hotly contested land whose status has been disputed by members of competing ethnic groups for decades. The Masisi region's volcanic soil is capable of producing up to three harvests per year, and North Kivu's varied climate is ideal for cattle ranching and dairy farming. There is money to be made in rural North Kivu. People were killing each other over the rights to this land before the wars and before mining was even a concern. Mobutu used North Kivu's land as an instrument of patronage; everyone from the Catholic Church to prominent politicians of whichever ethnic group was in his good graces at the time got a piece of land in exchange for loyalty to the regime. Under the RDC-Goma rebel government, land in Masisi and elsewhere was given to prominent Tutsis as a reward for supporting the regime and a means of securing Tutsi control of the region. When the Tutsis lost power, many also lost their land. Conflicts over land having nothing to do with minerals erupt all the time.
In other words, it's much more complicated than just the mineral trade. Which is why the argument that shutting down the mines will end all of this violence is fundamentally flawed. It is, quite frankly, based on incorrect assumptions and a lack of rigorously-analyzed evidence.
An all-encompassing focus on the mineral trade won't end violence in the eastern DRC. Assuming that it were even possible to track the Congo's minerals from source to market and that it would be possible shut down the militarized mineral trade (and, given the limits of technology and oversight, those are two mighty big assumptions), would the loss of income really force these armed groups to the negotiating table? These forces are already well-accustomed to terrorizing local populations to obtain the necessities of life. Would their behavior really change if they lost this income stream? I'm not sure. And, we must remember, there's the tiny problem of external financing of these armed groups (especially the FDLR) that the international community has until very recently completely ignored.
Then there's the lingering detail of the 1 million+ people who depend on the mineral trade for their livelihoods. Any program to shut down the mines have to take their employment into account. As Harrison Mitchell and Nicholas Garrett continue to point out, legitimizing the mineral trade is a far better idea than shutting it down.
I do not know a single scholar of the Congo who buys into the "cell phones cause rape" thesis. We all understand that the situation there is far too complex to be reduced by the activists to a simple resource war that could be solved if we just pressure Congress to stop the conflict mineral trade (How many of you are willing to give up your mobile phones to stand in solidarity with Congolese women? Keep in mind that there aren't any conflict-free cell phones.).
This doesn't mean that minerals don't matter. But the militarized mineral trade is a symptom of the disease of state failure, not the root cause of violence. Even setting aside all of the logistical issues with certification, controlling supply chains, taking physical control of the mines, developing the technology necessary to track minerals, finding livelihoods for newly unemployed miners, and creating a degree of consumer consciousness that's stronger than the desire for an iPhone, the violence won't end. It won't. There's no entity capable of stopping it.
Treating one symptom rarely cures a disease. We all want the people of the Congo to live productive, peaceful lives that are free from the constant threat of violence. We all agree that the eastern DRC is in many ways the linchpin for regional stability. But until there is serious security-sector reform, the Congolese government can actually control its territory, tax, and pay its soldiers, and the regional dynamics that drive much of the conflict over land, citizenship rights, and Rwanda's role in the region are settled, armed groups and civilians will continue to commit horrific acts of violence, simply because they can. "Doing something" about the mineral trade won't change that fact.
Policymakers would do well to focus less on oversimplified solutions to extraordinarily complex problems, and to instead turn their attention to giving the people of the DRC what they deserve and need: peace, public order, and a chance to make life better. That will require a long-term, sustained effort that doesn't pretend the peacekeepers only need to stay another six months or a year. It will require negotiating with unsavory non-state actors. It will require honest assessments of regional actors' territorial and sphere-of-influence ambitions. It will require the recognition of corruption in all its many varied forms, and of the need to directly target aid to its beneficiaries.
Above all else, it will require policies that are based on facts, not assumptions. The stakes are too high not to pursue policies that are data-driven and have a reasonable chance of success.
Then again, maybe it's easier to oversimplify things.