Exploring New Media’s Role in Conflict: The Case of Kyrgyzstan

Kyrgyzstan has witnessed two recent phases of social and political upheaval. I won’t get into the details of these episodes, but here’s the short version: the first phase occurred in April of this year, when President Bakiyev was overthrown in a coup and an interim government put in place. This was followed by ethnic clashes between ethnic Kyrgyz and the Uzbek communities in June, 2010. My attempt here is to understand the role that “new media” played in these episodes and their aftermath.

But first, some broader context:

It’s obvious even to the most casual observer that digital communication tools have made fairly fundamental changes in the way that individuals communicate with each other and consume information. With the emergence of mobile phones, social networking sites – what are being called “new media” – there’s also a great deal of interest in figuring out the role that new media are having in violent conflict and political change. From Iran to Kenya, there are good reasons to believe that they have played some role in large socio-political change. But figuring out the precise causal mechanisms of this role is a difficult task.

We are still in early days of this task: the technologies we’re researching, the data we have available, and even the number of cases we have on hand are all only beginning to emerge. But we’re already seeing two groups emerging amongst policy-makers and analysts. On one hand, we have those who believe that new media will play a key role in democratizing repressive regimes and resolving conflicts – the “cyberutopians.” [Read Larry Diamond’s recent article on ‘liberation technology’ (pdf)] The other camp — the “cyberskeptics” –sees regimes gaining an edge over activists for political reform, with the Internet acting as yet another means for surveillance, state repression, and violence. [The poster-child for this perspective is Evgeny Morazov].

As with many things, the truth of the matter is probably somewhere in the middle of the two. But without adding nuance, that statement is a mere platitude. Recent ‘Blogs & Bullets’ research released by the U.S. Institute of Peace’s Center of Innovation for Science, Technology & Peacebuilding, George Washington University, Harvard University’s Berkman Center for Internet & Society, and Morningside Analytics attempts to get us a little closer to that kind of nuance. They break down this argument into five levels where new media are having an effect in violent or contentious politics:

  • Individual transformation: how does online discourse affect the behaviour of individuals in violent conflict?
  • Inter-group relations: Do new media help create bonds across communities, or do they worsen inter-communal tensions?
  • Collective action: How do new media help individuals and groups overcome collective action costs to organize and protest?
  • Regime policies: How do repressive/violent regimes react to new media? Do they repress it? Co-opt it?
  • External attention: How do new media draw the attention of international media and political actors to contentious political issues and events in conflict zones?

Without getting too bogged down in explaining each of these factors, I’m going to explore the case of Kyrgyzstan along two of these vectors where new media appear to have had played some important role: inter-group relations and collective action.

Inter-group relations:
It appears that online discussion forums played a large part of the social media sphere in Kyrgyzstan in both April and June. We interviewed Elena Skochilo, a local citizen journalist, about the role of these online forums in mitigating or exacerbating violence between the two communities. It appears that the discussion forums and blogs were used to “spread racial hatred and many topics at the Diesel forum were blocked because of racial discussions. Moreover, Diesel forum were used to spread unjustified information to increase panic mood among the society.”

There were some sporadic instances of more positive discussions online:

“There was a special topic about positive events, like when one Kyrgyz family was hiding the Uzbek family during June events. The users of forum posted many positive stories but it was just a drop in an ocean of total racial discrimination. There were many social clips and projects also but I cannot say that they helped to establish a collaboration between Kyrgyz and Uzbeks.”

Of course, this begs the question of whether this kind of vitriol actually led people to perform acts of violence either in April or June. Also consider the counterfactual: would violence have erupted without this online discourse? There’s good reason to think so, but I have found little concrete evidence either way. But regardless, this is hardly an encouraging trend since it succeeds in inflaming an already tense situation, feeding into a frame of longstanding inter-ethnic grievances.

Collective Action
There are more heartening instances, with stronger causal linkages, of the use of new media in coordinating against violence in Krygyzstan. Local police, either due to a lack of capacity or willingness, were unable to prevent the widespread looting in the northern capital of Bishkek in the wake of the April riots. The Diesel forum was used to coordinate and recruit civilian militia groups that protected stores, homes, and people. Further, as Skochilo told us, “There were no such announcements through radio or TV. All recruiting was made through this Internet forum.”

However, there is no evidence to suggest that similar groups emerged during the June riots in the southern city of Osh. Skochilo points: “The reason is that southern part of Kyrgyzstan doesn’t have so much access to the Internet connection as northern part has. It was difficult for people to be organized in that chaos. There wasn’t even electricity to recharge cell phones, how can we talk about recruiting volunteers to be in the civilian militia in this situation?”

Information Vacuum
But it appears that new media in Kyrgyzstan fulfilled a role that conceptually ‘precedes’ the five levels laid out in the Blogs & Bullets research. As we have seen above, in some cases, they have behaved as active causal variables in the turbulent politics of Kyrgyzstan. But they played a larger, albeit more passive, role in those politics by filling in an informational space that was largely a vacuum, especially during events in June. As conventional news media succumbed to increased state control, forums like Diesel became the source of information for those wanting to hear about events on the ground, especially concerned diaspora. But new media filled this vacuum chaotically — rumours and misinformation were rampant, and the hashtag #freekg was overloaded with useless information. Thus, while new media have created new ways to fill information voids, they clearly suffer from a lack of verification that traditional editorial structures ensured.

*In the interest of full disclosure: Unlike my fellow bloggers, I still work with the US Institute of Peace. As with all my other posts, this reflects my opinion, not the Institute’s.


Fail Faire DC – Overall, a Win

I spent most of my evening listening to a group of honest folks describing their forays into the often treacherous world of information and communication technology (ICT) for development. Specifically, they spent their time being honest about a topic that so many in the development space have a tough time even raising: failure.

Fail Faire DC, put together by MobileActive.org and the World Bank, and hosted by the ever-buoyant Katrin Verclas, was created with the express purpose of learning from the numerous cases where ICT solutions had failed — tales of hubris, of unintended consequences, and of good intentions gone awry were all laid bare in an ostensibly safe space, where practitioners and policy-makers in a new field could learn from each others’ mistakes without finger-pointing and ridicule.

Presentations ranged from World Bank education projects to the National Democratic Institute’s election monitoring efforts in Gaza and Montenegro. Tim Kelly‘s presentation about infoDev’s Global Capacity Building Initiative for Regulators and its failure to even launch because of complicated structure and the divergent interests of its many funders was an instructive story for those who like to throw the word ‘global’ into their projects just because they can. Global scaling of projects isn’t easy. Local is hard enough. (PS: You’ll note that the last update to the page I linked to was made in 2007).

Michael Trucano from the World Bank breezed through a solid 50-slide presentation on the 9 things that cause tech-based educational programs to fail. Most were obvious, like failing to create educational content to match local needs, or only doing it long after the hardware had been deployed. But the fact that this list was based on real (though politely anonymous) programs brought home Trucano’s point that they are often missed in the enthusiasm to apply technology solutions to development problems. (For another great list, here’s MobileActive’s Definitive Guide to Failure)

There were even more telling tales of failure from Robert Kirkpatrick, now at UN Global Pulse and formerly employed at a “large software firm in the Pacific Northwest…” The meat of his speech was off-the-record and beyond the scope of this post. But that brings me to the real success of this event: that it even happened. Kirkaptrick made his presentation off-record for a good reason. The blowback for this kind of honesty in the face of failure can be drastic.

The naysayers about poor incentive structures in development aid have been around for a while. Bill Easterly and Tom Carothers, among others, have been beating that drum for years — there’s really no way for development organizations to honestly acknowledge their failures or learn from past experiences. But little has changed even in a sector so ripe with innovation and possibilities. Much of the discussion that followed the presentations centered on this issue. How do we create incentives for developers and deployers of ICTs for development to come to a place like Fail Faire and not have to be off-the-record? How do we make sure that institutions can learn from their own mistakes and those of others? Here are some of the points that came up, and good points all:

  • It’s easier to get people who have succeeded a little to talk about their failures
  • Have internal “fail faires” before coming out of the closet.
  • Look at the way the private sector, especially parts like quality control on manufacturing, deal with their failures. There’s a lot to learn from their business process.
  • Make a ‘failure clause’ a key part of the grant. Failure should not be incentivized, but if it happens, make sure that grantees are still rewarded for honest introspection.
  • Create a ‘sandbox’ where programs are allowed to fail and insulated from severe retribution. Such sectors already exist in government — think of failed rocket ships. Was NASA dismantled over them?
  • Create job security for those who you want to talk to about their failures.