I spent the past two days in Cambridge, primarily around MIT, and almost exclusively talking about the “Arab Spring” and what we’ve learned about social media and protest in authoritarian states. Early Wednesday morning, the MIT Museum hosted a “soapbox” session, which put Dr. Marlyn Tadros and me in dialog with Egyptian protesters and bloggers, including Mahmoud “Sandmonkey” Salem, who I was thrilled to meet virtually. Events via video are tricky, and there were some issues with sound quality for the folks watching in Cambridge, but the resulting video of the event is excellent.
The highlight of the two days in Cambridge was an event I hosted at the MIT Media Lab yesterday afternoon, a conversation called “Civic Disobedience“, which featured three of my favorite people, who also happen to be three folks extremely knowledgeable about social media and the Arab Spring.
Zeynep Tufekci is an assistant professor of sociology at UMaryland Baltimore County, where she studies social networks on and offline. Her blog, Technosociology, has become required reading with very insightful essays on Wikileaks, the Arab Spring and other recent intersections between online and offline social networks.
Clay Shirky has been doing some of the most interesting writing and thinking about the internet and human relationships, since 1996. He teaches at NYU in both the journalism department and in the Interactive Telecommunications Program, writes extensively online and has published two key books about the internet, participation, groups and social change.
Sami ben Gharbia is the director of Global Voices Advocacy, the free speech arm of Global Voices Online. He’s the co-founder of Nawaat.org, one of the central actors in the Tunisian dissident media space. He was exiled from Tunisia 13 years ago and returned home for the first time a few weeks ago, in the wake of Tunisia’s successful revolution. He is also one of the smartest critical thinkers about the limitations of our current understandings of internet and social change – his essay, The Internet Freedom Fallacy and Arab Digital activism, should be required reading for anyone expressing an opinion about “internet freedom”.
With these three folks on stage, I had virtually nothing to do as moderator. So I took notes, which I’ll share here, to tide you over until the session video is posted.
Sami opened the conversation by giving his view of how social media had helped enable protests in Tunisia. He offers three-part model that treats social media as part of a more complex ecosystem, involving Facebook as a publishing platform, multiple curation platforms (Nawaat, Global Voices, Twitter, Posterous) and broadcast platforms (AlJazeera and France24).
Facebook became central to the Tunisian media ecosystem because all other sites that allowed video sharing – YouTube, Daily Motion, Vimeo and others – were blocked by the Tunisian government, along with hundreds of blogs and dozens of key twitter accounts. This censorship, Sami argues, drove Tunisian users towards Facebook, and made it hard for the government to block it. The government tried in 2008, but the outcry was so huge, they reversed course. The main reason – usage of Facebook more than doubled during the 10 days of blockage as Tunisians found ways around the national firewall and onto the service.
Censorship, in general, because a unifying force in the Tunisian online sphere. Reacting to censorship taught Tunisians how to disseminate information through alternative paths and helped them use social media for advocacy in a time of crisis. For all the disagreements Tunisians have with one another, they can agree on censorship as a common enemy. This is why, when Ben Ali offered a final set of concessions to his people on January 13th in a desperate bid to hold onto power, one concession was the elimination of online censorship.
Facebook was an important platform for Tunisians for publishing, mobilizing and organizing, Sami tells us. But it’s a very limited platform. It’s closed, both technically and socially, which can make it extremely difficult for journalists to find people to interview about stories. And Tunisia can be linguistically closed, even to other Arabs – the Tunisian dialect is a mix of French, Berber, Italian and Arabic that can be very hard to penetrate. While Facebook was used to share videos, it also made it very hard to figure out the origins of those videos – when were they originally published and by whom? For Facebook to be useful for a wider audience than Tunisians, you needed Tunisian users to identify key pages and profiles and bring them out of Facebook’s closed system and into the open web.
That’s what curators did. Sites like Nawaat were critical in identifying content posted on Facebook, tagging, timestamping and categorizing it and making it accessible to other media organizations. Both Nawaat and Global Voices translated key pieces of content, and Nawaat used a Posterous blog to identify over 400 videos, many of which were used by Al Jazeera.
Once content made it onto Al Jazeera, it began filtering back into Tunisia, letting Tunisians who weren’t looking for content online understand what was unfolding. Jazeera has a huge audience in Tunisia, though it’s never been allowed to report there. (I’d been telling people that Jazeera had been forced to stop operating in Tunisia by Ben Ali – Sami tells me Ben Ali never let them in at all…) Jazeera, Sami argues, became an extension of the internet, publishing user-generated content and using it to educate Tunisian citizens about what was going on in their own country, and eventually the whole region. Tunisians knew how important Jazeera was once police officers began heading into cafes and begging owners to switch their TVs to another channel.
This three part model created an information cascade that Sami believes directly led to the revolution. He cites some key events that gave the media disproportionate power. One was the Tunileaks/Wikileaks cables. Tunileaks received cables about Tunisia sent from a dissident within Wikileaks who was upset that the group was cooperating only with mainstream media and not citizen media. Tunileaks released these cables well before Wikileaks released their archive of cables. (I asked Sami, “You’re involved with Tunileaks, right?” His response: “I am Tunileaks.” :-) Sami and friends used Google Appspot to publish the cables, knowing that the service rested on a set of IP addresses used by several other key Google services. This meant that, in blocking the cables, the Tunisian government was forced to block other key services, raising attention to the cables and encouraging more people to use firewall circumvention tools to access them.
Sami also cites the Anonymous attacks on Tunisia as another key turning point. They weren’t especially effective, but the story was so sexy, American media had to start paying attention.
Expanding on Sami’s analysis of the ecosystem, Zeynep offers the idea of analyzing social media and revolutions in terms of “meso-level causal mechanisms”. (After offering that phrase, Zeynep gives a disclaimer that she’s early in her analysis and just “thinking out loud”. That her thinking out loud includes phrases like “meso-level causal mechanisms” gives you a sense for why she’s so worth reading.) There’s a temptation, she says, to view social media as like other media, just faster. But that fails to see some of the key nuances.
There are network effects that come from social media. The shape of connectivity networks changes – people are more directly connected to one another, rather than being clustered into separate groups, linked by bridge figures. Tunisia, in particular, has an online social network “with one giant component, one big, heavily linked space, probably related to the anti-censorship campaigns Sami spoke about.” This network is big, tightly connected and fast, and information passes through it much more quickly than it passes through offline social networks.
There are field effects as well. When media reaches a broad audience, either through social media or through broadcast, it’s possible to affect the mood of the country all at once. And we see network to field effects: information cascades. The experience of Tunileaks was, in part, the revealing of hidden preferences. Tunisians knew they weren’t fond of Ben Ali, but discovering that no one liked him, including the US, had an important effect. When Egyptians looked at Tunisia and said, “We can do this, too!”, that’s also a network to field effect.
The meso-level mechanisms include increased participation. We don’t always like what we get when we see increased participation. Increases simply accelerate and strengthen dynamics that are already in place. In a polarized situation, increased participation often means increased polarization, which is what we may be seeing in Bahrain. That makes it hard for participation to lead towards coordinated action. In Egypt, near the end, “Mubarak’s dog didn’t like him. Much as we wish it was, that’s not the case in Iran or Bahrain…”
Another meso-level effect is faster information diffusion. This can mean the ways audiences are segmented change as well. Information that might have been accessible only to a literate class is not accessible to non-literate people as well. In much of the Middle East, there’s a big divide between the literate and non-literate public spheres – when those distinctions collapse, there’s the possibility of coordination between those two groups. On the other hand, the Habermasean pubic sphere (which may never have been as calm and reasoned as Habermas wished it was) can get downright emotional. The emergence of Mohamed Bouazizi as a rallying point helps show the emotional nature of the narrative in Tunisia.
One way to understand how big these changes are is to watch the shift in “coup etiquette”. In her native Turkey, Zeynep tell us, you can tell a coup based on what song is playing on the radio. “If you hear this one specific patriotic song, you know it’s time to go buy bread.” That’s because coup planners traditionally seized the radio and television stations first. In Egypt, there was a debate amongst Tahrir protesters about seizing a television station – in the end, they decided not to bother. The emergence of social media makes broadcast less relevant, though probably not irrelevant.
Authoritarian states are very experienced at trying to silence dissent, Zeynep reminds us. They are very good at playing whack a protest, and most of the time, they’re successful, using a “quarantine” model to separate protesters from the rest of the state. She cites a protest in Tunisia in 2008 in the mining town of Gafsa, which the Tunisian government successfully defeated, by surrounding and isolating the protesters. In Sidi Bouzid in 2010, enabled in part by social media, a very similar crackdown failed to stop the spread of the protest “virus”.
Sami added a key note to Zeynep’s model, pointing out that the Sidi Bouzid protesters appealed to the rest of the nation for support with their demands. The protesters in Gafsa focused their grievances on a local mining company, which made it very hard for the rest of the nation to join in supporting them. “They quarantined themselves, in a way.”
Given Clay’s extensive writings about social media and protest, I asked him to evaluate what he got right and wrong, in light of events in Egypt and Tunisia. Warning us that four months isn’t long enough to understand what’s actually gone on with these protests, Clay explains that he feels recent events have confirmed his thoughts about the importance of synchronizing groups. “Governments aren’t afraid of informed individuals – they’re afraid of synchronized groups.” In particular, they’re afraid of groups that have shared awareness.
With authoritarian states, there are three possible states. In the first, everyone knows the government is corrupt. In the second, everyone knows that everyone knows the government is corrupt. In the final stages, the ones where governments collapse, everybody knows that everybody knows that everybody knows. Clay argues that autocratic regimes can survive the first and second phases for years – that third stage, where shared awareness leads to synchronization, is more dangerous for autocrats.
What he got wrong, he says, was overemphasizing the use of tools for coordination for protest. “I concentrated too much on using tools to get people out into the streets. It turns out that bringing people out into the streets only works if it’s the end of a long process. It’s not a replacement for that process.” This, he believes, is why Egyptian protests were successful – they leveraged long-standing networks like Kefayah. But without those networks, going into the streets can be very dangerous. He cites an example worthy of Evgeny Morozov – when Sudan feared a revolution, “they used Facebook to call a revolution againt itself, then arrested everyone who came out, as they were the people most likely to make trouble.”
Referring to Zeynep’s mechanisms for action, Clay says he believes that social media “synchronizes opinion, coordinates action, and documents results.” The medium is less relevant than these processes – it’s not about mobiles versus Facebook versus Jazeera. If you want to know how seriously to take these effects, Clay suggests you look at the fact that both insurgents and autocrats believe these tools matter, and take risks to act on these beliefs. He offers the example of Libyan officials searching people fleeing across the Tunisian border for digital cameras and USB sticks. “Even Qaddafi doesn’t like letting documentation of murder reach the rest of the world.”
Clay shifts the conversation to the issue of “internet freedom”. Noting how influential Sami’s essay was on his thinking, Clay suggests that the US overestimates the value of access to information and underestimates the value of access to each other. If we wanted to promote internet freedom, we need to think more about synchronization and less about information in considering these tools.
I asked Sami if he’d softened his stance on US involvement with internet freedom from his earlier writings. He points out that US support for the Iranian protests helped Ahmedinejad make the argument that protests were instigated by outside agitators, when they were actually a legitimate domestic movement. “In Tunisia, we fought very hard to keep our movement independent from foreign interference, including avoiding those who were collaborating with the government.” That said, Sami acknowledges that there’s a big difference between public statements by the US State Department and actions behind the scenes, which is often very productive and positive. What Sami would like to see the US doing publicly is controlling the sale of censorware, not advocating for freedom while allowing some of the key filtering technologies to be sold to repressive governments. He notes that individuals are also capable of taking effective steps in solidarity with dissidents – hosting video archives and mirroring key content to help make it visible in Tunisia, smuggling communications hardware into Egypt and Yemen, even calling attention to protests through actions like those of Anonymous.
Zeynep suggests that we not dismiss the Iranian green revolution as a failure. Much as the failed Dean campaign helped elect Obama, the Iranian protests helped us understand how to use social media for revolutionary change. While she supports efforts to get the US to be more consistent on internet policy, she suggests the larger problem is getting US foreign policy to shift from supporting dictators. “I’m betting most, if not all, will be gone by the end of the decade.”
Clay suggests that watching other country’s revolutions matters enormously, in terms of bearing witness, moral support, and in the case of US citizens, influencing the policy of a superpower. He’s happy to admit an normative bias for democracy and free speech and to support a foreign policy that respects this. But this demands we push for consistency.
“I urge my students not to try to pay attention to the whole world, but to start by picking a country to care about. Mine is Bahrain, and I believe we need to make visible the tension between our politices and our current support for Bahrainm which is becoming an apartheid state run by Saudi Arabia.”
Clay doesn’t believe the US should stay out of fields like internet freedom. “We can’t. We need bilateral relationships with everyone.” But we need to recognize that we’ve lost the ability to speak in three separate voices – one directly to other states, one to the public and one to the cognoscenti. Twitter and Wikileaks have collapsed these channels, and as a result, the US may need to speak a lot less, at least in public.
As the discussion moved into question and answer, it became significantly more free-flowing, and I had to moderate rather than taking notes. I will mention a couple of exchanges that stuck in my memory:
- A questioner asked whether we’ll see social media playing an important role in governance as well as in revolution, suggesting that the social media revolutions that elected Deval Patrick and Barack Obama have been disappointing in terms of participatory governance. Sami made the point that Tunisians need to rebuild a vast range of institutions – an independent media, NGOs, transparency organizations, political parties, and that all were being rebuilt using new media and social media tools.
- A good deal of our discussion involved analogies to previous revolutions. Sami made a key point – the revolutions in Tunisia and Egypt were not trying to overturn existing systems of government – both states have been constitutional democracies. The revolution wasn’t to change the form of government, but to get it to be respected.
- Professor Ian Condry suggested that, if these revolutions took ten years to unfold, we need to think through what ten-year changes might be underway now. Clay pointed to Paul Ford’s essay “Why Wasn’t I Consulted” and suggests that the assumption of participation may be a key ten-year change.
- Nitin Sawhney pointed the audience to three examples that appear to contradict the relationship between communications technology and democratic revolution. The Islamic Revolution used a new technology – cassette tapes – to lead to non-democratic change. The Palestinian first and second intifadas were organized with virtually no technology and were effective forms of resistance. And in Bahrain, being heavily wired hasn’t led to a successful revolution. In each case, American foreign policy seems to have mattered more than communication technology. The panel responded by acknowledging that none think that communications was the key or sole factor in the changes in Tunisia and Egypt – however, Clay argued that states try to keep an equilibrium state between the utility of new tools and the inability of citizens to syncronize protest, and that new technologies may destabilize that equilibrium and offer an opportunity for change.
We should have video for this session soon – I will post it once it becomes available. Sincere thanks to my three friends for their wonderful talks and to the audience for a great conversation.