UPDATE: addition of Web Ecology Project research (Nov. 8, 2011)
UPDATE: David Faris on the role of blogs in framing processes (Nov. 9, 2011)
Since the Arab Spring began last December, both academics and amateurs have studied the effects of digital technology on the revolutions. At this point there have been so many insights that what is needed is a good curated aggregation of the best answers to the question: What effect did digital technology and social media have on the Arab Spring?
The purpose of this post is to serve as a dynamic compendium of the best answers to this question, and I hope to update it frequently as new insights are gained. Here is what I’ve come up with so far.
Step 1 – In the Beginning: a Safe(r) Space to Reveal Preferences
In a recent post on Gigaom, Mathew Ingram quotes open data advocate Aaron Swartz‘s quick and dirty summary of Jon Elster‘s theory of the steps of a revolution:
- A core group of committed activists get together to “do something completely crazy.”
- The government cracks down, and this behavior makes people who are sympathetic to the cause “rally to the support of the crazy ones.”
- As the protests continue and it looks as though they might have some tangible effect, at some point “it seems worth it even for just normal reasonable people to start joining in.”
- Eventually, the protests become so large that “even their opponents pretend to be part of them, so as not to be on the wrong side of history.”
Yet even before activists get together “to do something completely crazy” they need to simply get together. The Internet provides a safer space than offline for potential activists to meet and honestly share views.
Yes, it is true that repressive governments are getting increasingly savvy about this function of the Internet and are more actively tracking opposition activities online. Yet activists are becoming savvier too, such that identifying a digital activist who is truly skilled in anonymity is quite difficult to identify (witness the successful evasions of the hackers of Anonymous). At the same time, many governments lack the resources or understanding necessary to conduct thorough surveillance of the online activities of potential opponents, so simply being online may provide meaningful cover, even in the absence of sophisticated anonymity techniques.
This means that the Internet can help strengthen a nascent opposition movement by giving passionate individual opponents of a regime a place to meet one another, share and develop their views by revealing preferences, and build a collective identity that will make opposition to the government – and membership in the activist group – and ever stronger motivator of their behavior. (For more on like-minded groups and intensification of ideology see here.) This period of mutual sharing of discontent can last for years before different contextual factors align (changes in opportunity structure) and the time for mobilization is ripe.
At this point, the revolution has not even begun. The activists have not planned or executed any action. They are just talking to one another by using technologies like forums, blogs and social networks. If the government is relatively ignorant of how activist use the Internet, citizens can have these discussions on the public web, as Egyptian political bloggers did in 2004-2005. (The development of the quasi-political group Anonymous on the imageboard 4chan is another example of this phenomenon.) If the government is savvy, more sophisticated tools for anonymity will be necessary.
In a recent talk in Beirut, Zeynep Tufekci of the University of North Carolina at Chapel Hill noted that repressive governments prevent revolution by creating collective action problems for their citizens, specifically high individual cost of failure (torture, prison) and high cost of organization (difficult to meet, communicate).
The Internet can help citizens solve the first collective action problem of living in a repressive regime: the inability to communicate with others who share their feelings of opposition. This step actually involves two functions:
- The capacity to reveal ones’ preferences to others and
- The capacity to communicate with others who share that preference
Social platforms like blogs, forums, and social networks are very effective places to reveal preferences to others. While it is easier to maintain anonymity in a forum or chat room that on a social network, where social connections can be used to reveal identity, all three provide an opportunity for citizens to mutually share their frustration with the regime, either explicitly or in coded/indirect language. Because these platforms allow for many-to-many communication, it is possible to have mutual statements of preference. It’s not just one person standing on a digital soapbox and shouting into the darkness, it’s the capacity for others to respond, “me too!”
The Boston Review recently published an interview with Ahmed Saleh and Nadine Wahab, two administrators of the Facebook page We Are All Khaled Said. The page was a key meeting point (and then mobilizing point) for the young urban middle class before and during the Egyptian Revolution. It is named for Khaled Said, a young man beaten to death by police in June of 2010. Saleh describes how the page served as a freer meeting place for politically aware Egyptian youth:
The youths in Egypt, pre-revolution, lived two lives, one online and one off-line. The off-line life is very limited in access to information, freedom of speech and mobilization, and even in access to each other. For decades, it was illegal for five people to gather for any reason (per emergency law), although it was tolerated except when it was politically motivated. Online political activists used terms like “group,” “room,” and “comment” as if they had physical meanings.
Nadine Wahab, another page admin, concurs:
Since the Egyptian government had made the brick-and-mortar world so unfriendly to free expression and the Internet was so readily available to just tweet, update Facebook, or send a quick blog post, it became the space to express your thoughts or post a news item.
The page also was a place to reveal preferences about a different future for Egypt, all while protecting themselves through the use of false names. Saleh continues:
The Internet offered an open environment that politicized the youths, allowed them to raise awareness on possibilities of shaping their future, diversified their perspectives, anonymized their identities, gave them the taste of free speech, and pushed them to see through the regime propaganda and despise it.
Wahab explains how the admins explicitly attempted to use the Facebook page to create a public debate about the Mubarak regime, to create a readiness for mobilization against the regime (ie, framing processes):
As the people posted live, people would react live and a conversation developed. I believe 2010 was a tipping point for this interaction; we went from conversation to a public debate, and just not with activists but with a larger, less engaged tech-savvy population. Administrators were very deliberate in cultivating a relationship with this population.
David Faris of Roosevelt University, who is writing a book on digital activism Egypt, confirms the importance of digital tools in framing processes and also highlights the importance of blogs, particularly in using torture as an issue with which to challenge the legitimacy of the Mubarak regime.
Digital is indeed wonderful for framing processes, but taking the long view, what made Khaled Said possible in the first place were past attempts (largely digital with offline connections) to make torture part of the public sphere, and to introduce the tortured as claims-makers in Egyptian politics. So efforts like Torture in Egypt and Misr Digital were critical stage-setting efforts, made possible by digital tools, that transformed the discursive environment around torture long before [Khaled Said admin] Wael Ghonim ever set foot on the political stage.
This raising or awareness (and creation of collective identity around that awareness) was key in creating a group of people ready to be mobilized when the right moment hit. As research provided later reveals, people using a Facebook were among the first to turn up at the demonstrations in Tahrir Square. Though there were other factors in place, the key opening was the successful revolution in Tunisia, which showed that dramatic change was possible.
Step 2 – Solving the Next Collective Action Problem: Collaborative Planning
Once a group of people are known to one another and have mutually revealed their preference for political change, it will not be too long before someone suggests that they do something about it. This will involve a more involved form of collective action than collective identity development and group formation. Activists will now need to jointly identify and analyze contextual factor (political, social, economic) that will determine whether and how they can act. They will need to develop, delegate, and track tasks. They will need to develop a plan.
Fortunately, the Internet provides a plethora of tools for collaboration. Some technologies like mobile SMS, instant messaging, Skype, and email, can be used with the highest protection of anonymity by using various types of encryption. More public tools, where anonymity is guarded by using a false name and accessing the platform via a circumvention tool like Tor, can also be used in this stage.
However, while actions can be planned on a public platform like Facebook (and were in the case of the We Are All Khaled Said page), detailed campaign planning is best suited to the careful work of a small group, where coordination is easier, meaning SMS, IM, Skype, and email may be safer and more useful. From his base in Dubai, Wael Ghonim, the most famed admin of the We are All Khaled Said group, used GChat to communicate with other activists back in Egypt. There are also a variety of more specialized technologies like fundraising widgets and wikis and Google Docs that can be useful for aggregating information and jointly creating a plan.
Unfortunately, because these activities take place through private platforms like email and SMS – rather than public platforms like Facebook and Twitter – it is often hard to find examples of how technologies were used in collaboration without asking activists themselves.
Step 3 – Going Public and Getting Big: Mobilization, Information Cascades, and Media Narratives
Digital technology has been used in many ways to facilitate political change, and the revolution is not public yet. It is only now that the first step of Elster’s model occurs. Having used digital technology to mutually identify other people unhappy with the regime and to collaborative create a plan of action, now that core group of committed activists can get together to “do something completely crazy.”
At this stage the Internet and mobile phones become useful for spreading information to a broad public. This information can range from mobilizing calls to action (“go to Tahrir tomorrow!”) to anti-regime preferences (“Mubarak out!” “proud to be an Egyptian #jan25″). The Khaled Said Facebook page was used to mobilize the first wave of protesters, although many groups that were mobilized existed before and outside of the realm of digital activism. Ahmed Saleh explains:
If it weren’t for Facebook, the Egyptian revolution would have started anyway. The effect of a Facebook call to a timed revolution with a large outreach (that activated an organized political activist community that’s been in the making for decades) is making the revolution shorter, more organized, with fewer casualties and more theatrical.
Saleh also describes how quickly information spread and how quickly other activists began self-organizing, such that Facebook became less important:
Additionally it is my claim that in the afternoon of January 25, 2011 when the masses came out, the Internet and Facebook became irrelevant. In fact all of the administrators of the Facebook pages and even the political activists were surprised that the demonstrators continued protesting all over Egypt on January 26 and beyond, without any Facebook page calling for it or organizing it. The administrators were now on the receiving end of the news.
It is here that the important function of information cascades comes into play, which means that people observe the actions of others and, as a result, make a choice to undertake the same action. The classic example is the protests in East Germany that led to German reunification. (This case is described by Susann Lohmann in greater detail in this 1994 paper.) From a summary on Wikipedia:
Small protests began in Leipzig, Germany in 1989 with just a handful of activists challenging the German Democratic Republic. For almost a year, protesters met every Monday growing by a few people each time. By the time the government attempted to address it in September 1989, it was too big to quash. In October, the number of protesters reached 100,000 and by the first Monday in November, over 400,000 people marched the streets of Leipzig. Two days later the Berlin Wall was dismantled.
In this example, the mechanism of the information cascade was that citizens physically saw the protesters out in the streets. Each Monday the presence of the protesters made political change seem more possible and each Monday a few more people became convinced that if they joined it would make a difference. While in the beginning only the hardcore of activists were convinced, by the end large swaths of the population saw the end of a divided Germany as inevitable. The same occurred with the street protests in Tunisia and Egypt, though not only because citizens saw the protests in real life, but also through various forms of media.
In the Internet age, information cascades are networked and include multiple media types, jumping between citizens, traditional media outlets, and back again. In the case of Tunisia, activists used Facebook and sneakernets to transmit video of the Sidi Bouzid protests (and subsequent government crack-down) to international TV broadcasters like Al Jazeera, which have a much larger audience than any Facebook page. Al Jazeera reporters had been barred from entering Tunisia and they were eager to find a way to work around the blocks by using citizen media. Viewing these protests on TV – and hearing about them on the radio – was the mechanism by which many Tunisians learned about them and made the decision to start protesting in other cities. (Ethan Zuckerman also has a nice video describing Tunisian media protest dynamics).
Recent survey-based research analyzed by Zeynep Tufekci and Christopher Wilson and collected by Alix Dunn of The Engine Room has also shown that, as many surmised, Facebook was very important in bringing protesters to Tahrir Square on the first day of the Egyptian protests, before there was any revolution for the international media to report on. First-day protesters were also likely to use Twitter, another means of mobilizing supporters, though the overall use of Twitter was low among protesters in general and miniscule in the Egyptian population.
During the Arab Spring, social media information cascades had both a local and international function. For example, it could be argued that the primary value of Twitter was not as a tool for local mobilization but as a mechanism of international contagion and agenda-setting. From early on in their revolution, Tunisian activists used the Twitter hashtag #sidibouzid as a way to aggregate information about what was going on in their country and broadcast it to the world. During the Egyptian revolution, people across the world followed the #jan25 hashtag to get the most recent information on what was happening when international media outlets lacked the resources or insight to provide that information. Every country involved in the Arab Spring had an associated hashtag, either a specialized hashtag like #feb14, the revolutionary hashtag for Bahrain, or simply the name of country.
The exact mechanics of informatoin cascades also differed from revolution to revolution. Research on Twitter activity during the Arab Spring by the Web Ecology Project, a volunteer research collective not dissimilar from the Meta-Activism Project, revealed that:
In the case of Egypt, lots of information flows start from journalists, bloggers and activists, with bots as a lesser, but important, influence. In Tunisia, there were fewer flows started by journalists, more by bots and bloggers, and way fewer from activists. This may reflect the fact that the Tunisian story caught many journalists and activists by surprise – they were late to the story, and less significant as information sources than the bloggers who cover that space over time. By the time Egypt becomes a story, journalists realized the significance and were on the ground, providing original content on Twitter, as well as to their papers.
The fact that citizens could record their own content and share it with sympathetic international broadcasters like Al Jazeera also meant that they could control the narrative of their revolution as never before. And, if they lacked a sympathetic broadcaster in the mainstream media, they could just broadcast themselves internationally, by dropping the Twitter hashtag of their revolution into their tweets to gain the attention of an international audience. In this way, no individual activist had to build an individual following. They just had to attach their own content to an already popular hashtag to gain the attention of the world.
In our current media environment, information can enter the media system through an increasing number of entry points. This makes it much more likely that important information will spread broadly within the population for whom it is important (from local to global) and makes it much harder to maintain information vacuums. Even in places where the international media is forbidden, a single memory card with video of a protest can make its way onto the Internet or out of the country in an activist’s back pocket.
Conclusion: a Catch-22 for Repressive Regimes
Going back to Elster’s timeline of revolution from the beginning of this post, we can now see that Elster’s first step (activists doing “something crazy” in public) is actually the end of a long process of group formation and planning that is less visible. This may be why governments in Tunisia and Egypt were unsuccessful in beating back the revolutions. Once the revolutions were visible it was already too late.
Repressive regimes need to nip digital activism in the bud if they are to prevent digitally-facilitated revolution. But this is a catch-22 for repressive regimes. “Nipping digital activism in the bud” means controlling and punishing behavior that is not an explicit act of rebellion, like sharing a news story about government corruption or talking about political change in another country, both part of step 1. When governments control this speech they end up controlling the speech of more people, as not only hard-core activists but also mostly apolitical citizens might wish to discuss these topics. This means that they are creating resentment against the government by citizens who were not previously politically engaged. This may be what is happening in China.
Also, it is important to note that digital technology was not used as effectively in all the revolutions of the Arab Spring. Even when digital technology was used, as in the cases of Syria and Bahrain, much stronger contextual forces (a government willing to violently repress and Saudi support of the government, respectively) may have overwhelmed the positive effects of digital technology. In other cases, like Libya, low Internet penetration and significant blocking meant that digital technology played a limited role, though some international broadcast was still possible using citizen media (see Andy Carvin’s talk on citizen media in the Libyan revolution).
Please let me know in the comments if you agree with this analysis of the role of digital technology in the Arab Spring and if you have evidence that supports or contradicts it.