How Twitter’s New Policy Rewards Elite Activism

by Sarah Kendzior on 1/29/2012 · 13 comments

On Thursday, Twitter announced that it would begin to selectively block tweets on a country by country basis. The decision prompted an immediate outcry from free speech advocates as well as a more measured response from scholars of social media, several of whom praised Twitter’s relative transparency while noting that it has no choice but to comply with the regulations of individual governments.

One of the most passionate defenders of Twitter’s new policy is Zeynep Tufekci, who described it as an “excellent policy which will be helpful to free-speech advocates”. Tufekci sees Twitter’s selective censorship as an improvement over the broad censorship practiced by other internet companies, in which content deemed offensive by one is deleted for all. Under the new guidelines, a tweet deemed inappropriate by the leaders of a particular country will only be censored within that country. To the rest of the world, it will be labeled as “blocked”, a development she describes as “excellent” because it renders state attempts to suppress speech transparent.

Tufekci, a scholar and advocate of free speech in the Arab world, has been criticized by some who see her post as a rationalization of censorship. While I disagree that this is her intention, her article does prompt troubling questions about the nature and purpose of Twitter: both for activists and their supporters. Tufekci’s thesis proceeds from the assumption that local activists get global followings. It assumes that Twitter activists are internationally connected and have a network of trusted advocates who will notice, and care, when their words are censored. This view is reflected in guidelines for online activists released by the Electronic Frontier Foundation, on which Tufekci consulted, that place the onus of activism on international digital networks facilitated by interpersonal trust.

In some parts of the world, particularly in the Arab region that is Tufekci’s focus, these points might make sense. But for much of the world, they highlight a fundamental misapprehension of the role social media plays in activist networks. The weaknesses of Twitter’s censorship policy reflect the weaknesses of Twitter itself.

* * *

Last summer, two Uzbek journalists, Malohat Eshonqulova and Saodat Omonova, went on a hunger strike to protest censorship and corruption in Uzbekistan’s state-run media. For one month, they tweeted the details of their strike from the Twitter account @Malohat_Saodat. They spoke of their physical agony, of the challenges facing reporters in Uzbekistan, and of the hypocrisy and corruption of the government of Uzbekistan. They posted videos of their strike to YouTube, and tweeted the links. Most of their tweets were in Uzbek, although some were in Russian. By the time of their final post, they had tweeted 730 times, had amassed around 65 followers, and had attracted no international media attention or global outcry. Malohat and Saodat’s near death on Twitter scarcely merited a retweet.

Why did this happen? First, Malohat and Saodat were writing in Uzbek, which few outside Central Asia can read. I translated some of their tweets into English and encouraged people to follow them, and several people did in response. But their case still attracted little interest. This brings me to my second point – almost no one cared about Malohat and Saodat because they were Malohat and Saodat. They were two journalists from Uzbekistan, a country with which few profess familiarity and whose activists do not use Twitter as a primary social medium.

Malohat and Saodat were subject to Twitter’s inherent popularity contest, in which the very few command the attention of the very many. In this system, well-connected English-speaking activists serve as the Justin Biebers of political dissidence: they attract followers based on brand recognition, and give the illusion that self-made internet performance breeds success. Malohat and Saodat did many of the things their Arab counterparts did, shortly after their Arab counterparts did them. Yet the world remained indifferent to their plight.

* * *

This is a story of Twitter activism before Twitter censorship. Now imagine what will happen once Twitter begins selectively censoring. What will become of the activists who lack both clout and Klout? Tufekci argues that Twitter’s transparency arrangement will focus attention on maligned activists – but this assumes that people outside the censored region will care. In reality, it is people within a particular region who follow regional crises most closely. Malohat and Saodat were ignored by the world, but they were followed closely by Uzbek activists and Uzbek independent media. I do not know what Uzbekistan’s policy on Twitter will be, but given its long history of internet censorship it will likely be one of the countries that demands Twitter block controversial content. And so cases like Malohat and Saodat’s will disappear from Twitter entirely, hidden even from their limited target audience.

Tufekci is correct that Twitter’s policy is realistic – as she points out, “the Internet is not a ‘virtual’ space, and cyberspace is not a planet which can float above all jurisdictions forever.” But realistic does not mean right. Twitter’s policy privileges the already privileged, hurting nascent dissident movements and the regional activists who struggle to promote them.

Subscribe to receive updates from Registan

This post was written by...

– author of 21 posts on 17_PersonNotFound.

Sarah Kendzior is an anthropologist who studies politics and the internet in the former Soviet Union. She has a PhD in cultural anthropology from Washington University in Saint Louis and an MA in Central Eurasian Studies from Indiana University. Her research has been published in many academic journals and media outlets, including American Ethnologist, Central Asian Survey, Demokratizatsiya and the Atlantic. She is currently an instructor at Washington University, where she teaches a course called "The Internet, Politics, and Society." Follow her on Twitter.

For information on reproducing this article, see our Terms of Use


Zeynep Tufekci January 29, 2012 at 4:42 pm

On the contrary, I don’t see why anyone needs a global following to use Twitter’s system to combat censorship. What you do need is to be more than one person, but that is the point of being a movement. If you are not a movement, or at least a sizable group, well, Twitter can’t solve that.

Tweets will be “marked” as censored and accessible in multiple, fairly easy ways (change your country setting, use a proxy, use Tor, go to…)

You can argue that the government can then come after the person. Yup. But again, that’s not something Twitter –or any Internet platform– can solve. In places where one can conduct political free-speech campaigns, Twitter’s set-up is excellent. In places where repression is too severe to do so, neither Twitter nor anything else is the solution (though, of course, global following does help in such cases).

The fact that people don’t pay attention to Uzbekistan is a problem; but as you know, Uzbekistan’s main problem is that the state is too strong and the opposition is too weak. Again, the Internet can help to a degree in such situations but there are no easy solutions to such problems. Internet is one of the ways in which people in Uzbekistan can try to gain more attention but, yes, they have many things stacked against them. It was worse in the age of television.

There is a lot of “what about X” type questions raised about the Internet which are not, correctly analyzed, problems about the Internet. They are usually about things for which the Internet is useful but not a magic wand.

Another example from a country on the opposite end of the democratic spectrum: some people were complaining to me that Twitter’s new policy is hard to combat in the face of “superinjunctions” in the UK. Um, yeah, Twitter can’t solve that problem for you, either. People need to organize a mass-tweet/blog campaign against these superinjunctions. It’s hard for me to advocate that a company engage in a civil disobedience –defy a superinjunction– by itself when the obvious, and morally and politically correct route is for people to either get organized and get the law changed, or engage in civil disobedience themselves as people.

In the end, my realism is not about what I expect from Twitter in light of the fact that they are a company; rather, it is about the fact that political problems require political solutions which are much broader than what a single company can or should do. In the end, people around the world would likely continue to ignore Uzbekistan even if Twitter never pulled a single tweet in response to the government. (Heck, I’d even argue it might help their case if they got censored on Twitter. Lots of people told me that Tunisian government’s censorship of the Internet got more attention than its repression of dissidents. I don’t like this last fact but it is the reality of the world).

Finally, I do believe all companies need to draw certain lines on cooperating with repressive regimes but I believe pulling tweets reactively (no a priori filtering) in face of court orders in a transparent and easily circumventable manner seem to be the right line for me. It sets the people up to kick at the ball; that’s up to us.

Sarah Kendzior January 29, 2012 at 5:29 pm

Thank you for your thoughtful comments, Zeynep. I think we agree on most issues. I also don’t see the internet as a panacea to political problems and agree that the key problem is the state. I have no problem with Twitter in general and agree that their policy is preferable to those of other companies. However, given that Twitter has the potential to influence and structure activist movements, I think it’s important to analyze the limitations of the new policy.

You write that “If you are not a movement, or at least a sizable group, well, Twitter can’t solve that.” This is true. But that opens the question of how small groups become social movements. I don’t believe that social media inherently produces social movements – people produce social movements – but social media allows people to effectively communicate and build alliances, which is important in a place like Uzbekistan where the state actively tries to keep dissidents from inside the country from interacting with dissidents outside the country. The new policy damages people in this vulnerable position. I’m less concerned with whether Uzbekistan attracts global attention than whether Twitter’s policy hurts activists who use social media for regional issues – and whether the problems they may face will be glossed over by the fact that the policy is less harmful to more established, internationally connected activists.

I don’t have a better alternative to Twitter’s selective censorship, but I am interested in hearing ideas from you or others who are knowledgeable about this topic.

Jillian C. York (@jilliancyork) January 29, 2012 at 4:55 pm

You certainly make some fair points, but I have serious doubts that Twitter will cave to demands by the Uzbek government. Furthermore, as I’ve noted elsewhere: this isn’t a new policy, it’s a new mechanism. Uzbekistan may have already, or at least could have already, submitted legal requests to Twitter under the previous system. Had Twitter complied, they would have been globally blocked, rather than locally blocked – that’s the only difference.

I’m glad you touched on the language issue, particularly because I bristle at the use of your term “elite activist.” The reason Arabic-language Twitter users can easily attract thousands of followers is that the world has 310 million native Arabic speakers (not to mention the non-native speakers). Contrast that even to Russian (let alone Uzbek), which has less than half of that number, not to mention the fact that the Arab world tends toward unity on a number of issues, and you can see on a very basic level why the two regions are very different animals when it comes to online activism (and that’s before considering all of the other sociopolitical factors, of course).

Sarah Kendzior January 29, 2012 at 5:40 pm

Thanks Jillian – I’m glad you responded. Those are good points, especially on the relative size/significance of the Arab world. I use Uzbekistan as an example because it’s the country whose situation I know best. I don’t expect it to shape Twitter policy. But I suspect that there are many activists all over the world in similar situations to those in Uzbekistan – regional activists whose situations do not get discussed because we don’t know much about them, and we don’t know much about them because social media skews our understanding toward popular languages, successful causes, familiar geopolitical contexts, and media-savvy dissidents who know how to attract an international following.

I agree that Twitter is making a compromise that is understandable, but it is still an unfortunate situation for a lot of people – the people whose voices are already the least likely to be heard.

Also, you are right that “elite” is not the best word to express what I mean – “influential” or “established” would have been better. I apologize for any negative connotation that term implied.

Jillian C. York (@jilliancyork) January 29, 2012 at 5:57 pm

“I agree that Twitter is making a compromise that is understandable, but it is still an unfortunate situation for a lot of people – the people whose voices are already the least likely to be heard.”

I’m still going to push back on this – you’re implying here that Twitter is going to censor vulnerable activists, when my impression–from conversations with Twitter policy folks as well as the stated policy–is that Twitter intends to be very selective and careful in the orders it follows. So unless German neo-Nazis are a vulnerable group, I’m just not seeing it.

Is it possible? Yes, and I fear that Twitter’s policy, under a different “administration” so to speak, could be used that way. But unless Twitter opens an office in Tashkent, it seems highly, highly unlikely.

Katy Pearce January 29, 2012 at 11:13 pm

Great point WRT the linguistic minority issue.

Sarah Kendzior January 29, 2012 at 6:06 pm

I’m going by what Twitter said in the statement it released. If they clarify what they want to censor, and it’s limited to a small number of groups like neo-Nazi organizations, then I would be wrong in worrying about more widespread repercussions. But their statement says they “will enter countries that have different ideas about the contours of freedom of expression” and that they may “withhold content” in those countries. I don’t have a lot of confidence from this statement that the censorship will be limited to the sort of groups you mention.

Jillian C. York January 30, 2012 at 2:01 pm

That’s not exactly what they said.

Germany and France have different ideas about freedom of expression than the US. The rest of that paragraph, by the way, adds some clarity: “Some differ so much from our ideas that we will not be able to exist there. Others are similar but, for historical or cultural reasons, restrict certain types of content, such as France or Germany, which ban pro-Nazi content.”

I have confidence, but I think that comes from my proximity to Twitter. I can certainly understand why people don’t necessarily trust them, but I nonetheless feel confident that they won’t be responding to absurd requests from the Uzbek government.

Josh Shahryar January 29, 2012 at 6:23 pm


I think there are a couple of issues here that need be addressed. Although I don’t agree completely with Zeynep’s assessment of Twitter’s censorship policy, I also think your post addresses an issue that may not be as related to Twitter’s policy.

I agree that Twitter has become a valuable resource for sharing human rights-related topics, but sharing doesn’t necessarily mean that something will also get traction. For the most part, many ‘super-star tweeps’ – as some call them – get traction because people who care most about the issue – those facing similar circumstances in the same geographical location – are on Twitter or the social media on which the tweep is. These secondary sharers create a snowballing effect whereby the primary sharer’s concerns find a larger audience by extension of the secondary sharers. This is where tertiary sharers, the larger international audience comes in, who are generally receiving the info from the secondary sharers and it just blows up after that.


– Primary Sharer: most directly affected by the concern and almost always within the geographic location shares.

– Secondary Sharers: Indirectly also affected by the concern and usually within the same geographical location, reshares, creating a snowballing effect

– Tertiary Sharers: Not directly or indirectly affected by that concern and not within the geographical location, shares and finds international audience for the issue.

In the case of Uzbekistan, there are very few secondary and hence few tertiary sharers because Twitter is not as popular in Uzbekistan as it is in say Egypt.

Now there are other countries in a similar situation, like say Iran. They follow almost the exact same rules through other means, such as posting on and then letting it slowly trickle down to Facebook from where it’s picked up and shared on Twitter. There are other examples. But the main point is that for Uzbekistan or any country’s issues to find traction on the internet, you need secondary sharers, which Uzbekistan lacks, imo.

I for one know of very few tweeps from Uzbekistan, whereas you have thousands of Egyptians on Twitter who constantly bombard the internet with Egypt-related tweets from primary sharers. Until Twitter becomes popular in Uzbekistan and there is a sizable number of secondary sharers, human rights issues in Uzbekistan will get little traction on Twitter.

That being said, I’m sure even if Twitter does get completely censored, people will find a way to share issues on the internet. One of the main things that came out of the social media ‘revolution’ – I hate that word – is a realization that if there’s a will to get something traction on the internet and enough people put their time and effort into it, it will almost always find an audience. I for one would be very interested in a post about how Uzbek activists maybe using Russian language blogs and websites to share their situation with the world at large.

Sarah Kendzior January 30, 2012 at 6:55 am

Thanks for your comments, Josh – I agree with your assessment of how Twitter works. But I’m less concerned with whether tweets gain wide traction than that I am with whether their target audience is able to read them. For many activists, reaching an international audience is not a key goal – their main objective is to reach the regional audience that the state has rendered inaccessible. But because “success” is often measured in terms of international recognition, policies like Twitter’s tend to be defined – and defended, as in Zeynep’s piece — on similar grounds.

The conditions in Uzbekistan are so oppressive that social media networks are unlikely to have a major impact, at least at this point in time, even if they brought Uzbekistan’s problems to international attention. But social networks help information flow in and out of Uzbekistan and facilitate meaningful dialogue within small dissident communities. Twitter’s policy may threaten that.

Uzbek dissidents frequently use Russian, although many prefer Uzbek out of national pride or for literary reasons. (Many Uzbek dissidents are poets.) I’ve written several articles (and recently, a dissertation) on this subject; they’re available at my website:

Josh Shahryar January 30, 2012 at 7:45 am

You do have a point there.

But correct me if I’m wrong, won’t the Uzbek government block any social media website that it sees as gaining too much local traction?

And yes, as a journalist covering human rights, given the experiences I’ve had with many other countries’ extremely harsh crackdowns on websites – say the case of Iran -, I am a bit skeptical about social media gaining local following in such regimes because of the threat of government clamp downs.

Sarah Kendzior January 30, 2012 at 9:03 am

The Uzbek government blocks opposition, political, foreign news and religious websites, but so far they have kept social media networks relatively open, with the exception of a few temporary blackouts. They have also designed a state-sanctioned social networking site where citizens have to register using an Uzbekistani phone number — this prevents exiled Uzbeks from joining and allows the state to more easily monitor what is being said and by whom. Twitter has thus far played a minimal role in Uzbek politics. Facebook is more popular than Twitter, but the most popular sites are the Russian social networking sites (Odnoklassniki, VKontakte etc). That said, one of the biggest obstacles to effective mobilization via social media is the knowledge that the government is monitoring user activity. Self-censorship plays as big a role as censorship.

Sarah Kendzior February 1, 2012 at 7:47 am

A great new post on BoingBoing today reiterates many of my points:

“We keep talking of activism as content: it’s as privileged a viewpoint as you’ll ever get from the silicon tower of tech journalism, where the act of disclosure is more virtuous than having nothing to disclose, and where the West’s ethical feather-plucking is more real than the reality of politics in dangerous places.

Silicon Valley seems finally to be learning the lesson that if you sell yourself on virtue, the business will make you eat your words. Twitter’s U-turn on censorship teaches it another one: if you take credit for what activists do with your tools, you’ll end up eating their words, too. “

Previous post:

Next post: