Losh essay

Hashtag #Euromaidan: What Counts as Political Speech on Instagram?

Dr. Elizabeth Losh, University of California, San Diego

download the essay

#Tahrir, #Zuccotti, #IndiaGate, #Taksim, #Ferguson. All of these hashtags index a particular informational thread of protest within the larger cultural conversation. Such hashtags simultaneously mark the importance of a specific geographical location in which the body is at risk – of being jailed, deported, beaten, tortured, or executed – and in which the body becomes part of a political crowd that agitates for a more direct and less representative form of democracy. Malcolm Gladwell has famously insisted that “the revolution will not be tweeted,” because he believes that it is necessary to have “skin in the game,” as activists marching in the civil rights movement facing fire-hoses and police dogs did. Those who merely participate remotely in social movements by expressing solidarity with demonstrators through disembodied electronic means – on sites such as Twitter, Instagram, or YouTube – are labeled as “slacktivists,” who conveniently avoid “arrest, police brutality, or torture,” according to Evgeny Morozov – one of Gladwell’s sources. Morozov claims that those expressing solidarity with demonstrators may be “loud” enough to attract more media attention but too “lazy” to effect change.

Of course, those participating remotely may also have “skin in the game” if their IP addresses point to risky dissident activities. Sympathizers who curate, remix, and disseminate digital files with politically incriminating content may also find their in-real-life bodies at risk, if they are within reach of authoritarian regimes, as activists affiliated with human rights groups such as Witness or Global Voices remind us. Even if they avoid clashes with the police, those who are present live at political demonstrations may face after-the-fact consequences, now that many government forces sponsor facial recognition initiatives, including ones that rely on low-tech patriotic crowd-sourcing of the work of identifying potential culprits.

In Networks of Outrage and Hope Manuel Castells asserts that political power is expressed both through violence and through knowledge systems. I would agree that the relationship between the body that can be counted and the body that can be harmed is central to the cultural imaginaries of revolution in the age of computational media and digital networks, although the rhetorical functions of the body may be multiple and its relationship to knowledge systems critical. First, a body can serve as a pixel in the spectacle shaped by a larger crowd, as Latour and Weibel observe. One could say a body in this sense serves as a data point in a larger information visualization, such as a million man march seen from the air. Second, a body can be framed in a composition or posed in negative space to pay homage to the iconic images of photojournalism – classic shots now primarily encountered through the remixing and mashing up activities of blogs and social network sites rather than through traditional venues like newspapers and magazines. Such a body provides pattern in a particular media file, which may also have other distinguishing characteristics, such as its file format extension or time stamp. Third, a body of the present moment usually bears some form of wearable computing, generally a mobile phone in a pocket or a hand. This kind of body is able to participate in politics as a cyborg with a recording and communication device, simultaneously documenting and bearing witness to others.

All of these kinds of bodies are understood as capable of manifesting political agency. But what about other kinds of bodily performance in public spaces, which are characterized by the practices of everyday life identified by Michel de Certeau? In other words, how do we understand “politics with a small p,” which may be dictated not by participation in conventional party politics through highly ceremonial official processes such as voting in national elections but rather by seemingly contingent and minor consumer choices, language use, pedestrian mobility, traffic behavior, negative politeness (perhaps by giving a wide berth to those interacting with the police), or demonstrated affinity for particular songs or Internet memes, which might serve as anthems or emblems of protest. As Henry Jenkins and Clay Shirky have argued, even fan sites, social network platforms, or links to download music exist on the continuum of legitimate civic participation. However, Mimi Ito and Wendy Chun have pointed out that more rigorous analysis of individual cases with the tools of ethnography or close reading reveal more complex stories. For example, the protests by “candlelight girls” against U.S. meat exports to Korea that escalated to calls for president Lee Myung-bak's impeachment may represent more complicated dynamics among users of sites such as the Cyworld social network (Chun) or the online forums of Cassiopeia, a fan club for the boy band Dong Bang Sin Gi (Ito) than Internet positivism may indicate. Insights from geography about place-making and from human-computer interaction about distributed cognition are also transforming the discipline of media studies and its readings of political behavior. The turn toward attention to embodiment, materiality, situatedness, affect, and labor among those who study digital culture should change how we study something like speech acts marked as political on Instagram.

Activities in public spaces may carry multiple political valences simultaneously. In using the work of Hannah Arendt, dana boyd affirms that public spaces “allow people to make sense of the social norms that regulate society, they let people learn to express themselves and learn from the reactions of others, and they let people make certain acts or expressions 'real' by having witnesses acknowledge them.” boyd argues that “mediated publics,” such as social network sites, function much like the spaces of traditional “unmediated publics,” such as parks, malls, or coffeehouses, although she does acknowledge that conversations that take place in mediated publics can be searched, surveiled, mined, and archived in new ways with algorithmic technologies. The great late twentieth-century advocates for flexible uses of public space – Jacobs, Habermas, and Oldenburg – may not necessarily have written works that reflect the existence of such regulatory mechanisms introduced by tech companies or the design questions raised by sexual and ethnic diversity, but they do speak to the importance of considering everyday scenarios of conflict and consensus, including observations about workers ordering supplies, kids playing loud music, and patrons telling stories at bars.

Although the founding of the Euromaidan Facebook page is often highlighted in the origin stories of the Kyiv demonstrations, online activity doesn’t always translate neatly into street protests, like the ones that also took place in the squares of Cairo and Istanbul, which were perhaps too eagerly attributed to mobile access to distributed digital networks and Silicon Valley liberation theology. In actuality, the relationship between transnational digital communication networks and massive civic participation is remarkably complex. As users hack, share, remix, tag, map, visualize, and curate particular political memes, they rarely behave like disinterested citizen journalists or adopt an idealized hacker ethic. The YouTube viral video “I am a Ukrainian,” which features an impassioned call by Yulia Marushevska for viewers to gather communally and collectively at the city’s square, remixes footage and images of protestors and their clashes with police and has attracted over eight million views since being posted. But many digital files tagged with Maidan-related information often aim at much smaller audiences. For example, what should be made of a male Ukrainian youth posing with weapons in a space of private occupancy that expresses a visual logic of individual ownership on Instagram? How does his participation in a particular reputation economy oriented around gun culture speak to a shared code of masculinity and instrumentalism among his followers on Instagram that might not be common to all who converge on the square and upload photos to the site with the same hashtag?

In The Exceptional & The Everyday: 144 Hours in Kyiv Lev Manovich’s Software Studies Lab curates 13,6208 Instagram images shared by 6,165 people in central Kyiv during the 2014 Ukrainian citizen uprisings. The geolocating function of Instagram indicates that the accounts sharing images represent participants who were either present on-site or very near the public demonstrations in Independence Square (Maidan Nezalezhnosti in Ukrainian) against then-president Viktor Yanukovych rather than voyeuristic slacktivists engaging in cheerleading from abroad. Nonetheless, the raw dataset is remarkably heterogeneous. Photos shot by citizen journalists that document the tumultuous scene in the square are mixed in with text, graphics, and stock images that could come from anywhere. Exterior shots of crowds, encampments, or the buildings that loomed above the protestors are interspersed with domestic selfies and other photos capturing interior locations associated with private spaces. (This public/private dichotomy has been substantively interrogated by Paul Dourish and Genevieve Bell, who have written about how the ideologies of ubiquitous computing are shaped by the cultural imaginaries around domesticity that may repress the lived contradictions of gendered space.) Even among the set of images that clearly depict those assembled on the square in Kyiv, there are shots that could just as easily be tagged #streetfashion as #euromaidan.

Revolutionary politics isn’t just about producing bodies in public places, because twenty-first century activism also requires managing procedural logics that have taken a computational turn as the protocols of organization and disruption create new sorting systems in the political order. In other words, when content can be tagged in many different ways, the metadata is increasingly becoming the message. Without expert users – such as disciplined professional activists accustomed to managing message control – who perform the informational labor of tagging content and instructing others in proper naming conventions, the same hashtag may be used for multiple purposes or the same purpose may be expressed by deploying multiple hashtags. Furthermore pro-Putin #antimaidan trolls often use the tags of their political opponents, so that #antimaidan and #euromaidan may appear in the same posting. As Gabriella Coleman notices, hacker collectives such as Anonymous may pursue “lulz” that disrupt discourse rather than promote the rhetoric of a specific political agenda.

While activists who use Facebook or Twitter often set very specific expectations for how content will be labeled, Instagram is known as a site that enforces less message control and invites more personal idiosyncrasy. Admittedly, certain highly orchestrated hashtag campaigns launched elsewhere, such as the Huffington Post‘s #whattranslookslike or the corporate-sponsored #likeagirl, appear to produce relatively coherent image sets when migrated to Instagram, but this uniformity seems to be the exception rather than the rule. Common tags such as “followme,” “instagood,” or “love” in the dataset indicate that apolitical messaging impulses may coexist with political ones. As Manovich’s group notes, it may actually be productive to attend to this “noise” in the Instagram materials from Kyiv. Much as W.H. Auden observed that Breugel depicted how the drama of intense suffering often takes place while “someone else is eating or opening a window or just walking dully along,” The Exceptional & The Everyday shows a diversity of experiences happening near the same location simultaneously from the sublime to the banal.

In contrast, the Social Media and Political Participation Lab at NYU focused on sites such as Facebook, where discussions about the norms of encoding messages may be more common, to analyze the political sentiments of Kyiv. Although both labs may trace a similar correlation between the intensity of the street conflict and the activity of social networks – which must be adjusted to account for the rhythms of work and sleep – the two groups orient attention to different platforms for exchange and thus construct different narratives about how big data reveals patterns of participation in the era of ubiquitous computing. For example the NYU lab seems to infer that more data makes the political signal easier to interpret: “While there are obviously downsides to the massive amount of information that social media creates (it can be hard, at times, to sort through everything and to assess the accuracy of the information that is shared), we have also seen that the availability of these platforms gives protestors a loud voice internationally, independent of traditional media structures and of the government, and provides a space for organization outside of the square itself.” In contrast the Software Studies lab seems to put forward a more ambivalent reading of the aggregation of information, which directs readers to sources that question grand narratives and histories of great men. As Manovich’s group concedes, Instagram had far fewer users than the behemoth VKontakte in the Ukraine and far less political activity than Facebook and Twitter. (Twitter also saw a spike in the subscriptions of new users during the demonstrations.) Yet this absence of critical mass may make Instagram more of interest as a document of visual culture, according to The Exceptional & the Everyday. “Examining images and data we collected, it appears that unlike Facebook and Twitter, Instagram was not used systematically for communication by protesters, oppositional parties or the government. Our image set is not dominated by a few power users posting disproportional numbers of images.”

It is worth noting that governments use automated data mining initiatives not only to potentially identify the faces of political protesters for prosecution, which is currently often impractical as Kelly Gates reminds us, but also to map larger trends. Using technologies for “concept detection,” intelligence agencies strive to identify themes in so-called “open source” information that might be publicly available on the Internet for predictive analytics or further investigation of future threats. These same technologies have been explored by Manovich’s group, which has tried to collaborate with machine learning and computer vision experts working with very large libraries of very diverse images or video “in the wild” and training systems to generate tags such as “explosion,” “flag,” or “crowd” automatically. The potential use of concept detection in digital humanities scholarship certainly offers a fundamentally different approach to taxonomy from conventional bibliography, although now it is largely only emulating computer generated labeling with manual tagging.

After manually appending 29 tags to 818 images from the Kyiv dataset, CUNY’s Alise Tifentale asks, “What is the visual grammar of a revolution as represented in social media? What are the common subjects, and what are the patterns of these subjects over time and space?” She finds “crowd,” “fire”/”smoke,” and “flags” among the most frequent types of subject matter, but also identifies “candles,” “flowers,” and “debris” as common signifiers of political activity in this particular site for dissent. Despite its interesting findings, Tiefentale's analysis might miss one of the important aspects of these particular “non-violent” icons of the #euromaidan revolution, how candles, flowers, and debris function as tropes of nostalgia and sentimental commemoration. Thus, these images in the dataset seem to look toward the past as well as toward the future. Even as online discourse accelerates with real time rapid-fire posting mechanisms, the photographic record of events in the square acknowledges “slow media” activities of remembrance and accretion.

Obviously this kind of media visualization project could easily be used by the oppressive forces of state authority and lead to unintended long-term negative consequences for participants. Legible word clouds of user names and assemblages of recognizable portraits of individual people certainly could be dangerous if the government of Kyiv changes again. This poses a challenge not only to Manovich’s group, but to a number of other labs working on social media visualization, including R-Shief, which is also deriving results about political conversations on Instagram. Mindful of these issues, Manovich’s team has adopted a number of safeguards. The site asserts that “none of our visualizations, graphs, or their labels shows any usernames or any additional information we could have obtained by following the links to user profiles” and that “in our visualizations Instagram images always appear as small thumbnails.” They also offer to remove any images that users ask to keep private.

As protestors in Hong Kong document their street actions this month, scholars should perhaps consider how to collaborate ethically with NGOs and question why universities and government agencies should have aligned research interests around intelligence and policing, which is reflected in many computer science papers. Human rights activists are advising Hong Kong dissidents to protect themselves against police brutality and future prosecution while also urging them to authenticate their visual evidence of mass protest and the repression of freedom of assembly and speech by turning on GPS features and capturing details of street signs and landmarks. For example, in a section with the heading “Variety of shots” on the website for Witness, those with mobile networked technologies are given the following advice: “Capture wide shots to show the size of the crowd or position of the authorities, if possible, film from a balcony, rooftop or window to get an aerial perspective. Get medium shots to show activity in the crowd. When safe, record close-up shots to show details such as, injuries, police badges, the surrounding crowd, nearby vehicles.” Perhaps in considering how to tag and visualize images it is worth acknowledging that the politics of visual rhetoric may be forensic as well as deliberative or ceremonial and that judgment about how these images may function as evidence may be made by future arbiters.


Arendt, Hannah. The Human Condition. Chicago: University of Chicago Press, 1958.

Barberá, Pablo, and Megan Metzger. “Tweeting the Revolution: Social Media Use and the #Euromaidan Protests.”
Huffington Post, February 21, 2014. http://today.yougov.com/huffingtonpostwidget/live/webpollsmall1.html.

boyd, danah. “Social Network Sites: Public, Private, or What?” Knowledge Tree 13 (2007). http://www.zephoria.org/thoughts/archives/2007/05/07/social_network-3.html.

Brown, Penelope, and Stephen C Levinson. Politeness: some universals in language usage. Cambridge: Cambridge Univ. Press, 1994.

Castells, Manuel. Networks of Outrage and Hope: Social Movements in the Internet Age. Cambridge, UK; Malden, MA: Polity, 2012.

Certeau, Michel de, and Steven Rendall. The Practice of Everyday Life. Berkeley: University of California Press, 1984.

Chun, Wendy Hui Kyong. “Where’s the Beef: Cyworld versus Faceboo.” presented at the Communications Department Colloquium, University of Amherst, April 2009.

Coleman, Gabriella. Hacker, Hoaxer, Whistleblower, Spy: The Many Faces of Anonymous, 2014.

Dourish, Paul, and Genevieve Bell. Divining a Digital Future: Mess and Mythology in Ubiquitous Computing. Cambridge, Mass.: MIT Press, 2011.

Gates, Kelly. Our Biometric Future: Facial Recognition Technology and the Culture of Surveillance. New York: New York University Press, 2011.

Gilroy, Paul. Small Acts: Thoughts on the Politics of Black Cultures. London; New York: Serpent’s Tail, 1993.

Gladwell, Malcolm. “Small Change.” The New Yorker, September 27, 2010.

Gregory, Sam. “Cameras Everywhere: Ubiquitous Video Documentation of Human Rights, New Forms of Video Advocacy, and Considerations of Safety, Security, Dignity and Consent.” Journal of Human Rights Practice 2, no. 2 (July 1, 2010): 191–207.

Gregory, Sam, and Elizabeth Losh. “Remixing Human Rights: Rethinking Civic Expression, Representation and Personal Security in Online Video.” First Monday 17, no. 8 (July 18, 2012).

Habermas, Jürgen. The structural transformation of the public sphere: an inquiry into a category of bourgeois society. Cambridge, Mass.: MIT Press, 1989.

Hariman, Robert, and John Louis Lucaites. No Caption Needed: Iconic Photographs, Public Culture, and Liberal Democracy. Chicago: University of Chicago Press, 2007.

Hoskyns, Teresa. The Empty Place: Democracy and Public Space. Routledge, 2014.

I Am a Ukrainian, 2014. http://www.youtube.com/watch?v=Hvds2AIiWLA&feature=youtube_gdata_player.

Ito, Mizuko. “Statics: Media Literacy and Social Action in a Post-Pokemon World,” 2009. http://www.itofisher.com/mito/publications/media_literacy.html.

Jacobs, Jane. The Death and Life of Great American Cities. New York: Random House, 1961.

Latour, Bruno, and Peter Weibel. Making Things Public: Atmospheres of Democracy. Cambridge, Mass.; [Karlsruhe, Germany]: MIT Press?; ZKM/Center for Art and Media in Karlsruhe, 2005.

Losh, Elizabeth, Beth Coleman, and VJ Um Amel. “Will the Revolution Be Tweeted? Mapping Complex Data Patterns from Sites of Protest.” Selected Papers of Internet Research 3, no. 0 (October 27, 2013).

Losh, Elizabeth, and Lev Manovich. “Visualizing Global News.” Lausanne, 2014. http://dharchive.org/paper/DH2014/Paper-869.xml.

Lovink, Geert. Networks without a Cause: A Critique of Social Media, 2011.

MacKinnon, Rebecca. Consent of the Networked: The World-Wide Struggle for Internet Freedom. New York: Basic Books, 2012.

Morozov, Evgeny. “The Brave New World of Slacktivism.” Foreign Policy Blogs, May 19, 2009. http://neteffect.foreignpolicy.com/posts/2009/05/19/the_brave_new_world_of_slacktivism.

———. To Save Everything, Click Here: Smart Machines, Dumb Humans, and the Myth of Tecnological Perfectionism. New York: Perseus Books, 2013.

Oldenburg, Ray. The Great Good Place: Cafés, Coffee Shops, Bookstores, Bars, Hair Salons, and Other Hangouts at the Heart of a Community. New York; [Berkeley, Calif.]: Marlowe?; Distributed by Publishers Group West, 1999.
“Publications | Cees Snoek.” Accessed October 6, 2014. http://www.ceessnoek.info/index.php/publications/conference-papers/.

“R-Shief | A Knowledge Management System.” Accessed October 6, 2014. http://r-shief.org/.

Shirky, Clay. Cognitive Surplus: Creativity and Generosity in a Connected Age. New York: Penguin Press, 2010.

Terranova, Tiziana, and Joan Donovan. “Occupy Social Networks: The Paradoxes of Corporate Social Media for Networked Social Movements.” In Unlike Us Reader: Social Media Monopolies and Their Alternatives, n.d.

“Tips for Citizens Documenting Hong Kong’s Protests.” WITNESS Blog. Accessed October 6, 2014. http://blog.witness.org/2014/10/documenting-hong-kongs-protests/.

Zuckerman, Ethan. “Beth Coleman on ‘Tweeting the Revolution’ | ... My Heart’s in Accra.” My Heart’s in Accra, October 18, 2011. http://www.ethanzuckerman.com/blog/2011/10/18/4237/.