Center for the Study of New Media and Society
Twitter and the Russian Street
Memes, Networks & Mobilization
CNMS Working Paper 2012/1
May 2012 Moscow
Samuel A. Greene
Twitter and the Russian Street: Memes, Networks & Mobilization
Samuel A. Greene, Center for the Study of New Media & Society, New Economic Schoo l, Moscow
This paper reviews the role played by online ‘social media’ in political mobilization during Russia’s turbulent 2011-2012 election season. Focusing in particular on Twitter during the March 2012 presidential elections and subsequent protests, the study constructs oppositional, pro-government and neutral ‘meme’ distribution networks, reviews the dynamic and structural characteristics and draws early conclusions about the ways in which competing mobilizational groups used Twitter during the period of contention.
Center for the Study of New Media & Society The New Economic School 47 Nakhimovsky Prospect, Suite 1918 117418 Moscow +7 495 956 9508
[email protected] http://www.newmediacenter.ru http://www.nes.ru
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivs 3.0 Unported License. To view a copy of this license, visit http://creativecommons.org/licenses/by-nchttp://creativecommons.org/licenses/by-ncnd/3.0/ or send a letter to Creative Commons, 444 Castro Street, Suite 900, Mou ntain View, California, 94041, USA.
Twitter and the Russian Street Memes, Networks & Mobilization
Introduction
What role have online ‘social media’ played in political mobilization in Russia’s turbulent election season? This question has captured the attention of political scientists, sociologists and media and communications scholars since election-related protests first broke out in December 2011. While it has been fairly clearly demonstrated that on-line social media – including the use of blog platforms such as LiveJournal, microblogging platforms such as Twitter and social-network systems such as Facebook and VKontakte – played a significant role in mobilization (and counter-mobilization), both as a means of communicat communic at ion ion and of community-building, the specific content of this role has yet to be sufficiently explored. 1 This paper contributes to that effort by beginning to unpack the role played by one social media system in particular – Twitter – during a period of peak political contestation, namely the Russian presidential election held on March 4, 2012 and the pro- and anti-government protests that ensued on the following day. Interrogating a database of more than 11,000 Twitter messages sent by more than 8,500 users during that period, the Center for the Study of New Media & Society asks how Twitter was actually used by contestants on both sides, how the s tructure of the contesting communities shaped the patterns of communication (and vice versa), and whether those structures differ across political dividing lines. A ‘first cut’ at the data is presented in this paper reaches three key conclusions: •
First, Twitter performs multiple functions simultaneously and at differing stages of mobilization, including:
1
Суворов, Глеб. «Кто же все-таки был на Болотной и на Сахарова? Анализ профилей 20 000 участников http://basilisklab.com/boloto-analis-posetitelei.html;; Greene, Samuel. “The End of митинга». Basilisklab, http://basilisklab.com/boloto-analis-posetitelei.html Virtuality”. Center for the Study of New Media & Society. http://www.newmediacenter.ru/2012/01/18/the-endof-virtuality/
1
o
o
o
•
•
As an aggregator of information, ideas and memes relevant to the mobilization effort; As a broadcast medium to spread aggregated information, ideas and memes to a broader audience; As an ‘echo chamber’ that may help reinforce group solidarity and adherence to aggregated memes;
Second , the contesting political ‘camps’ display significantly differing social structures: The opposition networks are relatively more diverse and dispersed than those of o their pro-government opponents, are well linked with media o utlets that enjoy a significant Twitter presence, and, in part as a result, dominate the Twitter landscape during the period in question; The pro-government networks are relatively more concentrated, have more o intensive communication along established channels, and place heavier emphasis on formal role-players rather than informal participants; Third , throughout the Twitter landscape during the period in question, a key role is played by professional journalists and media out lets, full-time bloggers, and established civic groups, NGOs and political organizations, while the impact of ‘informal’ tweeters or ‘accidental’ tweets appears minimal.
This research, we hope, contributes to the growing body of analysis on the role played by online social media in the recent Russian political mobilization and, perhaps, more broadly. In particular, it adds a more dynamic, event-case element to research recently published by the Berkman Center for Internet and Society at Harvard University, Mapping Russian Twitter , whi which draws on data from 2010 and 2011 to draw static, stable clusters of communities in Twitter. 2 In some ways, this report refines the Berkman findings about the fairly discrete structure of politic politi cally opposing Twitter networks in Russia and begins to address the need, identified in that report, to “enhance our understanding of the structural flow of ideas in this medium.” In some ways, this work calls into question the findings of other researchers, who have approached similar issues in Russia from different directions. In particular, our findings somewhat contradict those of Gleb Suvorov, from Basilisk Labs, who argues based on data on the structure of communities in Facebook and VKontake that the protest m protest mobilizations in March were “not protests, but flashmobs” of atomized, largely unconnected citizens 3; this paper finds a relatively high degree of organization. Second, this paper, in finding that opposition memes dominate the Twitter landscape during the period of the study, differs from recent survey work by Ted Gerber, who finds no significant correlation between political opinion and online social media usage. 4 However, none of these differences, which will be discussed in more detail towards the end of th is paper, are irreconcilable; instead, they should be seen as opportunities for further investigation.
2
Kelly, John, Vladimir Barash, Karina Alexanyan, Bruce Etling, Robert Faris, Urs Gasser and John Palfrey. Mapping Russian Twitter . Berkman Center Research Publication No. 2012-3. Cambridge, MA: Berkman Center for Internet & Society, Harvard University, 2012. http://cyber.law.harvard.edu/node/7537 3 Суворов, Глеб. «Общество анонимных революционеров». Slon.ru, 20 March 2012. http://slon.ru/russia/obshchestvo_anonimnykh_revolyutsionerov-765700.xhtml 4 Неяскин, Георгий. «Интернет меньше любит Путина и Сталина». Slon.ru, 07 March 2012. http://slon.ru/russia/polzovateli_interneta_menshe_lyubyat_putina_i_stalina-762175.xhtml
2
Methodology
For the purposes of this analysis, we collected a database of 11,113 ‘tweets’ from a total of 8,565 unique accounts, the vast majority of which (99.84%) were sent during the period from 4-6 March 2012. (Due to the nature of the search algorithm, a small number of tweets fell outside this period; the earliest was sent at 12:23 Moscow Mo scow time on 28 February 2012, and the latest was sent at 08:54 Moscow time on 6 March 2012.) 2012. ) The data collection began from the identification of popular ‘memes’ – either keywords or Twitter ‘hashtags’ – that were clearly and uniquely associated either with the March 4 presidential election or with the ensuing pro- and anti-government demonstrations; from there, we employed a simple ‘snowballing’ technique, identifying other hashtags and keywords contained in the initial data ‘grabs’ and searching tweets containing those memes, until all memes delivering significant results had been captured; in all, 14 memes were captured, including nine hashtags and five keywords. Each search resulted in a data ‘grab’ of all tweets containing the search term sent by the most recent 1,000 accounts to use the term. (Grabs were thus smaller in instances where fewer than 1,000 accounts had ever used the term.) Each data grab generated a ‘meme network’ of user accounts and tweets containing the relevant search term, including the text of the tweet itself, the user name of the originator, whether and by whom the tweet was re-distributed, and the timing of all relevant activity. Presented here, in Figure 1, are three very basic metrics on each meme network: the number of tweets, the number of tweeters, and the ‘connectivity quotient’ (simply the percent of total tweets in the network that are that are including a list the top 5 tweeters, 6 re-distributed at least once by another user). 5 (The full data, including ranked by in-degree, or the number of other users re-distributing their tweets, are presented in Appendix I.) The memes were then divided into three broad categories, based on whether their use a priori identifies a tweet as belonging to either to the opposition or pro-Kremlin discourse; those that do not carry an a priori connotation were defined as neutral. 7 The resulting ‘composite meme networks’, presented in Figure 2, vary greatly in size, with the largest by far being the network of neutral memes, followed by the network of opposition memes, and then the network of proKremlin memes. (Again, the full data are presented in Appendix II.)
5
The ‘connectivity quotient’ is a shorthand heuristic used for the purposes of this paper as a simple measure of the amount of ‘socialness’ in each meme network; as such, it reflects the portion of people in the network whose communication who communicated directly with others in the network, although it says nothing about the intensity or nature of that communication. 6 The ‘Top 5’ Tweeters are presented purely for the purpose of illustration. There is no inherent statistical significance to the number and, in fact, their relative ‘centrality’ in their respective networks varies greatly from one meme to another. 7 It should be emphasized here that, in referring to these memes as ‘neutral’, I am making no judgment whatsoever regarding the perceived and received meaning of the communication involved, or the political orientation of the producers and recipients of that communication. Rather, it is an a priori label, indicating the fact that the word or words involved do not carry an inherent political connotation that would immediately align the communicator with one political camp or another.
3
Figure 1: Individual Meme Network Statistics 1800 1600
70,00% Number of Tweets Connectivity Quotient
Number of Tweeters 60,00%
1400 50,00% 1200 1000
40,00%
800
30,00%
600 20,00% 400 10,00%
200 0
0,00%
There are certain significant limitations t o the method described above, which clearly circumscribe our ability to analyze the resulting data. The first set of limitations concerns the data-collection process itself and centers around the fact that we cannot be assured that we have collected all of the relevant tweets and accounts. This is because of the pot ential for human error in selecting the sample of memes (Twitter, at the time the data was collected, did not provide aggregate data on meme popularity in Russia), as well as the fact that the software used and Twitter’s own limitations on automated searches meant that each search was limited to no more than 1,000 user accounts. The second set of limitations concerns the nature of t he resulting data, and mainly the fact that common network metrics – including various measures of centrality – cannot be compared in a straightforward manner across networks of varying structure and size. In further analysis, some of this can be overcome through the use of more sophisticated methods than are presented here; for the sake of this paper, we have simply compared the a priori composite networks of ‘opposition’, ‘Kremlin’ and ‘neutral’ memes memes as sub-components of the whole network to give some feel for the ways in which those networks may differ. As a result of these limitations limitatio ns it is important to underscore that we do not present the reader with a complete and comprehensive picture of all relevant activity on Twitter, nor of the underlying social structure that informs communication on political issues in Russia via Twitter or any other medium. We do believe, however, that we are able to explore broad patterns and identify important questions for further research.
4
Figure 2: Composite Meme Network Statistics 12000
50,00%
45,00% 10000 40,00%
35,00% 8000
6000
Vertices (Tweeters) Edges (Tweets)
30,00%
Connectivity Quotient
25,00%
20,00% 4000 15,00%
10,00% 2000 5,00%
0
0,00% Opposition Memes
Kremlin Memes
Neutral Memes
All Memes
The Data
As per the methodology described above, in the beginning was the meme, and the meme was a word. In generating the database, we began with Twitter ‘hashtags’ – keywords o r phrases preceded by a #, which allow Twitter users to follow a particular topic of interest – that could clearly be associated with the events in question, namely the Mar Ma rch 4 elections and t he competing rallies on March 5. Thus, for the elections, these were # выборы 8 and #выборы2012 9, as well as #4марта10. In addition, to allow for the fact that many many relev relevan antt messages messages may not inclu include de hashtags, we performed keyword searches on the terms путин and вброс 14. For the утин 11, уик 12, карусель13, and rallies, the relevant search terms were #5марта15, #митинг 16, #пушкинс пушкинская кая17, #мане манежка жка and 8
elections Elections2012 10 4march 11 putin 12 UIK, the Russian acronym for polling station 13 Carousel, a term used to denote the organized bussing of fraudulent voters to multiple polling stations 14 Ballot-box stuffing 15 5march 9
5
#манежная 18. Finally, two memes cut across both the electio ns and the rall rallies: ies: #стабильность19, associated (ironically and satirically) with the opposition, and # запутина 20, associated with the pro-government camp, as well as путин, mentioned earlier, which is neut ral. ral. Structure & Connectivity
The resulting meme networks and composite networks differ, of course, in size, but are bounded at the upper end by software limitations, which allow the collection of network data on no more than 1,000 users per search term. Most of the individual meme networks thus cluster at or near this size, although some are considerably smaller. This size differential carries over to the composite networks as well, due both to the size of the underlying meme networks and the number of memes associated with each composite meme group. Meme networks – including both individual i ndividual meme networks and composite meme networks – have a standard structure, with a core of more or o r less densely connected vertices (representing individual Twitter accounts) linked together by Tweets that reference one another, surrounded by a cloud of unconnected nodes who, at least in t he dataset as collected, have no connections with any other users in the set. This structure st ructure does not differ from one meme or meme group to another. These networks do differ, however, in the relative size of t he connected core versus that of the disconnected orbit. In some networks, this connected core is quite large, making up as much as 65.8% of total nodes in the case of o f #запутина , while in others it is extremely small, making up 14.8% in the case of #выборы or even 0.01% in the case of #манежка. This figure, which I refer to as the ‘connection quotient’, is 41.4% for the entire database, with opposition memes coming in at a higher figure (47.1%) and pro-government memes at a lower figure (33.4%), and neutral memes very close to the aggregate (40.7%). The relative meaning of these differences will be explored further in this paper. Suffice it for now to note, however, that the connectivity quotient itself says nothing about the nature of those connections: their diversity and intensity are also important to understanding the impact that such connections have on social relations, mobilization and other outcomes. Tweeting the Election
Tweets about the election – which, by and large, were sent on election day itself, or early on the following day – divide broadly into two categories: those that use a priori neutral memes, and those that use a priori oppositional memes. There are very few election-related tweets th at employ clearly pro-government memes, memes, which in and of itself is an important finding. That is not to say, however, that pro-government sources, including mass media sources, were not tweeting about the elections. Three of the most popular election-related memes – #4 марта, #выборы and #выборы 2012 – were dominated by Kremlin-linked media outlets, including Interfax in the former two cases and the RIA group (Russia Today TV and RIANovosti) in the latter o ne. In most of these cases, however, the connectivity quotient is considerably lower than for the database as a whole, and the bulk of the re-distribution of memes that does occur emanates from a relatively small number of large-scale sources (with television personality Tina Kandelaki joining the major, state-backed mainstream media organizations named above).
16
rally Pushkinskaya, the name of the square where the opposition rally was held 18 Manezhka and Manezhnaya, interchangeable names of the square where the pro-government rally was held 19 Stability 20 For Putin 17
6
Opposition-minded tweeters, including both formal and informal proponents of the opposition and the anti-government protest movement that emerged in December 2011, dominated Twitter communication that could be characterized c haracterized as more targeted and instrumental. This includes most prominently tweets involving the keyword уик, the Russian acronym for polling station, the vast majority of which involved allegations of fraud at one station or another, and many of which included links to videos or pictures providing evidence for those allegations. Similarly popular were tweets involving the work карусель, denoting the bussing of repeat voters from station to s tation. In both cases, top originators included prominent bloggers’ accounts, such as @adagamov, @mvelmakin, @solomonhaykin and @drandin, alongside election monitoring initiatives including Liga Izbirateley and Rosvybory, plus on-line media sites sympathetic to the opposition, including TV Rain. Here, levels of connectivity were also below those of the database as a whole, but higher than those of the broader election-related memes mentioned above. Tweeting the Rallies
Almost as soon as the election results were announced, attention turned to the rallies planned for March 5. The opposition was scheduled to gather on Pushkinskaya Push kinskaya Square, while the government organized a counter-demonstration on Manezhnaya Square. Communication on Twitter reflects this shift, as use of election-related memes drops drops off, and use of others o thers builds. Here, too, the data can be sliced in various ways. As above, the memes can be divided into those that are a priori neutral (such as #5 марта or #митинг), and those that are more clearly associated with one camp or the other (such as #пушкинская on one side or # манежная on the other). As in the case of the elections, the differences along this dividing line are stark. In particular, the role of professional journalism outlets shifts. Outlets such as Glavportal, Vedomosti, The New Times, Russky Reporter, Public Post and others are dominant both in the neutral meme networks and the opposition meme networks, but are almost absent on the pro-government side. Pro-government meme networks, in turn, are dominated during the rallies by those linked directly to the regime: United Russia’s official Twitter feed, plus United Russia parliamentarians Vladimir Burmatoff and full-time Kremlin-sympathetic bloggers, including @aragorn28, @eugenechigir and others. A different way of slicing the data, however, divides memes into those that are generic informative memes – i.e., hashtags that simply mark the tweet as having something to do with the rally – and those that carry a mobilizational message. In particular, the opposition coalesced on March 5 around #стабильность, which translates to ‘stability’ and was used to satirize the regime’s political message: typical tweets would refer to actions of the riot police, broken bones and other unpleasantries as indications of how the Kremlin, evidently, defined ‘stability’. On the other side, the mobilizational meme was #запутина , translating simply to ‘For Putin’. Both of these memes display a much higher level of connectivity (63.6% for # стабильность, 65.8% for #запутина ) than the more generic memes or the database as a whole. Both were also driven by the co ncentrated efforts of a few formal fo rmal players: in the case of the opposition, the o nline news source Lenta.ru and the blogger @zhgun, and United Russia and Burmatoff in the case of the regime.
Analysis Taking a step up in aggregation from the individual meme networks to the composite networks of the three meme groupings – oppositional, pro-government and neutral – allows us, hopefully, to 7
identify larger-scale patterns and underlying structures. As mentioned before, all three composite networks bear the same underlying structure, with a connected core and a disconnected free orbit. Reviewing the three networks in turn helps uncover some of the differences. The Opposition Meme Network
The opposition meme network, as a subset of the general database, includes 3,366 tweets sent by 3,037 individual accounts and has a connection quotient of 47.1%, significantly above the average for the sample as a whole. Its connected core is remarkably large and complex, dominated by @lentaruofficial (the Twitter feed of the popular news website Lenta.ru), but with strong roles played by other media outlets (The New times and TV Rain), oppositional civic groups (Rosvybory, Liga Izbirateley and others), and independent bloggers (@mvelmakin, @zhgun, @solomonhaykin, @drandin and others). Its complex and chaotic core (see Figure 1) includes i ncludes numerous interlinkages and bridges, and the largest connected component includes includes 932 nodes, or 30.7% of all of the users in the network. Figure 3: Opposition Meme Network
8
Notably, Figure 3 also illustrates that most of the edges – which represents tweets by one user mentioning another – have low weights (represented by the thickness of the edge). 21 This indicates a diversity of sources and communication, as each individual seems to consume and an d re-distribute information from a multitude of different sources, and most of the ‘top tweeters’ are connected through long pathways with most of the o thers. An important exception is the Twitter feed of the news website Glavportal, whose interlocutors do not communicate with the rest of the network; indeed, filtering out all nodes with one or fewer edges removes the rather large Glavportal cluster from the network entirely. (The full network maps are presented in the Appendices to this paper.) The Kremlin Meme Network
By contrast, the Kremlin meme network, apart from being smaller (1,876 tweets by 1,536 accounts), is also considerably less dense and connected (33.4% connection quotient). It is heavily dominated by State Duma Deputy Vladimir Burmatoff (@burmatoff), who in turn has key relationships with a small number of o f re-distributors (@monomakh34, @mantsur, @n28vova and others); Duma Deputy Robert Shlegel and the official United Russia Twitter feed also play notable but demonstrably secondary roles as meme distributors (see Figure 4). Figure 4: Kremlin Meme Network
21
Edge thickness is a standard metric in social networks, reflecting the relative intensity of communication between each pair of nodes.
9
Indicatively, the edges in the core of the Kremlin network are considerably ‘heavier’ (thicker) than in the opposition network, indicating more formalized and/or routinized channels of meme distribution and a higher degree of centralization, consistent with the use of Twitter as a broadcast and coordination mechanism, rather than as a forum with a strong social interaction function. The network’s largest connected component, meanwhile, is smaller than that of the opposition o pposition network, including 324 nodes, or 21.1% of the network’s members. The Neutral Meme Network
Finally, the ‘neutral’ meme network is by far the largest, with 6,896 t weets by 5,188 accounts, with a connection component of 40.7% (see Figure 5). It is i n many ways similar to the opposition meme network, with a complex core of numerous interconnected components (with the same caveat about the strangely unconnected Glavportal component). Given what we have seen earlier regarding the opposition dominance of a priori neutral memes, this is perhaps unsurprising. But the network also includes some meme distributors – such as Interfax and Tina Kandelaki – that have large clusters but are poorly connected with the rest of the network. The largest connected component nonetheless includes 27.7% of nodes, or 1,436 accounts. Figure 5: Neutral Meme Network
Interestingly, it is within this neutral network that some os tensibly influential opposition online actors – such as Alexey Navalny and Liga Izbirateley – show up, virtually fo r the first time, although with relatively small clusters. In the case of Navalny, this appears to be a result of the fact that th at he does not generally use hashtags and thus entered the database primarily through key- word searchers, which in turn are concentrated in the neutral meme category; it may also be a result of the fact that, on March 5, his level of involvement with the rally prevented him from being an active 10
tweeter. As for Liga Izbirateley (the League of Voters, an independent election monitoring group set up after the December parliamentary parliamentary vote), their relative (though not complete) marginality in the network may reflect the importance of organizational infrastructure and established networks for the effective use of Twitter as a mass communication tool. The fact that oppositional content dominates the ‘neutral’ meme network is in itself and important finding and serves to highlight something that could easily be misunderstood. The ‘neutrality’ of the memes, as mentioned before, does not refer to the intentionality of the communication or communicators involved; indeed, the content of the ‘neutral’ meme network is anything but neutral. But it is important to remember that Twitter is a mass-communication system, with elements of broadcast media, meaning that more people are likely to see any given Tweet than are likely to produce one. Thus, ‘neutrality’ in this case is defined from the perspective of such a passive reader: if the reader is oppositionally minded and interested in the events of March 4-6, (s)he may have searched for all Tweets containing, say, # карусель, a hashtag clearly associated with the opposition; conversely, if (s)he was well disposed to the Kremlin, (s)he may have searched for all Tweets containing # запутина . But if (s)he was undecided u ndecided or simply wanted to know what was happening, (s)he may have searched for # выборы, a hashtag linked to the elections that carries no group attachment in and of itself. It is thus all the more significant to note that the bulk of what that neutral or curious Twitter reader would have encountered was oppositional. Dueling Memes
The contrast between the opposition o pposition and pro-government meme networks comes into sharper focus in Figure 6. For the t he purposes of this comparison, a composite network was drawn of the opposition and Kremlin memes, and all nodes with fewer than two connections to other nodes were filtered out of the map for fo r the sake of clarity and visibility. Figure 6: Dueling Meme Networks
11
On the face of it, two clear camps emerge, with very different underlying structures. The opposition is ‘led’ by a relatively loose coalition of bloggers (@mvelmakin, @navalny, @vorewig, @drandin, @solomonhaykin, @zhgun), media outlets (The New Times, TV Rain, Russian Reporter) and civic groups (Rosvybory). There are numerous long pathways and cross-linkages (note, in particular, the degree to which the networks of @rosvybory and @mvelmakin duplicate each other), while the weight of each edge is low. The other si de is dominated by a single tweeter, @burmatoff, whose network is close-knit and coherent but not diversified and is characterized by ‘heavy’ or ‘thick’ edges, representing repeated communication between the same individual accounts. Between the two camps would appear to be the news website Lenta.Ru (@lentaruofficial), acting as a bridge and mediator. On closer inspection, however, this does not turn out to be the case. Rather, Lenta.Ru, as we have seen earlier, was a key distributor of o pposition memes, particularly the important mobilizational meme #стабильность, while the real bridges between the two camps are several nodes located between @lentaruofficial and @burmatoff on the map. The resulting picture, then, is one of a diversified opposition with complex and fluid networks squaring off against a more cohesive, concentrated and hierarchical pro-government network.
Discussion & Conclusions
The role that online social media such as Twitter can and do play i n mobilization and political contention is itself highly contested and, in many ways, susceptible to the minutiae of research design and methodological predilection. Almost all of the fundamental aspects of each study – the geographic, technological and temporal parameters of what is being studied and how data are collected and manipulated, to say nothing of the specifics of the questions ask – often differ from one investigation to the next to such su ch a degree that the results are difficult to compare. Generalizing results from one study to a broader context should be attempted only with the greatest of caution. This paper represents a first, tentative exploration of data on the use of Twitter during two and a half contentious days of Russian politics i n March. As such, it bears no direct relevance to the use of Twitter in other contexts, whether topical, geographic or temporal, or to the use of other online social media platforms. Due to the nature of the data and the process by which they were collected, we must also be careful in generalizing our conclusions even to the level of the broader social groups whose interactions we have attempted to observe here. That said, some broad patterns appear evident, among which are the following: •
The use of Twitter as a social media tool varies, even within a brief period, greatly, from one group to another and one situation to another . In some cases, particularly involving the use of mobilizational memes during the ‘rally’ phase of the March 4-6 period, Twitter clearly acted as a broadcast medium, in line with much of what h as been written about Twitter in the academic literature. However, the use of Twitter took on a much more dispersed and interactive character during the election phase, when the emphasis was on collecting and aggregating information on what was happening during the voting process, including evidence of fraud. Moreover, the transition, particularly within opposition networks, from more dispersed to more concentrated communication over time suggests a process by 12
•
•
which information is collected, co llected, aggregated, aggregated, processed and then re-broadcast for mobilizational purposes. It is also notewo note worthy rthy that the emphasis within the pro-government networks is more heavily on broadcast. 22 There are marked differences in the underlying social structures of the pro- and anti government online mobilization networks. The latter are demonstrably more diverse and more fluid – as represented by the multiplicity of weaker ties – while the former display a greater degree of hierarchy, concentration and institutionalization. This is consistent with the Berkman Center’s research on Twitter, although it is somewhat at odds with their research on the structure of blog networks. More fundamentally, however, it is consistent with broader sociological research on social movements and an d networks, most notably with Granovetter’s classic work on “The Strength of Weak Ties”. 23 Throughout all the networks explored here, professionals – whether professional journalists, politicians, bloggers or activists – play far far and away the most important role in the production and re-distribution of communication via Twitter . Again, this is consistent with much of the existing literature on Twitter in other contexts. But it also underscores both the importance of infrastructure and resources even in i n using ostensibly costless and ‘democratizing’ online resources, as well as the fact that the Russian opposition does apparently enjoy the support of a robust professional media network.
Despite all of the qualifications and caveats made above, regarding the ability to compare these findings to those of others, a close reading of these findings and those presented by colleagues studying the same general series of events raises important questions and contradictions. This most clearly arises regarding the work on Facebook and VKontakte done by Gleb Suvorov and colleagues at Basilisk Labs, who find that, “The protest movement has no structure or hierarchy. There are ‘leaders’ and masses, and between them only a yawning gap. … Each oppositional individual is in that gap. They are individualists, who do not need likeminded thinkers. Out of 20,000 participants, 8,000 have no friends who were on Bo lotnaya. … Each oppositional individual committed a solitary act of expression of dissatisfaction by going out into the streets. Alone. And he came back alone.” Leaving aside the rather glaring conceptual problem of equating on-line and off-line protest populations, this conclusion would seem to reflect a fundamental misunderstanding of the nature of social networks in general and protest networks in particular. The finding of at least 8,000 connected nodes within a universe of 20,000 individuals – or, i n the terms used in this paper, a connectivity quotient of 40%, more or less in li ne with the Twitter networks explored here – is actually a remarkably high degree of c onnectivity, and significantly higher than we wo uld expect to find in a randomized social network of the same size. Mo reover, at what is still an early stage in a protest movement with an uncertain future, it is unreasonable to expect a higher degree of solidarity (even if we make the leap of equating connectivity with solidarity) than we currently see. A greater challenge is posed by the work of Ted Gerber, in his study o f “electronic political engagement” in Russia, conducted through traditional polling techniques. Gerber’s broad findings – particularly of an increase new media use for political communication after the December 2011 Duma elections – are not in contradiction to those presented here. However, his finding that on-line presence, both in terms of online social networking and Internet-based “information seeking”, is
22
This diversity of Twitter usage is in line with findings in the comparative literature on Twitter communication; cf. Dann, Stephen. “Twitter content classification,” First Monday , 2010, 15(12). 23 Granovetter, Mark S. “The Strength of Weak Ties,” American Journal of Sociology , 1973, 78(6): 1360-1380.
13
“generally not associated with political views” would seem to be at odds with our f inding that that oppositional communication dominates the Twitter space during the period of contestation. 24 It should be remembered, of course, that we are working with very different data sets. Gerber polls a representative sample of 2,403 Russians before and after the elections, in 46 regions, while this report limits itself to those who actually communicated about politics on Twitter during the March 4-6 period – clearly a self-selecting sample. But this still does not resolve the contradiction: if t here is no clear tendency for “on-line” Russians to be oppositional, then why do they domi nate in our networks? The answer likely comes down to reporting, particularly self-reporting by survey respondents, and underscores the need for multi-method research. The data we present here obviously do not allow us to draw conclusions about the place of Twitter communication in Russia as a whole, and we can surmise that Twitter users are not representative of Russia at large. Unfortunately, Gerber’s poll did not ask a question specifically about Twitter usage, which would have provided some reference point; absent that, we are left with data from Yandex, which reports that there are some 3.03 million Twitter users in Russia, a significant number, to be sure, but one about which we know almost nothing in detail. 25 Polling data such as Gerber’s meanwhile, do not allow us to draw any conclusions about what p what people are actually doing on line. Until we make that connection and have much larger and more complex datasets, we need to be extremely careful in arguing cause and effect. Indeed, this paper makes no causal claims whatsoever: while we believe we have some idea of how and what people were communicating on Twitter March 4-6, and we can draw tentative conclusions as to why, we cannot extrapolate from that into an argument about whether Twitter made the protests larger, for example. And, in fact, studies of online social media and mass mobilization in other times and places suggest that such causal links are far from clear-cut. In the UK, for example, the “Reading the Riots” research team found that peaks in riot-related Twitter activity followed, rather than preceded, activity in the streets – perhaps not surprising, given that much communication tends to be reactive, but complicating any argument that Twitter was instrumental in bringing people out into the streets. 26 Moreover, as Navid Hassanpour demonstrates, Hosni Mubarak’s abrupt removal of on-line media from the mobilizational equation in Egypt – accomplis accompli shed by simply turning off access to the Internet – actually led to more mobilization, not less. 27 Thus, whatever answers are suggested by this paper, it points to even more questions for further investigation. In many respects, we echo the Berkman Center’s call for the systematic comparison of online communication across platforms, including blogs and networking sites; to this we would add multimedia sharing sites and the forums of established online news sites. To that we would add the need to combine on-line research with rigorous off-line approaches, including both large- and small-N studies. Within the framework of our own existing and growing database, we recognize the importance of content analysis, including i ncluding semantic and sentiment analysis, as well as link analysis. If we are truly to understand how and why social media are used in mobilization, we must combine this research with more sophisticated (and better theorized) qualitative and quantitative social research. 24
Неяскин, Георгий. «Интернет меньше любит Путина и Сталина». Slon.ru, 07 March 2012. http://slon.ru/russia/polzovateli_interneta_menshe_lyubyat_putina_i_stalina-762175.xhtml 25 http://blogs.yandex.ru/ 26 http://www.guardian.co.uk/uk/interactive/2011/aug/24/riots-twitter-traffic-interactive 27 Hassanpour, Navid. “Media Disruption Exacerbates Revolutionary Unrest: Evidence from Mubarak’s Natural Experiment,” presented at the Annual Meeting of the American Political Science Associati on, 2011.
14
Appendix I: Individual I ndividual Meme Networks Meme
#выборы
1456
Number of Tweeters Neutral Memes 857
#выборы2012
1349
775
27.72%
Путин
1114
935
16.89%
Уик
1492
1000
35.52%
#4марта
1513
895
41.18%
#5марта
1445
1000
51.63%
#митинг
1319
1002
52.69%
1205
Opposition Memes 1000
31.29%
160
119
33.75%
#пушкинская
1163
1000
40.41%
#стабильность
1457
1000
63.62%
#запутина
1589
Kremlin Memes 433
65.76%
#манежка
1459
1000
0.01%
Карусель
Вброс
Number of Tweets
Connectivity Quotient
14.84%
Top 5 Tweeters
@interfax_news @tokyohatsu @vmysila @zalezaka @volya_naroda @rt_russian @riapress @47shots @alexeyhm @golosameriki @dlindele @navalny @insider_chat @inosmi @userdie @adagamov @ligaizbirateley @osha_kot @kozlovsky @olevskiy @interfax_news @tina_kandelaki @zachistievybory @you_reporter @alyasha_alyasha @glavportal @vissevald @zghu2010 @vedomostilive @vitaswift @glavportal @ladyguri @narod_protiv_er @vitaswift @kurginyanru @rosvybory @mvelmakin @solomonhaykin @drandin @tvrain @letmefashion @letmefashion @zapyatoy @viktor_bad @yrasunrise @bajiek_ @glavportal @the_newtimes @vova_moskva @russianreporter @public__post @lentaruofficial @zhgun @vorewig @fleurklin @fontanka_spb @burmatoff @er_novosti @gattarov @mantsur @sadreeff @c1rcool
#манежная
187
134
36.90%
@leisha_xoroshij @ftzwxz @vitaly_russia @pknsht @aragorn28 @eufaraday @c1rcool @eugenechigir @d00h
Appendix II: Composite Meme Networks
Vertices (Tweeters) Edges (Tweets) Self Loops Connection Quotient Top 10 Tweeters
Opposition Memes 3037 3366 1780 47.12% @lentaruofficial @glavportal @rosvybory @mvelmakin @the_newtimes @zhgun @solomonhaykin @drandin @tvrain @vova_moskva
Kremlin Memes
1536 1876 1250 33.37% @burmatoff @er_novosti @mantsur @gattarov @sadreeff @aragorn28 @monomakh34 @shlegel @kirillschitov @n28vova
Neutral Memes
5188 6896 4091 40.68% @glavportal @ladyguri @interfax_news @vissevald @tina_kandelaki @adagamov @ligaizbirateley @dlindele @zachiestievybori @navalny
All Memes
8565 11113 6516 41.37% @lentaruofficial @glavportal @ladyguri @interfax_news @vissevald @burmatoff @tina_kandelaki @rosvybory @adagamov @mvelmakin
Appendix IV: Kremlin Meme Network
natagkabaranova
krotizs
sasphopmitfo jer0nym0 zshesht sundupova93 palamer2344kar
macquectuma veorshckov psjihoidus
dimbfuddy
sandcmharsunslo
tyawinlakta kosidonbo
zycetarpopib px7r
anti_vodklagranad_in_k nikaklybnikia
mhaxim42rus tokninaty
shibaevanatalia yzifunanaxenovaelena
pbepper_inc
k_a_b_ag_n
tnksvq
jakmsz dima_mgetalhead
paclegitir1988
kuafedelmoon
mars_lovje_you armareetto skandal_isgt ahsitsil silf_keblvar divkij_zveryuga fucvkthebass wpolfgang777 ponyslicue znitro143 aburafyj quekabobesn anekdjotskij slavjtan4eg pealognoco udav4ik197y9 apaahysj ilya_ternovoy life_msk schofinoatndir sluzrnom hey_michakel nppteuc nc_tv daf_mlasta kugplu efremakinchejv nik_bubudkin si_8p881
yplvfo
grinechka fflaresky
leviu_hou
efjcdr
cchi4ikuku
asflidisti
fhs_ruliot
fayt_2006
mr_kaerdxolder jepnonomn
mammiezn001gu vkaningtemva
smsulow
zvgsc
edpdi_gribl
chtitadriita skripnikaa
cizebsoak
nata_guseva
dmitrij_majurov
rhejaflow
lufandever esamweb
puunkpost
sachagslazkov ivan24_02
ptalgat_1
pribyshin sxp274 jewraldcopq seetecusubperkasper
salty_1970
ifsanych1977
dmitrij_frei
infodestinyru persoxnal_dan
miqgynovn23
fr_andrey
atlajntis83
bihanskyy
nepogostu
dgordgina ialexander13 elena_yakushina
svetaboiko ohoqoqo basegprm nnevgen
orevkov
fer0ce
dolbokril svirdikova
belovtimotej
igoryash
drmax_nn
imakhaev
natalyabutba olegmalcev sashamalygina
jonny_ozezab0t ex50dar
ojanyguqo
semadagaev
fadeichevpavel
gtlkef zheka_anashskin bonsaisa8z99a
alexander_rybin
casuaul_ryanskkilgishovan
sergeyignatev esh_brevnouv
bogdanlinch
alexkei1
wordovholics
jilohalym fucktbi
hatipov
once_in_july
mur_ov
mikhashenok pargina nucleusx7
larilcea
fuwaitegi
luberr3791
samangerlm
arshak_m
mfedpeg
ljfoska
rjabehnkosasha
daveryanov vhwvuy
kykyllionok
sablosontecrazznzy_mag nesabarwchema casyssffh
phvwwjberido
pavel_lukashpov jyajoloom
kozzadereza
armmragedonn
tmiakot
tupitnukibed
janeairp23 ya_iz_lmoskvy
nurdik
skrupskaya
vasya2kurolesov
artem_tiennco
19nick89 vadimvma jetzajac
ophthalmalbus _uylechka_
freeon_r1t2
t30p_i genesuisis
cmihail27
t310m9
ruqosidgnmeteorra_x_x
klimkaxrptzev
sergeypush34
fozeil_jon
spiteful_suslik ayaz_01_
t30p
srel_nik
kotoviv primakovd natalies19 _9737832830961 ostup_bender greek777 efremovsergey raevskaja
qgfyz
drweb_8u7 bpatyanya_55
citizen_veterr cnterlectdriv
zh77250 sibirsknijj
tasamoungle
minervvaxyygu
usurunif
zhekan8781
miannyxyyke nfucevyga bespe4nyjaingel
a3imbayev ivushkinn
rukyeaura
philipkravtsov
yav_sverxu kolhyusya
enotsava
meldauvvhgo andrezw_samick
shevadronova
sjksirwin
mstiteli_lji_ru
fuerfentchala eccewko
julee_voronkova
simon_dkumbo ysasonch4k
morsnews_ru
stervina
larpyba tonik1p4
grnqraz moskvindp
uwajywilo bfzsyv
dajlajtt
xaskvk
di_tuneless
novoseltsevv_m alex_seeattle
ur_insomnia ryzhkovak
flaish2883 projekt_cargos tesvter2011 vizi7 miszter_goodman
mmyrockmy sqlm3b4
rufust1960
angmriz
rvoytsex
cavejackall
chao_ous
xoroshrij_macho
zasrakta
roaat
bunusullk
orlondy76
nicocebakchilor aleksanya hctzzx
mioxasyanych akliks_first
valusha_tronova
zzmhkm xxxfasterrxxxx juloek7777 gzsmnj
mitmtz
pissaro_150p0
inoori_kito
basgdadskijvor
huptura
bzorisrizik
ferdor_ageev sk_ah_zi
iliua_oblomov yas_max14 fwalejuo
dyarenbnasu suffer_tlife
pmakarychev
sanya_z1988 kotv_mudrecc lehaexorn
hstepits
krupicdy
abakanskiy
nmiemevinmu qzeyodohakegirorun
azatx88888 szmallblackfox
nrxjtpoie cyamotasasm vgehetout pasha_leoktor
dvamas
ralternatus nexptunene
za_3_sqrok edeinica_1a
eihshjpnr
xgwosryyo alekscand_kr snakizatorrnr
fdrfry dyurij_gordeev m_ikpsanov prestonmlcc
ezywyy fqbxkcrxj
ghostdarknress alyoxao_72 r_snol
donnybimialkelex wycpkfjau
caarlbdfh
muajdan_seryj
laryfcz
makned0nski
serdfyulka daggets
ikmzav
u_bletjka kolya_fommin
himik_k3
desqswasey9bbc jfurli
bordeer13 sjasha_tokugava
evgelneelne vladchewd
jsyn_yastreba
jdhrais molot_sfeksa
liocsepudis
djinchik
hofibaqyge uflokodvf
cuissigbora
xxx_stringk_xxx
dead_opcerator
dtortugalec
malyshx_dzhon sergeybatiyr agolabap
prosto_dzhejbms cocc17 stjyleqw
lexarrshem
afoks_batler
uaohdoyfl heguzyhe
aoruetkni
impipourra
lashawntuwucz antonova67lqvbbu
mikkooykko
kfufjs djromeompaper cruscqhohemisc
cybapiliypet zzhrfw demonpsm
jsohoganri
overrhot mapzqsfulmer n_katin
ragionevolce
dmitrigj_09
offnendr
fpostcamoci1974 vigtalchik88 goutog
nywerestn
senhator_tmy vyborov_andreuj anonim_mmgo
alepxandr_pm113
rvobert_sun
stzimul77
tcaiba yozufiqst
betfnf
chyzek
pahan88p888 parabolation
xvqqsx
nepokoryaoemyj
zadoorcongsoi ugituameo hiomnz kijkvon
krqrrdufber
myndanpamgpher sgenbhydhueda
eaasmik bigboiss01
lfibiclew
esquzw
pikonsuda
fono2v0 rycar_maqstdaj
pubijca_xa smartxejrox
sicawfilne
ciufudaub
vomvax
stryjd_roker
nuipbusrini epi_ameat
gaisnfelcepma
art_d00t7 carexnigra
umoxrrist ihestgze
xtinaooh opzhaghye
nidiolie
ivahn_fakov
amskvn
lecubeantda dj_gavriok suinaseje
raozen1313 vorcavise
nsn_info
pulsj12
learntabdroker
konyushko
rikof1963 forevelbitlz
dvorik_eco
ne_ozloj_pk slafm_slamik
ddbdden
vderagrid gjlgjkrjdypbr
sanitskaya
fgreysun_56
cogntravober zantella servifse_10 sherlok_xolmns
deslodqer udkeqji
_dasha_ch etfboh
spybpatlpili
fail_lirfe
xlgbpeifi dbimka275yj
droftexr
rftrtew qcjsva
kbhrfzipn xsgmpu
ryqaqor nnybarg18 qbcnow alekszijk claspueceac
palayerix leonid_shevelev servsmounhena adrian_sticqi
twlixp
balotepkupyd evodolu nibkesan
rohecksurfvofl
shakdiman ihthqs
nutiplsug
demiuwrg222
churhkiwoin
bkeast321
onikcs_4o
oleg_matveeonko
xxlegalaxsxx densezx
astfvt
sunvita darismirnoff sabfv
jneodissidcnfor
husyzipoba
z0n4v1k
paavel_prometov
vesel4ak_y
alonen73
itasaarr
nlortonuke
life_i4_sk8
serzh_sledeopyt hcfdbhfxk
ramid_n72
aldenb5678qui tranereunun
ofktyd
aego_istko
serzh_tguru pudsapith korsapr7 tolr_iz_moskvy
gqlkkj
al_larin
poman199h0 asato_tcuzluki ueirug
xzjuhelx
negarl1
lstaunl
kleminayulia svhaldi uwityzico
vsitya06111990
mrsencefull
assk1ckesr
werliruos
flexxxlik yakushevaanna mini_grapefruit
apnna_moza
nimfomanypak
talibanisbhhe v36rus
kartnyn
oxhsupreeeeeee maroussia1985
yqzmts loargo_s
ccabamilow
rolbene
gfszmw
wntzks
itte_irasnhai
ciof_live
valynovk
prazdnikvkudsa
gagasos59
skrnadol
aragorn28
krisboldyreva
ozqprin parcnyaga_sega mahmudekrk krediftovanie
andlgeo frikazoljg
jonpaspid
ypodiduet
mazhoorishhe
garqi00
kcostik_83 maximusgmeo raskazchigk orespekt2010 kajchukabra sergej_kregtov yurii66i6999 edwardq_ro
demasfrost novikova_m
jspyzj
getdiota
olen7u7
akademlogik
merimorozova
raisitns_cake servise_0x8 hasilamizdt
puzikazuzika
dancer_min_fire lexa_sypecnaz fuckiyng_emo69 rimilxaos roman_chyuko lipskerov
tvrain
n0tv_so_f45t fantompmog
xeresavak
stefanvi_lave bryanskru
gonhgarmak
tfalkvarl
izbranha
sayko195p8
sergey_delnet
elvigubushiev alterego_ist
saboteur_2 jurgens710 skorohodik demonout tikhomirov_vl comandanteterzi strong1494 kipi_shin
nata_fox_
vvkusnovsem
mertvetsj_vkg
danik81slim shabanov_s_v
lunatik13
djb0b proudlyfly
afir_xofqf
luckysshzark ivhanovalex50
vcrqhh it_snowqs ketstantcraider btreatnot
nybavun
leisha_xoroshij
svistyshka
aslan_bisenrov
alnatoli70
stranger_kkuin
kanrinacadefg rvhpzqfxf
kaelevofcrt
sgerasimov
markoffka11
my_sadm_cry
vocv4ik007
diybydil cymorafoat
gorebihu ma_grunlge cdawdcicomar
japanecse_wrx
sexy_rbeal_slow imagadan_ru
msaryev
kwrtzp
vujaqyqer
perhecodown
akacmorpl
hjohifold
canheatwtherna
victorrusa onerkhymer
yankicolon ganbik
aleksraduga
ftzwxz
jkpyrpczc
shergei_34
moimabeat nimmjermal
pknsht d00h
maxtaksi
jashaten storm_axe
shurxshik_80 ottjo_life
dsmllh
tplisina cligvpasfore
dostoverkin
lakomkina topbot
t30p_0
ubushdevil
dripbdingdisro
iqokiboso
denis_leoqnchuk public__post
t30p_3
sloqn_not_sad vypagaju
mruzkzzz
ramsessanchez
tanyadm11 artembeloglazov
voffafa
forecasteir88 ehucodydiq
mrt_davo
samthebgest shxipovich
aquajcoli
ecrbriticin
russianews24
korobkov
hdo20
omhtpr fbreakru
34k5543
eugenechigir
xx_fhishkin_xx ghoniq
tosyriterqilyc
c_adrian
tagan77
_julia_v notearesout
matvey_sjpb summerssii d_rjynfxrnt abramowadasca
cpzjgm
hodjaevb
tatarinov
evshu
mmanlregd
drryazgov
nadysynablif chusyand
hellgqamadness
tramzekagid
funnycrime
ivanenko_i
gdelkh
kira0509
littlesivanov
pokyrejuk
ray_kjiri
vivintic
mishenkangis
evbgenii_lee ixeverlastingi
alyre32
kopylovrussia dmitrykryukov petrant0 alexandra_mtv general0317 evgeniy__s
er_insider
marfeukiyrsay mvereschagin aliantemarifno
zberg75 ualanaz001puzon
denisbeta
elaterus
goncharov_66 cryofrostzero
yudinkirill zloj_grishja
zloymopga
last7ru
olees_smaga
ciel_murdher
yanabogdanova
oksana_25 zlatafishka irishkapavlovna
nikolayvoronkov
freerhejrzreali unkemdo
adowoisi icttqf
alpostnikov
m_nastroenie
optashkin suzzankavesna mnik197796
wanderervn
er_novosti madinasageeva
aweann alisamokina
pacayozke
cockroach_brain evank_sp phylisbszz02ki kcoraleebapqq alex_bogatyrev
voice_of_true
misha_pfacan
alexcrafty
alexandre_mho
zewyelair tatianadambaeva
taushakkar
yiibhnp
virtu5
krymovvb shtefanrv
exkarlson1973 kirillschitov
tsemenov
iskhh_xajse
ritikyry
grizzlynohead koctypolevoy
eufaraday karimov_oleg
glerka_88
anpersimpningh
ufleearxw
aleksandrashv
kuanov_yury
1lesha1
desctertmiolos erminchik7
inkahegs
mishak_borshh
psdfsdfdx
eddzvaard
sladckovich konstantin_reva
kurpasov n_telepnev
memfis1962
beler77
ivlan2009i
sup_igor
ac_ovb
alexpolt2
zyablya
osman_0c5
alexchesno
vova_moskva
mikaelrus sherzakir
myshkinmr evasilkova
mila_niskanen
vbortnikova
inoyvi
kgirzaeren afanansmax
dutehuzsh
copoka7
2zesky ivan___durak
cccp2_0kassade
ignabi
batyonu
blhood_lenny epresfinider
jmewelcij
denniman_er
hohlow_a kazdalevsky
dimmironov
lihqoaxon
rssdiagest
ivan_kargalev
3abtpa
_2235646523811
mashina_s
givanisis
tpodarit
valerivolga
skarlet28ohara
vomolbo1
chybykov
eglassh_com_ua
evgeninyabest99
riboclovny qkkrng
tradparlohna
na_nitochke
franindcuchond
miva_iva rwedsilber145 qbrittneybrl
fausts91
denyschaynek
cojzee cbc2000
rykov
klon_1q20
dim4iok89
anatqolb_547 victgr1 ten_sergey meriko_20
shlegel
file_11
tolmachyov eparanoik_sleva
edinros34
maksimusrus
denf_bs
ilyakanakidi
akmpavel
life_drgift
uncleean_spirit
yazvilka dmitrykharlamov
korvin1973
dariadikon
ivan_okuzmenko
vaxkesfdi
blackv_bonnie katrin_aaaa
moscowck
gorshkovnikolai
yzblec
istwtc
garon0075
rybakoff
dorian_velikyi
mitinginfo
scroogeman84
molonya
ometn_darkwatch
lenazavrik
ylajyp
converter_being
akna_slesh
guiltu_jluke
kulkdodar
hikjjve
svetazaharova
beria_lp
monomakh34
helen_uvarova
ditya_bqashorga
karasevalexey
good_mum prodoc2 oligathor
ifuqeleri
lokotyvyf
kvmironov
vestovojw7678
teractordriver
zmqmsy
mgvardia_54
alleeex_g
tralabalah xoccrh esoulrocorest
i_unspacelobagpheaaga
vovansokol
krasauskas73
elenalip
leeroy_drunm
ent3m1
kuzoqteal
babsym_po_nebu
klakson102
micawely
ginnoff
cka65guf
maroipg
free_man_brelya
tivesude
shvidskoy_pahom blyambzambam
derytyl
serunim
wipakmips aniaslanyan
qgadr12
ne_almkogolik jdkfvg
ilyas_kalmikov heriiginth
tifflefvgaiflir
amra1993
sadreeff
juventa_29
booawlbdk
xepzae
sufdone
yvrgcf
lxokokaluga
stantfaranli sfpodus vsaske_uchiya katrinoros
zverogon
mellarossa
_lena_k
alekseigrachev
ladacipyma
uzhastpeg
tovpergifts
alina_dovgulya
arina_ra andre_soldatov
pedanov
bigagagicv
cynimux
isemnuus azbukywedy golchikov1
anenkovaanya kremlyanka
gytdvjybz jonvlasov
yt82svnap
khakimov_ilyas
chemerisss
xywycygu
mantsur
alena_arshinova
easnumtirva
tommi_venrsetti
starogaz
temkinmir
n28vova
sergei_gr_ussr
vitaliy_logos
otdedyxibr1985 multik6z233 ironmikka
vova_usuxarev
maqrorgolder
vespeurtilio7 murzaewav
aqucozyzo unujibygos
kolganodv_s
marlenka_rus
lyukshinnikita
akpp_otvetty bywijiqe akpp_otvett
molgvardia34
barboze
kurbatulya
missbysinka kcavkazec_arm
antesla
denisenkoden
er_ivanovo
izkotrsstbrss
onelovemoscowdivotelsedakun
a_galitsyn lyzeniz
izejobula
shushan4ik qwarker
kvol4_ok
ydau4abl
rimlywan ladiginocity
sasmarinaxlk
rus_tm
iamstinger
belov49
svetariesmeltory
rikhtman
alexs_ctvet bruryj_laki
allengolng
prushkalev trebunskih
n_romanenkova
skripinsckij
nloony_dnb
pfyhaus pcakaship
ireacan
zvezdo4ka6p66
akppotvety
burmatoff
c1rcool oganesyanarman psyangel
melon6f666
mkoditor
alexgizn
smolenkov telushkin_m disabyruqo
galanskii
timsdc
dzhebntlmen
s_shipunova
gattarov
lakkuchu29
brusni4ka76
annastarkova
bfead_one
mivck71
rolexx777 amareev
goshanazgul
akumyba
veraminicooper
vadstd
nashacanada
romaprud2015 akpp_otvety upasedu
zonchik deadejig
singertroya
uszbxs
dyadyushkea39
bav19802006
sretenskaya drupady albonumismatico
ifehily
olly_jolly
rybachek22 dubov1967
justbecaus
ss_i_vchk
gvbhnjmkjhyb
akppotvet
iletunul
maoreman10
screenctlean
xolorkostya
stas_kireev
ostarkova_irina
garaninandrei dutov_rossmi cazeruli icutopedu
artempushkin91
dmd13rus
artur_shamilov
sigl_pvml
wrmafq professcor6467
denisjara
ruslax
cetamez
diety_pohudety
akelubivy
negubami
blog_ulann_udebena_dzhojs
olegcheb ilja13132
icipigi
cyfocevil kyfocecihu
nufawela
sanvlord
rusachkaaaa
gepemoj awavaqe diety_pohydety antonyurtaev 1_kanal fyrigaryn logopedavr kspavlova burgphill chepk90 swy79ya olyjelukow m_kosheleva 2012terskovs tver0fadeeva ukinetowul vadim_0526 kholodow yuyanna201 umaruvo macggg1 g_ulanova joninvincible
topoprf
zhaqwer5
dsjqwpyqx aaawkkoo
velikastranarus
urolygizi ayrtonessaratiboardigor3711
vtrupa
ifgorleb
julewn1785
ferneclowieii
gazzz87 epzasn
volzt1989 m_mixalyvch_m
ehee535
junkoyoyzir
rostov35
dtjmug
aleshka_der
turow_a
ktxjvk
cingmysdewbi
shlurkin_roman
celtfowebtthear bohss8888
mmx_09
alekbsandr_bam
i_fjuma_fuma_i
parcdi100g bxinaellitr
myulij_dunaev moi_fotografivi
sztepan_fliger banlimit
agoltsov
rinkaza
proud_angeal
xlsexsx
nememlass dafsdgfaosdf
vugufyozorytir
xmozt
gopstygl
m77k263 huvegbogy
m_petko
vitaly_russia
deron66o6 bramosxcow
matan12f3
madfbun90
vmmandrika lufiopohe
napishi_putinu
maratakeno
nonistylecxlub
dgcfsk
keisy001
dhen9847
tmugick
tnucha1988 sxkywx
pashkov
racmstuning
ten_nyindzya
melisjs_fr
aleksej_saptundrpa123mr
tpirat53
forget_the_way
lovehaski
006t006
yklisho
vaklvul
aosmrf
mulejntiqj
ichbinuberalles
confnuthoduhotn
traxtenugarfild
hevrgandfeti
pvdovchenko
kolmakovann
broskomagazine
hasaminmai1979 richardj_asher
aleksej_8p1
pisudecroe
smirnov_ant
pavelkf
angel_sexa_uboy reedqrrdlenke chammenawthing
dawnieltdd
teddy_bearp23
tavreich
fragmeznt80
mrddfrkeeman bxngxy
hkelmxuv
ms_nobodya_666milliexwxyhy
altinwamye aleksmachetue
_i_a_d_
polufeanocr
sihehihly
chernogorskaya
zlobnyxj_yak
cdilax29
ryabov_mike
andresherdor
tofosuhmoz vanka_comsk
gniom10
videoksur
pvasenehy
_patriotrossii
jpqyjn
pureglpllitato arteym_mex
azimo_master
pravoved007unilisyva
galphastv ivan_zhuchfkov igyor_ovsksurt9966 ivan_govnov7i77
nterekhov
superhelvire
fakbtorxe
ihcparti
yurij_1980300x1
gallifreyfalls emo4ka1t9942007
sofo4ka
peretdoslsd aleks437v3
dead_szince
sdprbk
shon_fteris
sarnya_borodach ieshuualpdi herho19
vqlnqa
zapadnex
kukaanxamon leshae_kozlov
edjlyb
tclvbo
dbaf11 luk_tgs
gadskij_klotos
deshaunttgg
rumpelstiltsken andrey_rz
mintzru
nsevaspb
sidmponster
efremovra poskermastera
zemkuhru
dwhkth stuimul77
ser_andajrtuss
neomjcl
igicalbor
daniilkomkov kprostoamdrei
brazebrgold
rpovovsercju
paahkc
vertsinski tpaulocredi
elenapawlova
lyoshuka_chirik
smotrgdash
divide0 veselyputin
atlekander
romio199p1
aasdxcxzc
mbtobis noneastyxnee440 inavlvde
fakmcing_boy
gojspel81
amkusxqa zodomoc
anastasia_1308
catb1ack
ostaryj_goblin
npmljj
hfjnyb
ivwxmj
ofunccysrated
rtixij_shu
rooomcster
lwirndevio potjexin_artyom sca7788pyriore
sadoffnipk sobaaachdka
yoursmki cowboy757q7
serafim_n777
neuklyuzfhij
vatribapuv
godkunife
sorenburg777
kurssor
tmolikqwerty
pskysonny
ilchlfqow gdumqa
dauanf
olyam_roterman
gipbersex
alesnko7
thekenanytlt
livero_alias
skrwomnenkij
olegvasen
calexandromg25
maxim_volkov
vandaudvvnore slgavamail psdiholog2010
riapelphcopa
proselwriter jammes_cash
kooks_ann
ante_doribx fuebuculm
mraperliland
grusutnyj_shut shuphart2006
fnkqhk
vadikbogdanov alhavardo
cptekg rakitylya
dimoib
dmitry_kirienko
rjaneair23
tapokz_polet
ayul_student
dizmaxiz
fmjbid
xan_rbaf qeduthati
tocuamaja
volodyba1725
mhayskra
smokfi_mo_2
oxotonawuh sergmey72465
pypuwaqw
twkachellos
sabshadmb showx_2009
opeqotyke
dqkqvizpe
gserman_gott pkazumasa_is zhits_galimo mr_saslhka
pjsidz
bladem261
prezident_slk
lenorehesdseppq
black_wnemo iamsinichkin
guodknife orusrem
akarandin
santa200i606
rezhirm
sumemersii opasnee_voalka cordewlltee
razrubateel
funtik_195p5
agki478
oleg_einar katetretiakova
bvicser2 afebuwire
kbamillakwkkj
jejvj
shamean1996 pawiel_iwanov
gnyji921
kennfy_nw
kvidrs nicabanoa
nintipverti
boba_akao_eugen
michaelf_soyer
v1hyh1 milyan_grechkgo uljajmat kovstya77791
yadccuheo
ivankaanna coacpaftapzi greenfield_
ciecomdopho
vasfya_luzer
baranecte6lannfrzvw
xozyayin_zhizni qhausparinci
lina_indie
extravalex timazcimaz vvladk01
decembertale pajvlik_mudryj
rrfdav
ggamaydn
ailene67a78pul nikvolya41 vavilagn
arhdpyxa
rtuahr
fsunxdcut trboppik
ilya_kovalev
zsvpyn
bactalevaleksei
denis_dolganocv
pchedl
kukuesrhka
rybamchki
dsboris brgearius
befenmarggansna sxtesen extazzy_dyeh kwlerk1981 begazytafgik live_is_ksin
csydto dzyliak
nhoizersr stez3455nisly
hyogiharp io_noptwoc_ioi
aleix_ivamoto
piikoqut radisnlav25
mnekaz
freemanr_xxx
cuegvv adriaan49
turpji
alecsgeyyy klonk_136
rtetwtywgqwgw zeromcool_zz
Appendix VI: Dueling Meme Networks
pasl81 fleontiev tuganbaev
ganbik leisha_xoroshij
gkates ftzwxz yankicolon
orden_feniksa
elenapawlova chernogorskaya prozzak
diodoroff
er_ivanovo
reclusedotru rooliana osv723
witcharrr nickalleye
samklebanov loskoufa fadeichevpavel annastarkova alexander_rybin _i_a_d_ smirnov_ant
advokatkubany
doctorguinness jfislove
tokapb
dimazub
scut36 artpimanov
nagletssklyarov morsnews_ru
trrresh
playd
miketansky vesel4ak_y
sutkliff just_disobey
p4olki
vladislav21_
fleurklin
enotsava
zonkhoeva tplisina
senko v_milov eksvu t30p patriotrus1981 artemshitov greenlace
vmysila
msilab vika_morana
vakurov 4irikova
bryanskru
kendannedy amozharov hitechservice lecafard avrogachyov bro256 hasovsztb roman_primorye lovlimc doctorvii dneuymin nemo_47 dimmyboyk soulchva88 emdimon strizhev aleks1993ev 3aru80 ingenerishka katok666pk metallurg61 raulsangre hello_daria xthyjdjd sensimon1976 markevgenevich brand_michel texn1knikitkad rupetyaandrey_kubka andloban markwardog kulischew uralasbest aleyurch mvelmakin ligaizbirateley zhenyabahotskii ellabraigen potatoes_82 electionrussia rosvybory yuri_tyan egorscha proudlyfly go2berlin dalekodaleko galodium yodapunk
al_larin
klinkinm annabernstein lyudaluchik
russianreporter
szimniy varlamov
ndrevolt
the_prophf
mixei4 politrash
solomonhaykin salty_1970
tiranu_net tvrain paplic1987
pennupa _eversummer_
iile1
atomochka ostup_bender netstrannik _9737832830961 papa_anna aspiridonov efremovsergey lopouhiy bymth durdendaria eddinovak vladimirzakhesken_h zhgun grad_na_volge34 martin_fedor nevasya pinkus_ plazmatics romanosharov aragorn28 izagriv ru_kingpin the_newtimes kozlovsky k_golitsyn nijatguluzade nhacizade pellegrino7 fakemath bookmistress kotmoriz samagreene natalies19racoon_sergeeva vorewig nurislamovrinatdjb0b golden_oar shnumas gortolspb anna_veduta maxtaksi nblxa kirnifex svinina chaekrus kibanov forgotten_anger
77_rus nkukharenko kjdsar
egor_mq
yrasunrise
aleshru robotdemagog vgtrk_ olly577
zehnerson oleg_kozyrev s_eirena azikazimow g_kolomiets moscowlondon sokolovkirill
4annel drandin
navalny
ru_pirateparty
userpig
av_shikchanov
glavportal
lentaruofficial public__post
goldmandennis flowerflavor alexgoncharow grungebucks fontanka_spb 4mapta eufaraday juliyvchirkov tatoha56 anfirov chybykov a3ap shabanov_s_v sunsunych85 itimrussia dimmironov ain92ru vadinsar nyksen dmitrij7 kseniatr kissmyba riddler63 qloko eugenechigir kmrsfm funkodelic dracoola13 btf abugray evgeshik sadie_sponge kykyllionok danteshaneils regett kt_urban elizabaat reporter_tlt booshma tatarinov converter_being ruspoker kalininru svetamerenkova ignabi gytdvjybz korobkov ander511moveolocus baranova17 winzard d00h ataka440 lexusla kothates koctypolevoy mikhashenok tanyadm11 bogdanlinch antesla denisenkodenvrebyata voice_of_true sergeypush34 n_telepnev rykov optashkin topoprf tagan77 inopressa _uylechka_ hodjaevb shlegel modinmordin mashina_s _julia_v elenalip urfin_1648 vol1oleg kirillschitov memfis1962 yanabogdanova c1rcool turow_a kurbatulya ten_sergey barboze aniaslanyan monomakh34 3abtpa qwarker barnumix er_novosti kuanov_yury _lena_k dmd13rus alekseigrachev ivan___durak galanskii arina_ra evasilkova 2zesky jetzajac gattarov juventa_29 denniman_er wanderervn sherzakir vbortnikova ginnoff skarlet28ohara denisbeta rybakoff ress artempushkin91 vivintic klakson102 molgvardia34 valerivolga kurpasov shushan4ik n28vova hatipov mila_niskanen andre_soldatov ritikyry n_romanenkova madinasageeva trebunskih alexgizn temkinmir mantsur tjorrr vomolbo1 cojzee alena_arshinova vova_moskva zonchik sadreeff lyukshinnikita artur_shamilov murzaewav serunim rybachek22 smolenkov rus_tm gazzz87 azbukywedy chemerisss meriko_20 golchikov1 mikhail_golubev dorian_velikyi zverogon m_petko ayaz_01_ ra3new olly_jolly krasauskas73 siaynie dgordgina mellarossa basegprm molonya velikastranarus marlenka_rus imakhaev antonborisov vanuska_ ayrtonessa wer4ik mikaelrus
alessio_vee
tchernysheva
kepuby
burmatoff
1_kanal
pavelkf bihanskyy elvigubushiev twitchenko ausfan67 estraniero a3imbayev
zlaw
romario__agro
real_naumov
rostov35
katekotik russian_police zapyatoy slvbgtrv
andreydoronin bajiek_ anutka_1996 evgenyii
letmefashion
rashemi viktor_bad
mmx_09 napishi_putinu vertsinski