Memes and Narrative Strategies in the Russo-Ukrainian War: A Comprehensive Study

Authors: (1) Yelena Mejova, ISI Foundation, Italy; (2) Arthur Capozzi, Università degli Studi di Torino, Italy; (3) Corrado Monti, CENTAI, Italy; (4) Gianmarco De Francisci Morales, CENTAI, Italy. Table of Links Introduction Background and Related Work Methods Results Discussion and References 2 BACKGROUND AND RELATED WORK The 2022 Russian invasion of Ukraine happened at a time of wide adoption of social media, when the advancement of Internet infrastructure allows real-time high-resolution media sharing to a worldwide audience. As such, the actions on the battlefield and the reactions of those inside and outside Ukraine are streamed on Facebook, Twitter, Instagram, and other social media platforms. This constant stream also “proves to be an essential mechanism in getting on-ground real-time reporting of a dangerous event” for mainstream media sources [61]. Unlike in the Arab Spring during the early 2010s, Ukraine is not blocking social media, but is instead fostering a presence on various websites both through official governmental accounts [56] and via an informal “IT Army” [59]. Our first RQ aims to describe the content they produce. Several theoretical frameworks can help us frame such presence. First, as a cyber-influence campaign, which seeks to promote a narrative alternative to that of the Russian side [33]. In the case of the Russo-Ukrainian war, such cyber-warfare began even before the annexation of Crimea in 2014, and has involved contextualizing the ongoing events in narratives that include, for instance, historical events during World War II, terrorism, NATO, and the world economy [5]. Beyond contextualization, both sides directly address the messages of the other, and respond to perceived untruths, or “fakes”.[7] In this work, we use the narrative framework to study the content generated by prominent Ukrainian accounts [18]. The second framework is that of nation-building or nation-branding: two processes that were originally separated in the domains of the government (former) and PR campaign managers (latter). They have now coalesced into a single process that employs the dissemination of cultural artifacts via various media channels. Bolin and Ståhlberg [4] observe that the recent campaigns by Ukraine are “aimed to raise support for EU membership, arms deliveries”, and try to promote “Ukrainian capability in terms of decisiveness, bravery, and responsibility”, while targeting both international and domestic audiences. Such self-promotion can also be thought of as a use of soft power (or selfie diplomacy [45]), wherein the country relies on its stock of resources including culture, values, and policies to achieve favorable outcomes [48]. Whether such soft power is effective is determined by the success of the nation’s communication, which is the second RQ of our study. The third framework is that of community building during a crisis that “facilitates stakeholder support and builds relationship” [11, 20]. According to Jiang and Luo [32], successful crisis engagement depends on providing timely and accurate information, engaging with the public empathically, and reaching the audience via content forwarding and constant multimedia-enhanced conversations. Indeed, if Bolin and Ståhlberg are correct about the aims of Ukraine’s communication campaign, the intended audience of this communication spans not only the domestic media users but also potential allies and supporters. The third RQ of our study considers the international reach of the content produced by the Ukrainian Twitter accounts and correlates its popularity in a country with the actions taken by that country’s government. Since the annexation of Crimea in 2014, the social media presence of both Ukraine and Russia has been closely scrutinized by researchers in communications, media studies, and human-computer interaction. The separatist movement in the Donbass region has intensified a battle of narratives. A study of the Twitter messages with #SaveDonbassPeople shows that both pro-Ukrainian and pro-Russian sides employed five contextual frames: historical, geographical, religious, ethnic, and political [43]. According to Makhortykh and Sydorova, those on the pro-Russian side were also more likely to make extensive use of photos of children, the authors speculate, in order to “evoke compassion from the potential audience by using sentimental images”. Similarly, contextual framing has been studied on the Russian social media website VKontakte in 2014 [44]. Whereas the pro-Ukrainian users framed the conflict as a limited military action against local insurgents, those supporting the Russian separatists framed it as an “all-out war against the Russian population of Eastern Ukraine”. The authors speculate that the use of such divergent frames “led to the formation of divergent expectations in Ukraine and Russia concerning the outcome of the war in Donbass” [44]. Since the full-scale Russian invasion, several datasets have been made available to the research community. For instance, Chen and Ferrara [10] collected Twitter posts having various Russo-Ukrainian, war-related keywords. The volume of engagement with these keywords peaks shortly after the invasion, and gradually decreases after March 2022, thus pointing to a limited attention span of the users on the platform concerning this topic. Focusing on the accounts participating in these online conversations, Hare and Jones [29] track the use of the Ukrainian flag as a marker of support on Twitter in late February 2022. Such displays may be termed identity activism, which consists of the “prominent display of a social movement symbol within a space reserved for describing oneself” [29]. Such accounts are homophilic (more likely to follow others who also display the flag), and more likely to share U.S. Democrat-leaning messages. Unfortunately, automated accounts, or bots, are often part of a communication operation. Shen et al. [57] estimate the share of bot accounts around the beginning of Russia’s invasion in 2022. Using the tool Botometer [64], they identify around 13.4% of the tweets as likely to be generated by bots. Most of these tweets espoused a pro-Ukrainian position; it is worth noting that by then Russia had suspended access to Twitter for its citizens. Due to Russia’s unique position in this scenario, we exclude it as a potential audience of Ukrainian tweets. The term meme was originally defined by the evolutionary biologist Richard Dawkins in 1976 [17] as a replicator analogous to a gene in its ability to transmit information, including cultural artifacts and beliefs. Due to the vague nature of the term’s definition, and the fact that the very nature of memes is a “continuous mutation”, the study of memes, or Memetics, is applied broadly to the study of cultural information transfer [34]. In the age of social media, the concept of “Internet meme” breaks from Dawkin’s analogy to focus on the artifacts—texts, images, or videos—instead of abstract ideas [39]. Indeed, Reese et al. [52] point to the properties of the visual medium, including syntactic implicitness and iconicity, that make it particularly suited for “framing and articulating ideological messages”. As memes are defined by their capability of being modified and reshared, we consider all visual media as potential memes [67]. The ease of its perception compared to text [54], as well as a certain ambiguity [28] allows memes to permeate communication channels easily. Thus, memes have become an invaluable tool for the transmission and evolution of narratives on social media, and as such they have been extensively studied in the context of political communication. Visual communication, and memes specifically, have been studied during the 2016 US Presidential election [63], the 2019 Ukrainian election [51], in Brazil [13] and Hong Kong [21], in Germany’s [3] and US’ [14, 26] far right. Military actions in the past decade have been accompanied by meme-supported cultural expression by U.S. troops [58], by anti-Islamic State activists [47], and in the early days of the Russo-Ukrainian conflict around Crimea [62]. Finally, a notable precedent in the usage of memes by state military is that of the Israel Defense Forces [46]. Hero (benevolent, strong); Victim (benevolent, weak); Villain (malevolent, strong); and Fool (malevolent, weak). Finally, the popularity of memes has also been studied by using visual features, e.g., scale, inclusion of text, and attributes of its subjects [41]; we use these features as controls in our analysis. The Human-Computer Interaction (CHI) community has long been invested in promoting greater global solidarity and in increasing the diversity of researchers and research subjects [6]. Early calls for the CHI community to focus on war and peace have revolved around education, using new technologies to connect opposing factions, and exposing the horrors of war [31]. Additional attention has been paid to the processing of trauma, including via social media platforms [55]. Although most studies have focused on the individual or group experience of trauma, a similar view could be taken of a nation processing the ongoing trauma of war. As the government and people of Ukraine attempt to convey their experiences via social media, the same design principles are important: transparency [19], empowerment [30], and peer support [1]. In this study, we focus especially on the retweet action as an embodiment of support by the users of the platform. Further, CHI has had an important role in elucidating the relationship between social media and political discourse around the globe. Recent studies have revealed the role of demagoguery in civic engagement on Reddit [49], participating in online activism (“slacktivism”) in donation campaigns [40], and reach of anti-migration ad targeting on Facebook [7]. Although user bases of social media platforms are known to be non-representative of the larger society, their growing number makes it increasingly important to examine the efficacy of political communication and its possible real-world implications. [7] Examples of prominent websites self-identifying as fact-checkers are https://www.stopfake.org in Ukraine and https://waronfakes.com in Russia. This paper is available on arxiv under CC 4.0 license. Authors: (1) Yelena Mejova, ISI Foundation, Italy; (2) Arthur Capozzi, Università degli Studi di Torino, Italy; (3) Corrado Monti, CENTAI, Italy; (4) Gianmarco De Francisci Morales, CENTAI, Italy. Authors: (1) Yelena Mejova, ISI Foundation, Italy; (2) Arthur Capozzi, Università degli Studi di Torino, Italy; (3) Corrado Monti, CENTAI, Italy; (4) Gianmarco De Francisci Morales, CENTAI, Italy. Table of Links Introduction Introduction Background and Related Work Background and Related Work Methods Methods Results Results Discussion and References Discussion and References 2 BACKGROUND AND RELATED WORK The 2022 Russian invasion of Ukraine happened at a time of wide adoption of social media, when the advancement of Internet infrastructure allows real-time high-resolution media sharing to a worldwide audience. As such, the actions on the battlefield and the reactions of those inside and outside Ukraine are streamed on Facebook, Twitter, Instagram, and other social media platforms. This constant stream also “proves to be an essential mechanism in getting on-ground real-time reporting of a dangerous event” for mainstream media sources [61]. Unlike in the Arab Spring during the early 2010s, Ukraine is not blocking social media, but is instead fostering a presence on various websites both through official governmental accounts [56] and via an informal “IT Army” [59]. Our first RQ aims to describe the content they produce. Several theoretical frameworks can help us frame such presence. First, as a cyber-influence campaign, which seeks to promote a narrative alternative to that of the Russian side [33]. In the case of the Russo-Ukrainian war, such cyber-warfare began even before the annexation of Crimea in 2014, and has involved contextualizing the ongoing events in narratives that include, for instance, historical events during World War II, terrorism, NATO, and the world economy [5]. Beyond contextualization, both sides directly address the messages of the other, and respond to perceived untruths, or “fakes”.[7] In this work, we use the narrative framework to study the content generated by prominent Ukrainian accounts [18]. The second framework is that of nation-building or nation-branding: two processes that were originally separated in the domains of the government (former) and PR campaign managers (latter). They have now coalesced into a single process that employs the dissemination of cultural artifacts via various media channels. Bolin and Ståhlberg [4] observe that the recent campaigns by Ukraine are “aimed to raise support for EU membership, arms deliveries”, and try to promote “Ukrainian capability in terms of decisiveness, bravery, and responsibility”, while targeting both international and domestic audiences. Such self-promotion can also be thought of as a use of soft power (or selfie diplomacy [45]), wherein the country relies on its stock of resources including culture, values, and policies to achieve favorable outcomes [48]. Whether such soft power is effective is determined by the success of the nation’s communication, which is the second RQ of our study. The third framework is that of community building during a crisis that “facilitates stakeholder support and builds relationship” [11, 20]. According to Jiang and Luo [32], successful crisis engagement depends on providing timely and accurate information, engaging with the public empathically, and reaching the audience via content forwarding and constant multimedia-enhanced conversations. Indeed, if Bolin and Ståhlberg are correct about the aims of Ukraine’s communication campaign, the intended audience of this communication spans not only the domestic media users but also potential allies and supporters. The third RQ of our study considers the international reach of the content produced by the Ukrainian Twitter accounts and correlates its popularity in a country with the actions taken by that country’s government. Since the annexation of Crimea in 2014, the social media presence of both Ukraine and Russia has been closely scrutinized by researchers in communications, media studies, and human-computer interaction. The separatist movement in the Donbass region has intensified a battle of narratives. A study of the Twitter messages with #SaveDonbassPeople shows that both pro-Ukrainian and pro-Russian sides employed five contextual frames: historical, geographical, religious, ethnic, and political [43]. According to Makhortykh and Sydorova, those on the pro-Russian side were also more likely to make extensive use of photos of children, the authors speculate, in order to “evoke compassion from the potential audience by using sentimental images”. Similarly, contextual framing has been studied on the Russian social media website VKontakte in 2014 [44]. Whereas the pro-Ukrainian users framed the conflict as a limited military action against local insurgents, those supporting the Russian separatists framed it as an “all-out war against the Russian population of Eastern Ukraine”. The authors speculate that the use of such divergent frames “led to the formation of divergent expectations in Ukraine and Russia concerning the outcome of the war in Donbass” [44]. Since the full-scale Russian invasion, several datasets have been made available to the research community. For instance, Chen and Ferrara [10] collected Twitter posts having various Russo-Ukrainian, war-related keywords. The volume of engagement with these keywords peaks shortly after the invasion, and gradually decreases after March 2022, thus pointing to a limited attention span of the users on the platform concerning this topic. Focusing on the accounts participating in these online conversations, Hare and Jones [29] track the use of the Ukrainian flag as a marker of support on Twitter in late February 2022. Such displays may be termed identity activism, which consists of the “prominent display of a social movement symbol within a space reserved for describing oneself” [29]. Such accounts are homophilic (more likely to follow others who also display the flag), and more likely to share U.S. Democrat-leaning messages. Unfortunately, automated accounts, or bots, are often part of a communication operation. Shen et al. [57] estimate the share of bot accounts around the beginning of Russia’s invasion in 2022. Using the tool Botometer [64], they identify around 13.4% of the tweets as likely to be generated by bots. Most of these tweets espoused a pro-Ukrainian position; it is worth noting that by then Russia had suspended access to Twitter for its citizens. Due to Russia’s unique position in this scenario, we exclude it as a potential audience of Ukrainian tweets. The term meme was originally defined by the evolutionary biologist Richard Dawkins in 1976 [17] as a replicator analogous to a gene in its ability to transmit information, including cultural artifacts and beliefs. Due to the vague nature of the term’s definition, and the fact that the very nature of memes is a “continuous mutation”, the study of memes, or Memetics, is applied broadly to the study of cultural information transfer [34]. In the age of social media, the concept of “Internet meme” breaks from Dawkin’s analogy to focus on the artifacts—texts, images, or videos—instead of abstract ideas [39]. Indeed, Reese et al. [52] point to the properties of the visual medium, including syntactic implicitness and iconicity, that make it particularly suited for “framing and articulating ideological messages”. As memes are defined by their capability of being modified and reshared, we consider all visual media as potential memes [67]. The ease of its perception compared to text [54], as well as a certain ambiguity [28] allows memes to permeate communication channels easily. Thus, memes have become an invaluable tool for the transmission and evolution of narratives on social media, and as such they have been extensively studied in the context of political communication. Visual communication, and memes specifically, have been studied during the 2016 US Presidential election [63], the 2019 Ukrainian election [51], in Brazil [13] and Hong Kong [21], in Germany’s [3] and US’ [14, 26] far right. Military actions in the past decade have been accompanied by meme-supported cultural expression by U.S. troops [58], by anti-Islamic State activists [47], and in the early days of the Russo-Ukrainian conflict around Crimea [62]. Finally, a notable precedent in the usage of memes by state military is that of the Israel Defense Forces [46]. Hero (benevolent, strong); Victim (benevolent, weak); Villain (malevolent, strong); and Fool (malevolent, weak). Finally, the popularity of memes has also been studied by using visual features, e.g., scale, inclusion of text, and attributes of its subjects [41]; we use these features as controls in our analysis. The Human-Computer Interaction (CHI) community has long been invested in promoting greater global solidarity and in increasing the diversity of researchers and research subjects [6]. Early calls for the CHI community to focus on war and peace have revolved around education, using new technologies to connect opposing factions, and exposing the horrors of war [31]. Additional attention has been paid to the processing of trauma, including via social media platforms [55]. Although most studies have focused on the individual or group experience of trauma, a similar view could be taken of a nation processing the ongoing trauma of war. As the government and people of Ukraine attempt to convey their experiences via social media, the same design principles are important: transparency [19], empowerment [30], and peer support [1]. In this study, we focus especially on the retweet action as an embodiment of support by the users of the platform. Further, CHI has had an important role in elucidating the relationship between social media and political discourse around the globe. Recent studies have revealed the role of demagoguery in civic engagement on Reddit [49], participating in online activism (“slacktivism”) in donation campaigns [40], and reach of anti-migration ad targeting on Facebook [7]. Although user bases of social media platforms are known to be non-representative of the larger society, their growing number makes it increasingly important to examine the efficacy of political communication and its possible real-world implications. [7] Examples of prominent websites self-identifying as fact-checkers are https://www.stopfake.org in Ukraine and https://waronfakes.com in Russia. This paper is available on arxiv under CC 4.0 license. This paper is available on arxiv under CC 4.0 license. available on arxiv