Abstract
The bots’ activity is already frequently documented in the literature, and the war between Russia and Ukraine accentuated this scholarly interest for users’ sensemaking. Applying folk theories framework on 56 semi-structured interviews with users who tweet about “Russian bots,” I examine how bots might be understood as structural-computational entities, with complex roles in shaping digitally mediated realities. Findings reveal several theories associated with Russian bots. First, participants believe that these bots actively endorse users’ political enemies, which are mainly politicians from the participants’ countries. Second, such bots are considered to increase animosities between users, as participants actively unfollow their peers on Twitter and unfriend them in real life, based on their opinions regarding the war in Ukraine. Third, bots boost users’ digital activity, given the fact that participants consider them responsible for artificially increasing the popularity of certain accounts or, on the contrary, for systematic and aggressive attacks against others.
Introduction
The work of the Russian Internet Research Agency (IRA) is already well-known and documented in terms of its digital disinformation campaigns, both through trolls and bots (Dawson and Innes, 2019; Hern, 2017; Popken, 2018). It is already known that, in the context of the 2016 US presidential elections, Twitter revealed 37,000 Russian accounts whose tweets were viewed more than 300 million times (Cosentino, 2020; Graham, 2018). Of all the contributions focused on the IRA’s digital strategies, it is worth noting the Kremlin’s postmodern approach to Twitter and Facebook, designed to accentuate the permanent polarization between groups—whatever they may be—without attempting to nullify any particular group’s identity.
Unlike most other studies focused on identifying patterns of Twitter bots but also the strategies based on which they were programmed (Boshmaf et al., 2011; Haustein et al., 2016; Neff and Nagy, 2016) this study follows the main discourses and users’ folk theories when they encounter Russian bots on Twitter. Such an analysis is not limited to the mere identification of such bots but is rather about the perceived implications that these bots have on the daily political and social life of the users involved. Under social constructivism, I care less about whether these users correctly identified an account as bot, 1 so what matters are the ordinary experiences and valences that users associate with what they consider to be Russian bots.
While I share Brooker’s (2019) view regarding SNS bots as a sociological concern that should understand the social mechanisms behind the lines of code, I believe it is at least as important (and yet so understudied) for sociologists to examine the ways in which the realities associated with foreign bots are part of several systems of thought, especially in the context of an increasingly visible IRA activity in the digital environment, as a result of political and social events such as Brexit, Donald Trump’s victory in 2016 (Snyder, 2018), but also the invasion of Ukraine by Russia, from February 2022.
The main argument of this article is that users experience the presence of “Russian bots” on Twitter through several ways of sensemaking, along with the adoption of multiple systems of thought that guide their values and online experiences. As a result, I will highlight such systems as folk theories, that is, “intuitive, informal theories that individuals develop to explain the outcomes, effects, or consequences of technological systems, which guide reactions to and behavior towards said systems” (DeVito et al., 2017: 3165). Such an approach is essential for critical data studies, given the fact that the activity of such digital bots is recurrent and well-known on platforms such as X (formerly Twitter) and Facebook, being frequently associated with interference in Western political elections or other important social events. Also, the transnational character of this research allows the investigation of users’ cultured capacities (Siles et al., 2020), that is, the discursive and identity repertoires through which their cultural origins interfere with the adoption of certain theories to legitimize the activity of “Russian bots.”
Twitter bots: imaginaries, folk theories, and social sensemaking
As I will highlight in this study, users frequently associate positive or negative valences when it comes to social media bots, and this trait is even more visible in the case of what they consider to be “Russian bots.” This approach falls within the scope of prioritizing reflexivity in the study of digital entities (Geiger, 2016), which highlights the fact that the existence of social media bots is surrounded by continuous and collective sensemaking so that users outline their social realities as intersubjectively as possible when they encounter such bots. Thus, it is worth investigating complex projects such as what users call “Russian bots,” as they reflect users’ heterogeneous ways of sensemaking but which are strongly anchored in their political, cultural, and social realities.
The activity of bots on social media platforms was frequently identified and noticed by users, and this identification comes together with users’ own value judgments regarding these bots. Thus, Twitter bots were most often considered to be either legitimate or malicious (Chu et al., 2010), and the situation became even more complicated when the political role of these bots was made visible. As Woolley and Howard (2017) explain, the purpose of these bots to manipulate different political elections made them acquire such a high-performance design, so that users rarely identify the non-human character of these bots. However, these bots are not meant to polarize strictly on political issues, given their contribution to other polarizing issues, such as increased rights for sexual minorities, or for religious and black minorities in the United States (Beskow and Carley, 2020).
Twitter bots were defined as “algorithmically controlled accounts that can automatically perform a variety of actions including posting, retweeting, liking, responding, etc.” (Stukal et al., 2022: 844). The examination of social imaginaries specific to some computational components was previously discussed in the case of social media algorithms (Bucher, 2017). Like imaginaries, folk theories are of great relevance in this study, as they show us what users think when it comes to computational activities on digital platforms and how these thoughts are translated into specific ways of acting. I see folk theories as a coherent way of integrating particular data assemblages (Siles et al., 2020), that is, a way of relating to a specific data ontology.
More precisely, I see these folk theories as enacting data assemblages, which will be able to show in a flexible and efficient way how users associate the activity of political bots with other adjacent technological aspects, such as the affordances of the platform, but also the constant process of agency negotiation between user and machine. As Siles et al. (2020) discuss this relationship between algorithms and data assemblages in the context of the recommendation algorithm on Spotify, it is noticed that folk theories are also a useful framework for understanding the political activity of what users label as “Russian bots” on X platform. It is therefore important to examine why users use some theories over others, as well as the main social and cultural implications of each theory. To investigate these implications, it is important to investigate users’ cultured capacities in relation to these bots, given the fact that “Culture equips persons for action both by shaping their internal capacities and by helping them bring those capacities to bear in particular situations” (Swidler, 2001: 71–72). I thus argue that espousing some theories related to “Russian bots” might be seen as a way to enhance users’ capacities in specific cultural contexts.
While imaginaries refer to the ways in which users relate to digital or computational entities on social media, along with the related consequences when it comes to adopting certain behaviors (Bucher, 2017), folk theories insist that these imaginaries need to be studied as plural (Siles et al., 2020), given that theories actually reflect wider social and cultural realities.
In the context in which such bots constantly develop their surveillance tactics in the digital environment—by sending messages to the most suitable target groups—users also develop their own discursive imaginaries through which they sense such surveillance. Given the fact that users constantly perceive the role that certain digital instances have in profiling identities (Bucher, 2017), we can also observe the emergence of certain tactics of invisibility (Talvitie-Lamberg et al., 2024) in the digital environment, through which certain groups manage to challenge these digital expectations associated with constant profiling and surveillance. Given that users participate in this process of social sensemaking through what Merleau-Ponty (1962) calls invisibilities, it is important to investigate how opaque computational entities, such as these bots, legitimize ongoing imaginative processes in relation to their activity.
There are different categories of social media bots, and each category entails different perceptions among users. When it comes to self-declared bots that have the role of performing basic computational tasks, such as actively providing answers to users who follow such bots, there are great levels of engagement from those users, given the fact that: “followers enjoy conversing with character bots knowing full that they are automated programs and designed to behave as fictional characters” (Nishimura, 2017: 128). A similar orientation of the useful and playful character is observed by Massanari (2017) in the case of Reddit bots, where the author finds that the bots most appreciated by the user community are those who “are polite . . . useful and informative . . . unobtrusive . . . not one-off creations, but engage with the Reddit community over a long period of time” (Massanari, 2017: 123).
However, not all social media bots publicly assume the role of self-declared bots. Bots who need to pass as humans generate different expectations among users who encounter them (Woolley and Howard, 2017), since the political implications generated by these bots create contexts in which they are feared or avoided as much as possible, like staunch political opponents.
In such a context, folk theories represent an indicator of the impression that users want to create, that is, how they want to be seen. In this sense, certain studies focus on the notion of enactment (Seaver, 2017; Siles et al., 2020), given the fact that users resort to different discourses and practices through which they affirm their identity in relation to others through the different perspectives on the role of digital bots. Therefore, folk theories represent a useful framework in this research, as they favor the understanding of “Russian bots” as representing a complex imbrication between structural and computational. Thus, although they apparently operate mainly on the basis of particular lines of code, we observe that the various theories produced by users who encounter them reflect, rather, their own social, political, and cultural realities.
The ability of bots to pass as humans has helped to shape a contemporary moral panic, and this has led users to attach increasing importance to the social worlds constructed by these software entities. Thus, in the context of a visible relevance for the socially constructed realities that surround social media bots, it is stated that they “both elicit and operate within particular contexts and constraints that rely on the symbolic construction of reality” (Jones, 2015: 1). Therefore, it is necessary that any social reality associated with these bots is not interpreted as it is, but is realized by taking into account “their symbolic and affective dimensions, whether originating in humans or machines” (Jones, 2015: 2). When social constructivism is employed, it is found that interactions with bots do not necessarily define their ontological and technological existence, but rather say more about ourselves, given that everything we think about bots becomes a reality itself, according to our own interpretive lenses (Bollmer and Rodley, 2017).
In the process of enacting folk theories, users adopt objectified versions of their own social realities in relation to digital bots. This objectified interpretation of social reality is a permanent process that social actors carry out everywhere, and in order to do this, individuals need to be able to convey their own values and social expectations in order for them to become meaningful. Thus, explaining social realities to others is an important part of the legitimization process, through which we justify and explain to others why our own objective realities regarding a social phenomenon represent the right way to be and to do. The way in which people experience multiple pieces of software or code was also investigated by Taina Bucher, in her article on algorithmic imaginary. As in the case of bots, the analysis of the social realities surrounding social media algorithms proves to be meaningful in a socially constructed world, given the fact that such pieces of software “are not just abstract computational processes; they also have the power to enact material realities by shaping social life to various degrees” (Bucher, 2017: 40). Given the fact that most interactions between user and algorithm mainly follow the imaginative processes of users, folk theories prove to be a productive approach as they evaluate these “Russian bots” as data assemblages (Siles et al., 2020), thus examining how agency is a dynamic process, constantly between power and resistance in relation to these computational entities.
The way someone defines objectivity involves an important accumulation of experiences and impressions, as well as more complex identity categories, such as affect. Such a component is all the more important because, as Papacharissi (2014: 12) describes, it goes beyond “just emotions and feelings to describe driving forces that are suggestive of tendencies to act in a variety of ways.” In the following section, I will describe the process by which I reached the interviewees with whom I discussed about Russian bots, in the context of the intense activity of the IRA, with the Russian invasion of Ukraine.
A recent and relevant event in this context concerns the suspicious death of Yevgeny Prigozhin in August 2023. Although for Western audiences, it is clearly an intervention by Putin to punish Prigozhin’s failed coup, Kelly (2023) shows how the Kremlin’s narratives via social media have another effect on Russian citizens, who are skeptical about the true cause of Prigozhin’s death.
Methodology
To select my interviewees for the discussion about Russian bots on Twitter, I searched and manually selected all tweets that were posted about the activity of Russian bots with the start of the war in Ukraine on February 24, 2022. I chose that date because it was characterized by an exponential growth in the number of pro-Russia liking accounts, as seen in Figure 1 below.
The number of pro-Russia liking accounts (Graham, 2022). 2
Thus, I searched for all tweets posted for 10 months, related to Russian bots. Find below the Twitter syntax that I used to search for tweets. I kept only those tweets in English and only those that mentioned an opinion (whatever that might be) about the activity of Russian bots on X (Twitter). I also excluded those retweets that did not express any particular opinion. In the end, 477 relevant tweets were left (N = 477) posted in the first 10 months of the war. Given the expectations regarding Twitter research ethics (Fiesler and Proferes, 2018), I will not cite such tweets in my study, as I did not ask for permission to post them and because these tweets can easily lead to the identification of the authors. Also, I did not mention concrete tweets in this study, with the aim of protecting the identity of some minority groups who agreed to take part in the interview under the condition of anonymity. The syntax I used to search for relevant tweets in the first ten months of the war is as follows:
bot* (Putin OR russia*) until:2022-12-24 since:2022-02-24 -filter: replies
Later, I contacted the authors of those tweets, whose profiles seemed authentic: name, profile picture, the ratio between followers and following, but also the type of content that was tweeted. I contacted all these users (N = 419) between January 2 and 15 2023, mentioning the purpose of my research, my email address, 3 and also assuring them of the confidentiality of their data.
After the 2 weeks of contacting users, 58 agreed to participate in the interview, so that we set the date of the interviews. For this research, approval was obtained from the research ethics committee at the researcher’s university. The interviews were audio-video, and all of them, except for three, were recorded. In the case of the three exceptions, I wrote down as much relevant information as I could in my notebook. Although 58 had initially given their consent, two of them did not confirm their presence on the day of the interview, nor did they respond to the researcher’s messages, so the final number is 56 interviewees (N = 56). The participants’ sociodemographics are available in Table 1.
Participants’ sociodemographics (N = 56).
Main themes and subsequent dimensions. a
All these themes should be viewed rather interdependently, and not mutually exclusive, given the fact that most participants fall into more than one theme.
The interviews had a semi-structured format and took place between January 17 and March 6, 2023, in English, and their duration varied between 48 and 87 minutes, with an average duration of 62 minutes. Given the fact that the main purpose of my research is to identify the main folk theories that Twitter users associate when discussing Russian bots, I asked the participants to describe as many experiences with these bots as possible: if they consider that the war in Ukraine has an impact on them, if they consider that these bots have any notable influence on their lives as “European citizens,” or “American citizens.” The participants also talked about their main experiences and affects in relation to “what these bots really do.” Given the importance given to users’ folk theories regarding these bots, I asked the interviewees to explain how they think these bots work, along with how the realities around them have changed with these digitally mediated experiences. All these experiences are useful and must be understood at plural (Siles et al., 2020), which is why social structures and institutions must be examined in relation to these theories.
Given the inductive nature of this study, along with the grouping of user experiences into certain recurring categories, I used thematic analysis, following closely the stages described by Braun and Clarke (2006: 87): first, I familiarized myself with the data obtained, then I generated the first open codes that emerged from the interviews, and later I performed a continuous searching and reviewing the resulting themes. The process of open coding is an essential step in this analysis, given the fact that the goal is to outline “meaningful groups” (Braun and Clarke, 2006: 88).
During the interviews, I wrote various analytical memos that helped me keep track of the key concepts mentioned by the participants. All these qualitative data were collected continuously until I found that theoretical saturation was reached (Lofland et al., 2006). The most relevant empirical broad phrases were grouped by means of open coding, with the aim of highlighting important meanings from the initial data. Following closely the phrases resulting from open coding, I then made the decision to group this data based on some thematically relevant categories (Elo and Kyngäs, 2008). These categories were outlined according to the specific characteristics of axial coding, by which previously discovered codes are organized based on comprehensive conceptual dimensions (Corbin and Strauss, 2014). All submitted interviews were entered and processed in Quirkos (Turner, 2016) for a more coherent visual and thematic grouping of the data. All these themes, once grouped according to the relevant open and axial codes, were read and analyzed by two of my colleagues, who intervened with small adjacent observations but did not modify the organization of the main themes. Next, I will present the three major themes and their subsequent categories: endorsing the enemy, increasing polarization, and boosting digital activity.
Findings
Endorsing the enemy
Sarah is a 27-year-old Democrat, student from the Black community, who also has a part-time job at a media outlet. Ever since the 2016 presidential election, Sarah has tweeted a lot about Twitter bots that she believes are of Russian origin. For her, the purpose of these bots is as clear as possible:
They are the same bots that in 2016 brought Trump to power. Now they are trying to support the MAGA extremists, for a very simple reason: they will immediately stop funding Ukraine. That would be the biggest mistake that could ever happen, so it is our duty to protect our democracy.
Sarah is not the only participant tweeting about bots as a way to signal the assault on democracy. Jim is a 32-year-old Democrat who works as a software engineer. He says that the design of these bots perfectly illustrates their intentions, given that: “most bots that tweet about the war in Ukraine want to make you doubt Biden, and if you’re gullible enough, there’s a risk of falling into their trap.”
Not all US participants share the same views on who the Russian bots really support. Unlike the other two previously mentioned participants, Olivia (28 years old, Republican, writer) states that the impact of these bots on Twitter cannot be that important:
I’m not saying they [Russian bots] don’t exist . . . but I doubt they’re as numerous and powerful as it’s said . . . Anyway, you don’t really have anyone to explain these things to, given that almost everyone who discusses this is profoundly biased when it comes to American politics. When a candidate they don’t like wins, as Trump did, they’ll immediately blame it on anyone else: aliens, Russian bots, possessed objects . . .
For Olivia, these bots—if they exist—don’t encourage pro-Trump messages, but on the contrary, they encourage pro-Biden ones. Olivia says that most of the accounts posting pro-Biden tweets are “bots trying to legitimize the largest possible sums that the White House is giving to Ukrainians.” Olivia is certain that these Twitter accounts cannot represent real people, as no one in her circle of acquaintances supports Biden’s foreign policy. Thus, she is outraged by the fact that “everyone speaks about Russian bots: from CNN and New York Times, to any ordinary American user. However, why not also talk about Ukrainian bots who are most likely well-funded by Biden?”
It is not just American users posting about the multiple social realities associated with Russian bots. Steve is a 45-year-old British teacher at a high school in Birmingham. Steve considers himself a harsh critic of Brexit, and considers it “a catastrophic mistake that Boris Johnson should have eliminated, not deepened.” He also says that the attitudes of the Russian bots are quite ambivalent when it comes to Johnson’s foreign policy, and this ambivalence arises because:
Putin’s bots spam thousands of messages of support for Boris Johnson when he talks about Brexit . . . However, things changed when Johnson became a supporter of Ukraine. I might be wrong, but I’m left with the impression that the bots that once supported Johnson are now trying to dismantle him.
Given that the participants talked about how bots tend to support political people they dislike, the next theme refers to the different contexts that led to the increase in polarization between the participants in this study and their peer groups. As will be seen, such increased polarization on the subject of Russian bots led them to take measures to distance themselves from these acquaintances, that is, giving them “the reason to react,” as described by Stewart (2007: 16).
Increasing polarization
The participants included in this theme state that they have distanced themselves from certain followers on their Twitter list, because they do not share similar views when it comes to the activity of bots in the context of the invasion of Ukraine. Michael, a 22-year-old Latino student living in Oregon, says that he unfollowed about ten college colleagues because they were retweeting a lot of content from “suspicious accounts that had a picture of an American eagle on their profile, and bio a description of how true patriots must stop welcoming immigrants.” Asked why he thinks that such a Twitter profile is a bot, Michael says that “it’s a ghost account that follows over five thousand users, and is only followed back by ten or fifteen. Besides, all this account posts are tweets of Ukrainians killing civilians, and Russians saving those civilians . . . bite me! (laughs).”
Like Michael, Luciano is a 36-year-old Italian who works as a bartender. Right from the outset, he draws attention that “these bots already have a worrying popularity. Probably those who control them are well paid to make Russian propaganda, because many of my friends like or retweet some content full of lies.” Later, Luciano says that he could no longer bear to see how “the posts of Putin’s bots are redistributed by people I know from childhood,” which is why “I made a hard decision, that is, to unfollow them.” Most of the time, unfollowing others represents only an intermediate stage for a more drastic decision: that of giving up friendship with those people for good. This is the case of Walker, a 37-year-old Democrat who works as media researcher. As he states, a lot of pro-Putin content:
Was retweeted by colleagues and friends about whom I initially had a good opinion, but who showed how easily they can be fooled . . . I don’t know how it is done, but the “Christian patriots” [air quotes] believe that it is the fault of the Ukrainians because the war started. I can’t take this anymore.
Walker believes that this activity of Russian bots has an active profiling role (Bucher, 2017), in that “the very people I detest have turned out to be supporters of Putin, even after he is constantly slaughtering innocent women and children . . . I couldn’t be more satisfied than that when I blocked all these idiots [laughs].”
But not all participants gave up their Twitter friends so easily. Lena, a 26-year-old Democrat who works as journalist, believes that her experience with Russian bots was rather painful. This was happening when some of her former friends on Twitter “were sharing some obviously fake videos, in which it was said that Russian soldiers were bringing food to war-affected Ukrainians.” For Lena, life has changed significantly, as her number of friends has halved since the invasion of Ukraine began. Lena says that her political affiliation as a Democrat did not really affect her relationship with her Republican friend, but with the beginning of the war:
I could no longer ignore the differences between me and my Republican friends on Twitter, and that’s because they retweeted all those messages saying that if the US government gives financial aid to Ukraine, then that is a challenge to Russia . . . Initially, I thought that such different opinions would not affect our friendship too much, but I realized that actually nothing was the same. And I’m sure she feels the same way . . . because we don’t text or go out like we used to anymore . . . [sighs]
A similar polarization is felt by Luciano, whom I presented previously. In his case, the biggest disappointment was when he noticed two close friends talking on Twitter about the war in Ukraine using the hashtag #HaStatoPutin, an incorrectly grammatical statement that means “Putin did it.” Luciano says this hashtag is usually used ironically by those who side with Russia in this conflict. Luciano describes his disappointment this way:
It’s my fault too, because I don’t know my friends as well as I thought. My principles in life are quite simple: I don’t care what you eat, who you sleep with, or who you vote for . . . but if I see you write #HaStatoPutin, you have nothing to discuss with me!
After such an incident, Luciano says that he no longer keeps in touch with his former friends. However, unlike Lena, he has a more subtle approach: “I don’t want to unfollow them, because that would attract their attention. Either way, they’re never going to get another Twitter like from me . . . to be honest, I just hope they ignore me like I’m going to ignore them.”
These social imaginaries regarding bots show exactly how online social networks considerably affect exposure to content that amplifies polarization (Druckman et al., 2018), and this polarization is all the more visible following the 2016 political events. Such polarization caused by pieces of software leads to a result similar to that identified by Bucher (2017), that of ruined friendships. However, my results place the ruination of friendships in the sphere of participants’ agency, resulting from a conscious and voluntary decision to separate from peers who have different values and opinions about the war in Ukraine. Polarized content on social media, such as the Italian #HaStatoPutin, therefore contributes to the shaping of multiple “affinity spaces” (Dynel, 2022), which bring together opposing or complementary ways of constructing the social realities around bots.
Boosting the digital activity
The participants in this theme believe that Russian bots actively participate in the artificial numerical multiplication of different affordances on Twitter, whether they are manifested by likes, retweets, or comments. Thus, the participants of this category believe that Russian bots exercise a kind of advanced profiling techniques, through which they want to impose their domination. Mark, a 24-year-old American student and member of the Black community, believes that Russian bots constantly attack those who have the Ukrainian flag displayed on Twitter. When asked what exactly these attacks he refers to looked like, Mark replies:
. . . lots of racist comments, but which had no logic. For example, one such bot wrote to me that I should leave America because crows have no business there, and the same bot wrote to me a few days later that I am a neo-Nazi since I take the side of Ukraine . . . it’s a complete nonsense! How could I be both black and a neo-Nazi? [laughs]
A similar pattern is observed by Tony, a 37-year-old British accountant. As he states, he became the target of many offensive comments after posting several tweets in which he defended the Ukrainian refugees. However, Tony says that all these comments “do not belong to real people. Let’s be serious! Everyone knows that real people almost never use such brutal insults.” For Tony, the purpose of these bots is to discourage Western users from defending Ukraine on Twitter, a deterrence that manifests itself “through a constant spam of offensive messages, designed to make you doubt your own opinions.”
The “inflationary” activity of bots on Twitter is not only manifested in terms of comments against some users, but also through the illegitimate increase in the number of likes and followers. This is the opinion of Luc, a 33-year-old French who works in IT. He says that Russian bots not only participate in posts about the war in Ukraine but are also active in the “exaggerated increase in the popularity of far-right politicians.” Specifically, Luc details that:
Marine Le Pen’s popularity has risen considerably on Twitter ahead of the 2022 presidential election. Coincidence? I do not think so. Let’s not forget that the same thing happened with Trump’s popularity before 2016. . . The bots are simply giving tens of thousands of followers to extremists so that ordinary people will be confused and say “Hmm, if he’s so popular on Twitter, it means he’s not such a bad man.”
However, it is not just Democrats or centrist participants who consider themselves targeted by Russian Twitter bots. Klara, a 44-year-old Republican who works as a seller, says that such bots “rather tend to defend minority movements, such as Black Lives Matter or LGBTQ+, and this is true even if certain media channels say something else.” Klara believes that she was threatened by several accounts with the BLM flag on the profile, and the number of these accounts multiplied when she tweeted that Putin must pay for his crimes. For Klara “it is not very clear how many are bots and how many are real people . . . but it is clear that I was attacked by waves of aggressive comments that wanted to shut my mouth.”
However, other users believe that Russian bots are much more “refined” when selecting Twitter profiles to cause damage. In this sense, Pyke, a 29-year-old Democrat who works in market research, believes that Republicans have an advantage on Twitter precisely because of Russian bots, because:
Bots give a lot of likes to comments written by Republicans, because Republicans are defending Russia in this war . . . and as you know, the most liked comments end up being read by more people because of algorithms, and our comments will be read by no one.
Pyke believes that such a problem could be solved if the activity of bots were properly restricted, but he believes that this will not happen anytime soon because “Musk is a staunch Republican who would give anything to keep his wealth, so it’s in his best interest to keep Russian bots on Twitter to boost Republican popularity.” Thus, Democrat users like Pyke believe that social media platforms systematically disadvantage them, and this pattern is all the more visible in moments of intensive activity of Russian bots.
Without insisting on the need for a phenomenological incursion into the role that invisibilities—as discussed by Merleau-Ponty—have in the organization of our own effects, it is worth noting that users on platforms such as Twitter frequently feel the role of unseen elements (but sensed) in organizing their digital activity. More precisely, these participants assimilate the social realities related to algorithms (Bucher, 2017), which subtly organize their activity on Twitter. Also, for such users who encountered Russian bots on Twitter, the antagonism between them is done in an ideological way, through multiple strategies related to “weaponization of language” (Pascale, 2019). Therefore, it must be affirmed that the recognition of these Russian bots on Twitter corresponds to quite winding trajectories characterized by intersubjective social realities.
Discussion: encountering Russian bots as folk theories
The main functions that X/Twitter users attribute to “Russian bots”: endorsing the enemy, increasing polarization, and boosting digital activity, can be seen as relevant folk theories, for several reasons. First, these functions do not appear only as intimate imaginaries but come together with practical implications, through which users try—through both discursive and practical enactments—to elaborate different action strategies (Siles et al., 2020). More precisely, these actions reflect the constant processes of power and resistance in relation to computational entities on social media, through which users make constant efforts of identity negotiation with political bots within the digital platforms. Second, the diversity of users encourages the formation of cultured capacities (Swidler, 2001), through which they manage to perform different identities, negotiate their membership in certain social groups (Siles et al., 2020), but also to express their own values regarding social and political justice. Third, folk theories reveal constant processes of agency negotiation between users and “Russian bots” on X/Twitter, through which the political role of these bots is well internalized, while it is contested through several discursive enactments.
Based on my findings, it can be stated that Twitter users encounter these Russian bots rather as structural-computational entities, precisely due to the fact that such bots perform complex and heterogeneous functions, closely related to users’ social and cultural contexts. Therefore, this study complements previous efforts (Siles et al., 2020) to see such interactions in terms of assemblages, given the fact that these “Russian bots” participate in the process of legitimizing cultural spaces that are as different as possible. My findings thus contribute to questioning “the notion that engagement with digital media is based on and informed by a single culture” (Toff and Nielsen, 2018: 638), since these bots are perceived as being in close connection to users’ social, cultural, and political backgrounds. Through folk theories evoked by users, Russian bots also contribute to shaping what Kopelman and Frosh (2025: 4) call “algorithmic as if,” that is, those aspects of algorithmic culture that are computationally constituted and that constantly reproduce social, cultural, and political structures among social actors.
As I have shown, these three recurrent theories—endorsing the enemy, increasing polarization, and boosting digital activity—also highlight users’ cultured capacities (Swidler, 2001), given the fact that the tensions felt in relation to these structural-computational entities, although in appearance they are predominantly political, reveal in fact a cultural existence in relation to these bots. Precisely for that reason, this study lays the foundations of the comparative approach when it comes to the heterogeneous ways of encountering Russian bots on Twitter, by pointing out the cultural dimension of folk theories. Thus, further studies could also investigate other structural-computational aspects of digital platforms, but also other cultural spaces, as Siles et al. (2020) investigate folk theories in the Global South.
Perpetuating the entanglement between personal troubles and public issues, observed very well within the social realities associated with Russian bots, should give us relevant insights regarding the politics of unfinished conflict (Curtis, 2020) through the permanent opposition, mediated linguistically, between the different political affiliations. As Hardin (2003) states, most technologies are constructed in such a way that individuals have no way of knowing exactly how they work. However, this does not prevent social actors from projecting different imaginaries regarding the functionality of these technologies (Bucher, 2017), and this research exactly confirms that this pattern is successfully explained when it comes to Russian bots.
Limitations
This study has some limitations. First, this study is strictly limited to users who directly encountered “Russian bots” on Twitter through their tweets. Second, most respondents who agreed to take part in the interview are White (68%) and American (84%); thus, their interactions—albeit meaningful—with “Russian bots” are closely related to their sociocultural and structural profile. However, this study has multiple implications for communication studies and digital sociology, given the fact that it highlights exactly how the evolution of autonomous technological designs, used for political purposes, reflects different folk theories of the users who encounter such technologies.
Footnotes
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
