POPULISM, TWITTER, AND POLITICAL COMMUNICATION: an analysis of Jair Bolsonaro’s tweets during the 2018 election campaign

POPULISMO, TWITTER E COMUNICAÇÃO POLÍTICA: análise dos tweets de Jair Bolsonaro durante a campanha eleitoral de 2018

POPULISMO, TWITTER Y COMUNICACIÓN POLÍTICA: análisis de los tuits de Jair Bolsonaro durante la campaña electoral de 2018

ABSTRACT

The main objective of this paper is to understand populism and its relationship with social media, in particular, to analyze the official Twitter account of the former presidential candidate of Brazil, Jair Bolsonaro. This paper starts with a theoretical review of the definition of populism presented by classic authors such as Boyte and Riessman (1986) and Bell (1992), and by more contemporary authors such as Mudde (2004), Panizza (2005), and Müller (2016). This article also seeks to identify whether the key characteristics of populism were present in the political campaign narratives. Starting from a qualitative and quantitative analysis of content shared on Twitter, we identified populist messages in Bolsonaro’s election campaigns, particularly ones that tend to blame the elites and trigger society’s fears.

Key words
Populism; Social media; Brazilian elections; Election campaign; Jair Bolsonaro

RESUMO

O principal objetivo deste artigo é compreender o populismo e a sua relação com as redes sociais, em particular, analisar a conta oficial do Twitter do então candidato a presidente do Brasil Jair Bolsonaro. Partindo de um levantamento teórico sobre a definição do populismo, desde autores clássicos como Boyte e Riessman (1986) e Bell (1992), até outros da atualidade como Mudde (2004), Panizza (2005) e Müller (2016), o artigo procura, também, entender se as características-chave do populismo estiveram presentes nas narrativas da candidatura de Bolsonaro. Partindo de uma análise qualitativa e quantitativa do conteúdo partilhado no Twitter, demonstrou-se a presença de mensagens populistas nas campanhas deste político, em particular numa dimensão relacionada com a culpabilização das elites e o instigar dos medos da sociedade.

Palavras-chave
Populismo; Mídias sociais; Eleições brasileiras; Corrida eleitoral; Jair Bolsonaro

RESUMEN

El objetivo principal de este trabajo es comprender el populismo y su relación con las redes sociales, en particular, analizar la cuenta oficial de Twitter del entonces candidato a la presidencia de Brasil Jair Bolsonaro. Partiendo de una revisión teórica de la definición de populismo, desde autores clásicos como Boyte y Riessman (1986) y Bell (1992), hasta otros de la actualidad como Mudde (2004), Müller (2016) y Panizza (2005), el artículo busca, además, comprender si las características clave del populismo estuvieron presentes en las narrativas de la candidatura de Bolsonaro. A partir de un análisis cualitativo y cuantitativo del contenido compartido en Twitter, se demostró la presencia de mensajes populistas en las campañas de este político brasileño, particularmente en una dimensión relacionada con la culpabilización de las élites y la instigación de los miedos de la sociedad.

Palabras clave
Populismo; Redes sociales; Elecciones brasileñas; Carrera de caballos; Jair Bolsonaro

1 Introduction

Populism is not a contemporary phenomenon. It has always been analyzed by critics and social scientists since the 19th century to some degree, particularly in the context of studies on crowd psychology (Le Bon, 2001 [1896]), public behavior (Tarde, 1901), and the formation of public opinion (Lippmann, 1922, 1928, 1933). According to Lawrence Goodwyn (1981), political populism is crystallized in the Omaha Convention¹ of 1892. He states: “Equal rights for all and special privileges for none”. However, other authors argue that the rise in populism is connected to the conquest of the American West (Larson, 1986; Ostler, 1993) and the search for a policy of consensus (Key, 1967 [1961]), and even others who argue that it may have emerged out of discontent after the American civil war (Pollack, 1962) and the consequent use of the irrational (Hofstadter, 1955), or a revolt over the poor distribution of public income and power (Harris, 1973).

In this article, we are interested in understanding the relationship between populism and its current capacity on the internet, a theory that was pioneered in the late 1990s by Bruce Bimber (1998) at a time when the network was already gaining a certain relevance in the political sphere, particularly on the issue of information flow in political participation and interest groups. “The net, in the populist view, will decentralize access to communication and information, increasing citizen access. Individuals’ engagement in politics will increase, as will their influence” (Bimber, 1998, p. 139). A theory equally defended in Pippa Norris’ vicious circle concept (2000) and later in more detail in Manuel Castells’ network society (2009).

The more democratic access to information afforded by the internet has allowed populists to use it to directly address the media and to affect public opinion without going through the filter of gatekeepers (Engeser et al., 2017). Affecting public opinion is used heavily by populists since conventional politicians still seem to have a certain inability to work with social networks, according to Fitzi et al. (2019). Bartlett (2014, p. 106) complements this view well when he explains that the type of message conveyed by populists, characterized by simplifying complex problems, works well on social networks: “humor, outspokenness, pithy put-downs and catchy slogans: these are the DNA of cyber culture”.

2 Key features of populism

Before exploring the values (or lack of them) of the relationship between populism and social networks, we shall first take a brief overview of this political phenomenon. Works that study populism primarily consider it to be a phenomenon that does not have much uniting power, it is a chameleon, meaning it changes and adapts according to the regions, history, countries, and realities in which it appears (Wieviorka, 1993; Fitzi et al., 2019). What’s more, this term is contested in the political context (Panizza, 2005; Müller, 2016) and is also considered that “there is some confusion in the academy around characterizing elements of populism and those related to the host ideologies of populist parties” (Fernández-García & Luengo, 2018, p. 58). According to these authors, this non-definition results in the vague use of the term.

Populism can be a political style (Fitzi et al., 2019), a particular model of articulation of popular-democratic elements (Retamozo, 2014), or a thin ideology² that splits society into two opposed groups: the pure people and the corrupt elite, the idea being that politics should represent the general wishes of the people (Mudde, 2004). It can also be an ideology based on dualism (the good people versus the bad elites) and anti-pluralism (Matthews-Ferrero, 2018), a concept (and not an ideology) used by politicians to promote an opposition policy, according to Tormey (2019), it is the lack of an intellectual origin and the fact that not all populists have the same values that rule out the possibility of this being an ideology. For Laclau (2005, p. 11) “populism is simply a way of constructing the political”. In his controversial work “Populist Reason”, Laclau (2005) analyzes the formation of collective identities to affirm that populism “does not have any referential unity because it is not attributed to a demarcated phenomenon, but a social logic, the effects of which crosses over a variety of phenomena” (p. 11).

The division of society into two antagonistic groups, the meeting point for theories about populism, is part of the nature of this phenomenon, “because it is through antagonism that political identities are constituted, and radical alternatives to the existing order can be imagined” (Panizza, 2005, p. 28). The author adds that the meaning of populism is about the questioning of institutional order. In what is considered a more flexible approach, Galito (2018) states that populism can be both an ideology and discourse or political strategy. She claims that populism is always directed against the ruling elite and against the direction the country is taking, regardless of what that direction may be.

Müller (2016, p. 19), however, takes an even more distinctive view of populism, defining it as a particular moralistic imagination of politics where the other politicians represent an immoral and corrupt elite. Müller is one of several authors who consider the movement to be anti-elitist (provided that this elite is not itself) and anti-pluralist but admits that this is a condition for populism and not something that determines whether a certain politician is populist or not. Regarding the anti-pluralist aspect of populism, it is important to remember that it is not unchanging or unwavering; the pluralities that the particular movement opposes, whether based on culture, religion, ethnicity, or geography, have relation to the autochthonous factor, far beyond an ideology. For this author, what differentiates a populist politician from a democratic one is that the former sees him or herself as representing an absolute truth, while the latter accepts the debate, the possibility of being contested by institutions, models, etc. Also, according to Müller, populism is “a permanent shadow of representative politics” (Müller, 2016, p. 102).

According to Boyte and Riessman (1986), this notion of people is directly linked to the place in which these people live and, consequently, to their “legends, oral traditions, folk remedies, foods, rituals, music, ways of speaking, and remembering. A people has a moment of origin and birth, tales of evolution, spaces that it has filled in and become attached to” (p. 8). Panizza (2005), however, explains that the people – or the imaginary “real people” – is composed of those considered to be oppressed by the ones in power (or in this case, the elites). At the same time, the elites are the groups that consider themselves superior to others. Bell (1992), explains that the concept of elitism, in the context of populism, is related to the fact that these elites believe they are capable of making better decisions on issues that affect not only them but the entire political-social spectrum. According to this same author, there are at least seven categories of elites: knowledge, monetary, power, expertise, interest groups, professional segments, and internal elites. Instead of trying to give so many examples of so many types of elites, it is easier to understand them as groups at the top of a chain of hierarchy. These groups have decision-making powers and use them for their influence. Thus, they can be politicians, presidents of associations, judges, intellectuals from different fields, journalists, leaders of cultural, religious, and economic groups.

In addition to the number of differing views on what the definition of the concept of populism is, there is also a great variety of characteristics and key elements associated with it, most of which tend to revolve around two central points: anti-elitism and/or anti-pluralism. For example, Wieviorka (1993) points out some features to help define it. The first point is a feeling of separation between the public and the political and economic power, which in this case is associated with the so-called intellectual elites. Wieviorka goes on to say that populism speaks on behalf of the people and of a homogeneous community, as well as looks to the past for a tradition of national identity. “We need to understand that populism is trying to reconcile the past and the future, identity, and change, which makes it an objective that quickly reveals itself as contradictory...” (Wieviorka, 1993, pp. 95–96).

Populism speaking on behalf of the people can be controversial. This is because populists do not differentiate between individual opinions and those of the masses. It is a particular way for politicians to interpret the world (Matthews-Ferrero, 2018). Furthermore, representatives of this phenomenon see the public as singular, thus ignoring the pluralism and differences of opinion that society has within it (Tormey, 2019).

Given the historical background and the profile of populist politicians, what are some of the political characteristics they share? Engesser et al. (2017, p. 1.109) argue that there are five key ideological elements of populism: giving prominence to the sovereignty of the people, advocating for the people, attacking the elites, ostracizing the others, and looking back to the heartland”³. Tormey (2019) also adds opposing the political establishment, a leader with exceptional powers, and using direct language to these characteristics. Müller (2016), even though he agrees with the characteristics presented by these authors, stresses that a populist does not necessarily have all (or only) these attributes.

Populists often consider themselves the only ones capable of restoring the sovereignty of the people, of taking power away from the elites and other institutions. Another feature within the key elements developed by Engesser et al. (2017) is that populist politicians usually maintain a close relationship with people and place popular demands above any other issue in society. This results in populists excluding who the authors call “the others”, most often referring to immigrants, ethnic minorities, and religious groups. However, this viewpoint must be expanded according to the context in which these groups are inserted, after all, not all countries with supposed populist leaders face the same challenges. On the other hand, Galito (2018) adds that populist parties are characterized by their revolt against power and their defense of political equality and the supremacy of the people.

The transitive nature of populism, between the left and the right, comes from the chameleon-like power that it holds. Panizza (2005, p. 205) states that “populist politics allows for more than one political form depending on the ideological elements that compose them and the way they are socially constructed”. He concludes by saying that populism can be both authoritarian and nationalist, as well as a civil and popular democracy.

Left-wing populists, for instance, hold core beliefs that tend to be shared by the political parties that champion this ideology, usually based on the critique of social inequality and the search for a new model of society. In her work “For a Left Populism”, Chantal Mouffe (2019) highlights the features of equality and social justice in her support of left populism, which she believes is as an opportunity in the current “populist moment” that we live in to combat the xenophobic politics of right-wing populism. On the other hand, right-wing populists have anti-immigrant and anti-cosmopolitan ideals, they are against the media elite, they see cultural pluralism as a threat to indigenous values (religious, racial and national), and are also critical of inequality, although they understand it from a commonly romanticized view of the past (Bartlett, 2014; Fernández-Gracia & Luengo, 2018; Fitzi et al., 2019; Tormey, 2019). For both viewpoints, populism is about fighting the elites.

Of the many forms of populism, Norris and Inglehart (2019) in their work “Cultural Backlash” consider authoritarian populism to be “the most serious risk for liberal democracy (…) allowing strongman leaders claiming to speak for the people to step into the vacuum, while simultaneously endorsing social intolerance toward out-groups” (p. 65). They describe the characteristics of this authoritarian populism as a reaction against the silent revolution of values, economic grievances, and immigration.

So, what makes the populism movement inherent to the ideology? Populism is characterized by how it offers the solution to different problems: being against the political establishment: populists versus the dominant politics. Populists see themselves as the honorable (and only) defenders of the truth (Bartlett, 2014).

The table below (table 1) presents the key characteristics of right-wing populism as listed in this study. The characteristics converge between different authors and sometimes receive different nomenclatures, but we chose to use the terminology that best encompasses other corresponding characteristics. The table is a reference for our case study in which we compare these characteristics with the content published on the official Twitter account of our object of study: Jair Bolsonaro.

In addition to the framework, we considered two pre-conditions of populism when elaborating Bolsonaro’s profile, they are: 1) the fact that the politician is (or claims to be) an outsider of the political establishment and 2) a scenario of crisis/discontent with the current policy.

Table 1
Key characteristics of populism

Characteristic	Framework (what it represents)
Personify the people’s pain	Placing oneself as the sole moral spokesperson/defender of the “people’s” repressed interests. Representing the desire of a supposed majority, to rescue what was taken by the elites; discrediting plurality.
To be antipluralist	To suppress, ignore or devalue interests and opinions that do not conform to one’s own, whether in political or sociocultural terms.   Ostracizing the “others”. The “others” tend to be minorities or people of non-indigenous origin such as immigrants; also includes people who hold other sexual or religious beliefs or people from other ethnic or cultural backgrounds.
Blame the elites	Opposition politics, being contrary, antagonistic to the political agents of the current system, media, and institutions.
Trigger society’s fears	Nostalgically motivated defense of a morally ideal society, to regain the supposedly lost values of the nation. It is not necessarily linked to nationalism, but it opposes the sociocultural direction the country is taking.

• Source: authors’ own, based on the following authors: Bartlett (2014); Bell (1992); Boyte and Riessman (1986); Engesser et al. (2017); Ernst et al. (2017); Fernández-García and Luengo (2018); Fitzi et al. (2019); Galito (2018); Matthews-Ferrero (2018); Mudde (2004); Müller (2016); Panizza (2005); Retamozo (2014); Tormey (2019); Wieviorka (1993).

3 Populism, journalism, and social networks

The operating logic of the internet differs from traditional media, not only because of the interactive way in which it involves the public but also because of the flow of information and the fact that people are able to access what they choose (Bimber & Davis, 2003). What’s more, these authors claim that the internet offers politicians “the opportunity to bypass journalists and speak directly to voters, while maintaining control of the message in the process” (p. 21). For Levy (2002), the fact that the control of the message sidesteps the filter of journalism and is given to the issuing agent is one of the characteristics of cyberdemocracy, a result of the expansion of the public sphere. Teixeira (2018, p. 11), in turn, states that the internet is no longer just a more interactive, impersonal, plural, fast, and anonymous means of communication, it has become a path to what he defines as a “new sociopolitical dimension” in which “the rise of new social and political dimensions becomes a real possibility”.

Twitter is a network with important added values for both gatekeepers and the primary sources of political information; in other words, the parties, politicians, and public agents themselves. According to a study carried out by Parmelee (2013), political journalists from North American newspapers regularly use Twitter as part of their work process and consider this tool to be more important than Facebook or any other social network in their work routine. The Twiplomacy study 2018⁶, carried out by communication and public relations consultancy Burson Cohn & Wolfe, shows that 97% of the United Nations (UN) member states are currently active on Twitter, which includes 372 personal accounts (heads of state or foreign ministers) and 579 institutional accounts. These numbers make Twitter the most popular social network among government officials, even ahead of Facebook, with 93% of UN member states active on this network.

Aggio (2011) understands that this platform can function as a practical means to efficiently disseminate campaign information in addition to being a potential mobilizing platform for the public in digital and physical environments. In line with this thought, Highfield (2016) says that using the platform this way depends on a variety of personal and partisan factors but emphasizes that Twitter can serve as a platform to broadcast content from elected representatives to voters, thus facilitating this kind of dialogue.

According to Parmelee and Bichard (2012), publishing on Twitter, or tweeting, has several purposes for politicians, the main one being to communicate directly with the public without being filtered by journalists. For these authors, in terms of campaigns, politicians who successfully communicate through this social network are more likely to influence the public than they would through traditional media. “Leaders use social media to appear transparent to their voters, to explain votes and answer questions. Campaigns use social media to influence supporters to do free advertising” (p. 19).

Twitter’s influence on the public and its power to bypass the filter of journalists are aspects also addressed by López-Meri et al. (2017). According to these authors, this tool encourages voter participation and establishes a two-way form of communication between political agents and the public. These authors further explain that the possibility of performing what they call “mass self-communication”, meaning without having to go through the intermediation of journalists and the media, is one of the main uses of Twitter in politics.

Ernst et al. (2017) also add that social networks offer other characteristics that are extremely compatible with populist communication such as its large potential for personalized messages and the ability to direct those messages to specific groups. However, Fitzi et al. (2019) compare the idea of supposed direct communication with people (bypassing the journalists’ filters) to the frankness and/or authenticity with which populists express themselves when communicating on social networks. First, the authors argue that mediation does exist in these networks, although it is different from what is observed in the mass media with algorithms or limitations of the platforms.

Due to its operating logic, Twitter has a large potential for resonance. Users not only see what they follow but also what other users like or re-publish from other accounts, called “retweets”. This logic may be what leads Moore (2019) to point to the need for political communication to have a personality and to stand out, especially among the so-called digital natives. “In the politics that unfold on Twitter, the one who gets the most attention – and has the most influence – is the one who makes controversial statements, discredits the status quo, hurls personal insults, and gets into arguments” (p. 202).

Despite the indications about the positive relationship between populism and the internet, especially with the use of social networks, Fitzi et al. (2019, p. 31) argue that this is not a uniform fact and as such cannot be characteristic of the movement. “Even though social media has arguably reshaped politics – and that includes populism – there are good reasons to have reservations about the supposed relationship between populism and social media”. Cavassana de Carvalho and Urizzi Cervi (2018) conducted a comprehensive bibliographic review and consider that this relationship between populism and social networks does exist, and although it empowers and brings citizens closer to politicians “only regulated and therefore institutionalized environments tend to guarantee real political results and effects” (p. 98).

More recently, Letícia Cesarino (2020, p. 442) demonstrated that the “contemporary obsession with authenticity and spontaneity in politics can increase the anxiety of media coverage” resulting in a “perfect neoliberal storm”. Rodríguez-Andrés (2018) lists five factors that led to Trump winning the American election in 2016: the personal aspect of the tweets (Trump himself published them); the naturalness and spontaneity of his tweets; his break from the established conventions of political communication on Twitter; victimization against the system or establishment; and finally, the identification of an enemy who also frightened or intimidated his followers.

In an analysis carried out on Brazil, Aggio (2011) noticed that the 2010 presidential election (the first one where Twitter and other digital platforms were regulated by the country’s electoral legislation) had 15 categories for the tweets published by the presidential candidates at the time. We reduced the number of categories to 11 as we excluded some of the ones that Aggio had listed because they include, or do not include external links. Our 11 categories are the dissemination of the agenda; dissemination of campaign materials, interviews, and debates; mobilization; comment on events; interaction with comments; message replication; proposals of candidates; information about candidates’ past political achievements; attacks on opponents; opinion polls; and others, which the author classifies as “trivial”.

The following table summarizes the general aspects of a political campaign, comparing Aggio’s survey (2011) with classic literature on election campaigns from such authors as Stanley Kelly (1956, 1983), Theodore H. White (1964), or even Melvyn H. Bloom (1973), including more recent authors who study social networks like Twitter, such as Biju (2019), Taras and Davis (2019), Urman (2019), or Reimer et al. (2018).

Table 2
General aspects of a political campaign

Characteristic	Definition
Government proposals and candidate profile	Positions, ideas, and topics included in the proposed government program and/or introduce the candidate.
Acknowledgments	Response to demonstrations and visits by supporters.
Campaign actions	Organizing and participating in committees and popular mobilizations.
International relations	Dialogue with leaders/politicians from other countries, and topics of international interest.
Dissemination of personal material or third-party material on other media and/or formats	News, reports, and videos which do not include enough text to analyze.
Others	Separate images and videos, emojis, social media marks, and message greetings (good morning, good afternoon, good night).

• Source : authors own.

4 Case study: analysis of Jair Bolsonaro tweets during 2018 election campaign

4.1 Objective, corpus, and methodology

The objective of this article is to look at the key characteristics of populism (shown in table 1) and identify whether populist narratives were used on Jair Bolsonaro’s Twitter account during the 2018 election campaign (@jairbolsonaro). The analysis was conducted on the entire 2018 Brazilian presidential election: from August 16 to October 5, 2018 (first round of the election) and from October 12 to 28, 2018 (second round of the election). After this analysis, we then went on to analyze to what extent populism contributed (or not) to the efficiency of Bolsonaro’s campaign on this social network.

In terms of the methodology, after collecting data and texts published on the internet, we followed the quantitative and qualitative content analysis proposed by Daymon and Holloway (2002) since hypertext has particularities that the researcher cannot overlook, such as the temporary and concise nature of its content.

The tweets from the @jairbolsonaro accounts were collected one year after the election had ended, or, in October 2019. From a quantitative point of view, the total number of tweets was collected first, and then the daily number of publications was identified to determine the frequency with which Brazilian politicians used the social network. We also identified the number of tweets published by Bolsonaro that fit into the key characteristics of populism and divided them into categories. Replies to other users’ tweets and tweets containing grammatical errors were excluded from this analysis. For comparison purposes, table 2 was added to the quantitative and qualitative analysis in order to understand how Bolsonaro’s non-populist tweets are divided, in other words, the tweets that do not contain the key characteristics of populism. Lastly, the quantitative analysis was completed by calculating the interactions in the tweets (measured by the sum of comments + likes + retweets)⁴. This mathematical determination of the interactions was conducted to verify whether tweets on topics of populism had greater circulation and repercussion on the social network than those that did not contain this topic.

4.2 Results

4.2.1. Bolsonaro’s campaign on Twitter

During the election period, Bolsonaro released 484 publications on Twitter, 305 of which came in the first round, and 179 in the second round. The daily average, considering the 66 campaign days (51 days in the first round and 15 in the second), was 7.33 tweets per day.

Throughout his campaign on Twitter, Jair Bolsonaro’s account had some noticeable peaks and drops in terms of the daily number of tweets published. The biggest spike in publications observed in the first round (19 tweets) occurred on the Sunday before the first-round vote (09/30/2018), while the other big spike (20 tweets) occurred on the Sunday before the second-round vote (10/21/2018). The high number of publications in both cases was basically due to the retweeting of support messages. Figure 1 shows one of the most recurrent publications on these aforementioned dates, the account just identifies the location and gives thanks for the expression of support it received.

Figure 1
Tweet from 09/30/2018, during the first round of elections

On the other hand, the days that registered the least amount of participation on Twitter were also days on which Bolsonaro was attacked by a man wielding a knife during one of his campaigns in the city of Juiz de Fora, in the state of Minas Gerais. Only one post was made to his account on the day of this attack and the following day. The lowest average of publications during the entire election period (3.6 tweets per day) was recorded in the days following the attack (from 09/07/2018 to 09/21/2018).

4.2.2. Populist elements in Bolsonaro’s Twitter campaign

The collected data indicate that 176⁵ of the 484 tweets (or 36.36% of all publications) observed during the campaign period for the @jairbolsonaro account contain some of the four key characteristics of populism. Out of all these 176 publications, 42 deal with the key characteristics of the political phenomenon and the general aspects of a campaign. The example in figure 2 shows a tweet that, when decontextualized, appears to take on a literal meaning: Bolsonaro’s pledge to fight organized crime. However, this publication is also an attack on the elites, and in this case, it is a clear allusion to the opposition Workers Party (PT) and to the former president of Brazil Luís Inácio Lula da Silva (imprisoned at the time) and presidential candidate Fernando Haddad, referring to them as “employees”.

Figure 2
Tweet from 10/12/2018, during the second round of elections

Tweets like the one in figure 2 were repeated numerous times. There are others where Bolsonaro labels the PT a “criminal faction” and “red bandits” and refers to Lula as an “imprisoned leader”, using the hashtag #LulaPresidiário.

The attack, or the blaming of the elites, is the main characteristic of populism most often used by Bolsonaro on his Twitter account. A total of 122 publications contained this theme throughout the campaign (55 in the first round and 67 in the second). Other key characteristics (listed in table 1) are also present in the @jairbolsonaro tweets, although to a lesser extent. The other themes of populism addressed throughout the campaign, listed in descending order, were triggering society’s fears (33), personifying the people’s pain (22), and being anti-pluralist (13). It is important to point out that out of the total number of tweets with a populist theme, 14 of them are tweets that contain more than one characteristic of populism. The most notable of these is the personification of the people’s pain added to the blaming of the elites in a single tweet, which occurred on five occasions. It’s not hard to see why, it’s almost like simple math: use the establishment against the people. It is like saying that the current problem has a solution, that solution being the populist, who is responsible for putting an end to the social disorder.

Figure 3
Chain of tweets from 09/15/2018, during the first round of elections, combine blaming the elites with personifying the people’s pain

Although there were fewer of them, the key characteristics of populism (apart from just blaming the elites) played an important part in the first round of elections. To understand this movement, we need to briefly look back at the model of the Brazilian presidential election.

In 2018, there were a total of 13 candidates who ran in the Brazilian presidential elections. The first round involves candidates participating in various debates where they present their proposals, ideas, and solutions to the problems faced by the country. In the second round, the choice focuses on two figures who, in theory, present their plans and how they intend to govern the country for the next four years. Contrary to the norm, Bolsonaro was absent from the debates and only participated in one which was held by Rede Bandeirantes in the first round. The reason given for Bolsonaro’s absence was the state of his health; he had just recently suffered an attack on September 6. But it was also a strategy.

In the context of the Brazilian election, we can see how Bolsonaro’s use of populist themes became more immediate from the first round to the second. In the first round, his tweets were more diverse, that is, they addressed several key characteristics of populism; however, in the second round, the tweets tended to focus more on just blaming the elites. What’s more, the first round of elections had a greater number of political elites involved (parties and candidates). Once the second round began, and the dispute was just between Bolsonaro and Haddad, his attacks on Haddad became more direct and numerous. The graph in figure 4 demonstrates this.

Figure 4
Populist publications from the @jairbolsonaro account

Before looking at the details of Bolsonaro’s blaming of the elite, we have to take into account the other populist themes used in the campaign as they are often interconnected. The narrative of the tweets belonging to the trigger society’s fears category describes a chaotic landscape in Brazil, from the threat of freedom and sovereignty to the rescue of order and progress. On a number of occasions, Bolsonaro alluded to ideological indoctrination, one that is responsible for robbing the country at a sociocultural level. This inevitably was channeled into an attack on the elites, this same type of attack is also used to blame the elites for defending the supposed interests of the people.

Lastly, violence and corruption are two themes that are also quite present in triggering society’s fears or personifying the people’s pain. In 2018, the year of the election, the country was ranked 105^th in the Corruption Perception Index (CPI)⁷, published by the non-governmental organization International Transparency, which ranks 198 countries by their perceived level of corruption in the public sector. Brazil faced two crises in the previous two governments that undermined the credibility of political agents: the Mensalão scandal in 2005, and the Lava Jato (an expression that, in English, means Car Wash) scandal in 2016, which culminated in the trial, conviction, and/or imprisonment of many individuals from the political class, including stakeholders, deputies, ministers, and even a former president.

By blaming the elites, Bolsonaro was targeting a wide variety of opponents, but he did tend to focus his attention on certain ones. The targets of more frequent attacks changed as the campaign progressed and did not necessarily focus on any one specific party, at least initially. For instance, in the first round, the media and the press were the ones who Bolsonaro most targeted, accusing them of harassment and spreading fake news, he even personally offended professionals in the field of journalism. The first round had a total of 18 tweets from the blaming the elites category that were specifically directed toward the media.

Figure 5
Tweet from 09/28/2018 explicitly attacking the media during the first round of elections

Continuing with the blaming of the elites during the first round, Bolsonaro presented a constant state of war against the political class and against his opponents in general, whom he recurrently labeled as being corrupt and cowardly, calling them scoundrels and hypocrites. By doing this, Bolsonaro made himself appear like an outsider on the political spectrum, a common characteristic among populists (Panizza, 2005; Tormey, 2019), and that he could put an end to one of the issues he most complained about: corruption.

Figure 6
Tweet from 08/17/2018, the first round of elections, where Bolsonaro does not call out any politician or political party in particular but instead blames the establishment

However, by openly naming his opponents, his attacks were clearly being directed toward the PT (the party of the candidate he would compete against in the second round) and toward the PSDB, the party of candidate Geraldo Alckmin, who received the most television time and, at the time, was considered one of Bolsonaro’s main opponents. Past elections saw PSDB candidates competing with PT candidates in the second round, but the former never were successful.

Figure 7
Tweet from 09/06/2018, the first round of elections, showing Bolsonaro responding to Alckmin by retweeting on his official Twitter account

In the second round, when there were no other direct opponents other than the PT, the blaming of the elites became more standardized. There are no tweets about the PSDB and very few about the political class in general (10). Even the attacks on the press, which were most prevalent in the first round, were slowing down (7) compared to the tweets directed toward the PT, Fernando Haddad (the opposing presidential candidate), and Lula (50).

Of note is the fact that Bolsonaro’s tweets in the second round began to be addressed directly at Haddad, which he had only done once before during the first round. When talking about his opponent Bolsonaro usually preferred not to use Haddad’s proper name, instead, he opted to use a series of pejoratives such as “puppet of the corrupt”, “employee of the corrupt”, “under the thumb”, “creator of the gay kit” and “Andrade”, this last reference relating to a meme which was circulating at the time.

Figure 8
Tweet from 10/16/2018, the second round of elections, this time showing Bolsonaro’s retweet on Haddad’s official Twitter account

The 18 tweets attacking the press in the first round represented 32.7% of the total publications in the category of blaming the elites. However, this percentage dropped slightly by the end of the campaign. Adding the first and second rounds together, this category represented 20.5% of the total publications. On the other hand, even though attacks against the PT and its candidates in the first round were marginal (20% of the total tweets), the absolute predominance of attacks against them in the second round (74.6%) made it so that exactly half of the tweets from the @jairbolsonaro account were directed toward blaming the elites.

Figure 9
The different elites attacked by Bolsonaro and the change according to election round

4.2.3. Bolsonaro’s campaign beyond populism

If the populist theme represented a little more than 36% of Bolsonaro’s electoral campaign on his @jairbolsonaro account, what were the other themes that he addressed during that period? There were three main themes: presenting proposals (80); disseminating personal campaign material or from third parties, but not on Twitter (101); and finally, massive dissemination of manifestations in his support and his appreciation of them (103).

Some of Bolsonaro’s political proposals converged on the populist issue (28), such as the publication shown in figure 2. These were recurrent but were not the only issues he addressed. Bolsonaro also addressed issues such as education, public safety, fighting corruption, and the economy. Even still, he never really discussed these proposals in-depth, and ended up sounding more like abstract ideas, for example: “change the direction of education in Brazil”. When he wanted to talk more about these issues, Bolsonaro often released videos of interviews and links to news reports and live feeds on other social networks (Facebook and YouTube).

Figure 10
Tweet from 09/06/2018, in the first round of elections, urging followers to go outside the social network to find out more information on the issue he is presenting

Another fact about his Twitter campaign that deserves mentioning is the considerable decrease in the number of demonstrations of support, institutional visits, and trips across Brazil after the attack Bolsonaro suffered on September 7, 2018. In total, there were only 38 tweets about campaign actions, but what is interesting is that 19 of these occurred before he was attacked, 22 days into his campaign. The other 19 tweets were published during the remaining 44 days.

Apart from the populist theme, a quarter of Bolsonaro’s Twitter account space was used to thank third-party supporters (27.2%), while his proposals occupied even less space, representing just over 21% of his tweets. Figure 11 demonstrates this information.

Figure 11
Bolsonaro’s campaign outside of populism

4.2.4. Comparative results from Bolsonaro’s campaign

Jair Bolsonaro was elected president of Brazil after receiving the most votes in the first and second rounds. Given this context, it is important to understand which issues led Bolsonaro to be more noticed, commented on, republished, and liked. Were they general aspects of a campaign or populism? Did populism play a crucial role in this process?

As we have seen, the campaign on the @jairbolsonaro account covered several issues, although there was a predominant one. There were 176 tweets on populism, which represents just over 36% of all Twitter posts over the 66 campaign days in our analysis. The other 308 publications on the account did not contain populist elements. Which ones had the most interactions among followers?

In total, the @jairbolsonaro account had 15.979.736 interactions with the 484 tweets (note: this number varies if you add the non-populist and populist tweets together). This total number of interactions is divided between 12.678.477 likes, 2.786.949 retweets, and 514.310 replies. Thus, the average number of interactions per tweet in our analysis period was 33.016. The first analysis of this total number of interactions is due to the relationship between the total number of interactions with non-populist (8.389.933) and populist (7.589.803) tweets. As we can see in the graphs in figure 12, although there is a greater number of tweets with a non-populist theme, this difference drops considerably compared to the number of non-populist interactions versus populist interactions. Even though there are 43% fewer populist tweets, the number of interactions is almost equal.

Figure 12
On the left, the number of tweets; on the right, the number of interactions

As a result, we need to establish another parameter for comparison: the average number of interactions between populist and non-populist tweets. As illustrated in figure 12, populist tweets are more expressive regarding the number of interactions: they catch the attention of followers and are most likely to be republished, liked, and commented on – and not only by followers of the account but also by those who oppose it, as they too respond and comment on the tweet. After all, this is one of the particularities of this social network’s audience, as noted by Weller et al. (2014). Populist publications receive an average of 43.124 interactions, a much higher number than the 27.240 for non-populist publications.

Lastly, it is interesting to note how the interaction between the four themes of populism is divided. Listed in ascending order, these were the key features that received the most interactions: blaming the elites (4.826.471); triggering society’s fears (1.268.662); personifying the people’s pain (839.366), and being anti-pluralist (655.304).

5 Conclusions

Contrary to social and political movements that try to persuade voters to join or support their goals, populist rhetoric does not seek to convert the public or even to indoctrinate it with a cause. The populist narrative has a more cathartic purpose to it and seeks only to confirm what the public already knows or believes to be true; the experience described by a populist, similar to how horoscopes work, always meets the audience’s frustrations or expectations. Populism is therefore not a political ideology, at best, it is the fastest way to glorify simplistic proposals that almost always run over democracy and human rights. A politics of opportunity, opportunism, and convenience headed by protagonists who, using a colloquial tone, attack the elites and create a supposed common enemy and an audience of peers who do not question the simplicity of the electoral promises.

The populist tweets during Jair Bolsonaro’s candidacy had an average interaction rate of 1.6 times greater than non-populist publications, in other words, the populist theme generated a much greater result for Bolsonaro. Even though populist tweets make up 36.36% of the total number of publications, the relevance of these tweets is notable as they were responsible for 47.5% of all interactions. The key characteristics of populism were present on almost a daily basis on Bolsonaro’s Twitter posts and, as the growing number of interactions showed us, they were important in gaining the attention, anger, and paradoxical appreciation of users of this social network.

However, an extensive analysis of one of the most prominent names in populism today⁸ has shown us that populism grows in fragile environments and uses political instability and popular dissatisfaction in order to thrive. And it is the establishment that helps the populist gain notoriety. In addition to blaming the elites, the large-scale projection of negative narratives triggering society’s fears in Bolsonaro’s campaign was evident, not to mention the anti-pluralism and the personification of the people’s pain (although these last two characteristics were analyzed on a smaller scale).

Despite this landscape, we cannot confirm whether there is a rise of populism in the age of social networks as there is no pre-social network parameter to which we can compare populism. What we can see, however, is that Twitter is an open field of populist action; it is a rapid, vast social network with no mediation or journalistic filter. Twitter values originality, bravery, fearlessness, authenticity, and other attributes of populism, but it can also be perplexing, emphatic, and assertive, all within a certain timing and a good dose of pro-activity and consideration in the responses; this is how to obtain the best experience from this network, which operates almost as a real-time commentary on what is happening outside the virtual environment.

The populist narrative dialogues very well with Twitter, as demonstrated here in the comparison between the average number of interactions between populist and non-populist tweets. But the problem is not simply about being populist, but how populism is used to offend, discredit, deceive, oppress and distort public facts.

NOTES

• 1
  “The People’s Party”, commonly referred to as the “Populist Party”, held its first convention in Omaha, Nebraska, USA, in 1892. The party elected James K. Weaver as president and ratified the Omaha Platform, which included proposals for the graduated income tax, secret ballot, direct election of U.S. senators, the eight-hour workday, and other reform measures (Goodwyn, 1981).
• 2
  The “thin ideology” concept is defended by several authors and is related to the fact that populism is combinable with other political ideologies (socialism, nationalism, communism, etc.) as it is open to the left or the right and operates more as a discursive strategy than a concrete proposal.
• 3
  The idea of “looking back to the heartland” can be better understood as a nostalgic look at the supposed values of a country.
• 4
  Retrieved from https://twiplomacy.com/blog/twiplomacy-study-2018/
• 5
  All tweets were characterized and categorized with the name in the format: day-month-year-number. In the case of threads and dialogues, each case was evaluated separately: tweets that did not make sense individually were added to those that make sense, while the others were differentiated.
• 6
  Number of individual tweets. If the intersections between populist publications are included, as shown in figure 5, this total reaches 190.
• 7
  Retrieved from www.transparency.org/en/cpi/2018/results/bra
• 8
  According to article in The Guardian, May 1, 2019, written by Dom Phillips, Jason Burke and Paul Lewis entitled “How Brazil and South Africa became the world’s most populist countries”, retrieved from www.theguardian.com/world/2019/may/01/how-brazil-and-south-africa-became-the-worlds-most-populist-countries

• TRANSLATED BY: LEE SHARP
• One review used in the evaluation of this article can be accessed at: https://osf.io/wdpnf/ Following BJR’s open science policy, the reviewer has authorized this publication and the disclosure of his/her name.

REFERENCES

• Aggio, C. (2011). As campanhas políticas no Twitter: Uma análise do padrão de comunicação política dos três principais candidatos à presidência do Brasil em 2010. Proceedings of the IV Encontro da Compolítica. Rio de Janeiro. Retrieved from www.compolitica.org/home/wp-content/uploads/2011/03/AGGIO-Camilo.pdf
  » www.compolitica.org/home/wp-content/uploads/2011/03/AGGIO-Camilo.pdf
• Bartlett, J. (2014). Populism, Social Media and Democratic Strain. In C. Sandelind (Ed.), European Populism and Winning the Immigration Debate (pp. 99–114). Retrieved from https://eprints.whiterose.ac.uk/104297/1/Korr%20igen.pdf
  » https://eprints.whiterose.ac.uk/104297/1/Korr%20igen.pdf
• Bell, J. (1992). Populism and Elitism, politics in the Age of Equality Washington, DC: Regnery Gateway.
• Biju, P. R. (2019). Political internet: State and Politics in the Age of Social Media Routledge: London.
• Bimber, B. (1998). The Internet and Political Transformation: Populism, Community, and Accelerated Pluralism. Polity, 31(1), 133–160. DOI: 10.2307/3235370
  » https://doi.org/10.2307/3235370
• Bimber, B., & Davis, R. (2003). Campaigning Online: The Internet in U.S. Elections New York: Oxford University Press.
• Bloom, M. H. (1973). Public Relations and Presidential Campaigns: A crisis in democracy New York: Thomas Y. Crowell Company.
• Boyte, H. C., & Riessman, F. (1986). The New Populism: The Politics of Empowerment Philadelphia: Temple University Press.
• Castells, M. (2009). Communication Power Oxford: Oxford University Press.
• Cavassana de Carvalho, F., & Urizzi Cervi, E. (2018). Mais Populismo, Menos Representatividade: monitoramento e lógica populista da comunicação política em redes sociais online. Revista Estudos Políticos, 9(1), 86–103. DOI: 10.22409/rep.v9i17.39852
  » https://doi.org/10.22409/rep.v9i17.39852
• Cesarino, L. (2020). Como as mídias sociais proporcionam uma política populista: observações sobre liminaridade com base no caso brasileiro. Trabalhos em Linguística Aplicada, 59(1), 404–427. DOI: 10.1590/01031813686191620200410
  » https://doi.org/10.1590/01031813686191620200410
• Daymon, C., & Holloway, I. (2002). Qualitative research methods in public relations and marketing communications London and New York: Routledge.
• Engesser, S., Ernst, N., Esser, F., & Büchel, F. (2017). Populism and social media: how politicians spread a fragmented ideology, Information. Communication & Society, 20(8), 1109–1126. DOI: 10.1080/1369118X.2016.1207697
  » https://doi.org/10.1080/1369118X.2016.1207697
• Ernst, N., Engesser, S., Büchel, F., Blassnig, S., & Esser, F. (2017). Extreme Parties and populism: an Analysis of Facebook and Twitter across six countries, Information. Communication & Society, 20(9), 1347–1364. DOI: 10.1080/1369118X.2017.1329333
  » https://doi.org/10.1080/1369118X.2017.1329333
• Fernández-García, B., & Luengo, Ó. G. (2018). Populist Parties in Western Europe. An analysis of the Three Core Elements of Populism. Communication & Society, 31(3), 57–76. DOI: 10.15581/003.31.3.57-74
  » https://doi.org/10.15581/003.31.3.57-74
• Fitzi, G., Mackert, J., & Turner, B. S. (2019). Populism and the Crisis of Democracy, Politics, Social Movements and Extremism (Vol. 2). New York: Routledge.
• Galito, M. S. (2018). Populism as a political phenomenon. JANUS.NET E-journal of International Relations, 9(1), 53–69. DOI:10.26619/1647-7251.9.1.4
  » https://doi.org/10.26619/1647-7251.9.1.4
• Goodwyn, L. (1981). The Populist moment: a short history of the agrarian revolt in America New York: Oxford University Press.
• Harris, F. R. (1973). The new populism New York: Saturday Review Press.
• Highfield, T. (2016). Social Media and Everyday Politics Cambridge: Polity Press.
• Hofstadter, R. (1955). The Age of Reform: from Bryan to F.D.R New York: Vintage Books.
• Kelly, S. (1956). Political Campaigning. Problems in Creating an informed electorate Washington: The Brookings Institution.
• Kelly, S. (1983). Interpreting Elections Princeton: Princeton University Press.
• Key, V. O. (1967). Public Opinion and American Democracy (4ª ed.). New York: Alfred A. Knopf. [1961]
• Laclau, E. (2005). La razón populista Buenos Aires: Fondo de Cultura Económica.
• Larson, R. W. (1986). Populism in the Mountain West Albuquerque: University of New Mexico Press.
• Le Bon, G. (2001). The Crown: A Study of the popular mind (9ª ed.). Kitchener: Batoche Books. [1896]
• Levy, P. (2002). Ciberdemocracia Lisboa: Instituto Piaget.
• Lippmann, W. (1922). Public Opinion New York: Harcourt, Brace and Company.
• Lippmann, W. (1928). American Inquisitors: A Commentary on Dayton and Chicago New York: The MacMillan Company.
• Lippmann, W. (1933). A Preface to politics New York: The Macmillan Company.
• López-Meri, A., Marcos-García, S., & Casero-Ripollés, A. (2017). ¿Qué hacen los políticos en Twitter? Funciones y estrategias comunicativas en la campaña electoral española de 2016. El profesional de la información, 26(5), 795–804. DOI: 10.3145/epi.2017.sep.02
  » https://doi.org/10.3145/epi.2017.sep.02
• Matthews-Ferrero, D. (2018). Populismo europeo contemporáneo y la vuelta de la historia. Revista CIDOB D’Afers Internacionals, (119), 85–112. DOI:10.24241/rcai.2018.119.2.85
  » https://doi.org/10.24241/rcai.2018.119.2.85
• Moore, M. (2019). Democracia manipulada Carcavelos: Editora Self.
• Mouffe, C. (2018). For a left populism London: Verso.
• Mudde, C. (2004). The Populist Zeitgeist. Government and Opposition, 39(4), 541–563. DOI: 10.1111/j.1477-7053.2004.00135.x
  » https://doi.org/10.1111/j.1477-7053.2004.00135.x
• Müller, J.-W. (2016). What is populism Philadelphia: University of Pennsylvania Press.
• Norris, P. (2000). A Virtuous Circle: Political Communication in Postindustrial Societies Cambridge: Cambridge University Press.
• Norris, P., & Inglehart, R. (2019). Cultural Backlash. Trump, Brexit, and Authoritarian Populism Cambridge: Cambridge University Press.
• Ostler, J. (1993). Prairie populism Kansas City: University Press of Kansas.
• Pamerlee, J. H., & Bichard, S. L. (2012). Politics and the Twitter Revolution: How Tweets Influence the Relationship between Political Leaders and the Public Plymouth: Lexington Books.
• Panizza, F. (2005). Populism and the Mirror of Democracy London: Verso.
• Parmelee, J. H. (2013). The agenda-building function of political tweets. New Media and Society, 16(3), 434–450. DOI: 10.1177/1461444813487955
  » https://doi.org/10.1177/1461444813487955
• Pollack, N. (1962). The populist response to industrial America: midwestern populist thought Cambridge: Harvard University Press.
• Retamozo, M. (2014). Populismo en América Latina: Desde la teoría hacia el análisis político. Discurso, sujeto e inclusión en el caso argentino. Colombia Internacional, (82), 221–258. DOI: 10.7440/colombiaint82.2014.09
  » https://doi.org/10.7440/colombiaint82.2014.09
• Riemer, N., Simon, Douglas W., & Romance, J. (2018). The Challenge of Politics; An Introduction to Political Science (5th ed.) London: SAGE / CQ Press.
• Rodríguez-Andrés, R. (2018). Trump 2016: ¿Presidente gracias a las Redes Sociales? Palabra Clave, 21(3), 831-859. DOI: 10.5294/pacla.2018.21.3.8
  » https://doi.org/10.5294/pacla.2018.21.3.8
• Taras, D., & Davis, R. (2019). Power Shift? Political Leadership and Social Media Routledge: London
• Tarde, G. (1901). L’Opinion et la Foule Paris: Felix Alcan.
• Teixeira, V. C. (2018). Competição eleitoral no cenário brasileiro utilizando a internet: Ágora ou Clientela. Esferas, (12), 9–18. DOI: 10.31501/esf.v1i12.8267
  » https://doi.org/10.31501/esf.v1i12.8267
• Tormey, S. (2019). Populismo uma breve introdução São Paulo: Editora Pensamento-Cultrix Ltda.
• Urman, A. (2019). Context matters: political polarization on Twitter from a comparative perspective. Media, Culture & Society, 42(6), 857-879. DOI: 10.1177/0163443719876541
  » https://doi.org/10.1177/0163443719876541
• Weller, K., Bruns, A., Burgess, J., Mahrt, M., & Puschmann, C. (2014). Twitter and Society (Vol. 89). New York, NY: Peter Lang Publishing.
• White, T. H. (1964). The Making of the President 1964 New York: Atheneum Publishers.
• Wieviorka, M. (1993). A Democracia à prova. Nacionalismo, Populismo e Etnicidade Lisboa: Instituto Piaget.

Desk Review Editor: Tania Rosas-Moreno

Publication Dates

• Publication in this collection
  15 Apr 2022
• Date of issue
  Sep-Dec 2021

History

• Received
  29 Mar 2021
• Reviewed
  01 June 2021
• Reviewed
  16 Aug 2021
• Reviewed
  20 Oct 2021
• Accepted
  21 Oct 2021