Peculiarities of Modern Nationalist Messages in Online Political Communication: The Analyses of Donald Trump’s and Jair Bolsonaro’s Election Campaigns

As digitalized election campaigns are a new phenomenon, there are almost no studies defining the peculiarities of modern nationalist messages in online political communication. This article seeks to identify some communication patterns and recent innovations in delivering online nationalist messages. These patterns are regarded in conflation with nationalist and populist approaches by political leaders during their digital election campaigns. The literature review approach is chosen to explore the articulation of nationalist and populist messages during Donald Trump’s (The United States), and Jair Bolsonaro’s (Brazil) election campaigns. Overall, the study boils down to an analysis of the populist and nationalist signifiers in social media posts, and the degree to which their structures of meaning revolve around the vertical down/up or the horizontal in/out axis. As a result, some common traits of modern nationalist messages in online political communication are identified and future areas of research are proposed.


Introduction
The victories of Donald Trump in 2016 (USA) and Jair Bolsonaro in 2018 (Brazil) were marked by several novelties from the point of view of mechanisms of online political communication. According to the data, Trump's campaign messages were three times more retweeted and five times more shared on Facebook than those from the opposite candidate (Persily, 2017). Based on a sample of 17 million tweets, Woolley and Guilbeault (2017) demonstrated how bots were able to occupy central positions in mediating information on Twitter during elections. Jair Bolsonaro, in the same vein, with just six seconds of TV daily, left other candidates behind; those candidates used traditional models of communication, such as mass media, debates and voter support (Piaia & Alves, 2019). The victory of the Brazilian political-fringe, divisive and intolerant political candidate (who joined the party only some months before the election, without a robust organizational structure in the states, and without a party alliance that guaranteed television time for electoral propaganda) had been seen as completely unlikely before the first round of the 2018 election. Nonetheless, Jair Messias Bolsonaro was the first case of building a candidacy mostly supported by digital channels in Brazil.
These two examples show that the digital campaigns, once relegated by marketers to the background in the mass media, have gained relevance. These practices have been professionalizing and diversifying due to the technological innovations that appear during each cycle, such as websites, emails, blogs, Twitter 1 , Facebook 2 , Instagram 3 , YouTube 4 and, more recently, WhatsApp 5 (Aggio, 2014). These significant changes in the mediated political communication system have given rise to scientific research devoted to the investigation of messages, posts, retweets etc., for defining modern communication styles successfully used by politicians in social media. As a result, flamboyant speeches, extremist rhetoric, emotional appeals and other techniques were attributed to populism by most researchers (Fitzduff, 2017;Gonawela et al., 2018;Iasulaitis & Vieira, 2019;Ituassu et al., 2018).
The nation-state remains the primary context for democratic political representation and public debate, making references to "the nation" unavoidable for most political discourses. Traditionally, operating within a national context, politicians, even populists that do not endorse a nationalist program, tend to speak in the name of a people defined at the national level. These connections between populism and nationalism have led to a conflation of populism and nationalism, which is traditionally reflected in political communication styles (de Cleen, 2017, p. 342).
Much valuable work has been done on how populism and nationalism come together in particular movements and parties (Canovan, 2005;Mény & Surel, 2000;Stavrakakis, 2005). However, explicit conceptual reflections on the relationship between populism and nationalism in the online communicational styles of politicians that could 1 Twitter™ is a trademark of Twitter Inc., registered in the U.S. and other countries. 2 Facebook™ is a trademark of Facebook Inc., registered in the U.S. and other countries. 3 Instagram™ is a trademark of Instagram Inc., registered in the U.S. and other countries. to a certain space, and that is constructed through an in/out (member/nonmember) opposition between the nation and its outgroups" (de Cleen & Stavrakakis, 2017, p. 308). This does not mean that nationalists exclusively use the word "nation". They also refer to "the people", as well as, "state", "land", "freedom", "democracy" and "culture", which acquire meaning in relation to the signifier 'nation' (Freeden, 1998, p. 750).
Conventionally, nationalism has been distinguished between an ethnic and a civic variant. Civic nationalism is based on citizenship and the ability of individuals to join the nation, whereas ethnic nationalism is based on the myth of common descent and is thus less inclusive (Kohn, 1944). According to Florian Bieber (2018), although nationalisms differ around the world, there are still some common patterns. The fear of immigrants, linked to the fear of a threat to the identity of the receiving country, is one of these patterns. The wish for homogeneity is an important element that fuels nationalism, and traditionally it has always been weaker in countries with a multifaceted identity and a high level of immigration that contributed to shaping this identity.
Even more polysemic and controversial than nationalism is the concept of populism, which refers to a wide range of empirical phenomena. It has been defined as a rhetorical style of political communication, a thin-centered ideology, a form of political behavior, and a strategy of consensus organization (Mudde, 2007).
Populism is a dichotomic discourse in which the people are juxtaposed to the elite along the lines of a down/up antagonism, in which the people are discursively constructed as a large, powerless group through opposition to the elite, who are conceived as a small and illegitimately powerful group. Populist politics claims to represent the people against elite that frustrates their legitimate demands, and presents these demands as expressions of the people's will (Laclau, 2005;Stavrakakis & Katsambekis, 2014). Thus, three elements can be considered as the common denominators of these successive historical and scholarly theoretical shapes of populism. Populism (1) always refers to the people and justifies its actions by appealing to and identifying with the people; (2) is rooted in anti-elite feelings; (3) considers the people as a monolithic group without internal differences except for some very specific categories who are subject to an exclusion strategy.
If populists define the people in national terms (a tempting strategy in both nationstates and in multinational states) populism and nationalism merge; whereby, the corrupt elite can be either a minority accused of holding a disproportionate amount of political or economic power, or an elite accused of being beholden to foreign interests.
Conceiving nationalism and populism as ideologies that are articulated discursively by political actors and media actors bridges existing literature from political science and communication science (de Vreese et al., 2018). From a distinct political communication point of view, the focus now shifts from what constitutes the ideology of nationalism and populism to how it is communicated. In order not to overlook the vast research area of nationalism and populism studies, further understanding of, and approach to, nationalism and populism, thus, centers around communication and social media. Due to the fact that the advent of more sophisticated communication technologies, and the rise of social media, are seen to have created new opportunities and platforms for nationalists and populists to spread their messages, the main accent in this study is paid to messages produced exactly in social media, skipping other mediums Gerbaudo, 2018).
There is now a growing body of literature on populism, and in the emerging field of populist political communication, which has mainly concentrated on media effects and mediated populism in Europe (Aalberg et al., 2016). Empirical examples in recent years have demonstrated this in Switzerland (Ernst et al., 2017), the Netherlands (Hameleers et al., 2016), Hungary and Italy (Moffitt, 2016). Hameleers et al. (2018) report on an unprecedented sixteen-country experiment testing the effects of populist communication on political engagement. Some empirical tests for the presence of populism features in media content have been conducted (Bos & Brants, 2014;Rooduijn, 2014), and, based on the use of content characteristics and style features, one can distinguish different types of populism such as complete populism, excluding populism and empty populism (Aalberg & de Vreese, 2016;Jagers & Walgrave, 2007). Elena Bloch and Ralph Negrine (2017) provided a framework, which assesses the relevant features of specific populist actors' communicative styles, whereas Christian Fuchs (2020) defined the role of social media in the communication of nationalist ideology. Nationalist communication on social media seem to be predominant, and a great change after years in which liberal culture used social media in a more effective way (Adriani, 2019). L. Hagen et al. (2019) investigated the frames and meanings of emoji characters expressed by actors in defining their own identities involved in the white nationalist conversation in Twitter. Ki Deuk Hyun and Jinhee Kim (2015) in their study contend that online political expression facilitated by news consumption enhances support for the existing sociopolitical system, both directly and indirectly through nationalism.
Although researchers find the combination of populism and nationalism style a "powerful cocktail" (Sheranova, 2018, p. 2) as these features of communication style are defined to be used in online message building, only a few studies can be traced in the area of nationalism and populism conjunction as a communication phenomenon. Arzuu Sheranova (2018) considered the empirical case of the populist-nationalist leadership of Hungary, and demonstrated the interplay between populism and nationalism. Whereas, Robert Schertzer and Eric Woods (2020) found that ethnonationalist and populist themes were, by far, the most important component of Trump's tweets, and these themes built upon long-standing myths and symbols of an ethnic conception of American identity.
However, this work tends to be quite limited. It often focuses on identifying discursive strategies that are broadly shared across movements and content, rather than on how nationalist populist communication is articulated between each other in the online messages of politicians. These studies are limited by the analysis of single country representatives, and doesn't express the wide peculiarities, which could be a framework for further research. As a result, they can miss what makes these ideologies meaningful and, therefore, why they resonate with voters.
This article adopts a literature review approach to analyzing how nationalism and populism are articulated in the two cases of online election campaigns, those from The United States and Brazil, in ways, which surface important processes of sensemaking and message construction, and highlight issues of conflation. First, I provide a choice explanation of Donald Trump and Jair Bolsonaro's election campaigns, and give an account of my qualitative methods of data collection and an interpretive mode of data analysis. Second, related shared generalizations centered on aspects of nationalist and populist conflation are elaborated. Finally, I discuss the implications of my findings and methodological approach for understanding the processes of online political communication.

Methodology
The study aims to identify the peculiarities in online political communication, which are seen in conflation of nationalism and populism approaches, through comparative empirical investigation of the parameters, as well as the operative variables in order to verify if there are specificities and regularities.
For exploring the collective evidence in articulation of nationalist and populist messages in online communication, I selected the literature review approach (Bass & Wind, 1995) to make generalizations, rather than meta-analysis, because of the diverse design of the studies considered and the international scope (as it examines research by academician and practitioners in the United States and Brazil). The empirical generalizations are made by analyzing the interrelations among nationalist and populist messages that researchers have studied in defining the specific style and way messages were disseminated, which were popular among target audiences in Trump and Bolsonaro's election campaigns.
Studies eventually included were selected from an extensive database using investigations of the two presidential election campaigns at different angles of study. The study selection criteria were the following: (1) Each study should focus on the online communication effects of election campaigns. Thus, studies pertaining to general political issues and communication effects after elections were excluded; (2) The study should report empirical results or discuss empirical results of other studies; (3) The study should be in English, Portuguese or Russian; (4) The majority of studies were published in communication journals. Relevant books and conference proceedings were also included. The key consideration was whether the study contributed to the stock of knowledge of how the new type of message is being built in political communication on the social media. In order to get a sample of research articles and case studies, the following snow-ball procedure was applied; I started with a well-known and widely cited set of research articles on Trump's and Bolsonaro's communications on social media, traced references in those articles, back-tracked the cited works in the references, and so on, until I was able to find no new relevant articles. This process resulted in a total of 54 research articles.
I make no claim that this selection is complete, and there are practitioners and academic papers I missed, but I believe these studies are enough to shed a light on the peculiarities of nationalist messages in online communications in America and Brazil. Therefore, to trace commonalities in the diversity, the comparative method (Esser & Pfetsch, 2016) has been used, by which the election campaigns of Donald Trump and Jair Bolsonaro are analyzed.
These two cases were chosen because there are clearly similarities between the phenomena of "Trumpism" and "Bolsonarism" that do not seem to be mere coincidence. Trump's and Bolsonaro's marketers have invested heavily in creating the image of a "new" politician.
In the last decade, the re-emergence of nationalist populism has sparked much research into possible reasons for its renewed attractiveness and its implications worldwide (van Hooft, 2016, p. 30). US President Donald Trump's rhetoric and policies are especially considered as signs of, and impetus to, a renationalization of global politics, in which national interests are prioritized and international challenges and partners subordinated (Kagan, 2018). Other leaders with nationalist messages all around the globe have been riding on the wave of nationalist messages that Trump has been spreading since his election campaign in 2016. One especially successful propagator of nationalist appeals is Jair Bolsonaro, who was elected Brazil's President in 2018. During his election campaign, he promised to fight globalism, and has been called the "Trump of the Tropics" based on their similarly nationalist rhetoric (Shear & Haberman, 2019). These phenomena pushed the scientific community to deeply study the peculiarities of these presidents' election campaigns and their populist messages in social media.
Thus, considering these similarities and the presence of numerous studies devoted to Trump's and Bolsonaro's identity, rhetoric and use of media, this research is mainly concentrated on analyzing the conflation of nationalism and populism in their online messages during election campaigns, rather than on political regimes and the type of communication strategy after successful elections.

Data analysis
The empirical generalizations have been developed from content analysis of empirical studies results due to the fact that this type of analysis is an appropriate method when the phenomenon to be observed is communication, rather than behavior, or a physical object (Malhotra, 1996). This study carefully followed the procedures recommended in content analysis literature (Harris, 2001).
One of the most fundamental and important decisions is the determination of the basic unit of text to be classified (Weber, 1990). Six units that have been commonly used in content analysis literature are word, word sense, sentence, theme, paragraph and whole text (ibid.). Sections such as Analysis and Results in articles were chosen as the units of analysis. The aim was to find the data used for analysis and the final results in those studies.
The main codes were connected with central elements of populist and nationalist communication. In populist communication three elements are central: (1) reference to "the People" (2) a battle against the "corrupted elite" and with a possible extension of (3) the identification of an out-group. These defining elements have been emphasized by several scholars of populism (Aalberg & de Vreese, 2016;Kriesi, 2014;Mudde, 2004). For nationalism, they are: (1) trust in other nationalities (perceived out-groups); (2) pride in the Nation; (3) isolationism; (4) the rank of national identity over other identities; and (5) State vs. ethnic identity (Bieber, 2018).
During the process of analysis, special attention is paid to the content and how the nodal points of populism (the "people-as-underdog" and "the elite") and nationalism (the ("people-as-nation") acquire meaning through the articulation of populism and nationalism with their down/up and in/out structure in the candidates' online messages (de Cleen & Stavrakakis, 2017, p. 312)

Results
The success of the recent election campaigns of the American and Brazilian presidents exemplifies the combination of nationalist and populist approaches in their online communication rather than a clear nationalist or populist strategy. The content of the messages was cumulatively intertwined with the features of both ideologies and distinguished as a more efficient one in the process of involvement. Thus, according to No. 1, 5, 9, and 12 in the Table 1, and the content peculiarities of candidates' messages, the following generalizations can be made.

G1: A positive association between candidates' personal messages in their online presence, and claims of being an outsider and patriot of the country, are considered an indirect indicator of the predictive potential of online interactions with respect to electoral results.
The notion of an outsider in the candidates' messages refers to the elevation of their status as the representatives of the people and defenders against the others, which is the strategy of populist rhetoric. To reinforce the positive connotation of this word, references to patriotism and national values are applied. Words like father, mother, son, family, boy and school refer to the narrative of the narrator's past and family connections (Santos et al., 2019, p. 5). It is an attempt to tell the story itself. This rhetorical construction resumes its personal and political trajectory, detaching itself from the main actors of negative reputation and approaching or, at least, illuminating its connections with actors of positive reputation according to their speech.
Regarding the structure of society, Trump and Bolsonaro vehemently defended nationalism using national symbols, such as the flags and their colors, to compose the visual identity of their campaigns. Referring to the phrases "Brazil above all, God above all" or "Make America Great Again", such words as Brazil, God and America are central nationalist indicators. Other words also gain relevance, especially the verbs believe, change and want that appear in argument sentences. The verb change, for example, appears more than 55 times in Bolsonaro's posts, like "[…] really change the destiny of our Brazil". This content is the construction of an idealized, fraternal and pious nationalism mobilized through an emotional rhetorical approach, using terms such as friend, hug, heart and love (Santos et al., 2019, p. 6). It is worth mentioning frank, personal posts on social nets that blur the boundaries between a politician and ordinary people. The candidate's personal involvement was also underlined by the account name, for instance, @real_DonaldTrump, as though to underline the tweets, which came directly from Trump himself and were not managed and crafted solely by his campaign (Enli, 2017). Single-source and experimental studies have repeatedly and independently verified that the combination of a satirical tone with negative and aggressive meaning, used with simple, informal language repeated per involvement cycle, were the language mechanisms of getting closer to ordinary people and braking barriers between them.
Researchers also emphasize the fact that highly personalized and hyperbolized messages on social networks are intended to transform political facts and events into a spectacle. The new technologies contribute to this process. The hyperbolization of broadcast messages on social networks (Facebook, Twitter) is another efficient mechanism used in the image building of these two politicians. The use of epithets appeals to the emotions of voters by means of representing everything that happens in the country in a superlative form, such as the worst, the most terrible, nightmare, catastrophe, etc. (Passos de Azevedo, 2019, p. 8).

G2: Exacerbation of nationalist mobilization in the process of giving the oligarchy's power back to the people through the candidate.
According to the statements 3, 8, 9, and 12 in the Table 1, this positioning is part of a larger delegitimizing strategy that targets the oligarchies as opposed to the people (de la Torre, 1992, p. 386), an oligarchy that can encompass various actors, including the social elite, the political class and the media. These politicians have used language to create otherness, with varying degrees of success. Similar tactics of linguistic othering is a prominent example of how they create and disseminate a populist frame.
Trump's strategy, like Bolsonaro's, was to present the mainstream media as the one chasing them. The candidates repeatedly claimed that the country's main media outlets were campaigning against them, defending their main opponent and even publishing lying news that compromised their images (Iasulaitis & Vieira, 2019). The intention is to exacerbate nationalist mobilization with the promise to give back power to the people, through the candidate of course (Santos et al., 2019).
For Trump, others' supportive quotes or endorsements were the most common content. As much as 39.3% of his tweets belonged to this category, followed by criticism or attack on other (25%). The attack tweets took up a quarter of Trump's tweets including attacks on other candidates (16.9%). The attacks were often coupled with incivility; one out of ten (10.5%) Trump's tweets and retweets included uncivil wordings attacking other candidates (clowns, jabronis, corrupt liar, disaster, Ambien for insomnia), journalists (outright liar, really dumb puppets), and debates (clown competition, garbage) (Lee & Lim, 2016, p. 850). Such criticism of the main political actors and institutional mechanisms perceived in the two campaigns under analysis highlight the anti-establishment brand (Iasulaitis & Vieira, 2019).

G3: Such themes as, (a) violence/public security; b) health; c) the economy are the three key intermediate effects which lead to the success in elections.
Statements 10, 13, 14, 15, 16, 20, 21, and 24 in the Table 1 empirically reconcile both types of rhetoric employed by the candidates. Researchers come to the conclusion, based on Trump's and Bolsonaro's online messages, that politicians who managed to control the themes that are more prominent on the public agenda, covering these subjects in its favor, are more likely to win the elections (Magalhães & Veiga, 2019).
Thus, nodal nationalist words in politician accounts referred to security. Commitment, government, and citizen place a strong emphasis on the topic of public security, combating crime and other related matters. In other words, the posts of the candidates present a vocabulary that often associates the idea of citizen/people to the issue of security, including the use of commitment as a way to tie government actions with improving the individual's life via security. The words that make up this class are aggregated around the idea of "security" (Carreiro & Matos, 2019, p. 4).
In the sphere of foreign affairs, both candidates exercised the defense and apology for the Armed Forces in Brazil and the United States. In addition, the candidates strongly defended the legalization of the possession of weapons (including a weapon becoming a symbol of Bolsonaro's campaign) triggered through images, gestures and emoticons.
Both candidates emphasized national security, the review of trade relations with China, strong animosity towards socialist countries, and their insertion in international organizations. Bolsonaro even proposed to remove Brazil from the UN Security Council, "for taking a stand against Israel and only defending what is no good" and defended the establishment of commercial relations without what he calls "ideological bias", which refers to relations with countries with leftist governments (Iasulaitis & Vieira, 2019, p. 5).
Regarding the thematic plan, the analysis of relational pairs demonstrated that Trump's strategies in the immediately preceding election served as a showcase and demonstration effect for Bolsonaro's campaign (from the point of view of the agenda adopted and the style, and the strategies of using Twitter for the candidate) filled with political humor techniques, sarcasm, irony, scorn, negative adjectives and nicknames of opponents, to carry out a negative campaign.

G4: The concept of splitting the country into Christian or Family groups (to further their unification with the candidate's help), and the Minorities group, is applied as a paradigm of digital political communication in contemporary electoral contexts.
According to the statements 9, 12, 22, and 25 in the Table 1, populists claim that the will of the people is the only legitimate source of authority, but the concept of people inherently depends on there being some other groups that are opposed to "us". The politicians use social media to assert themselves as the defenders of the in-group against all out-groups, which is the in/out structure of nationalist rhetoric. Though it is fundamental for populists to position themselves as the defender of the people, it must be clear who they are defending the people against (Caiani & della Porta, 2011, p. 187). Populists often target and attack minority cultural and ethnic groups, both within their sovereign borders and beyond their borders. In these cases, Bolsonaro defended morality and the traditional family, and opposed abortion under any circumstances, whereas Trump gave a voice to Christians and other proponents of traditional family valuesies (Iasulaitis & Vieira, 2019). Though, on the surface, this color-blind and gender-blind language might make Bolsonaro seem to be treating all people equally; in fact, color-blindness and gender-blindness are counterproductive ideologies ignoring the systematic discrimination and prejudice that women and people of color face (Trujillo-Pagán, 2018).
Both politicians use language to position themselves as the defenders of a constructed in-group against a constructed out-group. Presumably, in-groups and outgroups are constructed because the people positioned in each group are not inherently interconnected. The LGBTQ community, indigenous people and the political left are not intrinsically related. Rather, Bolsonaro and Trump groups them together as the out-group in opposition to the in-group, which is similarly a constructed group.

Conclusion
Social media has rapidly accelerated the mediatization of politics by allowing political leaders to disseminate individualized content and directly interact with constituents, as well as creating niche interest networks where people can constantly interact with others who share their ideals (Mazzoleni & Bracciale, 2018). Such opportunities may develop additional features of message building in the process of communication with voters.
This article represents an attempt at identifying some communicative patterns and recent innovations in delivering nationalist messages, which are regarded in conflation with nationalism and populism approaches by political leaders during their election campaigns. When a politician articulates (elements of) different discourses, these are not simply added, one on top of the other. Rather, through the articulatory process (and the elements thereof) each acquires a particular meaning. This means that when different political agents articulate populism and nationalism, this can lead to very different results (de Cleen, 2017).
Donald Trump and Jair Bolsonaro are the perfect examples of leaders who have capitalized on the mediatization of politics in order to gain political relevance in a relatively short period of time, and they are seen by many researchers as the candidates making use of populist or nationalist rhetoric. The analysis of particular politicians' messages shows the articulation of populism with nationalism and applied literature review approach is aimed at uncovering the specificities of these cases in question.
Overall, the study of the articulation of populist and nationalist online messages boils down to an analysis of the location of the populist and nationalist signifiers in their messages on social media, of the degree to which that structure of meaning revolves around the vertical down/up or the horizontal in/out axis, and of the signifying relations forged between the populist and nationalist signifiers, and between the down/up and in/out axes.
As a result, the developed generalizations uncover populist messages revolving around the exclusion of certain groups of people from the nation, from the nation-state, and from political decision-making. This style of content creation reinforces the fact that the "people-as-underdog" is a sub-group of the ethnically and culturally defined nation and includes no ordinary people of foreign descent (which are marked as another outgroup, called Communists, in the case of Bolsonaro). Moreover, the "people-as-underdog" is constructed in opposition against migrants and other nationalist out-groups. Indeed, populist messages have interpellated ordinary people primarily as an underdog, using the exclusionary nationalist terms Family and Christian.
Ordinary people, they argue, are the prime victims of multicultural society. Thus, Bolsonaro asked to "Remove Brazil from the UN Security Council", and Trump demanded to "construct a wall from illegal immigration" (Caiani & della Porta, 2011, p. 188). Also, the main argument presenting "the elite" as lacking legitimacy is that it pampers ethnic-cultural minorities and does not take the interests of the ordinary people who suffer from diversity to heart (de Cleen, 2016a, 2016b). Trump and Bolsonaro disseminate an exclusionary frame through messages on social media in order to establish themselves as the single representation of the constructed in-group and the defender against out-groups. What I see here is how positions on the in/out axis (membership of the nation, as well as, serving the interests of the nation) come to determine positions on the down/up axis.
One important consideration for future research is the question of method. The selection of material to illustrate the conflation was drawn from many sources devoted to the analysis of media-based messages, and specifically in the social media, but other types of media were not included. Politicians traditionally exist in mediatized and nonmediatized environments, and future research should be sensitive to the many ways, in which leaders communicate and establish themselves as leaders through speech, gestures, symbols and actions. It is the totality that is important, and future research should go beyond, for example, analysis of one form of data only, be it news items or manifestos.