Navigation – Plan du site

AccueilNuméros1Hors-thèmeConnecting the American People: T...


Connecting the American People: The Effects of New Modes of Communication during the American Presidential Campaign of 2020

Louise Anglès d’Auriac


Cette étude analyse la notion de « peuple américain » sur les comptes Twitter de Joe Biden (@JoeBiden) et de Donald Trump (@realDonaldTrump) pendant les élections présidentielles de 2020 aux États-Unis. Cette observation quantitative et qualitative souligne la manière dont différentes stratégies de communication ont transformé la définition que les candidats ont donnée du peuple américain, et les implications que ces stratégies ont eues sur la représentation démocratique et sur l’action en ligne. En premier lieu, une étude lexicale et rhétorique formelle permet de relever les différents modes d’ « émergence » du peuple (Rosanvallon 7) dans les publications des candidats. L’analyse approfondie du pronom we suggère que les candidats tirent parti de la polysémie de la notion de généralité, dans un but stratégique. En second lieu, la création d’in-groups et d’out-groups crée un cadre strict, qui donne forme au peuple et suggère un type de la « représentation-narration » qui caractérise la politique contemporaine (Rosanvallon 192). En troisième lieu, à la suite de la typologie développée par Lance Bennett et Alexandra Segerberg (1), cette étude montre que Biden a perpétué sur Twitter la logique traditionnelle d’ « action connective », tandis que la stratégie de Trump mettait en évidence certaines caractéristiques d’une communication « hybride ». Cette étude suggère que, sur les réseaux sociaux, Trump a moins d’influence que Biden sur la voix de ses supporters, malgré ce que sa personnalité dominatrice a souvent laissé penser.

Haut de page

Texte intégral

Fig. 1.

Fig. 1.

Screenshots of the X-Twitter platform. Tweets published by @JoeBiden and @realDonaldTrump in May 2020.

  • 1 Although it was renamed X in July 2023, this article refers to the platform as Twitter, as it was c (...)

1These two tweets (Fig. 1) were published a few days apart, in the midst of the 2020 presidential campaigns in the United States, by the two candidates in the race. Both messages define the American people. Yet, they display very different communication strategies. Although Donald Trump and Joe Biden mobilize the same notion of “the people,” the definition given to this entity differ markedly: Trump is obviously belligerent (“warriors”), while Biden relies on values and qualifiers (“strong, tough, brave, and big-hearted”). Analyzing the various ways in which candidates defined the American people online, this study highlights how communication strategies framed a specific notion of the American community, which established distinct modes of representation during the 2020 presidential campaigns on X (formerly Twitter).1 Communication on social media implied two different forms of popular action, depending on the candidates’ communication strategies.

2Defining “the people” has proved a problematic endeavor in the history of the United States. Gordon Wood (xi) underlines that, in the eighteenth century, the new colonies lacked the basics of national unity, such as an American ethnicity, historical institutions, and a stable national government. Therefore, the rhetorical construction of the country has been crucial to give the American people a “feeling of society” (Pye and Tailor). According to Cody Delistraty (31), the American national construction is a narrative process. The United States is founded on a set of “characters,” “subnarratives,” and “overarching myths” (Danforth 203).

3For many reasons, including the gradual expansion of the country towards the West (the Frontier) and the tensions between Northern and Southern states in the nineteenth century, citizens developed a strong attachment to the particularisms of their local states. Today, these are still very strong: according to Jefferey M. Sellers, Anders Lidström, and Yooil Bae (314), the United States is a “civic localist” democracy. Civic identities appear at the local level, while the federal structure loosely attaches the states to the federal government. Simultaneously, claims for the rights of certain social groups such as women, African Americans, and LGBTQI+ people have led to the polarization of the electorate. This is amplified in the contemporary “post-broadcast era” (McKinney 469): “common political knowledge” about the people is broken down while “individual-based [...] micro-political realities” prevail (Gillies 14). Society is fractured along economic, cultural, and partisan lines. The combinations of these factors create communities of like-minded individuals, who share similar political views (Edsall 71).

4Therefore, Americanness strongly relies on the construction of a political bond, often through myths and discourses. During political campaigns, the stakes of defining American identity are very high: the dynamics that run through the apparently unified nation depend on the definition that the candidates’ communication gives the people of the United States. Because the essence of the nation’s unity is not to be found in a common language or history, Americanness is “unfinished” (Walzer 614): “Our mission of holding together a diverse people in one indivisible nation has been a work in progress” (Danforth 70).

5In this context, campaigns are key moments of theorization and construction of the notion of American people. Today, social media have provided new opportunities for the diffusion of a message on the people. Communities are specifically targeted since identity politics have become the norm; both Republicans and Democrats tailor their rhetoric to the interests of particular groups (Edsall 30). Social media have generated new opportunities and challenges for fundraising, communication, and recruiting activists (Iyengar 147). On the citizens’ side, social media have transformed “civic engagement” (123) and exposure to news (126). Dane Pye and Mary Anne Taylor speak of a “feeling of society” (6). Online, individuals assert their particularities, which largely determines their political awareness and their votes.

6This research deals with three sets of questions:

71– How do candidates give “a feeling of society” on Twitter, in the context of political campaigns?
  1a– Between April 2019, when Biden announced that he was running for president, and January 2021, how did the candidates’ communication strategies on Twitter frame the notion of “the American people”?
  1b– Did the polysemic pronoun “we” imply an agglomeration of individual entities?

82– In terms of communication, how did candidates use framing and storytelling?
  2a– Did the candidates’ communication target specific communities? Can we identify the main characters—communities, in-groups and out-groups, “heroes” and “villains”—that played a part in the notion of American people? Did the framing and storytelling of the American people favor identity politics?
  2b– Did their attitudes confirm the shift towards “representation as narrative” that Rosanvallon (192) notices in contemporary politics?

93– In terms of organization, how did these rhetorical strategies shape a specific mode of representation?
  3a– What rhetorical position did candidates take online? Was one candidate more controlling than the other towards the voice of his base, or, on the contrary, did they both embrace the discourse that seemed to emerge from their supporters?
  3b– Did candidates use Twitter merely as a communication tool, or did they aim at generating a particular civic action offline? Did candidates generate the “collective” or “connective” action online schemes theorized by Lance Bennett and Alexandra Segerberg (1)?

Literature review

The notion of “the people”

10Most of the research studying the accounts of the candidates during presidential elections emphasizes the change that Donald Trump brought about in 2016 (Jamieson and Taussig 23). Some of the main traits underlined by the analysts are the Republican candidate’s spontaneity, his Manicheism, his rejection of scientific and democratic institutions, and his reliance on “alternative facts” (Coe et Griffin 9; Pizarro-Sirera 163; Austermuehl 528). The populist aspect of speech has been extensively described (Mélandri; Jamin, 2017; 2018; Zuckerman), but less research is available on the relational dimension of the candidates’ rhetoric.

11Pierre Rosanvallon’s categorization is of particular interest as he identifies three “modes” of the people (130). Contrarily to other theorists of the “people” (Bras; Butler; Mény and Surel), Rosanvallon defines the people depending on the democratic institution in which it emerges. The first one—the “electoral people”—manifests itself in the polls and is shaped by the distinction between a minority and a majority. It is different from the second mode of the people—the “social people”—, which is an addition of minorities, either active or passive, that have been wronged or struggle in their attempt to create a common world. Finally, Rosanvallon identifies a third mode—“people as principle”—defined as democracy’s main project, which is equality. This mode of emergence of the generality manifests itself in the US Constitution or in the Universal Declaration of Human Rights. Based on these categories, this study shows that Joe Biden and Donald Trump concealed the plurality of the “social” people, to generate a unified “electoral” people, offline in the polls and online on Twitter.

The relational aspect of campaign rhetoric

  • 2 “Grandissant, diminuant, s’étendant exagérément, s’évaporant.” Translated from the French by the au (...)

12The modes of “emergence” of “the people” characterize “the people” based on the democratic principle that gives it shape. Discourse, in that sense, is at the heart of any social structure, as it creates a space of social relations. This space is constantly changing, as the commentators Peter Meyers and Nancy Struever note, it is “growing, shrinking, excessively broadening, fading away” (23).2

13Furthermore, in the United States, discourse emerges in a polarized context. Shanto Iyengar analyzes how polarization can be seen in the American political culture, which trickles down into all social relations before American citizens even reach the age to vote (174). Iyengar also considers the enunciative context that is specific to presidential campaigns: they are strategic moments when the attitudes of the candidates are conditioned by the imperatives of momentum. Political discourse is cautious, attentive to the reaction of the audience, and linked to the significant events which punctuate the process (Iyengar 168-178). Campaigns allow the “positioning” of a candidate in the political sphere, and the construction of legitimacy in the voters’ representation of the race (Anglès d’Auriac 9).

14The discourse of the candidates has to adapt to this polarized context. Rhetorical tools such as framing and storytelling are used to refer to the fragmented American people. Through speech, candidates “frame” communities. The concept of speech frames originates in the sociological theories of Erving Goffman (1986), George Lakoff (2002) and Herbert Blumer (1969). A frame is a selection of “some aspects of a perceived reality [to] make them more salient in a communicating text” (Entman 52). These frames allow the candidates to identify “types”—heroes and villains—that play a part in the American story (Salmon). Storytelling generalizes the internal stories of the people (Salmon 11, 13). Although the technique is far from new in American political history, the massive diffusion of discourse online has contributed to the growing importance of this rhetorical strategy (Bordeau 93). This study identifies the frames that the candidates used to create in-groups and out-groups on social media. It notes the importance of storytelling as a mode of generalization of the American people and suggests that storytelling has become a mode of representation in itself.

Social media as action-generating platforms

15Social media have changed the rules of campaigning. Many theorists have underlined the constant modification of the media landscape since the twentieth century and the recently acquired role of social media in American political life. Defined as an “instrument of social infomediation of news” (Rieder and Smyrnaios 105), social media platforms have become the place where information circulates in a quicker, more efficient way (Smyrnaios et Ratinaud 125; Compagno et al. 93). Every user is a producer and a consumer of information, on the model of “produsage” developed by Axel Bruns (1). Furthermore, social media discourse has to comply with specific constraints, especially on Twitter, where the character limit went from 140 to 280 in 2017.

16Yet, Twitter is not only a stream of communication but also an organizational tool. Bennett and Segerberg distinguish between collective and connective actions online. A collective action is the “more familiar action logic”: “[It] emphasizes the problems of getting individuals to contribute to the collective endeavor that typically involves seeking some sort of public good (e.g., democratic reforms) that may be better attained through forging a common cause” (12). The collective action frame can be analyzed in the light of the negotiations that have to be made so that a society finds a consensus on its collective identity, often in a heated context. By contrast, the core of connective action recognizes “digital media as organizing agents”: “Taking public action or contributing to a common good becomes an act of personal expression and recognition or self-validation achieved by sharing ideas and actions in trusted relationships” (14-15).

17This study aims at expanding on this idea, by showing that social media campaign strategies can take into account the connective potential of a platform and effectively use it to create a powerful movement.

A qualitative and quantitative methodology

18This study is a rhetorical take on the notion of “social people” devised by Pierre Rosanvallon. It observes how candidates made use of the ambiguity of the word “we” online in 2020, to conceal the plurality of the American people behind the appearance of unity. Using the theories devised by Shanto Iyengar and Christian Salmon, it shows that the American people was composed of in-groups—the “assemblage of minorities” that Rosanvallon refers to—and out-groups. Thanks to an analysis of the imperative forms of verbs, it also shows that candidates used Twitter both as a communicational and an organizational tool.

19My approach relies on an extensive analysis of the rhetorical and structural construction of “the American people” on the Twitter accounts @realDonaldTrump and @JoeBiden. In order to determine whether candidates gave a “feeling of society” on Twitter during the campaigns, I perform a lexicometric analysis of the frequency of the personal pronoun “we.” This quantitative take on discourse is completed by the qualitative and formalist methodology of the Groupe Mu (Bonhomme 2010). Observations made on the context of the election and the specificities of the platforms (Thiault 68) help define how the blurry notion of “the American people” was defined by each candidate.

20Secondly, to analyze the different particular communities that compose the American generality, I rely on theories of “storytelling” (Salmon 6), following a two-step methodology: first, I observe the type of communities to which candidates referred, based on a frequency analysis of their posts. To support this computerized approach, I turn to a semantic analysis of these terms to identify how communities were framed, positively or negatively. This methodology is key to understand how a narrativization of the American people structured groups, and how rhetorical means led to the creation of in-groups and out-groups.

21Thirdly, in order to determine the relation of claims-making and representation that the candidates’ discourse generated online, I base my analysis on “representation as narrative” on social media (Rosanvallon 192). I argue that language is a tool that produces, manages, and transforms power relationships (Fairclough and Scholz 115). A morphosyntactic analysis of the imperative modes of verbs—which indicate a command and suggests a control of the voice and movement of an individual over another—aims at understanding how discourses entailed specific movements online. With a focus on references to voting, the voice of the voters, and the election, I define the civic movement generated by the candidates and distinguish between “collective” and “connective” actions online (Bennett and Segerberg 12).

22It should be noted that the identities of the authors of the posts can be problematic when observing online discourse. During presidential campaigns, linguistic elements are created by the candidates’ communication teams. Therefore, although Trump’s posts appear as spontaneous, the content published on the account @realDonaldTrump was probably generated both by his campaign team and by himself. This is even more plausible since Dan Scavino was probably the author of a majority of Trump’s tweets in 2016 (Draper). Yet, this study analyses the discourse of the candidates through the lens of their ethos, or characters. The ethos of the candidates are two carefully devised entities, built by the campaign team in an effort to present voters with a coherent character behind whom they can rally. This study posits that these coproduced ethos were the semi-fictional authors of their tweets during the campaigns—at least in the minds of the American people.

Corpus and data collection

23This three-step analysis is applied to the messages published on the official Twitter accounts of the candidates. The platform Twitter was chosen because it has become one of the most important political arenas since Barack Obama’s 2008 campaign. For politicians, it is an inexpensive means of diffusion which allows candidates to “circumvent traditional media keepers” (Merry 3). In the case of Donald Trump, his personal account @realDonaldTrump was chosen, although the incumbent president had an official account as a federal official—@POTUS. This choice is explained by the popularity of the account @realDonaldTrump and the considerable number of its followers at the time (between 80 and 90 million during the campaign).

24The tweets of @JoeBiden were collected with the Python programming language tweepy package, in connection with a developer API. The candidate’s timeline was saved, including retweets, from the official campaign launch date (April 18, 2019) to his inauguration (January 20, 2021). In total, this part of the corpus amounts to 5,219 posts, including 3 retweets. During the campaign, the follower count of the candidate oscillated between 10 and 15 million subscribers.

25The tweets of @realDonaldTrump were not available on the Twitter API at the time this study was made, due to the platform’s decision to remove all the tweets published by the former president. Therefore, this part of the corpus was extracted from the website The Trump Archive.3 As was the case with the tweets published by @JoeBiden, the tweets and retweets of the candidate were streamed from his campaign announcement (June 18, 2019) to his account’s suspension (January 8, 2021). This part of the corpus amounts to 16,637 posts, including 8,289 retweets.


The polysemy of “we”

Fig. 2.

Fig. 2.

Frequency analysis of personal pronouns in Donald Trump and Joe Biden’s publications on Twitter (April 18, 2019 to January 21, 2021 for Joe Biden; June 18, 2019 to January 8, 2021 for Donald Trump).

26First, this study focuses on the pronoun “we” on the Twitter accounts @JoeBiden and @realDonaldTrump. The pronoun “we” is always referring to an auxiliary entity. In its most general sense, this pronoun is a grammatical word used by a locutor who situates her or himself in a group of people and who takes the role of the speaker in a conversation exchange. On Twitter, the pronoun “we” is particularly interesting: it is part of many other references to the generality, such as the terms “people,” “America,” and “nation.” Yet, its reference is rarely explicit and remains open to interpretation. When it is identifiable, it is often implied by subtle references in the co-text—the discourse surrounding the word “we.” Therefore, it allows communication to reach as wide an audience as possible, as anyone might feel entitled to be included in the group.

27The observation of the frequency of pronouns in the candidates’ discourse on Twitter indicates that Joe Biden used it significantly more often than other personal pronouns (41% of all pronouns) during his campaign. Trump used pronouns in relatively equal proportion, with a slight preference for “I” (24.20%) and “you” (21.17%).

28An observation of the occurrences of the pronoun “we” indicates that Trump’s and Biden’s definitions of this grammatical word were very blurry in 2020. First of all, “we” could refer to different entities, varying from the campaign team, the presidential ticket, or the American people. Secondly, when referring to the American people, the definition of the generality was itself unclear. For example, on August 29, 2020, Biden published three different tweets, all including a radically different definition of the word “we.”

We have a president more interested in beating people at golf than beating COVID-19.
@JoeBiden, Aug 29, 2020 – 46.5K Retweets – 21.8K Quote Tweets – 293.2K Likes4

Folks, we’re just a few days away from our biggest end-of-month deadline yet and we need your help to hit our goal. Every donation—big or small—makes a big difference. Please chip in today.
@JoeBiden, Aug 29, 2020 – 2542 Retweets – 2282 Quote Tweets – 8751 Likes5

America was built on a simple idea: All men—and women—are created equal. It’s time we ensured all people are treated equally as well.
@JoeBiden, Aug 29, 2020 – 17.5K Retweets – 8007 Quote Tweets – 94.5K likes6

29In the first post, “we” can be defined as the group of citizens who are part of the federal republic of the United States and who are represented by federal officials, such as the president. The pronoun “we” in the second post refers to Biden’s campaign team, which is caught in the race and calls for financial support from its supporters. Finally, the third instance of “we” is ambiguous: it refers to a political agent which can “ensure” the fair treatment of citizens. Therefore, it can designate political representatives who effectively pass legislation, or citizens who can have an impact by casting a vote.

Reference: dividing the American people between in-groups and out-groups

30To better understand the “people” that is defined by the candidates’ communication, I analyze the co-text of the pronoun “we” to identify the in- and out-groups. During the 2020 campaigns, the creation of the generality in discourse led to the identification of antagonistic entities. Our study observes the lexical and semantic elements which contributed to the creation of such groups inside the supposedly unified American people.

Fig. 3. Nouns referring to communities and individuals mentioned by Donald Trump and Joe Biden.

Fig. 3. Nouns referring to communities and individuals mentioned by Donald Trump and Joe Biden.

The percentages indicated under the number of occurrences are determined in relation to the total amount of words in the publications (386,772 for Trump and 170,911 for Biden). Capitalized and non-capitalized letters were harmonized, and plural derivatives were included in the count.

31As indicated in Figure 3, the characters that mostly appeared in Donald Trump’s discourse were Democrats—or “Dems”—and Republicans. Elected officials (politicians, congressmen and congresswomen) and members of security forces (police) were mentioned 5 to 6 times more by Trump than by Biden. Observing the occurrences of those words allows us to define the antagonists (the Democrats, mentioned 15 times more by Trump), as this post exemplifies:

The only thing the Dems have going for them is their very close relationship with the Fake News Lamestream Media!
@realDonaldTrump, May 26, 2020 – 9,493 Retweets – 821 Quote Tweets – 56,8K Likes7

Fig. 4. Nouns referring to communities and individuals mentioned by Joe Biden.

Fig. 4. Nouns referring to communities and individuals mentioned by Joe Biden.

For each candidate, the percentages indicated under the number of occurrences are determined in relation to the total amount of words in the publications (386,772 for Trump and 170,911 for Biden). Capitalized and non-capitalized letters were harmonized, and plural derivatives were included in the count.

32On the other hand, Biden put more emphasis on social status and class. As shown in Figure 4, the term “middle-class” was mentioned 43 times more by Biden than by Trump. The Democratic candidate was also 27 times more likely to refer to somebody as a “wealthy” individual. Biden’s rhetoric divided society between the “super-wealthy” who were privileged by Trump’s economic policy, and the suffering “middle-class” which needed to be rebuilt:

As president, I’ll make sure giant corporations and the super-wealthy pay their fair share in taxes — and then invest that money in growing a stronger, more inclusive middle class.
@JoeBiden, Dec 5, 2019 - 745 Retweets – 737 Quote Tweets – 3.2K likes8

Fig. 5. Terminology used to qualify members of security forces

Fig. 5. Terminology used to qualify members of security forces

For each candidate, the percentages indicated under the number of occurrences are determined in relation to the total amount of words in the publications (386,772 for Trump and 170,911 for Biden). Capitalized and non-capitalized letters were harmonized, and plural derivatives were included in the count.

33A lexical analysis of the reference to various communities highlights the vocabulary that was used by the candidates to refer to members of the American people. When pointing at traditional members of his base, such as Republicans, the military or the police, Trump’s lexicon was mostly meliorative, as shown in Figure 5. Trump mentioned “law enforcement” 10 times more than Biden, which put an emphasis on the legitimacy of the police institution. This term “deem[s]” the actions of the police “legal, since it is on the side of the law” (Bhatia 86). It also recalls the Republican motto of “law and order.” For example, on July 25, 2019, Trump stated:

We love our Law Enforcement Officers all around this great Country.
@realDonaldTrump, Jul 25, 2019 – 16K Retweets – 7.8K Quote Tweets – 93K Likes9

34Conversely, while referring to undocumented immigrants, Trump’s vocabulary was openly disparaging, alternating between “illegal immigrants” and “illegal aliens.”10 On the other hand, Biden’s references to the Spanish-speaking community were favorable. The candidate went as far as tweeting in Spanish, for instance on November 1, 2020, two days before election day:

Un número récord de latinos son elegibles para votar en esta elección, así que ve a votar para lograr que tu voz se escuche. No dejes que los demás decidan por ti.
@JoeBiden, Nov 1, 2020 – 1773 Retweets – 243 Quote Tweets – 9110 Likes11

35Another example is the way candidates tackled the issues of protest and affirmative action. During the protests that followed the assassination of George Floyd, in May 2020, Joe Biden referred to the groups as “peaceful protester[s],” whereas Trump used the word “rioters.”12

My Dad used to say there’s no greater sin than the abuse of power. Whether it’s an officer bloodying a peaceful protester or a President defending him with a conspiracy theory he saw on TV.
@JoeBiden, June 9, 2020 17.4K Retweets – 2,121 Quote Tweets – 92.2K Likes13

36The lexicon of the candidates was loaded with value judgements, either negative or positive, which effectively implied the integration of certain communities into the American people, and the rejection of others out of it.

37Finally, Trump’s notion of the people of the United States was divided into political parties, while Biden’s was divided into classes. As shown by Figure 3, during the campaigns, Trump mentioned Democrats 1,912 times, and Republicans 716 times. Biden only mentioned Democrats 56 times and Republicans 60 times. He preferred the notion of “class” and designated himself as the candidate of the middle-class, mentioned 154 times, as opposed to 8 occurrences for Trump as shown by Figure 4. This division was nonetheless always hidden behind an apparently unified “we.” The following tweet is an example of this paradox:

Republicans are fighting for citizens from every background, & from every race, religion, color & creed. We are a movement for ALL AMERICANS who believe in fairness & justice, equality & dignity, opportunity & safety. We are a big tent, & a big party, w/ BIG IDEAS for the future!
@realDonaldTrump, Jan 15, 2020 – 8,652 Retweets – 1,047 Quote Tweets – 46.2K Likes14

38In this post published by Donald Trump, the pronoun “we” in the second sentence creates a division in the people by referring specifically to the Republican Party, judging by the co-text and the mention of the word “Republicans” in the first sentence. Yet, the call to “ALL AMERICANS” and the capital letters evoke a homogenized people, which is characteristic of populist rhetoric (Jamin, 2017).

Typification: creating heroes and villains

39In the candidates’ publications, the various groups that compose the American people were either heroicized or villainized in order to create in-groups and out-groups. Heroes are positive types, themselves defined as ideal role models with a set of properties and values. Villains, on the other hand, are antagonists that heroes must fight. During the political campaign, this binary logic was used by both candidates. For example, Trump used labels, i.e., adjectives meant to sum up the main quality of an individual, as part of his process of generalization:

The great people of Pennsylvania want their freedom now, and they are fully aware of what that entails. The Democrats are moving slowly, all over the USA, for political purposes. They would wait until November 3rd if it were up to them. Don’t play politics. Be safe, move quickly!
@realDonaldTrump, May 11, 2020 – 21k Retweets – 83k Likes15

40In this post, Trump gives the people of Pennsylvania a personality and a thought process, as if they were thinking in unison (“they are fully aware of what that entails”). Furthermore, they are described as struggling against the antagonistic force that the Democrats represent. In this tweet, Trump takes the role of the storyteller of his people, just as Biden does in the following post:

Brien usually works at the Ford Plant in Rawsonville, Michigan, driving a semi. But when the plant started assembling much-needed ventilators, he courageously stepped up and volunteered to help. His selfless actions will no doubt save countless lives.
@JoeBiden, Apr 26, 2020 – 208 Retweets – 1 Quote Tweet – 972 Likes16

41In this post, Biden tells the story of an everyday man, Brien, with plenty of details which add to the veracity and credibility of his story (“driving a semi”). His persona represents the values of “selfless[ness] “and “courag[e]” that characterize the self-reliant citizen and the individualistic ideology that has been at the heart of Americanness since Benjamin Franklin (Bellah et al. x). According to Biden’s campaign rhetoric, the “middle-class” is close to a “working class” of industrious people. In these posts, heroes and villains reinforce the divide that Biden and Trump highlight within the American people: Brien is a hero of the middle-class, while the people of Pennsylvania are fighting against the Democratic enemy.

Addressing the American people: generating online movement

Fig. 6. The use of the imperative mode in the tweets published or re-published by Donald Trump and Joe Biden during their campaigns.

Fig. 6. The use of the imperative mode in the tweets published or re-published by Donald Trump and Joe Biden during their campaigns.

For each candidate, the percentages indicated under the number of occurrences are determined in relation to the total amount of words in the publications (386,772 for Trump and 170,911 for Biden). Capitalized and non-capitalized letters were harmonized, and plural derivatives were included in the count.

42The third part of my analysis defines how the communication strategy of the candidates aimed at controlling discourse and at triggering civic action. To that end, I have studied the imperative modes of the posts, indicated in Figure 6. In 2020, Biden’s use of imperative verbs was remarkable relative to the overall number of words in his posts. The candidate asked his followers to “chip in,” i.e., to contribute financially to the campaign, 325 times more than Trump. He encouraged them to “tune in” to an online event 439 times more, and to “head to” a different online or offline place 93 times more than his opponent. Most verbs called for the realization of each voter’s “dutiful citizenship” (Butkowski 2400)—such as “vote” or “unite.” Other verbs strongly encouraged citizens to take part in the campaign (“chip in,” “tune in,” “head to,” for example). Yet, none of these imperative verbs called for “self-actualizing citizenship”—an “expressive” and “individualized” form of political participation which seeks to express one’s critical views on political matters (Butkowski 2400).

43On the other hand, Trump used fewer imperative modes. Aside from voting, the action that the incumbent president emphasized was “watch[ing]” content, i.e., often replays of his rally speeches.


A “feeling of society” on Twitter

44Both candidates displayed different “feeling[s] of society” on Twitter (Pye and Taylor 6). Their discourses demonstrated a plural and heterogeneous conception of the American generality. Yet, the plurality of communities and minorities existing in the American people was hidden behind the polysemic pronoun “we.” Therefore, the people built by Biden’s and Trump’s discourses navigated between homogeneity and atomization. This paradox is a structural pattern of contemporary American politics: “In the last third of the twentieth century we evolved from a nation of joiners to a nation of isolated individuals” (Danforth 70). This evolution is favored by racial, religious, and gender-related tensions.

45While Trump divided the nation along partisan lines, Biden used the categories developed by the tradition of class theory (“middle-class”). “Classes” are the social entities which constitute the workforce and shape modern capitalism (Rosanvallon 46). In the 1930s, this notion was incorporated in the rhetoric of the Democratic Party, as its interests and those of the labor movement united during the passage of the New Deal by Franklin Roosevelt (Tarrow 151; Battistini 93). Nowadays, the apparition of “new strata of workers,” with “short-term contracts” led to “a revival of the kinds of class-based movements” (Tarrow 151). Yet, it has become an “out-of-vogue symbol” (Battistini x) which designates the workers who struggle in the context of an economic crisis, striving to reach the American Dream. The absence of any lower or higher class in Biden’s discourse confirms Matteo Battistini’s assertion that “middle class is reduced to a rhetorical figure of an American identity” (167). It is an Aristotelian “happy medium” (Badiou 29), opposed to the aristocracy of the “super-wealthy” that Biden mentions in his tweets.

46In 2020, Biden’s reference to the middle-class and his conception of the parties as porous, permeable entities had a strategic objective: his chances of being elected relied on his ability to broaden his base (Meeks 18). On the other hand, Trump chose to secure his Republican electorate (Hart 8) and focused on the Republican party. Although it has been observed that the Democratic party has abandoned the “politics of classes” that was traditionally favored by the Left since the New Deal and the Fair Deal (Edsall 30), this study shows that this rhetorical tradition was still a key aspect of Biden’s rhetoric in 2020.

47Based on Rosanvallon’s distinction of the modes of emergence of the people, this study argues that the blurriness of the pronoun “we” was part of a broader strategy, which aimed at basing the candidates’ discourse on the illusion of a unified group. The candidates constantly pointed at a specific antagonist, and this discourse shaped in-groups and out-groups. Unity was a “feeling” created by the reference to broad entities such as the “middle-class” and “we,” but it disguised a divisive rhetoric. The “feeling of community” identified by Pye and Taylor (6) was threatened, because, in America, the people and its antagonist—in-groups and out-groups—coexist side by side.

The characters of the American myth: tailoring the story to fit electoral strategy

48Framing and storytelling strategies were used to target specific communities, favoring identity politics. The vocabulary used to designate the communities that constitute the people is never neutral: it always carries meaning, whether negative or positive, regarding a segment of the population. In 2020, the candidates used rhetoric to identify the in-groups and out-groups that defined the “people.” The frontiers that separated communities flowed through the notion of people and structured it. In Trump’s and Biden’s discourses, the American people was not the frame which held communities together; communities were the American people.

49The frequency analysis underlines this phenomenon. While Biden spoke positively of minorities such as Latinos and African Americans, Trump told the story of law enforcement heroes. Shanto Iyengar, in Media Politics (144-145), states that the language of a political campaign is extremely constrained. Ideological elements contained in the message serve a specific electoral strategy. Biden’s choice to publish in Spanish was a way for him to signal to the Spanish-speaking community, which represented a growing, Republican-leaning part of the electorate, that they could rally behind him. Conversely, Trump’s rejection of “illegal aliens” confirmed previous studies, which have shown that this group was the primary target of Trump’s white nationalist rhetoric of exclusion (Sirin et al. 167). As Thomas Edsall notes, Trump targeted specific communities and adopted the identity politics that had gradually become the norm in the Democratic Party since the Civil Rights Movement, thus following the political advice of his advisor Steve Bannon. His discourse focused on “Nixon’s silent majority, the Christian Right, the Reagan Democrats, the angry white men, and the Tea Party” (Edsall 30). Thus, candidates tailored their rhetoric to frame specific heroes and villains, according to their electoral strategies.

Storytelling: the creation of representation frames

50In 2020, the use of storytelling on Twitter did not only play a part in identity politics, but it also fostered a new type of representation. For Trump, the platform was more than a mere vector, a transportation system which brings the message to the audience, it served as an organizational tool.

51According to Rosanvallon, the contemporary “social form” is a “society of particularity” (64). Therefore, candidates have to navigate between their attention to particularity and the necessity to generate a wide movement. To do so, leaders use storytelling to acknowledge and narrate the identity of the American people. Following this idea, Rosanvallon develops the notion of “representation as narrative” (192), which is a new form of the politics of presence (196). In 2020, telling the story of the people of the United States on a day-to-day basis on Twitter was not only a means of communication, it was already a form of representation.

52Observing storytelling on Trump’s and Biden’s Twitter accounts highlights how candidates became the voices that formulated the claims of their people. Both candidates singularized stories into myths and generalized difficulties into shared experiences. The processes of typification and heroization are elements which prove that storytelling has become a key element of the contemporary “politics of presence.” Indeed, in 2020, the candidates’ categorization of the American people implies a specific representation. Again, the frequency of the use of words referring to certain strands within the American people suggests that Trump’s heroes were Republicans, while Biden’s were members of the broad “middle-class.” The Democratic candidate created characters that fit into the myth of the American self-made man—“Brien” the selfless worker, for example—whose work and striving for equality made him an American citizen, regardless of his party, culture, or creed.17

Twitter: between connective and collective action

53Judging by his rhetoric, and especially his use of the imperative form (Figure 6), Biden used the platform in a classic way, which Bennett and Segerberg would link to a “collective” scheme (739). On the other hand, Trump’s communication on Twitter was an example of a “hybrid scheme.”

54In 2020, Biden’s method was heir to Obama’s use of the media (Tarrow 152). As head of the executive, the president has a paternal relation to citizens: he or she has to be a role model who brings the people’s will to its full potential and maturity. This conception of leadership can be detected in the use of imperative verbs calling for the realization of “dutiful citizenship,” rather than “self-actualizing citizenship” (Butkowski). In 2020, Biden did not ask for each citizen’s individual, self-directing collaboration, but merely for their support.

55To move towards a theoretical understanding of that process, we could say that Biden’s communication aimed at transforming the mode of the “social people” into the mode of the “electoral people” (Rosanvallon 130). Plural voices had to be transformed by the institutions of representative democracy, moving away from their heterogeneous state into a consensual unity. If we link this posture with the “representation as narrative” (Rosanvallon 192) that is now the core of the relationship between the people and its representatives, the modality of that narration is not an annihilation of the contrary opinions into a homogenized totality, but rather a pacified coexistence of the antagonists; what David Axelrod (322) calls a “moving mosaic of national unity.” Biden’s communication strategy on Twitter was therefore paternalizing, but not unifying per se. Biden was the expert of the American people, the one who could transform it.

56Thus, in Biden’s communication strategy, the campaign was a tool. The real goal was offline: the election. Twitter was merely the place where one had access to the voice of the leader, and transformed into an active, offline, voting citizen:

You heard @KamalaHarris: Your voice is your vote. So get prepared for the election and make a plan for how you’ll cast your ballot. #DemConvention
@JoeBiden, Aug 20, 2020 – 3,549 Retweets – 189 Quote Tweets – 17.9K Likes18

57Trump’s conception of the platform, on the other hand, was hybrid. The “hybrid” scheme of action is defined as follows:

[It] involves formal organizational actors stepping back from projecting strong agendas, political brands, and collective identities in favor of using resources to deploy social technologies enabling loose public networks to form around personalized action themes. (Bennett and Segerberg 758)

58Trump and Biden shared some similarities in the way they used the platform to communicate. Although the organizational infrastructure of both campaigns differed from the strict “conventional configuration of a vertical party model” (Bennett and Livingston 1672), they followed a classical “up-to-bottom” dynamic, which corresponded to the definition of a “collective action scheme” (Bennett and Segerberg 758). Twitter was a “community system” for Trump too. Although more rarely than Biden, Trump shared elements from his policy agenda, congratulated Republican officials for their nominations, and announced rallies and events which took place at the local level. Trump’s campaign was strongly coordinated by his campaign staff (Tarrow 189).

59Yet, the social network was not only a vector to convey Trump’s paternal voice and urge his supporters to act offline. Rowland goes as far as saying that Trump “extend[ed] the rally atmosphere into virtual space for core supporters” (248). Thus, action was also online. If we follow Biden’s definition of the vote—“your voice is your vote”19—Trump’s conception of social media allowed Twitter itself to become a ballot box.

60In order to fully grasp this concept, it is necessary to clearly state Trump’s anti-democratic threat that loomed over the 2020 elections. The assault on the Capitol on January 6, 2021 was testimony to the constant questioning of democratic institutions by Donald Trump. Back in 2016, before he was even elected, the candidate hinted that he would not accept the results of the election. In a democracy which is grounded in Christian values (Wolfe et Katznelson), desacralizing the moment of communion that is the vote was a strong symbol: it invalidated majority rule. The interests of “the greater number” could no longer be identified with those of the majority, because the “people” remained a “heterogeneous mass,” linked by denials, frustrations, and tough situations (Rosanvallon 5). Trump’s campaign rhetoric did not transform the people into a positive voting force.

61Trump’s discourse on Twitter in 2020 suggested that the civic act was online. It was the “sharing,” as Bennett and Segerberg state, “of already internalized and personalized ideas, plans, images, and resources with networks of others” (754). This explains the overwhelming amounts of retweets that are too often overlooked by the studies of Trump’s accounts. Retweeting, which is the act of republishing somebody else’s post, was a way for Trump to encourage the self-expression of his base and to widen its audience: around 30% of his total tweet count was retweets. Trump’s communication scheme did not imply a hierarchy between his influential voice and those of his supporters, as Biden’s did. Rather, he used his Twitter account as a megaphone, where the personal initiatives of his followers were amplified and given credit. Although Trump is often described as a controlling persona, his rhetoric in 2020 gave more room to his supporters to express themselves.


62The “people” is a complicated notion, where multiple communities flow and dissonant voices are striving to be heard. In 2020, depending on their electoral strategy, the candidates adopted various communication schemes which all led to a clear definition of their conception of the American people. The communication of the candidates did produce a sense of generality and, ultimately, a “feeling of society” (Pye and Taylor 6).

63Yet, the communications of the candidates implied the construction of two very distinct versions of this generality. Through frames and storytelling, the candidates became the voices of different in-groups and made different claims. While Trump’s world was divided in terms of political parties and ideological points of view, Biden’s was structured by class and economic status.

64Both candidates targeted specific communities, in accordance with the contemporary shift to identity politics. Framing and storytelling were used as tools to shape specific in-groups and out-groups, but the use of narrativization went beyond identity politics. It favored the establishment of “representation as narrative” (Rosanvallon 192). Therefore, this study shows that the elections of 2020 demonstrated innovative uses of social media, to establish a new link with citizens. While Biden’s use of Twitter remained traditional, in line with the “collective action” underlined by Bennett and Segerberg (739), Trump allowed users to create a collaborative discourse that corresponded to a hybrid scheme of both collective and connective action. Trump’s distrust of the institutions of democracy—the Capitol and the election, in particular—led to a redefinition of citizenship.

65Contrary to popular opinion, the relation between the candidates and people on Twitter in 2020 suggests that Biden’s rhetoric was more paternalizing than unifying. Trump, however, used Twitter as an alternative ballot box. In a time when people felt that voting had become ineffective, making one’s singular voice heard on social media became a civic act. It remains to be determined, however, whether this virtual ballot box genuinely allowed the citizens’ voices to be heard, or whether it merely gave them the feeling that they were. In What Women Really Want, the book that Kellyanne Conway co-authored in 2005, the former manager of Trump’s 2016 campaign explicitly sums up this phenomenon: the internet is a place of “solitary connectedness” (Lake and Conway 24).

66Much remains to be done to complete this study. The intentions of the candidates or the dynamics that their message generated can be analyzed by looking at the meta-data associated with their posts: the number of retweets and replies can indicate a specific dynamic induced by the candidates. Interviewing the campaign teams of the candidates can also tell us a lot about their communication strategies. Yet, analyzing the text itself clearly shows how the candidates framed their discourses to fit their electoral strategies.

Haut de page


ANGLÈS D’AURIAC, Louise. “La rhétorique des peuples. La construction et la transformation du peuple américain dans les campagnes présidentielles de Brack Obama (2008) et Donald Trump (2016). ” Master’s thesis, École normale supérieure de Lyon, 2020.

AGNEW, John. “Anti-Federalist Federalism: American ‘Populism’ and the Spatial Contradictions of US Government in the Time of Covid-19.” Geographical Review, vol. 111, no. 4, 2021, p. 510-527.

AUSTERMUEHL, Frank. “The Normalization of Exclusion through a Revival of Whiteness in Donald Trump’s 2016 Election Campaign Discourse.” Social Semiotics, vol. 30, no. 4, 2020, p. 528-546.

AXELROD, David. Believer: My Forty Years in Politics. New-York: Penguin Press, 2015.

BADIOU, Alain. “Twenty-Four Notes on the Uses of the Word ‘People.’” What Is a People? Translated from the French by Jody Gladding. New York: Columbia University Press, 2016, p. 21-31.

BATTISTINI, Matteo. Middle Class: An Intellectual History through Social Sciences. Leiden: Brill, 2022.

BELLAH, Robert Neelly, Richard MADSEN, William M. SULLIVAN, Ann SWIDLER, and Steven M. TIPTON. Habits of the Heart: Individualism and Commitment in American Life. Oakland: University of California Press, 1985.

BENNETT, Lance. “The Uncivic Culture: Communication, Identity, and the Rise of Lifestyle Politics.” PS: Political Science and Politics, vol. 31, no. 4, 1998, p. 741-761.

BENNETT, Lance, and Shanto IYENGAR. “A New Era of Minimal Effects? The Changing Foundations of Political Communication.” Journal of Communication, vol. 58, no. 4, 2008, p. 707-731.

BENNETT, Lance, and Steven LIVINGSTON. “The Disinformation Order: Disruptive Communication and the Decline of Democratic Institutions.” European Journal of Communication, vol. 33, no. 2, 2018, p. 122-139.

BENNETT, Lance, and Alexandra SEGERBERG. “The Logic of Connective Action.” Information, Communication & Society, vol. 15, no. 5, 2012, p. 739-768.

BHATIA, Aditi. Discursive Illusions in Public Discourse. New York: Routledge, 2015.

BIMBER, Bruce. “The Internet and Political Fragmentation.” Domestic Perspectives on Contemporary Democracy. Ed. Peter F. Nardulli. Urbana: University of Illinois Press, 2008, p. 155-170.

BIMBER, Bruce. “Digital Media and Citizenship.” The SAGE Handbook of Political Communication. Eds. Holli Semetko and Margaret Scammell. Los Angeles: SAGE, 2012, p. 115-126.

BLUMER, Herbert. Symbolic Interactionism: Perspective and Method. Upper Saddle River: Prentice-Hall, 1969.

BONHOMME, Marc. “La rhétorique des figures : entre formalisme et énonciation.” Protée, vol. 38, no. 1, 2010, p. 65-74.

BORDEAU, Jeanne. “La véritable histoire du storytelling.” L’Expansion Management Review, vol. 129, no. 2, 2008, p. 93-99.

BRAS, Gérard. Les voies du peuple. Éléments d’une histoire conceptuelle. Paris: Liens Socio, 2018.

BRUNS, Axel. Blogs, Wikipedia, Second Life, and Beyond: From Production to Produsage. New York: Peter Lang, 2008.

BUTKOWSKI, Chelsea P. “‘If You Didn’t Take a Selfie, Did You Even Vote?’: Embodied Mass Communication and Citizenship Models in ‘I Voted’ Selfies.” New Media & Society, vol. 25, no. 9, p. 2399-2418.

BUTLER, Judith. Rassemblement. Pluralité, performativité et politique. Translated from the English by Christophe Jaquet. Paris: Fayard, 2016.

CHEVALLIER, Jacques. L’État post-moderne. Paris: Librairie Générale de Droit et de Jurisprudence, 2004.

COE, Kevin, and Rachel Alicia GRIFFIN. “Marginalized Identity Invocation Online: The Case of President Donald Trump on Twitter.” Social Media + Society, vol. 6, no. 1, 2020, p. 1-12.

COMPAGNO, Dario, Arnaud MERCIER, Julien MÉSANGEAU, and Kamel CHELGHOUM. “La reconfiguration du pluralisme de l’information opérée par les réseaux socionumériques.” Réseaux, vol. 5, no. 205, 2017, p. 91-116.

DANFORTH, John C. “Holding Ourselves Together.” Our American Story: The Search for a Shared National Narrative. Ed. Joshua A. Claybourn. Lincoln: University of Nebraska Press, 2019, p. 66-73.

DELISTRATY, Cody. “America’s Broken Narrative of Exceptionalism.” Our American Story: The Search for a Shared National Narrative. Ed. Joshua A. Claybourn. Lincoln: University of Nebraska Press, 2019, p. 201-207.

DRAPER, Robert. “The Man Behind the President’s Tweets.” The New York Times, April 16, 2018. Accessed March 26, 2024.

EDSALL, Thomas Byrne. The Point of No Return: American Democracy at the Crossroads. Princeton: Princeton University Press, 2023.

ENTMAN, Robert. “Framing: Toward Clarification of A Fractured Paradigm.” The Journal of Communication, vol. 43, 1993, p. 51-58.

FAIRCLOUGH, Norman, and Ronny SCHOLZ. “‘Critical Discourse Analysis as “Dialectical Reasoning”: From Normative Critique towards Action, by Way of Explanation’. Interview with Norman Fairclough Conducted by Ronny Scholz.” Mots. Les Langages Du Politique, no. 122, 2020, p. 113-123.

GILLIES, Jamie. Political Marketing in the 2016 U.S. Presidential Election. New York: Palgrave Macmillan Springer, 2017.

GOFFMAN, Erving. Frame Analysis: An Essay on the Organization of Experience. Boston: Northeastern University Press, 1986.

GROUPE MU. Rhétorique générale. Paris: Seuil, 1998.

HART, Roderick P. “Why Trump Lost and How? A Rhetorical Explanation.” American Behavioral Scientist, vol. 66, no. 1, 2022, p. 7-27.

IYENGAR, Shanto. Media Politics: A Citizen’s Guide. Fourth edition, New York: W.W. Norton, 2018.

JAMIESON, Kathleen Hall, and Doron TAUSSIG. “Disruption, Demonization, Deliverance, and Norm Destruction: The Rhetorical Signature of Donald J. Trump.” Political Science Quarterly, vol. 132, no. 4, 2017, p. 619-650.

JAMIN, Jérôme. “Le populisme aux Etats-Unis : la question du peuple. Première partie.” Amnis. Revue de civilisation contemporaine Europes/Amériques, no. 16, 2017, Accessed March 30, 2024.

JAMIN, Jérôme. “Le populisme aux Etats-Unis : le rejet de l’élite. Deuxième partie.” Amnis. Revue de civilisation contemporaine Europes/Amériques, no. 17, 2018, Accessed March 30, 2024.

KRUGMAN, Paul. “The Spiral of Inequality.” Mother Jones, Dec. 1996, p. 44-49.

LAKE, Celinda, and Kellyanne CONWAY. What Women Really Want: How American Women Are Quietly Erasing Political, Racial, Class, and Religious Lines to Change the Way We Live. New York: Free Press, 2005.

LAKOFF, Robin Tolmach. The Language War. Berkeley: University of California Press, 2002.

MCKINNEY, Mitchell. “Consistency and Change in Political Campaign Communication: Analyzing the 2012 Elections.” Communication Studies, vol. 64, 2013, p. 467-469.

MEEKS, Lindsey. “When Two Become One? Examining Kamala Harris and Joe Biden’s Campaign Themes from Primary to General Election.” Presidential Studies Quarterly, vol. 52, no. 2, 2021, p. 313-339.

MÉLANDRI, Pierre. “La rhétorique populiste aux États-Unis.” Vingtième Siècle. Revue d’histoire, no. 56, 1997, p. 184-200.

MÉNY, Yves, and Yves SUREL. Par le peuple, pour le peuple. Le populisme et les démocraties. Paris: Fayard, 2000.

MERCIER, Arnaud. “Twitter, espace politique, espace polémique.” Les Cahiers du numérique, vol. 11, no. 4, 2015, p. 145-168.

MERRY, Melissa K. “Trump’s Tweets as Policy Narratives: Constructing the Immigration Issue via Social Media.” Politics & Policy, vol. 50, no. 4, 2022, p. 752-72.

MEYERS, Peter Alexander, and Nancy S. STRUEVER. “Esquisse sur la modernisation de la rhétorique comme enquête politique.” Littérature, no. 149, 2008, p. 4-23.

PIZARRO-SIRERA, Margalida. “Toxic Masculinity in American Politics: Donald Trump’s Tweeting Activity in the US Presidential Election (2016).” European Journal of American Culture, vol. 39, no. 2, 2020, p. 163-181.

PYE, Danee, and Mary Anne TAYLOR. “Backward and in Converse: Artful Political Communication; How the Biden/Harris Presidential Ticket Reverses the 2-for-1 Campaign.” American Behavioral Scientist, vol. 68, no. 1, 2021, p. 56-79.

RIEDER, Bernhard, and Nikos SMYRNAIOS. “Pluralisme et infomédiation sociale de l’actualité: le cas de Twitter.” Réseaux: Communication, Technologie, Société, no. 176, 2012, p. 105-139.

ROSANVALLON, Pierre. Democratic Legitimacy: Impartiality, Reflexivity, Proximity. Translated from the French by Arthur Goldhammer. Princeton: Princeton University Press, 2011.

ROWLAND, Robert C. The Rhetoric of Donald Trump: Nationalist Populism and American Democracy. Lawrence: University Press of Kansas, 2021.

SALMON, Christian. Storytelling. La machine à fabriquer des histoires et à formater les Esprits. Paris: Découverte, 2007.

SELLERS, Jefferey M., Anders LIDSTRÖM, and Yooil BAE. Multilevel Democracy: How Local Institutions and Civil Society Shape the Modern State. Cambridge: Cambridge University Press, 2020.

SIRIN, Cigdem V., Nicholas A. VALENTINO, and José D. VILLABOBOS. Seeing Us in Them: Social Divisions and the Politics of Group Empathy. Cambridge, UK: Cambridge University Press, 2021.

SMYRNAIOS, Nikos, and Pierre RATINAUD. “Comment articuler analyse des réseaux et des discours sur Twitter. L’exemple du débat autour du pacte budgétaire européen.” tic&société, vol. 7, no. 2, 2013, ticetsociete/1578. DOI : 10.4000/ticetsociete.1578. Accessed May 9, 2024.

TARROW, Sidney. Movements and Parties: Critical Connections in American Political Development. Cambridge, UK: Cambridge University Press, 2021.

THIAULT, Florence. “Le produsage des hashstags sur Twitter, une pratique affiliative.” Questions de communication, no. 28, 2015, p. 65-79.

WALZER, Michael. “What Does It Mean to Be an ‘American’?” Social Research, vol. 71, no. 3, 1990, p. 633-654.

WOLFE, Alan, and Ira KATZNELSON, eds. Religion and Democracy in the United States: Danger or Opportunity? Princeton: Princeton University Press, 2010.

WOOD, Gordon S. “Can the United States Be One People?” Our American Story: The Search for a Shared National Narrative. Ed. Joshua A. Claybourn. Lincoln: University of Nebraska Press, 2019.

ZUCKERMAN, Ethan. “Six or Seven Things Social Media Can Do For Democracy.” My Heart’s in Accra, May 30, 2018, Accessed March 26, 2024.

Haut de page

Table des illustrations

Titre Fig. 1.
Légende Screenshots of the X-Twitter platform. Tweets published by @JoeBiden and @realDonaldTrump in May 2020.
Fichier image/png, 39k
Titre Fig. 2.
Légende Frequency analysis of personal pronouns in Donald Trump and Joe Biden’s publications on Twitter (April 18, 2019 to January 21, 2021 for Joe Biden; June 18, 2019 to January 8, 2021 for Donald Trump).
Fichier image/png, 110k
Titre Fig. 3. Nouns referring to communities and individuals mentioned by Donald Trump and Joe Biden.
Légende The percentages indicated under the number of occurrences are determined in relation to the total amount of words in the publications (386,772 for Trump and 170,911 for Biden). Capitalized and non-capitalized letters were harmonized, and plural derivatives were included in the count.
Fichier image/png, 112k
Titre Fig. 4. Nouns referring to communities and individuals mentioned by Joe Biden.
Légende For each candidate, the percentages indicated under the number of occurrences are determined in relation to the total amount of words in the publications (386,772 for Trump and 170,911 for Biden). Capitalized and non-capitalized letters were harmonized, and plural derivatives were included in the count.
Fichier image/png, 133k
Titre Fig. 5. Terminology used to qualify members of security forces
Légende For each candidate, the percentages indicated under the number of occurrences are determined in relation to the total amount of words in the publications (386,772 for Trump and 170,911 for Biden). Capitalized and non-capitalized letters were harmonized, and plural derivatives were included in the count.
Fichier image/png, 95k
Titre Fig. 6. The use of the imperative mode in the tweets published or re-published by Donald Trump and Joe Biden during their campaigns.
Légende For each candidate, the percentages indicated under the number of occurrences are determined in relation to the total amount of words in the publications (386,772 for Trump and 170,911 for Biden). Capitalized and non-capitalized letters were harmonized, and plural derivatives were included in the count.
Fichier image/png, 76k
Haut de page

Pour citer cet article

Référence électronique

Louise Anglès d’Auriac, « Connecting the American People: The Effects of New Modes of Communication during the American Presidential Campaign of 2020 »Transatlantica [En ligne], 1 | 2024, mis en ligne le 01 juin 2024, consulté le 12 juillet 2024. URL : ; DOI :

Haut de page


Louise Anglès d’Auriac

Université Sorbonne Nouvelle

Haut de page

Droits d’auteur


Le texte seul est utilisable sous licence CC BY-NC-ND 4.0. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.

Haut de page
Search OpenEdition Search

You will be redirected to OpenEdition Search