Populism on the Web: Presidential Elections in Bolivia, Ecuador, Peru, and Colombia (2020–2022)

Abstract

Populism has become one of the main features of political action worldwide. This research aims to characterize the populist discourse in the tweets of presidential candidates in the Andean Community in recent elections (2020–2022). Accordingly, we analyze the characteristics of their social network profiles, as well as the content and latent discourse of their tweets. We demonstrate that the differences and similarities of their discourse go beyond their right and left association. The differences result from how they construct their identity and establish their relationship with their electorate. Our analysis reveals that this type of discourse is ideological as well as performative. It is ideological because, in the candidates’ discourse, they recontextualize the actual meanings of “us” and “them”. It is performative because it is carried out by a charismatic leader who acts in a specific way to define himself or herself as the embodiment of “the people” and “the good”.

1. Introduction

1.1. Latin American Populism in Its Communicative Dimension

In recent years, worrying movements and dysfunctional leaders have emerged worldwide, with some authors describing them as “populist” [1]. These leaders, under the pretext of restoring power to the “people”, question institutions, constitutional practices, and the essential elements that characterize a liberal democracy [2,3].

Latin America has a long populist tradition arising from both left- and right-leaning parties. It is clearly manifested in their political communications [4]. The concentration of economic and political power among a small minority has caused such “populist” discourse to be attractive to many. In this discourse, a “fraudulent oligarchy” or “corrupt elite” is generally identified as a group acting against the wishes and needs of the “people”. Therefore, in free and fair democratic elections, voters express their dissatisfaction by choosing populist leaders [5].

Carrión [6] argues that Latin American populism is opposed to political representation, a characteristic of traditional democracy. Traditional political representation, unlike populism, allows pluralism and does not seek to delegitimize opponents through an “us vs. them” approach. Nevertheless, according to Carrión [7], populism can operate in democracies, with elections being necessary to achieve and exercise power. Therefore, populism is a political strategy and is not necessarily associated with nondemocratic regimes. Jiménez and Patarroyo [8] affirm that populism manifests itself mostly as a performative act... and its effective action manifests itself in a discursive dimension.

In Latin America, there have been three populist waves [5]. The first wave began with the Great Depression in 1929 and extended to 1960. The characteristic of this wave was an increase in rural migration to urban areas with the implementation of economic reforms to foster industrialization. These measures led to demands for political and social rights, with socialism and communism acquiring great strength. In this period, the ideas of the “pure people” and the “corrupt elite” were constituted. De la Torre and Srisa-nga [3] call this stage “classic populism”, and it is represented by, e.g., Domingo Perón (Argentina), José María Velasco Ibarra (Ecuador), and Lázaro Cárdenas (Mexico). In the 1930s, we find, as well, the first manifestations of populism in the president of Brazil, Getulio Vargas. The second wave, the neopopulist stage [3], was shorter and emerged in 1990. At that time, many Latin American countries were suffering from a deep economic crisis; hence, leaders such as Carlos Menem (Argentina), Fernando Collor de Mello (Brazil), and Alberto Fujimori (Peru) won their elections after blaming the economic situation on the political and economic elite. However, these leaders cooperated with the International Monetary Fund (IMF) to implement neoliberal reforms [5,9]. Finally, the third wave, radical left populism [3], began with Hugo Chávez (Venezuela) in 1998 and expanded with the electoral triumphs of Evo Morales (Bolivia), Rafael Correa (Ecuador), and Daniel Ortega (Nicaragua). These leaders employed anti-imperialist rhetoric as the backbone of their discourse. In Latin America, there has been a traditional propensity to exploit socialist and communist ideas [5], which has led to a series of nationalizations.

It is worth considering the current characteristics of political leaders: do leaders from the right or the left express novel populist features? In this article, we analyze the discourse of eight Latin American political leaders to derive empirical evidence to determine these characteristics. The main aim is to compare different features of populism and contrast to what extent left- and right-wing candidates might vary in that regard.

We selected these countries based on a criterion of geographical nature (the area shared by the Andean Mountain range in South America), called “Andean countries”, because they have common sociopolitical, economic, and identity aspects [10].

1.2. Twitter as a Populist Space?

Nowadays, social networks in emerging countries provide venues to maintain genuine political discussions [11]. Campos-Dominguez [12] argues that on Twitter, spontaneity and immediacy prevail, which can promote the fluid exchange of conversation and political debate among its users, although political actors typically maintain low involvement in any debate they motivate. However, Arroyas-Langa, Martínez-Martínez, and Berná-Sicilia [13] affirm that the brevity and immediacy that characterizes Twitter hinder the necessary contextualization for the proper understanding and reflection on actual events in the public sphere.

In general, social networks [14] structure digital communities, orienting topics, and establishing a particular dynamic for political competition. In terms of Latin America, López-López and Vásquez-González [15] studied the role of thematic agendas and Twitter in the 2015–2017 presidential elections in Latin America, finding a cohesive theme that pervaded the left–right and national axes. Regarding European populist leaders, Alonso-Muñoz and Casero Repolles [16] analyzed the agenda setting and the “more is less” effect on Twitter. They showed that on Twitter, there was a low degree of thematic fragmentation, the launch of proposals instead of attracting voters, and a strong negative correlation between the number of published tweets and user interest.

Research demonstrates that political polarization arises from the discursive strategies employed by political actors in building their agenda [17]. Populism is expressed in a very fragmented way across social networks. It is usually used (less on Twitter than on Facebook) by political extremists, particularly from the opposition [18].

Populist communicative practices have developed, mutated, and adapted in today’s hybrid media society [19]. Likewise, political actors generate a hybridization of new and old media via Twitter [20]. Their Tweets embody a performative ideology that requires further analysis to reveal its full potential [21].

Thus, it is necessary to interrogate how the populist discourse on Twitter functions as populist rhetoric in the Latin American context. Accordingly, we follow Waisbord and Amado [22], who suggest that Twitter does not cause significant changes in political communication. Instead, it facilitates a vertical approach to populism without debate or dialogue. Its use has primarily involved harassing critical journalists, social network users, or citizens.

In this study, we explore the populist discursive strategies of candidates in Andean Community electoral processes, specifically, the two top-ranked candidates in the latest elections in Bolivia (2020; Luis Arce and Carlos Mesa), Ecuador (2021; Guillermo Lasso and Andrés Arauz), Peru (2021; Pedro Castillo and Keiko Fujimori), and Colombia (2022; Gustavo Petro and Rodolfo Hernández). Our analysis will identify the nuances of their discourse and the populist dimensions within it.

1.3. Theoretical Approaches to Populism

One of the main problems of populism, at a theoretical level, is what Sartori [23] described as “conceptual stretching”, that is, using the same concept to apply it to objects with different attributes. One way to overcome this problem is to situate it on a strictly political plane, which can be summarized in three approaches.

First, from the ideational approach, Mudde and Kaltwasser define populism as a thin-centered ideology that separates society into two homogeneous and antagonistic camps: the “pure people” and the “corrupt elite” [5,24]. This definition allows a certain malleability since it provides a thin center. It explains why populism combines with other complete or “thickly” centered ideologies. These thick or strongly centered ideologies (on the left or right) thus assimilate populist ideology.

Within this definition, populism constructs the identities of the “people” and the “ elite” as opposite and homogeneous poles. In this opposition, the “people” are common people who embody the nation having sovereign power. They are constituted as an expression of the “general will.” The concept of “elite” is defined from a “moral” perspective. The “elite” not only ignores people’s interests but also acts against them. From this perspective, the reality is constructed in a Manichean way (black vs. white; people vs. elite) and lacks gradation.

Second, there is a strategic approach that focuses on the pursuit and maintenance of power through political action and mobilization. It is a behavioral approach that differs from the more ideological vision [25]. Through this focus, De la Torre [2] also affirms that populism polarizes society into two antagonistic fields; however, this definition pays more attention to political discourse and strategies. This author does not present populism as an “ideology” but as a “populist strategy” that aims to break down the entire institutional system through polarization. Because the “people” comprise an amorphous collective seeking to solve problems, they need a “charismatic leader” to mobilize and organize them. This leader is seen as the one who expresses the “true voice” and represents the “true people”. He or she defines who is “part” of the nation that must govern the “whole”, embodying the voice of the excluded “majority”. Anyone who dares to question the leader is classified as an “enemy” of the nation [26,27]. The leader’s goal is to restore power to the people. To achieve this, he or she constantly confronts the “enemy” using a discursive repertoire full of symbolic terms.

Third, from the discursive–performative approach, the people–elite division becomes “us” vs. “others”. The struggle for hegemonic power generates this division [28]. Among the authors that most influenced this perspective, we find Laclau and Mouffe [29,30,31]. Laclau’s theory focuses on how a populist “discourse” shapes and defines a social reality constructing a political agency. In this context, discourse is a social practice that implies a dialectical relationship between a particular discursive event and situations, institutions, and social structures [32]. This populist discourse does not have a certain a priori normative content, as proposed by the ideational approach [27]. Instead, the concern is how popular identities and identifications are constructed throughout the discourse [1].

For Laclau [33], the leader’s identity crystalizes through discourse. The possibility that the leader’s name does not mean anything increases, thus becoming an “empty signifier”. This makes the leader politically powerful: he or she allows the people to project many meanings, emotions, and demands on him or her. Thus, the leader will mean different things to different groups in different contexts. Moreover, each group will believe that they grasp the “true” meaning.

More recently, Ostiguy and Moffitt [34,35] revised this idea: the populist leader functions as an “overflowing signifier”. The leader is not an empty signifier because he or she already possesses certain particularities that connect him or her to the people. The people, in turn, possess multiple interpretations and readings of the leader. Thus, the authors highlight the “relational” aspect of the leader–people dyad, which involves constant feedback. Therefore, this perspective incorporates not only ideological aspects but also sociocultural, stylistic, and performative elements [21].

2. Materials and Methods

This is a descriptive analytical study that explores the characteristics of the populist political discourse of presidential candidates voiced on Twitter. We focus on the two top-ranked candidates from the last presidential elections in the Andean region: from Bolivia, Luis Arce and Carlos Mesa (2020); from Ecuador, Guillermo Lasso and Andrés Arauz (2021); from Peru, Pedro Castillo and Keiko Fujimori (2020); and from Colombia, Gustavo Petro and Rodolfo Hernández (2022) (

Table 1. Object of study.

Candidate	Twitter	Time Analyzed	Tweets Posted	Tweets Analyzed
BOLIVIA
Luis Arce	@luchoxbolivia	20 September to 18 October 2020	292	87
Carlos Mesa	@carlosdmesag		106	44
ECUADOR
Guillermo Lasso	@lassoguillermo	14 March to 11 April 2021	2356	164
Andres Arauz	@ecuarauz		284	125
PERU
Pedro Castillo	@pedrocastillote	9 May to 6 June 2021	84	31
Keiko Fujimori	@keikofujimori		72	37
COLOMBIA
Gustavo Petro	@petrogustavo	29 May to 19 June 2022	851	200
Rodolfo Hernández	@ingrodolfohdez		239	168
		Total	4284	856

Source: Own elaboration.

). The selection of only two candidates is justified in three aspects: first, the presidential logic of the political system itself; second, the polarizing reasoning of the social network; and finally, the need to delimit the object of study.

Our analysis follows a mixed methodology based on two techniques: (1) the characteristics of their social network (Twitter) via the Fanpage Karma tool [36] and (2) content and discourse analyses using previously defined categories derived from Charaudeau [37] and De Bruijn [38]. These authors define the characteristics of populism and politicians’ expressions. We summarize them in

Table 2. Categories and definitions.

CATEGORY	DEFINITION
1. Catastrophe (crisis)	The populist discourse tries to exploit existing resentment in the population; the populists discuss economic the crisis and social and political chaos due to corruption. The aim is to describe a disastrous situation and define oneself as “the savior”.
2. Victims	“People” or the candidate are victimized by the existing conditions or actions carried out by their opponent(s) or the current government.
3. Villains	Villains are characterized as those who have acted or act against the “people” or the candidate to cause them physical, moral, or economic damage.
4. Heroes	The candidate or his/her group defines himself/herself as a savior for the crisis.
5. Candidate’s values	Positive values of the candidate are exalted to build a positive representation of himself or herself.
6. Opponent’s values	Negative values of the opponent are exalted constructing a negative representation of the other.
7. Playing with opposite perspectives	Positive aspects of the candidate are exalted, while negative aspects of the other are exalted.
8. Exaltation of national values	Positive national values are exalted.
9. Exaltation of regional people’s values	Positive regional values are exalted.
10. Exaltation of the local people’s values	Here, we must differentiate “pueblo” from “gente”. “Pueblo” is a collective entity that highlights the people’s town, village, and/or their working-class values. Left-wing politicians frequently use the term “pueblo”. However, some right-wing politicians use the word “gente” to refer to a confederation of individuals. “Gente” emphasizes individual civil rights.
11. 3P Model: Principles, politics, personal experiences	The personal experiences and principles of the candidate as if they were those of the “real people”, which gives him/her the “real authority” to carry out policies. Here, the identity of the candidate or opponent is related to their possible future actions in the government.
12. Meta-framing	It consists of changing how arguments are framed. Moving, for example, from what affects the candidate to what affects the rival. The focus of attention thus shifts from candidate to rival.
13. Emotions	Discourse generates positive emotions toward the candidate or negative emotions toward the opponent.
14. Promises	The candidate offers to make changes (social, economic, or structural) through actions, policies, etc.
15. Clientelism	The candidate offers excessive money, goods, etc., to specific groups in exchange for their votes.

Source: Charaudeau [37] and De Bruijn [38].

: Categories and definitions. We follow the methodology described by van Dijk [39,40,41] for our discourse analysis.

The software Fanpage Karma generates several indicators to analyze a Twitter network: the number of fans or followers, number of posts, total reactions, comments shared, and follower growth. It also provides a measure of engagement, defined as the average amount of how often a fan interacts with a page’s post, calculated by dividing the daily number of reactions, comments, and shares by the number of fans. To measure post interactions, it provides an average number of all interactions, comments, and shares per fan per post.

The corpus for our analysis was established for each candidate’s account with the registry in Fanpage Karma. This record was then saved in Microsoft Excel. In the first stage, we reviewed all messages issued by a candidate. In the second stage, tweets without value or not related to the process were discarded (thanks, greetings, and names). Thus, the specific corpus of our analysis was set. We processed all the information through a template (derived from Table 2), allowing us to note each category of analysis. In this manner, we defined the quantitative (to make the content analysis) and qualitative (to make the content and discursive analysis) aspects of each candidate’s discourse to delineate their common patterns. To compare the populism categories of each candidate and their relative weights, we performed bivariate correlations using the Pearson correlation coefficient for an ordered relationship based on previous investigations [42,43,44].

We analyze the tweets from four weeks before the last elections, regardless of if the election was the first or second round. However, in the case of Colombia, we only used three weeks since it was the time between the first and second rounds.

Our research questions, based on our theoretical approach, are as follows:

• How is the “identity” of the political actors (leader, people, and others) constructed in the candidates’ discourse on Twitter?
• What actions, purposes, norms, and values are attributed to the political actors (leader, people, and others) in the candidates’ discourse on Twitter?
• What are the characteristics of the leaders’ discourse on Twitter when it refers to the people, the region, or the nation?
• How are the relationships of leader–people and people–others constructed in the candidates’ discourse on Twitter?
• Are there any differences in the discourse based on ideological or national axes?

Charaudeau [37] has highlighted the characteristics of populism, which are in accordance with what the authors propose in the ideational, strategic, and performative-discursive theory. These are the characterization of an economic crisis and social and political chaos due to corruption, the victimization of the candidate and what they consider the “true people”, and the characterization of the candidate as a hero and of the others as villains. These must be regarded as the core features of populism, which creates an “us vs. them” discourse. Nevertheless, various discursive strategies are additionally used to achieve the objective of building this polarized world. These are the exaltation of the candidate’s positive values, the exaltation of the opponent’s negative values, the playing with opposites’ perspectives [38], the exaltation of people’s values, nationalism, regionalism, the 3P model [38] (the personal experiences and principles of the candidate as if they were those of the “real people”, which gives him or her the “real authority” to carry out policies), meta-framing [38], emotions, promises, and clientelism. The core discursive characteristics of populism and the other discursive strategies have been considered categories (Table 2) for analyzing the candidates’ discourse.

The left/right positioning of the candidates has been determined by the self-identification of the candidates and their voters and historical-analytical aspects defined by academics [45]. What usually defines the left in Latin America is the approach of a solid and interventionist state in social and mainly economic aspects. The extreme manifestation of this interventionism is the nationalization of private companies, for example, mining companies, arguing that the raw material they extract from the subsoil belongs to the people. Instead, the right tends to raise the free market and, as far as possible, the slightest intervention of the State. Conservatism, in social terms, characterizes the extreme right.

3. Results

3.1. Context: Characteristics of the Candidates’ Twitter Account (Table 3)

We summarize our results measuring the characteristics of each social network in Table 3. We display them explicitly for each candidate.

Table 3. Basic indicators of social media content.

Profile	Fans	Engagement	Post interactions	Total: Reactions Comments Shared	Number of Posts	Follower Growth (Absolute)	Follower Growth (In%)
Luis Arce	53.107	10.99%	1.09%	142,784	292	11,673	28.1%
Carlos Mesa	635,083	0.46%	0.13%	84,197	106	5826	0.3%
Andres Arauz	113,056	27.02%	2.76%	836,444	284	12,661	12.61%
Guillermo Lasso	716,529	4.37%	0.05%	874,052	2356	41,098	6.08%
Pedro Castillo	85,307	16.84%	5.81%	325,332	84	-	-
Keiko Fujimori	1,152,959	1.61%	0.65%	529,593	72	36,830	3.3%
Gustavo Petro	5,230,212	9.33%	0.24%	10,444,622	851	302,644	6.14%
Rodolfo Hernández	391,721	26.51%	2.44%	2,011,890	239	170,646	77.19%

Fanpage Karma [36].

Table 3. Basic indicators of social media content.

Profile	Fans	Engagement	Post interactions	Total: Reactions Comments Shared	Number of Posts	Follower Growth (Absolute)	Follower Growth (In%)
Luis Arce	53.107	10.99%	1.09%	142,784	292	11,673	28.1%
Carlos Mesa	635,083	0.46%	0.13%	84,197	106	5826	0.3%
Andres Arauz	113,056	27.02%	2.76%	836,444	284	12,661	12.61%
Guillermo Lasso	716,529	4.37%	0.05%	874,052	2356	41,098	6.08%
Pedro Castillo	85,307	16.84%	5.81%	325,332	84	-	-
Keiko Fujimori	1,152,959	1.61%	0.65%	529,593	72	36,830	3.3%
Gustavo Petro	5,230,212	9.33%	0.24%	10,444,622	851	302,644	6.14%
Rodolfo Hernández	391,721	26.51%	2.44%	2,011,890	239	170,646	77.19%

Fanpage Karma [36].

In the Bolivian case, Arce’s number of posts is about three times Mesa’s number of posts (292 vs. 106), but Mesa has a much larger number of followers (635,083 vs. 53,107 for Arce). Nevertheless, the total number of reactions, comments, and shares is much higher for Arce than for Mesa. This latter feature translates to a greater engagement of Arce’s followers (here, engagement is the average obtained from the total posts’ interaction on the number of followers). Furthermore, Arce’s follower growth increased at a faster rate. These statistics thus suggest that Arce’s followers identified more strongly with him than the followers of Mesa.

In the Ecuadorian case, the number of followers of Guillermo Lasso is also larger than that of Andrés Arauz (716,529 vs. 113,056). Considering the total reactions, comments, and shares, the difference between the candidates is not as radical as in the Bolivian case. In the Ecuadorian case, these totals are significant, even though they are slightly larger for Lasso than Arauz. The number of Arauz’s followers is much lower, but the number of interactions among them is very high. This entails much higher engagement with this candidate. Likewise, Arauz’s follower growth rate was much faster than Lasso’s; however, it did not lead to a larger number of followers. Lasso differs from all candidates by the large number of posted tweets (2354, with the next one being Petro from Colombia with 851 tweets). However, this is not reflected in engagement (4.37% vs. 27.02% for Arauz). His many tweets are primarily personal appreciation and responses to his followers. He creates a new tweet rather than responding to an existing one. Therefore, these new tweets have a low value as interactions. Then, although Lasso’s engagement is not reflected directly in the statistics, it should not be immediately concluded that Lasso lacks strong connections with his followers.

In the Peruvian case, Fujimori’s number of followers is much greater than that of Castillo (1,152,959 vs. 85,307). This may be due to the time Fujimori has been on Twitter. Despite her number of followers, Fujimori has not used Twitter extensively (she only has 72 posts), although Castillo also has a comparable number of posts (84). The total number of reactions, comments, and shares is larger for Fujimori (529,593 vs. 325,332 for Castillo), but Castillo’s follower engagement is far more intense (16.84% vs. 1.61% for Fujimori). Castillo’s fans are much more engaged, considering their number of followers. Perhaps because of the impact a new leader (an “outsider”) has on people’s hopes.

For the Colombian case, we find that Petro has a much larger number of followers than Hernández (5,230,212 vs. 391,721). The total number of reactions, comments, and shares for Petro is huge compared to all of the candidates under study (10,444,622). The number of tweets posted is also higher for Petro than for Hernández (851 vs. 239). However, Hernández’s follower engagement is almost ten times larger (26.51% vs. 0.24% for Petro). This could also be (as in the case of Castillo) because he appears as a “new leader”, an “alternative” to other politicians. Although Hernández’s follower growth rate is higher than Petro’s, the total number of followers remained low, perhaps because Petro has had his Twitter account for much longer.

We noticed no horizontal or two-way dialog between candidates and followers, although they have a high number of reactions, comments, and shares. The relevant reactions and comments of the followers are for tweets posted vertically by the candidates. Usually, the candidates are not interested in discussing fundamental social or economic problems with them.

3.2. Results by Category

We use our previously defined categories to carry out the discourse and content analysis. In

Table 4. Populist categories of left-wing candidates.

Rank	Castillo		Arce		Araoz		Petro
	Category	Percentage	Category	Percentage	Category	Percentage	Category	Percentage
1	Heroes	71%	Emotions	49%	Emotions	80.8%	Candidate’s values	66%
2	Candidate’s values	71%	Promises	48.3%	Candidate’s values	77.6%	Emotions	47%
3	Promises	71%	Candidate’s values	37.9%	Promises	73.6%	Catastrophe	30.5%
4	Catastrophe	51.6%	Catastrophe	34.5%	Catastrophe	49.6%	Heroes	28.5%
5	Emotions	48.4%	Victims	24.1%	Heroes	46.4%	Opponent’s values	27%
6	Playing with opposite perspectives	41.9%	Exaltation of regional values	16.1%	Victims	42.4%	Promises	25.5%
7	Villains	38.7%	Exaltation of local people’s values	14.9%	Villains	40%	Villains	24%
8	Victims	29%	Exaltation of national values	14.9%	Opponent’s values	35.2%	Playing with opposite perspectives	23%
9	Opponent’s values	29%	Villains	14.9%	Playing with opposite perspectives	20%	Exaltation of local people’s values	21.5%
10	Exaltation of local people’s values	25.8%	Heroes	14.9%	Exaltation of local people’s values	16%	Victims	16%
11	3P Model	16.1%	Opponent’s values	6.9%	Exaltation of regional values	11.2%	Exaltation of regional values	12%
12	Exaltation of national values	9.7%	3P Model:	2.3%	Clientelism	4%	Exaltation of national values	2.5%
13	Meta-framing	6.5%	Playing with opposite perspectives	1.1%	3P Model	2.4%	Clientelism	2.5%
14	Exaltation of regional values	3.2%	Meta-framing	0	Exaltation of national values	1.6%	3P Model	1.5%
15	Clientelism	0	Clientelism	0	Meta-framing	0	Meta-framing	1%

Source: own elaboration.

and

Table 5. Populist categories of right-wing candidates.

Rank	Fujimori		Mesa		Lasso		Hernández
	Category	Percentage	Category	Percentage	Category	Percentage	Category	Percentage
1	Heroes	59.5%	Catastrophe	59.1%	Emotions	23.2%	Candidate’s values	55.4%
2	Candidate’s values	59.5%	Villains	59.1%	Exaltation of national values	23.8%	Promises	36.9%
3	Emotions	54.1%	Emotions	52.3%	Promises	20.1%	Emotions	21.4%
4	Promises	40.5%	Opponent’s values	50%	Candidate’s values	17.7%	Heroes	18.5%
5	Catastrophe	35.1%	Promises	38.6%	Heroes	4.3%	Villains	17.9%
6	Victims	32.4%	Heroes	36.4%	Catastrophe	2.4%	Playing with opposite perspectives	16.7%
7	Clientelism	32.4%	Candidate’s values	27.3%	Exaltation of regional values	1.2%	Opponent’s values	11.9%
8	Opponent’s values	29.7%	Playing with opposite perspectives	22.7%	3P Model	1.2%	Victims	7.7%
9	Villains	18.9%	Victims	13.6%	Meta-framing	0.6%	Catastrophe	6.5%
10	Playing with opposite perspectives	13.5%	Meta-framing	11.4%	Playing with opposite perspectives	0.6%	Exaltation of national values	6%
11	Exaltation of local people’s values	5.4%	Exaltation of local people’s values	6.8%	Victims	0	Meta-framing	3.6%
12	Exaltation of regional values	2.7%	Exaltation of regional values	6.8%	Villains	0	Clientelism	3.6%
13	Meta-framing	2.7%	Exaltation of national values	2.3%	Opponent’s values	0	Exaltation of regional values	1.8%
14	Exaltation of national values	0	Clientelism	2.3%	Exaltation of local people’s values	0	Exaltation of local people’s values	1.2%
15	3P Model	0	3P Model	0	Clientelism	0	3P Model	1.2%

Source: own elaboration.

, we have put the categories for the left- and the right-wing candidates from the highest percentage to the lowest. The “rank” represents the position of the category in relation to each candidate’s other categories. Number “1” is the category with the highest percentage and number “15” is the category with the lowest percentage.

As we can see in Table 4 and Table 5, in the first eight positions of the rank are what we have defined as the core of populism. It determines the “us vs. them”. In these positions, we see the importance of representing society and the economy as catastrophic, the existence of villains, the victimization of the people, and the candidate as a hero.

This analysis (Table 4) shows that a high percentage of tweets from left-wing candidates express an ongoing catastrophic situation in their countries: Castillo (51.6% with a rank of 4), Arce (34.5% with a rank of 4), Arauz (49.6% with a rank of 4), and Petro (30.5% with a rank of 3). This is one of the fundamental characteristics of populist discourse: to draw attention to corruption, the economic crisis, or a lack of democracy. Using this description of reality, the candidates present themselves as “saviors” and “heroes”; particularly, Castillo and Arauz stand out in this category (Table 4). Castillo also tries to form the idea that he personifies the “people” (“el pueblo”) and that he himself, together with the “people”, will write history.

In this populist discourse, the description of the catastrophic situation leads to the definition of “culprits” (the “villains”). We find high numerical values for this description, mainly for Castillo (38.7%, although with a rank of 7), Arauz (40% with a rank of 7), and Petro (28.5%, but with a rank of 4). Generally, in these tweets, the “culprits” are the opponents or groups perceived as supporting these opponents.

In several cases, the blame goes to a “corrupt elite”. This is the case of power groups and the binomial Keiko/Alberto Fujimori for Castillo; “bankers” and the “rich” for Arauz; and “corrupt” and “fascists” for Petro. Thus, the opposition of “us” vs. “them” is generated. The populist discourse “victimizes” the addressed group, using these symbolic representations of heroes and villains. We find that all candidates from the left victimize their target voters to a greater or lesser extent.

Populists construct an identity of themselves by alluding constantly to their own values. For example, they exalt their “honesty”, their “capacity for work”, or how they share similar values with the “people”. Regarding this, we find a very high percentage among all the candidates: Castillo 71% with a rank of 2, Arce 37.9% with a rank of 3, Arauz 77.6% with a rank of 2, and Petro 66% with a rank of 1. They also draw attention to their opponent’s values but to a lesser extent. Nevertheless, this characteristic is present among almost all candidates, mainly Castillo (29% with a rank of 9), Arauz (35.2% with a rank of 8), and Petro (27% with a rank of 5).

In some tweets, the candidates compare their positive features with the negative aspects of the opponents and their group. In this manner, all candidates “play with opposite perspectives”. This occurs mainly among Castillo (41.9% with a rank of 6), Arauz (20% with a rank of 9), and Petro (23% with a rank of 8).

The case of Arce is slightly different, as he makes fewer direct allusions to his opponent. However, he does it indirectly, repeatedly mentioning that “we have to recover what was lost”. This is reflected in lower scores for the categories of “villains” (14.9% with a rank of 9), “opponent’s values” (6.9% with a rank of 11), and “playing with opposite perspectives” (1.1% with a rank of 13). Nevertheless, what is essential for Arce is a strategy to obtain identification with national and autochthonous symbols, as shown in his scores on regional values (16.1% with a rank of 6), popular values (14.9% with a rank of 7), and national values (14.9% with a rank of 8). The exaltation of the people and the region is also used by Arauz and Petro, but not as much as Arce (Table 4). It is remarkable that for Castillo, the category “local people’s values” has a higher score than regional or national values. He constantly seeks to merge himself with his potential electorate (mainly from the inland) by exalting the values of the “village people” and the working class (“El Pueblo”). In symbolic terms, he constantly tells us, “I am the people” (“Yo soy el pueblo”). For example, he speaks of his personal experiences and principles being a “rondero” (a traditional countryside patrolman) and teacher as he presents himself as “of the people”. This conveys the idea that he has the authority and courage to carry out specific policies. Perhaps, for this reason, the “3P model” category has a relatively high score for Castillo (16.1% with a rank of 11).

Finally, all candidates scored very high in the “emotions” category. This reflects a strategy that seeks identification with the candidate and/or rejection of the opponent. We can find this mainly for Arce (49.4% with a rank of 1), Arauz (80.8% with a rank of 1), and Petro (47% with a rank of 2). The “promises” category is also well represented among Castillo (71% with a rank of 3), Arce (48.3% with a rank of 2), and Arauz (73.6% with a rank of 3).

Regarding the right-wing candidates (Table 5), Lasso’s case is distinctly different. His speech lacks some of the characteristics of populist discourse. For example, Lasso does not victimize the population or refer to his competitor in his tweets. In this sense, he avoids the “us vs. them” discourse, preferring to highlight his values and national values. However, the other right-wing candidates reflect many characteristics of the populist discourse.

Mesa presents himself as a “savior” (36.4% with a rank of 6) and repeatedly defines his opponent as a “villain” (59.1% with a rank of 2). Fujimori extols her personal values (59.5% with a rank of 2) and describes herself as a “hero” who stands up to the supposed “communist” onslaught (59.5% with a rank of 1). Hernández also makes a constant exaltation of his own values (55.4% with a rank of 1). He defines himself as a “hero” (18.5% with a rank of 4), mentioning those he considers “villains” (17.9% with a rank of 5).

To present themselves as “heroes,” they describe the country’s situation as “catastrophic”. They frequently allude to how “corrupt” and “violent” their opponents are. The right-wing candidates constantly warn that a disaster will take place if the other candidates are elected. For instance, they affirm that the opponents will increase poverty by implementing their “communist” policies. We see this mainly among Mesa (59.1% with a rank of 1) and Fujimori (35.1% with a rank of 5).

The exaltation of “national”, “regional”, and “local people’s” values is very different in left- and right-wing candidates. They are less present on the right (except for Lasso) than on the left. For Fujimori, the exaltation of “national” values is not “stated” in her tweets; however, she attempts to generate a specific nationalist identification by wearing the national team’s shirt. She thus seeks identification with the “nation” at a performative level. This is a type of approach also used by Castillo, who wears his village’s hat in every situation.

Finally, “emotion” is decisive for all right-wing candidates (Fujimori 54.1% with a rank of 3, Mesa 52.3% with a rank of 3, Lasso 23.2% with a rank of 1, and Hernández 21.4% with a rank of 3), as well as the “promises” category. What distinguishes Fujimori from all the other candidates is using “clientelism” in her discourse (32.4% with a rank of 7).

To compare discursive patterns and the affinity between left- and right-wing candidates, bivariate correlations have been used depending on the intensity of each variable. This methodology has been tested in previous research with interesting results [42]. We found a strong correlation (

Table 6. Correlation (N = 18) using Pearson’s correlation coefficient.

		Castillo	Arce	Arauz	Petro	Fujimori	Mesa	Lasso	Hernández
Castillo	Correlation	1
	Sig. (Bilateral)
Arce	Correlation	0.54 *	1
	Sig. (Bilateral)	0.037
Arauz	Correlation	0.86 **	0.75 **	1
	Sig. (Bilateral)	0.00	0.00
Petro	Correlation	0.86 **	0.64 *	0.18 **	1
	Sig. (Bilateral)	0.00	0.01	0.00
Fujimori	Correlation	0.79 **	0.51*	0.89 **	0.82 **	1
	Sig. (Bilateral)	0.00	0.05	0.00	0.00
Mesa	Correlation	0.73 **	0.71	0.77 **	0.81 **	0.64 *	1
	Sig. (Bilateral)	0.00	0.08	0.00	0.00	0.01
Lasso	Correlation of	0.71	0.61*	0.11	0.39	0.30	0.22	1
	Sig. (Bilateral)	0.08	0.02	0.13	0.15	0.28	0.44
Hernández	Correlation	0.83 **	0.86	0.81 **	0.79 **	0.78 **	0.71 **	0.47	1
	Sig. (Bilateral)	0.00	0.07	0.00	0.00	0.00	0.00	0.08

Source: own elaboration. * Correlation is significant at the 0.05 level (bilateral). ** Correlation is significant at the 0.01 level (bilateral).

) between Arauz and all the candidates, except for Lasso, which is logical due to the dynamics of national competition. The greatest intensity occurs with the Peruvian leftist candidate (0.92 **). He has significant correlations with all others except for Guillermo Lasso. Fujimori has a strong correlation with Castillo, Arauz, Petro, and Hernández. In the dimension of significance, those with the lowest correlations are Lasso and Arce. We find indistinct correlations among candidates from the left and the right because populism is present in all of them; however, the significant one is slightly higher in terms of the ideological key. The only candidate who obtains low levels of significance with the rest is Guillermo Lasso, which can lead to two interpretations: (a) his “way” of constructing populist discourse is different, or (b) he does not directly express populist discourse.

3.3. Results by Competition Axes

3.3.1. Bolivia

Luis Arce: Nationalism and Identity

Luis Arce generates an identity not around himself but around Bolivian culture. Arce exalts, throughout his tweets, regional and ancestral values. This is achieved through mentions and visuals of traditional symbols used in popular and indigenous art. This gives his campaign an intensely emotional aspect.

The candidate exalts himself by using praise and association with Bolivian cultural diversity. For example, in several tweets, we can see Arce singing popular music with a guitar in hand, merging, in this way, with symbols with which people identify. The candidate’s strategy is to empower himself by constantly appealing to regional and national symbols.

Arce constantly mentions that the economy is in crisis and that once he is elected, “wealth… will return to Bolivian families without distinction” (Arce, 10.-14-20). He presents the economy in relation to past nationalizations. In a tweet, he posts a video praising the nationalizations during the government of Evo Morales. He says that Bolivia “will” become a producing country and imports will be substituted, alluding, in some way, to what has been “lost”. He affirms that the past “was better” (with Evo Morales) and that it is necessary to “recover” it, of course, through him since he embodies Morales’ vision and values. He also refers constantly to the “lost democracy”. A recurrent phrase is: “recover democracy and the homeland”, and then “happiness” and “smiles” will be recovered too.

In many tweets, he presents himself as a victim of the current government, labeling it a “de facto” government. He maintains that this government does not respect human rights, denouncing political persecution against him. He also affirms that the government does not facilitate transparency in the electoral process, floating the idea of potential fraud. Despite this, he calls for national unity.

Carlos Mesa: Citizen Community vs. Corrupts

In Carlos Mesa’s tweets, there is no nationalist discourse as in Arce’s tweets. It is worth pointing out that Mesa does not refer to the Bolivian population as “el pueblo” but as “la gente”, in a “citizen community”. “Pueblo” highlights the people’s town, village, and/or the working-class values. Left-wing politicians frequently use the term “pueblo”. However, “gente” mainly refers to a confederation of individuals. Mesa, as well as other right-wing politicians, use this term to emphasize individual civil rights.

Mesa does not make a recurrent appeal to Bolivian culture or popular art, such as music, as Arce does. Instead, he creates an identification with the electorate in a different way. To instill a particular emotional identification, he insists that the “citizen community”, “la gente”, are “responsible” for “change” for ending fraud, corruption, and the socio-economic crisis. Mesa poses himself as “the savior” who will save Bolivia from the abyss. Through this kind of discursive construction, Mesa puts forward the idea that he has the authority to “ask” for the population’s vote.

He repeatedly makes references to the government of Evo Morales. Mesa remarks that Bolivia today is in crisis because of the policies implemented during 14 years of Evo Morales’ government. He identifies Morales with Arce; for him, they are the same. Mesa affirms that if Arce is elected, Morales will return. He characterizes Arce/Morales as abusers of power who are violent and produce division in Bolivian society. In this version of reality, according to Mesa, he will prevent disaster from happening again, ensuring that those responsible will be justly punished for their actions.

3.3.2. Ecuador

Guillermo Lasso: The Least Populist

Guillermo Lasso is the focal candidate with the least number of populist characteristics. He is also the candidate who made continuous and systematic use of the social network, having identified 2353 tweets. Many of these are brief acknowledgments or responses. Using this constant and personal gratitude, he tries to build the idea that he appreciates his electorate.

He seeks identification with the other, speaking more of “us” than “me”. In doing so, he presents himself as part of a group with common goals expressing phrases such as: “we will achieve”, “we will guarantee”, and “with us”. However, there are certain moments when the candidate speaks of himself, establishes personal traits, and takes ownership of his proposals.

One of the most recurrent notions in Lasso’s discourse is “change”. He even uses “change” as part of a frequently used hashtag. He establishes that “change” should be a national value, a necessity, and a collective goal. For Lasso, this historical change must imply a “new stage” and “a worthy future”. He proposes what he calls “a better Ecuador”, implying that under the other candidate, it will be worse than it already is.

Lasso constantly uses the image of a country progressing towards “a dream”. He maintains that Ecuador should be a space for encounters, freedom, prosperity, and opportunities. A recurring feature is his concept of unity when talking about various social, regional, or minority groups (“together”, “united”, and “You and I dream of it”). He alludes simultaneously to the ideas of diversity and equality. At the same time, he incorporates in his discourse minorities or vulnerable groups, proposing recognition of their rights and freedoms.

Lasso tries to connect emotionally with voters. He is always grateful. He presents himself as someone who can be trusted and will work hard to create opportunities and well-being. Lasso proposes himself as a hero, affirming that he will defend and monitor the democratic process.

Andres Arauz: Rich vs. Poor

Andres Arauz changed strategies in the second round of the election [46]. In his tweets, Arauz used a strategy of “us” vs. “them”. He personified this in the case of “bankers” vs. “people” or, more directly, in the case of “rich” vs. “poor”. He associates the “bankers” with the president at that time, Lenin Moreno, and his opponent, Guillermo Lasso. For Arauz, Moreno/Lasso are the ones who unleashed the economic crisis. To characterize the “bankers” as villains, he accuses them of abandoning the poor and subjecting Ecuadorians to new debts. He constantly mentions that the Ecuadorian population is a victim of the “bankers” and defines the people as “abandoned”, “in pain”, “lacking quality services,” “indebted,” and “suffering.” In this way, he creates a people’s representation of helplessness, appealing to a deep sense of vulnerability. Another way he frames the “bankers” and the “rich” as “villains” is by asserting that their main goal is to keep their “privileges” without any consideration for the well-being of Ecuadorian society.

Arauz builds the idea that he will be the “change”. Therefore, he will bring economic progress and social well-being. He presents himself as an ally and “savior” of the people. However, he generally does not speak in the first person, using “we” to address the audience (“we” are going to be the agents of change). In addition, Arauz tries to legitimize his claims, pointing to his advanced degree and experience as a former banker. This should make him knowledgeable in these matters.

He makes recurrent allusions to the various regions of Ecuador, describing Ecuador as a “plurinational and intercultural state”. In this manner, he tries to establish a close relationship and identification with the indigenous population.

3.3.3. Peru

Pedro Castillo: The People vs. the Elite

In Pedro Castillo’s campaign, the leader’s identity plays a fundamental role. His discursive strategy consists of exalting his values positively, contrasting with the negative values of his opponent. Therefore, his discourse has a constant play of opposite perspectives.

In his tweets, he tries to build the idea that he and the people are one or the same. He defines himself by his many roles: a father, a teacher, a “rondero” (or patrolman), and a farmer. In addition, he defines himself as “a family man”, a person that shows solidarity with the people of his town, a man of tradition who, at the same time, strives for progress and modernization. His main goal is to achieve identification with the “people”.

He presents himself as an energetic actor facing crime and corruption. He repeatedly affirms that the past was “dark” and that history must be ruptured. He presents himself as a “hero” who will reformulate the socio-economic scenario and change the existing conditions. He defines himself as a “savior” of Fujimori’s past victims, the only one who dares to face this “evil”. Therefore, he will repair what has been done by the Keiko/Alberto Fujimori binomial.

Castillo constantly represents Keiko Fujimori as a “villain” using the play of opposite perspectives. He presents Keiko Fujimori as the past (“dark ages”) who will carry out forced sterilizations, disappearances, and human rights violations as her father did. He even refers to Keiko Fujimori as a “bad daughter” by accusing her of being an accessory to the torture and mistreatment her father inflicted on her mother. On the contrary, Castillo presents himself as the one who embodies morality (good father, good husband, honest, etc.) while identifying with those who have been victims of corruption and abuse. He raises the idea that he and Fujimori’s regime victims are one or the same. With this argument, he tries to attract the anti-Fujimori vote.

Castillo uses the 3P model (personal experience, principles, and politics). For example, in a tweet, he identifies with those who suffered forced sterilization, pointing out that his family has suffered the same crime (personal experience). Therefore, he will ensure justice and reparations (principles), making the State recognize its responsibility (political). In another tweet, he affirms that as a farmer and rondero (personal experience), he defended people from the mining companies hurting the environment (principles) and will continue to do so (politics).

In this way, he seeks identification with the “people” for who he was and is and what he does. To achieve identification, he also generates the opposition of “us” (Castillo = Pueblo) vs. “them” (Keiko/Alberto Fujimori and the corrupt elite). He presents “them”—Keiko and Alberto Fujimori and all power groups—as corrupt. Business groups such as CONFIEP, and corporations such as Odebretch, together with congress members and politicians, are a single corrupt unit. Castillo defines himself as the alternative to corrupt and hegemonic power.

Peru is characterized as a “being” capable of writing its own “history.” Peru is the people, and the people are Castillo. Ultimately, Castillo is the one who will be able to “write history”. Castillo identifies only with the “people”, with those who have been “forgotten”. Therefore, the rest of the Peruvian population is left aside. “They” are not Peru.

Keiko Fujimori: Democrats vs. Communists

Keiko Fujimori tries to win voters using the old strategy of clientelism. In this sense, she does not propose building a close and horizontal relationship with the electorate; she does not intend to merge herself with the “people” as Castillo does. Through this clientelist strategy, she establishes a vertical relationship where she presents herself as the one with the power to “give” money, goods, and services to citizens as if it were a gift and not their unquestionable right.

In our analysis of videos posted in her tweets, we find her exalting the conservative values of Peruvian society, presenting herself as a “good mother”, “good wife”, and a defender of the family unit in traditional terms. She also presents herself as the only one who will defend peace and “unity” among Peruvians. Keiko Fujimori constantly calls for the “unity of Peruvians”.

Like Castillo, Keiko Fujimori plays with opposite perspectives. She creates the idea that if you vote for Castillo, for example, everything will go up in price because he will stop imports and, consequently, there will be greater poverty. With her, however, the population will receive money, tablets, computers, land, mining distributions, etc.

Her goal is to raise fear in the population. She represents Castillo as promoting “class struggle”. She associates Castillo with “violence”, the “radical left”, and “communism”. In a video posted by her on Twitter, she calls on women to “save the country” by voting for her, repeating it several times throughout the video. She tells them to “not be afraid”; she would be a “savior”, as the one who faces the “evil”. The image established is of support in the face of imminent danger. In some way, she assumes the figure of the “hero” in the face of a threat.

Throughout her tweets, she claims that she is a victim not only of Castillo but also of the judiciary. The money laundering investigations on her are just another form of persecution, according to her. However, she builds the image that she is confronting Castillo, complying with the law, but being treated unfairly by the judiciary.

Keiko Fujimori compares the electoral competition with a soccer match. She always wears Peru’s national team shirt, constantly writing in her tweets that “we are going to turn this game around”, thus appealing to emotion. She intends to assume the symbol of the nation as her own. The team’s colors correspond to the Peruvian flag, a trait of nationalism.

At various times, she questions the current government and the neutrality of the National Electoral Jury (Jurado Nacional de Elecciones, JNE) in the elections. There are constant allusions to possible foul play, a narrative that was later used to question the process.

3.3.4. Colombia

Gustavo Petro: The Working Class vs. the Fascists and the Corrupt

Throughout his tweets, Gustavo Petro establishes a close relationship with the “working class” population. In his videos on Twitter, we find him in inland regions, in agricultural areas, or visiting various factories where he interacts with workers. They explain to Petro the production processes and the problems they encounter. Petro shows great empathy with them, raising possible solutions when faced with a problem. In this way, he tries to achieve identification with “each group” in particular. Therefore, Petro does not treat the “people” as a homogeneous group; instead, he seeks identification with each working-class group dealing with specific problems. He does the same with each different group or town. After presenting himself as someone who understands a particular situation, he makes a promise. In this way, the promise does not appear “hollow”, “empty”, or “clientelist” but as a possible course of action if elected.

He shows authority advising on how to solve the problems, but at the same time, he tries to merge with the other: “he is like one of us”. In his tweets, he shows photographs of himself living and sleeping in the inland regions the way local people do. Symbolically, this states that Petro and people living anywhere are “equals”. He also creates this sensation by affirming that “he will always be with them”.

He tries to define himself not as a radical leftist politician but as one who respects democracy, the constitution, and private investment. In this way, he deflects the criticisms from his opponent.

Petro plays with opposite perspectives. He accuses Hernández of being “corrupt”, “fascist”, and “a bad professional”. In contrast, he defines himself as “honest”, “a worker who understands other workers”, and “a person who understands each group” with their own “peculiarities and problems”. Additionally, he criticizes the electorate that supports Hernández and classifies them as admirers of “Nazism”, “irrational”, and “retrograde”. In opposition to those who follow him, who are “progressive” and rational.” He sets up the idea that only with him there will be a “historical change”; everything will be different. Jobs will be available and, above all, there will be peace. He also dedicates many tweets to defending himself against the accusations and threats of the opposition.

Rodolfo Hernández: The Outsider

Rodolfo Hernández is defined by his opponents as antipolitical, a feature that seems consolidated in his speech. He considers himself as an “outsider” from a badly entrenched political class.

In Hernández’s case, the leader is the axis. According to him, he represents the country directly, distinguishing himself from the right- and left-wing politicians. He sets up the idea that he is a leader who follows his own path. Faced with the possibility of alliances in the second round, he quickly and repeatedly distances himself from former president Alvaro Uribe, strongly asserting his autonomy.

The permanent use of the first person becomes evident. He constantly personalizes every action and purpose. He often uses more “me” than “we”. The idea of “teamwork” rarely appears in his discourse. Only when the candidate wants to identify himself with the nation or the people, he uses “we”.

He accuses the opponents of being guilty of all the country’s problems. They are responsible for all crises, violence, poverty, and corruption and for disrupting the country. He constantly places corruption as a feature of the other. To affirm this, he mixes diverse actors, sometimes politicians in general, and sometimes a particular group. In contrast, Hernández presents himself as the one who will achieve change and defeat that enemy. He affirms that the system, the state captured by his enemies, must be defeated. He will represent all Colombians (entrepreneurs, the people, women, young people, etc.), ending all abuse of power.

In his discourse, he sees two extreme positions: one positive as a symbol of what is desirable and the other linked to hatred and rupture. National unity becomes a central theme, a brand that defines him. He promises an ideal future, moving away from an evil “past” defined by either right- or left-wing policies. He creates the image of an opponent who attacks, transgresses, and lies. On many occasions, Hernández victimizes himself, claiming his life is in danger from the enemy.

Faced with previous misogynistic expressions and discriminatory words towards the LGBTQ community and minorities, Hernández tries to turn the criticisms around. Instead, he frequently mentions that he is a defender of women’s rights and vulnerable minorities, portraying himself as a guarantor of their rights.

In most tweets, Hernández makes promises by speaking in the first person. He portrays himself as a generous leader who will personally support those in need. In some cases, his tone sounds patronizing.

In some ways, he is a contradictory figure. On the one hand, he appeals to religion and traditional family values. On the other hand, he is for egalitarian marriage, gender equality, and the legalization of abortion.

4. Discussion

All the candidates, both from the right and the left (except for Guillermo Lasso), express almost all the characteristics of populist discourse in their tweets. The difference between the right and left is conveyed differently. In the case of leftist candidates, national, regional, and/or local people’s values are exalted; this is an essential factor in building their identity. In the tweets of Arce, Arauz, and Petro, the people are not homogenized; however, in those of Castillo, there is a tendency to homogenize the people by symbolically merging them with him (into what he is). For the right-wing candidates, however, only Lasso expressly highlights national values.

To a greater or lesser extent, there is a catastrophic description of reality: the villains are guilty, the “people” are victims, and the candidate is a “savior”. Therefore, the populist strategy of “us vs. them” is one of the most common characteristics of their tweets. These elements could be part of future research addressing how these aspects could be related to the pernicious polarization McCoy and Somer [47] pointed out, especially the construction of the other in terms of a villain. From the authors’ perspective, a degree of polarization is natural and is part of the spaces of distinction between the different parties. The most worrying aspect of what is called “pernicious polarization” is its prolongation which reinforces prejudices and biases between social groups, becoming the “other”, an “enemy” to be defeated.

We must conclude, agreeing with Waisbord and Amado [22], that Twitter by itself does not necessarily promote horizontal dialogue and interactivity; instead, as a medium, it has taken on specific characteristics that reinforce a top-down relationship of the candidates with their audience. Our analysis of Twitter reveals a polarized discourse that does not help the consolidation of democracy but quite the opposite. For Mansbridge and Macedo [48], however, the core of populism can be good for democracy. For them, the core elements are “(a) the people, (b) in a morally charged (c) battle against (d) the elites”. But, as they say, the associated characteristics are often dangerous for democracy. They are “(a) homogeneous people, (b) an exclusive people, (c) greater direct popular rule, and (d) nationalism”. They affirm that three other characteristics accompany these elements: “(a) the embodiment of the people in a single leader, (b) opposition to vilified vulnerable out-groups, and (c) the valorization of the authentic folk knowledge of the people with the devaluation of deliberation and expertise”. We can see most of the characteristics these authors consider dangerous for democracy when we analyze the discourse expressed on the candidates’ Twitter accounts (Section 3.3).

Substantial differences are present in how the candidates build their identities and establish relationships with the population. In the case of Pedro Castillo, for example, the identity factor is set up around him. Castillo presents himself as belonging to the “people” (he “is” the “people”); therefore, he has the authority to speak and act on their behalf. In contrast, he suggests that Keiko Fujimori and her group do not have this authority because they are not from the “people.” In this way, he delegitimizes the voice of Fujimori as being one of “them”.

Should we consider the “us vs. them” division an “ideological” one? Ostiguy, Panizza, and Moffit [35] suggest that this division is political and sociocultural rather than ideological. For him, it is also strategic because it operates as a discourse with a type of logic and rhetoric. Kissas [21] asserts that populism is ideological as well as performative. In their discourse, the populists manifest an ideology through which they build their subjective identity. Therefore, populism must be conceived as a performative ideology. It is ideological because, in this discourse, there is a recontextualization of the existing meanings about who “we” are and who “our” enemies are. It is performative because it is carried out through a charismatic leader who defines himself or herself as the one who manifests or embodies, through speech and symbolic representations, the “us”, the “people”. In this sense, there is no dichotomy between performativity and ideology. As Kissas states, performativity is ideological.

Castillo’s representation is in line with not only the “strategic approach” [3] but also the “ideological performative approach”. Castillo’s voice is constructed as “the true one”, as the one who embodies the “true people”. His body is the expression of the struggle for liberation. Symbolically, he represents the people through his clothing. His hat, seen in the photographs, alludes to the inland country. His words are directed at the excluded, the forgotten, the discriminated, and the “pure people” who fight against a “corrupt elite”, as Mudde and Rovira Kaltwaser [5] would say.

The discourse expressed in these tweets should not be considered simple words. Their analysis shows how the leader establishes his or her relationship with the population. For Castillo, this identity is drawn from who he is. However, for the other candidates, the identity of the “people” and the leader are built “in the relationship”. This is mainly the case for Lasso, Arce, Arauz, and Petro. The discursive–performative–ideological theoretical proposal can be of great help: it is not about the existence of pre-given identities but rather those that are patiently established through the representations of the relationship [1], as well as the actions of the candidates. We see, for example, Arce interacting with groups of the population using their vehicles of identification (their culture, music, rituals, etc.). He sings their songs and participates in their rituals. Therefore, in this relationship, he is not defined in isolation but in relation with the other. The same thing happens with Petro.

The “way” of the candidate’s relationship with the population is undoubtedly based on “content”. Fujimori relates to the electorate through a clientelist practice, using a style and rhetoric that expresses superiority over the other and a vertical power relationship. Alternatively, there is a discursive strategy that generates emotion. For example, “fear” creates an identity of a “defenseless” population, which inevitably needs a “savior”. This “way” of relating is at the same time “content”; in this manner, a type of subjectivity is built in the electorate. This reiterates the claim that performativity is ideological.

Therefore, what distinguishes populism is the imaginary [49] construction of an “us vs. them” (be it right or left). It is on this basis that a specific relationship is established. Populism is not necessarily focused on the leader alone but constitutes an ongoing relationship that develops over time. This could be a future line of investigation for populism in social networks.