POPULISM IN BRAZIL’S 2018 GENERAL ELECTIONS: AN ANALYSIS OF BOLSONARO’S CAMPAIGN SPEECHES

POPULISMO NAS ELEIÇÕES PRESIDENCIAIS DE 2018: UMA ANÁLISE DOS DISCURSOS DE CAMPANHA DE BOLSONARO

Abstract

Through an analysis of Bolsonaro’s speeches during his official campaign, we aim to identify the presence of populist traits in his discourse. Preliminary results suggest that Bolsonaro’s discourse have, compared to its predecessors, higher levels of populism. As a theoretical framework, we use the ideational approach to populism. The data was collected and analyzed by Team Populism using the “holistic grading” textual analysis method. Results revealed that, despite his anti-elite, polarizing, and Manichean speech, Bolsonaro is an incomplete populist. In his rhetoric, populist traits vie for space with patriotic elements.

Keywords:
Ideational Approach to Populism; Populism; Bolsonaro; Populist

Resumo

Por meio de uma análise dos discursos de Bolsonaro durante sua campanha oficial, nosso objetivo é identificar a presença e intensidade de traços populistas em seus discursos. Os resultados sugerem que o discurso de Bolsonaro apresenta, em comparação com seus antecessores, níveis mais altos de populismo. Como referencial teórico, usamos a abordagem ideacional do populismo. As falas foram analisadas por meio do método de análise textual conhecido como “classificação holística” e os dados coletados e analisados pelo Team Populism. Os resultados revelaram que, apesar de apresentar um discurso antielite, polarizador e baseado em uma visão de mundo maniqueísta, o populismo, no discurso de Bolsonaro, tem presença moderada. Na sua retórica, traços populistas disputam o espaço com elementos fortemente patrióticos.

Palavras-chave:
Abordagem Ideacional do Populismo; Populismo; Bolsonaro

Introduction

With a long history of populist leaders and movements, Latin America is often considered by scholars to be the land of populism (de la Torre, 2017). In Brazil, as in other Latin American countries, populism dates back to the middle of the 20th century, being one of the primary expressions of the predominant personalism in the politics of the region.

In recent years, however, populism has not attracted much attention from Brazilian scholars. Despite this lack of interest, there is reason to suspect that the last general election in 2018 signals the resurgence of populism in Brazil, aligned with a global wave of rising populist politicians, such as Donald Trump in the United States, Viktor Órban in Hungary, Recep Erdogan in Turkey, the “Five Star Movement” in Italy, or Rodrigo Duterte in the Philippines.

When it comes to the success or failure of populist leaders, the literature divides explanations into two dimensions: demand and supply. Understanding the demand for populism is crucial if we want to understand how citizens become more likely to accept and defend populist ideals and, hence, elect populist leaders and parties. However, populist attitudes present in a society only have behavioral effects in specific contexts, where leaders operate as a “catalyst in the activation of populist attitudes” (Hawkins and Rovira Kaltwasser, 2019, p. 15).

The present study focuses on the supply aspect, analyzing campaign speeches to answer the question: To what extent is Jair Bolsonaro, the winning candidate of the Brazilian general election of 2018, a populist politician? An issue with no consensus among scholars, journalists, and politicians. For this, we cover his official electoral period, which goes from mid-July to October 26, the day of the second round of voting.

Preliminary analyses indicate that Bolsonaro’s campaign speeches present a mix of populist, patriotic and nationalist traits. His average populist score in the campaign (and even during his first months in office),¹ is higher than that of other Brazilian presidents over the past 20 years. However, it does not reach the level seen in other countries with highly populist leaders because elements of patriotic discourse crowd out the populism in most of his speeches.

The data used for the speech analysis were collected as part of an effort led by Team Populism, a team of scholars from Europe and the Americas. Collected from official campaign events and Facebook livestreams, the speeches were coded by one of the authors with the assistance of Caio Marques, a student from IE University, Spain. The present analysis was enriched by Kirk Hawkins’ comments during the coding process.

This paper proceeds as follows: the first section briefly defines populism according to the ideational approach; the second section describes the methodology; and the final section presents our findings and analyses of Bolsonaro’s speeches, starting with a score analysis and complemented by a qualitative one.

Populism: Ideational Approach

Populism is not something new. According to authors Norris and Inglehart (2019) “its historical roots can be traced back to the Chartists in early-Victorian Britain, Narodnik revolutionaries in late-nineteenth century Tsarist Russia, Fascist movements in the inter-war decades, Peronism in Argentina, and Poujadism in post-war France” (Norris and Inglehart, 2019, p. 4). As a contemporary political phenomenon, populism receives growing attention with the rise of political figures like Donald Trump in the United States, Viktor Órban in Hungary and Andrés Manuel López Obrador in Mexico. Indeed, in 2017 the Cambridge Dictionary elected “populism” the word of the year (‘POPULISM’…, 2017).

However, despite gathering much attention over the past year, as the literature shows, the concept still lacks a consistent definition. Driven by the recent re-emergence of populist forces, scholars worldwide have shifted their focus towards trying to understand populism from its core ideas (Hawkins and Rovira Kaltwasser, 2019; Mudde, 2017). Known as the ideational approach, it states that all forms of populism share one similar element: seeing the world as a Manichaean and moral struggle between good (the people) and evil (the conspiring elite). This is the concept of populism we adopt.

To Mudde (2004), populism frames the world as being ultimately divided in two different and morally antagonistic groups, “‘the pure people’ versus ‘the corrupt elite’, and argues that politics should be an expression of the volonté générale (general will) of the people” (Mudde, 2004, p. 543). Therefore, populism is born of three necessary and sufficient conditions: (i) a Manichaean and moral cosmology; (ii) the creation and defense of “the people” as a homogenous and virtuous community; and (iii) the framing of an “elite” as a corrupt and self-serving entity (Aguilar and Carlin, 2017, p. 2; Hawkins and Rovira Kaltwasser, 2019).

Populist ideas are better expressed in the rhetoric of its leaders. It is a moral discourse, potentially used by any political party or actor, since it is a centralized ideology. Meaning that populism “necessarily appears attached to… other ideologies” (Mudde and Rovira Kaltwasser, 2017, p. 6). By uniting itself with other full ideologies (like socialism and conservatism), populism can take on different forms. The main difference between these subtypes would be whom the movement portrays as “the people” and “the corrupt elite,” as they are malleable and change depending on the context (Aguilar and Carlin, 2016; Hawkins and Rovira Kaltwasser, 2019; Mudde and Rovira Kaltwasser, 2017; Müller, 2016; Kessel, 2016; Reinemann et al., 2017; Taggart and Rovira Kaltwasser, 2015).

Methodology

Bolsonaro’s speeches were coded and analyzed using theholistic grading method of textual analysis, where coders interpret whole texts instead of counting content at the level of words and phrases. We opted for this technique because it is suitable for diffuse, latent textuals meanings, common in polical discourses. As Hawkins writes: “a holistic approach works by assessing the overall qualities of a text and then assigning a single grade without any intervening calculations” (Hawkins, 2009, p. 1049).

After designing a simplified guide for evaluating the speeches (rubric), two to three coders conducted the coding process and analysis, making reliability tests along the way to ensure the results (Hawkins, 2009). For Bolsonaro’s speeches, Team Populism assigned two trained coders who participated in The Guardian’s “The New Populism” project.²

Team Populism grades the speeches on a scale of 0 to 2, where 0 is a speech with few if any populist elements and 2 is a highly populist speech, being close to the ideal populist discourse (Hawkins, 2009, p. 1062). This study follows a newer version of this scale, presented on the project made by Team Populism in partnership with The Guardian, that classifies 0 as “not populist,” 0.5 as “somewhat populist,” 1.0 as “populist,” 1.5 as “very populist” and leaves 2.0 open for what we will call “perfect populist.”

Team Populism’s rubric comprised six criteria that represent essential dimensions of populist discourse, according to the ideational approach:³

1. A Manichean view of politics and the world;

2. Use of cosmic proportions to emphasize moral significance and justify arguments;

3. Exaltation of the “will of the people,” the “people” as a legitimate source of moral and political authority;

4. Labeling minorities or the opposition as the enemy, as being evil;

5. Arguments in favor of a systemic change often expressed as “revolution” or “liberation”;

6. Adopting an “anything goes” approach towards the “enemies.”

Regarding the corpus, we selected speeches from both official events and Facebook live streams. With the exception of July (that had fewer official pronoucements), we chose at least two speeches per month: one from July, two from August, three from September (one before the knife attack against Bolsonaro and two after), and four from October, the month of the elections.⁴ The choice of grading videos does not affect the average score (Hawkins and Castanho Silva, 2019).

Results: Quantitative Analysis

Before looking at Bolsonaro’s score on the populism scale and provide a qualitative analysis of his speeches, we will briefly discuss the reliability of the process using Krippendorff’s alpha.

Krippendorff’s alpha is a coefficient developed to measure the agreement between observers and show that the results are not a product of chance. According to Krippendorff (2011), α = 1 indicates “perfect reliability”, and α = 0 the absence of reliability. We chose this method because it applies to any number of observers, categories, scale values or measures, and can also use nominal and ordinal data, as well as intervals (Krippendorff, 2011).⁵

For Krippendorff, the social sciences should rely only on variables with reliability above α = 0.8, and variables with reliability between α = 0.667 and α = 0.8 used only for “drawing tentative conclusions” (Krippendorff, 2004, p. 241). Since our α for the campaign speeches was α = 0.88, we can assume our data and analysis are trustworthy.

Table 1 shows the individual score given to each discourse, by each coder, and their average score. A quick analysis shows that Bolsonaro’s speeches grow in populism: his campaign begins with an average populist score of 0.5 and ends with 0.9, an 80% increase. Our final unit of analysis is his total average campaign speech score (0.5).

Table 1
Speech Scores

Date	Coder A	Coder B	Average / Final Score (rounded)	Title	Source:
July 7, 2018	0.4	0.5	0.5	1- PSL conference: Official launch of his candidacy for President	YouTube
August 23, 2018	0.6	0.5	0.6	2 - Bolsonaro's speech at Araçatuba	YouTube
August 31, 2018	0.1	0	0.1	3 - Bolsonaro's speech at Porto Velho, Rondônia	YouTube
September 6, 2018	0.3	0.2	0.3	4 - Speech at the Commercial Association of Rio de Janeiro (ACRJ)	YouTube
September 16, 2018	0.3	0.3	0.3	5 - Speech after being stabbed	YouTube
September 30, 2018	0.5	0.3	0.4	6 - Speech at Avenida Paulista	YouTube
October 6, 2018	0.3	0.2	0.3	7 - Speech one day before the first round of elections	Facebook Live
October 16, 2018	0.7	0.6	0.7	8 - Speech right after the first round of elections	Facebook Live
October 22, 2018	1	0.8	0.9	9 - Speech at Avenida Paulista	Facebook Live
October 27, 2018	0.8	1	0.9	10 - Speech the day before second round voting (Facebook Live)	Facebook Live
Total Average	0.5	0.44	0.5

• Source: Team Populism.6

Moderately populist, his discourse scores noticeably low at some points, and becomes more populist after October 6. The graphic below (Graph 1) illustrates this progression.

Graph 1
Progression of Populism on Bolsonaro’s Campaign Speeches

The red line indicates the turning point in his speeches, October 6, the first round of the elections. After the first round of voting (October 7), his discourse became gradually more populist going from an average 0.3 on October 6, to a 0.9 on October 27, one day before the second round.

For comparison, Table 2 shows the average scores of other Brazilian presidents over the past 24 years, as well as other international leaders. Their average scores were calculated considering speeches from their presidential terms. Team Populism coded four speeches from different categories: one famous, one international, one ribbon cutting, and one campaign speech (except for Michel Temer).⁷

Table 2
Brazil past presidents and international leaders scores

Leaders	Term	Average / Total Score
Luiz Inácio Lula da Silva	2003 - 2011	0.3
Dilma Rousseff	2011 - 2016	0.2
Michel Temer	2016 - 2018	0
Fernando Henrique Cardoso	1995 - 2003	0
International Leaders
Hugo Chávez	1999 - 2013	1.8
Donald Trump	2017 - today	0.8

• Source: Data produced by Team Populism.

His predecessors’ average scores vary from 0 to 0.3, not enough to be considered populist. However, these scores are not entirely comparable, since this study only considers Bolsonaro’s campaign speeches. Nevertheless, these scores, built on original data, offer the first overview of presidential political discourses in Brazilian politics.

About Lula, with a 0.3 average score,Hawkins (2009) writes:

none of his speeches have much in the way of a Manichaean quality. Instead, Lula tends to focus on narrow issues and avoids any kind of cosmic proportionality or the mention of historical figures. He consistently emphasizes consensus and negotiation and, while briefly criticizing some individuals or opposition groups (e.g., former president Cardoso and wealthy Brazilians), he avoids characterizing these as evil. He does make brief mention of a popular will in some of his speeches, reminding the audience of his own working-class origins and telling them that he understands their needs (Hawkins, 2009: 1056).

Dilma’s speeches, with an average score of 0.2, focus on specific issues, and often do not vilify her opponents. However, by the end of her second term, during the process of impeachment, she constantly references enemies that might be conspiring against her, subverting the system, and usurping democracy. Temer and Cardoso, with an average score of 0, do not require any detailed analysis since their scores indicate the absence or insignificant presence of populist elements in their discourses.

The second part of Table 2 (international leaders) brings a wider perspective to our analyses, as Hugo Chávez and Donald Trump are considered populist leaders. Compared to them, Bolsonaro’s campaign level of populism is low. We believe that this is due to the strong presence of patriotic and nationalist traits in his speeches.

Chávez, with a 1.8 average score, presents an actively populist discourse, close to the “perfect populism.” It displays a moral and Manichaean division between good and evil with cosmic proportions. It praises the popular will and the “people” as the true sovereigns and frames the elite as the enemy responsible for subverting the system and harming the people.

Trump presents 0.8 as the average score, closer to Bolsonaro’s. Although he displays a Manichaean view of politics and the world, praises the “people”, and is consistently against the political elite in power before his election, he is inconsistent, as Hawkins and Rovira Kaltwasser (2018) and a recent article published by The Guardian (Smith et al., 2019) reveal.

However, the present study works only with campaign speeches, not Bolsonaro’s speeches in office. Also, the selected campaign speeches were given at different places, on different platforms, and to different audiences, so they might have different framing effects that one should consider.

Even if our analyses are limited to the campaign, Bolsonaro’s speeches bring to the political scene something that was absent from Brazil for a while: populism. Nonetheless, Bolsonaro’s average score indicates that he is not a perfect or pure populist. In the next section, we analyze passages from his campaign speeches that support our statement and show the patriotism and nationalism traits that prevent him from scoring higher.

Qualitative Analysis: Bolsonaro’s populism

The qualitative analysis focuses on the three main dimensions of the ideational approach (Hawkins and Rovira Kaltwasser, 2019): (i) the praise of the popular will or the “people”; (ii) the framing of an elite as corrupt and selfish; and (iii) a Manichaean view of politics and the world. We illustrate each dimension with a few excerpts of Bolsonaro’s speeches, but the discourses were analyzed as a whole. Subsequently, we highlight nationalist and patriotic elements in his speech, further explaining why these traits are incompatible with populism in Bolsonaro’s case.

The People, The Good

One of the main dimensions of populism is the belief and praise of popular sovereignty. As the highest principle possible, the “popular will” should be respected and followed above all.

We are indeed different from those who ruled over us over the past 20 years - PT and PSDB. With us, you will be in the first place; you will be our bosses! Together we can change Brazil; we won’t have another opportunity!⁸

Nevertheless, this “people” is something malleable; it is a group that changes and reshapes according to context (Reinemann et al., 2017). In his campaign, Bolsonaro builds “people” from a cultural perspective, from traditional and religious values.

Brazil is ours, “good citizens”, workers, conservatives, Christians that preserve family values; that don’t want gender ideology in classrooms; that want Brazil doing business with the entire world, without an ideological bias.⁹

He uses “we” and “our” to highlight his identification with popular ideas among his supporters, which could reflect an attempt to build a homogeneous populist people and their “imagined community.” However, the “people” occupy a secondary position in Bolsonaro’s discourse. Its use is inconsistent and often implicit, playing a supporting role to other preferred terms (e.g., “Brazilians,” “our country,” “our nation,” “(our) Brazil”). Therefore, the “people” end up overshadowed by other elements unsuitable to this people-centrism, which we will soon discuss.

The Elite, The Enemy

Besides people-centrism, to classify someone as a populist, their speech must vilify an elite as the source of all “evil,” selfish and corrupt, responsible for conspiring against the people to usurp their power. According to Hofstadter (1996): “this enemy is clearly delineated: he is a perfect model of malice, a kind of amoral superman: sinister, ubiquitous, powerful, cruel, sensual, luxury-loving” (Hofstadter, 1996, as cited in Hawkins, 2009, p. 1044).

In his campaign discourses, Bolsonaro openly acknowledges the opposition (the left and PT) as being his enemies, addressing the PT government as corrupt, inefficient, and responsible for executing a plan to spread its ideology while in power. Also, Bolsonaro holds PT accountable for the undermining of the traditional family and its values. Bolsonaro uses belligerent language when addressing PT, becoming more aggressive as the elections approach, and openly defends non-democratic means to defeat his political enemy.

Petralhada,¹⁰ you will all go to the edge of the beach,¹¹ you won’t have any more shots in our homeland, because I will cut off all of your luxuries. You won’t have any more NGO’s to satisfy your hunger for mortadella.¹² It will be a purge never seen in the history of Brazil!¹³

At times, populists might build “the enemy” in an obscure way; however, one may identify “the enemy” by analyzing the communicative context (Reinemann et al., 2017). In the following example, it is possible to determine that the enemy is PT: “No one is going to leave this homeland, because this homeland is ours [and] not this gang with a red flag and “brain-washed” [sic].¹⁴

Manichaean Division

Praise of popular sovereignty and identifying an enemy are both necessary characteristics, but insufficient to classify someone as populist. A populist speech also depends on dividing the world into two morally opposed poles: good and evil.

But right now is polarized: it’s us and PT; it’s the Brazil green and yellow, and them, that represent Cuba, represent the Venezuelan government, with its flag that is red with a hammer and sickle on top of it. Let’s change Brazil!¹⁵

For Bolsonaro, there is a moral distinction between those who “defend the traditional family and values” and those who oppose it.

After all we are left with only two paths: the one of prosperity, freedom, family, of being on God’s side, by the side of those who have a religion and those who do not have, but are also competent [sic]; and the other we are left with the Venezuelan way. We don’t want that for our Brazil. The other candidate [the one from PT], we know who surrounds him and who he seeks advise from and where. We don’t want that kind of people back on the Palácio do Planalto [The Presidential Palace].¹⁶

Even when being vague, his use of certain expressions (e.g., “our side and their side,” “good citizens”) denotes a moral division: “The other side is the return of the past, is the corruption, the lies, the contempt of family, is the approximation of dictatorships”.¹⁷

He sees and treats the opposition as the people’s enemy, with no in-between: either you are with them, or against them.

You, Petralhas, will see a Civil, and Military police with legal rearguard to uphold law on your back. Bandits from MST,¹⁸ bandits from MTST,¹⁹ your actions will be typified as terrorism; you will not terrorize the countryside and the cities any more! Either you fit in and fall in line or you will keep company to the “drunkard” in Curitiba!²⁰

Although present, the Manichaean division is inconstant. At times, likely as a reaction to criticism, his speeches highlight the unity of the Brazilian society:

Let’s unite white and black, straight and gay, also trans, there is no problem, everyone does whatever they want, be happy [sic]. Let’s unite northeasterners and southerners, we will silence these small separatist movements we see in Brazil, unite rich and poor… Let’s unite employers and employees, not boost the discord between them. Indeed, one needs the other.²¹

Patriotism and Nationalism

Although showing the main elements of populism, Bolsonaro’s discourse does not score high on the populism scale, mainly because his speech also contains patriotic and nationalist traits inconsistent with his populism. These elements are not necessarily incompatible with populism; it is possible to identify subtypes of populism that derive from interactions between these discourses. Norris and Inglehart (2019) classify Trump’s populism as an authoritarian populism, while Jenne, Hawkins, and Castanho Silva (2019) test speeches from several world leaders for ethnopopulist frames. That is not the case for Bolsonaro. His patriotic and nationalist traits compete with populism, leaving considerably less room for the “people” in his discourse.

Both patriotic and populist speeches can present a Manichaean view of the world and politics, and an anti-elite rhetoric. The main difference is that patriotism, unlike populism, emphasizes the State. As Hawkins, Amado and Cranney (2010) state:

The State has an existence that is more independent of the individuals in it. We must all revere the state, and the state, in turn, protects and blesses us, but no one of us is a bearer of the state in the same way that we are the bearers of our national identity. (Hawkins, Amado and Cranney, 2010, p. 14)

As O’Donnell (1979) writes, the State stands above society. Nothing is above the State (Hawkins, Amado and Cranney, 2010); it goes beyond territory and institutions, it is rooted in traditions, values, and symbols like the national anthem, the flag, and its colors. Not coincidentally, Bolsonaro’s campaign motto was “Brazil above everything, God above all.”

Bolsonaro’s campaign does not claim to be the agent of the people or the guardian of the popular will, but rather Brazil’s savior, protector of the nation and the State against the enemy that has been in power for a decade. However, it is unclear whether he is referencing “the nation” or “the State” - sometimes it can be either. According to Jenne, Hawkins, and Castanho Silva (2018), nationalist discourse restricts the boundaries of a sovereign ethnos to the group considered nationally dominant. Thus, “when political actors utilize national rhetoric, this has the effect of re-inforcing or adjusting these boundaries in the public imagination” (Jenne, Hawkins and Castanho Silva, 2019, p. 8).

Drawing from these ideas and the nationalism rubric in Jenne, Hawkins, and Castanho Silva²², we identified the presence of patriotic and nationalist traits in Bolsonaro’s speeches. First, there is subtle praise of the virtues and distinctiveness of what he identifies as the “nation’s core”:

We the Brazilian people (Brasileiros) say that there is something more, way more important than our lives: our freedom! Because men or women imprisoned have no life! With your strength, let’s make Brazil! I am here because I believe in you, you are here because you believe in Brazil, this Brazil is ours! Our flag is green and yellow! [sic].²³

And what I want, if that’s God’s will, is from next year’s January, not to be an army captain anymore, but to be a soldier of our Brazil [sic].²⁴

I’m here because I believe in you, you are here because you believe in Brazil!²⁵

Table 3 illustrates this difference:

Table 3
Differences between populism, patriotism and nationalism

	Populism	Patriotism	Nationalism
Greatest value	The people	The State	The nation
The leader is	Agent of the people	Protector of the State	Savior of the Nation

• Source: Author’s elaboration.

Ultimately, the core element of Bolsonaro’s speeches is not the people, but the state and the nation. Terms like “Brazil,” “our flag,” and “nation” appear repeatedly, leaving less room for the “people”; “we” and “the people” are interchangeable with “the nation.” Bolsonaro also repeatedly references the nation’s name, “Brazil.” We argue that this is mostly because words like “people” are heavily associated with the left, to PT - his main opposition -, and becomes a strategy to distance himself from what he frames as the “enemies.”

I know what is at stake in this approaching moment: it is the destiny of this great nation called Brazil.²⁶

[Voting for Bolsonaro] You are saving mine, yours, our Brazil!²⁷

We are only one country, one homeland, only one nation, only one green and yellow heart. Together we can really make Brazil a great nation [sic].²⁸

Also present is the rhetorical frame that argues for protecting the national group status to save the whole nation (Jenne, Hawkins and Castanho Silva, 2019). This referenced group, seen in the previous quotes, is not the currently politically dominant, but the real Brazilians who are, in a profoundly conservative way, in favor of the traditional family. “We have fought against fascism, and we are fighting PT now, which is a fascist party. They lie and try to blame me for things that are their fault [sic].”²⁹

As shown, nationalist and patriotic elements coexist with populism in Bolsonaro’s discourse, but do not combine to form an “ethnopopulism” or “patriotic populism.” Although scoring 0.5 means he is “somewhat populist,” his use of “people-centrism” is inconsistent, as illustrated by our analysis.

Concluding remarks

In light of all the analyzed data, can we describe Bolsonaro as a populist leader? Although presenting a low average populist score of 0.5, Bolsonaro’s campaign speech exhibits all three main dimensions constitutive of populism. As Mudde and Rovira Kaltwasser (2013) state, the peculiarity of populism as a set of ideas lies precisely in the combination of these three elements.

However, Bolsonaro is far from being a “perfect populist.” The patriotic and nationalist elements in his discourses eclipse the “people”, lowering his populist score.

Our analysis focuses on his presidential campaign and should not be used to draw any final profile regarding his ongoing presidential term. As Hawkins, Amado and Cranney (2010) argue, campaign speeches may be more populist than speeches in government since both frame conditions and external contexts are prone to change. Discourses given on different platforms may exhibit different levels of populism since they are designed for different audiences and, therefore, written for different purposes (Wiesehomeier, 2019).

Our next step involves expanding our analysis to cover Bolsonaro as president. Initial research by Team Populism and The Guardian show that he maintained a score of 0.5 even after elected; compared with other Brazilian presidents (Temer, Dilma, Lula, and Cardoso), in his first months in office, he already scores higher than his predecessors.

Finally, examining Bolsanaro speeches is the first step towards a better understanding of what happened in Brazil in the polarized 2018 general elections. Brazil might be a case in which demand for populism meets supply, but this can only be understood by comparing the supply of populists with the demand for populism among voters, what requires analysis of public opinion data. We hope that our study can contribute to the debate about populism and its different manifestations both in Brazil and worldwide.

Bibliography

• AGUILAR, Rosario; CARLIN, Ryan. 2017. Ideational populism in Chile? A case study. Swiss Political Science Review, v. 23, n. 4, pp. 404-422.
• DE LA TORRE, Carlos. 2017. Populism in Latin America. In: ROVIRA KALTWASSER, Cristóbal; TAGGART, Paul; ESPEJO, Paulina O.; OSTIGUY, Pierre (ed.). The Oxford handbook of populism. Oxford: Oxford University Press. pp. 250-274.
• HAWKINS, Kirk A. 2009. Is Chávez populist? Measuring Populist discourse in comparative perspective. Comparative Political Studies, v. 42, n. 8, pp. 1040-1067.
• HAWKINS, Kirk A.; AMADO, Mayavel; CRANNEY, Stephen. 2010. Beyond populism: a typology of elite political discourse with some examples. Paper presented at the Annual Meeting of the American Political Science Association, Washington, DC, September 1-5.
• HAWKINS, Kirk A.; CASTANHO SILVA, Bruno. 2019. Textual analysis: big data approaches. In: HAWKINS, Kirk A.; CARLIN, Ryan; LITTVAY, Levi; ROVIRA KALTWASSER, Cristóbal (ed.). The ideational approach to populism: concept, theory, and method. New York: Routledge. p. 27-48.
• HAWKINS, Kirk A.; ROVIRA KALTWASSER, Cristóbal. 2018. Measuring populist discourse in the United States and beyond. Nature: Human Behaviour, v. 2, pp. 241-242.
• HAWKINS, Kirk A.; ROVIRA KALTWASSER, Cristóbal. 2019. Introduction: the ideational approach. In: HAWKINS, Kirk A.; CARLIN, Ryan; LITTVAY, Levi; ROVIRA KALTWASSER, Cristóbal (ed.). The ideational approach to populism: concept, theory, and method. New York: Routledge . pp. 1-41.
• JENNE, Erin; HAWKINS, Kirk A.; CASTANHO SILVA; Bruno. 2018. Populism and Nationalism in Central and Eastern Europe. Paper presented at the workshop “Future of Political Parties”, Berlin, May 17-18.
• JENNE, Erin; HAWKINS, Kirk A.; CASTANHO SILVA; Bruno. 2019. What’s in a speech? Mapping populist and nationalist rhetoric of government leaders across Europe and North America. Paper presented at APSA Annual Meeting 2019, Washington, DC, August 29-September 1.
• KESSEL, Stijn van. 2016. The surge of ‘exclusionary’ populism in Europe. Comparative Politics Newsletter, v. 26, n. 2, pp. 81-86.
• KRIPPENDORFF, Klaus. 2011. Computing Krippendorff’s alpha-reliability. Departmental Papers, pp. 1-10.
• KRIPPENDORFF, Klaus. 2004. Reliability. In: KRIPPENDORFF, Klaus. Content analysis: an introduction to its methodology. 2. ed. Thousand Oaks: Sage. pp. 211-257.
• LEWIS, Paul; CLARKE, Seán; BARR, Caelainn. 2019. How we combed leaders’ speeches to gauge populist rise. The Guardian, 2019 Mar. 6. Available from: Available from: https://bit.ly/34fw7SW Access in: 2020 Apr. 6.
  » https://bit.ly/34fw7SW
• MUDDE, Cas. 2004. The Populist Zeitgeist. Government and Opposition, v. 39, n. 4, pp. 541-562.
• MUDDE, Cas. 2017. Populism: an ideational approach. In: ROVIRA KALTWASSER, Cristóbal; TAGGART, Paul; ESPEJO, Paulina O.; OSTIGUY, Pierre (ed.). The Oxford handbook of populism. Oxford: Oxford University Press . pp. 45-70.
• MUDDE, Cas; ROVIRA KALTWASSER, Cristóbal. 2013. Populism. In: FREEDEN, Michael; STEARS, Marc (ed.). The Oxford Handbook of political ideologies. Oxford: Oxford University Press . pp. 578-600.
• MUDDE, Cas; ROVIRA KALTWASSER, Cristóbal. 2017. Populism: a very short introduction. Oxford: Oxford University Press .
• MÜLLER, Jan-Werner. 2016. What is populism? Philadelphia: University of Pennsylvania Press.
• NORRIS, Pippa; INGLEHART, Ronald. 2019. Cultural backlash: Trump, Brexit, and authoritarian populism. Cambridge: Cambridge University Press.
• O’DONNELL, Guillermo. 1979. Tensions in the bureaucratic-authoritarian state and the question of democracy. In: COLLIER, David (ed.). The new authoritarianism in Latin America. Princeton: Princeton University Press. pp. 285-318.
• ‘POPULISM’ revealed as 2017 Word of the Year by Cambridge University Press. 2017. University of Cambridge: News, 2017 Nov. 30th. Available from Available from https://bit.ly/2xOxFao Access in: 2020 Apr. 3rd.
  » https://bit.ly/2xOxFao
• REINEMANN, Carsten; AALBERG, Toril; ESSER, Frank; STRÖMBÄCK, Jesper; VREESE, Claes H. 2017. Populist political communication: toward a model of its causes, forms, and effects. In: AALBERT, Toril; ESSER, Frank; REINEMANN, Carsten; STRÖMBÄCK, Jesper; VREESE, Claes H. (ed.). Populist political communication in Europe. New York: Routledge . pp. 12-29.
• SMITH, David; LEWIS, Paul; HOLDER, Josh; HULLEY-JONES, Frank. 2019. The Teleprompter Test: Trump’s rightwing populism not his own. The Guardian, 2019 Mar. 6. Available from: Available from: https://gu.com/p/akc76?CMP Access in: 2020 Apr. 6.
  » https://gu.com/p/akc76?CMP
• TAGGART, Paul; ROVIRA KALTWASSER, Cristóbal. 2015. Dealing with populists in government: some comparative conclusions. Democratization, v. 23, n. 2, pp. 345-365.
• WIESEHOMEIER, Nina. 2019. Expert surveys. In: HAWKINS, Kirk A.; CARLIN, Ryan; LITTVAY, Levi; ROVIRA KALTWASSER, Cristóbal (ed.). The ideational approach to populism: concept, theory, and method. New York: Routledge . pp. 90-112.

• 1
  Research conducted by Team Populism together with The Guardian where Bolsonaro’s first month in office speeches were coded.
• 2
  Project that coded speeches from leaders around the world and produced the “Global Populism Database”: the “most up-to-date, comprehensive and reliable repository of populist discourse in the world” according to The Guardian website (Lewis, Clarke and Barr, 2019). The group (44 coders) coded 886 speeches in total, and achieved a high level of intercoder reliability, with a 0.824 Krippendorff’s alpha.
• 3
  For more information, see the complete rubric at Hawkins (2009).
• 4
  For access to the data, contact one of the authors.
• 5
  For more information on Krippendorff’s alpha, see Krippendorff (2011).
• 6
  To simplify, we rounded the average score to only 1 decimal value.
• 7
  For more information about discourse classification, see ^{Appendix 1}.
• 8
  Bolsonaro (2) - Araçatuba (2018).
• 9
  Bolsonaro (9) - Av. Paulista (September, 2018).
• 10
  Reference to PT affiliates, a wordplay with the words “Metralha” from the Brazilian Portuguese translation of the Beagle Boys (mobsters) + Petista (member of PT).
• 11
  Reference to a place where political prisoners were taken during the military dictatorship.
• 12
  Leftist activists are also called “mortadella sandwich”.
• 13
  Bolsonaro (9) - Av. Paulista (October, 2018).
• 14
  Bolsonaro (9) - Av. Paulista (October, 2018).
• 15
  Bolsonaro (7) - One day before first round (2018).
• 16
  Bolsonaro (8) - Right after the first round of elections (2018).
• 17
  Bolsonaro (10) - One day before second round of elections (2018).
• 18
  Landless Worker’s Movement.
• 19
  Homeless Workers Movement.
• 20
  Bolsonaro (9) - Av. Paulista (October, 2018).
• 21
  Bolsonaro (1) - Official launch of his President candidacy (2018).
• 22
  For the complete classification and the rubric, see Jenne, Hawkins and Castanho Silva (2019).
• 23
  Bolsonaro (1) - Official launch of his President candidacy.
• 24
  Bolsonaro (2) - Araçatuba (2018).
• 25
  Bolsonaro (9) - Av. Paulista (October, 2018).
• 26
  Bolsonaro (1) - Official launch of his President candidacy.
• 27
  Bolsonaro (9) - Av. Paulista (October, 2018).
• 28
  Bolsonaro (2) - Araçatuba (2018).
• 29
  Bolsonaro (1) - Official launch of his candidacy for President.

Appendix 1

Criteria for Selecting Speeches

Generally, we need a speech that is at least 2-3 pages long, or about 2,000 words, in order to have enough text to analyze. We will use an extremely long speech (>5 pages) if it is the only one available in the category or is clearly the right speech for that category (as in the case of a famous speech), but given a choice, we prefer something shorter to make our work a little easier. We will also use an extremely short speech (1 page or less), but only if it is the only speech available. When the leader has been in office several years (for example it is the last year in a 6-year term) and there are a variety of speeches available for a category, we generally prefer the most recent ones because they are the easiest to find. To ensure comparability of coding across speeches and leaders, we need to have transcriptions rather than video recordings.

Campaign

Here we ask for a speech given during this chief executive’s latest campaign for office. Keep in mind the above criteria, especially length. Campaign speeches are often the hardest to find because they were given before the person was elected, and so they are usually not recorded on any government website. Be prepared to call the political party or the office of the chief executive to speak to someone who was involved in the campaign. If it is impossible to get a speech for the person’s own campaign, we will take a speech that he/she gave for some other candidate’s campaign (for example, for members of the legislature during a mid-term election). If several speeches are available, we prefer the closing speech of the campaign to the opening speech, and a speech given to a large public audience over one given at a party convention.

Ribbon-cutting

This is a speech given at some kind of public ceremony dedicating a government building or project, typically a road, park, or building. You will likely find a number of these on the government website. Given a choice, look for a speech that is given to a small, local audience rather than a national one, and to a domestic audience rather than an international one-we prefer something obscure in order to see whether the chief executive uses a populist discourse in settings with little apparent significance. If you have a lot to choose from, pick the most recent.

International

Here we are looking for a speech whose primary audience, or a significant part of the audience, consists of citizens from other countries-leaders, diplomats, or even ordinary people. There will be quite a few international speeches available, including on non-government websites. For consistency, we encourage you to look for a speech given outside the country, with as small of a domestic audience as possible. UN speeches are especially good as long as they are long enough.

Famous or most popular

In this category, we seek for a speech that is widely regarded as one of the best-known and most-popular speeches given by this leader. Of course, some leaders don’t give very popular speeches, but we at least want one of their best-known speeches. As someone who knows this country well, you are in a good position to pick what you think is a particularly appropriate speech. We encourage you to contact the office of the chief executive or the political party and ask them for a recommendation. They will often suggest an inaugural speech (when the chief executive actually took office) or an annual report to the nation, but not necessarily. You should not feel obliged to use one of these particular speeches if you know of another one that is more famous (or notorious). Talk to a couple of people if you feel unsure.

Source: Team Populism

Publication Dates

• Publication in this collection
  05 June 2020
• Date of issue
  Jan-Apr 2020

History

• Received
  11 Feb 2020
• Accepted
  19 Mar 2020