Rage within the machine: the engagement and radicalization of party members and leaders in Brazil¹

Abstract:

What are the drivers of party engagement in the context of increasing distrust and hostility toward political parties? In this article we argue that hatred for political parties is a significant factor that influences party engagement in today’s political landscape, in which parties are often seen as illegitimate by the public. We collected observational and experimental data through an extensive survey of members of 32 Brazilian political parties to examine how incentives can condition the participation of party members and leaders. The study’s primary findings reveal that hatred against a political party can be a powerful driver of engagement. By highlighting this contemporary feature of party activism, we hope to contribute to the debate on the challenges that democratic systems face with the rise of the far right.

Keywords:
political parties; party engagement; selective incentives; collective incentives; far right

Resumo:

O que motiva o engajamento partidário em um cenário marcado pelo desgaste da imagem dos partidos políticos? Neste artigo, argumentamos que uma dimensão de natureza negativa atua como vetor relevante para a compreensão do engajamento em sistemas políticos contemporâneos, marcados pelo aprofundamento do descrédito dessas organizações junto à sociedade. A partir de dados observacionais e experimentais, coletados por meio de um survey inédito aplicado a membros de 32 partidos políticos brasileiros, examinamos como incentivos reconhecidos pela literatura condicionam a participação de filiados e dirigentes. Os resultados evidenciam que o ódio a um partido opera como um influente catalisador do engajamento, contribuindo para o debate sobre os desafios enfrentados por sistemas democráticos com a ascensão da extrema direita.

Palavras-chave:
partidos políticos; engajamento partidário; incentivos seletivos; incentivos coletivos; extrema direita

Resumen:

¿Cuáles son los vectores que impulsionan la participación partidaria en el contexto de una creciente desconfianza y hostilidad hacia los partidos políticos? En este artículo, sostenemos que un elemento dañino opera como un vector relevante para comprender la participación partidaria en los sistemas políticos contemporáneos, donde las organizaciones de partidos son vistas como actores ilegítimos a los ojos del público. Con base en datos observacionales y experimentales obtenidos a través de una encuesta sin precedentes que reunió respuestas de miembros de 32 partidos políticos brasileños, examinamos cómo los incentivos reconocidos en la literatura pueden condicionar la participación de miembros y líderes de partidos. Los principales resultados muestran que el odio contra un partido político puede ser un catalizador influyente para la participación en actividades del partido. Al reconocer esta característica contemporánea del activismo partidista, pretendemos contribuir al debate sobre los desafíos que enfrentan los sistemas democráticos con el ascenso de la extrema derecha.

Palabras clave:
partidos políticos; compromiso partidista; incentivos selectivos; incentivos colectivos; extrema derecha

Introduction

The rise of the far right in countries that seemed less susceptible to authoritarian leadership has become increasingly frequent in contemporary democracies. Among other factors, the widespread distrust of political parties and the growing discredit of representative institutions in society have been identified as causes of a phenomenon that has destabilized several political systems (Gidron; Adams; Horne, 2020; Ignazi, 2014; Mair, 1997, 2013; Webb; Scarrow; Poguntke, 2022).

Amid this scenario, the case of Brazil has highlighted in recent years a deepening in the cartelization of political parties (Katz; Mair, 1995) and the distrust in party organizations (Moisés, 2011; Ribeiro, 2013), a situation that resulted in the victory of a far right-wing candidate in the 2018 presidential contest, who was elected as the leader of a party that had had no national expression until then (PSL).

Contrary to the expectations created by public opinion, which is unfavorable to political parties, recruitment and membership have gradually progressed in the country. We argue that a dimension of a negative nature - engagement motivated by hatred - has been a relevant vector of our understanding of party activism in political systems marked by distrust and hostility between competitors (Gidron; Adams; Horne, 2020). Therefore, we examine the extent to which extreme rejection of a party acts as a catalyst for a more intense participation by members.

Our work identifies a negative conditioner for high-intensity engagement with a party, characterized as when a member dedicates more than 30 hours per month to party activities. From this perspective, the results show that hatred of a political party, a phenomenon that goes beyond the idea of simple competition between opponents, conditions the intensity of the involvement of the members, which has potential consequences for the functioning and operation of party organizations. In this sense, engagement motivated by hatred acts as a catalyst of party activism and increases participation of members in a context of deep hostility between antagonistic parties and their delegitimization.

To identify the individual conditions that influence engagement, we conducted the first-ever national survey with party members and leaders in Brazil, which is considered to be a critical case for understanding the rise of those who call into question institutions that are fundamental to the functioning of representative democracies, including here the parties themselves.

It is important to point out that the legitimacy crisis in parties and the growing hostility between opponents today affect different political systems, including the oldest ones (Ignazi, 2014; Gidron; Adams; Horne, 2020; Svolik, 2019; Webb; Scarrow; Poguntke, 2022). The investigation we conducted for this article reflects, therefore, on a broad phenomenon, which is the transformation of party organizations amid the rise of ideational populism (Mudde, 2017). Populism has advanced into countries with different historical and institutional contexts based on a common repertoire that belittles political institutions essential for supporting representative democracies.

Using an experimental approach, we compared the influence of hatred to other incentives that are recognized as relevant in party activism literature (Ribeiro; Amaral, 2019; Scarrow, 2015). Among the main results, we highlight the critical role played by a negative dimension of engagement, which is still underestimated in the literature. The findings show that hatred of a party potentially winning the next presidential elections has a more significant influence on the engagement of members than aspirations for internally more democratic processes, for example.

After this brief introduction, we present in the following sections the main theoretical debates and set out the methodological procedures used in conducting the research. We then examine descriptive data and the results of a binomial logistic regression model that explain the puzzling phenomenon that arises: the parties do not lose members despite a scenario marked by severe damage to the image of party organizations.

We then analyze the results of a within-subjects experimental approach, in which members indicate, at different and randomly presented stages during the questionnaire completion, their willingness to become highly engaged given the hypothetical scenarios depicting incentives that are recognized in the literature on party activism.

Using experimental data, we assessed whether scenarios involving the virtual success of their own party - or of the party they most reject - in the next presidential elections, and the possibility of exerting a greater influence on the organization’s internal processes, would boost their high-intensity engagement as measured in the control scenario.

Therefore, the research identifies a catalytic element of engagement between members that can operate as a potential vector for transforming party organizations in an environment marked by profound distrust and growing rejection of opponents. Since individuals have beliefs and attitudes that shape political parties and impress on them the essential traits that condition their actions, we are looking to contribute to the debate on the risk of the erosion of the democratic consensus that was historically forged and is now under threat in Brazil and in other countries that are experiencing the rise of the far right.

The link between the parties and society

Since the second half of the 20th century, the erosion of Western populations’ trust in democratic institutions has been a phenomenon frequently observed in comparative studies in Europe and the Americas (Dalton; Wattenberg, 2000; Katz; Mair, 1995; Mair, 2013; Moisés, 2011; Norris, 1999).

Various diagnoses have pointed to the growing perception of illegitimacy in the representative arrangements of contemporary democracies, including rigid party systems, dissatisfaction with the efficiency of institutions, systemic corruption, distrust of authorities, low turnout rates in elections, decreasing levels of party identification, etc. (Moisés, 2011; Norris, 1999).

Although parties are considered core elements in the intermediation of demands and tensions between society and the State in representative democracies (Kirchheimer, 2012), public opinion’s overall negative assessment of party organizations is evidence of a widespread process of distrust in contemporary democratic arrangements (Dalton; Wattenberg, 2000).

Even though political parties are understood to be indispensable for the functioning of representative democracies, literature draws our attention to the considerable challenges arising because of how they have distanced themselves from society (Ignazi, 2014, 2017; Katz; Mair, 1995).

As with other political systems, the case of Brazil shows the dissatisfaction that exists with institutions that are essential to the functioning of democratic government, a dissatisfaction that crosses different ideological and socio-demographic segments (Moisés, 2011; Moisés; Carneiro, 2008).⁶

The widespread perception of the crisis in legitimacy being experienced by party organizations in recent years suggests the extent to which the link between parties and civil society has been weakened:

Parties are perceived as an indispensable tool channeling the voice and the demands of the citizenry but, at the same time, quarrelsome and costly machinery composed of prime donne and parasites, social climbers, and rent-seekers. Although few argue against the party per se, many express low levels of confidence and esteem in them. This widespread disregard has fostered anti-party sentiment in most consolidated democracies (Ignazi, 2014, p. 160).

Recent data from the World Values Survey (WVS) (Inglehart et al., 2014) corroborate the trajectory of discredit of Brazilian party institutions in public opinion. According to Graph 1, in 2006 fewer than half of all Brazilians (45.8%) reported that they did not trust political parties at all, while by 2014, 58.6% were showing this same level of distrust, which reached its peak in the historical series in 2018 (66.5%).

Graph 1
Proportion of the interviewees who in no way trust Brazilian political parties (2006-2018)

Ribeiro (2013), on the other hand, draws attention to an increasing trend in the total number of party members in the country since 2006, as attested to by recent data from the Superior Electoral Court (TSE). Graph 2 below illustrates this scenario of notable association between voters and party organizations in Brazil, as evidenced by the significant number of members in the time frame we examined.

Graph 2
Total number of political party members in Brazil between 2006 and 2022 (in millions)

The evolving distrust of parties is in opposition to the significant number of links that have been established between citizens and party organizations in the country. Given this counterintuitive picture, this article suggests investigating the elements that condition party engagement in an environment marked by a significant deterioration in the image of political parties and growing hostility between opponents (Gidron; Adams; Horne, 2020; Webb; Scarrow; Poguntke, 2022).

Incentives and party engagement

Duverger (1985) suggests that the links established between parties and individuals result from the different characteristics assumed by party organizations. According to the concentric circles model, the author classifies the political involvement of individuals based on their position in the orbit of influence within the organization.

Engagement can thus be thought of as a function of the intensity of citizen involvement in the activities of a party organization: they become more militant as these activities increasingly occupy their daily lives (Duverger, 1985).

In this article we investigate the conditions necessary for the high-intensity engagement of members, which is associated not only with a formal link with a given party, but above all with active involvement in the life of the party organization. This article clarifies the motivations of members who commit a significant part of their time and other resources to this type of political activism (Duverger, 1985; Scarrow, 2015; Van Haute; Gauja, 2015).

Highly intense participation, classified as that which exceeds 30 hours per month, therefore, involves a cost-benefit calculation on the part of the elites and party activists (Scarrow, 2015). On the one hand, the calculus of the elites tends to consider the usefulness of encouraging this form of participation (demand-side explanations), while on the other, activists assess the meaning of joining a party according to their individual objectives and resources (supply-side explanations).

On the demand side, the literature understands that engagement is conditioned by asymmetries in accessing material, associative, and cognitive resources and attributes. According to this perspective, a strong relationship exists between participatory dispositions and educational variables, which are generally associated with specific strata of income and professional occupation (Scarrow, 2015; Verba; Schlozman, Brady, 1995). According to this theoretical approach, individual attributes, such as gender and education, are among the conditions necessary for political engagement.

If we admit that individuals on the supply-side are seeking to maximize their benefits when they act, then the paradox of collective action emerges, which is present given the possibility that collective decisions may benefit those who did not get involved or assume any participation costs (Scarrow, 2015; Seyd; Whiteley, 1992). Olson (2011) argues that positive (or selective) incentives act as mechanisms that encourage participation in democratic contexts.

In line with Panebianco (2005), Seyd and Whiteley (1992) present a theoretical approach with their General Incentives Model that identifies the selective, collective, and psychological incentives that condition party membership and high-intensity engagement with the party. In short, they claim that selective incentives can be considered in two categories: process incentives and outcome incentives. Process incentives are those related to individual benefits and the degree of satisfaction derived from participating in party activities.

In the case of outcome incentives, what stands out when it comes to understanding this phenomenon are the material and status elements. Material incentives are associated with individual financial compensation, including assistance networks and various forms of patronage. Status incentives, in turn, come from individual aspirations, such as personal prestige and public renown, which are dimensions that potentially catalyze party engagement (Scarrow, 2015; Seyd; Whiteley, 1992; Van Haute; Gauja, 2015).

When Ribeiro and Amaral (2019) researched party members in the State of São Paulo, they found that a significant amount of party activism is motivated by selective process and outcome incentives. Both combined correspond to around 44% of all highly engaged members.

In the case of collective incentives, aspects of identity are recognized as motivating the participation of individuals, and are usually linked to the level of adherence to the party’s political program or project. In their theoretical model, therefore, Seyd and Whiteley (1992) identify a stimulus to engagement that is aimed at gains for the group to which the member belongs, which is something that goes beyond a merely individual perspective for understanding the phenomenon. According to the authors, collective incentives are either positive or negative in nature. The former are linked to engagement motivated by adherence to agendas that are defended by the party, while negative collective incentives derive from a notion of dissimilarity or opposition to a party association (Scarrow, 2015; Van Haute; Gauja, 2015).

A feeling of rejection, therefore, is a strong stimulus for party engagement, because it constitutes collective identities and social bonds that are sustained by opposition to a set of ideas espoused by opposing groups or political parties. The perception of threat coming from a rejected rival group, therefore, reinforces the individual’s identification with the group, thus triggering a more intense engagement, which is associated with the need for differentiation and defense of their party or political group (Iyengar et al., 2019).

We argue that when partisan identities are perceived as significantly threatening, this conflict dynamic - which emerges especially in contexts marked by deep hostility between opponents - arouses extreme emotional reactions, such as hatred, and stimulates high-intensity party engagement.

Recent research undertaken in Brazil shows the growth of the phenomenon of an anti-partisanship and anti-PT (PT - Brazilian Workers’ Party) sentiment among the electorate (Fuks; Ribeiro; Borba, 2020; Paiva; Krause; Lameirão, 2017; Ribeiro; Carreirão; Borba, 2016; Samuels; Zucco, 2018). As Fuks, Ribeiro, and Borba (2020) observe, some studies consider this phenomenon to be associated with rejection of the Workers’ Party, as indicated by the evolution of anti-PT party sentiment (Samuels; Zucco, 2018). Recent findings, however, also point to a widespread picture of discredit and negative assessment that encompasses a broader spectrum of political parties (Paiva; Krause; Lameirão, 2017).

Also according to Fuks, Ribeiro, and Borba, (2020), in comparison with those who do not widely reject the parties in a system, anti-party voters tend to perceive social minorities, representative institutions, and the democratic regime itself in an unfavorable light.

In such a scenario, the examination of the catalytic factors driving intense party engagement, as outlined in this article, aims to enhance our comprehension of the transformational process through which party organizations evolve within political systems marked by escalating tensions and hostility among opponents.

Methodology

To investigate whether hatred of a party motivates people to dedicate themselves to party activities, we conducted an original survey of party members and leaders. The research included the participation of members linked to 32 Brazilian political parties,⁷ who completed a 52-question survey that was sent to them by email, using the Survey Monkey platform.

Between March and May 2020, 78,417 emails were sent out, and by the end we had received a total of 3,266 responses, of which 2,514 were from party members and leaders. This corresponds to a response rate of 4.16%, which is common for surveys carried out under similar conditions (Kertzer; Renshon, 2022). Considering the total number of members registered with the TSE when we started collecting data, we have a sampling error of 2%, with a 95% confidence level.

To address the questions that motivated this work, the survey results were analyzed in two stages. First, we used observational data to identify factors associated with greater participation in activities related to the functioning of party organizations.

To do so, we proposed a binomial logistic regression model, the results of which are discussed in the following section. The model’s dependent variable is high-intensity engagement with the party, when the member dedicates more than 30 hours per month to party activities (a category that has the value 1, with all the other replies indicated as a reference category).⁸

Among the explanations we included: gender; education; ideology; holding any position in the party; and the reason for becoming a member. The original response categories of the first two variables were recoded with the idea of testing, respectively: whether male party members (1) demonstrate a greater probability of high-intensity engagement with the party, which is a hypothesis that is in line with the underrepresentation of women in institutional politics; and whether members who had or had not completed higher education (1), and who are generally more exposed to political debates, actually participate more intensely in party activities.

With the ideology variable we investigated whether center-right members (who ranked between 6 and 10 on a scale of 1 to 10) tend to engage less than members who are not located in this range on the ideological spectrum. Conversely, we hypothesize that members who currently hold a position in the party (1) dedicate more time per month to party activities.

The main explanatory variable of the model assesses whether the interviewee considered the potential opposition to a specific political party as a very important reason for becoming affiliated. We aimed to determine if individuals who identified this motivation as very important were more deeply involved than others. This approach allowed us to probe whether observational data might indicate a form of negative engagement, indicated by a correlation between high-intensity engagement and party affiliation driven by the rejection of another party’s political project.

Using experimental data we then examined whether the hypothetical scenarios presented to respondents could act as the catalyst of high-intensity engagement with the party. At this stage, we employed a within-subjects experimental design, where various measurements taken from the same interviewee while they completed the questionnaire allowed us to assess their reactions to the treatment. In this approach, control and treatment conditions, typically associated with separate groups, corresponded to consecutive measurements on the same individual.

In the within-subjects experimental approach conducted in this research, the control scenario corresponds to the interviewee’s spontaneous response regarding the number of hours a month currently dedicated to the party. The three scenarios, randomly presented in the questionnaire, investigate whether incentives aligned with dimensions acknowledged in the literature, along with engagement stemming from hatred, can trigger high-intensity engagement with the party.

With two treatment scenarios we investigated whether the possibility of the party’s victory in the next presidential elections and the possibility of exerting greater influence on the party’s internal processes would boost engagement with the party.

With the other scenario, we tested the main hypothesis that motivated the research: that hatred toward a political party serves as a cognitive shortcut capable of amplifying high-intensity engagement among party members. This would manifest as a larger proportion of highly engaged members given the possibility of their most rejected party winning in the upcoming presidential elections.

The article examines, therefore, whether a negative dimension drives high-intensity engagement. It enhances our comprehension of activism, especially in a context characterized by the significant erosion of parties’ credibility with society and the ascent of far-right figures in numerous nations, even in places previously thought to be less prone to authoritarian solutions.

To interpret the results, we used paired t tests and compared whether there were statistically significant differences in means relative to the proportion of highly engaged members in the control scenario and the three treatment scenarios.

Negative incentives and high-intensity party engagement

In the following table, we present the results of the proposed binomial logistic regression model based on observational data collected using a national survey of party members and leaders. At this point, we seek to identify the variables that are potentially associated with a greater likelihood of high-intensity engagement with the party.

Table 1.
Binomial logistic regression model (dependent variable: high-intensity engagement with the party)

	Odds ratio	Std. Error	P-value
Gender	1.491*	0.17276	0.02076
Education	1.028	0.12943	0.82615
Ideology	0.746*	0.12196	0.01667
Position	1.614**	0.15928	0.00264
Opposition	1.277*	0.11220	0.02903

• Source: Prepared by the authors from survey data.

The coefficients indicated above were transformed to better interpret them in terms of the odds ratio. Therefore, the second column of the table shows the ratio between the probability of high-intensity engagement occurring and the complementary probability associated with the non-occurrence of the event.

The results corroborate the hypotheses previously presented, and provide evidence of an important association between affiliation motivated by opposition to a political party and high-intensity engagement, thus adding to our ability to recognize a negative collective incentive (Seyd; Whiteley, 1992).

Regarding the gender variable, we identified a 49.1% greater probability of high-intensity engagement with the party among men. This means that the probability of a more intense involvement in party activities among male members is 59.85%, while that of women is less significant (40.15%). The results match our expectations, considering the overrepresentation of men in institutional politics. Since parties are the collective actors that lead the political processes essential to representative democracy, and also considering the underrepresentation of women in politics in Brazil, the data indicate considerable obstacles to overcoming female exclusion from the political-institutional scenario.

The education variable did not show a significant association with the result. Contrary to expectations, therefore, it is not possible to identify any differences between members who have or have not completed higher education. The variable was used as a proxy for exposure to political debates, something we might think of as encouraging more intense political participation by members who have more years of formal education.

The findings, however, indicate that education is not related to greater engagement by party members, which points to the importance of investigating other instances of socialization and involvement in political activities beyond the university environment.

The ideology variable indicates that there is an unfavorable odds ratio for high-intensity engagement by center-right members who are positioned between 6 and 10 on a scale that goes from 1 (left) to 10 (right). We noticed an approximately 25% lower probability of this type of engagement among these members when compared to others.

In other words, we estimate the probability of center and right-wing members of becoming highly engaged in party activities at 42.75%, while the other interviewees are associated with a probability of 57.25%. Thus, the results suggest that there is a tendency for greater involvement in party activities among members who are located on the left of the ideological continuum.

The variable that indicates whether the member holds a position in the party bureaucracy worked as an important control and predictor. As we expected, the probability of a member being highly engaged because they play a role in the party organization exceeds the corresponding probability relative to the reference category by more than 60%.

In other words, the probability of intense engagement among those who participate more directly in the party organization is around 61.74%, while the probability of being only highly engaged occurs in much lower proportions among affiliated activists (38.25%).

Finally, the odds ratio associated with the main explanatory variable of the model indicates a 27.7% greater probability of highly engaged members among those who indicated that the possibility of opposing a party was a very important reason for them to become affiliated with a party.

The result suggests that intense participation in party activities is connected to a dimension of negative engagement (Seyd; Whiteley, 1992). The data indicate that aversion to a party can play a relevant and still underestimated role in party members’ engagement. This is because the probability of high-intensity engagement is 56.09% among those who indicated that their opposition to a party was a very important reason for their affiliation. On the other hand, the reference category is associated with a more modest probability of highly engaged affiliates, 43.9%.

In the following section, we use experimental data to investigate elements that potentially induce greater engagement with the party. By examining scenarios that interact with incentives recognized in the literature, we also test the possibility that hatred of a party acts as a mechanism capable of driving high-intensity engagement.

Hatred as a catalyst of high-intensity engagement

When compared to the use of observational data, an experimental survey approach offers analytical advantages that support a more reliable evaluation of the causal relationships that are of interest to this investigation. The essential element of an experimental research design comprises the possibility of exogenously manipulating the explanatory variable, which is characterized by the researcher directly intervening during the data collection process (Druckman et al., 2011).

This inherent characteristic of the experimental design makes for a more secure causality assessment when compared to what is possible using observational data. Through exogenously manipulating the variable of interest, researchers can compare the measured outcomes between treatment (subject to intervention) and control groups. This comparison enables more precise conclusions about the direction and intensity of the causal relationship between the two variables. (Druckman et al., 2011).

The within-subjects experimental design used in this article allows the control and treatment scenarios to refer to the same individual, who is subjected to different stimuli or treatments when completing the questionnaire. According to Gerber and Green (2012, p. 273), “within-subjects experimentation refers to studies where a single person or entity is tracked over time and random assignment determines when a treatment is administered”.

When the questionnaire was being applied, each interviewee indicated their willingness to engage with the party in a highly intense way in three hypothetical scenarios. The treatment scenarios occurred after the spontaneous measurement of the total monthly hours currently dedicated to the party (control) and were presented in random order throughout the questionnaire.

Using an experimental approach, we assessed the proportion of highly engaged members given the possibility of: their own party’s success in the next presidential elections; exerting greater influence over the internal processes of the party organization; and the victory of the party they most hate (or have a strong feeling of rejection for) in the next elections for president.

Regarding the advantages inherent in the use of this research design, we emphasize the following:

The allure of within-subjects designs is their capacity to generate precise treatment estimates with a single subject. Precision derives in part from the fact that individuals or entities are compared to themselves, which means that background attributes hold constant. Another feature of most within-subjects experiments is that over-time comparisons are made under controlled conditions, so that the principal source of over-time variation is the treatment (Gerber; Green, 2012, p. 273).

The graph below presents the results corresponding to the treatment scenarios and the control scenario of the within-subject experiment:

Graph 3.
Percentage (%) of highly engaged members given the scenarios investigated

We draw attention, initially, to the observation that approximately 1/5th of the survey respondents (19.09%) are highly engaged, as they currently dedicate more than 30 hours per month to the party. To a large extent, this evidence contradicts a widespread but little investigated notion about the absence of inner workings and militancy of Brazilian political parties.

Table 2.
Proportion (%) of highly engaged members

	Control	Own party victory	Victory of the party they most reject	Intraparty democracy
All parties	19.09	41.73*	36.62*	32.08*
Traditional parties	19.06	39.43*	34.91*	30.00*

• Source: Prepared by the authors from survey data

In the table above, using paired t-tests, we compare whether the differences in the measurements of the treatment and control scenarios are statistically significant. The results point to the effectiveness of the treatments. We observed that the three hypothetical scenarios serve as catalysts for driving high-intensity engagement among members. These effects are evident in both the overall sample, which includes respondents from all 32 parties, and a subgroup limited to members of traditional parties.

Regarding the main hypothesis that motivated this article, the findings align with our expectations: hatred toward opposing political parties significantly influences the high and intense engagement of party members. This outcome particularly resonates with recent discoveries by Gidron, Adams and Horne (2020), Svolik (2019), and Webb, Scarrow and Poguntke (2022), especially as we identify the less-than-virtuous aspects of partisan engagement in a context of escalating hostility.

In the hypothetical scenario in which the party or candidate most rejected by the interviewee could win the next presidential elections, we observed a significant increase in the proportion of highly engaged members: 91.82% greater than for the spontaneous scenario. Faced with the possibility of a victory by the party they most reject, highly engaged members represent 36.62% of those interviewed.

The result draws attention to the potential risks arising from extreme hostility between opponents in party political disputes, which is similar to the warning of Gidron, Adams and Horne (2020) regarding voters in general. It also coincides with Svolik (2019, p. 23) by suggesting that “in polarized electorates, voters are willing to trade off democratic principles for partisan interests.” What becomes apparent is a hostile dimension of activism that arises within a context where public opinion perceives political parties negatively. Other hypothetical scenarios, such as the chance of victory for their own party in the presidential elections and the greater possibility of influencing internal processes, both linked with incentives conventionally acknowledged in the literature, likewise resulted in a notable increase in the proportion of highly engaged members (who, as a result, now represent 41.73% and 32.08% of the interviewees, respectively).

In the general picture, when we consider all the parties, the possibility of victory for their own party motivates members to a greater extent than the rejection of another political group; these two scenarios are the ones that exerted the most influence on those interviewed throughout the questionnaire.

But there are differences when we disaggregate by party. We can identify, for example, two groups with different engagement characteristics. In the first group, we note that engagement resulting from the possibility of the victory of one’s own party in the presidential elections serves as a stronger stimulus than the possibility of victory by the party most rejected by the respondent. This group includes interviewees from the PT, PDT, PSB, PL and PC do B (mostly left-wing Brazilian political parties). In another group, it is impossible to distinguish which of the two scenarios operates in a more influential way. In other words, love and hatred exert the same weight on the engagement of members of a group of important traditional parties in Brazilian politics, such as PSDB, MDB, DEM, PP, PTB, and the PSL.⁹

Table 3.
Proportion (%) of highly engaged members by party.

	Control	Own party victory	Victory of the party that they most reject	Intraparty democracy
PT	22.34	47.34	40.43	31.91
PSDB	13.60	31.20	31.20	26.40
MDB	17.39	32.92	30.43	26.71
DEM	25.00	42.05	39.77	37.50
PP	19.81	27.36	29.25	23.58
PL	19.18	45.21	34.25	34.25
PSL	13.33	28.33	35.00	33.33
PDT	20.34	50.00	43.22	34.75
PSB	18.75	42.19	31.25	29.69
PTB	13.70	34.25	32.88	27.40
PCdoB	23.29	50.68	38.36	31.51

• Source: Prepared by the authors from survey data

With the purpose of examining whether rejection is particularly directed towards one party or group of parties, we asked those interviewees who said that they rejected at least one party to indicate the parties that triggered their aversion. We summarize these results in the following table:

Table 4.
Percentage (%) of rejection, by party (multiple alternatives)

Party	Rejection (%)
PT	34.06
PSOL	27.33
PSL	25.14
PCdoB	24.78
PSDB	23.37
DEM	20.81
MDB	20.66
PP	12.52
PDT	8.29
PSB	7.67
Outros	11.11

• Source: Prepared by the authors from survey data

In line with the literature on vote choice and anti-partisanship in Brazil (Amaral, 2020; Fuks; Ribeiro; Borba, 2020; Paiva; Krause; Lameirão, 2017; Ribeiro; Carreirão; Borba, 2016), the PT (Workers Party) is also at the forefront of rejection by members. The positions held by PSOL, PSL, and PC do B align with the level of attention and prominence these parties garnered during the peak of the radicalization phase, characterized by the ascent of the far right in Brazil. Other parties hold positions reasonably in line with their visibility and organizational structure in the ranking of those that were rejected. In conjunction with the experimental data, the results in Table 4 reveal that high-intensity engagement motivated by hatred of a party does not merely coincide with the phenomenon of anti-PT sentiment. Although the Workers’ Party does, in fact, top the list as the most rejected (by approximately 34% of all respondents), the data show that rejection also affects other parties. While the Workers’ Party does indeed rank highest as the most rejected (by roughly 34% of all respondents), the data illustrates that this rejection extends to other parties as well. Apart from the PT, six more parties are rejected by over one-fifth of all respondents (MDB, DEM, PSDB, PC do B, PSL and PSOL).

Considering the findings from both observational and experimental data, we identify that engagement driven by hatred is a more widespread phenomenon than solely an anti-PT sentiment.

Final considerations

Drawing from data collected in the first nationwide survey applied to members of 32 party organizations in Brazil, the article examines how incentives acknowledged in the literature shape the participation of political party members and leaders. The primary findings underscore that hatred of a party is an influential catalyst for high-intensity engagement in the current tense landscape in Brazilian politics. Moreover, we have identified a widespread rejection of parties across different spectrums and measured the weight of vectors outlined in the literature as influencing intra-party participation.

The findings shed light on how engagement driven by intense aversion to opponents enriches our comprehension of party activism within an environment where parties face discredit and representative institutions lack trust. The situation is also marked by growing hostility between competitors.

We emphasize, therefore, that Brazil is a typical case for examining the effects of ideational populism on the transformation of political organizations. From a comparative perspective, the results highlight the present or latent risks in different democratic systems sharing analogous contexts of extreme hostility among adversaries. The article contributes to a research agenda focused on understanding the challenges faced by contemporary democracies amidst the rise of far-right leaders.

Referências

• AMARAL, O. E. The victory of Jair Bolsonaro according to the Brazilian electoral study of 2018. Brazilian Political Science Review, v. 14, p. e0004, 2020.
• COPPOCK, A.; LEEPER, T. J.; MULLINIX, K. J. Generalizability of heterogeneous treatment effect estimates across samples. Proceedings of the National Academy of Sciences, v. 115, n. 49, p. 12441-12446, 2018.
• DALTON, R.; WATTENBERG, M. Parties without partisans: political change in advanced industrial democracies. New York: Oxford University Press, 2000.
• DRUCKMAN, J. N.; GREEN, D. P.; KUKLINSKI, J. H.; LUPIA, A. Experimentation in political science. In: DRUCKMAN, J. N.; GREEN, D. P.; KUKLINSKI, J. H. (orgs.). Cambridge handbook of experimental political science, New York: Cambridge University Press, 2011. p. 3-11. [capítulo 1].
• DUVERGER, M. Os partidos políticos. Rio de Janeiro: Zahar, 1985.
• FINDLEY, M. G.; KIKUTA, K.; DENLY, M. External validity. Annual Review of Political Science, v. 24, p. 365-393, 2021.
• FUKS, M.; RIBEIRO, E.; BORBA, J. From antipetismo to generalized antipartisanship: the impact of rejection of political parties on the 2018. Brazilian Political Science Review, v. 15, p. e0005, 2020.
• GERBER, A. S.; GREEN, D. P. Field experiments: design, analysis, and interpretation. New York: WW Norton, 2012.
• GIDRON, N.; ADAMS, J.; HORNE, W. American affective polarization in comparative perspective. Cambridge: Cambridge University Press, 2020.
• IYENGAR, S.; LENKES, Y.; LEVENDUSKY, M.; MALHOTRA, N.; WESTWOOD, S. J. The origins and consequences of affective polarization in the United States. Annual Review of Political Science, v. 22, p. 129-146, 2019.
• IGNAZI, P. Power and the (il)legitimacy of political parties: an unavoidable paradox of contemporary democracy? Party Politics, v. 20, n. 2, p. 160-169, 2014.
• IGNAZI, P. Party and democracy: the uneven road to party legitimacy. Oxford: Oxford University Press, 2017.
• INGLEHART, R., C.; HAERPFER, A.; MORENO, C.; WELZEL, K.; KIZILOVA, J.; DIEZ-MEDRANO, M.; LAGOS, P.; NORRIS, E.; PONARIN, B.; PURANEN, B. et al. (eds.). World Values Survey: Round Six - Country-Pooled Datafile Version. 2014. [Madrid: JD Systems Institute]. Disponível em: Disponível em: http://www.worldvaluessurvey.org/WVSDocumentationWV6.jsp Acesso em: 15 abr. 2024.
  » http://www.worldvaluessurvey.org/WVSDocumentationWV6.jsp
• KATZ, R. S.; MAIR, P. Changing models of party organization and party democracy: the emergence of the cartel party. Party Politics, v. 1, n. 1, p. 5-28, 1995.
• KERTZER, J. D.; RENSHON, J. Experiments and surveys on political elites. Annual Review of Political Science, v. 25, p. 529-550, 2022.
• KIRCHHEIMER, O. A transformação dos sistemas partidários da Europa Ocidental. Revista Brasileira de Ciência Política, Brasília, n. 7, p. 349-385, abr. 2012.
• MAIR, P. Rulling the void: the hollowing of western democracy. Londres: Verso, 2013.
• MAIR, P. Party system change: approaches and interpretations. Oxford: Oxford University Press, 1997.
• MOISÉS, J. A. Political discontent in new democracies: the case of Brazil and Latin America. International Review of Sociology: Revue Internationale de Sociologie, v. 21, n. 2, p. 339-366, 2011.
• MOISÉS, J. A.; CARNEIRO, G. P. Democracia, desconfiança política e insatisfação com o regime: o caso do Brasil. Opinião Pública, Campinas, v. 14, n. 1, p. 1-42, 2008.
• MUDDE, C. Populism: an ideational approach. In: KALTWASSER, C. R.; TAGGART, P.; OSTIGUY, P. The Oxford handbook of populism. Oxford: Oxford University Press, 2017. p. 27-47. [capítulo 2].
• NORRIS, P. Critical citizens: global support for democratic government. Oxford: Oxford University Press,1999.
• OLSON, M. A lógica da ação coletiva. São Paulo: Edusp, 2011.
• PAIVA, D.; KRAUSE, S.; LAMEIRÃO, A. P. O eleitor antipetista: partidarismo e avaliação retrospectiva. Opinião Pública, Campinas, v. 22, n. 3, p. 638-674, 2017.
• PANEBIANCO, A. Modelos de partido: organização e poder nos partidos políticos. São Paulo: Martins Fontes, 2005.
• RIBEIRO, E.; CARREIRÃO, Y.; BORBA, J. Sentimentos partidários e antipetismo: condicionantes e covariantes. Opinião Pública, Campinas, v. 22, n. 3, p. 603-637, 2016.
• RIBEIRO, P.F. El modelo de partido cartel y el sistema de partidos de Brasil. Revista de ciencia política, Santiago, v. 33, n. 3, p. 607-629, 2013.
• RIBEIRO, P.F.; AMARAL, O. E. Party members and high-intensity participation: evidence from Brazil. Revista de ciencia política, Santiago, v. 39, n. 3, p. 489-515, 2019.
• SAMUELS, D.; ZUCCO, C. Partisans, antipartisans, and nonpartisans voting behavior in Brazil. Cambridge: Cambridge University, 2018.
• SCARROW, S. Beyond party members: changing approaches to partisan mobilization. New York: Oxford University Press, 2015.
• SEYD, P.; WHITELEY, P. Labour’s grass roots: the politics of party membership. Oxford: Clarendon, 1992.
• SVOLIK, M. W. Polarization versus democracy. Journal of Democracy, v. 30, n. 3, p. 20-32, 2019.
• TSE. Tribunal Superior Eleitoral. Estatísticas eleitorais. 2024. Disponível em: Disponível em: https://sig.tse.jus.br/ords/dwapr/r/seai/sig-eleicao/home?session=310732171615820 Acesso em: 15 abr. 2024.
  » https://sig.tse.jus.br/ords/dwapr/r/seai/sig-eleicao/home?session=310732171615820
• TSE. Tribunal Superior Eleitoral. TSE aprova registro do partido União Brasil. 8 fev. 2022. Disponível em: Disponível em: https://www.tse.jus.br/comunicacao/noticias/2022/Fevereiro/tse-aprova-registro-do-partido-uniao-brasil Acesso em: 11 maio 2024.
  » https://www.tse.jus.br/comunicacao/noticias/2022/Fevereiro/tse-aprova-registro-do-partido-uniao-brasil
• VAN HAUTE, E.; GAUJA, A. (orgs.). Party members and activists. New York: Routledge, 2015.
• VERBA, S.; SCHLOZMAN, K.L.; BRADY, H.E. Voice and equality: civic voluntarism in American politics. Cambridge, Mass.: Harvard University Press, 1995.
• WEBB, P.; SCARROW, S.; POGUNTKE, T. Party organization and satisfaction with democracy: inside the blackbox of linkage. Journal of Elections, Public Opinion and Parties, v. 32, n. 1, p. 151-172, 2022.
• WVS. World Values Survey, 2023. Disponível em: Disponível em: https://www.worldvaluessurvey.org/WVSContents.jsp Acesso em: 28 maio 2023.
  » https://www.worldvaluessurvey.org/WVSContents.jsp

• ¹
  For data replication: https://doi.org/10.7910/DVN/UOB2XJ
• ⁶
  Considering factors like income, religion, ideology, age, gender, etc. (Moisés, 2011; Moisés; Carneiro, 2008).
• ⁷
  See the table in the appendices for the general characteristics of the sample, such as the proportion of respondents by party and the function or position held. Although it largely mirrors the proportion of members per party on the national scene, the investigation was carried out based on convenience sampling. We would highlight, however, that according to Coppock, Leeper and Mullinix (2018) convenience samples can produce results that are substantially better than those obtained from surveys with representative samples. Within the terms suggested by Findley, Kikuta, and Denly (2021), therefore, we emphasize the extent and relevance of the findings of the article for an understanding of the picture of political parties in Brazil.
• ⁸
  The data were originally collected in seven categories. We particularly chose this approach to collection and not on a continuous scale because it was highly unlikely that the interviewees would remember or register with any degree of exactness the number of hours a month they had dedicated to party activities. In presenting the reply alternatives in categories (in bands or intervals that can be easily compared) we sought to ensure that the data we collected would be precise and reliable, which would allow for an interesting ordering and scaling of the level of engagement of the respondents.
• ⁹
  Two of these political parties (PSL and DEM) were merged after the questionnaire had been applied and gave rise to the União Brasil party. See: TSE (2022).

APPENDICES

General characteristics of the sample

Table A
Proportion (%) of respondents by party organization

PT	9.19%
MDB	8.39%
PSB	7.00%
PSDB	6.52%
PP	6.28%
PDT	5.69%
PSD	4.81%
DEM	4.53%
PTB	3.90%
PL	3.78%
PV	3.74%
PCdoB	3.58%
CIDADANIA	3.54%
PODE	3.26%
PSL	3.26%
REPUBLICANOS	3.22%
SOLIDARIEDADE	2.63%
PSC	2.43%
PSOL	2.31%
PROS	2.11%
PATRIOTA	1.99%
AVANTE	1.67%
PRTB	1.43%
REDE	1.11%
PMB	0.84%
PTC	0.72%
DC	0.68%
PMN	0.60%
NOVO	0.32%
UP	0.24%
PSTU	0.16%
PCB	0.08%

• Source: Prepared by the authors from survey data

Table B.
Proportion (%) of respondents by position or function in the party

President of the municipal executive	46.01%
Activist	16.16%
Member of the municipal board	7.57%
Secretary of the municipal executive	7.48%
Other	6.32%
Vice-president of the municipal executive	5.91%
Treasurer of the municipal executive	3.80%
Speaker of the municipal executive	1.20%
Member of the state board	1.16%
Member of the foundation linked to the party	0.91%
Secretary of the state executive	0.58%
President of the state executive	0.54%
Member of the municipal theme secretariat	0.45%
Member of the party’s municipal council	0.45%
Member of the national board	0.33%
Treasurer of the state executive	0.25%
Member of the state theme secretariat	0.25%
Vice-president of the state executive	0.17%
Speaker of the state executive	0.12%
Member of the party’s state council	0.12%
Speaker of the national executive	0.08%
Secretary of the national executive	0.04%
Member of the national theme secretariat	0.04%
Member of the party’s national council	0.04%

• Source: Prepared by the authors from survey data

Table C.
Proportion (%) of respondents according to their ideological position

Ideology	Proportion
1 (left)	15.15
2	5.81
3	7.47
4	7.05
5	17.91
6	8.02
7	7.05
8	7.61
9	5.19
10 (right)	18.74

• Source: Prepared by the authors from survey data

Single subject experiment questions

(Control scenario) Thinking about your current party, how many hours a month do you dedicate to party activities? (Response: how many hours dedicated a month)

(Treatment scenario) Considering the following hypothetical scenarios, how many hours a month would you dedicate to the activities of your party?

If the party or candidate you most hate (or reject strongly) could win the next presidential election (Response: number of hours dedicated per month)

If you had more opportunities to influence party proposals, and select internal leaders and candidates (Response: number of hours a month dedicated to it)

If your party had a real chance to win the next election for president (Response: number of hours dedicated per month)

Publication Dates

• Publication in this collection
  06 Sept 2024
• Date of issue
  2024

History

• Received
  23 Nov 2022
• Accepted
  04 Mar 2024