Over the past two decades, levels of political polarisation have shown an upward trend in almost every region of the world. Particularly in Western democracies, there is growing concern across various sectors of society regarding the consequences of this phenomenon. In this essay, I argue that political polarisation is positively associated with the occurrence of political violence. In societies with higher levels of polarisation, and where significant signs of affective polarisation are evident, there is a greater tendency for political violence to occur. To illustrate my argument, I will conduct an analysis focused on Latin America, through case studies on Brazil, Mexico, and Colombia. For these case studies, I will use indices of democracy and political polarisation, as well as official data on political violence (subject to availability in each country), and I will estimate affective polarisation based on public opinion surveys.
Literature Review
The concept of affective polarisation was first introduced in the 2012 article Affect, Not Ideology: A Social Identity Perspective on Polarization (Iyengar, Sood, and Lelkes), as the tendency not only to support one group but also to openly display aversion towards another. According to the authors, this represents the most accurate way to assess mass polarisation within a society, as opposed to traditional measurement methods such as divergent stances on public policy, ideological distance from opponents, and opinion polls. In the article, the authors argue that affective polarisation is a more diagnostic indicator of mass polarisation in a society, to the extent that “partisan identification represents a meaningful group affiliation”, based on the assumption that social identification requires a positive view of one’s own group and a negative perspective of the opposing groups. Since partisan preferences relate inconsistently to public policy preferences, a more representative measure of mass polarisation would be the dislike partisans have to individuals belonging to the out-group. They propose that identification with a political party is sufficient to generate rejection of and aversion to the opposing party,
Following this study, other authors began exploring the causes, consequences, and various manifestations of affective polarisation across different regions and contexts. In a 2023 study, Berntzen, Kelsall, and Harteveld examined the consequences of affective polarisation in the United Kingdom and Norway, seeking evidence of correlations with intolerance, social withdrawal, and support for violence within these societies. There is a persistent concern in the literature that political polarisation can lead to the erosion of social norms and democracy, political gridlock, and ultimately political violence; however, much of the research in this field still focuses on the United States. Berntzen, Kelsall, and Harteveld extend the scope of research on affective polarisation to another region and find little evidence that mere dislike of an opposing political group is sufficient to generate intolerance and social withdrawal, and completely rule out political violence. The authors suggest that other elements must be considered when weighing the likelihood of more severe consequences of polarisation within a society. Like other studies, they highlight, in particular, an extreme dualism (a Manichean mindset); the perception of societal injustice, specifically, belonging to a group perceived as being disadvantaged by another; and individual personality traits that may lead to more violent stances.
For this essay, I adopt a perspective aligned with political psychology studies addressing hostility between societal groups. In-group and out-group dynamics have been central to many debates in political science and international relations for decades, explaining a range of political behaviours, from policy alignment to ethnic conflicts and civil wars. Political polarisation, as described by Iyengar, Sood, and Lelkes, is fundamentally rooted in in-group and out-group dynamics, that is, the perception of belonging to one group in opposition to another. In the 2003 article Prejudice and Intergroup Hostility, John Duckitt describes how prejudice and intergroup hostility arise and manifest in society. Although he does not directly address issues of political polarisation, focusing instead on dynamics that are primarily ethnic, racial, and social in nature, Duckitt explains how intergroup threat, or the perception of a threat, helps determine the levels of hostility between groups. This threat can manifest in various ways, but the article highlights, in particular, realistic threats to a group’s power, well-being, and resources; symbolic threats related to values and beliefs; and threats to a group’s identity. In a political context, this perception of the opposition as a threat is further exacerbated in scenarios of polarisation, which could explain a level of intergroup hostility that goes beyond mere political disagreement, a phenomenon described by the concept of affective polarisation.
The relationship between political violence and polarisation, specifically affective polarisation, has been increasingly explored by academia in recent years, yet most studies are limited to the United States, and the overall body of research remains quite restricted. In a 2023 article, J. Piazza focuses on the relationship between Democrats and Republicans in the United States but extends his research to Brazil, Turkey, and Bangladesh. Drawing on both qualitative and quantitative approaches, the author argues that one of the ways political polarisation can lead to instances of political violence is through the dehumanisation and demonisation of opponents. When a group comes to view its adversary as both inferior and a threat, violence directed at that adversary becomes more socially acceptable. The study finds sufficient empirical evidence to support the thesis that the likelihood of political violence is higher in polarised democracies, yet it does not establish a causal relationship between polarisation and political violence.
The literature on this topic presents certain limitations, primarily because it focuses largely on the American political landscape. The United States possesses a very particular political model and society, which can make it difficult to extend these studies to other regions. Therefore, drawing on existing literature regarding affective polarisation and political psychology, I aim to expand research on the relationship between polarisation and political violence to Latin America. In line with the articles mentioned here and others used throughout my research, I will limit my scope to democracies, albeit ones presenting diverse political and socioeconomic contexts. My objective is to analyse whether the relationship observed between these two political phenomena in Latin America corresponds to that observed in the United States by Piazza, or in Norway and the United Kingdom by Berntzen, Kelsall, and Harteveld, for example. I will adopt the definition of affective polarisation proposed by Iyengar, Sood, and Lelkes, that “the mere act of identifying with a political party is sufficient to trigger negative evaluations of the opposition”, and I will describe in more detail how I intend to measure this phenomenon in the case studies in the next section.
Research Design
To illustrate my argument, I will develop three case studies focusing on Latin American countries. The rationale for this choice lies in the high levels of political polarisation in South and North America, the highest in the world, respectively, and the variation in the quality of their democracies. The preliminary selection of cases within Latin America was based on data regarding the incidence of political violence, the quality of democracy, and polarisation, as well as public opinion surveys, where available, that I will use as a proxy for affective polarisation. My intention was to select countries where it would be possible to conduct a deep and extensive analysis of the political and social landscapes relevant to my research question, after considering data availability and prior research in the field.
The first case selected is Brazil, with a democracy index of 6.49 and a polarisation index of +3.22. The significant polarisation in the country, which intensified between 2013 and 2022, combined with the fact that Brazil is one of Latin America’s most consolidated democracies, makes it an interesting case for this study, particularly for analysing the potential democratic consequences of affective polarisation. The Bolsonaro administration represents a pivotal moment in this regard, given that the former president himself was a target of political violence during his 2018 campaign and was recently convicted in an attempted coup d’état and plots to assassinate political rivals. Although Brazil has always experienced a certain degree of political polarisation, the emergence of the far right led by Bolsonaro caused affective polarisation to become much more pronounced in society; therefore, I will focus primarily on this period for the case study.
The second case focuses on Mexico, which has a democracy index of 5.32 and a polarisation index of +2.13. Although Mexico has historically exhibited high levels of violence, largely attributed to the presence of criminal organisations and cartels, there is significant evidence of political violence driven by competition between opposing political groups. The high levels of violence during the 2024 elections, reported by various international organisations, offer an excellent opportunity to understand the country’s complex internal dynamics.
The third case is Colombia, which has a democracy index of 6.35 and a polarisation score of +1.89. Political violence has been a problem in Colombia for a few years now; however, in 2024 and 2025, the country experienced a sharp rise in polarisation, and incidents of politically motivated attacks or attacks against candidates increased significantly. The 2025 assassination of presidential pre-candidate Miguel Uribe during a rally stands out as the most prominent example of this escalating political violence. Given the recent surge in polarisation and political violence, Colombia offers an interesting case for analysing the potential triggers behind this escalation.
As mentioned above, the issue of organised crime is highly prominent throughout Latin America, and the countries selected for the case studies are significantly affected. A potential limitation of my analysis is accurately distinguishing whether cases of violence reported as political are, in fact, acts of political violence rather than the result of organised crime in these regions or the country’s overall level of violence. I attempt to minimise this ambiguity by selecting data solely on violence directed at governors, mayors, village chiefs, local councils, local government bodies, election workers, and other administrators. Another strategy adopted to address this issue is the use of opinion polls capable of indicating levels of affective polarisation and signalling social dynamics that are difficult to identify through data alone.
However, to incorporate the significant element of organised crime into my analysis, I intend, where applicable, to identify the stances adopted by different parties on this issue. In Colombia, for instance, the agenda of combating organised crime was highly prominent in recent elections, whereas in Brazil, it is generally a key issue for the right and the far-right. The positions political actors or parties take regarding organised crime can potentially exacerbate political polarisation or even serve as a driver for acts of violence against political actors. Although this is not the study’s primary focus, I consider it an important factor that could significantly influence my analysis and conclusions.
The data presented above were collected from the 2025 V-Dem dataset on political polarisation, the 2025 Economist Intelligence Unit dataset on democracy, and reports from ACLED (2024) on violence and the Edelman Trust Barometer (2025). For the case studies, in addition to these datasets, I will use reports from international organisations, news regarding recent events, opinion polls, and, where applicable, statements and pronouncements by political actors.
Case Studies and Results
Brasil
Brazil is one of Latin America’s largest and most consolidated democracies, having enjoyed a long period of relative stability since the country’s return to democracy in 1985. In 2013, however, the Dilma administration faced a turbulent period of public dissatisfaction and protests that spread across the country. Although re-elected in 2014, the President was impeached in 2016. Following more than a decade of Workers’ Party (PT) rule, and in response to the turmoil that marked the Dilma administration, the political right gained significant ground, featuring new actors and a new platform. Jair Bolsonaro, then a congressman, emerged as the symbol of the right-wing and the country’s rising far-right, announcing his candidacy for the 2018 presidential election. Domestic tensions escalated sharply, driven by a right-wing movement of unprecedented strength in recent years, while the leftist leader and former President Lula remained imprisoned and ineligible to run due to his involvement in one of the biggest corruption scandals in Brazilian history.
About a month before the 2018 elections, then-presidential candidate Jair Bolsonaro was attacked during a rally. Shortly thereafter, the perpetrator was arrested and charged with a “personal attack motivated by political dissent.” For years, Bolsonaro’s supporters questioned whether the attack was an isolated incident or part of a larger conspiracy against the candidate, but the Federal Police concluded that it was, in fact, an isolated act. Bolsonaro was elected President of the Republic in that year’s election.
The Bolsonaro administration involved several controversies, instability, and internal political disputes. During his term, former President Lula was released from prison, and the case in which he had been convicted was annulled, restoring his eligibility to run for office. In 2022, Bolsonaro lost the presidential election to Lula. The 2022 election and transition period were marked by significant tension and mistrust across the country, with Bolsonaro and his supporters constantly questioning the legitimacy of the election and protests gaining momentum. Many of his supporters called for a military intervention, and encampments outside army installations spread across the country. On January 8, 2023, Bolsonaro supporters stormed and vandalised public buildings in the capital, Brasília, with many being arrested. Despite the popular unrest, there was no military intervention, and Lula took office.
It was revealed over the following three years that high-ranking officials of the Bolsonaro administration had indeed contemplated a coup d’état following the election defeat, in addition to plotting the assassination of Lula, his vice-president Geraldo Alckmin, and other figures deemed political enemies. In September 2025, Bolsonaro, members of his administration, and involved allies were convicted by the Supreme Federal Court of attempted coup d’état and attempted violent abolition of the democratic rule of law, among other crimes.
After these events, the country is deeply divided between a far-right faction that still supports former President Bolsonaro and an established left-wing bloc supporting President Lula and the PT. In 2025, 17% of the Brazilian population approved of threats and acts of violence as a means to bring about political change. Other opinion polls from 2024 and 2025 also indicate a significant rise in the perception of incivility among people holding opposing ideologies, pointing to growing affective polarisation in the country. These data indicate that, although the country is no longer at the peak of polarisation seen in 2022, a strong social division persists, causing people to even alter the way they relate to one another. In 2024, there were 102 acts of violence recorded as political violence; however, as noted in the previous section, this figure cannot be considered absolute. When analysed alongside previous data, it is possible to infer that some of these acts are indeed politically motivated, driven by the country’s overall political situation, yet there is a risk of contamination by other phenomena.
Regarding the issue of organised crime, which is highly salient in certain regions of Brazil, the public security agenda is largely dominated by the country’s right and far-right. Bolsonaro was a staunch advocate of gun ownership for the general population, and most governors, deputies, and senators elected through his party or among his close allies place the fight against organised crime at the core of their mandates. Because the discussion is heavily concentrated on the Brazilian right, it has become a key point of debate, particularly given the left’s lack of a clear stance on the issue. As organised crime-related violence has escalated in some regions, support for the Lula administration has declined, while approval ratings for right-wing governors have risen.
Mexico
Political violence in Mexico has been significant for years, with election periods proving extremely violent and attacks against political opponents occurring frequently. In 2018, the country recorded 145 murders of individuals directly linked to the electoral process, and 88 during the 2021 elections. In 2024, there were 342 acts of violence against governors, mayors, village leaders, local council members, local government officials, election workers, and other administrators, according to ACLED data. Within this figure, ACLED reports 102 political assassinations, kidnappings, enforced disappearances, assassination attempts, and other attacks targeting family members, staff, and others.
In the case of Mexico, however, the issue of organised crime is so deeply rooted in all levels of society that it becomes virtually impossible to distinguish between acts of political violence driven by polarisation and those driven by criminality. This is particularly true given the extensive direct involvement of organised crime in Mexican politics; consequently, violence perpetrated by organised crime often effectively becomes political violence. Political and criminal groups are simultaneously associated with and competing against each other, not merely for political office, but for control over entire regions. The interests of cartels and criminal organisations come to dictate the direction of politics and society, especially in areas where public offices are held by members or allies of these groups.
Due to all these factors, the case of Mexico is extremely difficult to analyse within the scope of this essay. The country experiences high levels of political polarisation and extremely high rates of political violence; however, the involvement of organised crime in politics makes it exceedingly complex to disentangle the country’s political and social dynamics. In many instances, political polarisation stems from conflicts between rival criminal groups, and the same applies to political violence. Disputes between cartels and criminal organisations blur with and overlap the conflicts between political parties, candidates, and ideologies.
However, despite the case’s inherent obstacles, it was still possible to identify interesting points regarding Mexico relating more directly to this essay. According to 2025 Edelman Trust Barometer research, 20% of the population approves of threats and acts of violence as a means to bring about political change. A significant segment of the population is dissatisfied with the political and social situation in Mexico, particularly regarding the feeling of insecurity caused by cartels. Protests against the government and calls for change are frequent, yet this dissatisfaction appears to be directed more at the state of the country as a whole than at the rivalry between political groups, as is the case in Brazil.
Colombia
The case study for Colombia proved very similar to that of Mexico, with a high level of involvement of organised crime and cartels in the country’s politics, making it extremely complex to distinguish the dynamics at play. In 2024, there were 74 acts of violence classified as political, and the 2025 assassination of presidential pre-candidate Miguel Uribe during a rally demonstrates how violence in the country has escalated significantly in recent years. Uribe was an opposition pre-candidate, part of the Democratic Centre party, and a fierce critic of former left-wing President Gustavo Petro.
However, the primary source of polarisation in the country remains the issue of public security and the fight against organised crime, particularly due to persistent allegations of cartel involvement in national politics. Yet, much like in Mexico, public dissatisfaction appears directed at the state of the country as a whole, and only recently has the situation begun to resemble the political polarisation seen in Brazil. In the past few months, the matter of organised crime in the country gained once again more relevance in the face of the threat of American intervention to fight drug cartels.
Conclusion
Based on existing literature on the subject and the case studies presented here, it is possible to observe the dynamics of the relationship between political polarisation and political violence, particularly when affective polarisation is present within society. This relationship is especially evident in the case of Brazil, given the pronounced nature of political polarisation and the fact that recent events concretely demonstrate how it can escalate into political violence. The emergence and growth of affective polarisation within society are also observable, particularly in the country’s opinion polls and in the nature of its political division.
In the case of Mexico, and to some extent Colombia, the issue of organised crime made analysis nearly impossible, not only because it coexists with political issues (as in Brazil) but because it is directly intertwined with politics in those countries. This was a challenge encountered during the research, one that proved greater than I had initially anticipated. I believe, however, that this phenomenon recurs across much of Latin America and that, in this regard, Brazil is the exception. The question of the quality of democracy becomes particularly relevant here, given that Brazil possesses a much more stable democracy, with far greater safeguards against external interference in its electoral processes, than other countries in the region. This factor may even help explain why Brazil does not exhibit the same dynamics regarding organised crime as Mexico, Colombia, and other Latin American nations. I believe this represents a limitation of my essay, as well as an opportunity for future research focused on the subject.
Finally, the case of Brazil, in particular, illustrates how high political polarisation can undermine the quality of democracies, as can more severe instances of political violence, such as those witnessed in Brazil and Colombia. Yet, once again, the initial quality of the democracy proves significant, given that Brazil managed to maintain considerable political and institutional stability despite the assaults on its democracy.
References
ACLED. (2025). Political Violence During Brazil’s 2022 Presidential Runoff. Available at: https://acleddata.com/report/political-violence-during-brazils-2022-presidential-runoff
ACLED. (2025). Violence Targeting Local Officials | 2023 Annual Report. Available at: https://acleddata.com/annual-report/violence-targeting-local-officials-2023-annual-report
ACLED. (2024). Violence Targeting Local Officials | 2024 Annual Report. Available at: https://acleddata.com/annual-report/violence-targeting-local-officials-2024-annual-report
Asmann, P. (2024). Mexico’s Extreme Election Violence Explained. InSight Crime. Available at: https://insightcrime.org/news/mexico-extreme-election-violence-explained/
Campos, N. and Federico, C. (2025). A New Measure of Affective Polarization. American Political Science Review, pp.1–19. doi:https://doi.org/10.1017/s0003055425000255
Comerford, R. (2025). Thousands protest against government in Mexico over violent crime. BBC. 15 Nov. Available at: https://www.bbc.com/news/articles/cn8vm30rr78o
Druckman, J.N., Klar, S., Krupnikov, Y., Levendusky, M. and Ryan, J.B. (2020). Affective polarization, local contexts and public opinion in America. Nature Human Behaviour, 5(5), pp.1–11. doi:https://doi.org/10.1038/s41562-020-01012-5
Duckitt, J. (2003). Prejudice and Intergroup Hostility. doi:https://doi.org/10.13140/2.1.4326.0165
Edelman Brasil. (2025). 2025 Edelman Trust Barometer. Available at: https://www.edelman.com/br/trust/2025/trust-barometer
Edelman Latam. (2025). 2025 Edelman Trust Barometer. Available at: https://www.edelman.com/lat/trust/2025/trust-barometer
Gazetadopovo.com.br. (2025). O que se sabe até hoje sobre a facada em Jair Bolsonaro. Available at: https://www.gazetadopovo.com.br/vida-e-cidadania/adelio-inimputavel-e-inqueritos-fechados-o-que-se-sabe-ate-hoje-sobre-a-facada-em-jair-bolsonaro/
Igreja, F. (2025). Desaprovação do governo Lula aumenta e aprovação cai após megaoperação no Rio, diz Quaest. CBN. Available at: https://cbn.globo.com/politica/noticia/2025/11/12/desaprovacao-do-governo-lula-aumenta-e-aprovacao-cai-apos-megaoperacao-no-rio-diz-quaest.ghtml
Iyengar, S., Sood, G. and Lelkes, Y. (2012). Affect, Not Ideology: A Social Identity Perspective on Polarization. Public Opinion Quarterly, 76(4), pp.819–819. doi:https://doi.org/10.1093/poq/nfs059
Kleinfeld, R. (2023). Polarization, democracy, and political violence in the United States: What the research says. Carnegie Endowment for International Peace. Available at: https://carnegieendowment.org/research/2023/09/polarization-democracy-and-political-violence-in-the-united-states-what-the-research-says?lang=en
Laboratorio Electoral. (n.d.). Resultados finales del monitoreo de la violencia electoral. Available at: https://laboratorioelectoral.mx/violencia
Lars Erik Berntzen, Kelsall, H. and Eelco Harteveld (2023). Consequences of affective polarization: Avoidance, intolerance and support for violence in the United Kingdom and Norway. European Journal of Political Research, 63(3). doi:https://doi.org/10.1111/1475-6765.12623
McMurtrie, B., Philipp, M., Hebden, R. and Williams, M. (2024). Development and Validation of the Affective Polarization Scale. International Review of Social Psychology, 37(1). doi:https://doi.org/10.5334/irsp.926
Molfese, L. (2025). ‘Punhal Verde e Amarelo’: entenda plano para matar Lula, Alckmin e Moraes. CNN Brasil. Available at: https://www.cnnbrasil.com.br/politica/punhal-verde-e-amarelo-entenda-plano-para-matar-lula-alckmin-e-moraes/
Our World in Data. (2025). Political polarization score. Available at: https://ourworldindata.org/grapher/political-polarization-score
Our World In Data (2025). Democracy Index. Our World in Data. Available at: https://ourworldindata.org/grapher/democracy-index-eiu
Parker, N. and Eisler, P. (2023). Political violence in polarized U.S. at its worst since 1970s. Reuters. 9 Aug. Available at: https://www.reuters.com/investigates/special-report/usa-politics-violence/
Piazza, J.A. (2023). Political Polarization and Political Violence. Security Studies, 32(3), pp.476–504. doi:https://doi.org/10.1080/09636412.2023.2225780
Svolik, M. (2019). Polarization versus Democracy | Journal of Democracy. Journal of Democracy. Available at: https://www.journalofdemocracy.org/articles/polarization-versus-democracy/
