Introduction: The New Polarization in Latin America

Abstract

Mounting evidence suggests that Latin American democracies are characterized by politics and societies becoming more divisive, confrontational, and polarized. This process, which we define here as the “new polarization” in Latin America, seems to weaken the ability of democratic institutions to manage and resolve social and political conflicts. Although recent scholarship suggests that polarization is integral to contemporary patterns of democratic “backsliding” seen in much of the world, this new polarization in the region has not yet received systematic scholarly attention. Aiming to address this gap in the literature, the different contributions in this special issue revise the conceptualization, measurement, and theory of a multidimensional phenomenon such as polarization, including both its ideological and affective dimensions, as well as perspectives at the elite and mass levels of analysis. Findings shed light on the phenomenon of polarization as both a dependent and an independent variable, contributing to comparative literature on polarization and its relationship to democratic governance.

As Latin America’s “third wave of democracy” (Huntington 1991) navigates its fifth decade, there is mounting evidence that democratic regimes across much of the region are in trouble, plagued by a combination of poor performance, fragile representative institutions, and diminished adherence to liberal democratic norms and procedures. Although notable democratic gains have recently been made in Guatemala, on a regional scale, Mainwaring and Pérez-Liñán (2023, 156) starkly assert that “Latin America is now contributing to the global decline of democracy.” Not surprisingly, the region’s democratic deficits have become focal points of social mobilization and political conflict, with new political parties or movements emerging on the both the left and right flanks of mainstream party systems, and traditional parties experiencing decline. These dynamics are fueling concerns that both Latin American politics and societies have become increasingly divisive, confrontational, and polarized (see Carothers and Feldmann 2021). The region is seemingly on the cusp of—if it hasn’t already entered—a dangerous, recursive feedback loop whereby democratic deficits spawn polarized conflict, which feeds back to paralyze and weaken democratic institutions themselves (McCoy and Somer 2019a). Deepening polarization may thus be both a driver and consequence of democratic dysfunction. Such recursive processes can undermine the capacity of democratic institutions to channel, process, and regulate social and political conflicts.

Political polarization is hardly a novelty in the region, however, and it may in fact be the historical norm (Murillo 2022). Polarization was integral to Latin America’s wave of revolutionary challenges, democratic breakdowns, and authoritarian takeovers in the 1960s and 1970s (O’Donnell 1973; Valenzuela 1978; Weyland 2019). However, contemporary forms of polarization seem to be different from previous polarization processes in three respects. First, polarization in the 1960s and 1970s often occurred between minority, non-democratic actors, that is, groups that supported armed revolutionary movements versus those that supported the armed forces and military coups. By contrast, the new polarization typically occurs among actors within the democratic arena—that is, between bitter partisan rivals or, in some cases, between mainstream parties and broad sectors of the citizenry mobilized in opposition to the political establishment. Political contestation between these actors unfolds within democratic institutions, as neither the parties nor the partisans necessarily think of a revolution or a coup d’état. Nevertheless, this polarization may sometimes erode democratic norms and adherence to democratic procedures.

Second, contemporary forms of polarization were preceded by—and lie in sharp contrast to—the relatively tranquil years of the 1990s, when democratic regimes were becoming institutionalized across most of the region and largely converged—at least at the level of political elites and mainstream parties—on the so-called “Washington Consensus” for economic liberalization (Williamson 1990; Edwards 1995). Quite differently, the violent polarization of the 1960s and 1970s was preceded by—and was, arguably, the culmination of—the turbulent record of post-1945 democratic regimes in Latin America and their highly contested efforts to incorporate labor-based populist and leftist parties into democratic politics in countries like Argentina, Brazil, Chile, and Uruguay (see O’Donnell 1973; Collier and Collier 1991). Third, and posed in terms that we explain below, the previous polarization was highly ideological, centered on the left-right divide and epic Cold War struggles between capitalist and socialist models of development and the international blocs that supported them. Under the new polarization, ideological conflict between rival camps has been tempered (though hardly eliminated) by the demise of socialist alternatives, but conflicts with important affective and cultural components centered on different “us versus them” identities have become highly divisive.

This new polarization has not yet received systematic scholarly attention, even though conventional wisdom suggests that political moderation and ideological depolarization were vital to the consolidation of Latin America’s new democratic regimes in the late twentieth century (Mainwaring and Pérez-Liñán 2014), and recent scholarship suggests that polarization is integral to contemporary patterns of democratic “backsliding” seen in much of the world (Levitsky and Ziblatt 2018; McCoy and Somer 2019a; Haggard and Kaufman 2021). Why, then, after the end of the Cold War and an extended period of political moderation (or even convergence), have Latin American polities become so deeply divided or polarized once again? Are mass publics themselves truly polarized, or is polarization largely a function of elite competition and conflict? (see Murillo 2022). Are polarization processes driven from the top-down by elite strategies to demonize their rivals and mobilize supporters, or by bottom-up patterns of social mobilization, protest, and tribalism—or both? And, given the historical track record, what are the implications of this new polarization for democratic representation and the vitality and stability of democratic regimes? This special issue brings together scholars from both Latin America and the United States to address these questions from a range of different perspectives. As the contributions make clear, polarization itself is a complex, multidimensional political phenomenon, one that is subject to different conceptual frameworks, levels of analysis, and modes of measurement and empirical inquiry. The articles employ diverse methodological toolkits to examine different types of data at both elite and mass levels of analysis, as well as the interaction between polarization dynamics at these different levels. Therefore, in line with the multidimensional nature of polarization, including its patterns, levels, and dynamics, this special issue proposes a multimethod approach and extensive coverage of topics and cases. It incorporates a comparative approach and case analyses, as well as qualitative and quantitative data collection and analytic techniques. The latter include public opinion surveys and social media data. As the editors of this special issue, we strongly believe that a multimethod approach is a key strength of our collective work, given the intricate and multifaceted nature of polarization processes.

The special issue begins with a cross-national, region-wide assessment of mass- level polarization dynamics based on survey data, which is followed by a series of country or case-based analyses of Brazil, Chile, Mexico, and Colombia. The articles cover a wide range of topics and countries, ranging from ideological polarization in the region as a whole (Paolo Moncagatta and Pedro Silva’s article) to the link between populist storytelling and negative affective polarization in Mexico (Rodolfo Sarsfield and Zacarias Abuchanab’s contribution), and the polarizing and depolarizing messages of political elites on Twitter in Colombia (Laura Gamboa, Sandra Botero, and Lisa Zanotti’s article). Other articles also explain why the increasing ideological polarization at the elite level in Chile does not automatically map onto the mass level, with an observed significant dealignment among citizens (Juan Pablo Luna’s article), and how polarization is exacerbated by partisan stereotyping in Brazil (David J. Samuels, Fernando Mello, and Cesar Zucco’s contribution). Jennifer McCoy concludes the special issue with a comparative assessment of the dominant patterns identified in the contributions, and their implications for the broader study of polarization and democracy in contemporary global politics. This special issue also includes a review essay by Alfred Montero on major recent books that examine different dimensions of polarization in contemporary Latin American politics.

At the mass level, these articles examine trend lines in public opinion and mass political identification, as well as relationships between ideology, partisan brands and stereotypes, and socio-cultural identities and political affect. On one hand, when analyzing the changes in mass-level trends in Latin America, Moncagatta and Silva find a region-wide process of ideological restructuring and an increase in ideological polarization taking place in the region during the second decade of the twenty-first century. On the other hand, when studying the relationship between partisan stereotyping and polarization in Brazil, Samuels et al. show that the greater the bias in perceived partisan group composition, the greater the feelings of social distance towards the partisan out-group.

Articles also analyze social mobilization and protest dynamics and their impact on mass attitudes toward political institutions, including anti-establishment or anti-institutional sentiments. Proposing a new conceptualization to analyze Chile’s case, which is characterized as “disjointed polarization,” Luna points out that polarizing trends at the elite level can coexist with significant dealignment at the mass level. Some actors may polarize along the left-right axis (e.g., established political parties), whereas other actors can polarize along other dimensions, such as the competition between system insiders (the establishment) and outsiders (anti-establishment politicians).

At the elite level, other articles use data from Facebook and Twitter messages to examine populist discourse, misinformation, and issue framing; the diffusion of both polarizing and de- polarizing messages; and elite activation or stimulation of polarized attitudes in the public at large. On the one hand, proposing the concept of “negative affective polarization,” Sarsfield and Abuchanab explore the relationship between populist storytelling and polarized replies on Twitter (now X). Sarsfield and Abuchanab find that certain specific populist stories indeed tend to induce more negative polarized attitudes among citizens in Mexico. On the other hand, analyzing the characteristics of Twitter usage among political candidates in Colombia, Gamboa, Botero and Zanotti show that the visibility of a candidate does not necessarily correspond to a greater use of Twitter, an increased deployment of polarizing rhetoric, or the abuse of negative emotions in their messages.

Taken together, these articles contribute to the comparative study of polarization both at the level of descriptive inferences (i.e., conceptualization and measurement) and at the level of causal inferences (i.e., polarization as an independent and dependent variable). In so doing, they shed new light on the multiple dimensions of polarization processes, the factors that encourage and potentially inhibit polarization, and the impact of polarization on democratic representation and contestation in Latin America.

What Do We Mean by Polarization—and How Should We Measure It When We See It?

With the rise of research on polarization in recent years, the meaning of the concept has become a subject of debate. Several definitions of polarization have been offered by different scholars such as Alan Abramowitz, James Druckman, Morris Fiorina, Shanto Iyengar, Yphtach Lelkes, Matthew Levendusky, Jennifer McCoy, and Murat Somer, among others. Given the expansion of polarization processes to polities across different regions of the world, and the inherent multi-dimensionality of the phenomenon (Roberts 2022), there has been a proliferation of alternative forms of the concept, including a number of examples of polarization “with adjectives,” such as “societal polarization” (e.g., McCoy and Rahman 2016), “affective polarization” (e.g., Iyengar, Sood, and Lelkes 2012), and “pernicious polarization” (e.g., McCoy and Somer 2019b). Complicating the conceptual landscape, other scholars propose the ideas of “social polarization” (e.g., García-Guadilla and Mallen 2019), “geographical polarization” (e.g., Rohla et al. 2018), and “populist polarization” (e.g., Enyedi 2016). The lack of consensus and the multidimensional character of polarization processes become quickly noticeable when these different conceptualizations are examined.

Within this complex conceptual debate, however, we argue that most definitions of polarization can be grouped along two principal analytical dimensions that allow the development of a typology for the conceptualization of polarization. The first dimension focuses on the nature of polarization, distinguishing between ideological and affective polarization. The second dimension refers to the level of analysis at which polarization occurs (e.g., elite or mass polarization). Although this typology is not exhaustive, it allows the ordering of different definitions of polarization that have been used in the literature—particularly those for which instruments of measurement have been developed and systematically adopted for empirical research.

Regarding the first dimension, we find two major approaches to the concept of polarization. The first approach locates polarization on an ideological plane, and examines ideological polarization by focusing exclusively on parties’ and voters’ ideological views or policy preferences (e.g., Abramowitz and Saunders 2008; Fiorina 2017; Fiorina, Abrams, and Pope 2006; Fleisher and Bond 2001; Hetherington 2001; McCarty, Poole, and Rosenthal 2016). According to this approach, polarization occurs when ideological views and/or policy platforms move spatially toward the left and right poles, rather than converging near the center. Parties or voters, therefore, move further apart spatially and compete on the basis of sharply differentiated programmatic alternatives. Conversely, a political community does not show polarization when the majority of political leaders, parties, and citizens adhere to centrist positions in their views on major policy questions. In ideological terms, therefore, polarized competition has a centrifugal spatial logic, whereas non-polarized competition is centripetal in character.

The second approach to the nature of polarization, labeled affective polarization, conceptualizes the phenomenon in terms of emotional feelings or attitudes rather than ideological preferences (e.g., Iyengar, Sood, and Lelkes 2012; Iyengar et al. 2019; Druckman and Levendusky 2019). According to this definition, polarization is the extent to which partisans or rival socio-political camps view each other as a disliked and distrusted out-group. In this vein, it is necessary to distinguish between two types of affective polarization that are analytically distinct: in-party/out-party affective polarization, that is, affinity with the in-party elite and hostility toward the out-party elite, and in-group/out-group affective polarization, that is, affinity with in-party partisans and hostility toward out-party partisans (Druckman and Levendusky 2019). In empirical terms, these two forms of affective polarization may or may not coexist. This understanding of polarization is based on the classic concept of social distance (Bogardus 1947), and it “requires not only positive sentiment for one’s own group, but also negative sentiment toward those identifying with opposing groups” (Iyengar, Sood, and Lelkes 2012, 406). Affective polarization occurs when partisans of one side dislike and distrust or even fear those from the other (Iyengar et al. 2019), that is, when we find mutual animosity between supporters of a party A and those of a party B.

It should be recognized, however, that affective polarization can exist even in the absence of strong positive attachments to any given party, if some voters are motivated primarily by their antipathy for a certain party or political leadership they especially dislike. Samuels and Zucco (2018) provide powerful evidence of such forms of “negative partisanship” in their analysis of polarizing opposition to the Workers’ Party (PT) in Brazil. Meléndez (2022) suggests that polarizing negative affect may also form in opposition to a trans-partisan political elite or establishment, often derided as a casta política (political caste) in populist rhetoric. In short, whereas ideological polarization refers to the spatial distance in policy preferences between parties or voters (e.g., Roberts 2022), affective polarization tries to capture the degree of mutual animosity between them—to the point where one group may not recognize the other as legitimate actors in the democratic arena (e.g., McCoy and Rahman 2016). These two types of polarization are analytically distinct and potentially independent of each other, as affective polarization can exist in the absence of well-defined ideological distancing. In practice, however, the two kinds of polarization may coexist and reinforce each other—i.e., the greater the ideological distance between rival political camps, the more they are prone to dislike and distrust each other. Building on the classic work of Sartori (1976), who associated polarization with the presence of anti-system parties, Schedler (2023, 338) argues that the analytic core of contemporary polarization is not ideology or affect per se, but rather the breakdown of “basic democratic trust” that occurs when major actors question the democratic commitments of their rivals and their adherence to the accepted “ground rules of democratic dispute resolution.”

Considering polarization’s second dimension, that is, the level of analysis at which it occurs, the literature has proposed a distinction between elite and mass polarization. Elite polarization refers to ideological and/or affective polarization between the leaderships of the major parties or political movements vying for state power (Baldassarri and Gelman 2008). Polarization may sometimes be an intentional strategy employed by political leaders to demarcate rival political camps, discredit their opponents, and mobilize supporters (Penfold and Corrales 2007). Research on US politics, for example, has used congressional voting records to chart the steady increase in the ideological distance between Republican and Democratic members of Congress in recent decades (McCarty, Poole, and Rosenthal 2016, 33–34).

Mass polarization, on the other hand, refers to ideological and/or affective polarization between a social in-group (e.g., members of one’s own party or social bloc) and a social out-group (e.g., members of the opposing party or social bloc) (Levendusky 2009; Mason 2018). Mass polarization can be identified along multiple fronts, including public opinion, voting behavior, party and civic activist networks, and social movement mobilization. It is important to recognize that these multiple fronts do not necessarily coincide; public opinion surveys, for example, may show most citizens grouped around centrist positions, even while voting behavior skews toward the left and right poles despite the presence of centrist alternatives in multi-party systems. Likewise, grass-roots activist networks and social movements rarely mobilize a majority of the electorate, but the minorities they do mobilize may have disproportionate influence in pulling political competition toward the poles because of their relatively high levels of political engagement, organization, and ideological commitment (Gillion, 2020; McAdam and Kloos 2014; Schlozman 2015). The mobilization of such minorías ruidosas (loud minorities), and the frequent counter-mobilization of their most ardent opponents, helps to explain why polarizing dynamics are not always reflected in aggregate public opinion surveys.

Following a similar idea for the same dimension (the level at which polarization occurs), some scholars suggest a typology of polarization distinguishing political from societal polarization. Thus, according to Oosterwall and Torenvleid, “political polarization points at large differences […] between opposed ‘camps’ or ‘coalitions’ of political parties, and high similarities […] within these coalitions,” whereas “societal polarization is the equivalent for the existence of a few, large groups in society” (Oosterwaal and Torenvlied 2010, 261). So conceived, political polarization is specific to parties and/or political coalitions, whereas societal polarization is specific to social groups.

The different definitions of polarization not only address typological distinctions but also differences in the degrees of polarization. Thus, some authors use the idea of pernicious polarization—labeled also as “toxic polarization” (Boese et al. 2022)—to refer to extreme forms of political and/or societal polarization with a type of Manichean struggle between “friends” and “enemies” that hinders social interaction and blocks cooperation. With pernicious polarization, “there is a propensity to view the ‘other’ group as essentially homogenous and treat the members of that group according to some stereotypical notion” (McCoy and Rahman 2016, 8; see also Sarsfield and Abuchanab, and Samuels, Mello, and Zucco in this special issue). Each group questions the legitimacy of the other, viewing the opposing group as a threat to their way of life. This is a key point, as it is the starting point for democratic institutions not being able to process political conflict (Bobbio 1996; Schedler 2023). Democracy can process conflicts when rival groups recognize the other side as legitimate interlocutors for societal interests and preferences in the democratic arena; when the other side is viewed as illegitimate or threatening, however, competition assumes an existential character, democratic norms start to break down, and anti-democratic practices are likely to follow.

After reviewing the different conceptualizations of polarization used in the literature, the next step is to consider how they have been measured. Several different indicators have been developed to measure ideological and affective polarization, at both elite and mass levels of analysis. Some scholars measure ideological polarization via surveys using standard questions on ideological self-placement and the distance between parties, leaders, or voters on a left-right scale, or a liberal-conservative scale (Conover and Feldman 1981; Hinich and Munger 1994; Malka and Lelkes 2010; Moncagatta and Silva in this special issue). Other authors contend, however, that not all ideological polarization is amenable to measurement on a left- right scale (e.g., McCoy and Rahman 2016; Vegetti 2014). ¹ Thus, other indicators that assess voters’ issue positions on different policy items have been used, such as attitudes toward government spending on social security, support for a government health insurance plan or other government services, the expectation that government should guarantee jobs and a standard of living, and support for defense spending (e.g., Iyengar, Sood, and Lelkes 2012; Abramowitz and Saunders 2008). In addition, other scholars propose that polarization can be measured not only along a left-right economic axis (i.e., state-market), but also on a cultural axis that would include contentious issues such as abortion and women’s rights, gay rights, minority rights, environmental protection, and law and order (Roberts 2022; Ostiguy 2017). We return to this topic below.

Regarding affective polarization, scholars have measured it using different survey instruments (Iyengar et al. 2019). One of the most used survey instruments is known as the feeling thermometer rating. This instrument asks respondents to rate how cold or warm they feel toward a liked in-group (typically, one’s own party) and toward a disliked out-group (typically, the opposing party) (Druckman and Levendusky 2019, 115; Lelkes and Westwood 2017, 489; Tajfel and Turner 2001). A second instrument asks respondents to rate how well a set of positive and negative traits describe one’s own party and the opposing party (Iyengar, Sood, and Lelkes 2012; Garrett et al. 2014). A third type of instrument, known as social-distance measures, gauge how comfortable people feel with having close friends or neighbors from the other party, or having their children marry someone from the other party (Druckman and Levendusky 2019, 115–16; see also Iyengar, Sood, and Lelkes 2012; Levendusky and Malhotra 2016).

The Complex Relationship Between Polarization and Democracy

The relationship between polarization and democracy is highly complex, as polarization can be both an independent variable that shapes democratic outcomes, and a dependent variable that is conditioned by democratic competition. Adding to that complexity, polarization is widely considered to be a threat to democracy, but it can also serve as a potential corrective to certain democratic deficiencies. Somer and McCoy’s work provides a good synthesis of these debates, as they recognize that polarization can be associated with both democratic strengthening and democratic erosion (Somer and McCoy 2018). Under certain conditions, polarization “can help to strengthen political parties and institutionalize party systems because it enables them to mobilize voters around identifiable differences” (McCoy and Somer 2019b, 235). Scholarship on both Europe (Berman and Kundnani 2021) and Latin America (Roberts 2014) has suggested that partisan programmatic convergence can weaken parties as representative agents and leave party systems vulnerable to the rise of populist contenders. Some degree of polarization is arguably necessary for parties to challenge entrenched social hierarchies and effectively represent the full range of societal interests, preferences, and identities found in any modern polity. Extreme polarization, however, may prevent rival political actors from coordinating to address commons social problems, and it may even lead them to deny the democratic legitimacy of their rivals.

Regarding the complex relationship between polarization processes and democratic systems, conceived as both causes and effects of each other, Somer and McCoy (2018) postulate three possibilities:

1. 1) erosion and democratic backsliding as a result of polarization

2. 2) emergence of polarization as a result of democratic crises, and

3. 3) strengthening of democracy as a result of polarization.

The first scenario usually occurs when there are groups that mobilize in order to build coalitions to defend a social, economic, political, or institutional agenda in response to the perception that there is another group that represents an existential threat to their interests or beliefs. In some cases, these groups are mobilized by the rhetoric of a leader who intentionally seeks to polarize the electorate to demonize opponents and mobilize supporters (Penfold and Corrales 2007). Regardless of whether the rhetoric used for mobilization is intentionally polarizing or not, there is usually a counter-mobilization of the group that represents an antagonistic social or political identity. This creates a recursive feedback loop that raises the level of polarization, and the possibilities of de-escalating social tensions are undermined. From the perspective of the state, when there are two social or political blocs with great pressure capacity but transversally different programmatic preferences and expectations, the ability to respond to social demands is weakened. When there are high levels of polarization, competing groups may perceive that undesired outcomes are due to state capture or paralysis by the “enemy.” Ultimately, elites and citizens belonging to these polarized groups may become more likely to employ undemocratic—or even violent—methods to exclude rivals and impose their agendas.

In the second scenario, democratic crises constitute the antecedent that prepares the ground for the emergence of polarization, making polarization the effect more than the cause of democratic dysfunction. In these scenarios, the inability to generate a consensus to cope with a country’s political, social, or economic problems gives rise to new tensions and contradictions between different groups. Ultimately, these differences form polarized positions on the expectations and even the procedures of democracy (for example, support for liberal principles vs. potentially illiberal practices based on populist and plebiscitary notions of empowering “the people” against a crisis-ridden democratic establishment). Over time, some of the groups could even seek a profound institutional reconfiguration to establish a majoritarian political order in accordance with their preferences, which could include clauses of exclusion or a lack of protection for the “other.”

The third scenario refers to the positive effects that polarization could generate in democratic systems. From this perspective, moderate degrees of polarization are capable of revitalizing the incentives for popular participation, inclusive democratic representation, and broad-based political cooperation. Different political and social actors who normally do not articulate their interests due to factors such as social atomization or political apathy are more likely to become politically activated if a new party emerges to champion their cause, or if an adversary attempts to impose its public agenda on one or more issues. As well, in the field of party competition and electoral preferences, a certain degree of polarization around programmatic differences could allow citizens to have clearer information and rely on simpler cues to make decisions about which type of political party or candidate best articulates their priorities based on a particular cultural identity, socio-economic interests, etc. Finally, under authoritarian regimes, pro-democratic actors can use polarization as a tool for articulating opposition efforts and mobilizing forces to make their opposition to the government more effective (Somer and McCoy 2018).

Other authors have highlighted potential positive effects of political polarization. In a Laclauian twist to the understanding of polarization, Stavrakakis (2018) argues that under certain conditions, polarization may be positive for democracy, as it can act as an agent of popular mobilization and inclusion. Using Greece as a case study, the author argues in favor of a “benign” conception of polarization, based on the potential for building a constructively agonistic and pluralistic democracy (Stavrakakis 2018). LeBas (2018), basing her analysis on four cases in sub-Saharan Africa, claims that whenever there is a balance of forces between groups on both sides of the political divide, and in the absence of a history of formal group exclusion, a relatively high degree of polarization can have positive institution-building effects for new democracies in the long run. Similarly, Levitsky et al. (2016) suggest that polarization and conflict are integral to successful party building in Latin America.

Attention has also been paid to the relationship between populism and polarization, since populism invariably entails the binary division of the political field between “the people,” however conceived, and some sort of establishment or power elite (Laclau 2005). Most scholars who have studied this relationship consider populism to be a cause of polarization, which in turn is often seen as detrimental to democracy. De la Torre and Ortiz Lemos analyze the regime of Rafael Correa in Ecuador to illustrate “how the interactions between a strong populist government and a weak, divided, and inefficient internal opposition in a context of weak liberal institutions could lead to what Guillermo O’Donnell termed ‘the slow death of democracy” (De la Torre and Ortiz Lemos 2016, 221). Enyedi (2016, 217) studies the recent process of dramatic democratic backsliding in Hungary to conclude that “populist polarization”—understood as “the combination of the intense and aggressive competition between party blocs, the concomitant rejection of the division of power, the focus on the question of who the ‘people’ are, and the central role of relatively stable and strong parties—is the principal cause of the difficulties in the consolidation of liberal democracy in the country (and in the Eastern European region in general).” Roberts (2022: 695–96) argues that increasing polarization may well be the most consistent effect of populism, but recognizes that “populist polarization would appear to be a two-edged sword for democracy.” Under certain circumstances, it can have the “potential to broaden and invigorate democratic representation,” but it can also acquire “an inertial, self-reinforcing quality that makes it ‘pernicious,’ in the terminology of McCoy and Somer (2019a), and corrosive for democracy itself.” In a similar vein, Stavrakakis argues that “populism may or may not be a trigger of pernicious polarization and a threat to democracy, depending on its inclusionary or exclusionary character” (Stavrakakis 2018, p. 52).

Due to the existence of different types and levels of polarization, it is necessary to identify which are the most detrimental to democracy, as well as their dynamics and consequences. McCoy and Somer (2019b) identify “pernicious” polarization as especially harmful to democratic systems. This level of polarization occurs “when the normal multiplicity of differences in a society increasingly align along a single dimension, cross-cutting differences become instead reinforcing, and people increasingly perceive and describe politics and society in terms of ‘Us’ vs. ‘Them’” (McCoy, Rahman, and Somer 2018; see also Sarsfield and Abuchanab, in this special issue). In this conception, each political group perceives its counterpart to be an existential threat to the nation or to the ways of life they consider superior (Somer, McCoy, and Luke 2021). Furthermore, certain sectors of the population are predisposed to apply non-democratic strategies to defend their interests and their access to, or retention of, political power. Among its consequences, high levels of mistrust in public sphere activities and institutions can be evidenced, as well as a generalized lack of cooperation between actors belonging to different groups. This, in turn, can negatively affect inclusive socio-economic development and the efficiency of private economic activities (Keefer and Scartascini 2022).

The Varieties of Democracy (V-Dem) Project reports increasing levels of what they have labeled “toxic polarization” that threatens democracy in many countries (Boese et al. 2022). According to the 2022 V-DEM Democracy report, “polarization is increasing to toxic levels in 40 countries” (Boese et al. 2022, 7). This contributes to autocratization processes, as polarization can leave a fertile ground for electoral victories by extremist and anti-pluralist leaders and the empowerment of their agendas. While there is no exact threshold to identify when polarization becomes “toxic” for V-DEM, they mention that it “becomes toxic when it reaches extreme levels,” and camps of “Us vs. Them” start questioning the moral legitimacy of each other (Boese et al. 2022, 30). According to V-DEM’s analyses, polarization and autocratization form a mutually reinforcing, vicious cycle.

In their exhaustive review of the literature on the relationship between social media, political polarization, and political disinformation, Tucker et al. (2018) observe increasing levels of affective polarization or “negative partisanship” that have generated remarkable hostility toward opposition party identifiers (Iyengar, Sood, and Lelkes 2012; Webster and Abramowitz 2017). According to these scholars, these “negative feelings threaten to undermine norms of civility and mutual respect in political debate. The strongly negative affective reactions that opposition partisans now inspire create a constituency for the winner-take-all political tactics […] and undermine the incentives for elites to engage in civil discourse and policy compromise” (Tucker et al. 2018, 50). Social media can exacerbate these polarizing tendencies when citizens are uniformly exposed to highly partisan and often misleading political messaging aimed at vilifying rivals or sowing distrust of regime institutions. In the worst-case scenario, citizens may get locked into antagonistic media echo chambers that insulate them from alternative perspectives and sources of information (Persily and Tucker 2020).

Ultimately, polarization dynamics at the elite and mass levels may interact and reinforce each other, with negative consequences for the quality of democratic governance even when not necessarily threatening the survival of a democratic regime. The deepening of differences between political leaders, for example, can lead to increasing trust problems, thus preventing them from negotiating policies and reaching minimum agreements on pressing social problems. In extreme cases, situations of “political gridlock” can be generated, with the paralysis of the public administration, as well as the promotion of negative campaign agendas that only seek the repeal of policies promulgated by partisan rivals. This level of polarization results in political parties that represent the interests of polarized social and political identities, promoting increasingly distant agendas in spatial terms. Therefore, within the electoral offer, agendas that seek to capture the median voter through moderate proposals become increasingly scarce (Downs 1957).

As mentioned earlier, polarization can also become “societal,” when it extends from the political sphere to a large part of social relations, exporting the friend vs. enemy logic to normally apolitical spheres. Among the spaces that have become politicized are the family sphere, educational institutions, and churches (Somer and McCoy 2018). In extreme cases, this type of polarization can feed the perception that coexistence with groups adopting different identities is unfeasible, thus significantly eroding bridging social capital (Putnam 2000).

Contemporary Polarization in Historical Perspective

In historical perspective, political polarization has ebbed and flowed in Latin America, and the levels and types of polarization have varied considerably across countries. Following the rise of labor-based mass politics in the early to mid-twentieth century, only Chile developed a mass-based partisan left—both Socialist and Communist—to structure its party system along a conventional (by European standards) left-right ideological axis. In several other countries, however, populist labor and social mobilization sharply divided the political arena between elite and mass-based parties, creating highly polarized competition between, for example, Peronists and anti-Peronists in Argentina, or Apristas and anti-Apristas in Peru. While less clearly structured along ideological lines, these socio-political cleavages led to the frequent proscription of populist contenders from the democratic arena (Collier and Collier 1991). Even in countries with lower levels of leftist or populist labor mobilization, intra-elite competition between rival partisan patronage machines was capable of generating intense conflict, ranging from authoritarian political closures to civil war. This could be seen in the historical conflicts between Liberals and Conservatives in Colombia, Liberals and Nationalists in Honduras, Blancos and Colorados in Uruguay, or Liberals and Colorados in Paraguay.

Polarization in Latin America reached unprecedented levels of intensity and ideological structuring in the aftermath of the Cuban Revolution, which inspired a plethora of left-wing insurgent movements in the 1960s and 1970s, along with a virulent right-wing backlash that spawned highly repressive forms of military authoritarianism across much of the region (O’Donnell 1973; Weyland 2019).

Contemporary patterns of political polarization appear relatively tame in comparison to that violent era, when major actors on both the left and the right abandoned democratic institutions and took up arms to advance or defend their political goals. The trauma of political violence, however, along with the realities of political gridlock, exerted a sobering effect that tempered ideological conflict and paved the way for democratic transitions across the region in the 1980s (O’Donnell and Schmitter 1986). Although these transitions varied in their levels of elite-level pacting and bottom-up mobilization, they reflected a generalized convergence among political actors that liberal democracy had become the “only game in town” (Linz and Stepan 1996). Key elements on the left abandoned revolutionary armed struggle, relaxed their commitments to socialist development models, and embraced the protections for human rights and civil liberties afforded by democratic institutions. With the Soviet model of socialism entering its death spiral, and labor movements in Latin America decimated by authoritarian repression and economic crisis, conservative actors could dispense with their military protectors and tolerate forms of democratic competition that posed little threat to elite economic interests.

Indeed, Latin America’s debt crisis and hyperinflationary spirals forced political actors across most of the ideological spectrum to accept neoliberal structural adjustment programs by the end of the 1980s, producing a de facto programmatic convergence that was a far cry from the acute ideological polarization between capitalism and socialism of the recent past (Williamson 1990; Edwards 1995). It is largely in comparison to this perhaps “exceptional” period of political moderation and programmatic convergence that contemporary Latin America appears so polarized.

That convergence, however—which included the adoption of neoliberal reforms by center-left and labor-based parties that were historic bastions of statist and redistributive development models in countries like Argentina, Mexico, Venezuela, Bolivia, Ecuador, and Costa Rica—was highly contingent politically. At the elite level, it rested on the narrowing of viable economic platforms and a virtual eclipse of social democratic alternatives during a period of severe, crisis-induced global market constraints. At the mass level, it depended on the widespread disarticulation and demobilization of labor unions, peasant associations, and other social actors during the initial stages of the region’s free market revolution (Kurtz 2004).

The “Washington Consensus” for neoliberal reform was not destined to survive the repoliticization of social and economic inequalities at the turn of the century, however, and the attendant strengthening of new social movements and leftist alternatives that increasingly challenged neoliberal orthodoxy following the defeat of hyperinflation in the early to mid-1990s (Weyland, Madrid, and Hunter 2010; Levitsky and Roberts 2011). Indeed, the notion of a programmatic “consensus” masks the fact that neoliberal orthodoxy itself had polarizing properties, as it pushed regional political economies further toward the pro-market right pole on the economic axis. In so doing, it left ample space for policy contestation to reemerge even in the absence of well-defined socialist alternatives.

This policy contestation was not always highly polarizing, at least in the short term. Where a major party of the left remained in opposition during the period of structural adjustment, it could channel societal resistance to neoliberal orthodoxy into the democratic arena, strengthen electorally over time, and lead relatively moderate, highly institutionalized political shifts to the left in the early twenty-first century—a pattern seen in Brazil, Chile, and Uruguay. Polarization was more pronounced in countries where structural adjustment policies were imposed by traditional labor-based or center-left parties, causing party systems to converge on neoliberal orthodoxy and channel societal dissent into extra-systemic forms of mass social protest. In countries like Venezuela, Ecuador, Bolivia, and Argentina, the surge of new social movements—including indigenous movements, unemployed workers, and the urban poor (Yashar 2005; Silva 2009; Madrid 2012; Rossi 2017)—helped bring down elected presidents, undermine mainstream parties, and open the door to new and more radical populist challengers to the left of established party systems (Roberts 2014).

A pattern thus emerged whereby mass social protest served as a precursor to mass electoral protest—that is, the election of left populist “outsiders” who rejected the entire political establishment, demonstrating the interrelated processes of elite and mass polarization. Hugo Chávez in Venezuela, Evo Morales in Bolivia, and Rafael Correa in Ecuador were staunch critics of the neoliberal economic models adopted by mainstream parties in their countries, and they advocated varied forms of statist or even socialist development models and ambitious redistributive social programs as alternatives (Ellner 2022). More fundamentally, they used popular referendums to claim and exercise constitutive powers, convoking constitutional assemblies to refound national democratic regimes. In so doing they concentrated powers in the hands of chief executives and their newly-mobilized popular majorities, albeit in ways that often posed highly polarizing threats to traditional political and economic elites, minority political rights, and regime checks and balances (Corrales 2018; Velasco 2022; Carrión 2022). As such, regime contestation was grafted onto the left populist vs. establishment divide, and opposition forces—largely relegated to the margins of regime institutions—sometimes resorted to extra-institutional strategies for removing populist rulers, ranging from mass protests and strikes to threats of secession and attempts at military coups (Velasco 2022; Gamboa 2022; Ellner 2022).

In contrast to the United States, where polarization is often characterized as asymmetric, driven largely by the radicalization of the Republican Party (Grossman and Hopkins 2016; Lieberman, Mettler, and Roberts 2022), it has more pronounced bilateral tendencies in Latin America. It is not, however, unidimensional, arrayed along a single left-right competitive axis defined by positioning on economic issues (i.e., preferences on the left pole for statist and redistributive policies, vs. an emphasis on free markets and private property rights on the right pole). Although polarization along this economic axis has surely increased since the 1990s, new forms of polarization have also been heavily conditioned by political conflict around cultural issues which do not necessarily map onto the same left-right axis. Indeed, cultural dimensions of political contestation are at least potentially orthogonal to the left-right economic axis, as portrayed in Figure 1 above. By challenging traditional social hierarchies and conventions, political mobilization around gender equality and abortion rights, LGBTQ rights, environmental protection, and indigenous self-determination invariably elicits counter-mobilization by conservative defenders of the traditional social order, typically positioned on the political right. It may also, however, clash with the cultural conservatism and developmentalist logic of many left populisms as well. The new, movement-based lefts that ushered into power leaders like Gabriel Boric in Chile and Gustavo Petro in Colombia, who have taken strong stands on issues of gender equality, indigenous rights, and the environment (placing them in the top left quadrant of Figure 1), are quite different from the more populist lefts associated with leaders like Rafael Correa in Ecuador, Andrés Manuel López Obrador in Mexico, and Pedro Castillo in Peru (located in the bottom left quadrant of Figure 1).

Figure 1. Polarization in Two-Dimensional Space

A new, more radical right—at times with populist and markedly authoritarian tendencies—has also emerged in contemporary Latin America by politicizing the lower pole on the cultural axis in Figure 1 (Mayka and Smith 2021). ² Leaders like Jair Bolsonaro in Brazil and José Antonio Kast in Chile have appealed to conservative religious values, patriarchal norms, ethno-nationalist identities, and law-and-order demands to mobilize support among popular constituencies who might not otherwise favor a right-wing candidate on economic issues. Such politicization of the cultural axis provides a potential response to the “conservative dilemma,” i.e., the electoral challenges encountered by parties of the right whose core constituencies are comprised of economic elites (Gibson 2001; Ziblatt 2017). But it has exacerbated polarization and aligned the right with a highly exclusionary if not openly authoritarian brand of politics—one that is hostile towards feminists and sexual minorities and committed to mano dura (iron fist) anti-crime policies that threaten to erode civil liberties while criminalizing the poor, immigrants, and racialized minorities (Mayka, forthcoming). The open identification of leaders like Bolsonaro and Kast with national military regimes of the recent past—not to mention Bolsonaro’s efforts to delegitimize and then overturn Brazil’s 2022 presidential election—brings to the surface the authoritarian current that continues to run deep within the Latin American right (Hunter and Power 2019), much like the gradual descent into full-fledged autocracy in Venezuela under Nicolás Maduro and Nicaragua under Daniel Ortega exposes its counter-current on the left.

In ideological terms, therefore, polarization is best conceived in a two- dimensional space, recognizing the existence of political conflict along cultural dimensions that are potentially independent of left-right positioning on the economic axis, even if in practice they often map onto that horizontal axis (on the latter, see Martínez-Gallardo et al. 2022). The Latin American experience also demonstrates that polarization cannot be understood with a narrow fixation on either elite or mass-level political dynamics; it inevitably involves a complex, reciprocal interaction between mass publics, civic activist networks, social movements, party organizations, and political elites. This reciprocal interaction is heavily conditioned by political contestation around democratic deficits—in particular, the limited reach of democratic representation and citizenship rights in contexts of extreme economic inequalities, entrenched social hierarchies, rampant corruption, and runaway crime. Economic hardships and inequalities have repeatedly spawned mass protest movements—in Venezuela, Argentina, Ecuador, and Bolivia before and after the turn of the century, and in Chile, Honduras, Peru, Ecuador, and Colombia more recently—that set the stage for the rise of new populist or movement-based electoral alternatives well to the left of the established party systems of the neoliberal era. Even a relatively well-functioning and institutionalized democratic regime like that of Chile has been rocked by repeated social uprisings, or estallidos, that politicized the limits of social citizenship rights in a neoliberal order with highly privatized and profit-based education, health care, and pension systems (Rhodes-Purdy and Rosenblatt 2023).

Likewise, indigenous movements in the Andean region and Guatemala, along with feminist movements in countries like Argentina, Mexico, and Chile, have challenged traditional social hierarchies and claimed new social citizenship rights that are stridently opposed by conservative defenders of the traditional order. Institutionalized corruption has bred anti-establishment political mobilization across the ideological spectrum, while citizen insecurity in the face of rampant crime and criminal violence has fostered the rise of a new far right pledging highly punitive forms of law and order, following leaders such as Nayib Bukele in El Salvador, Bolsonaro in Brazil, Kast in Chile, and Javier Milei in Argentina.

Polarization, then, can be seen as both a cause and an effect of democratic failings. Acute polarization can undermine democracy by making rival actors mutually intolerant and less willing to recognize each other’s political rights or legitimacy to govern. But autocratic attempts to concentrate power, exclude rivals, or weaken democratic checks and balances are also highly polarizing in their own right, as they dramatically raise the stakes of democratic contestation by threatening to make transitory victories—and losses—cumulative and permanent. Finally, democracy’s many limitations in Latin America generate forms of discontent that are conducive to anti-establishment or even anti-systemic forms of political mobilization, from a range of different ideological perspectives, at both elite and mass levels of analysis.

It is important to recognize, however, that polarization is not simply a response to democratic failings or dysfunctions. The mobilization of far-right forces in the region in defense of traditional social hierarchies has elements of a backlash against the democratic advances made by women’s and indigenous movements in a number of Latin American countries. Paradoxically, even the modest gains of the “inclusionary turn” (Kapiszewski, Levitsky, and Yashar 2021) in third-wave democracies can have polarizing effects in a region where social and economic inequalities are deeply entrenched, and the socio-cultural norms surrounding traditional patterns of subalternity are vigorously defended by dominant groups.

Overview of the Volume

The articles that follow provide ample evidence of these varied causes and consequences of polarization. Moncagatta and Silva develop a novel measure of mass ideological polarization using individual-level Americas Barometer survey data. They demonstrate a moderate rise in aggregate levels of ideological polarization across the region in the twenty-first century, while highlighting significant variation in the intensity and trend lines of polarization dynamics across different countries. Samuels, Mello, and Zucco Jr. also work with survey data to examine the impact of partisan stereotyping on polarization processes in Brazil. Their findings indicate that both PT and anti-PT citizens misperceive the social composition of their opponents and supporters, and these stereotyped views contribute to exaggerated perceptions of partisan extremism, social distance, and programmatic differences.

The next two articles shift attention to political elites’ use of social media to disseminate polarizing political messages to their mass constituencies. The article by Botero, Gamboa, and Zanotti examines Colombia’s 2022 presidential race to see whether polarizing Twitter messages spread further, faster, and deeper than their de-polarizing counterparts. The article by Sarsfield and Abuchanab analyzes the social media dissemination of populist narratives or “storytelling” in Mexico. It finds that stories of conspiracy and ostracism have a significant impact on citizens’ attitudes, as they are associated with negative affective polarization.

Finally, Luna explores the impact of social protest and a crisis of representation on polarization dynamics in Chile. Although indicators of polarization are present, Luna suggests that they are of secondary importance to the failures of political intermediation or representation that have deepened the divide between Chilean citizens and the state. In her conclusion, McCoy explains what Latin America can teach us about the politics of polarization, and compares the regional patterns to those found in other parts of the world. Lastly, a review essay by Montero examines the most recent books on polarization in Latin America, enriching the portrait drawn by this special issue.