Populism and Polarization in Comparative Perspective: Constitutive, Spatial and Institutional Dimensions

Abstract

Polarization may be the most consistent effect of populism, as it is integral to the logic of constructing populist subjects. This article distinguishes between constitutive, spatial and institutional dimensions of polarization, adopting a cross-regional comparative perspective on different subtypes of populism in Europe, Latin America and the US. It explains why populism typically arises in contexts of low political polarization (the US being a major, if partial, outlier), but has the effect of sharply increasing polarization by constructing an anti-establishment political frontier, politicizing new policy or issue dimensions, and contesting democracy's institutional and procedural norms. Populism places new issues on the political agenda and realigns partisan and electoral competition along new programmatic divides or political cleavages. Its polarizing effects, however, raise the stakes of political competition and intensify conflict over the control of key institutional sites.

The rise of populist parties, movements and leaders can have a number of important consequences for national political systems, but their most consistent effect may well be that of increasing polarization. Takis Pappas (2019: 212), in fact, proclaims ‘extreme polarization’ to be ‘the absolutely most important element of populist rule’. Although these polarizing effects are widely recognized (see, for example, Enyedi 2016; Handlin 2018; Urbinati 2019), they have received little explicit theoretical attention from scholars, despite the routine attachment of polarizing descriptors – such as radical, extreme or anti-systemic – to political actors identified as populist on both the left and right sides of the spectrum. Recent work has provided some much-needed cross-fertilization of the largely separate literatures on populism and polarization (McCoy and Somer 2018; Pappas 2019; Stavrakakis 2018), but the multiple dimensions of populism's polarizing effects remain poorly understood or too easily conflated in much of the field. A deeper understanding of populism's polarizing logic can shed new theoretical light on the political conditions most conducive to its rise, as well as its effects on democratic institutions.

This article addresses these issues by analysing the system-level dynamics of polarization under populism and developing a conceptual framework to identify how they structure and realign political contestation. Building on Jennifer McCoy et al. (2018), I differentiate between primary and secondary dimensions of populist polarization. The primary dimension is constitutive in nature, as it inheres in the constitution of a populist subject and its dichotomous cleaving of the political field. The secondary dimensions are spatial and institutional in character, as they reflect ideological or programmatic distancing within that field – that is, the politicization or radicalization of issue positions in democratic competition – and heightened contestation over democracy's institutional and procedural norms. The constitutive dimension of polarization is intrinsic to – indeed, definitional of – populism; the spatial and institutional dimensions, on the other hand, are routine, but not essential, consequences of populism. Combining these three dimensions, polarization can be defined as the binary division of society into antagonistic political camps, pushed further and further apart.

So conceived, polarization is congenital to populism – an essential component of its political DNA – but also variable in its form and content. By adopting a macro-level, cross-regional comparative perspective – analysing populism in Europe, Latin America and the US – this article explores how polarization affects the social bases and programmatic content of partisan competition, as well as the functioning of democratic institutions. It also provides a conceptual framework to facilitate more micro-analytic studies of populist attitudes and beliefs. The cross-regional perspective makes it possible to analyse a wider and more diverse set of cases where populist leaders or movements have come to power at the national level, not only as coalition partners, but as governing parties that control national executive office and command legislative majorities.

The comparative analysis suggests that populism typically arises in contexts of mainstream party convergence with low political polarization – the US being a major, if partial, outlier. It has the effect, however, of sharply increasing polarization by constructing an antagonistic frontier between establishment and anti-establishment forces (see Laclau 2005: 153–154) and, in many cases, by politicizing new policy or issue dimensions and contesting the institutional arrangements for democratic pluralism. Populism, therefore, generates a new political cleavage, places new issues on the political agenda and realigns partisan and electoral competition. In so doing, it has the potential to provide ‘voice’ or representation to previously excluded or marginalized sectors of the national polity. But as Yannis Stavrakakis (2018) emphasizes, populism also elicits a counter-mobilization by anti-populist forces that intensifies polarization. Its impact on governing institutions, therefore, can vary considerably depending on the competitive equilibrium between populist and anti-populist actors and their relative control over institutional sites that offer countervailing checks and balances. Polarization raises the stakes of political competition when it politicizes these institutional sites and transforms them into instruments of partisan advantage.

Populism, pluralism and constitutive polarization

The study of populism and polarization have both experienced growth spurts of late (see, for example, de la Torre 2019; Mason 2018; McCoy and Somer 2018; Mudde 2017; Rovira Kaltwasser et al. 2017), and while these fields of study intersect, the complex interactions between the two political phenomena cry out for greater theoretical integration. The simple claim that populism is intrinsically polarizing, for example, is complicated by the fact that polarization occurs along several distinct and potentially independent analytical dimensions. As stated above, polarization involves the binary division of society into antagonistic political camps, pushed further and further apart. But even this simple definition contains distinct constitutive and spatial dimensions. As explained below, the constitutive dimension refers to the consolidation and constitution of political subjects that divide society into binary and antagonistic political camps. This dichotomization is well-captured by McCoy et al. (2018: 18), who define polarization as ‘a process whereby 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” versus “Them”’.

Constitutive polarization, however, does not tell us how far apart the rival camps are located in ideological or programmatic space – that is, how polarized they are spatially. Neither does it tell us whether the rivalry between these camps transgresses or threatens the functioning of democratic institutions. The ideological distancing that was central to Giovanni Sartori's (1976) influential conceptualization of polarization, as well as to much of the American politics literature on the topic (Abramowitz 2011; McCarty et al. 2016), can certainly reinforce and help define rival political camps, but it is not essential to their construction; social and political cleavages can form and ‘sort’ mass constituencies in the absence of major ideological differences. Indeed, the logical independence of the constitutive and spatial dimensions of polarization is reflected in Sartori's linkage of polarization to multiparty pluralism, rather than political dichotomization. Fragmented multiparty systems, he believed, allowed small radical parties to exist on the ideological fringes of party systems, creating centrifugal – or polarizing – competitive dynamics (see Sartori 1976: Chapter 6). Spatial models of politics, in fact, generally expect dichotomous or two-bloc competition to have a depolarizing, centripetal dynamic, since it incentivizes parties to compete for the median voter while penalizing ideological radicalism (Downs 1957).

If constitutive and spatial dimensions of polarization are logically independent, so also is polarization independent of populism. Populism may be intrinsically polarizing, but not all forms of polarization are constructed along a populist line of sociopolitical demarcation. Polarization along a class cleavage between workers and capitalists, for example, or between rival ethnic or religious groups cohabiting a national political community, need not be articulated in populist forms – although, to be sure, such societal divisions can provide raw material or sociological bases for a populist articulation of the ‘true’ and authentic ‘people’.

Populism, however, is not a political expression of fixed or predefined sociological categories. It is a constitutive process that constructs a new, unified popular subject out of myriad grievances or claims in a context of social heterogeneity and generalized discontent. Populism thus emerges from a pluralist social landscape containing multiple subjects, fluid identities and cross-cutting interests, but it is a pluralist landscape where diverse societal interests are not meaningfully represented in established institutions. Whereas pluralism is the political expression of social complexity, the political logic of populism is radically different: it condenses that complexity by constructing a binary divide along a new antagonistic frontier between ‘the people’, however conceived, and a power elite or establishment. ‘The people’, then, is plural in its social composition but unitary as a political construct; as Ernesto Laclau (2005: 74) argues, the constitution and ‘symbolic unification’ of this new popular subject require ‘an equivalential articulation’ of disparate anti-establishment demands that makes ‘the emergence of “the people” possible’. Under populism, ‘the people’ subsumes a wide range of subject positions aligned together on the same side of a binary divide, but the content of this cleavage, like the composition of ‘the people’ and ‘the elite’, can take a variety of different forms (see Caiani and Graziano 2019). A populist subject can be spatially positioned on the political left or the right, or even in the centre, since spatial polarization – unlike constitutive polarization – is not essential to populism. The construction of a populist subject is, therefore, a context-specific product of political conflict, mobilization and discourse.

So conceived, populism is a subset of polarized politics, but its constitutive logic is intrinsically polarizing as defined by McCoy et al. (2018: 18). This logic involves a two-step process of constituting a new, unified popular subject out of myriad claims and dividing the political field between this subject and its elite or establishment adversaries. Under populism, pluralism's multiple, cross-cutting interests and identities are condensed and welded together through a logic of equivalence (Laclau 2005: 78), then realigned into rival binary camps on opposite poles of a new axis of political contestation.

Populist mobilization is inevitably a process of cleavage construction, one that is intrinsically polarizing because the constitution of ‘the people’, as Nadia Urbinati (2019: 5) stresses, is simultaneously inclusionary and exclusionary. Populism creates a strong and highly inclusive sense of belonging among ‘the people’, especially given the pre-existing political marginalization of the social groups that comprise this subject. But the elite or establishment ‘other’ across the internal frontier is external to the ‘authentic’ people and in conflict with their just interests. Polarization is intentional, in the sense that populist actors strategically employ polarizing rhetoric and tactics as instruments of political identification, demarcation (or ‘branding’) and constituency mobilization. The polarizing and exclusionary constitutive logic realigns the entire political field in ways that force other issues and conflicts to map onto the central cleavage, while positioning different actors on one side or the other of the political divide. As such, intermediary positions, buffer zones and cross-cutting interests and identities dissipate, dissolve or get relegated to the political margins. This exclusionary antagonism typically assumes ‘a Manichean and moral dimension’, as it ‘paints conflict among groups in black and white, good and evil terms’ (McCoy et al. 2018: 20). Populist narratives stress the betrayal and victimhood of ‘the people’, generating powerful resentments that lend their invocations of popular sovereignty – their pledges to empower ‘the people’ – a highly redemptive character (Pappas 2019: 114). This affective component of polarization – what Noam Gidron et al. (2020: 1) characterize as ‘distrust, dislike, and contempt’ for political opponents – is nearly always reciprocated by anti-populists, intensifying the constitutive divide.

This conceptualization of populism and its polarizing effects helps to explain the phenomenon's political resurgence in recent decades in much of the world. If populism thrives in a pluralist milieu, it follows that it has been structurally enabled by the progressive disarticulation of class-based political subjects and cleavage alignments in post-industrial and neoliberal capitalist societies, and the resulting pluralization of societal interests and political identities (Laclau and Mouffe 2001). In particular, the weakening hold of socialist and social democratic parties over working-class constituencies in Europe and Latin America has created ample opportunities for more populist articulations of political subjectivity, on both the left and right flanks of mainstream party systems (Berman 2016; Roberts 2014). As explained below, these left and right populisms both challenge mainstream party systems that converged in their support of market globalization, but they do so along different, orthogonal axes of political contestation, one economic and the other cultural.

This suggests that structural conditions alone do not tell the full story, as populist mobilization has institutional preconditions as well – namely, failures or crises of political representation (Laclau 2005: 136–7; Roberts 2017; Stavrakakis 2017). These more specific representational failures weigh heavily on the different subtypes of populism, as they shape the grievances or demands that populism can politicize, and thus allow for distinct articulations of ‘the people’ and ‘the power elite’ along opposite poles of a competitive axis. As such, they condition the forms of political realignment and polarization that populism generates in different national and regional contexts.

Spatial polarization and subtypes of populism

Although populism is intrinsically polarizing, it does not necessarily emerge in polarized environments. The rise of Donald Trump as a right-wing populist figure in an increasingly polarized US political system is something of an anomaly, one attributable in part to the institutional particularities of American politics: presidentialism combined with a two-party system, plurality elections and an open primary system that allows insurgent outsiders to challenge established party elites for presidential nominations (see Lee 2020). These institutional features allowed Trump to serve as a political vessel for pre-existing polarizing tendencies at work in American politics (Mason 2018; McAdam and Kloos 2014), rather than their catalyst, and to transform the Republican Party into a vehicle for his populist leadership. In the absence of these institutional traits, Trump's highly polarizing candidacy would have required the formation of a new populist party – the standard recipe for radical right populisms in Europe (Mudde 2017) – or an independent ‘outsider’ bid for presidency, as often seen in Latin America.

More typically, in Europe and Latin America it is a mainstream transpartisan convergence, rather than partisan or ideological polarization, that serves as the political milieu for populist eruptions (Berman and Kundnani 2021; Roberts 2014). This party convergence in the political field is not at odds with the pluralist social milieu that is structurally conducive to populism; indeed, it often reflects it, as a pluralist social landscape allows parties to appeal to fluid and cross-cutting social constituencies, rather than anchoring themselves in deeply entrenched social cleavages with sharply differentiated political identities and ideological positions. Pluralist social landscapes that diffuse social cleavages may thus encourage mainstream parties to converge programmatically and become indistinguishable in their ‘catch-all’ electoral constituencies. In so doing, however, mainstream parties dilute their distinctive partisan ‘brands’ (Lupu 2016), making it easier for aspiring populist challengers to lump them together as part of a transpartisan establishment elite, or a casta política, as it is aptly labelled in Southern Europe. This lumping together – a ‘logic of equivalence’ on the anti-populist side of the binary divide – is integral to the construction of an antagonistic frontier, as it allows ‘the people’ to be constituted among all those who feel alienated from, or unrepresented by, mainstream political parties.

In pure form, an anti-establishment populist subject does not have to induce spatial polarization by adopting ideologically radical positions. Its programmatic positions may not even be markedly different from those of mainstream parties; it polarizes politics constitutively, by drawing a line in the sand to separate insiders from outsiders, and by challenging the political legitimacy or authenticity of those who wield power. Party systems discredited by performance failures – such as severe economic crises or corruption scandals – are especially vulnerable to this type of anti-establishment populism, as they allow populist challengers to promise more competent and effective governance on ‘valence’ issues (such as clean government and a strong economy) that are broadly supported within the electorate, without being ideologically polarizing (Stokes 1963: 373). Although severe crises are not a necessary condition for populism, they may generalize discontent and provide a proximate cause – a detonator, so to speak – for populist mobilizations that have been structurally enabled by underlying processes of social pluralization and mainstream party convergence.

The rise of Italy's Movimento 5 Stelle (M5S – Five Star Movement) after 2009, in the midst of Southern Europe's financial crisis, was a prominent example of this anti-establishment but ideologically eclectic populism. This movement-party, founded by the comedian and political blogger Beppe Grillo, staked out moderate positions on a range of issues that drew eclectically from both left- and right-wing influences (Della Porta et al. 2017; Mosca and Tronconi 2019). As such, the M5S did not polarize Italian politics spatially along an ideological spectrum, but rather along a central, constitutive divide between established parties – with which it refused, initially, to enter into alliances – and ‘the people’, who it claimed to empower by means of digitally enhanced direct democratic participation.

More typically, however, populist alternatives arise not only by challenging a mainstream party establishment, but also by spatially polarizing issue-based competition. In particular, populist challengers politicize issue dimensions that mainstream parties have neglected or declined to compete on – that is, dimensions along which mainstream parties do not adopt meaningfully different stands. Populist challengers, therefore, may break a mainstream transpartisan consensus that does not reflect the full range of preferences found in the electorate, whether latent or activated. Such populisms are both spatially and constitutively polarizing: they challenge the establishment not only on valence grounds, but also on ‘positional’ issues that starkly divide the body politic (Stokes 1963: 373). In so doing, they outflank traditional parties on one pole or another of a competitive axis.

Regrettably, our conventional political lexicon obscures the multidimensional character of spatial polarization under different types of populism. The primary forms of spatial polarization cannot be located on opposite poles of a unidimensional left–right competitive axis. Instead, they are located on economic and cultural axes that are both analytically distinct and spatially orthogonal. As shown in Figure 1, the horizontal economic axis has a statist/redistributive pole on the left and a free market/private property pole on the right. This economic axis is bisected by a vertical cultural axis that has a cosmopolitan/universalist upper pole and a lower pole where ethnic, religious and/or nationalist particularisms are located (variations on these axes can be found in Bornschier 2010; Kriesi 2008; Ostiguy 2017). These orthogonal competitive axes allow left and right subtypes of populism not only to construct ‘the people’ in very different ways, but to do so along alternative axes of contestation that, in fact, render ‘right’ populism something of a misnomer, as explained below.

Figure 1. Economic and Cultural Axes of Competition

For left populism, ‘the people’ is defined in terms of economic disadvantage and political marginality, and is constructed near the left pole on the horizontal axis, where statist measures can be employed for redistributive purposes. What is generally regarded as ‘right populism’, on the other hand, constructs ‘the people’ around particularistic – often nativist – national and cultural identities near the lower pole on the vertical axis, in opposition to cosmopolitan, multicultural and universalist values near the upper pole. This construct, however, does not require spatial location to the right of centre on the horizontal (economic) axis; as explained below, radical ‘right’ populisms in Europe have shifted leftwards over time on the economic axis. They differentiate themselves from mainstream parties primarily through their positioning on the lower pole of the cultural axis, creating a high–low sociocultural and political cleavage that is captured well in Pierre Ostiguy's (2017) conceptualization of populism as the ‘flaunting of the low’. Rather than a radical ‘right’, they are spatially positioned as a radical ‘low’ brand of populism. Spatial polarization, therefore, can occur along either the horizontal or the vertical axis – or, in theory, along both – depending on which type of populism is constructing ‘the people’ and ‘the elite’. And since polarization, like populism, is a continuous rather than a discrete indicator (Caiani and Graziano 2019), populist movements can be located at variable positions along these competitive spectrums, closer to or farther from the polar extremes.

Prominent examples of populist polarization on the left flank occurred in Latin America in the early 2000s, following the region-wide adoption of neoliberal austerity and structural adjustment measures in response to the debt and hyperinflationary crises of the 1980s and 1990s. Although most Latin American countries elected leftist presidents during the first decade of the 21st century, that ‘left turn’ did not always take a populist form, even if Latin American party systems – which are more volatile and less institutionalized than those found in Western Europe – are susceptible to populist mobilization. In countries such as Brazil, Chile and Uruguay, established centre-left parties were able to channel societal opposition to neoliberal reforms adopted by conservative rulers, and after winning elections, they took office through highly institutionalized alternations in power. These parties supported moderate social reforms, but they did not construct polarizing binary divides between ‘the people’ and an elite establishment; instead, they took national office as integral members of the partisan establishment (Levitsky and Roberts 2011; Weyland et al. 2010).

By contrast, populist left turns were more polarizing and institutionally disruptive in Latin America (see Handlin 2017; Roberts 2014). They occurred – in countries such as Venezuela, Bolivia and Ecuador – where established centre-left parties played a major role in the imposition of neoliberal austerity and structural adjustment programmes, causing mainstream party systems to converge on market orthodox positions (i.e. moving towards the right pole in Figure 1). This mainstream partisan convergence channelled opposition to the neoliberal model into extra-systemic forms of mass social protest, leading to the impeachment or resignation of elected presidents in all three countries. Social protest then served as a prelude to mass electoral protest, as voters abandoned traditional parties in droves and threw their support to new, anti-neoliberal, left-populist charismatic figures – Hugo Chávez in Venezuela and Rafael Correa in Ecuador – or the movement-based left-populist party Movimiento al Socialismo (MAS) led by Evo Morales in Bolivia. After winning national elections, these left-populist alternatives upheld their pledges to refound national democratic regimes by plebiscitary acclamation: they employed popular referendums to circumvent established legislative and judicial bodies, convoke constituent assemblies, ratify new constitutions and reconstitute executive and legislative institutions under the majoritarian control of their own partisan vehicles. As the old order crumbled, in each case national polities sharply polarized along a new populist/anti-populist cleavage that reconfigured the traditional left–right axis (Anria 2018; Levitsky and Roberts 2011; Madrid 2009; Weyland et al. 2010). In an index of polarization developed by Samuel Handlin (2017: 28–29), Venezuela, Bolivia and Ecuador ranked as the most polarized party systems in South America, as their scores increased sharply following the rise of left-populist movements in the early 2000s.

Graphical depictions of mainstream party convergence and populist polarization can be seen in Figures 2 and 3. In Venezuela, the Christian Democratic Party (COPEI) was the dominant party of the right in the country's post-1958 democratic regime, while Democratic Action (AD) was the major left-of-centre party (Coppedge 1994). AD, however, was in power when Venezuela imposed neoliberal ‘shock treatment’ in response to the debt crisis in the late 1980s, and the party progressively shifted rightwards, converging on the pro-market stands of the conservative COPEI. As seen in Figure 2, in a survey of deputies in the 2000–5 National Assembly, AD legislators located themselves and their party to the right of centre on a 1–10 ideological scale (5.56 and 5.6, respectively), relatively close to COPEI at 5.83 and 7.0. This right-of-centre ideological convergence opened vacant political space on the left flank for a populist contender – Hugo Chávez – who assailed both the political establishment and neoliberal orthodoxy. With ideological self-placements and party placements of 3.64 and 4.11, respectively, legislators from Chávez's Fifth Republic Movement (MVR) were well to the left of those from AD and COPEI.

Figure 2. Left Populism and Polarization in Venezuela

Source: Survey of Venezuelan legislators, Élites Parliamentarias Latinoamericanas, Estudio #35 (Encuesta a Diputados Venezolanos, 2000–2005), PELA-USAL, University of Salamanca, Instituto de Estudios de Iberoamérica y Portugal, https://oir.org.es/pela/en/.

Note: Scale of 1–10, 1 = most left, 10 = most right.

Figure 3. Left Populism and Polarization in Bolivia

Source: Survey of Bolivian legislators, Élites Parliamentarias Latinoamericanas, Estudio #9 and #47 (Encuesta a Diputados Bolivianos, 1997–2002 and 2002–2007). PELA-USAL, University of Salamanca, Instituto de Estudios de Iberoamérica y Portugal, https://oir.org.es/pela/en/.

Note: Scale of 1–10, 1 = most left, 10 = most right.

Similarly, in Bolivia neoliberal shock treatment was imposed in 1985 in the midst of a hyperinflationary crisis by the party which led Bolivia's 1952 revolution, the Revolutionary Nationalist Movement (MNR). Neoliberal reforms continued in the 1990s under the leadership of another party with leftist roots, the Movement of the Revolutionary Left (MIR), as well as Bolivia's main conservative party, Nationalist Democratic Action (ADN). As shown in Figure 3, deputies from ADN and MNR located themselves and their parties on the right in legislative surveys, while those in the MIR identified slightly left of centre. The convergence of these mainstream parties around neoliberal orthodoxy, however, clearly vacated political space on the left flank for a party with roots in Bolivia's formidable anti-neoliberal protest movements. The MAS of Evo Morales, formed by the coca-growers’ peasant union (Anria 2018; Madrid 2008) forged close ties to protest movements and articulated a plethora of grievances against the neoliberal model, locating itself well to the left of the three mainstream parties (with deputies’ self-placements and party placements at 1.57 and 2.14, respectively).

The Latin American experience found parallels in Southern Europe as part of the political fallout from that region's post-2009 financial crisis. In Greece, Spain and Portugal mainstream centre-left parties played major roles in the adoption of neoliberal austerity and adjustment measures during the early stages of the crisis. Party systems, therefore, largely converged on orthodox responses to the crisis under the pressure of bond markets and European financial institutions. As in Latin America, this neoliberal convergence set the stage for mass social protest and anti-establishment electoral protest by citizens opposed to market orthodoxy. Waves of anti-austerity and anti-elite social protest – the so-called indignados movements – swept across the region in 2011, classic examples of what Paris Aslanidis has characterized as ‘bottom-up populism’. Following a populist logic, these movements claimed to ‘represent a social whole rather than the interests of particular strata’, and they sought ‘wholesale reform of the political regime to restore the sovereignty of the people’ (Aslanidis 2017: 305, 307). Mass social protest, then, was a prelude to the weakening of mainstream parties and the electoral ascendance of movement-based parties on the left flank of the party systems in Greece (Syriza) and Spain (Podemos), along with the formation of a new coalition between centre left and radical left parties in Portugal (see della Porta 2015; Portos 2020; Stavrakakis and Katsambekis 2014).

Like the M5S in Italy, the rise of Syriza and Podemos injected a new populist challenger into national party systems during a period of crisis, one that claimed to represent the sovereign people in confrontation with an unresponsive political elite.¹ In contrast to the M5S, however – and more like the populist left in Latin America – Syriza and Podemos also realigned their national party systems programmatically by outflanking mainstream parties on the left (even if Podemos tried to avoid the leftist label, considering it a remnant of class-based politics that did not fully encompass the social breadth and diversity of ‘the people’). In short, through their strident critiques of austerity measures, neoliberalism and economic inequality, along with their staunch defence of social safety nets, Syriza and Podemos repoliticized the left pole on an economic axis largely vacated by the rightward drift of traditional socialist parties towards pro-market positions. The spatial polarizing effect of left populist movements and parties, then, lay in their ability to break the mainstream partisan consensus about neoliberal orthodoxy, broaden the range of programmatic contestation, and channel societal dissent that was previously on the margins of the democratic order.

This rise of left populism followed a political logic similar to that laid out in Simon Hug's (2001) work on the formation of new party organizations around novel issue claims or positions. As Herbert Kitschelt (2017: 362) states, ‘New parties must assume issue leadership to crystallize a potential electorate and mobilize it around a new cluster of political demands.’ Although this can be an elite-driven, supply-side process in which populist appeals politicize new issues, sway public opinion or activate latent cleavages, the left populist experience in Latin America and Southern Europe strongly suggests that dynamics on the mass politics/demand-side are critical in certain contexts. In both regions, mass social protest politicized novel issue positions and demonstrated societal demand for political changes that became crystallized in new populist alternatives. These alternatives carried myriad grievances articulated by protest movements into the electoral arena, unifying them through shared opposition to austerity, neoliberalism and the casta política. In so doing, they forced mainstream parties to defend and compete on policy positions that were previously taken as a fait accompli – that is, as the only ‘responsible’ course of action in contexts of acute market constraints and transnational political pressures (Mair 2013).

The rise of the populist radical right (PRR) in Europe also followed Hug's (2001) logic, but increasingly these parties do not politicize the right (pro-market) pole on the same economic axis where left populists compete. Instead, in both Western and Eastern Europe, they politicize the lower, nativist or ethno-nationalist pole on the cultural axis that mainstream parties largely abandoned during an era of globalized market liberalism (see Berman and Kundnani 2021; De Cleen 2017). Although Western Europe's PRR parties initially adopted pro-market, anti-tax platforms in the 1980s and 1990s (Kitschelt and McGann 1995), these neoliberal platforms were more of a ‘political weapon’ than an ‘economic programme’, as they buttressed the parties’ visceral hostility toward ‘established political institutions’ (Betz 2017: 342). Many of these parties gravitated over time away from market fundamentalism in order to support more nationalistic and protectionist economic policies, thus reinforcing their cultural nativism on the lower pole of the vertical axis (and accounting for the arrow pointing leftwards in the radical right populist oval in Figure 1). As such, they became defenders of ‘welfare chauvinist’ positions that made the defence of generous national welfare states a political rationale for denying citizenship rights to ‘undeserving’ immigrants, as well as a plank to attract support from working-class voters threatened by the insecurities of market globalization (de Lange 2017).

The PRR in post-communist Europe followed a similar play script, in contexts where mainstream parties had also converged around support for European integration and market liberalization (Binev 2018; Snegovaya 2018). PRR parties thus articulated a range of positions – such as hostility towards immigration and ethnic or religious minorities, opposition to transnational European economic and political institutions, support for trade protectionism, etc. – that lay outside the mainstream partisan consensus, but they exerted stronger polarizing effects downwards on the cultural axis than rightwards on the economic axis. Their polar opposite, therefore, is not the radical (statist) left, but the cosmopolitan cultural ‘high’ in Figure 1 – that is, parties that defend more culturally liberal, cosmopolitan and multicultural visions of the political community (Ostiguy 2017).

In short, the gravitational pull of market globalization in the late 20th and early 21st centuries induced mainstream parties to shift upwards and rightwards in the two-dimensional space in Figure 1. Their convergence in a narrowed programmatic space dealigned the class-based left–right cleavage structures that Seymour Martin Lipset and Stein Rokkan (1967; see also Bartolini and Mair 1990) famously credited with ‘freezing’ European party systems during the industrial era. Mainstream convergence largely vacated the lower left quadrant in Figure 1, traditionally a political space occupied by organized labour and its social democratic allies. It also opened programmatic space for new populist contenders to compete for votes by politicizing economic inequality and precarity (on the left pole) or ethno-nationalist identities (on the lower pole). The latter, in particular, drew many working-class voters to PRR parties as they backed away from market orthodoxy and embraced protectionist and welfare chauvinist positions (Harteveld 2016). In short, both left- and right-wing populist challengers cleaved the political field between ‘the people’ and the partisan establishment, and they polarized it spatially by expanding the range of programmatic contestation. This contestation, however, occurred along distinct, orthogonal axes of political differentiation.

This helps to explain why Matthijs Rooduijn and Tjitske Akkerman (2017) find a strong association between radicalism – on the right or the left – and populist discourse in their study of party manifestos in five West European countries. As they state, ‘whether a party is left or right does not affect its degree of populism; it is its radicalness that determines how populist the party is … both radical right parties and radical left parties are inclined to employ a populist discourse’ (Rooduijn and Akkerman 2017: 193). Parties are radical, or spatially polarizing, because they adopt ideological and programmatic stands that lie outside the mainstream party consensus; they are populist because they assail that political establishment in the name of ‘the people’. Indeed, their radicalness validates their outsider status – that is, their non-membership in the political establishment. It follows, then, that in Rooduijn and Akkerman's (2017: 198) measure of populist discourse, parties of the radical left (7.06) and radical right (10.33) score much higher than mainstream parties (0.65). Likewise, in their research on parties’ media press releases, Laurent Bernhard and Hanspeter Kriesi (2019) find populist appeals to be concentrated on the radical left and the radical right, with the latter politicizing cultural issues, and both types of radical parties politicizing economic issues in a populist manner.

Populism, polarization and the rise of Donald Trump: is the US an anomalous case?

If a two-dimensional conceptualization of political space helps identify the political logics of left- and right-wing populisms, so also does it help to explain the anomalies of the contemporary US case, which are two-fold. First, although right populist leader Donald Trump was surely a highly polarizing figure, he emerged in a context of serious partisan and ideological polarization, rather than a context of mainstream convergence as in the European and Latin American cases examined above. In that sense, Trump served as more of a political vessel and force multiplier for polarizing tendencies rather than their political catalyst. Second, although Trump was a consummate political outsider who stridently opposed both major US party establishments, he was able to capture one of the traditional parties and transform it into an instrument of his insurgent populist project. In contrast to populist counterparts in Europe and Latin America, therefore, Trump did not have to found a new political party or run an independent campaign to access executive office; he was able to use the machinery of the GOP to win a presidential election, exercise power and insulate much of his norm-breaking behaviour from institutional oversight and accountability. Trump's routine flouting of democratic norms, and the GOP's craven willingness to aid and abet it, intensified political polarization in the US, making it abundantly clear that polarization was no longer contained at the level of stark ideological or policy differences. It had, instead, metastasized to infect the very democratic institutions set up to process, regulate and ameliorate conflicts of interest or values.

Even before Donald Trump left reality television to enter the political fray, scholars and pundits were decrying the steady increase in political polarization in the US, a generation-long process of sociopolitical mobilization and counter-mobilization that drove a deeper wedge between the Democratic and Republican parties and their liberal and conservative constituencies, respectively (Abramowitz 2011; McAdam and Kloos 2014; McCarty et al. 2016). Polarization moved the parties further apart in their ideological worldviews and policy platforms, made them more internally cohesive and sorted their electorates into mutually antagonistic partisan camps divided along racial, ethnic, religious and geographic lines (Grossman and Hopkins 2016; Lee 2016; Mason 2018). This polarization, however, has conventionally been understood by scholars of US politics as a unidimensional left–right (or liberal–conservative, in American political parlance) phenomenon, given the proclivity of the parties – the GOP in particular – to bundle social, economic and cultural issues together in their advocacy of conservative (or liberal) orthodoxy. The rise of Trump as a populist figure not only disrupted these orthodoxies but drew back the curtain to reveal them for what they really are: political constructs that are contingent in nature and subject to contestation and realignment.

A two-dimensional conceptualization of political space provides a very different framework for understanding polarization in the US and its relationship to radical right – and surely exceedingly ‘low’ – populism. On the horizontal economic axis in Figure 1, it is highly debatable whether polarization was underway in US politics prior to the rise of Bernie Sanders on the left flank of the Democratic Party. Viewed from a comparative perspective, the pre-Sanders Democratic Party was not a party of the left, and its centrist positioning in the Clinton era led the party to adopt a series of pro-market reforms, from financial deregulation to free trade accords and cutbacks in social welfare programmes. Spatial polarization on the economic axis, therefore, was largely driven by the GOP's steady drift towards increasingly fundamentalist positions on the right (pro-market) pole, a process more appropriately characterized as unilateral radicalization rather than bilateral polarization. As Matt Grossman and David Hopkins (2016: 108) argue in their study of polarization in the US, even before the rise of Trump the GOP adopted ‘the most consistently conservative positions of any political party in the world’. Republican radicalism aside, bipartisan pro-market convergence, rather than interparty polarization, was the dominant pattern in several major spheres of economic policymaking, including international trade and financial liberalization.

Polarizing dynamics, however, were more clearly present on the vertical cultural axis, the site of the so-called ‘culture wars’ in US politics, including the ‘formative rift’ (McCoy and Somer 2019) of racial inequality that has plagued American democracy since its founding (Hajnal 2020). The political cleavage over race did not structure interparty competition in the post-Reconstruction era, as it cut across the two main parties – dividing them internally – rather than separating them into opposing camps. This changed with the civil rights movement of the 1960s, the passage of the Civil Rights Act in 1964 and the conservative counter-mobilization against them. With the Democratic Party taking the lead in desegregation, racial animosities drove the socially conservative South towards the GOP, and the party system was realigned to group economic and cultural liberals in the Democratic Party, and their conservative counterparts in the GOP (Schickler 2016). This cleavage was subsequently reinforced and deepened by other cycles of sociopolitical mobilization and counter-mobilization: anti-war, women's rights, gay rights, immigrants’ rights and environmental movements pulled the Democratic Party towards the ‘high’ pole on the vertical axis in Figure 1, while the evangelical, anti-abortion, anti-busing, anti-gay rights and gun rights movements pulled the GOP towards the lower pole (Schlozman 2015). The anti-Obama Tea Party movement deepened this polarization on both axes of contestation, as it pushed the GOP towards more extreme forms of market fundamentalism in the midst of the Great Recession, and also gave voice to strengthening anti-immigrant and white identity currents at the grassroots of the GOP (Blum 2020; Parker and Barreto 2013; Skocpol and Williamson 2013). Social mobilization on the right flank of the GOP demonstrated, once again, that activist networks can generate demand-side pressures for partisan polarization (Abramowitz 2011), reinforcing the inflammatory effects of racially charged supply-side rhetoric from politicians like Trump (Schaffner 2020).

The Tea Party movement was also staunchly critical of the Republican Party establishment, setting the stage for Trump's presidential campaign as a populist outsider who attacked party elites on both sides of the divide while competing within the Republican primaries. Trump's polarizing effects, however, were not limited to his attacks on the political establishment. His anti-immigrant rhetoric and policy stands sharply politicized the lower pole on the cultural axis of contestation, pushing the GOP towards increasingly nativist or ethno-nationalist positions. Ironically, this was the exact opposite of the political strategy charted by the party establishment in the aftermath of Mitt Romney's defeat by Barack Obama in 2012. In its so-called ‘autopsy’ of the electoral defeat, the project report of the Republican National Committee called for the party to recognize that ‘America is changing demographically, and unless Republicans are able to grow our appeal … the changes tilt the playing field even more in the Democratic direction’. Claiming that the party sounded ‘increasingly out of touch’, the report said ‘many minorities wrongly think that Republicans do not like them or want them in the country’. The party's future electoral viability thus required a ‘more welcoming conservatism’, one that would ‘embrace and champion comprehensive immigration reform’ (Republican National Committee 2013: 4–8).

Trump's populist gambit, however, pushed in the opposite direction, appealing explicitly to anti-immigrant and white ethno-nationalist currents at the base of the Republican Party. In announcing his candidacy for the presidency, Trump declared that:

When Mexico sends its people, they're not sending their best. They're not sending you … They're sending people that have lots of problems, and they're bringing those problems with us. They're bringing drugs. They're bringing crime. They're rapists … I would build a great wall, and nobody builds walls better than me, believe me, and I'll build them very inexpensively, I will build a great, great wall on our southern border. And I will have Mexico pay for that wall. (CBS News 2015)

Indeed, Trump politicized the lower pole on the cultural axis in myriad ways, both as a candidate and in the White House. He advocated a border wall and other harsh measures to block immigrants from entering the US and went so far as to separate immigrant children from their parents at the border and place them in detention camps. He also embraced the evangelical community and its policy platform; declined to disavow white nationalist currents inspired by his political project; and made ‘America First’ and ‘Make America Great Again’ cornerstones of his redemptive national-populist appeal. This appeal had a pronounced isolationist streak that marked a sharp break with the post-World War II bipartisan foreign policy consensus in support of multilateralism. Trump loathed international institutions, alliances and treaty obligations; he criticized NATO and US allies, embraced foreign autocrats, withdrew from international climate change accords and the World Health Organization, and heaped scorn on international trade accords.

Indeed, Trump's disdain for free trade and economic globalization – cornerstones of a market economy long supported by both major US parties – revealed the contingent character of the multidimensional political construct associated with conservative orthodoxy in the US. By doubling down on the lower, ethno-nationalist pole on the cultural axis, Trump – like the populist radical right parties in Europe – at least partially decoupled it from the right pole on the economic axis, demonstrating the potential orthogonality of the two-dimensional space. In short, Trump showed that the bundling together of Christian and white ethno-nationalist identities with market fundamentalism is neither necessary nor inevitable. He also buttressed the long-term trend that has shifted white working-class votes from the Democratic to the Republican Party since the 1990s (Carnes and Lupu 2020: 6). As in Europe, Trump could appeal to working-class voters in the lower-left quadrant of Figure 1 by relaxing the market fundamentalism of Republican orthodoxy and adopting more nationalist and protectionist stands, thus taking advantage of the Democratic Party's embrace of free trade and its shift towards the ‘high’ pole on the cultural axis. Whether Trump's economic policies in office actually served these working class voters’ interests is another question; since protectionist stands were packaged alongside other, more orthodox Republican positions such as massive tax cuts and deregulatory policies, Trump catered to the party's traditional business constituencies as well, lending his populism a markedly transactional and plutocratic air (Pierson 2017).

Given Trump's ability to channel populist currents at the grassroots of the GOP, the US primary system facilitated his takeover of the party organization. It allowed him to defeat the party establishment in the 2016 primary elections, then use the threat of primary challenges against incumbent Republican officeholders to discipline the party and enforce personal fealty to his presidential leadership. In so doing, Trump completed a political transition that had been underway for the better part of a political generation: the transformation of a mainstream conservative party – a pillar of the US political and economic elite for well over a century – into a populist radical right party such as those more widely recognized in Europe (Mudde 2017).

Institutional polarization and democracy: taming effects v. inertial properties

How does a focus on populism's polarizing logic help us understand its impact on democracy as a system of governance? Can democracy transform populism, or ‘tame’ its polarizing effects? Populist polarization would appear to be a two-edged sword for democracy. On one hand, it clearly has the potential to broaden and invigorate democratic representation, giving expression to the identities and concerns of citizens who previously felt neglected by or disconnected from mainstream party elites. That is especially the case where mainstream parties have converged in programmatic space and ceased to offer meaningful programmatic alternatives that reflect the full range of preferences found in the political community. Populist challengers can place competitive pressures on mainstream parties to address issues and concerns that were previously left off the political agenda, whether those are issues related to economic inequality on the horizontal axis or immigration policies on the cultural axis.

On its own, spatial polarization may be healthy for democracy by expanding and aligning representation on the issues and, in the process, making party systems more responsive to the full range of preferences in society. The field of comparative politics has long seen conflict and polarization as the structural foundations on which stable party systems are built, as they allow parties to sink deep roots in social cleavages that anchor party organizations, ‘sort’ the electorate and align electoral competition (Bartolini and Mair 1990; Levitsky et al. 2016; Lipset and Rokkan 1967). Indeed, theorists of democratization often see conflict and polarization – not a liberal normative consensus – as the breeding ground for democracy itself (Rustow 1970), with the latter conceived as a cluster of rules designed to mitigate sociopolitical conflict by safeguarding political rights and institutionalizing competition, constraints on power and alternation in public office. Examples can certainly be found where democratic institutions have tempered conflicts, restrained populist actors and induced them to moderate their positions. That can be seen in Western Europe when populists have made a transition from the electoral arena, where polarization is a political weapon that can enhance voter mobilization, to the policymaking arena, where populist parties in multiparty coalitions are typically forced to compromise in order to govern. For that reason, Michael Minkenberg (2017: 443) speaks of the ‘taming effect’ of governing responsibilities on radical right populist parties in Western Europe. A similar story can surely be told about Syriza in Greece, which tempered its left radicalism considerably after winning national elections in 2015 and assuming governmental responsibilities in the midst of a severe financial crisis. The Greek case cautions against facile assumptions that populist movements in power are necessarily bound to implement ‘irresponsible’ policy measures or to clash with liberal democracy's pluralistic norms and institutionalized checks and balances.

Nevertheless, democracy's taming effects are not automatic, and its built-in institutional constraints do not always contain populism's polarizing dynamics. The other side of populism's two-edged sword is the possibility that polarization may acquire an inertial, self-reinforcing quality that makes it ‘pernicious’, in the terminology of Jennifer McCoy and Murat Somer (2019), and corrosive for democracy itself. This is especially likely not only where populist actors embrace policy stands that lie outside the mainstream – a form of spatial polarization that democratic institutions are designed to process and assuage – but also where they are polarizing because they harbour illiberal tendencies that lead them to deny the legitimacy of rival actors and contest essential democratic norms or procedures in their pursuit of political power. Institutional polarization, therefore, entails frontal conflict over the basic rules of the political game – rules that were designed and evolved, in large part, to process and manage political conflict itself.

Democratic checks and balances are designed to limit the damage that illiberal actors can do so long as their electoral appeal consigns them to a minority status, but liberal democracy has no ready-made solution to the challenge of restraining illiberal actors who have sufficient electoral strength and political cohesion to capture state powers and coordinate across branches of government.² Simply put, every institutional site designed, on paper, to be an institutional check on abusive or unwarranted concentrations of power is also, potentially, an institutional lever that can be wielded as an instrument of partisan advantage by an illiberal actor. Such instruments can be used to tilt the democratic playing field to the advantage of incumbents by stacking the courts, rewriting or reinterpreting laws and constitutions, manipulating electoral institutions, intimidating the media and civil society, neutralizing watchdog agencies or employing them as political weapons to undercut opponents. Populist governments on both the left (Venezuela) and the right (Hungary) have resorted to these kinds of tactics to consolidate authority and undermine opponents (see Grzymala-Busse 2019; Levitsky and Loxton 2013), using democratic levers to transform or refound regimes.

Such forms of institutional weaponization magnify polarization far beyond that which policy differences alone can produce. They dramatically raise the stakes of democratic contestation by challenging its iterative character: where temporary electoral or institutional advantages are used to tilt the playing field or ‘stack the deck’ to the advantage of incumbents, opponents are hard-pressed to compete in future iterations of the democratic process, and temporary imbalances can be locked in indefinitely. Stripped of their temporal boundaries and iterative character, democratic contests become existential in character, and the ‘contingent consent’ that induces losers to accept electoral and policy defeats on the condition that they can compete to reverse them in future iterations is shattered (Schmitter and Karl 1991: 82). This is when polarization becomes truly inertial and pernicious for democracy, metastasizing from the policy domain to a wide range of institutional and civic sites that are otherwise at least partially insulated from partisan contestation.

Under this type of pernicious polarization, the notion that there might exist institutional ‘guardrails’ that protect the regime itself, or a non-partisan professional civil service that belongs to the state and serves the larger public interest, becomes seriously contested. Such institutional sites may be treated by populist figures as constraints on ‘the will of the people’, as ‘deep state’ sediments that preserve the ‘old order’ or as levers to be captured and weaponized against opponents. What made the US case under Trump so chilling was the extent to which these institutional guardrails and public institutions – from the federal judiciary to the Justice Department, the Federal Bureau of Investigation, the State Department, congressional oversight and investigative commissions, inspector generals in a wide range of agencies, the electoral machinery and even the Census Bureau and Postal Service – became so thoroughly infused with partisan politicization. That process reached extreme lengths under Trump's populist leadership, but it began well before his election to the presidency and, crucially, the Republican Party clearly provided political cover for Trump as he accelerated the pace. If threats to democracy had been limited to Trump's personal whims and autocratic impulses, the highly developed checks and balances of US democratic institutions could have safeguarded the regime more effectively; to be operative, however, those checks and balances require a governing party that places regime loyalty and democratic principles ahead of partisan self-interest and policy or ideological commitments. Otherwise, they can be readily neutralized by a ruling party that opportunistically transforms the levers it controls into instruments of partisan advantage, or shields for presidential impunity.

Conclusion

Populism is intrinsically polarizing, as it is predicated on the binary division of political space between ‘the people’ and an elite, a ‘we’ versus a ‘them’. Polarization can take a number of different forms, however, dividing political space along diverse antagonistic frontiers, and it does not have uniform effects. In particular, the effects of polarization may be quite different for a populism in opposition and a populism in power. Where mainstream parties have converged programmatically and ceased to offer voters meaningful alternatives on issues of public concern, populist challengers may expand democratic representation, bringing new voices to the political table and placing new issues on the political agenda. That issue expansion can be polarizing and contentious, but democratic competition is designed to institutionalize such conflicts and force policy compromises, potentially taming populist challengers. But polarization can also assume pernicious forms, when rival actors differ not only on policy responses to the issues of the day, but diverge sharply in their commitments to the very rules of the democratic game itself. Where populists gain access to power and use the institutional levers they control as instruments of partisan advantage, tilting the democratic playing field, polarization can take on inertial qualities that imperil democracy itself.

This autocratic temptation is hardly exclusive to populism, and different populisms may not be equally susceptible to it. Populism challenges liberal democracy, but does so in the name of democracy itself, alternatively conceived as an unbridled expression of popular sovereignty. Its moralistic fervour, binary worldview and redemptive imagery arguably incline it towards hegemonic constructions of popular sovereignty that are silent or ambivalent with respect to minority political rights. Populist leaders and movements rarely articulate what happens to ‘them’ – the elite or establishment ‘Other’ and the social groups they represent – if and when they are defeated politically by the sovereign ‘people’. Do those who have been vilified and demonized on the other side of the antagonistic frontier – the ‘anti-people’, by default – retain political rights once removed from power? Are they allowed to regroup and contest democratic outcomes again, on a playing field that is level and fair? In Gramscian terms, where, precisely, is the boundary line between hegemony – a consensual form of domination exercised through cultural and ideological influences – and domination exercised by means of coercive exclusion? Polarization taken to its logical extreme is ill-equipped to raise, much less answer, these questions. But populist actors can ill afford to ignore them if they seek to empower ‘the people’ without curtailing their rights.