Abstract
From Hugo Chávez in Venezuela to Geert Wilders in the Netherlands and Sarah Palin in the US, populist leaders claim to offer more power to ‘the people’. However, most scholars argue that populism is in fact a democratic pathology, because it seeks to build a political system devoid of the rule of law. While it is true that populism maintains an ambivalent relationship with liberal democracy, little attention has been paid to the legitimacy of the questions raised by populist forces. Drawing on the work of Robert Dahl, I argue that current manifestations of populism are offering specific responses to two dilemmas that do not have a clear democratic solution: the boundary problem (how to define the people?) and the limits of self-government (how to control the controllers?). My article shows that populist forces are posing legitimate questions about the current state of democracy in Europe and the Americas, although their solutions tend to be more controversial than helpful.
Populism seems to be a pervasive phenomenon in the contemporary world. From North to South America and from Eastern to Western Europe populist leaders, movements and parties have become increasingly influential in the last decade. Although they vary in their definitions of populism, many authors maintain that populism is first and foremost a democratic disease or pathology (e.g. Abts and Rummens, 2007; Alonso et al., 2011; Rosanvallon, 2008; Taggart, 2002; Urbinati, 1998). In this view, tainted though it is by medical metaphors, populism is damaging for democracy: it is a syndrome that permits the rise of chauvinistic and fundamentalist forces seeking to dismantle checks and balances, jettison the rule of law and establish a new regime where political power relies on the unified will of the people rather than on representatives and unelected bodies.
While it is true that populism has an ambivalent relationship with liberal democracy, depicting the former as ‘bad’ and the latter as ‘good’ produces more problems than it resolves. This way of thinking overlooks the fact that the very institutional design of the liberal democratic model is anything but perfect, particularly in terms of maintaining an equilibrium among the so-called checks and balances, as well as between them and the power of the demos (Armony and Schamis, 2005). Behind this moral language used by many scholars is the tacit assumption that populism has nothing substantial to offer to democratic theory and praxis. In contrast, in this article I argue that populism raises issues that are entirely legitimate and, in consequence, we should avoid treating populism as an irrational impulse. In consonance with the work of Benjamin Arditi (2004; 2005) and Margaret Canovan (1999; 2004; 2005), I maintain that populism deserves more intellectual attention, because it poses fundamental questions that are difficult, and to a certain extent impossible, to solve in a democracy. That said, my article does not intend to offer a romanticised view of populism, but rather tries to develop an original approach regarding how to think about populism vis-à-vis democracy.
This is a theory-driven article, which develops a novel framework for studying the ambivalent relationship between populism and democracy, and which offers some empirical illustrations to highlight the usefulness of the proposed framework. The article is structured in three sections. In the first part, I examine Robert Dahl's work with the aim of showing the existence of two dilemmas that do not have a clear democratic solution: the boundary problem (how to define the people?) and the limits of self-government (how to control the controllers?). The second part of the article is focused on the solutions that contemporary populism is prone to offer when it comes to dealing with the two democratic dilemmas sketched above, and it will illustrate some of the differences and similarities between the predominant manifestations of populism that Europe, Latin America and the US are experiencing today. Finally, I conclude by providing some suggestions for rethinking the relationship between populism and democracy.
Dahl's Democratic Dilemmas
Without a doubt, Robert Dahl is one of the most important political thinkers of the twentieth century. Unlike many other authors, his work has been influential for both empiricist and theoretical researchers. Dahl's notion of ‘polyarchy’ is probably the most commonly used concept for measuring and operationalising democracy. His list of institutional guarantees that a regime has to comply with in order to be considered democratic is the minimal core of almost all empirical approaches. Moreover, his work is characterised by historical and heuristic reflections about not only the nature but also the normative grounds of democracy. As Michael Saward (2003, p. 48) points out, Dahl is particularly interesting ‘because he both absorbs and extends the dominant Schumpeterian narrative, on the one hand, and in more normative mode is one of the most trenchant critics of the limits of American democracy, on the other’.
In sum, at the heart of the Dahlian perspective is the very idea that both the empirical and normative aspects of democracy should be combined in a single theoretical approach. How can this be accomplished? Dahl's answer to this question is that we have to be aware of the dual meaning of democracy: it refers not only to an ideal, but also to actual regimes (i.e. polyarchies) that fall considerably short of the ideal. For instance, although no one really doubts that contemporary Spain is a democratic country, the indignados have good reasons for claiming that the political system is not functioning properly and that the principle of self-government is not being met. After all, ‘the invisible hand of the market’ and supranational institutions are imposing reforms at odds with the will of the Spanish people. As this example reveals, the concept of democracy can be used to evaluate whether a regime is democratic or not, and to criticise existing democracies for their difficulties, and sometimes inability, in guaranteeing substantive outcomes.
According to Dahl, the double meaning of democracy is often confusing, but also indicative of a productive tension between empirical and normative approaches. Whereas the former seek to measure whether and to what degree a country is democratic, the latter aim to show that democracy is justified as a political system that is responsive to all its citizens. Given that the gap between existing democracies and the democratic ideal cannot be avoided, a continuous process of critical assessment and monitoring takes place, whereby injustices are denounced, institutional reforms are promoted and social movements are legitimised (Norris, 2011). While this dual understanding of democracy has been adopted by many authors, who agree that democracy should be conceived of as a dynamic and open-ended process that always remains incomplete (e.g. Dunn, 2005; Tilly, 2007; Whitehead, 2002), less academic attention has been paid to Dahl's work on certain assumptions made by democracy advocates. However, these ‘half-hidden premises, unexplored assumptions, and unacknowledged antecedents form a vaguely perceived shadow theory that forever dogs the footsteps of explicit, public theories of democracy’ (Dahl, 1989, p. 3).
In the next sections I present two of these assumptions, which I call ‘democratic dilemmas’ for a simple reason: these are problems that do not have a clear democratic solution. In other words, the dilemmas that I explore denote situations in which a difficult choice has to be made and there is no way of satisfying the democratic process, that is, the principle of ‘rule by the people’. These two dilemmas are at the heart of Dahl's democratic theory, and are crucial for understanding his critique not only of other approaches (e.g. Schumpeter's work) but also of the imperfections of modern-day democracies (e.g. the US model).
How to Define the People? The Boundary Problem
The first dilemma identified by Dahl is the so-called boundary problem (Whelan, 1983), which alludes to a basic question: what constitutes ‘a people’ for the purpose of democratic government? Neither empirical nor normative theories tend to deal with this question, because they normally assume that ‘a people’ already exists (Dahl, 1970, pp. 45–79). The reason for this is that we normally take for granted that those who are living within the (nation) state have the right of self-determination. Although this seems to be a plausible answer, it is bound to be paradoxical. People cannot be defined by state boundaries because the former constitute the supreme political authority in charge of creating the institutions of self-government. In other words, the notion of popular sovereignty that is at the centre of modern democratic theory assumes that the people authorise the establishment of the state, and not the other way round (Yack, 2001, p. 523). Certainly this leads to a vicious circle or an infinite regress, since individuals cannot unify their wills through reason to constitute ‘a people’ because any such unification presupposes an already existing people (Ochoa Espejo, 2011, p. 43). In fact, collective self-determination has proven to be a dangerous principle, whereby a supposedly existing peoplehood has undemocratically imposed the formation of a (nation) state on larger populations, which are then assimilated into the dominant identity promoted by the (nation) state, or simply forced out and murdered (Mann, 2005).
Unlike most political thinkers of the twentieth century, Dahl is aware of this problem, one which in his opinion cannot be solved democratically. Arbitrary outcomes such as invasions, revolutions and wars are the key mechanisms for defining the people who have the right to participate in the collective decision-making process. In his own words: ‘In the real world, then, answers to the question, what constitutes “a people” for democratic purposes? are far more likely to come from political action and conflict, which will often be accompanied by violence and coercion, than from reasoned inferences from democratic principles and practices’ (Dahl, 1989, p. 209). To be sure, this is a disturbing proposition for all those who (like Dahl himself) believe in the superiority of democracy over other forms of government, because it reveals that in democratic regimes the people is always born out of a combination of coercive force and persuasive storytelling, rather than of a rational discussion leading to the formation of a voluntary association (Näsström, 2007, p. 629).
To cope with this ‘dark side of democracy’ (Mann, 2005), or better said, to develop a satisfactory argument about who ought to be a member of the demos, Dahl has formulated the so-called all-subjected principle. This is based on the idea of political equality and states that ‘[t]he demos must include all adult members of the association except transients and persons proved to be mentally defective’ (Dahl, 1989, p. 129). While it is true that defining transients and those suffering from a severe mental disability is not straightforward, Dahl's categorical principle seeks to develop a criterion of inclusiveness that is clear enough ‘to distinguish between a political system that is democratic in relation to its own demos and one that is democratic in relation to everyone subject to its rules’ (Dahl, 1989, p. 130). In this sense, the argument developed by Dahl opposes the view of Joseph Schumpeter (1942, pp. 243–5), who maintained that it is up to the people to define who is entitled to participate in the democratic process. Nonetheless, the determination of the criteria of the electoral body is a matter that is logically prior to the operation of the majority principle, and cannot be solved by it (Whelan, 1983, p. 15). In the words of Ivor Jennings (1956, p. 56), ‘[t]he people cannot decide until somebody decides who are the people’.
With the exception of those who are sympathetic to the all-affected principle advanced by cosmopolitanism, Dahl's all-subjected principle is the starting point of almost all democratic theories. This means that ‘a people’ is taken as given and is constituted by all those living within the territory of the (nation) state. Because they are subjected to the law, they are the members of the political community and have the right to participate in the democratic process. As we have seen before, this solution is inconsistent with the very idea of popular sovereignty and, in consequence, it falls into the infinite regress trap. The gist of modern democracy is that political power comes from the demos, implying that the people is the only authority that has the right to enact laws to which it would be subjected. Yet the all-subjected principle can be challenged not only in theoretical, but also in normative and empirical terms.
First, from a normative point of view, assuming that the state boundaries delimit who is entitled to participate in the collective decision-making process is anything but a neutral argument (Yack, 2001, p. 529). As Sofia Näsström (2011a, p. 128) indicates, ‘[s]ubstituting the task of justification with the force of history and tradition it indirectly sustains a condition of status quo. It affirms the boundaries of the present’. The situation of diaspora communities and foreigners is a paradigmatic example of this, since they are usually excluded from the democratic process because they do not reside in their own states (e.g. Chilean exiles during Pinochet's dictatorship who are still living abroad), and/or because they do not possess citizenship of the state in which they actually live (e.g. children born to Gastarbeiter coming from Turkey and living in Germany). Moreover, history tells us that defining state borders is usually an arbitrary and bloody enterprise. Not by coincidence, Charles Tilly (1993) has shown that the modern history of Europe is characterised by the double process of war making and state making.
Second, from an empirical point of view, the plausibility of the all-subjected principle turns out to be unsatisfactory when we consider the current state of democracy, particularly after the third wave of democratisation. In their famous book, Problems of Democratic Transition and Consolidation, Juan Linz and Alfred Stepan (1996) maintain that democracy cannot endure in places facing serious problems of stateness. In their own words, ‘Democracy is a form of governance of a modern state. Thus, without a state, no modern democracy is possible’ (Linz and Stepan, 1996, p. 17). Interestingly, stateness is defined by the authors in question not only as the existence of a sovereign state able to collect taxes and monopolise the legitimate use of force within its territory, but also as the presence of basic agreements about how to define ‘a people’ for the purposes of democratic government. 1 Hence, they follow the Dahlian perspective according to which democracy can be established only if – or after – those living within a territory have reached an agreement about who is entitled to participate in the collective decision-making process. Put simply, stateness is conceived of as a prerequisite for democracy.
However, the current world is marked by the existence of many democratic regimes, which are able to grant free and fair elections, but cannot maintain more difficult-to-achieve conditions that are normally associated with the liberal understanding of democracy dominant in contemporary Western political thought. As Jørgen Møller and Svend-Erik Skaaning (2011, p. 17) point out, stateness seems to be ‘more of a prerequisite for the rule of law and social rights than for electoral rights and political liberties. [This means that] the problems of stateness are more relevant for the thicker types of democracy (liberal democracy and social democracy) than for the thinner types of democracy (electoral democracy and polyarchy)’. Seen in this light, neither high state capacity nor the presence of basic agreements about how to define the demos are mandatory for the establishment of democratic regimes, at least those understood as Dahlian polyarchies.
This is not a trivial remark. Oddly enough, many tend to forget that the old democracies of the Western world have a long history of defining the people for the purpose of democratic government in a way that often seems to be achieved by questionable if not repugnant means. In addition, several of these democracies have endorsed a narrow concept of the people, to the extent that many of the populations living in their state territories explicitly or implicitly have been excluded from the demos (e.g. blacks, immigrants, women, etc.). Therefore, we should not be surprised that the new democracies of the non-Western world are not fully democratic in relation to everyone subject to its rules. One of the lessons of the historical Western route to democracy is that defining who belongs to the people is anything but a neutral political endeavour, but rather a breeding ground for some of the bitterest human animosities and vicious behaviours (Smith, 2003).
How to Control the Controllers? The Limits of Self-Government
The second dilemma identified by Dahl alludes to the complex relationship between constitutionalism and popular sovereignty. While Dahl is aware of the fact that it is important to have independent and unelected institutions specialising in the protection of fundamental rights that are intrinsic to modern democracy (e.g. rights to express one's views freely, to assemble, to vote, to form political organisations, etc.), he is also concerned about the downsides of these institutions, particularly when it comes to the people's capacity to monitor their functioning and enact the popular will. After all, the very notion of democracy implies that the ultimate political authority is vested in the people and not in divine powers or unelected bodies composed of experts. Not by coincidence, Dahl refers to constitutional courts and similar institutions as ‘quasi guardians‘. This language is particularly ironic, since for Dahl ‘guardianship’ is synonymous with a specific form of authoritarian rule, which distrusts the moral and intellectual competence of ordinary people, and advocates the formation of a government controlled by a minority of wise persons, who allegedly have superior knowledge about how to achieve the common good. Dahl's scepticism towards a vigilant judiciary empowered to strike down policies adopted by the executive and legislature (i.e. democratically elected powers) can be summarised in a basic principle:
There is necessarily an inverse ratio between the authority of the quasi guardians and the authority of the demos and its representatives. If the authority of the quasi guardians were comprehensive the demos would alienate its control over the agenda of public affairs and the democratic process would be gutted…. The more the scope of the rights and interests the quasi guardians are authorized to protect, the more they must take on the functions of making law and policy (Dahl, 1989, p. 188).
In other words, putting limits on the theory and praxis of self-government is a highly problematic measure, because then political power is delegated to authorities that are insulated from the people's influence and, in consequence, the problem of who controls the controllers comes to the fore. Or should the demos simply trust that the quasi guardians seek the common good rather than the interests of powerful minorities? Dahl's answer to this hypothetical question is rather negative and can be found in his devastating assessment of the US Constitution (Dahl, 2003). For the sake of simplicity, I will refer to two criticisms characteristic of the Dahlian perspective which are crucial for understanding the dilemma of the limits of self-government.
The first problem raised by Dahl is that the very process of constitution making is not just a neutral and technical operation, in which the debate between rational actors leads to the formulation of a complete institutional design. Quite the contrary: the constitution-making process is first and foremost a political battle characterised by a hard bargain between different, and to a certain extent antagonistic, forces. The necessity of reaching agreements among conflicting interests results in several democratic shortcomings (e.g. the protection of unequal representation through the senate in the case of the US) that are almost impossible to overcome later on, due to supermajority requirements, which tend to privilege the status quo and preserve existing injustices against reform by a majority decision. Being aware of the political struggle involved in the formulation of any constitution helps us to think more seriously about their democratic character. As Melissa Schwartzberg (2007, p. 19) indicates, ‘[i]f we preserve a vision of a constituent assembly as at its core a legislature – with the strengths and weaknesses associated with democratic decision making – the notion that fundamental constitutional norms ought to be entrenched becomes much less attractive’.
Second, even if the constitution-making process has ended in the production of a well-functioning institutional design, there is no democratic argument for impeding the ability of the people to play an active role in (re-)founding and updating the higher legal norms and procedural rules that regulate the exercise of political power. In fact, the ability to change existing laws is an essential means by which the demos can not only craft novel institutional solutions to current problems, but can also ensure that the laws remain just. In Dahl's own formulation: ‘No majority should have the right, moral or constitutional, to foreclose decisions by future majorities’ (Dahl, 2003, p. 148). This problem has been analysed by contemporary scholars under the rubric of ‘the paradox of constitutionalism’, whereby the insoluble tension between constituent power and constitutional form is pointed out (Loughlin and Walker, 2007). While the former alludes to the people as an active agent of change whose legitimacy is connected to the democratic possibilities of self-authorship and self-authorisation, the latter refers to the existence of an important set of institutions specialising in dividing, constraining and exercising governmental power. What is more, there are good reasons to think that the constitutional form has become increasingly influential, but at the cost of the constituent power. A paradigmatic example of this can be found in the recent history of Europe, where deep distrust of popular sovereignty underlay the political reconstruction of Western Europe after the Second World War in general, and the project of the European Union in particular (Müller, 2011).
It is worth noting that the two problems sketched above – the political nature of the constitution-making process, and the insoluble tension between constituent power and constitutional form – are at the centre of attention of contemporary democratic theory. This is particularly evident in the work of Jürgen Habermas (1992; 1996), who is one of the most vigorous defenders of constitutionalism. According to him, the solution relies on the so-called ‘co-originality thesis’, which states that the rule of law and popular sovereignty are created simultaneously and maintain a harmonious relation. This means that the constitution must be seen as a continuously self-corrective mechanism that is able to grasp the changing ideas and interests of the people:
each citizen of a democratic polity can at any time refer to the texts and decisions of the founders and their descendants, in a critical fashion, just as one can, conversely, adopt the perspective of the founders and take a critical view of the present to test whether the existing institutions, practices, and procedures of democratic opinion- and will-formation satisfy the necessary conditions for a process that engenders legitimacy … On this premise, each founding act also creates the possibility of a process of self-correcting attempts to tap the system of rights ever more fully (Habermas, 2001, pp. 775–6, emphasis in original).
At first glance, the solution offered by Habermas seems appealing, because it conceives of constitutionalism as a dynamic system, which goes hand in hand with the collective learning process experienced by the people across time. Nevertheless, the empirical and theoretical plausibility of this argument can be contested in several ways. To begin with, Habermas' line of reasoning is teleological. He does not think about the future as something contingent and open-ended, but rather as an ongoing ‘self-correcting’ process leading to human progress. Intentionally or not, this position legitimises the status quo, as Habermas presumes that the shortcomings of any constitutional order are going to be solved in the near future, but he is silent about whether the people considered as constituent power should play an active role in this process. Thus, Bonnie Honig is right in stating that:
Of the four modes of relation to a constitution – application, interpretation, amendment, and revolution – Habermas treats mostly the first two and reserves those for courts (which may, however, be influenced by episodic political action). Amendment and revolution are more obviously modes of relation to a constitution that might introduce something new, an event, a rupture, a cry…. His constitutional democracy is fed (through discriminating sluices) by the publics and counter-publics of a decentred and dispersed popular sovereignty that he repeatedly characterizes as ‘wild and anarchic’ (Honig, 2001, p. 799).
In addition, Habermas takes for granted that popular sovereignty and the fundamental rights that are normally entrenched in a liberal constitution are co-originally constituted, but he is not able to demonstrate the empirical underpinnings of this thesis. This is not a minor point. The history and current situation of both ‘developed’ and ‘underdeveloped’ countries is characterised by the existence of inhabitants who may agree with the idea of popular sovereignty, but reject, or only partially believe in, the fundamental rights. Extreme examples in the contemporary world range from al-Qa'eda militants to spree killers in the US and neo-fascist activists in Europe. As a consequence, these inhabitants consider the fundamental rights as an arbitrary imposition that constrains the people's capacity to rule themselves. Not by coincidence, liberal democratic theories are haunted by the spectre of ‘constitutional paternalism’ (Ochoa Espejo, 2011, p. 90). Instead of defending fundamental rights by appealing to universal moral values, Dahl's position is more realistic and is based on an empirical observation: ‘no constitution can ensure democracy in a country where the conditions favourable to democracy are absent’ (Dahl, 2003, p. 142). This means that constitutional entrenchment cannot sustain democracy in the absence of a shared commitment to certain fundamental rights, but neither can it bring about such a commitment (Schwartzberg, 2007, p. 205).
In summary, given that Dahl holds that the more power independent and unelected bodies have, the more constrained the people, the quasi guardians should have a very clear and narrow room of manoeuvre. Their primary task should be the protection of the democratic process, particularly with regard to two aspects: on the one hand, the defence of basic rights such as free speech and free association, and on the other hand, the respect of the all-subjected principle according to which no majority may unreasonably use its political rights to deprive a minority of its political rights. However, Dahl is aware of the fact that there is no perfect way of dealing with the tension between constitutionalism and popular sovereignty. From his perspective, the American solution of a strong supreme court with the faculty to strike down national legislation that violates fundamental rights should be seen as a system of quasi guardianship, which is neither necessary nor – based on the American experience – sufficient. In the absence of a universally best solution, specific solutions need to be adapted to the historical conditions and experiences, political culture, and concrete political institutions of a particular country (Dahl, 1989, p. 192). 2
The Responses of Populism
After having laid out two key democratic dilemmas that are at the heart of Dahl's theoretical framework – the boundary problem and the limits of self-government – it is time to turn our attention to the responses populism offers these dilemmas. However, this is not an easy task, because populism is a contested concept in the realm of the social sciences. Based on previous research (Mudde and Rovira Kaltwasser, forthcoming), I will first offer a succinct conceptual clarification. Then I will explore the ways in which populist forces in Europe, Latin America and the US are trying to solve the democratic dilemmas in question. That said, my aim is not to develop a comprehensive empirical analysis, but rather to illustrate the kinds of solutions that contemporary populists are prone to advance.
Defining Populism
Scholars agree neither on how to define populism nor on its essential features. To address this theoretical endeavour, Cas Mudde has developed a minimal concept, which is able to identify the core aspects of the phenomenon in question. According to him, populism should be conceived of ‘as a thin-centred ideology that considers society to be ultimately separated into two homogeneous and antagonistic camps, “the pure people” and “the corrupt elite”, and which argues that politics should be an expression of the volonté générale (general will) of the people’ (Mudde, 2004, p. 543). Above all, this definition stresses that populism is a moral set of ideas that is based on a Manichean distinction, whereby the elite is seen as a pathological entity, and the people is depicted as a homogeneous and virtuous community.
When defining populism as a thin-centred ideology, Mudde is applying Michael Freeden's (1996) morphological framework. 3 This seeks to study the internal structure of ideological belief systems, which can be classified into two groups. Whereas thick-centred (or macro) ideologies, like liberalism and socialism, are overarching networks of ideas that offer solutions to all the important political issues confronting a society, thin-centred (or micro) ideologies, like nationalism and populism, are world views limited in their ambition and scope, because they rely on a restricted morphology that may answer some (but not all) political questions that societies generate. Hence, one of the key characteristics of thin-centred ideologies is their capacity to sever themselves from wider ideational settings and adapt to specific contexts that enable them to attract the interest of large political groups (Freeden, 1998). Not by coincidence, populism has a chameleonic nature, and can be both right wing and left wing (Taggart, 2000). As Ben Stanley (2008, p. 100) has aptly noted, ‘there is no Populist International; no canon of key populist texts or calendars of significant moments; and the icons of populism are of local rather than universal appeal’.
Nevertheless, the ‘thinness' of the populist ideology should not distract us from the existence of certain elements that are inherent in all manifestations of populism. Mudde's definition makes clear that populism has three core concepts: the pure people, the corrupt elite and the general will. Interestingly, neither the pure people nor the corrupt elite are actual entities, but rather – paraphrasing Benedict Anderson (1983) – ‘imagined communities’ that are constructed in very different ways by specific populist actors and constituencies. In effect, the classification of who belongs to the pure people and the corrupt elite varies not only from one populist manifestation to another, but also throughout one populist experience over time. As I will show in the next section, examining how each of these ‘imagined communities’ is defined by different populist forces is a fitting method for getting a better picture of the contrasts between the current manifestations of populism in Europe, Latin America and the US.
The volonté générale is the third core concept of the populist ideology. It plays a major role since it allows for developing a specific notion of politics: given that the people is the sovereign, nothing should constrain its will. Certainly, this approach is akin to the Rousseauian critique of representative government, according to which the latter is an aristocratic form of power whereby a minority of elected authorities rule. In this sense, populist forces understand political representation in terms of imperative mandate: ‘the delegate operates under a fiduciary contract that allows the principal (the citizens) to temporarily grant an agent their power to take specified actions but does not delegate the will to make decisions, which is retained by the principal’ (Urbinati and Warren, 2008, p. 391, emphasis added). Unsurprisingly, populist forces normally support direct democratic mechanisms and other strategies that allow an unmediated relationship between the constituencies and the leader.
Finally, it is important to point out that there are two direct opposites of the definition presented here: elitism and pluralism. To a certain extent, elitism can be seen as the mirror image of populism's Manichean distinction between the people and the elite, because it takes for granted that the people is nothing else than a dangerous mob, while the elite is intellectually and morally superior. Pluralism, in turn, is opposed to the monist perspective of both elitism and populism, since it maintains that societies are characterised by the coexistence of a great variety of groups with different ideas and interests. According to this view, the people is a constantly changing ensemble of individuals and, in consequence, the unified will of the people is a fictional entity. Thus, pluralism assumes that not a collective entity in singular (i.e. ‘a people’) but rather a multiplicity of individuals and groups exert influence within the polity (Plattner, 2010, p. 89).
In sum, populism should be conceived of as a specific set of ideas that has three core concepts – the pure people, the corrupt elite and the general will – which are the necessary and sufficient conditions for classifying a phenomenon as populist. In other words, leaders or movements that are just anti-elitist, or simply allude to the general will, are not examples of populism. Therefore, although populism seems to be very widespread today, most political actors and parties in the Western world still accept the pluralist world view of liberal democracy (Mudde, 2004, p. 545). Moreover, relying on a similar concept of populism, Kirk Hawkins (2010) has empirically demonstrated that, even in contemporary Latin America, only a minority of chief executives can be categorised as populists. This means that populist actors are not omnipresent.
The Populist Way Out of the Boundary Problem
How can we define ‘a people’ for the purpose of democratic government? When confronted with this question, populist forces are prone to advance two kinds of answer, which are not inevitably mutually exclusive: ethnos and plebs. The first option is to resort to an organic conception of the pure people, according to which individuals sharing clear racial and cultural attributes compose national communities. Seen in this light, a singular ethnic group living in a specific territory has the right of self-determination. The logical consequence of this definition is discrimination against ethnic out-groups, which at the extreme may be excluded and/or cleansed from the territory (Mann, 2005, p. 55). In contrast, the notion of the plebs relies on a stratified concept of the pure people, which alludes to the multitude of ordinary folks vis-à-vis the elite (Canovan, 1984, pp. 315–6). The peculiarity of this conception is that it makes all inhabitants of the governed territory members of the people, lowering the status of the gentry and raising the status of the underdogs (Yack, 2001, pp. 521–2).
Looking at the contemporary world, the most evident examples of populist forces employing an ethnic definition of the pure people can be found in Europe. As Mudde (2007) has analysed in detail, populist radical right parties in Europe should be seen as the mirror image of the European greens. Whereas the former support a nativist view regarding who belongs to the demos and thus have clear xenophobic tendencies, the latter are staunch defenders of multiculturalism and the very idea that the coexistence of different national groups represents an enrichment of rather than a danger to the state. Interestingly, European populist radical right parties have been changing the way in which they define the pure people. Over the course of time they have started to develop more sophisticated approaches, such as the notion of ‘cultural differentialism’, which states that there are different cultures in the world and each of them has the right to defend its own identity (Betz and Johnson, 2004, p. 317). Hence, foreigners who do not assimilate into their new cultures represent a threat, since they challenge the integrity of the pure people. In fact, after the 9/11 terrorist attacks, the demonised out-group is related less to non-European immigrants (an ethno-national definition of the ‘aliens‘), and much more to Muslims as such (an ethno-religious definition of the ‘aliens’) (Betz, 2007).
In contemporary Latin America the situation seems to be radically different. Unlike the exclusionary identity politics approach used by populist forces in Europe, Latin American populist actors follow a discourse centred on material redistribution. Even in the case of Evo Morales in Bolivia, the populist set of ideas does not appear in combination with an exclusionary definition of the people (‘we, the indigenous‘), but rather with an inclusionary rhetoric that appeals to a variety of ethnic groups (‘we, the mestizos’) at odds with the establishment (Madrid, 2008; 2012). By framing a unified conception of the people as the underdogs, Latin American populist actors portray the less privileged majority of the society as a political community with a common life stretching back into the past and sharing a common destiny in the future. Given the high levels of socio-economic inequality existing in Latin America, this conception of the people is helpful for mobilising excluded sectors and assigning them an active role in politics (Rovira Kaltwasser, forthcoming). As Carlos de la Torre (2007) argues, the populist ideology is being used in Latin America today by leftist leaders who endorse a radical model of democracy in which intermediary institutions like political parties are distrusted, while mass rallies, the occupation of public spaces and a process of continuous campaigning are seen as legitimate mechanisms to strengthen the ‘voice of the voiceless’.
With regard to the current state of populism in the US, although the populist uprisings are only recently in the making and, in consequence, there is almost no research to draw upon, there are good reasons to think that the most clear example comes from the right side of the political spectrum: the Tea Party movement. 4 This is a case of populism in which both bottom-up and top-down dynamics are present (Formisano, 2012). On the one hand, its rise is to a certain extent related to the resources coming from corporations, billionaires and right-wing media. On the other hand, its emergence is also related to grassroots organisations and the existence of an important number of outraged activists who share the idea that politics has escaped popular control. Besides the question of what forces are driving the Tea Party movement, there is little doubt that it is based on an organic and nativist concept of the people, according to which the demonised out-group is composed not only of illegal immigrants and undeserving freeloaders but also Muslims (Skocpol and Williamson, 2012, pp. 68–77). At the same time, one of the Tea Party movement's central claims is that the people have been shut out of power by corrupt politicians, who support big government and public spending instead of promoting individual liberty and a balanced budget (Price Foley, 2012, pp. 26–9).
At first glance it might seem that populist forces employing an organic and ethnic concept of the people represent a democratic threat, while those who use the notion of the plebs should be seen as beneficial for democracy. Indeed, some scholars have developed a similar argument to make a distinction between ‘bad’ and ‘good’ populism (e.g. Kazin, 1995). Nonetheless, all kinds of populist forces are problematic for democracy in the sense that they utilise a moral language whereby the possibility of reaching agreements is extremely difficult, if not impossible – remember that populists believe not only that the people are pure and the elite is corrupt, but also that politics is about enacting the putative will of the people. By saying this I am not denying that, under certain circumstances, populism can have a positive impact on democracy. As I have argued elsewhere in more detail (Rovira Kaltwasser, 2012), my point is rather that we need more empirical research, and fewer normative and theoretical analyses, in order to gain new insights about the conditions that may determine when, why and how populism works as a threat to or a corrective for democracy. For instance, as paradoxical as it might appear, populist radical right parties in Europe have probably triggered a learning process by which mainstream political parties have started to become more responsive to societal demands and to rethink their policies in key areas such as corruption and immigration (Mudde and Rovira Kaltwasser, 2012b, p. 214).
The Populist Solution to the Limits of Self-Government
How to control the controllers? Although populist forces tend to argue that there should be no limits to self-government, they have different approaches when it comes to offering an answer to this question. In more specific terms, the answers that they normally develop are related to the way in which they think about the power of the corrupt elite. When populist forces do not have major complaints about the underlying constitutional order, they claim that the establishment is not acting properly and, in consequence, that those who belong to the pure people should be in charge of the independent institutions specialising in the protection of fundamental rights. In short, the silent majority should be the quasi guardians. In contrast, when populist forces claim that the existing constitution has been made to protect the interests of the establishment, they are prone to promote constitutional amendments or reforms with the aim of developing new institutions which are able to enact the popular will. In this case, the problem is not only the corrupt elite as such, but also the constitutional framework that the latter has designed in order to constrain the people's power.
Most European populist radical right parties do not have major problems with their national constitutions, and in fact many of them portray themselves as loyal supporters of the rule of law, particularly when it comes to denouncing Muslims as a threat to the liberal values of democracy. For instance, the Austrian Freedom Party (FPÖ), one of the oldest and most established populist radical right parties in Europe, maintains that:
we commit to a secular society, and the separation of church and state. Freedom of religion involves not only the free choice of religious affiliation but also the protection of citizens against extremists, which due to allegedly religious beliefs are not willing to respect the Austrian constitution and our laws. … Forced marriage, female circumcision and the oppression of as well as violence against women are by no means backed by freedom of religion in our constitutional state (FPÖ, 2011, p. 53, author's translation).
At the same time, European populist radical right parties are increasingly troubled by the influence of the European Union, since this implies a frontal attack on the principle of popular sovereignty (Berezin, 2009). Accordingly, they normally claim that international agreements affecting national constitutions and the people's capacity of self-determination must be approved by referendum. 5 This does not mean, however, that European populist radical right parties nurture liberal constitutions. Given the populist nature of these parties, they defend the majority principle at any cost. Indeed, most populist radical right parties in Europe end up proposing an ethnocracy, that is, a model of democracy in which the state belongs to an ethnic group, so that other ethnic groups can only live there if they accept this group's dominance (Mudde, 2007, pp. 142–5). This view certainly threatens the underpinnings of liberal constitutionalism, because it endangers the rights of (‘non-native’) minorities that do not necessarily share the values of the dominant ethnic group and/or of those who are in power.
Populist forces seem to be more radical in the case of Latin America. They are promising to activate the constituent power in order to craft new constitutions that are in tune with the ideas and interests of the people rather than of the establishment. By way of illustration, once in power, Hugo Chávez in Venezuela, Evo Morales in Bolivia and Rafael Correa in Ecuador have each set up a constitutional assembly which has drafted a totally new constitutional order (Silva, 2009). Certainly, these reforms have not only challenged the power of the elites but have also sought to construct a more participatory democracy with the aim of promoting the inclusion of voices that have been silenced from time immemorial. While it is true that these constitutional reforms can be seen as problematic from a liberal democratic point of view due to their tendency to weaken checks and balances, it is important to point out that the populist forces in question cannot be depicted as authoritarian per se. As Maxwell Cameron and Kenneth Sharpe note, ‘a significant feature, important to understand and often overlooked, is that the [populist] left in Venezuela, Bolivia and Ecuador is pursuing its agenda through constitution making, not through revolt, violence, or revolution. It is not pursuing its agenda extra-legally but by changing the basic legal structure’ (Cameron and Sharpe, 2010, p. 65, emphasis in original). Anyway, it is clear that contemporary populist forces in Latin America are invoking what Canovan (1999) has called the redemptive face of democracy, that is, a secular notion of salvation which implies that only a ‘government by the people’ can overcome the existing political alienation and realise the democratic ideal.
Finally, the main contemporary manifestation of populism in the US advocates a very specific approach to deal with the limits of self-government, namely constitutional originalism. Indeed, the Tea Party movement is characterised by a radical defence of the founding documents and, in consequence, the belief that nobody has the right to cast stones at the United States Constitution, which is glorified as a truly national icon. Whereas constitutional originalism is seen as the right answer, judicial interpretation is depicted as an evil mechanism whereby corrupt elites can govern against the wishes of the electorate. In the words of Elizabeth Price Foley (2012, p. 188), ‘[i]f ordinary people aren't ready to desegregate schools or legalize abortion, sodomy, or same-sex marriage, the progressive intelligentsia needs a mechanism to bypass our republican form of government and hand the issue off to another, more “educated” group. Living constitutionalism is the perfect solution‘. The Tea Party movement sees such living constitutionalism as a grave problem, caused by the influence of leftist intellectuals and politicians who are attacking the foundations of the heartland and seek to interpret the law in their own favour. There is no better example of this than the position of the Tea Party movement on the debate about the health care reform promoted by the Obama administration. Given that this reform is considered a ‘socialist’ takeover and thus a frontal attack on the US Constitution, the Tea Party movement argues that the Supreme Court should nullify the laws associated with the health care reform (Lowndes, 2012).
In summary, contemporary populist forces do not have a common stance when it comes to proposing an answer to the question of how to control the controllers. Three different strategies seem to be predominant. First, some populists are prone to defending the existing constitutional order. In this case, they normally invoke a particular interpretation of the constitutional texts, according to which the main problem stems from certain sectors of the establishment that are acting against constitutional principles (e.g. the Tea Party movement in the US). Second, when populist forces realise that international treaties or specific legal changes challenge the will of their constituencies, they claim that a plebiscite is the only legitimate mechanism to enforce relevant constitutional amendments (e.g. populist radical right parties in Europe). Third, the most radical option is when populist actors use the notion of the constituent power to defend the undertaking of major constitutional reforms and/or the writing of a new constitution by a democratic assembly designed for this purpose (e.g. radical leftist populist leaders in Latin America).
Concluding Remarks
Populism is neither democratic nor anti-democratic in itself. Instead of defining populism as a democratic disease or pathology, I have argued in this article that we should consider it as a thin-centred ideology which raises legitimate questions that are difficult, if not impossible, to solve by democratic means. In addition, I have shown that different manifestations of populism are offering specific solutions to two democratic dilemmas – the boundary problem and the limits of self-government – that are particularly pressing in Europe and the Americas. While it is true that the solutions offered by populist forces to those democratic dilemmas are controversial, the question about the impact of populism on democracy should be answered first and foremost empirically rather than based on normative and/or theoretical arguments (Rovira Kaltwasser, 2012). By way of comparison, consider nationalism, another thin-centred ideology that also has an ambivalent relationship with democracy. Interestingly, there is an ongoing debate about the impact of different variants or historical manifestations of nationalism (e.g. anti-colonial, civic, ethnic nationalism, etc.) on both democracy and the process of democratisation, but almost no one would go so far as to argue that nationalism per se is intrinsically dangerous to democracy.
Drawing on the work of Arditi (2004; 2005) and Canovan (1999; 2004; 2005), I would like to finish this article by providing some tentative ideas about how to rethink the relationship between populism and democracy in a more fruitful way. To begin with, those authors who conceive of populism as a democratic disease or pathology try to ‘externalise’ it and consider it as a ‘foreign body’. In contrast, I maintain that populism is something internal to democracy. Given that the core concepts of the populist ideology – the pure people, the corrupt elite and popular sovereignty – can be easily used to refer to the gap between democratic ideals and real existing democracies, we should not be surprised at the rise of populist actors who seek to enact the redemptive side of politics, and re-politicise those problems that intentionally or unintentionally are not being addressed by the establishment. Specific national and regional contexts offer different political opportunity structures for the emergence of subtypes of populism which, depending on the situation, can work as a threat to or corrective for democracy (Mudde and Rovira Kaltwasser, 2013).
If it is true that populism is not a democratic disease or pathology per se, but rather an ideology that follows democracy like a shadow (Arditi, 2004; Canovan, 1999), let me conclude by suggesting one promising avenue of future research, namely the complex relationship between populism and political representation. Scholars who see populism as anti-democratic in itself usually argue that populist forces are at odds with the very process of political representation. For instance, authors have maintained that populism is ‘hostile to representative politics’ (Taggart, 2002, p. 66), constitutes ‘pseudo-representation’ (Alonso et al., 2011, p. 11) and should be conceived of as a ‘perverse inversion of the ideals and procedures of representative democracy’ (Rosanvallon, 2008, p. 265). Although populist forces indeed tend to promote the use or introduction of plebiscitary instruments, and usually attack elected politicians and unelected officials, they are not against representation per se, but rather want to see their own representatives in power.
This means that it is flawed to assume that populists misrepresent the people. In order to gain new insights about the complex relationship between populism and political representation, the approach developed by Michael Saward (2006; 2010) is particularly interesting, because it helps us to think that the populist ideology fosters a particular mode of political representation. 6 With the aim of transcending orthodox accounts of representation, Saward has advanced a theory which stresses that, to study political representation properly, we should see the process of claim making as the core of representation (i.e. the way in which political figures act for others) and analyse the performative rather than the institutional side of representation (i.e. the aesthetic and cultural elements involved in the process of political representation). 7 Seen in this light, populism is less about ‘demagogic actors’ appealing to ‘foolish people’, but more about the dynamic relationship between political leaders and constituencies who share a common vision of society and how politics should function. Therefore, populism involves a creative process whereby concepts, diagnoses and images of the pure people, the corrupt elite and the general will are produced. Not by coincidence, the history of any manifestation of populism is full of examples of institutional improvisations and experiments, which seek to enact the world view shared by those who see politics through the lens of populism.
Footnotes
I am grateful to Sofia Donoso, Cas Mudde, Sofia Näsström, Kenneth Roberts, Paul Taggart and the three anonymous reviewers for helpful comments on earlier drafts. The usual disclaimer applies. Previous versions of this article were presented at the Political Studies Association Annual International Conference, held in Belfast, 3–5 April 2012, and the International Congress of the Latin American Studies Association, held in San Francisco (California), 23–26 May 2012. The research leading to these results has received funding from the European Community's Seventh Framework Programme (FP7/2007-2013) under grant agreement no: PIEF-GA-2010-273525.'
1
Since this aspect is often overlooked in the debate about stateness, it is worth evoking Linz and Stepan's own formulation: ‘When thinking about transitions to democracy, many people tend to assume that what is challenged is the nondemocratic regime and that with democracy a new legitimate system is established. However, in many countries the crisis of the nondemocratic regime is also intermixed with profound differences about what should actually constitute the polity (or political community) and which demos or demoi (population or populations) should be members of that political community. Where there are profound differences about the territorial boundaries of the political community's state and profound differences as to who has the right of citizenship in that state, there is what we call a “stateness” problem. Aspirant modern democracies can vary immensely on this variable from those polities that have no stateness problems to those where democracy is impossible until the stateness problem is resolved’ (Linz and Stepan, 1996, p. 16).
2
A similar argument has been put forward recently by Adam Przeworksi, who maintains that ‘there is no evidence that institutional systems that give more power to majorities are more abusive or capricious than those that impede governments from acting by checks and balances. In the end, although some limits may be generic, each democracy suffers from its own deficiencies. … Indeed, some facts are still as shocking as they are puzzling. How is it possible that a country that has had representative institutions for more than 200 years would have the highest degree of economic inequality among the developed nations and the highest proportion in the world of people languishing in jails, higher than most repressive autocracies? How is it possible that almost one-half of the electorate in this country does not vote even once in four years? How is it possible that the people in this country tolerate the flagrant influence of money over politics?’ (Przeworksi, 2010, p. 170).
3
At this stage, it is worth indicating that although Mudde's concept has some affinity with the approach developed by Laclau (2005), they differ in an important aspect: while the former is useful for undertaking empirical research, the latter is above all a highly abstract and normative framework that might be useful for thinking about the ambivalent relationship between populism and democracy, but is problematic for analysing this relationship from a theoretical and empirical point of view (Mudde and Rovira Kaltwasser, 2012a, pp. 13–5).
4
Some scholars have proposed that the Occupy Wall Street movement is another example of contemporary populism in the US (see, for instance, Walzer, 2012). However, according to the definition of populism proposed in this article, this is a borderline case. Although the Occupy Wall Street movement identifies a ‘pure people’ (the 99 per cent) and a ‘corrupt elite’ (the financial sector and its allies), I have serious doubts whether the defence of the volonté générale at any cost forms part of its discourse. Otherwise stated, because many – if not the majority of – supporters of the Occupy Wall Street movement adhere to pluralism, it is probably wrong to argue that this movement as a whole is an example of a populist uprising. I am grateful to an anonymous reviewer at Political Studies for directing my attention to this.
5
In fact, the FPÖ claims that the recent coming into effect of the Lisbon Treaty shows that ‘the strategies of those political forces interested in the construction of a super-state akin to the US American model are being implemented against the will of the citizens of the European member states. … Yet, the Austrian Freedom Party … demands a reversal towards a Europe of fatherlands and sovereign nation states. Hence, there must be a new European constitutional agreement, which addresses precisely this reversal. A new constitutional agreement of this kind comes into consideration for the Austrian Freedom Party only if a binding plebiscite in Austria takes place’ (FPÖ, 2011, p. 271, author's translation).
6
There is growing literature on political representation which could be drawn on to examine the ways in which populist actors and constituencies not only understand but also enact political representation. For a useful overview of some of this literature, see Näsström, 2011b.
7
Although it is outside the scope of this paper to discuss the conceptual and theoretical approach advanced by Benjamin Moffitt and Simon Tormey (2013 forthcoming) in a recent piece published in Political Studies, I completely agree with them in that it is key to analyse the performative dimension of populism and that for this purpose the work of Michael Saward can be of utmost usefulness. However, I have serious doubts about whether the notion of ‘political style’ is helpful not only for thinking more thoroughly about the performative elements of populism, but also for undertaking empirical research on populism. Suffice to say here that the very conception of populism as a set of ideas or a thin-centred ideology does not preclude the study of the performative dimension of populism. Moreover, given that Moffitt and Tormey's notion of ‘political style’ draws on Wittgenstein's family resemblance approach, it does not enable the development of a minimal concept whereby the necessary and sufficient attributes of populism can be identified in order to advance an approach that avoids the problems of conceptual travelling and stretching. For more details on the advantages of a minimal concept over the family resemblance approach, see Mudde and Rovira Kaltwasser 2012a, 2013, forthcoming; 
.