Defending democracy: Militant and popular models of democratic self-defense

With the electoral victories of authoritarian populists in a range of parliamentary democracies in recent years, there has been a growing unease with the ability of existing democratic institutions to keep such authoritarian threats under control. The election of authoritarian leaning figures in countries such as Hungary, Poland, Philippines, Brazil, Russia, and the United States has led many to doubt the capacity of the institutions of parliamentary democracy to protect themselves against democratic backsliding. This perceived inability for democratic self-defense has led to a resurgence of academic interest in the idea of militant democracy in recent years (Abts & Rummens, 2010; Cappocia, 2013; Kaltwasser, 2019; Kirshner, 2014; Malkopoulou & Kirshner, 2019; Müller, 2012; Sajo, 2012).

The concept of militant democracy was originally coined by the German-Jewish émigré and constitutional scholar Karl Loewenstein, who in two articles in APSR in 1937 sought to develop ways in which representative democracies could respond to the emergence of fascism. Loewenstein's argument was that free and equal political elections could open the path for a fascist dismantling of representative democracy via democratic means. Consequently, democracy had to become militant and safeguard itself by compromising with its foundational principles of freedom and equality by prohibiting extreme political parties and by curtailing the political rights of extremists (Loewenstein, 1937a, 1937b). As such, it is not difficult to see why contemporary scholars want to revive Loewenstein's idea of militant democracy as a response to populism. The main threat to present-day democracies, many argue, does not stem from revolutionary movements, which seek to subvert democracy through insurrection (Runciman, 2018, pp. 2–3; Levitsky & Ziblatt, 2018, pp. 5–6), but rather from the gradual erosion of democratic norms and institutions by elected political leaders.

Contemporary political leaders such as Donald Trump in the United States, Victor Orban in Hungary, Silvio Berlusconi in Italy, Nicolás Maduro in Venezuela, Jaroslaw Kaczynski in Poland, Recep Erdogan in Turkey, and Vladimir Putin in Russia have all ascended to power via more or less legitimate electoral channels and have—to a varying degree—centralized power, dissolved institutional checks and balances, and rolled back political rights. Moreover, contemporary populists display an antipluralist, anti-institutional, and authoritarian interpretation of popular sovereignty, insofar as many populist leaders claim to be the true representative of the people, denying the political legitimacy of political opposition and constitutional limits to the executive (Finchelstein, 2017; Müller, 2016a; Rummens, 2017)¹. Although militant democratic measures were developed to combat fascism in the 1930s, neo-militant models try to contain contemporary right-wing populism and prevent further democratic backsliding. Consequently, neo-militant democrats have developed institutional and juridical ways of limiting the political influence of elected populists and populist movements (Abts & Rummens, 2010; Kirshner, 2014; Müller, 2012; Tyulkina, 2015). The remedy to right-wing populism from such neo-militant democrats often involves restricted access to the political sphere either in the form of party bans (Bourne, 2012), restrictions on individual and political rights (Abts & Rummens, 2010), increased electoral threshold, or the strengthening of independent institutions like constitutional courts (Mounk, 2018, p. 257). In How Democracies Die, for example, Steven Levitsky and Daniel Ziblatt argue that the central historical and institutional precondition for the election of a populist such as Donald Trump was the demise of antimajoritarian, gatekeeping institutions and the removal of the “filtering role” of political parties in presidential nominations after the McGovern-Fraser Commission in 1971 recommended binding primary elections (Levitsky & Ziblatt, 2018, pp. 48–52). Implied in their argument is that without the gatekeeping, antimajoritarian functions performed by the “smoke-filled room” (Levitsky & Ziblatt, 2018, p. 41) of nonelected, unaccountable, elite officials, “the people” are free to, and will eventually, elect a demagogue like Trump. For Yascha Mounk (2018, pp. 257–259), the best way to contain a populist in office is to rely on constitutional courts as guardians of the constitution—a core feature of militant democracy.

In addition to his work on populism and the intellectual history of 20th century democratic ideas, Jan-Werner Müller has also been an important analyst of the nuances and problems of the militant democracy strategy in the postwar era (Müller, 2012, 2016b). Faced with the recent rise of populism, Müller identifies a form of “soft militant democracy” as a response to the present authoritarian danger. He contrasts such “soft” version with “the ultimate ‘hard’ measure of banning a party or restricting rights to certain kinds of speech,” as the “soft” version merely “leaves a party in existence – but officially limit its possibilities for political participation, or de facto make life for the party difficult” (Müller, 2016b, p. 259). As such, many intellectuals, who worry about the fate of liberal democracy, conceptualize one important source of liberal democracy's crisis as residing in extreme popular movements and populist parties, as an unreasoned and dissatisfied population, attracted to the dangerous political ideologies of right-wing populists, who promise the unrealistic restoration of an unbridled national sovereignty. The crisis of liberal democracy, in this line of thinking, emerges through choices and actions of an unbridled majority, and as such the remedy is to limit the popular access to the political sphere and count on antimajoritarian institutions like constitutional courts or legal obstacles like new party legislation to curb populist forces. These measures resemble the original militant democratic strategies developed by Loewenstein (1937a, 1977b). As such, to save liberal democracy, some critics of populism argue, the values of freedom and equality on which this regime is founded must be temporary suspended for certain political groups and demands (Abt & Rummens, 2010).

This way of countering the potential authoritarian threats to democracy has some limitations. As such, we argue that the policy prescriptions and modes of analysis associated with both hard and soft versions of militant democracy can productively be supplemented with other, less antimajoritarian and elite-driven approaches to democratic self-defense. The problem with the modes of democratic self-defense inspired by militant democracy is twofold.

First, on a normative level, we will argue that the idea of defending democratic institutions by limiting popular participation and expression is questionable as its rests on a depoliticizing, elitist, and exclusionary understanding of politics, relying on handing power to unelected and potentially unaccountable technocrats or jurists. Second, on an empirical level, we will argue that a militant approach to democratic self-defense risks, on its own terms, being counterproductive, as the exclusion of certain popular demands by the political elites might only intensify the political narrative on which populists are already harvesting votes. Insofar as the militant model of democratic self-defense depends on creating a conflict between a popular majority and political elites such as representatives, judges, or other unelected magistrates, it risks backfiring by politicizing the cleavage between an authentic people and technocratic elites, through which authoritarian populist projects tend to thrive.

To remedy these shortcomings, we propose a supplement to the militant democratic approach. Recognizing, as militant democrats point out, that existing institutions of parliamentary democracy have trouble dealing with authoritarian threats from within, we propose amending the militant model with a popular model of democratic self-defense. This popular understanding of democratic self-defense, drawing from both popular republican and socialist imaginaries, relies on institutional ways of deepening, rather than restricting democratic participation. Such a popular understanding of democratic self-defense involves not only an awareness of the dangers to democracy stemming from potentially authoritarian demagogues, but also to threats stemming from unaccountable economic elites, and to the inadequacy of liberal democracy to resist the translation of economic wealth into political power (McCormick, 2007). Instead of relying predominantly on the potentially depoliticizing and exclusionary strategy of militant democracy as a remedy to the contemporary crisis of democracy, we propose an “anti-oligarchic” strategy, which reintroduces the idea of institutions of collective power in order to combat excessive elite domination. It is important to stress that actually existing political systems might utilize both militant and popular instruments in defense of their democratic constitution. Hence, a democratic polity might strengthen its constitutional court (a militant instrument) while simultaneously establishing a second chamber of “ordinary” citizens with certain veto powers (a popular instrument). In this article, though, we are mainly interested in the conceptual, normative, and political differences between militant and popular models of democratic self-defense as models, that is, the ways in which the different models rely on either restricting or increasing popular participation as a means to defend the democratic constitution.

In order to advance this argument, the article is structured the following way: We begin, first, by revisiting the classical and contemporary arguments for militant democracy as democratic self-defense. Second, by reconstructing the genealogy of liberal democracy, we argue that the depoliticizing strategy of militant democracy is not a last resort of a liberal democracy in crisis; instead, depoliticizing and exclusionary strategies are integral to liberal democracy, and as such, militant democracy does not represent a perversion of liberal democracy, but rather a radicalization of tendencies already rooted in the liberal tradition. Third, we outline the historical trajectories of an alternative mode of democratic self-defense through a historical engagement with institutional solutions in the republican and socialist tradition. Lastly, we argue how these insights might form the basis of a supplementary, popular model for the defense of democracy that in contrast to the militant model does not seek to restrict but rather expand popular participation in democratic processes.

Militant democracy is a broad term for different legal and political mechanisms employed to prevent political extremism to emerge in a constitutional state with a representative government. The core idea of militant democracy is that democracies, in order to protect themselves, might under certain circumstances restrict the rights and access to the political system for those who seek to undermine democracy (Müller, 2012, 2016b). As noted above, some who deem populism an undemocratic, quasi-authoritarian political phenomena have turned to some version of militant democracy in order to contain the threat of populism (Müller, 2016b; Abt & Rummens, 2010).

Loewenstein's classic account of militant democracy was formulated in two articles from 1937 in which he analyzes how democratic systems can counter the threat posed by fascist movements. As such, militant democracy is an attempt to counter a specific political problem that emerges in the 20th century along with the spread of representative government, mass politics, and universal suffrage: If democracy essentially consists of free elections, universal suffrage, and majoritarianism along with the freedom of speech, assembly, and press, then antidemocratic movements can use the democratic process to subdue democracy itself. As Nazi minister of propaganda Joseph Goebbels once observed, “it will always be one of the best jokes of democracy that it gives its deadly enemies the means to destroy it” (Goebbels in Fox & Nolte [1995, p. 1]). By upholding a naïve loyalty to the democratic principles of free and equal access to the political sphere, Loewenstein argues, “fascist exponents systematically discredit the democratic order and make it unworkable by paralyzing its functions until chaos reigns” (Loewenstein, 1937a, p. 424). Through such “democratic fundamentalism,” democratic systems are effectively tolerating the “Trojan horse” of authoritarian movements using the democratic process of elections to subdue democracy (Loewenstein, 1937a, p. 424). In order to fight fascism, democracy itself must instead become militant, meaning that “if democracy believes in the superiority of its absolute values over the opportunistic platitudes of fascism, it must live up to the demands of the hour, and every possible effort must be made to rescue it, even at the risk and cost of violating fundamental principles” (Loewenstein, 1937a, p. 432). Such “every possible effort” involves a general constraining of democracy by banning subversive parties, heightening the electoral threshold, restricting freedom of the press in the form of criminalization of editorial subversive propaganda, restricting freedom of speech by prohibiting incitement to violence and hatred against particular groups of the population as well as prohibiting derogatory statements against democratic institutions, republican symbols, and high officials of the state (Loewenstein, 1937b, pp. 651–652). As such, by applying this militant democratic legislation, democratic states have begun the “deliberate transformation of obsolete forms and rigid concepts into the instrumentalities of ‘disciplined’, or even—let us not shy away from the word—‘authoritarian’ democracy” (Loewenstein, 1937b, p. 657, italics added). In short, militant democracy as a strategy of democratic self-defense involves—in classic formulation by Loewenstein—the transformation of the democratic ideal itself into a kind of political rule, which draws extensively upon the exclusionary strategies of the authoritarian ideologies, is to combat.

Contemporary neo-militant democrats have certainly moderated Loewenstein's original framework and shy away from authoritarian measures in order to make the restrictions on basic rights of political participation compatible with the fundamental principles of liberal democracy (Capoccia, 2013, p. 219). Some neo-militant democrats distinguish between antidemocratic actions and antidemocratic ideas, and limit militant measures to the former and not latter (Bourne, 2012, p. 209; Capoccia, 2005, p. 57). Others argue that antidemocrats have other legitimate political interests, which make exclusion illegitimate as long as they do not violate the right to participation of other citizens (Kirshner, 2014, pp. 40–41); yet others argue for a two-track strategy, where the threat of political exclusion increases as antidemocrats move closer to public offices and political power (Abt & Rummens, 2010).

Militant democracy has to a large extent set the parameters of the debates around countering populism today. As such, we argue that normatively militant democracy is elitist in its conceptualization of the political problem it seeks to remedy—insofar as the major threat to the democratic order primarily, though not exclusively, emanates from the dissatisfaction of ordinary people—potentially exclusionary, and depoliticizing in its responses to this problem and potentially ineffective, if not counterproductive in its results. The last issue concerning the effectiveness of militant democratic instruments is indeed an empirical question, one that we cannot do full justice to in this article, although we will provide some exemplary discussion.

First, militant democracy is an elitist strategy, as the task of combating political extremism is assigned to elected politicians, bureaucrats, or unelected, antimajoritarian institutions. The problem is most often associated with mass politics, which is deemed potentially volatile and violent. As Malkopoulou and Norman (2018) have recently argued, militant democracy is “a fundamentally anti-participatory and elitist logic … of anti-extremist politics” (p. 444), which regards mass participation as a potential threat that political elites are to counter by restricting the public sphere, constraining the democratic system and disciplining its culture. As highlighted in an overview article by Jan-Werner Müller (2016b, p. 254), while some understand militant democracy as part of a “transitional constitutionalism,” where the new elites are normatively justified in using strong juridical measures to defend the new democratic constitution against its enemies, others argue for a more fundamental normative justification by which a political system, and its governing elite, can never allow antidemocratic forces to come to power. Whether one operates with context-specific or fundamental justifications of militant democracy, the heart of the matter is that democratic self-defense in the militant register is the task of political elites.

Second, militant democratic strategies of self-defense are exclusionary, and potentially depoliticizing. Instead of facing political opponents in open political struggle, hereby emphasizing the pluralistic, conflictual, and agonistic nature of democracy (Lefort, 1988; Mouffe, 2013), militant democracy depoliticizes conflict and transposes it from the realm of politics into the legal realm, where exclusionary means like party bans and restrictions on rights of speech and assembly are used in order to stifle political conflict. In short, although the alternative model of democratic self-defense that we develop below encourages political conflict by empowering the citizenry through different institutional means, the militant model discourages political conflict by complicating the access to the public sphere for certain groups.

Third, the strategy of militant democracy is not only elitist, and depoliticizing, but also potentially ineffective and counterproductive. Many commentators on contemporary populism argue that the primary rhetorical strategy of populists is to highlight a conflict between the “pure” and “uncorrupted” people and the “self-interested” and “deeply corrupted” elites (Finchelstein, 2017; Mounk, 2018, pp. 41–46; Müller, 2016a, pp. 2–3, 103–104). By excluding certain parties from the political process as well as certain opinions from public debate, political elites might give further credibility to the “elite-versus-people”—narrative on which contemporary critics of liberal democracy are already mobilizing. It is obviously an empirical claim whether the use of militant democratic measures is responsible for creating political dissatisfaction. But, we argue, militant modes of self-defense posit a conflict between a popular majority against political elites, hereby potentially politicizing the cleavage between an authentic people and technocratic elites, through which authoritarian populist projects tend to thrive.

Based on these arguments, we find there are good reasons to supplement militant democracy with different modalities of democratic self-defense, which avoids some of the pitfalls of militant democracy. Such supplements entail procedures that are citizen driven as well as open to legitimate contestation and political struggle. One might object that militant democracy is only an extraordinary mechanism and that the moment democracy's enemies are defeated, the ordinary politics of liberal democracy will continue with its non-exclusionary, open, and egalitarian political processes. This is, for example, the argument of Ruti Teitel (2007, p. 49), who argues that “militant constitutional democracy ought to be understood as belonging to transitional constitutionalism, associated with periods of political transformation that often demand closer judicial vigilance in the presence of fledging and often fragile democratic institutions; it may not be appropriate for mature liberal democracies.” Here, we disagree. As we shall argue below, we regard the elitist, exclusionary, and depoliticizing elements of militant democracy as a radicalization of already existing tenets of liberal democracy, not the temporary suspension of liberal democracy's core ideals. Hence, we agree with Jan-Werner Müller (2011) that democracy as it has been institutionalized in the postwar constitutional settlements is indeed a “constrained democracy,” hereby making postwar liberal democracy and militant democracy members of the same species rather than fundamentally different.

In order to demonstrate how militant democracy is not a deviation, but rather a radicalization of tendencies in the liberal tradition, we revisit below the historical origins of liberal democracy. While proponents of “liberal democracy” like to trace its roots back to John Locke and the early modern period, the concept is of a relative recent pedigree. Duncan Bell has recently shown how the term “liberal democracy” did not come into regular use until the interwar years in conjunction with an understanding of a growing threat to the liberal order and a dichotomy between liberalism and totalitarianism (Bell, 2014). In this context, liberalism and democracy was increasingly tied together as not only connected, but mutually constitutive. This idea of liberal democracy, however, obfuscates the real political history, where liberalism and democracy, understood as broad-based popular sovereignty, have distinct histories, and have in most historical periods been in conflict.

Before the modern period, the term “democracy” was not principally used to specify a set of political institutions. Rather, democracy was defined as a type of social class rule, namely, the rule by the popular class—the poor—as opposed to the nobles or the propertied classes. In Aristotle's famous typology of state forms, “democracy” was defined by social class, rather than in institutional terms, as government in the interest of the poor (Aristotle, 1995, III, v. 4 [1279B]). Indirectly, as the poor constituted a majority, democracy involved majority rule, but the social definition was nevertheless central. This equation of democracy as majority rule with the political power of the “Party of the Poor,” and hence with egalitarian policies, can be seen through Western history. Andreas Kalyvas describes how democracy until the 19th century was seen as the “politics of the assembled poor” (Kalyvas, 2019, p. 539), finding the equation of democracy and the political power of the poor in figures from Xenophon to Thomas Aquinas, and Marsilius de Padua. This identification of democracy with the power of the poor reappears both with opponents of popular rule, such as the aristocratic republican Cicero (Wood, 2008, p. 143), and with the early modern proponents of democratic constitutions, such as popular republicans in the North Italian City States (McCormick, 2011) or the Levellers and Diggers of the English Civil War (Robertson, 2007; Rees, 2016).

The term “liberalism,” instead, was coined in the early 1800s, designating an ideologically centrist position on the constitutional question, in the spectrum between radical republican democrats and conservatives who defended absolutist monarchy: Liberals favored keeping monarchs, but curtailing their arbitrary power through constitutions (Fawcett, 2015). The new ideology of liberalism is built on a century-old tradition of liberal thought, represented by thinkers such as Locke and Montesquieu who voiced a critique of monarchy without demanding a fully democratic-republican constitution (Domènech & Raventós, 2008; Wood, 2008). Liberal political thought of course changed in the course of the 19th and 20th century, with classical liberals being pressured into accepting universal suffrage by movements of workers, women, and other excluded groups (Therborn, 1977). This gradual democratization of liberalism ended in the situation, where in the first decades of the 20th century it was possible to construct the idea of the eternal connection between liberalism and democracy that emerged in the interwar years. Despite this, however, some of liberalism's skepticism toward popular power remains. Paradoxically this can be seen in the liberal response to threats against liberal democracy itself. Here, defenders of liberal democracy, both in its militant and nonmilitant forms, have inherited a skepticism toward the popular masses that have survived the “democratization” of the liberal tradition.

The main problem of liberal theory, from this perspective, is that it has traditionally primarily been able to imagine threats against democracy as coming from either the state or the mob. David Held, for example, describes classical liberal democracy as essentially a form of “protective democracy” (Held, 2006, p. 99). This protection means on the one hand using the state to protect life and property against the mob, and on the other hand using the division of power, rule of law, and (limited) representation to protect the individual against the state. Of special concern was what Alexis de Tocqueville called the “tyranny of the majority” (De Tocqueville, 2003, p. 286), which followed from the introduction of representative government in the 18th and 19th century. This fear of the “tyranny of the majority” as a result of representative government is, as we have argued in the above, similar to the problem Loewenstein's militant democracy set out to solve a century later. With elected governments, the two dangers of the state and the mob could be combined by a poor majority using the power granted by general suffrage to confiscate property or tax away the wealth of the rich minority. When early advocates of what would become the liberal tradition like Madison or Montesquieu advocated for a mixed constitution, and opposed the notion of democracy, it was precisely in order to make sure that popular power was balanced with elements of elite rule. As Madison famously argued in federalist paper no. 10, “democracies have ever been spectacles of turbulence and contention; have ever been incompatible with personal security or the rights of property; and have in general been as short in their lives as they have been violent in their deaths” (Madison et al., 1961, p. 76). In short, Madison forcefully condenses the liberal fear of the people, insofar as he regards popular rule as volatile, unruly, and insecure as well as threatening to private property. For that reason, Madison argued in paper no. 63 that the defining characteristic of the American Constitution “lies in the total exclusion of the people in their collective capacity” (Madon et al., 1961, p. 385). Instead, liberals like Madison proposed representative government and division of power as explicitly nondemocratic means of governing the polity and preventing “the tyranny of the majority.” John Stuart Mill, arguably one of the greatest proponents of inclusive government within 19th century liberalism, also used the term widely and advocated for limiting the democratic elements of the constitution and create institutions that should be “protecting minorities by admitting them to a substantial participation in political power” (Mill, 2008, p. 302). In this way, proponents of liberalism envisioned protective institutions as necessary in order to protect individuals against the state, and protect executive state power against democracy, that is, the political power of the poor. This resulted in a set of antimajoritarian institutions, such as powerful political courts with appointed (elite) officers and constitutional limits to democracy.

One way to further demonstrate the relation between the elitist and depoliticizing strategy of militant democracy specifically and liberal democracy more generally is to note how liberal democrats interpreted the rise of the fascism in the first half of the 20th century. This gives us a good indicator of how liberal and militant democrats understand political problems and potential remedies alike. Interestingly, when liberal democracy was reinvented in the wake of the Second World War and the experiences of fascism, it was comparable forms of antimajoritarian institutions that were set up, as when liberals in the 18th and 19th centuries were trying to diminish the direct popular influence on their newly established constitutional states. The experiences of fascism were largely interpreted as a case of the excess of popular majority power, and the need was therefore to rein in democracy, creating a “disciplined democracy” (Müller, 2011, p. 39). This development was especially prominent in the European context. Michael Wilkinson describes how in the postwar era “European elites attributed the collapse of interwar liberal democracy to over-politicization” and that the relationship between state and mass democracy therefore had to be “reconstituted through a process of internal depoliticization” (Wilkinson, 2021, p. 74). What was different in the postwar period was that instead of conceptualizing these antimajoritarian institutions as limits to a popular majority appropriating private property, they were now construed as necessary safeguards for protecting democratic majorities against their own antidemocratic proclivities. Parallelly, even though the issue of minority protection was now cast in terms of protections for ethnic and minority rights, the type of institutional setup proposed to remedy these threats was to large extent similar to the antimajoritarian institutions that 19th-century liberals had envisioned for protection of the wealthy minorities (Moyn, 2018).

This is interesting, because the interpretation that interwar fascism sprang from an excess of popular power hardly seems the only explanation—or even, as shall argue below, an especially convincing explanation. Similarly, although some commentators link the electoral success of contemporary populism to various pathologies within the citizenry (i.e., the emotional, short-sighted, and irrational nature of the “the people”), other researchers show how populism is essentially an elite phenomenon (Herman & Muldoon, 2018; Mondon & Winter, 2019).

If we trace Mussolini's rise as a political figure, we see a situation where elite actors play a far more central role than electoral victories. The emergence of the fascists as a political force merged out of the political and economic turbulence after WWI. Here, economic elites, especially landowners in the Po Valley, turned toward protofascists organizations for protection against a wave of labor organizing and unrest (Paxton, 2007, pp. 73–86). When Mussolini was eventually appointed to prime minister, it was not as a product of a sweeping electoral victory, or the forceful seizure of power through the “march on Rome” as implied by fascist mythology, but rather through a series of deals with central actors in the economic and political establishment (Lyttelton, 2004, p. 94). Figures such as Franco in Spain or Horthy in Hungary likewise emerged against, or in spite of, electoral majorities and popular power. The best case for the rise of a fascist leader through the ballot box is of course Hitler. But here as well, his actual ascension to state power was not a product of an outright electoral victory, but rather by backroom dealings with representatives of factions of the traditional elite, notably the military, landowners, and sections of industry (Kershaw, 2014; Paxton, 2007). As Levitsky and Ziblatt (2018, p. 15) argue, historically, political elites have often helped fascists and semi-authoritarians into power by striking a “devil's bargain,” thinking they could control such political actors while simultaneously achieving short-term electoral gains. Given these examples, one could as reasonably have drawn very different lessons from the rise of fascism than the ones prevalent among constitutional architects of the postwar era. Instead of “disciplining” the popular masses, and creating a set of expert-run, antimajoritarian institutions, one could as reasonably have argued for expanding democracy and creating a set of institutions to “discipline” economic and political elites. When this did not happen, we argue that it is because in the liberal imaginary, threats to constitutional governments primarily stem from an excess of popular power.

What happens if we assume that the liberal approach to protecting democracy is really elitist and potentially antidemocratic? Does it necessarily follow that these protections are dangerous and unwanted? Might it not be the case that in order for democracy to function, it needs non- or even antidemocratic protections? Must democracies turn militant or die?

We will argue that this is not the case. Rather there are alternative, prodemocratic, or popular ways of conceptualizing the defense against authoritarian threats that potentially avoid some of the shortcomings associated with militant democracy. Hence we want to offer a perspective that can supplement to the militant model. But this requires a change of perspective. The liberal perspective, as we have argued above, primarily sees threats as stemming from either the state or the mob, leading to solutions involving divisions of power and antimajoritarian (often elite-led) institutions. But historically, other political traditions have conceptualized the problem differently. We argue that we can find the intellectual resources for a democratic notion of political self-defense in the popular republican and the socialist tradition.³ These traditions share the liberal skepticism toward state power and traditional hierarchies. But in contrast to liberalism, they identify the main threats to democracy as stemming from societal elites, rather than the masses. These perspectives on democratic self-defense can properly be understood as “anti-oligarchic” in nature, meaning that their main concern is defending democracy from being taken over by elites from within or outside of the state. This anti-oligarchic perspective aims to defend democracy in both direct and indirect ways. Directly, a popular, anti-oligarchic perspective aims to avoid unaccountable elites making devil's bargain with undemocratic forces in order to overturn democracy (as seen in Italy and Germany). Indirectly, a popular, anti-oligarchic perspective aims to drain away popular support for antidemocratic movements by keeping democracy more responsive to popular demands, less prone to elite takeover, and thus avoiding the sort of discontent that leads large numbers of people to turn toward antidemocratic movements. As Hannah Arendt argued already in The Origins of Totalitarianism (1951), a central precondition for tyrannical and totalitarian government is isolation and loneliness of individuals, as—in her vocabulary—these sentiments destroy the public realm of appearances and the distinctly political opportunity of acting in concert (Arendt, 1968, pp. 474–475). For our purposes, Arendt's argument points to the dangers of the depoliticizing effects of militant democratic and antimajoritarian models of democratic self-defense. By restricting the political realm and excluding groups or questions from the political process, these attempts risk further strengthening the sort of political isolation and alienation that draw people to authoritarian movements. Instead, the popular model of democratic self-defense developed below relies on the institutionalization of collective power as the principal mechanism of self-defense—that is, to counter individual isolation by creating public spaces, where acting in concert becomes politically possible. In short, if contemporary antidemocratic sentiments gain support from a critique of liberal democracy as elitist and exclusionary, it might be counterproductive to confront such antidemocratic sentiments with further depoliticization, as entailed in the militant model. What we lay out below are not institutional blueprints for a popular model self-defense but attempts at drawing alternative theoretical groundings for democratic self-defense from two political traditions, which allow for a shift in perspective from the liberal understandings of democratic self-defense.

It is one thing to argue that different traditions of democratic self-defense exist, it is another, more demanding task to argue that republican and socialist models of self-defense—what we group together as popular models of self-defense—can offer a viable supplement to militant democratic instruments given the contemporary crisis of democracy and rising right-wing authoritarianism.

Table 1 summarizes the analysis by distinguishing between two ideal typical models of democratic self-defense—the liberal and popular model. We are well-aware that real-world polities might productively combine instruments from the two models, and that upholding a democratic constitution might require attention to antidemocratic movements as well as to oligarchic elites. What we are interested in here is the difference between the models as models, that is, at the general and abstract level. Other interventions in this debate could productively explore specific, combined approaches to context-specific situations. While the main ambitions and primary instruments of each of the three strategies of democratic self-defense are explained in our analysis above, we want to highlight in Table 2 how each strategy of self-defense confronts three major problems intimately related to the defense of democracy, namely: (1) the election of a dictator, (2) the defense of minority rights, and (3) corruption of the state by elected politicians and bureaucrats.

In order to exemplify the popular defenses against the different threats outlined here, we have in addition to the theoretical discussion below also added some reflections on the contrasting practical implementation of the models.

First, the three strategies differ in their approach to the potential threat of an election of a dictator. All three approaches agree on the possibility of such an election, as the free, equal, and open nature of the democratic system makes possible the election of a dictator or a “would-be autocrat” as Levitsky and Ziblatt coin the term. The liberal approach of militant democracy is predominantly preemptive in its approach, insofar as militant democrats seek to counter “would-be autocrats” by restricting political rights and liberties in order to repress potentially subversive movements before they gain widespread popular support. Contemporary critics of populism often combine this approach with a call for stronger courts and softer instruments such as increased civic education in order to curb the extremist inclinations of ordinary voters. As such, such strategies somewhat shy away from political conflict by employing preemptive instruments of restriction and education. In contrast, the popular models are more reactive in their responses to the potential election of a dictator. Popular republicanism has employed political trails and ostracism and the socialist tradition has developed electoral recall as mechanisms of superseding dictators, hence relying on institutions of collective power in which a political conflict can take place rather than suppressing conflict and transposing it into the realm of law like militant democrats propose.

Our claim when it comes to preventing the election of a dictator is not that popular instruments of democratic self-defense would necessarily be more effective in the short term, as the institutional innovations we put forward might very well fail in protecting democratic institutions, as could be the case of the institutions of militant democracy. We argue, however, that by building on democratic empowerment, popular instruments might fail in a more productive way. Whether successful or not, the proposals we put forward would strengthen, rather than limit popular participation. As such, the reconstructed popular model might not be more successful at curtailing the rise of illiberal authoritarianism in countries like Hungary or Poland. A potential campaign to recall Viktor Orban or other Fides elected officials in 2010 might have worked no better at curtailing his rise than the current “European militant democracy” (Larsen, 2021). But the process of popular mobilization needed for such a campaign would have strengthened the potential for future challenges to the Fides regime, rather than giving Orban the role as defender of the popular will against unelected, foreign technocrats and jurists.

Second, the three strategies also agree that defending individual and minority rights is crucial for upholding a democratic regime. The reason to fear an election of an autocrat is among other things that individuals, groups, and minorities are at the mercy of such unlimited, sovereign power. Again, the liberal and popular models differ at the ideal typical level in their approach to this problem. Militant democracy is as argued above an offspring of what Müller has called a “self-disciplined democracy” (Müller, 2011, p. 125), namely, the restricted elite democracy that grew out of the experiences with fascism after WWII. Such “self-disciplined” democracies have primarily relied on antimajoritarian institutions for protecting individual and minority rights like supreme courts, constitutional courts, heightened electoral thresholds, and party bans. The popular strategies of democratic self-defense in republicanism and socialism disagree with the argument that antimajoritarian institutions are effective to protect the rights and freedoms of individuals and minorities. Because such approaches view the main threat as coming from elites, oligarchs, and the rich, they view popular institutions—that is, the institutionalization of collective power—as the most efficient way to protect rights and freedoms. By having institutions like the people's tribunes or workers’ councils individuals and minorities can be protected. These two approaches also testify to the different conception of political rights employed by the liberal and popular models of democratic self-defense in the first place. Although the liberal model understands rights as strictly individual hereby establishing a host of antimajoritarian institutions in order to curb the potential danger of mass politics, both republican and socialist models of democratic self-defense understand rights as social and as developed (and protected) through collective action. Hence, individual citizens and minorities need institutions of collective action to safeguard their rights, not only an abstract constitutional matrix as in the liberal model.

In a similar vein, the ideas of popular Tribunes or class-specific political institutions might not in themselves be ironclad guarantees for minority protection. A tribunate can, as laid out by McCormick, represent the great majority out of power, in his terms members of the bottom 90% of the income distribution that have not held elected office (McCormick, 2006). But it could also represent specific minorities who are un- or underrepresented in existing political frameworks, such as Dalits in India, Romas in East and central Europe, or residents without citizenship. The idea behind class-specific offices is to give marginalized minorities (or marginalized majority groups) direct access to independent power resources, rather than just right enforcement by courts and state elites. A controversial, contemporary case that illustrates the two models’ different approach to minority protection could be the potential decision by the U.S. Supreme Court to over Roe v. Wade (Liptak, 2022). While the militant model expects antimajoritarian institutions like supreme or constitutional courts to be the best guardian of individual rights, the popular model views institutions of collective power as the best safeguards of minority rights. In the case of abortion rights, the establishment of a “tribune of Women” could grant the relevant “minority” certain veto powers in order to protect their rights.

The third problem, which both liberal and popular models of democratic self-defense confront, is how to challenge the corruption of the state—that is, the turning of the state institutions toward the personal ends of elected magistrates and unelected bureaucrats. The liberal model understands such corruption through the classic prism of the tyranny of the majority. As representative government and universal suffrage has introduced the masses into politics, the chief danger is the new form of tyranny practiced not by the singular tyrant, but the electoral majority, using the institutions of the state to their own enrichment on the expense of the commonwealth. As we have seen, this fear is integral to liberalism, as in the classical liberalism of 18th and 19th centuries, the fear was that the poor would threaten the property of the rich as well as their wealth through increased taxation. As argued by Duncan Bell, this fear of the poor was transformed into a fear of the people or the electoral majority, as 20th totalitarianism was interpreted through the liberal schema. Popular models of democratic self-defense, instead, conceptualize the potential corruption of the state as stemming not from the majority, but instead from oligarchs and elites extending their personal economic power into public institutions. Such approaches provide institutional mechanisms for combating oligarchic influence by making elites accountable and removable. One contemporary example of the popular model's attempt to prevent the threat of the corruption of the state might be the spread of sortition-based climate assemblies in European countries such as Ireland, France, Scotland, England, and Denmark. If we bracketed for moment the fact that such sortition-based climate assemblies are often created by the state as a symbolic act without much legislative power (Mulvad & Popp-Madsen, 2021), such assemblies could be interpreted as a second (legislative or consultative) chamber consisting of ordinary, nonelite citizens charged with policy making on an issue, which political elites have not been able to confront. While the population at large in many European countries are in favor of a green transition and have to a certain degree voted accordingly, political elites have not responded with the conviction. Such inability to confront substantial threats to biodiversity has been discussed as a product of an inherent flaw of liberal democratic systems (Blühdorn, 2013). From the perspective of the popular model of democratic self-defense, the central problem is that organized interests representing carbon-based parts of the capitalist class are able to complicate democratic decision-making regarding climate change (Klein, 2015). The popular model's response to the inability and disinterest of the political elites, inspired by popular republicanism and the plebeian assemblies imbued with veto powers, is to advocate for the institutionalization of collective power—here exemplified in sortition-based assemblies—where ordinary citizens have the chance of preventing the capture of the common good by the special interests of the elites.

Our claim is not that the popular models provide guarantees against the undermining of democracy by a determined executive with consistent popular support over time (just as the liberal models provide no such guarantees, as have been demonstrated by recent events in Hungary or Poland). What we argue instead is that the popular model politicizes different forms of partisan cleavages. Militant modes of self-defense posit a conflict between a popular movement against political elites such as representatives, judges, or other unelected magistrates, potentially politicizing the cleavage between an authentic people and technocratic elites, through which authoritarian populist projects tend to thrive. Alternatively, the popular models’ reliance of countermeasures based on institutionalized popular contestation will move the political struggle to other terrains, less favorable for such authoritarian populist agitation. Contemporary populists operate with an anti-institutional and anti-pluralist interpretation of popular sovereignty, which essentially leads to an anti-representative understanding of democracy—the populist leader knows what the “true people” wants, even though electoral results show something different (Müller, 2016a). In contrast, the popular model rests on a plurality of institutional, representative, and delegatory mechanisms, run by the citizenry itself, making the populist rhetorical strategy of the corrupt elite versus pure, unified people difficult to sustain.

Recent years have seen almost unparalleled electoral success of predominantly right-wing populism across both the global North and global South. Such populist parties have enjoyed electoral success, and even election into political office, due to a deep-seated dissatisfaction with the political status quo, and the perceived inability of liberal democracy to tackle the political issues of the day. One response by critics of contemporary populism has been to revitalize the discussion of Loewenstein's idea of militant democracy, in which Loewenstein proposed the creation of a “constrained democracy,” in which political extremism would have difficulties to emerge, as uncontrollable popular passions were held in check by clever, depoliticizing institutional design meant to undermine parliamentary sovereignty (Müller, 2011, pp. 146–150). The core concern of a militant democratic model of self-defense is that undemocratic movements and parties can use the free, equal, and open electoral process to undermine the democratic regime from within. Contemporary militant democrats, though, shy away from the quasi-authoritarian commitments of Loewenstein's original proposal, and often subscribe to what Jan-Werner Müller has called a “soft” version of militant democracy that instead of outright banning extreme parties seeks to limit their room of maneuver.

While we agree with the general conviction that democracy needs defense mechanisms against antidemocratic or authoritarian forces, we argue that the proposals put forward by militant democracy should not stand along but can be productively supplemented with instruments and institutions from what we call popular models of democratic self-defense, resources to which can be found in republican and socialist traditions. We argue that liberal models of democratic self-defense such as militant democracy are normatively elitist, and exclusionary and empirically potentially ineffective approaches to democratic self-defense. Moreover, such strategies are not momentary, extraordinary deviations from the ideal of liberal democracy in which they are exercised, but instead militant democracy radicalizes already existing commitments of liberal “constrained democracy,” including the fear of popular subjects and its individualizing solutions to problems of political extremism. However, because liberal models of democratic self-defense rely on elite institutions such as constitutional courts, representative bodies, and nonelected bureaucrats, they have a tendency to demobilize the populace and restrict popular participation. As such, they potentially weaken the long-term resilience of democracy. Instead, we argue that a popular model of self-defense can offer potential solutions on several levels.

First, on the analytical level, the popular approach to democratic self-defense offers a different lens through which to view the issues of democratic backsliding, authoritarianism, and the populist resurgence. By conceptualizing the threats of authoritarian takeover as predominantly emanating from the elite, the popular model offers a broader array of potential answers to the defense of democracy. Second, on a political level, the popular model offers modalities of collective practices through the popular institutions that empower ordinary people rather than elites. This has the direct effect of strengthening the long-term resilience of democracy, and the indirect effect of not provoking the sort of populist backlash—the cleavage between “the elite” and “the people” that populists successfully campaign on—that reliance on elite institutions can have. As argued by Arendt in The Origins of Totalitarianism, for example, one central precondition for the success of totalitarianism was its ability to spellbound already isolated and atomized individuals. The popular model of democratic self-defense works to counter such atomized individualism by relying on the institutionalization of popular power as the modus operandi of combating political extremism. Third, on a normative level, the popular approach offers a model of democratic self-defense that is based on broadening rather than limiting channels of popular participation. As such, it is a form of democratic self-defense that offers a deepening rather than a restriction of political participation.

The article does not lay out a full program or institutional blueprint for the construction of a popular model of self-defense, and much work still needs to be done in order to transform these ideas into a functioning institutional setup. Instead, the article offers a shift in perspective, where threats to democracy are conceptualized as stemming not only from the popular masses, but also from political elites, wealthy oligarchs, “would-be autocrats,” and aspiring tyrants. In order to offer such a perspectival shift, we draw on resources from the nonliberal parts of the democratic tradition, namely, popular republican and socialist imaginaries. These traditions, unencumbered by the fear of popular politics integral to liberal modes of democratic self-defense, allow for a more democratically robust defense against threats to democratic institutions and individual rights. In these imaginaries, we discover concrete and historically tested institutional mechanisms—such as electoral recall, imperative mandate, sortition, political trails, and class-specific offices—that have placed the citizenry as such, and not just its representatives or nonelected elites, at the center stage of democratic self-defense.

Authoritarianism is certainly on rise across the West and around the globe. The timely and paramount question is: What to do about it? The core, normative conviction of the popular model of democratic self-defense that we have developed in this article is that the citizenry ought to have ways of continually testing the legitimacy and support of their elected political leaders. An election every fourth, fifth, or sixth year is simply too long of an interval, as this gives “would-be autocrats” and aspiring tyrants way too much time to cripple the courts, manipulate the electoral laws, gerrymander electoral districts, restrict minority rights, and change the constitution. The popular model of democratic self-defense, instead, provides institutional mechanism through which the citizenry can confront their political leaders the moment they begin to undermine the freedom of the polity. This does not imply, though, that if the popular model was in place, then “would-be autocrats” would never be successful in dismantling democracy. But it at least ensures that an institutionalized political conflict can arise on whether this or that change to the constitutional setup of the polity is legitimate or not. A noninstitutionalized mode of confronting tyrannical rulers would be that of revolution—John Locke's famous “appeal to heaven.” The popular model of democratic self-defense thus strikes a balance between an “overpoliticizing,” noninstitutionalized revolution as a mode of overthrowing tyrannical government and the depoliticized, bureaucratized attempt to preclude tyrannical government from emerging, as implied by militant democratic approaches. By having regular and constitutionally secured institutions of popular power and control, the popular model of self-defense is able to confront attacks on the democratic constitution continually and as they happen. This conceptual shift of perspective, we argue, can prove of vital importance to democratic forces in the coming times of political uncertainty.