1. Introduction
Despite having left a rich oeuvre in the field of legal philosophy and constitutional theory Ernst-Wolfgang Böckenförde is, among theorists, mostly known for one sentence: ‘The liberal, secularized state is sustained by conditions it cannot itself guarantee’ (Böckenförde Reference Böckenförde2020a: 167). A seemingly simple reply to a rather complex problem: whether a liberal, secularized state requires some form of pre-political grounding or ethical substance (‘sittliche Substanz’) to sustain itself. His answer came to be known as the Böckenförde dictum, dilemma or paradox. On the one hand, Böckenförde argued, a state dedicated to liberty ‘can only survive if the freedom it grants to its citizens is regulated from within’ (Böckenförde Reference Böckenförde2020a: 167). Citizens need to be willing to support and sustain the liberal state that they are part of; they need a certain disposition or attitude. Religion can be a source for such an ethos but it is certainly not the only one. On the other hand, however, the liberal state cannot resort to ‘the instruments of legal coercion and authoritative command’ to provide such an ethos ‘without abandoning its liberalness’ (Böckenförde Reference Böckenförde2020a: 167). That is why the full quote with which this article started is: ‘The liberal, secularized state is sustained by conditions it cannot itself guarantee. That is the great gamble it has made for the sake of liberty’.
Interestingly enough, a somewhat related paradox as Böckenförde formulated in 1964 can be found in the writings of Immanuel Kant. Addressing Kant’s principle of publicity Howard Williams discusses the question whether an ‘effective sovereign must rule in the spirit of freedom’ (Williams Reference Williams1983: 151). Kant’s answer is positive according to Williams. In On the Common Saying Kant writes: ‘In every commonwealth there must be obedience under the mechanism of the state constitution to coercive laws (…), but there must also be a spirit of freedom, since each, in what has to do with universal human duties, requires to be convinced by reason that this coercion is in conformity with right, lest he fall into contradiction with himself’. (TP 8:306) Discussion and argument in that sense can create consent and consent can lead to a stable form of obedience to law and sovereign. Williams, however, claims that there is a weakness in this line of argumentation. Kant assumes that the sovereign is of good will and would, therefore, not wrong his citizens (Williams Reference Williams1983: 151).Footnote i But what if reality turns out to be different? What if protesting citizens face a tyrant who is unwilling to change the law to their benefit or guarantee their right to criticize? In such a situation the public use of reason will be quite ineffective. In those circumstances, in other words, where the right to criticize government is most needed, no resources are available to keep it alive (Williams Reference Williams1983: 157). At this point, Williams argues, we are confronted with a paradox. Liberalism ‘presupposes a settled, mature society which encourages argument and debate, and where this condition is not realized it is forced into silence and political impotence (…)’ (Williams Reference Williams1983: 157).Footnote ii
To what extent, however, is this really an either-or situation? Either there is a ‘settled, mature society’ or there is ‘political impotence’? Either the sovereign is willing to effectively rule in the spirit of freedom or citizens are ‘forced into silence’? Or is Böckenförde right that a liberal state cannot guarantee the conditions that sustain itself, but can try to secure them? His paradox, after all, is often misunderstood to imply that the state cannot create the conditions needed to sustain itself. But that is not what Böckenförde actually claimed. He wrote ‘by conditions it cannot itself guarantee’. The state, according to Böckenförde, is capable of supporting and protecting the necessary ethos and basic convictions (Böckenförde Reference Böckenförde2020b: 379). But where can such a ‘sustaining and supportive foundation’ be found (Böckenförde Reference Böckenförde2020c: 200)? Is a form of shared nationality or shared religion needed to protect liberal societies or can the shift towards an ethical state – a liberal state sustained by a living common ethos among its citizens and officials – also take place ‘in and through the democratic political process’, in the interplay between active citizenry and representatives, and by a democratic ethos sustained and affirmed on the basis of liberal democratic principles as Böckenförde and later Jürgen Habermas have argued (Böckenförde Reference Böckenförde2017: 106). These questions become all the more relevant in an age in which we are witnessing a global rise in autocratic governance and liberal democracies that are increasingly being threatened by internal ‘illiberal forces’ (Freedom House 2022: 1).
Solving Böckenförde’s dilemma and Williams’s paradox of liberalism is too big a task to accomplish in one article. Not only regarding the available length of this paper, but also because both paradoxes are very similar in certain regards but also importantly different in others. Böckenförde’s paradox, for example, presupposes the willingness of political representatives to rule in the spirit of freedom, whereas Williams’s paradox does not. Böckenförde’s paradox poses the question how a state can secure the conditions under which citizens are willing to effectively use their public reason, whereas Williams’s paradox also raises the question of how the conditions can be secured under which the sovereign is willing or perhaps forced to rule in the spirit of freedom. Solving Böckenförde’s paradox, in that sense, might imply solving Williams’s paradox and perhaps also the other way around.
My aim in this article is more limited: to use both paradoxes as a framework to critically analyse the current theoretical literature that deals with a specific instantiation of this global increase in autocratic governance – how liberal democratic societies can defend themselves against ‘elected public authorities [that] deliberately implement governmental blueprints which aim to systematically weaken, annihilate or capture internal checks on power with the view of dismantling the liberal democratic state and entrenching the long-term rule of the dominant party’ (Pech & Scheppele Reference Pech and Scheppele2019: 10). I argue that the current theoretical literature is too strongly focused: (a) on the problem of militant democracy (thereby neglecting the promise of its counterpart, the concept of militant constitutionalism) and (b) on the formal institutional safeguards that are needed to make liberal democracies resilient (thereby partially neglecting what Böckenförde calls ‘the ethos of modern democracy’ (Böckenförde Reference Böckenförde2020d: 61-76). In addition (c), I claim that such an ethos not only implies a democratic attitude but also a patriotic or constitutional attitude.Footnote iii This article has therefore two main goals: to sketch the main framework of a theory of militant constitutionalism (1) and to indicate how liberal democracies can create the democratic and constitutional ethos necessary to sustain themselves (2).
Section 2 analyses the concept of militant constitutionalism on the basis of a comparison with militant democracy. Section 3 analyses which preconditions need to be met to make liberal democracies more resilient and stable. It first of all uses Jürgen Habermas’s discourse theory of law and democracy to sketch how democratic processes and practices can function as a source for political motivation. Second, it uses Jan-Werner Müller’s interpretation of constitutional patriotism (constitutional patriotism as containing an element of ‘militancy’) to argue for the importance of a constitutional attitude or constitutional ethos. Section 4 addresses some concluding remarks and returns to the broader context of the notion of ‘the paradox of liberalism’. With regard to liberal democracies, it will argue that Böckenförde’s paradox can be relatively successfully addressed, whereas Williams’s paradox can only be partially solved.
2. Militant democracy and militant constitutionalism
There is a growing body of work on democratic regress and rule of law backsliding. According to the Global State of Democracy Report 2023 – a yearly report compiled by the Institute for Democracy and Electoral Assistance – the currently ‘most worrying political trend’ is ‘the erosion of checks and balances and constitutional freedoms in nominally democratic societies’ (Institute for Democracy and Electoral Assistance 2023: 4). Examples of this erosion can be found in Poland, Hungary, the United States and Slovakia. 2022, according to the report, ‘was the sixth consecutive year in which more countries experienced net declines in democratic processes than net improvements’ (Institute for Democracy and Electoral Assistance 2023: 8). According to Freedom in the World 2023 – the main annual report of Freedom House – this period of decline is even longer. In 2022 ‘[t]he gap between the number of countries that registered overall improvements in political rights and civil liberties and those that registered overall declines was the narrowest it has ever been through 17 consecutive years of deterioration’ (Freedom House 2023: 2). The threat from ‘incumbent leaders’ is of specific concern: ‘assuming office through elections, these leaders rejected the established democratic process and sought to rewrite the rules of the game to maintain their grip on power’ (Freedom House 2023: 5) The World Justice Project, finally, has a similar warning in its 2023 report: ‘over 6 billion people now live in countries where the rule of law is declining. Amid widespread institutional stagnation, a majority of countries continue to experience rule of law backsliding characterized by executive overreach, diminished human rights, and justice systems that are failing to meet people’s needs’ (World Justice Project 2023: 8).
Not a new question but still a hugely important one, is how liberal democracies can deal with these challenges. Within political and legal theory, this question received considerable attention from the 1930s onwards. Over the last decade, however, interest has increased leading to numerous studies on resilience. Strikingly enough, however, these studies mainly focus on one of the two pillars of constitutional democracy, democracy, while the other pillar, constitutionalism and rule of law (both principles for limiting and binding governmental power), has received far less theoretical and philosophical attention. While it is true that within legal and political science much is written about the problem of rule of law backsliding and constitutional retrogression, a theory of militant constitutionalism – as the necessary counterpart of the theory of militant democracy – is seldom discussed. Legal theory, according to András Sajó, is short of an answer to the problem of the ‘self-incapacitating tendency of the rule of law’; respect for the rule of law can block measures ‘to restore the rule of law’ and change perverted legal systems (Sajó: Reference Sajó2023a). Even the notion ‘militant constitutionalism’ (or ‘militant rule of law’) is rarely used; exceptions are Andras Sajó, Jerg Gutmann and Stefan Voigt.Footnote iv One of the goals of this article is, therefore, to sketch the main framework of such a theory. To accomplish this goal, the idea of militant constitutionalism will be reconstructed (§2.2) on the basis of a comparison with the idea of militant democracy (§2.1). Important to emphasize at this point is that militant constitutionalism does not replace militant democracy, both are needed to make liberal democracies more resilient.
2.1 Democracy and resilience
‘Militant democracy’ was first used by German constitutional law scholar Karl Loewenstein in the mid-1930s. In two articles, ‘Militant Democracy and Fundamental Rights, I’ and ‘Militant Democracy and Fundamental Rights, II’, Loewenstein described how European democracies threatened to turn into autocracies and why the Weimar Republic could not defend itself against the rising National Socialist dictatorship. ‘Democracy and democratic tolerance’, according to Loewenstein, ‘have been used for their own destruction’ (Loewenstein Reference Loewenstein1937: 423). The cause, he argues, lies in a strong form of relativism and formalism: ‘It is the exaggerated formalism of the rule of law which under the enchantment of formal equality does not see fit to exclude from the game parties that deny the very existence of its rules’ (Loewenstein Reference Loewenstein1937: 423).
Within legal and political theory, an interesting example of this defenselessness against democratic self-annulment can be found in the work of a contemporary of Loewenstein: in Hans Kelsen’s justification and procedural interpretation of democracy. According to Kelsen, a strong link exists between democracy and relativism. The antagonism between philosophical absolutism and philosophical relativism, he claims, is ‘in many respects analogous to the fundamental opposition between autocracy and democracy as the representatives of political absolutism on the one hand and political relativism on the other’ (Kelsen Reference Kelsen1948: 906).Footnote v If absolute or objective moral values are inaccessible to us, we should not only respect our own political opinions and beliefs but also those of others as equally valuable (Kelsen Reference Kelsen2006a: 241).Footnote vi All political actors, in other words, should have an equal voice, and there is no standard for determining whether one voice carries more weight, value, or truth than another. With regard to the dangers of autocracy, Kelsen’s procedural conception of and epistemological argument for democracy – Loewenstein prefers to speak of ‘democratic fundamentalism’ (Loewenstein Reference Loewenstein1937: 424) – leads to the harsh conclusion that a democracy ceases to be a democracy when it tries to defend itself against a people that wants to abolish it. To even pose the question whether a democracy should be defended against an undemocratic majority is already to deny it. ‘Man muss seiner Fahne treu bleiben’ writes Kelsen, ‘auch wenn das Schiff sinkt; und kann in die Tiefe nur die Hoffnung mitnehmen, dass das Ideal der Freiheit unzerstörbar ist und dass es, je tiefer es gesunken, um so leidenschaftlicher wieder aufleben wird’ (Kelsen Reference Kelsen2006b: 237). Democratic self-destruction is, therefore, a real possibility.
A similar thought is expressed by an important contemporary of Kelsen, German legal scholar Gustav Radbruch. Like Kelsen, Radbruch argues that the value and justification of majoritarian democracy follow from a combination of moral relativism and political equality. And like Kelsen, Radbruch recognizes the problem that relativism can lead to self-destruction: ‘The freedom to renounce freedom is contained in the idea of freedom itself. Therefore, a dictatorship can be established in democratic forms’ (Radbruch Reference Radbruch1990: 20). Unlike Kelsen, however, Radbruch argues that in doing so we are not confronted with an unsolvable contradiction (Radbruch Reference Radbruch1990: 20). Although relativism implies a general form of tolerance, it does not imply tolerance in the face of intolerance. Given that absolute opinions and absolute decisions are incompatible with relativism democratic societies should not be allowed to abolish themselves once and for all: ‘relativism can tolerate any opinion – except the opinion that claims to be absolute. Hence, the attitude of the democratic state towards anti-democratic parties’ (Radbruch Reference Radbruch1990: 21).
Both Radbruch and Kelsen, in that sense, link relativism, individual freedom and political equality (Kelsen’s second argument for majoritarian democracy) to the value, justification, and resilience of democracy. Radbruch, however, claims that defensive measures against anti-democratic parties can be justified: democracy can be ‘militant’. But how to define militant democracy? A general definition of militant democracy is given by Jan-Werner Müller. Militant democracy, he writes, ‘refers to the idea of a democratic regime which is willing to adopt pre-emptive, prima facie illiberal measures to prevent those aiming at subverting democracy with democratic means from destroying the democratic regime’ (Müller Reference Müller, Rosenfeld and Sajó2012a: 1253). A less ‘militant’ definition is given by Samuel Isscharoff. It refers to ‘the ability of democratic regimes to restrict forms of debate, political organization, or political participation that pose an existential threat to democracy itself’ (Issacharoff Reference Issacharoff2015: 18).
What particular conception of democracy is presupposed by a theory of militant democracy is not a question we need to address here. Important, however, is to indicate that the current literature on militant democracy emphasizes that a coherent theory of militant democracy presupposes a substantive conception of democracy and not a formal or purely procedural one. It makes little sense, it is claimed, to call for militant measures if democracy is merely a decision-making procedure in which the majority principle determines which decisions are legitimate and which are not. According to formal conceptions, democracy is a neutral procedure for arriving at political decisions. Democracy constitutes ‘a “marketplace of ideas,” in which all ideas are equal’ (Rijpkema Reference Rijpkema2018: 8). According to substantive conceptions democracy is more than just a procedure, it also contains substantive moral values – like equality or mutual respect. Thus, not all ideas and opinions are equal.
Important to emphasize at this point is that not all forms of democracy that include a commitment to substantive moral values are by definition militant in the relevant sense as has been discussed above. The main goal of militant democracy is to prevent the destruction of democratic regimes by democratic means and not simply to prevent unjust outcomes from being the result of democratic processes. Although all substantive conceptions of democracy will limit the outcomes of the process of democratic opinion- and will-formation their main goal is not simply to ‘restrict forms of debate, political organization, or political participation that pose an existential threat to democracy itself’.Footnote vii Militant democracy, in that sense, is aimed at a specific goal and proposes specific means. Among the means offered to defend democracy are party bans, the restriction of political freedoms and democratic debate. Means whose main aim is to identify the enemies of democracy ex ante and prevent them from coming to power.
2.2 Rule of law and resilience
The ‘enemies of liberal democracy’ – as the authors of the Freedom House report of 2022 put it rather firmly (2022: 1) – focus their arrows not only on democratic governance but also on the rule of law; both can be attacked or fall into disrepair. This makes it highly relevant to also ask the conceptual question of what the rule of law is, the normative question of what value the rule of law has, and the political-legal question of how to make the rule of law more resilient. A coherent theory of militant constitutionalism should address all three questions. In this section, however, I merely describe the main framework of a theory of militant constitutionalism. I will do so by starting with the problem that a theory of militant constitutionalism should address. Following Jeremy Waldron’s characterization of the rule of law as a ‘solution-concept’ instead of an ‘achievement-concept’, I argue something similar with regard to the concept of militant constitutionalism: ‘it is the concept of a solution to a problem we’re not sure how to solve; and rival conceptions are rival proposals for solving it or rival proposals for doing the best we can in this regard given that the problem is insoluble’ (Waldron Reference Waldron2002: 158).
What is the problem that a theory of militant constitutionalism tries to solve? Different answers are possible. I will briefly mention two. The first deals with the means that may be used by governments to deal with national emergencies, like the September 11, 2001, attacks. How to prevent ‘politicians from exploiting momentary panic to impose long-lasting limitations on liberty’ (Ackerman Reference Ackerman2006: 3)? How to protect liberal democracies, in other words, against domestic and foreign terrorists without sacrificing the fundamental importance of civil liberties. The democratic paradox that proponents of a theory of militant democracy try to solve – ‘the possibility of a democracy destroying itself in the process of defending itself’ (Müller Reference Müller2012b: 536-7)Footnote viii – has its counterpart in a constitutional paradox: ‘the possibility of a constitution destroying itself in the process of defending itself’. The second answer, the one that is the focus of this article, is the rise of illiberal democracies, specifically democratic and rule of law backsliding in liberal democracies. To rephrase the quote I cited earlier, the focus of militant constitutionalism is then on defending liberal democracies against ‘the process through which elected public authorities (…) implement governmental blueprints which (…) systematically weaken, annihilate or capture internal checks on power’.
Although both militant democracy and militant constitutionalism are aimed at protecting liberal democracies, three differences can be distinguished: differences dealing with timing (1), the choice of means (2), and actors involved (3). Whereas militant democracy – Gutmann and Voigt argue with regard to the issue of timing – mainly aims to prevent non-democrats from gaining power, militant constitutionalism also becomes relevant after populist politicians come to power (Gutmann and Voigt Reference Gutmann and Voigt2023). The timely identification and containment of the ‘enemies’ of constitutional democracy – before they seize power – is, after all, not always possible. The goal of militant constitutionalism is to curb their power and prevent damage to legal and political institutions. Sajó makes a similar claim: ‘Once populists grab power, the issue is how much resistance can be provided by mechanisms of constitutional self-defense institutions’ (Sajó Reference Sajó, Malkopoulou and Kirshner2019: 187).
The appropriate means, second, are not party bans or restrictions on political freedom – a ‘toolkit of rights restrictions limiting the possibility that putative enemies of democracy will gain power’ (Sajó Reference Sajó, Malkopoulou and Kirshner2019: 193) – but separation of powers, limitation of terms of office, curtailment of government power to make executive orders, the constitutional protection of judicial independence, a strong form of constitutional review, the constitutional enshrinement of fundamental rights and liberties, and perhaps ‘eternity clauses and doctrines of unconstitutional constitutional amendments’ (Suteu Reference Suteu2021: 264)Footnote ix; so ‘measures that would limit the possibilities of abuse of government power, (…) aimed at perpetuating the government’ (Sajó Reference Sajó, Malkopoulou and Kirshner2019: 193). With regard to the actors charged with implementation, finally, militant constitutionalism is not only aimed at actors within the government but also ‘actors well beyond the government’ (Gutmann and Voigt Reference Gutmann and Voigt2023: 379), like NGOs or the media. These actors need to be ‘protected by the constitution against would-be autocrats’ (Gutmann and Voigt Reference Gutmann and Voigt2023: 381) but they also share in the responsibility of containing the power of would-be autocrats, for example by criticizing power abuse.
Concerning the conceptual question of what the rule of law is and the normative question of what value the rule of law has, this section needs to be brief. Formal and substantive conceptions of rule of law can be distinguished: ‘formal theories focus on the proper sources and form of legality, while substantive theories also include requirements about the content of the law (usually that it must comport with justice or moral principle)’ (Tamanaha Reference Tamanaha2004: 92). That more is at stake than just a conceptual or linguistic issue becomes clear when we look at the following statement by Joseph Raz: ‘A non-democratic legal system, based on the denial of human rights, on extensive poverty, on racial segregation, sexual inequalities, and racial persecution may, in principle, conform to the requirements of the rule of law better than any of the legal systems of the more enlightened Western democracies’ (Raz Reference Raz1979: 211). Those who believe that the rule of law only requires that governments act on the basis of laws and that law must meet certain formal conditions (such as being general, clear and consistent) will not be surprised by this. But those who believe that the rule of law is incompatible with racism, sexism and unequal treatment must show that the rule of law contains a moral core that at least affirms the moral equality of all regardless of race, color, sex, gender, religion etc. And it is precisely this moral core that makes the question of the resilience of the rule of law so relevant and fascinating. Militant constitutionalism based on a substantive conception of rule of law will entail a broader scheme of defensive measures than militant constitutionalism based on a formal conception.
Unlike a theory of militant democracy, however, a theory of militant constitutionalism need not automatically presuppose a substantive conception. Formal legality, after all, need not be a morally neutral instrument that can be used for both good and bad (Raz Reference Raz1979: 226). It can contain moral value. Given the focus of this article on internal ‘illiberal forces’ threatening liberal democracies, however, a substantive conception of rule of law – based on the importance of individual rights and democracy – will form the background of the following sections.
Independent arguments – not directly related to the usefulness of a substantive notion for developing a theory of militant constitutionalism – can, of course, be given for the value of such a substantive notion. Habermas, whose work on law and democracy plays an important role in the following sections, develops a convincing account of some of these arguments in Faktizität und Geltung (1992). Following Kant’s theory of right, Habermas’s discourse theory of law and democracy starts from the internal connection between the ideas of right and state. The organizing, sanctioning, and executive powers of the constitutional state are ‘not just functionally necessary supplements to the system of rights’ according to Habermas ‘but implications already contained in [these] rights’. Political power, he argues, is both ‘presupposed by law’ and ‘constituted in the form of basic rights’ (Habermas Reference Habermas1996: 134). But unlike Kant, Habermas denies that law derives its full normative sense from a priori moral premises. Legitimacy can only be reached through a democratically organized procedure of lawmaking (Habermas Reference Habermas1996: 135). Rule of law and democracy, in other words, are internally related.
Notwithstanding such an independent argument for a substantive notion of the rule of law, it is important to keep in mind, as Sajó rightly indicates, that a simplistic understanding of the rule of law can block necessary actions against illiberal politicians and illiberal democracies. Respect for the rule of law can operate as a barrier for ‘measures to restore the rule of law’ (Sajó Reference Sajó2023a). The rule of law, furthermore, ‘can be easily highjacked and turned into rule by law; it is also a technique of those who grabbed state power lawfully to impose their will in a seemingly legitimate manner’ (Sajó Reference Sajó2023a).
3. Coercion in conformity with right
What means are available to defend liberal democracies and make them more resilient? In the previous section, we saw that militant democracy and militant constitutionalism recommend two different toolkits. The first a ‘toolkit of rights restrictions’ that is aimed at ‘limiting the possibility that putative enemies of democracy will gain power’ (Sajó Reference Sajó2023a). The second a toolkit of formal legal safeguards aimed at limiting ‘the possibilities of abuse of government power, (…) aimed at perpetuating the government’ (Sajó Reference Sajó, Malkopoulou and Kirshner2019: 193). Both toolkits are needed to give a partial answer to the two paradoxes that we started this article with. Militant democracy and militant constitutionalism precisely show that liberal democracies need not be based on the assumption that the sovereign is of good will and ‘would not consciously wrong the citizens of his state’ (Williams Reference Williams1983: 156). Liberal democracy, in fact, is partly based on the idea that the exercise of power is bound by the rules and principles of law through a constitution. As Krygier argues: ‘(t)empering power’ is the point of the rule of law, ‘the arbitrary exercise of power’ its specific antipathy (Krygier Reference Krygier2019: 760). The constitution in that sense contains limitations to the power of the government and democratic majorities. Kant’s claim that the ‘civil constitution in every state shall be republican’ embodies this thought. Republicanism, according to Kant, ‘is the political principle of separation of the executive power (…) from the legislative power; despotism is that of the high-handed management of the state by laws the regent has himself given, inasmuch as he handles the public will as his private will’. (TPP 8:349, 8:352) This principle, moreover, is part and parcel of Kant’s larger program of political change. This program, Williams argues, ‘requires that societies gradually become republics with full separation of powers, leading ultimately to a peaceful worldwide society’ (Williams Reference Williams2003: 162).
Institutional safeguards for tempering and binding power provide necessary but not sufficient means to make liberal democracies stable and resilient. Legal and political institutions can only fulfill their protective role insofar as they are sustained and supported by officials and citizens alike; insofar as democracy and rule of law also have a cultural and societal anchorage. With regard to constitutions – even republican constitutions – it is easy to see that they ‘are (…) only pieces of paper that derive their power from the intersubjective views of elites and citizens’ (Huq & Ginsburg, Reference Huq and Ginsburg2018: 68). Institutional safeguards and constitutional design (A) are important for constitutional self-defense, but also a constitutional culture and ‘level of the entrenchment of the constitution in society and politics’ (B) (Jonssen Cornell Reference Jonssen Cornell2021: 2 and 12).
Although increasing attention is being paid within the domain of militant democracy to the need for a broader pallet of democratic defense tools (Müller Reference Müller, Rosenfeld and Sajó2012a: 1262)Footnote x – ranging from party bans and political rights restrictions to stressing the importance of public deliberation and civic education – the main focus of the current theoretical work is still on (A). More needs to be said, consequently, about the political and constitutional culture and rights practices that are required to sustain and protect liberal democracies (B). Habermas’s discourse theory of law and democracy provides a good starting point for both. In section 3.1 I argue that Habermas’s theory not only shows how ‘obedience under the mechanism of the state constitution to coercive laws’ can be combined with a ‘spirit of freedom’ (TP 8:306) but also entails a partially convincing response to the Böckenförde Paradox. ‘Partially’ because Habermas’s notion of constitutional patriotism in its current form is primarily focused on the ‘democratic quality of political culture’ and not on the defense of liberal democracy (Müller Reference Müller2017: 30). Section 3.2 will, therefore, address the question whether constitutional patriotism should also include a protective constitutional ethos or attitude.
3.1 In and through the democratic political process Footnote xi
Böckenförde argues that a liberal, secularized state as a liberal state cannot seek to guarantee the necessary ‘inner regulatory forces’ on the basis of ‘the instruments of legal coercion and authoritative command’ (Böckenförde Reference Böckenförde2020a: 167).Footnote xii It would otherwise destroy the state ‘as the order of liberty’ (Böckenförde Reference Böckenförde2020a: 167). But what others means are available for the state to support and protect such a shared ethos? Böckenförde mentions, among else, education in school (‘obligatory classes on ethics in school – alongside religion classes, not instead of them’), committing state financed public institutions (like public broadcasting) ‘more strongly to a cultural and educational mission’ and finally making sure that government officials abide by and effectively realize the law (Böckenförde Reference Böckenförde2020b: 380). Moreover, he seems to suggest that the shift towards an ethical state can only take place ‘in and through the democratic political process’ and by a democratic and constitutional ethos sustained and affirmed on the basis of liberal democratic principles (Böckenförde Reference Böckenförde2017: 106). Whether this is actually the case, is a difficult matter. Künkler, for example, notes that: ‘Böckenförde did not believe that joint participation in the democratic process alone was sufficient for this agreement to emerge and sustain itself’ (Künkler Reference Künkler2020: 34). For the line of argumentation in this and the next sections, however, we don’t need to answer this question.
What is important, though, is to look at Habermas’ response to Böckenförde’s paradox. His response can, for instance, be found in the conversation on the ‘pre-political moral foundations of a free state’ (Schuller Reference Schuller, Habermas and Ratzinger2006: 15) that he had with Joseph Ratzinger. Habermas does not immediately dismiss the paradox out of hand. On the one hand, he claims that the liberal state ‘depends in the long run on mentalities that it cannot produce from its own resources’ (Habermas Reference Habermas2008: 3). Citizenship ‘is “embedded” in a civil society that is nourished by spontaneous and, if you will, “prepolitical” sources’ (Habermas Reference Habermas2008: 105). On the other hand, this dependence on ‘prepolitical’ sources does not imply ‘that the liberal state is incapable of reproducing the motivations on which it depends from its own secular resources’ (Habermas Reference Habermas2008: 105). The legal order can legitimate itself in ‘a self-reflexive fashion through a democratic legal procedure’ (Habermas Reference Habermas2008: 104) and does not need to be grounded in a ‘shared religious background, a common language, [or] (…) national consciousness’ (Habermas Reference Habermas2008: 106). Democratic processes, in short, can help sustain the liberal state without having to rely on religion, national identity or a ‘Leitkultur’ (Müller Reference Müller2018: 202).
What kind of democratic processes are required to sustain liberal democracies and make them more stable and resilient? The legitimacy and persuasive power of democratic processes and practices depends, according to Habermas, on two sources: participatory and epistemic sources. All those affected by the potential outcome of democratic processes need to be included in the process of political opinion- and will-formation. Equal participation makes it possible for citizens to see themselves both as addressees and authors of the law. The decisions made on the basis of these processes also need to be ‘dependent on the more or less discursive character of preceding deliberations’ (Habermas Reference Habermas and Prattico2022a: xiv).Footnote xiii This discursive character grounds ‘the presumption of rationally acceptable results’ (Habermas Reference Habermas and Prattico2022a: xiv) and makes it possible for citizens to see the results as rationally acceptable and to recognize their individual will in collectively binding decisions (Holst and Molander Reference Holst and Molander2015: 549).
More, however, is needed according to Habermas. In order for these democratic processes and practices to be able to function as a source for political motivation three conditions need to be met. Political processes need to be: rooted ‘in the citizens’ own implicit beliefs’, embedded in a democratic constitution, and accompanied by effective human rights practices (Habermas Reference Habermas2022b: 48).Footnote xiv
An active citizenry requires first of all a liberal political culture, a ‘delicate fabric of attitudes and taken-for-granted cultural assumptions (…) of the democratic constitutional principles, which remains for the most part implicit’ (Habermas Reference Habermas2022b: 54). Such a basic agreement can grow on the basis of a process of political socialization and political education. The core of such a liberal political culture is the ‘willingness of citizens to reciprocally recognise others as fellow citizens and democratic co-legislators endowed with equal rights’ (Habermas Reference Habermas2022b: 54). Echoing Böckenförde Habermas claims that such a political culture has to be sustained and affirmed but ‘cannot be produced by legal and administrative means’ (Habermas Reference Habermas2021a: 550). It is a ‘web of historically saturated political values founded on constitutional patriotism that can only arise in passing’. At best, he argues, ‘it can emerge from a civic practice in which all citizens are already engaged’ (Habermas Reference Habermas2021a: 550).
A second condition for an active and loyal citizenry is a sufficient degree of social equality. All citizens need to be effectively able to participate in the process of democratic opinion- and will-formation. A last condition concerns the fragile relation between democratic state and capitalist economy. A liberal state should be a welfare state guaranteeing an equal status for all (Habermas Reference Habermas2022b: 54). A ‘practice of rights protection’ (Vallier Reference Vallier2021: 103) is needed in which basic rights and freedoms are not merely enshrined in the constitution but where it is clear to all citizens that these are also successfully protected in everyday life. This requires effective public policies aimed at ensuring real equality of opportunity and a dignified and decent life for all citizens (Vallier Reference Vallier2021: 22).
The previous can be rephrased somewhat differently. If rational morality, as Habermas often notes, is motivationally weak, the questions arises what can ‘compensate for the motivational weakness of good reasons’ (Habermas Reference Habermas2021b: 71). within liberal democratic states. A partial solution is provided by positive law. Positive law tells us – on the basis of sanctions backed by force – what to do or not to do. Law, however, also needs to be legitimate in order to be supported. As Habermas claims, there can be – in modern societies at least – ‘no mass loyalty without legitimacy’ (Habermas Reference Habermas1999: 62). And in this context participation and deliberation are crucial, which in turn implies that citizens need to be willing and see it as their responsibility to make use of their political rights. A democratic constitutional state ‘expects that its citizens should adhere to an ethics of citizenship that goes beyond mere obedience to the law’ (Habermas Reference Habermas2017: 222). Such an ethos, as emphasized before, cannot be morally ordered or legally imposed on the basis of the instruments of ‘coercion and authoritative command’ but needs to be fostered on the basis of learning processes that themselves also need to be protected and supported (Habermas Reference Habermas2009: 75).Footnote xv It is an ethos, moreover, that emerges from democratic practices in which citizens are ‘already engaged’ (Habermas Reference Habermas2021a: 550). Rainer Forst expresses this dynamic succinctly: ‘modern forms of democratic Sittlichkeit (…) institutionalize a form of legal and political order that reflexively generates the duty to improve on itself, procedurally and substantively, by establishing superior forms of democratic organization, of securing and interpreting human rights, and by aiming at transnational forms of democratic cooperation. It is thus a Sittlichkeit that is present and at the same time yet to come’ (Forst Reference Forst2021: 22). Politico-juridical institutions, in that sense, are a necessary but not sufficient condition to generate a political (democratic) ethos.
3.2 Constitutional patriotism and the duty to support just institutions
Of the three conditions that, according to Habermas, need to be met in order for democratic processes and practices to be able to function as a source for political motivation, the first condition seems to be the one that is most affected by the paradox that Williams refers to: liberalism presupposing ‘a settled, mature society which encourages argument and debate’. The liberal political culture needed to sustain liberal democracies and make them more resilient consists of a ‘web of historically saturated political values founded on constitutional patriotism that can only arise in passing’. The legitimacy and stability of liberal democratic states, in that sense, depend on democratic practices and a political ethos. The second and third condition don’t seem to presuppose to the same degree what Habermas calls ‘a civic practice in which all citizens are already engaged’ (Habermas Reference Habermas2021a: 550). In fact, these two conditions partially create the circumstances necessary for effective participation and a necessary level of trust to arise; the trust that citizens need to have in the functioning of their legal and political institutions. Properly functioning legal and political institutions and a vibrant liberal political culture are, in that sense, partially mutually reinforcing spheres. Whether this dynamic reintroduces the paradox of liberalism with full force – support for militant constitutionalism depending on the successful implementation of militant constitutionalismFootnote xvi – is a point that will be discussed in the conclusion.
To summarize, Böckenförde’s ‘prepolitical foundation’ can be reflexively generated in and through a discursive political and juridical order that guarantees citizens a status as legal persons and embeds democratic processes in constitutional institutions. Effectively securing rights and equal participation, in that sense, are two of the preconditions for a political ethos to develop. The democratic and constitutional state, consequently, is more than just a ‘“civilizing force”, “taming” coercion, containing despotism, and furnishing “legal and democratic domestication of brute political power”’ (Chambers Reference Chambers, Allen and Mendieta2019: 69).
And yet one element of militant constitutionalism needs stronger emphasis to realize an effective form of constitutional self-defense. Effective protection implies legal and political practices in which, as Jonsson Cornell rightly claims, ‘the constitution is considered a legally binding document by a society as a whole. Its values are entrenched in society, its words and spirit respected by politicians, civil servants, judges, business, civil society and the population’ (2021: 10). Such a form of self-defense not only implies a political (democratic) ethos but also a constitutional one. Politicians, civil servants, and judges need to accept, support, and try to effectively secure the values, rights, and liberties entailed in the constitution of their liberal democratic state. In addition, they also need to know that others are doing the same. As Sadurski argues: ‘What is required is not just knowledge of the norms, but also knowing that others know them, and that they will abide by them, and, if they do not, they will know that they have violated them’ (Sadurski Reference Sadurski2019b: 256). A shared constitutional ethos creates legal trust.
Constitutional patriotism seems a likely candidate for understanding the content of such a constitutional ethos. Although the core of this concept (at least within a Habermassian framework) is formed by the idea that the constitution can act as the focal point for cohesion and solidarity in liberal democracies – the constitution as the basis for collective identity – it does not need to be restricted to this focus and can operate within the framework of both militant democracy and militant constitutionalism. Jan-Werner Müller points at this possibility in his analysis of constitutional patriotism. German constitutional patriotism, according to Müller, contains two distinctive features: ‘memory’ and ‘militancy’. Memory in the sense of a ‘self-critical remembering of the Holocaust and the Nazi past’.Footnote xvii Militancy as a consequence of the use of party bans and restrictions on free speech to keep at bay ‘the enemies of democracy’ (Müller Reference Müller2007: 10). Of the two most well-known German theorists working on this topic, Dolf Sternberger more clearly embraces this link between constitutional patriotism and militancy than Habermas (Sternberger Reference Sternberger1990). Habermas’s main focus is on the ‘democratic quality of political culture’ and not on the ‘defense of democracy’ (Müller Reference Müller2007: 30). Müller characterizes Sternberger’s Verfassungspatriotismus, therefore, as a form of ‘protective constitutional patriotism’ aimed at ensuring peace and political stability (Müller Reference Müller2007: 24).
Memory and militancy, Müller claims, are not mere ‘accidental forms of particularity associated with constitutional patriotism; rather, there is an inherent normative connection to the universalist kernel of constitutional patriotism’ (Müller Reference Müller2007: 11). He immediately adds, however, the warning that it is wise not to put too much stress on both features. Memory and militancy can have ‘an illiberal side’ by reinforcing ‘“identity” through negative contrast – with the past that is being repudiated, or with anti-democratic political actors in the present’ (Müller Reference Müller2007: 11). But this does not imply that constitutional patriotism should not be linked to moderate versions of militancy: constitutional patriotism as implying a form of loyalty to the particular way in which universal morality and values are embedded in the constitution of our liber democratic state and a willingness to ‘to draw a legal and political line so far as the possibility of endangering these values is concerned’ (Müller Reference Müller2007: 41). Constitutional patriotism, in that specific sense, requires citizens (but also politicians, civil servants, and judges) ‘to support and to comply with just institutions that exist and apply to us’ (Rawls Reference Rawls1999: 99).Footnote xviii
To be clear, the object of attachment for a theory of constitutional patriotism is not a national culture but the constitution; not a ‘particular constitution in all its specificity at a particular point in time’ (Müller Reference Müller2007: 54)Footnote xix , but the way in which universal norms – human rights and democracy – are being interpreted and re-interpreted in a specific constitutional culture or project. It is this culture that is the ‘focus of citizen’s emotional attachment and the source of their civic solidarity’ (Huw Rees Reference Huw Rees, Allen and Mendieta2019: 67), and it is this culture that is also linked to an obligation or duty to support and protect the values, rights, and liberties entailed in the constitution. Such an obligation or duty can be a Kantian one. Regarding the reasons for attachment, Müller claims, different answers can be given depending on one’s ‘moral background theory’ (Müller Reference Müller2007: 62). ‘In a more strictly Kantian version’, he writes, the ‘reasons for attachment could be construed exclusively as duties to uphold a just political order; in a different version which stresses the importance of democracy for individual dignity, there might be instrumental reasons to sustain a constitutional culture or identity which allows for collective agency and collective learning processes’ (Müller Reference Müller2007: 62).
Müller here seems to presuppose a negative answer to the question whether Kantian theorists should argue that patriotism and cosmopolitanism are mutually exclusive ideals and a positive answer to the question whether they can accept that cosmopolitan universalism can be reconciled with ‘the attribution of “intrinsic significance” to state boundaries’ and the ‘agent-relative responsibility’ to uphold the just institutions of a specific state (Kleingeld Reference Kleingeld2000: 313-4). Pauline Kleingeld presents a convincing case for such an answer. According to Kleingeld, there is an argument that Kantians can use to fill the gap between ‘my general duty to promote justice and my particular duty to my particular state’. This argument, however, presupposes that ‘a just state is a democratic state’ (Kleingeld Reference Kleingeld2000: 327). A just and properly functioning democratic state, Kleingeld claims, ‘does not just need citizens who can comply with the laws and pay their taxes’, it also needs ‘the political involvement of citizens in order for democracy to function properly’ (Kleingeld Reference Kleingeld2000: 327). The ‘duty to promote just states’, she claims, is incoherent without a ‘duty of civic patriotism’ (Kleingeld Reference Kleingeld2000: 328). Such a duty, one could further argue, not only applies to citizens but more strongly also to politicians, civil servants, and judges.
Whether a strictly Kantian account of the duty to support and protect the values, rights, and liberties entailed in the constitution of a liberal democracy is the most convincing and coherent one, is not a question that needs to be answered here. What is important, however, is to emphasize that a theory of militant constitutionalism should include an account of the constitutional culture and ethos that is needed to make liberal democracies resilient and stable. A broadly Kantian account offers a very fruitful starting point.
4. Concluding remarks
This article had two main goals. First, to argue that both militant democracy and militant constitutionalism are necessary to defend liberal democratic states against political leaders that want to abolish democracy or co-opt or circumvent ‘the norms and institutions meant to support basic liberties’ (Freedom House 2022: 1). This goal was addressed in section 2 of this article. In this section the idea of militant constitutionalism was reconstructed on the basis of a comparison with the existing literature on idea of militant democracy. Both concepts were distinguished on the basis of a comparison regarding timing, means and actors involved. Given that the resilience of liberal democracies depends both on institutional safeguards and a legal and political ethos, the second goal of this article was to show how liberal democracies can try to secure the conditions that sustain themselves, thereby making them more stable and less vulnerable to forms of democratic regress and rule of law backsliding. This goal was addressed in section 3 of this article mainly on the basis of Habermas’s discourse theory of law and democracy. In addition Habermas’s notion of constitutional patriotism was reformulated in order to show that it can also entail a protective constitutional ethos or attitude. Section 3, in that sense, provides a partial answer to the Böckenförde paradox with which this article started. Liberal democracies cannot guarantee the conditions by which they are sustained but they can try to create these conditions.
It is important to emphasize in this regard that militant democracy and militant constitutionalism should not be understood as expressing full-blown theories of justice. Their aim is not to sketch the groundwork for a just liberal democracy. Militant democracy and militant constitutionalism are normatively dependent concepts that build on wider theories of democracy and rule of law ‘to gain substantive normative content’.Footnote xx Their focus is on the barriers that need to be erected and the practices and ethos that need to be sustained to defend liberal democracies.
What about the paradox of liberalism that Howard Williams described in his book Kant’s Political Theory (1983)? To what extent do liberal democracies presuppose settled, mature societies that encourage ‘argument and debate, and where this condition is not realized [are they] forced into silence and political impotence (…)?’ (Williams Reference Williams1983: 157). This article only provided a partial response to this paradox. Section 2 and 3 taken together lead to the conclusion that the rule of law entails more than just a system of institutional constraints. ‘Many aspects of the rule of law’, Krygier for example claims, ‘are intended to generate virtues of moderation and thoughtful self-knowledge, that are encouraged by constitutional and rule of law practices and institutions, not contained or constrained by them’ (Krygier Reference Krygier2019: 782). Institutional forms of moderation (a balancing or blending of power) and personal virtues like temperantia, he argues, are not two separate entities but internally linked. A similar and related line of thinking, this article has shown, applies to the idea of democracy. Correctly institutionalized and functioning democratic processes and rights practices can encourage democratic loyalty and ethos. Ethical substance and political and legal institutions, in that sense, are connected in different ways.
So on the one hand, liberal democracies do not need to be built on the assumption that politicians are of good will. Liberal democracies are not ‘forced into silence and political impotence’ if they aren’t. The rule of law can entail clear institutional forms of moderation and constraint. On the other hand, however, these institutional forms of moderation and constraint need to be sustained by a political and legal ethos that itself is partially dependent on the successful institutionalization of specific democratic processes and socioeconomic rights practices. Politicians, in that sense, do need to be willing to secure these processes and practices. Without this willingness, militant constitutionalism and militant democracy will lose part of their striking power. If these processes and practices function properly, however, a stronger barrier exists that can prevent elected public authorities from deliberately trying to dismantle the liberal democratic state.
Note, however, that the aim of this article was not to fully solve the various forms the paradox of liberalism can take. This article did not, for example, try argue that liberalism should be seen as a philosophy that ‘fights for bringing about an order of political justice in which everyone’s equal dignity receives public recognition’ (Bielefeldt Reference Bielefeldt1997: 547). Nor did it argue, as Heiner Bielefeldt has, that Kantian liberalism entails a form of fighting liberalism that ‘requires one to take on the challenges of moral self-responsibility and republican commitment’ whenever ‘human dignity is at stake (Bielefeldt Reference Bielefeldt1997: 525). This article was not about liberalism as a philosophy but about the defense of liberal democratic societies.Footnote xxi
