The Democratic Self-Defence of Constitutional Courts

1 Introduction: Protecting Democracy Through Constitutional Court Self-Defence

A paradox emerges in contemporary attempts to reform institutions endowed with powers of constitutional review, particularly in the context of judicial appointments. The United States has witnessed passionate debates surrounding the composition of the Supreme Court, prompting suggestions of ‘expanding’ or ‘packing’ the court to address the political crisis surrounding the institution.^[1] Ironically, however, court-packing, once deemed a strategy followed by illiberal governments, has been observed in countries like Poland to exert control over constitutional courts.^[2]

The unexpected parallels between debates on the politics of judicial appointments in the US and Poland highlight a broader challenge in reforming these institutions: it becomes challenging to perceive a legitimate reform of a constitutional court, acceptable from a democratic perspective, from an illiberal attack on constitutional review demending counteraction. While these theoretical and comparative dimensions warrent academic discussion,^[3] the issue transcends theoretical and scholarly considerations, manifesting a clear practical dimension. For institutions primarily designed to safeguard democracy, such as courts (constitutional or supreme), the potential risk to democracy posed by reforms to their own institutional design should matter.^[4]

This article tackles this complex issue, particularly addressing the conditions under which constitutional courts should veto reforms of their institutional design. The argument of this article is thus complementary to, but distinct from, previous debates in the field that seek to elucidate how constitutional courts should respond to attacks on democracy,^[5] or the effectiveness of constitutional court action in protecting democratic systems.^[6]

The response to the question addressed in this article is the theory of democratic self-defence of constitutional courts. In essence, the theory posits that constitutional courts should always declare the unconstitutionality of reforms on their institutional design when such changes serve as instruments for democratic decay. In this article, we develop such a theory and explain how its main elements should be interpreted. The article not only aims to clarify an obscure question in constitutional law but also endeavors to provide doctrinal tools for real-world courts to safeguard democracy by defending themselves against illiberal and authoritarian reformers.

The structure of this article is as follows: Following this introduction, we present some methodological considerations, explaining the main conceptual and methodological priors underpinning our argument. Subsequently, we address the background problem justifying this research: the takeover of constitutional courts by illiberal or authoritarian actors preceding processes of democratic decay. The following section presents the theory of democratic self-defence, detailing three main aspects that delimit its content and applicability: why it is a duty (not merely a right) for these institutions to protect themselves, which types of attacks are relevant to this theory, and when should an attack on the constitutional court constitutes an attack on democracy. The following section anticipates and responds to potential criticisms of the theory. The penultimate section of the article applies the theory to the Israeli case study, while the final section concludes.

2 Methodological Remarks

This article seeks to integrate theoretical and doctrinal approaches in discussing the idea of self-defence of courts. While the theory and doctrine of democratic self-defence of courts are distinct, they are inherently linked and intertwined. The theory of democratic self-defence of constitutional courts, being normative, involves an assessment of whether and when constitutional courts should safeguard themselves to uphold democracy. Operating on an abstract level, it seeks to provide a common denominator applicable to all courts in democratic polities, even if the ideas generated are often political and legal in nature.

Conversely, the doctrine of democratic self-defence of courts utilizes law and legal authorities to serve the theory. It seeks to interpret existing legal systems to implement the core aspects of the theory through a sound legal reasoning. That being the case, this doctrinal approach is inherently specific to each legal system, so rather than speaking of a general doctrine of democratic self-defence of courts, it is more accurate to speak of a Spanish doctrine of democratic self-defence, a French doctrine, a Chilean doctrine, an Israeli doctrine etc (Table 1).

This article deals with the relationship between constitutional courts, democracy and democratic decay. Thus, for reasons of clarity and transparency, and for the sake of methodological rigour, these three concepts need to be defined before delving into our discussion.

In this article, the term constitutional court denotes any judicial-type organ that has the final say on the constitutionality of legislation. Thus, this definition includes not only Kelsenian-style constitutional courts that monopolize the control of constitutionality, such as the German Federal Constitutional Court, but also apex courts in systems that follow the model of diffuse review of legislation, such as the US Supreme Court or the Supreme Court of Israel. The notion of ‘constitutional court’ is used loosely in this article.

Democracy, as defined in this article, adopts a ‘thick’ sense, including a system of government that enables the governed to remove incumbents from power through free and fair elections. However, this definition goes beyond mere electoral processes, incorporating additional elements such as respect for the rule of law, protection of basic rights, and the presence of checks on governmental power.^[7]

Democratic decay, as articulated by Daly, is adopted in this article, denoting ‘the incremental degradation of the structures and substance of liberal constitutional democracy’.^[8] The incremental nature of democratic decay, that Daly defines as a ‘subtle, step-by-step hollowing out of democratic governance’,^[9] aligns with the purposes of this article. Attacks on democracy, particularly those directed at constitutional courts, often exhibit this incremental nature.^[10] While the arguments presented in this article can be relevent to abrupt democratic breakdowns, the discussion is primarily developed with situations of democratic decay in mind.

On the methodological priors, the research rests on several of assumptions, and while space constraints prevent an exhuastive elaboration of the normative justification for each, they are transparently presented as the background framework of our argument. The research has, in particular, three methodological priors:

Democratic desirability. The first methodological prior posits the normative assumption that democracy is inherently valuable. It is considered a desirable form of political organization, not solely for instrumental reasons such as its potential to foster economic prosperity or ensure peace, or not only because of them. Democracy is a desirable form of political organizations for reasons that are inherent to it, and that have to do with human dignity. It is based on the expression of pluralism, a fundamental element of free human societies.^[11]

Militant democracy. Second, democracies can and must defend themselves. Democracies are not only normatively desirable but must also, because of that, create the arrangements to ensure their own viability and resilience vis-à-vis authoritarian threats. This prior is thus inspired by the ideas of militant democracy, that were best systematized by Karl Loewestein in the 20th century.^[12] However, recent literature highlights the inadequancy of traditional tools of militant democracy against contemporary forms of authoritarianism.^[13] This article takes note of these observations. Acknowledging this insufficiency of the traditional toolkit of militant democracy, but vindicating its telos, this article aims to provide a new tool for democracy protection, contributing to update militant democracy.

Pluralist legal constitutionalism. The final methodological prior relates to the ongoing debate between legal and political constitutionalists and the broader discourse on the constitutional powers of judicial institutions. Legal constitutionalism asserts that judicial enforcement of a normative constitution is essential for the democratic order’s viability.^[14] In contrast, political constitutionalists, constitutional review skeptics, generally argue that the ultimate decision on policy matters should rest with democratically elected branches of government.^[15] This article adopts legal constitutionalism as a methodological starting point, recognizing the necessity of an institution with constitutional review powers for the theory of democratic self-defence of courts to have relevence. However, the article adopts a pluralist approach to legal constitutionalism. While legal constitutionalism requires constitutional review as a necessary condition, a pluralist approach acknowledges diverse design possibilities and varying degrees of court powers.^[16] This prior allows for the acceptance of constitutional review forms that, for some extent, accommodate the assertions of political constitutionalists regarding policy choices made by democratically elected politicians.^[17] The only condition is that the constitutional court possesses the capability to enforce rules protecting democracy, aligning with our two previous methodological priors.

3 The Problem to Tackle: Constitutional Courts and Democratic Decay

The implementation of constitutional review in many countries is driven by the objective of protecting democracy.^[18] However, to fulfill this role, constitutional courts accumulate a wide range of powers, frequently intersecting with the political process.^[19] Consequently, the capture of courts by illiberal politicians posses a significant threat. As suggested by Ginsburg and Huq, under such control, courts can transform into a powerful instrument in the hands of illiberal politicians to ‘turn the law loose on their enemies’.^[20]

This scenario appears to have unfolded in a numerous countries around the world, spanning from Poland to Venezuela, Hungary to Turkey. Increasingly, literature characterizes political control over constitutional courts as a catalyst of democratic decay and rule of law backsliding.^[21]

The danger is twofold: firstly, once courts are weakened, captured, or controlled, undermining other democratic values and institutions becomes markedly more fesible. As stated in the introduction to these special issues:

Comparative experience shows that weakening the ability of courts to supervise the government and review its actions is a central pillar of democratic erosion, as it removes a crucial obstacle for a government seeking to consolidate its power, and makes it easier for it to make further changes later. In other words, because the judiciary is the last line of defense of democracy, in order to capture and undermine democratic institutions the first thing that governments do, once in power, is to threaten, limit and if possible, to capture and court. Once the court is weakened or captured, it then becomes easier to weaken or capture other democratic institutions.^[22]

Secondly, once courts are captured, they themselves become instrumental in the process of democratic decay by affirming and advancing the anti-democratic agenda of the government. Indeed, the paradox lies in the fact that institutions initially established to protect democracy find themselves serving processes of democratic decay. This phenomenon is termed by Rosalind Dixon and David Landau ‘abusive judicial review’.^[23] The theory of democratic self-defence of courts aims to address this predicament by offering remedies. It does so by presenting a general justification for vetoing attacks on constitutional courts. Furthermore, it provides a nuanced analysis distinguishing between democratically legitimate reforms to the design of constitutional courts and those that are not, warranting overturning.

4 A Theory of Democratic Self-Defence of Constitutional Courts

The theory of democratic self-defence of constitutional courts argues that it is a duty for these institutions to defend themselves from attacks that are part of a larger process of democratic decay. This simple presentation of the theory involves three elements that necessitate a more in-depth explanation.

5 Some Potential Objections Against the Theory: a Preventive Response to Critics

So far, we have argued that constitutional courts should defend themselves from attacks that can endanger democracy. But we understand that our argument might raise some objections.

A first potential criticism against our argument has to do with the epistemic problem that we acknowledged above, as well as with the acknowledged existence of ‘difficult cases’ in which the teleological and the consequentialist criteria for the determination of a risk for democracy do not match. For instance, sometimes it will not be easy for constitutional courts to know the intentions or the outcomes of a certain reform on their institutional design. We already acknowledge that, in these cases, the application of the theory of democratic self-defence encounters difficulties, as it faces a trade-off: either to declare unconstitutional a reform of its design that could not be in the end a threat for democracy, or declaring the constitutionality of a reform that could turn out instrumental to a process of democratic decay. But the problem is actually more acute in contemporary democracies, as given the strategies followed by illiberal actors, such difficulties are doomed to arise very frequently. The criticism could thus be that the theory of democratic self-defence is unable to meet one of its declared goals: to update militant democracy and make it more functional at a time in which democratic decay – as opposed to abrupt democratic breakdown – becomes more frequent.

The theory of democratic self-defence of constitutional courts can only respond to this criticism by reminding what it can still do and by simultaneously acknowledging its limitations. The theory of democratic self-defence of constitutional court faces indeed difficulties in these scenarios. Such difficulties would need to be solved on a case-by-case basis, with whatever level of information is available to the court. The trade-off described above will be however, in many instances, unavoidable. But even in these cases, the theory of democratic self-defence of constitutional courts is useful for two reasons. First, it will be helpful to remind constitutional courts that one of the criteria against which they have to assess the constitutionality of a reform of their own design is the extent to which it can be instrumental to democratic decay. Second, in making explicit the trade-off that courts might confront in many cases, the theory helps the constitutional court think about the choice they are down to make and its potential implications.

Besides, not all cases will be difficult cases. The theory of the democratic self-defence will have a rather straight-forward application in many instances, when the court has enough information to conclude that a reform of its design is both intended at and will likely cause democratic decay. Such cases will be the most blatant ones, and thus the most necessitated of a conceptual apparatus like the one developed in this article. The greater challenge will arise when democratic backsliding is subtle and incremental.^[43]

The theory of democratic self-defence of constitutional courts is designed for scenarios in which the court is attacked via a reform of its institutional design. But very often, attacks on constitutional courts do not require of such reforms. This might occur in two different scenarios. First, it might happen because attacks on the court operate through the existing institutional design of the court, which in certain circumstances allow illiberal politicians to neutralize or control the court. Second, it might also occur because the attack on the constitutional court proceeds not through formal but through informal mechanisms, such as corrupting constitutional judges.

Both these scenarios might be thought of as a failure of institutional design. Constitutional courts ought to be designed in such a way that control of the court cannot occur, neither through the abuse of the existing formal mechanism nor through the use of informal mechanisms.

In any case, because there is no formal amendment of the design of the court, there is no option to declare such an amendment unconstitutional, and thus the theory will not be of direct use in these cases. This is not to say, however, that the theory cannot be of some help even in these scenarios. At the very minimum, some of its elements – even if not the theory in its entirety – can help provide for an assessment of the situation and prompt action. If the telos of constitutional courts is to protect democracy, and constitutional courts have a duty to counter attacks on them that are instrumental to democratic decay, this duty applies not only when formal amendments are considered but also when informal mechanisms are applied.

Another possible objection that the theory might encounter is of a practical nature. The theory of democratic self-defence of constitutional courts expects these institutions, in certain cases, to declare the unconstitutionality of reforms of their own design. However, because of the complex context this entails, it will be easy for authoritarian political actors to argue that the constitutional court is behaving unlawfully.

Imaging for instance an illiberal reform aimed at packing a constitutional court with judges loyal to an illiberal government. Existing constitutional judges might declare such reform unconstitutional. But the illiberal political actors behind the reform might respond that the declaration of unconstitutionality is invalid because it has been issued by the old constitutional judges who, according to the new legislation, are no longer the legitimate members of the institution. Or consider a reform that makes it more difficult to strike down legislation, for example by requiring a super-majority of the panel; a declaration of unconstitutionality by an ordinary majority might be regarded as invalid.^[44]

We have two responses to this criticism. The first is that problems of a practical nature are not enough to invalidate the theory at the conceptual level. The second is that, at the practical level, these problems might well exist. In the worst case, they provoke a constitutional crisis, in which different constitutional organs have different, irreconcilable views on the constitutional way ahead. But a constitutional crisis will often be better than the suppression of liberal constitutionalism altogether.

A related challenge concerns the legitimacy consideration. When the court is to apply the self-defence theory to strike down legislation, and even more so to strike down a constitutional amendment, the court may be seen as acting to preserve its powers, its superiority over other branches of government and to enhance its self-interests, and be criticized for that. Such claims of misuse of the unconstitutional constitutional amendment doctrine have been made regarding judicial application of the doctrine in India, Pakistan, and Bangladesh in cases concerning judicial appointments and removal.

However, as Po Jen Yap and Rehan Abeyratne demonstrate, not every judicial review of constitutional amendments concerning the powers and status of the judiciary is problematic, when the theory is used to block interference by the executive in judicial functions and considering the legal and political context of the country. When judicial independence is in danger, judicial intervention even in constitutional amendments concerning the judiciary itself may be justified, especially in fragile democracies.^[45]

If preserving judicial independence is vital to the survival of the democratic order, then protecting the court seems more important than perceived legitimacy; but more than that, if the court does not intervene in such cases, it may lose its independence and thereby allowing democratic decay. The result of such non-interference might thus also have negative effects on the court’s legitimacy. If the choice is between suffering legitimacy harm and protecting democracy or suffer legitimacy harm and undermining democracy, the former is preferred.

A final criticism is that such judicial intervention will not prevent democracy decay anyway.^[46] Out of all the criticisms explored in this article, this is probably the most consequential one. In a way, it poses an existential criticism to the theory of democratic self-defence. Since the idea behind the theory is for constitutional courts to protect themselves in order to protect democracy, whenever democracy protection through the theory fails, the very theory becomes futile.

Against this backdrop, is it still worth wasting any efforts on such an idea as the theory of democratic self-defence of constitutional courts? we believe that is indeed the case, for at least the following reasons.

First, it is not entirely true that the theory is only worth if it achieves the desired effects. This is a consequentialist argument and, while it is powerful, it is not the only way to approach the theory. To this consequentialist argument, we could contrapose a different, deontonic argument: this theory is simply the logical corollary of the ideas implicit in the very concept of constitutional review. Constitutional courts are conceptually designed to protect democracy and thus have a duty to pursue democracy protection through the enforcement of the constitution regardless of whether they achieve this goal.

Second, taken to the extreme, the criticism that formal checks on power are unable to deter democratic decay when political actors do not commit to democracy leads to a somehow absurd scenario. This is because, if that were the case, why have formal checks on power at all? We could in this scenario dispense not only with constitutional courts but also with rigid constitutions, ombudsmen, supranational democratic clauses, organs for judicial governance, and a long list of other formal and institutional checks on power. Democracies, however, generally do rely on these sorts of arrangements.

In Israel, a similar argument against authorizing the court to review Basic Laws (ie, laws of a constitutional status) was made, according to which courts cannot protect democracy because, in any case, when a basic law that would truly harm democracy arrives, its invalidation by the court would simply be ignored. As former Minister of Justice, Ayelet Shaked, stated, ‘If Parliament would enact a law that says “all redheads must be hanged” – the court will not be able to assist because society has become so corrupted’, or ‘If the Knesset were to pass a law rescinding the voting rights of women or red-haired people… this would signal the collapse of our democracy. In such a case, I don’t think that even the court could save us from ourselves.’^[47]

With all respect, this argument falls to the false dichotomy of democratic failure: either we have a perfectly functioning democracy or a complete failure, Weimar style. But that is never the case; the court’s influence on democracy is not measured in the final hypothetical law that transforms a country from a democratic one to an autocracy but incrementally and gradually in a series of judgments in the life of a nation. The dichotomy between a perfectly functioning democracy and a complete failure is a false one. Between these two extremes, there is a vast spectrum in which courts can function as a useful stop sign or a speed bump against constitutional and legal reforms aiming to undermine or erode the democratic order.^[48]

To say that institutions do not matter mischaracterizes the problem and underestimates the role of formal institutions, precisely, in regulating the behaviour of political actors through incentives, constraints and values.

For that reason, the theory of democratic self-defence of constitutional courts can be helpful. At a very basic level, it has the capacity to change the cost-benefit calculus by illiberal political actors. Insofar as it gives the constitutional court a doctrinal tool to protect itself from attacks and fightback, it creates an incentive for illiberal political actors to be restrained. Thus, even before a constitutional reform is enacted, the mere possibility of its judicial invalidation has an ‘anticipatory effect’ that assists in protecting democracy.^[49] As Mark Tushnet correctly notes, the mere existence of unamendability doctrines ‘may serve as a political check on the amendment process, as a “sword of Damocles” that, because it occasionally drops, cautions political actors against devoting too many resources to attempting to alter the existing specification of some component of the [constitution’s] basic structure.’^[50]

Paradoxically, the theory would be most successful in cases in which it is not utilized. Of course, if a court can use the theory to break the causal chain that links a reform of its design with a process of democratic decay, that would be a success. But the theory can be particularly successful if it is not used because the very awareness of its existence deters illiberal politicians from seeking an assault on the court.

Finally, there is the issue of the assumptions. We should indeed not assume that the theory will be something akin to a universal solution to prevent democracy decay everywhere. But the assumption that it would not work also lacks empirical grounding. Because this is a novel theory, we have little evidence about how it would perform in real-world scenarios. It might well be that the theory, if put to the test, fails to protect democracy in all cases, or that it is very effective in fostering democratic resilience, or that its effectiveness depends on its specific application in specific settings. Because of the novelty of the theory, we cannot have enough evidence to argue in favour or against its effectiveness. We do have, however, instances of constitutional courts that tried to resist authoritarian assaults and of courts that managed to protect democracy.^[51]

6 Case Study: Israeli Supreme Court

So far in this article, we have provided a rather theoretical outline of the idea of democratic self-defence of constitutional courts. In this section, we aim to sketch how this idea could apply in practice. To do so, we use the case study of Israel. We chose Israel as a case study because, as highlighted in the introduction to the special issues, the country has witnessed in 2023 a judicial overhaul by a right-wing and populist government, proposing major constitutional reforms to the judicial system. These reforms included limiting the power of constitutional review, enacting an override clause allowing the enactment of unconstitutional laws notwithstanding court decisions, limiting the court’s authority of administrative review, weakening the status and independence of government legal advisors, and changing the manner by which judges are selected.^[52] In this section, we shall focus on two elements within this reform (although all the elements have accumulative effect and the whole is greater than the sum of its parts): changing the manner by which judges are selected and limiting administrative review.

Judicial selection: On January 4, 2023, Israel’s Minister of Justice, Yariv Levin presented his plan to overhaul the judiciary by reducing its powers and giving the government more control over judicial selection. The current committee for selecting judges is composed of nine members: three Supreme Court judges, two Ministers, two Knesset Members, and two members of the Israeli Bar Association. To appoint a judge to the Supreme Court, a super-majority of 7 out of 9 is required, ensuring that no branch controls that process entirely. Instead, the proposal sought to change the composition of the committee so that the process for selecting judges (to all courts in Israel, not only to the Supreme Court), would be controlled by the government and the coalition that supports it in the Knesset. The explanatory notes of the bill stated that intention was to ‘strengthen the influence of the public’s elected officials – the representatives of the executive and the legislature’ in the selection of judges, in order for those judges to reflect ‘the values of the public’, yet the proposal would effectively allow the government to capture the judiciary: ‘Rather than democratic legitimacy and judicial accountability – principles that are legitimately involved in judicial appointments and justify legislative and executive involvement – what the bill promoted was the capture of the courts, through absolute executive control of judicial selection and promotion.’^[53]

The bill passed the first reading and was prepared for the second and third reading, but the legislative process was paused in response to the unprecedent and massive public protest against the reform.^[54] However, if the constitutional amendment had been enacted, we argue that the Supreme Court could have applied the self-defence doctrine and declare the bill as an unconstitutional constitutional amendment. Given Israel’s relatively weak system of checks and balances,^[55] the capture of the judiciary by the government would have granted it almost absolute powers and put the very democratic character of the regime at risk. The consequences for the democratic character are clear, and even the intentions of the legislatures make this case an easy one, revealing the entire package of proposed reforms which revealed the government’s intentions: to weaken external checks and balances on its power, and granting it absolute powers of law-making without limitations. Considering the ongoing process of democratic erosion in the country,^[56] this attack was part of a larger process of democratic decay. Accordingly, the court would have been justified, in our view, to strike down this reform had it been enacted.

The second example is the law that limited the court’s authority of administrative review. This law – Amendment Number 3 to Basic Law: The Judiciary, stated that no court would hear cases about or issue an injunction against the government, the Prime Minister, or the ministers based on the reasonableness of their decisions, including appointments and inaction. Again, the aim was to eliminate judicial review and grant the government unlimited power.^[57] One may claim that the amendment does not completely grant the government absolute powers without administrative oversight because other doctrines, such as proportionality conflict of interest, still apply. However, these do not necessarily overlap with the reasonableness doctrine, are more difficult to prove, and without reasonableness, important gatekeepers such as legal advisors or the attorney general may be replaced with less independent once, allowing the government to continue with the process of democratic decay.^[58] It is also important to note that this amendment is extremely broad in its scope; unlike some suggestions to remove certain policy areas from reasonableness scrutiny (such as immigration, economic or security), it completely removes the court’s authority from supervising the reasonableness of any decision by the Prime Minister, the Ministers, and the Cabinet. Furthermore, the entire body of jurisprudence of the court regarding restrictions imposed on transitory government (a caretaker government has a narrower range of reasonableness than an ordinary government),^[59] in order to maintain competitive elections, is based precisely on this reasonableness standard. Accordingly, the amendment could have far-reaching implication to the democratic order.

This law was enacted in July 2023. It is the only part of the Minister of Justice’s package of reforms that was enacted thus far. Multiple challenges were submitted to the Supreme Court, which heard the petitions in a full bench of 15 judges, for the first time in its history.

Should the court intervene in this case? This example is more challenging than the previous one. In the beginning of the article, we have claimed that we accept that policy issues should belong to democratically elected branches of government. This is true. Removing the court’s authority to review the reasonableness of government’s actions aims to allow the government to advance its preferred policies. Thus, prima facie, the court should not veto the amendment. However, a deeper observation that includes the larger context of democratic decay process, the aims of the government (revealed by the other components of the reform), and the importance of the reasonableness standard to checks on the governmental powers in the Israeli context, leads to the conclusion that this reform (if interpreted as a complete ban on any form of review of reasonableness or arbitrariness), ‘could be viewed as a significant erosion of the institutional checks and balances on executive power – an erosion of the democratic minimum core.’^[60] Accordingly, the court may be justified if it declares this amendment as unconstitutional as part of the democratic self-defence doctrine of constitutional courts. Indeed, on January 1, 2024, the Israeli High Court of Justice declared the amendments as unconstitutional and invalidated it. A majority of Eight to Seven judges held that the amendment severely harms the core characteristics of the State of Israel as a democratic state, especially separation of powers and the rule of law.^[61]

Of course, this analysis is specific and context-based. A different conclusion may be reached if the analysis applies to different jurisdictions, considering local conditions, institutions and context.

7 Conclusions

As explained earlier in this article, the theory of democratic self-defence of constitutional courts is largely tributary of the work of Karl Loewestein. As he himself put it: ‘Democracy stands for fundamental rights, for fair play for all opinions, for free speech, assembly, press. How could it address itself to curtailing these without destroying the very basis of its existence and justification? At last, however, legalistic self-complacency and suicidal lethargy gave way to a better grasp of realities. A closer study of fascist technique led to discovery of the vulnerable spots in the democratic system, and of how to protect them’.^[62] Loewestein’s argument was, we believe, that democracies should not allow authoritarians to take advantage of democratic freedoms in their quest to undermine democracy itself. Democracies should protect themselves.

The theory of democratic self-defence of constitutional courts is an application of this idea to the question of reforms of these institutions. It seeks to provide the theoretical and doctrinal basis that should allow constitutional courts to protect themselves from attacks by authoritarian politicians, not in order to protect court per se, but precisely in order to protect democracy. The militant-democratic character of the theory is particularly clear when it comes to abusive constitutional reforms: reforms of the design of constitutional courts that look formally constitutional but which are linked to a risk of democratic decay. In these situations, the theory of democratic self-defence of constitutional courts provides for a solid basis that allows for action in defence of these institutions and of democracy. The theory seeks to afford such democratic protection while preserving the margin of democratically elected politicians to engage in legitimate reforms of the architecture of the court: those that do not pose a risk for democracy. The overall aim of the theory is thus to facilitate protection of democracy through the protection of certain aspects of the constitutional court, without fully entrenching all aspects of the design of these institutions. Since courts are the final gatekeepers of democracy, democracy should have a ‘militant protection’ of courts.