How to Detect Abusive Constitutional Practices

Abstract

The challenge of distinguishing between abusive and non-abusive constitutional practices – Main ways of detecting abuse: analyses of intent and effects – Obstacles to detecting bad faith intent in constitutional law – Structured and focused analysis of the effects of abusive constitutionalism: introducing the foreseeable effects test – Normative benchmark: substantial diminishment of accountability – Step 1: probability of harm, analysing the tested constitutional measure’s foreseeable effects on operability and autonomy of an accountability mechanism – Step 2: seriousness of potential harm, analysing the constitutional measure’s effects in the broader constitutional context (including interaction effects) – Step 3: harm mitigation, analysing the existence and adequacy of harm-mitigating safeguards accompanying the constitutional measure – Strengths and limits of the foreseeable effects test

Introduction

How can we tell abusive constitutional practices from non-abusive ones? When Canada and the UK introduced a system of judicial review of legislation which gave the last word to the legislature, it was celebrated as an attractive model squaring democracy with rights protection. ¹ When the Israeli government in 2023 announced a plan to give the Knesset the power to override the Supreme Court, this was viewed with dismay and further democratic deterioration was feared. ² Or consider an example from Europe. The German judicial system, often celebrated for its professionalism and efficiency, is largely governed by the Ministry of Justice. Yet, post-2015 Polish judicial reforms were criticised as excessively politicised when strengthening the executive’s mandate in court administration. ³

Abusive constitutionalism is understood as ‘the appropriation of liberal democratic constitutional designs, concepts, and doctrines in order to advance authoritarian projects’. ⁴ Some scholars see bad faith anti-democratic intent as a major indicator of abuse. ⁵ From this perspective, the architects of the new Commonwealth model of judicial review aimed to create a genuine dialogue between courts and legislatures and find a solution to the counter-majoritarian difficulty. In the Israeli case, however, the fear was that the introduction of the override clause was driven by the bad-faith intention to get rid of limitations on the government’s power. ⁶ Regarding the European example, the purpose of the German ministers’ involvement in judicial governance is to provide courts with democratic legitimacy and secure their accountability. In the case of Poland, however, many feared that the executive’s intention was to capture and neutralise courts as checks on its policies. ⁷

This article puts less emphasis on intent and, instead, contributes to the debate on detecting abusive constitutionalism by zooming in on effects of abusive practices. While I do not deny that most abusive constitutional practices are probably driven by bad faith, its detection in the constitutional realm can be rather impracticable at times. ⁸ Bad faith is an elusive concept, concentrating on the actor’s state of mind. In constitutional politics, the rulers’ internal motives are not easily accessible, mixed intentions are frequent in collective political bodies, and ulterior partisan purposes are common too. Even if we replace subjective with objective bad faith (unreasonableness), we still face the hurdles of legitimate disagreements about the desirable structure of the constitutional system and of significant varieties in constitutional design across democracies. These features complicate devising a practicable and persuasive method of detecting abusive constitutional practices and distinguishing them from non-abusive ones.

The concept of abusive constitutionalism, however, is not based solely on intent. The second part of its definition stresses the harmful effects on democracy. This article argues that going deeper into the analysis of mechanisms and effects of abusive constitutionalism provides a major opportunity to improve the possibilities of detecting abusive constitutional practices early on. With several notable exceptions, ⁹ however, the effects side of the concept of abusive constitutionalism is rather underdeveloped, probably due to a greater emphasis on intent. Refocusing on this effects side of abusive constitutionalism, I introduce the foreseeable effects test as a scholarly device designed for an early detection of harmful constitutional practices that are likely to advance democratic decay. It is conceived as a transparent and structured framework for analysing foreseeable effects of a constitutional practice and its abusiveness. The test is based on insights from multiple jurisdictions and recent theories of democratic decay. The test’s aim is not to evaluate the reasonableness or proportionality of constitutional practices. Rather, it is a framework designed to facilitate calling a constitutional practice (non-)abusive. It can work as a complementary or, possibly, alternative tool to intent analysis when detecting abusive constitutionalism.

Using the benchmark of substantial diminishment of accountability in the system, the foreseeable effects test consists of three steps. ¹⁰ Step 1: probability of harm analyses the prospect that the constitutional practice in question will substantially impair an existing accountability mechanism, particularly its autonomy and operability. Step 2: seriousness of potential harm analyses the foreseeable effects of the constitutional practice in the broader constitutional context. It asks whether adequate accountability mechanisms would remain in place once the tested practice is implemented. Particular attention is paid to the interaction of the tested practice with the surrounding constitutional structure and past and parallel constitutional developments. Step 3: harm mitigation checks whether adequate safeguards mitigating the identified harm accompany the constitutional practice. The three steps are inherently interconnected and the final verdict results from the overall assessment of the three steps together. I demonstrate the steps of the test on examples from various parts of the world hit by democratic decay.

While the foreseeable effects test is admittedly no panacea and has its own weaknesses, it contributes to the discussion about abusive constitutionalism as it can complement the intent-focused approaches in important respects. The test’s focus on foreseeability and on the interplay of various accountability mechanisms facilitates the early detection of harm, which is a crucial challenge in the context of the current incremental erosion of democracy. Next, the test guides us towards asking productive questions about abusive constitutionalism. It directs us towards the strengths of comparative constitutional studies: a thorough and contextualised institutional analysis, with an infusion of legal realism and increased sensitivity to constitutional politics. The foreseeable effects test makes the reasoning about what makes a constitutional practice abusive more transparent, structured, and open to contestation. The test, especially in synergy with intent analysis, can thus improve our understanding of abusive constitutional practices. These practices emerge as both a manifestation and a cause of the current crisis of constitutional democracy. They represent a crucial infrastructure for achieving shifts away from democracy towards hybrid regimes. As the first step to tackling those, however, a thorough method of detecting abusive constitutional practices is necessary.

The rest of this article proceeds as follows. The first part (Bad faith as an element of abusive constitutionalism), defines abusive constitutional practices and shows that some important approaches to their detection are based on uncovering the bad faith intent. The subsequent part (Obstacles to detecting the bad faith intent) explains possible challenges to detecting bad faith in constitutional law and politics. To bridge (some of) these downsides, the next part redirects our attention to the effects of abusive constitutional practices. It introduces the foreseeable effects test as a complementary or alternative method to intent analysis. It sets and justifies the test’s normative benchmark, explains its steps, and accounts for its limits. Finally, it demonstrates the test’s feasibility on a case study from Hungary. The illustrative case study of the (unimplemented) reform plan to introduce a separate system of administrative courts in Hungary shows the challenges of detecting abusive constitutionalism early on and the advantages of the foreseeable effects test in addressing some of them. The last part concludes.

Bad faith as an element of abusive constitutionalism

In the past decade, the ‘feel good ethos’ surrounding constitutionalism has been weakening. ¹¹ The post-Cold War era was optimistic about constitutionalism facilitating the flourishing of liberal democracy. Accordingly, scholars focused on the global spread of liberal constitutionalism and judicial review, the emergence of global constitutionalism, and the migration of constitutional ideas. ¹² More recently, however, optimism has faded. Reflecting the developments in socio-political reality, scholarship focusing on the rise of authoritarian populism, abusive constitutionalism and ensuing democratic decay, deterioration of the rule of law and human rights has started to dominate the field. ¹³

Scholars have provided crucial insights into the mechanisms of contemporary democratic decay, understood as an ‘incremental degradation of the structures and substance of liberal constitutional democracy’. ¹⁴ Two important characteristics of the process have been stressed: legalism and incrementalism. First, compared to earlier authoritarians, contemporary architects of democratic decay rely extensively on constitutional and legal forms rather than on violence and extra-legal measures. They engage with constitutional tools and use them to advance their goals, often in contradiction to the values underlying these constitutional concepts. ¹⁵ That is often accompanied by the use of constitutional democratic rhetoric as a smokescreen. Traditional notions such as popular will, popular sovereignty, and constitutional identity are reinterpreted and deployed to justify a subversion of constitutional governance. ¹⁶ Importantly, these trends are not limited to new or weak democracies. They have affected numerous countries, including those previously labelled as well-consolidated democracies. ¹⁷ Second, coups and sudden reversals are no longer the main mechanisms of dismantling a constitutional democracy. They have been replaced by an incremental erosion, consisting of a series of steps, each of which might be tolerable on its own. Their combination, however, leads to a considerable weakening of democratic checks. ¹⁸ Such incrementalism makes it extremely difficult to identify the breaking point of democratic resilience.

Reflecting upon these findings, the assumption about a natural affinity between constitutional tools and flourishing democracy, the rule of law, and human rights has weakened. A more realistic view sees constitutional concepts as double-edged swords. ¹⁹ Such an approach acknowledges that the instruments of constitutionalism have done good on numerous occasions in numerous places, but also recognises their darker side. While different scholars have coined various labels to capture the latter, this article uses the umbrella term abusive constitutional practices. ²⁰ Borrowing Dixon and Landau’s definition, I understand these as ‘the appropriation of liberal democratic constitutional designs, concepts, and doctrines in order to advance authoritarian projects’. ²¹

While it is possible to imagine abusive constitutional inertia, I mostly focus on instances of constitutional change, broadly understood. Abusive constitutionalism can take the form of a constitutional amendment or replacement, ²² sub-constitutional legislative changes, ²³ judicial reinterpretation, ²⁴ or factual practices employed by political actors. Importantly, the abusive practices rely on mechanisms that exist in well-functioning constitutional democracies, with Western constitutionalism and international human rights law as the main sources of inspiration. ²⁵ However, the abusers tend to decouple the substance from the form of the borrowed constitutional concepts. While taking advantage of the form, they do not follow the values and principles underlying the concept. ²⁶ Abusive constitutionalism is based on decoupling the constitutional tools from their purpose. That opens the door to using the constitutional tools to erode democracy. The core of the practice is surface-level compliance with the constitutionalist ideas while producing effects that the same constitutional institutions are supposed to block, minimise, or discourage. Such an approach allows the abusers to claim allegiance to constitutional democracy or, at least, makes criticising them for undemocratic practices much harder. In sum, abusive constitutional forms aim to reduce the costs of advancing democratic decay.

Despite all the advancements in the understanding of democratic decay, however, there is a serious issue with the primary step of detecting abusive constitutional practices and distinguishing them from non-abusive ones in an objective and persuasive way. Abusive constitutionalism bears fruit only if seen as somewhat legitimate by outside observers. Accordingly, the instrumental logic of abusive constitutionalism dictates that the abuse should not be too obvious. In addition, the way these practices play out in reality is highly context-sensitive. As Tushnet put it, ‘almost every specific constitutional development can be a valuable reform in some contexts and something that weakens constitutional democracy in other contexts’. ²⁷ The incremental nature of democratic decay further exacerbates this problem. The huge significance of context thus precludes the possibility of devising an abstract list of abusive practices. ²⁸

How then do we know that a constitutional practice is abusive? Some important approaches in the existing scholarship put a lot of weight on intent. Abusive constitutional practices are distinguished by the bad faith anti-democratic intent that accompanies them – ‘the intention to do harm’. ²⁹ What we get as a result is a ‘misappropriation’ of constitutional concepts. ³⁰ The new autocrats use constitutional law with the intention to undermine democracy and advance their anti-democratic projects. ³¹ Sajó thus argues that illiberal regimes are orders of ‘cheating and deceit’ – they are based on using legal forms with the intention to cheat through law. Working with pretention and hypocrisy, illiberal leaders present themselves as norm observers while disregarding the deeper principles of the rule of law. They rely on standard constitutional institutions, the rule of law, and human rights, but do so only instrumentally and without genuine commitment. ³² Drinóczi and Bień-Kacała concur that ‘instrumental and opportunistic use of the law’ is crucial for achieving and sustaining illiberal constitutionalism. ³³ In such regimes, values of liberal constitutionalism are formally accepted but the real intent is to act against them – to disrupt the structures and mechanisms of constitutional democracy. ³⁴

Dixon and Landau, in their seminal book on abusive constitutional borrowing, mostly work with the intent-oriented approach. ³⁵ They generally approach the concept as ‘intentional use of borrowing as a tool for eroding democracy’. ³⁶ However, they also pay attention to the anti-democratic effects of abusive practices and make a conceptual distinction between thin and thick abuse:

The thinner version focuses only on the effect of relevant practices: in this understanding, constitutional or legal changes could be viewed as ‘abusive’ when they have a significant adverse effect on the minimum core of democracy. A thicker understanding, in contrast, requires that would-be authoritarians knowingly or intentionally take aim at the democratic minimum core—or forms of constitutional ‘bad faith’. ³⁷

In sum, bad faith represents an important element of the concept of abusive constitutionalism and of some of the existing approaches to detecting abuse. This article does not aim to cancel intent-centred approaches, as I do not deny that most abusive constitutional practices are likely driven by bad faith. Indeed, there is an agent-relative element of abuse as some of the new authoritarianisms depend on particular political actors’ ability to gain popularity for their practices. However, there are a number of problems, as bad faith is an elusive concept that is hard to detect in an objective and persuasive way. These issues are further exacerbated in the realm of constitutional law and politics, the objects of which are complex institutional arrangements characterised by diverging intentions and legitimate disagreements about the desirable structure of the political system. In addition, we have recently been witnessing a trend of normalisation of abusive constitutional practices, ³⁸ which seems to reduce the significance of the agent-relative element of abuse. As a result, we may face serious obstacles at times when trying to distinguish abusive constitutional practices from non-abusive ones based on intent. Therefore, I take Dixon and Landau’s thin account of abusive constitutionalism as a starting point for further inquiry. The rest of this article refocuses from intent to effects and suggests that a more structured focus on effects of abusive constitutionalism can help to overcome some of the challenges of distinguishing between abusive and non-abusive practices. I introduce the foreseeable effects test as a feasible approach facilitating the diagnosis of abusive constitutionalism. Depending on one’s understanding of abuse (thick or thin), the test can work in synergy with intent analysis or, possibly, on its own. First, however, the next section elucidates the challenges related to bad faith and explains how the effects test could complement intent analysis and improve our understanding of abusive constitutionalism.

Obstacles to detecting the bad-faith intent

According to the Cambridge Dictionary, bad-faith action is understood as an activity ‘done in a dishonest way with the intention of tricking someone’. ³⁹ Relying too much on bad-faith intent in detecting abusive constitutional practices can be tricky, though. Bad faith is an elusive concept – it is subjective, often inaccessible, and unreliable. The domain of constitutional law makes this even more challenging, as widespread legitimate disagreement about the desirable organisation of political power casts doubt on the traditional proxies of bad faith such as unreasonableness.

Bad faith is an inherently subjective internal category. It focuses on the actor’s state of mind. It involves improper motives, deception, malicious purpose or belief, or fraud. ⁴⁰ Investigating one’s state of mind, however, often gets very difficult and implies epistemic difficulties. It is especially troubling for constitutional scholars who are neither psychologists nor investigative journalists equipped to analyse the ruler’s state of mind or their inner motives. In this respect, subjective bad faith is difficult to establish and ascribe to particular actors, especially if they are multimember organisations. The field of comparative constitutional studies lacks a widely accepted methodology for uncovering true intent. Accordingly, there is a danger that too much focus on intent in detecting abusive constitutionalism can turn into a ‘I know it when I see it’ exercise, with all its downsides: dangers of arbitrariness, subjectivity, and self-referential thinking based on gut feeling.

To be fair, authors provide some guidelines for making judgements about intent. Dixon and Landau, for instance, argue that, when ascertaining the relevant actors’ intent, we can look at what they say and what they do, seek indications of procedural irregularities, and check the broader context of a particular constitutional change. ⁴¹ It is true that these may be helpful indicators of malicious intent. Yet, while the idea of focusing on context is crucial, it is rather vague and potentially all-encompassing. Procedural irregularities can often accompany abusive constitutionalism, but not necessarily. The most powerful rulers in particular have the political resources to comply with the legally foreseen procedures. Analysing the discourse surrounding a constitutional reform can be helpful, but also overinclusive. Politics is opportunistic and the practice of changing positions to take advantage of a situation is not limited to constitutional abusers. More generally, these indicators are still confronted with the hurdle of the inherently internal and largely inaccessible nature of bad faith. Extensive focus on the actors’ state of mind also risks excessive personalisation After all, not everything a populist government does is automatically bad or unconstitutional. ⁴² In sum, the risk is ‘not only to wade into an evidentiary and epistemic morass but also to invite endless finger-pointing’. ⁴³

Some of the troubles generated by the internal nature of bad faith may be resolved by refocusing from subjective to objective bad faith. Objective bad faith comes close to the standard of (un)reasonableness. It concentrates on unreasonable conduct, tested against the norms of a legally relevant community. ⁴⁴ The idea is to resolve the difficulties of operationalising subjective bad faith by referring to the conceptions and understandings predominant in a given community. ⁴⁵ In the constitutional realm, it focuses on unwarranted deviations from constitutional standards. Objective bad faith thus requires an inquiry into constitutional standards in a particular constitutional area in a given region.

The main problem with objective bad faith and unreasonableness is that there are considerable varieties in constitutionalism(s). ⁴⁶ A regional inquiry will often reveal different and complex constitutional designs. Clearly, there are exceptions characterised by a transnational consensus, such as presidential term limits. That still leaves us with numerous unclear cases. Consider an example of the recently divisive area of judicial independence in Europe. Well-established democracies span from highly self-governing judiciaries (Italy, France) to executive-governed ones (Germany, Austria). A similar variety in constitutional design can be found in other areas, including those often targeted by constitutional abusers: electoral systems, executive-legislative relationships, constitutional amendment, the availability and design of constitutional review, fourth branch institutions, and the regulation of political parties. Objective bad faith thus cannot solve all the problems, as the current world of constitutionalism offers a reasonable blueprint for a wide array of designs, institutions, and doctrines. As Pozen put it, ‘It can be hard to ascertain which legal positions are unreasonable in a field where vague standards predominate and everyone is constantly debating the proper way to do law.’ ⁴⁷

An additional problem is the non-binary nature of bad faith. While legal norms often employ either the good or the bad faith distinction, in reality bad faith is a scale. ⁴⁸ Moreover, constitutional policies involve complex decisions by complex and often collective actors. ⁴⁹ Consider the story of the 99^th amendment of the Indian Constitution. To reform judicial appointments to higher courts the amendment established the National Judicial Appointments Commission. The aim was to increase judicial accountability, strengthen the executive’s voice in judicial selection, and weaken the voice of judges. ⁵⁰ This reform, however, was seen by some scholars as a democracy-eroding measure. ⁵¹ It was eventually abolished by the Supreme Court for violating the principle of judicial independence as a part of the Constitution’s basic structure. ⁵² From the perspective focusing on intent, it is important to add that the Modi government did not push the amendment alone. The amendment had considerable cross-party support. ⁵³ It is unclear what the broader support for the reform – which suggests a plurality of intentions – means for the intent analysis.

The Indian example reflects a more general feature of mixed intentions likely underlying many constitutional practices. After all, constitutional politics is still politics. It is often partisan and opportunistic. The processes of making, interpreting, and applying constitutional law usually include the interaction of diverging intentions. Accordingly, there is an additional challenge of setting a threshold as to how much bad faith is necessary (and how bad it must be) for a practice to qualify as abusive.

None of this is to downplay the perils of abusive constitutionalism. On the contrary, this section demonstrated the elusiveness and additional dangers related to abusive constitutional practices. Neither do I suggest treating abusive constitutionalism as a legitimate practice. The fact that bad faith is an elusive concept should not make us stop our inquiry and simply admit that abusive constitutionalism is elusive and hard to detect. The trend makes for a critical mechanism in the process of democratic decay. And it is necessary to understand the practice properly to design effective responses. But the nature and logic of abusive constitutionalism make even the very first step necessary to tackle this phenomenon – its detection – highly challenging. Recognising these problems, the next section sketches a possible way forward. I devise the foreseeable effects test as, hopefully, a viable alternative or a complementary tool to intent analysis.

Harmful effects as an element of abusive constitutionalism: towards the foreseeable effects test

Constitutional practices are instrumental. They serve as means to achieve ends. The ends are usually related to good governance principles such as democracy, effectiveness, or transparency. In the case of abusive constitutionalism, however, the use of constitutional means is accompanied by bad-faith intent to achieve a covert or ulterior end that harms constitutional democracy (see Figure 1). As showed above, when detecting abusive constitutionalism, some approaches put a lot of weight on the bad-faith intent. The intent analysis is important, as it focuses on the agent-relative element of abuse and checks the congruence between the proclaimed and the real aim of the constitutional practice. However, as the previous part argued, detecting bad-faith intent may face challenges. In constitutional politics, the rulers’ internal motives are often inaccessible, mixed intentions are frequent in collective bodies, and ulterior partisan purposes are common too. Even if we replace subjective with objective bad faith (unreasonableness), we still face the hurdle of significant variations in constitutional design across democracies. As a result, it may be hard at times to identify bad faith in an objective and persuasive way.

Figure 1. Situating the foreseeable effects test

I believe that one way of improving our abilities to detect abusive constitutional practices is to analyse the foreseeable effects of these practices in a more structured and focused manner. However, in terms of effects that make a constitutional practice abusive, the existing conceptions have not been too specific – with some exceptions – probably due to the focus on intent. Dixon and Landau’s account is the most transparent, as it sets the concept of the democratic minimum core as the threshold for assessing abusive constitutional effects. Essentially, democratic minimum core denotes an electoral account of democracy. It consists of free and fair elections and a set of institutions, procedures, and rights necessary to maintain a competitive democracy. These, Dixon and Landau explain, necessarily vary from country to country and, therefore, specific context plays a huge role. ⁵⁴

I agree that context is fundamental to the exercise of detecting abusive constitutionalism. However, the prominence of abusive constitutional practices as the legal infrastructure of democratic decay pushes us to go deeper towards a more detailed course of action for analysing the effects of constitutional practices. While context dependence surely precludes a comprehensive abstract list of abusive practices, we can still try to generalise and devise a more specific and structured method based on the common abusive patterns cognisable in decaying countries across regions and contexts. As explained below, this article argues that the major common pattern lies in the incumbents’ efforts to cement themselves in the positions of power by accountability-avoidance strategies. The normative benchmark and all parts of the foreseeable effects test are based on this feature.

The foreseeability of a constitutional practice’s effects is not straightforward either, still less in the context of democratic decay. Probabilities may be hard to ascertain in specific cases. The legalism and incrementalism inherent in the current wave of democratic erosion make things even more complicated. Most critically, incrementalism creates interaction effects. While the individual constitutional measures may look unproblematic, their interaction may lead to harmful effects: composed of ‘various perfectly reasonable pieces’, the constitutional system’s ‘monstrous quality comes from the horrible way that those pieces interact when stitched together’. ⁵⁵ In order to detect such ‘Frankenstate’ strategies, Kim Lane Scheppele suggested employing forensic legal analysis, understood as a ‘series of “what if?” questions to test for interaction effects’. ⁵⁶ This approach points in the right direction. However, it lacks a clear normative benchmark and is not specific enough to provide a comprehensive course of action for detecting abusive constitutionalism. ⁵⁷

In response, this part proposes the foreseeable effects test, primarily intended as a scholarly instrument. It is designed for the early detection of harmful constitutional practices likely to lead to democratic decay. The test is based on insights from multiple jurisdictions into the theory and practice of abusive constitutionalism. Rather than providing a device for clearcut answers, it is conceived as a framework for more structured and transparent reasoning about what makes a constitutional practice abusive and why (on the effects side), attuned to the disciplinary strengths of constitutional studies.

Setting the normative benchmark: substantial diminishment of accountability

The first task in establishing the effects-centred test for detecting abusive constitutionalism is to set the normative benchmark. I set the benchmark as a substantial diminishment of accountability in the constitutional system. It is a relative, floating standard. Constitutional democracy is a matter of degree. Relatedly, democratic decay has different starting points, trajectories, and varies in pace across countries. Given the incremental nature of decay, a significant drop in democratic quality may be worrisome and abusive even if it does not cross the line of a minimal standard for what still counts as a constitutional democracy. Therefore, rather than focusing on a democratic minimum core, I concentrate on relative shifts in accountability.

I chose accountability as the key regulatory ideal behind the test as it is a core value of democratic constitutionalism shared by democrats of all stripes. ⁵⁸ It is crucial for securing nondomination in the constitutional system, as it aims to exclude any power imbalance that would allow the rulers to exercise personal rule. ⁵⁹ Thereby, accountability enables partisan alternation in office, which is essential for democratic self-government. ⁶⁰ However, accountability mechanisms are regularly targeted by the architects of democratic decay and abusive constitutionalism. ⁶¹ Their accountability-avoidance practices increase the costs of unseating incumbents. If substantial, they likely erode partisan alternation and facilitate one-partyism and personal rule. I focus on accountability rather than a narrower category of checks and balances, reflecting the ways in which current democratic decay and its countering work. Existing research suggests that halting democratic decay requires the interplay of various types of accountability actors, including popular mobilisation and strategies of opposition parties, rather than checks and balances on their own. ⁶²

Still, accountability is a complex concept which requires unpacking. I understand accountability as a mechanism, ‘an institutional relation or arrangement in which an agent can be held to account by another agent or institution’. ⁶³ In this sense, it is important to explain who is accountable to whom and, relatedly, based on what kind of obligation. ⁶⁴ As to the first question, I consider the rulers – the political executive and, in parliamentary systems, the parliamentary majority that backs them – as actors obliged to render an account. Executives seem to be the main drivers of the contemporary wave of democratic decay, since executive aggrandisement manifests itself as the common denominator of backsliding trends around the globe. ⁶⁵ As to the second question, I consider electoral/vertical accountability to voters through free and fair elections, horizontal accountability to oversight actors checking and balancing the rulers, and diagonal accountability to civil society and independent media. ⁶⁶

Since the current wave of democratic decay is typical in that it targets horizontal accountability mechanisms, ⁶⁷ I pay particular attention to this dimension. I focus on the role of courts, legislatures (especially parliamentary opposition), fourth branch institutions (such as human rights commissions, ombudspersons, or electoral commissions), and international checks such as international courts and quasi-judicial bodies. I do not see these mechanisms as replacing electoral accountability, but rather as instruments enabling and facilitating it. Given the power asymmetry between the rulers and citizens, electoral accountability requires the empowerment of people. ⁶⁸ It is a collective mechanism. While the people can use the election to hold usurping governments accountable, such actions have huge information and coordination costs to be effective. It can happen only if the people have reliable information about the government’s behaviour and if they are able to act together in a coordinated way. ⁶⁹ Horizontal accountability actors offer a significant reduction in both the information and coordination costs of monitoring the rulers. Courts’ and other independent institutions’ outputs provide accessible, public, and structured information about the government’s behaviour. Such outputs also have a signalling function and can serve as focal points for the people’s coordination. ⁷⁰ Hence, horizontal accountability mechanisms empower people and facilitate vertical accountability by providing them with information and a locus for coordination. ⁷¹ If captured, however, these mechanisms can themselves be turned into promoters of abusive constitutional practices.

Nonetheless, not all adjustments of accountability mechanisms should be seen as abusive. There may be legitimate reasons to modify such mechanisms. Accountability ‘overkill’ can discourage the innovative thinking that may be necessary to meet the challenges of this complex era. ⁷² I recognise that state actors need power and room for political action to be effective and meet the tasks of the state; my goal is not to protect an ultra-minimal state. ⁷³ That is why the threshold of the foreseeable effects test is set as a substantial diminishment of accountability. The question is when the diminishment qualifies as substantial. Due to its context dependency, it is difficult to answer this question in the abstract. The three steps of the foreseeable effects test provide more detailed guiding questions. On the most general level, however, I view diminishment of accountability as substantial if it gives actor A the power unilaterally to endanger other actors’ performance of their core duties of holding A accountable.

Step 1: probability of harm

The first step of the test focuses on the probability of harm, i.e., the prospect that the constitutional practice in question will substantially impair an existing accountability mechanism. It considers the design of the tested constitutional measure and asks about the foreseeable situation after the implementation of the measure: to what extent would the constitutional measure affect the possibility of holding the rulers accountable through a particular accountability mechanism? In the case of horizontal accountability actors – the main targets of accountability-avoidance strategies – the key concern is accountability avoidance via institutional capture (by large-scale personnel changes), paralysis (by defunding, interferences with internal procedures, etc.), or other ways of setting accountability mechanisms aside (such as resistance against international supervision). Accordingly, it is important to pay attention not only to the links between vertical and horizontal accountability but also to the crucial preconditions of horizontal accountability actors’ effectiveness: operability, and autonomy. As O’Donnell explained:

For this kind of accountability to be effective, there must exist state agencies that are authorised and willing to oversee, control, redress, and if need be sanction unlawful actions by other state agencies. The former agencies must have not only legal authority but also sufficient de facto autonomy vis-à-vis the latter. ⁷⁴

I understand operability as the degree to which an actor is sufficiently resourced to have the capacity to deliver its task of holding another actor accountable. Typical measures impairing operability include defunding, jurisdiction stripping, meddling with internal procedures, appointment failures, and other measures designed to paralyse the institution. In Poland, for instance, the Prawo i Sprawiedliwość government adopted a series of amendments to the Constitutional Tribunal Act that changed its decision-making procedure and other internal features. Sadurski reported that the reform temporarily paralysed the Tribunal and effectively exempted the new legislation from constitutional review. ⁷⁵ Several international courts, such as the SADC Tribunal and the WTO Appellate Body, were temporarily paralysed due to a member state’s opposition to judicial appointments. ⁷⁶ Not even legislative bodies have escaped paralysation attempts, as the case of proroguing the UK Parliament shows. During the crucial phase of Brexit negotiations, Boris Johnson’s government tried to get Parliament suspended by way of an unusually long prorogation for five weeks. While Parliament reconvened after the UK Supreme Court’s intervention, Johnson’s move was seen as a way to avoid accountability to the legislative branch. ⁷⁷

Besides being able to hold the rulers accountable, the actors must be willing to do so. For that, they must enjoy sufficient autonomy, i.e. freedom from control by a recognisable outside faction or interests. They must be ‘capable of (i) developing and (ii) expressing preferences that are substantially distinct … from those of a single dominant outside actor …’. ⁷⁸ Autonomy goes beyond formal independence, as it is also a matter of nonpartisan orientation and professional independence. ⁷⁹ The typical measure interfering with autonomy is institutional capture, which appears when an accountability actor is redirected from following public interests to promoting partisan interests. This is frequently achieved through personnel changes, particularly by hand-picking appointees who are likely to remain loyal to the rulers and less likely to hold them accountable. The National Electoral Council in Venezuela illustrates this. After Hugo Chávez won the presidential election, he started a radical constitutional overhaul of Venezuela. Within the field of electoral management the new constitution created the National Electoral Council as a non-partisan five-member body. Nonetheless, Hawkins reported that during the phase of Chávez’s power consolidation the executive managed to obtain control over the Council’s composition: ‘by 2005 any pretense of nonpartisanship had been abandoned and the board consisted of a majority of Chavistas’. ⁸⁰ Subsequently, the Council was said to have contributed to an ‘electoral environment plagued by irregularities and governed by a biased regulatory agency’. ⁸¹

In sum, the first step of the test focuses on the foreseeable consequences of the constitutional measure in terms of operability and autonomy of the accountability mechanism. The following questions are crucial: what would be the resulting institutional links between the accountability actor and the rulers? What would be the consequences for the accountability actor’s capacity to deliver its tasks? What would be the consequences for the accountability actor’s ability to develop and express its autonomous voice?

Step 2: seriousness of potential harm

While the previous step of the test identifies the likely effects of the constitutional practice on a specific accountability mechanism, this step concentrates on the effects in the broader constitutional context. The focus is on the practice’s interactions with the existing constitutional structure, constitutional history including past and parallel reforms, and on the consequences for the overall system of accountability mechanisms. The core question is whether the implementation of the tested constitutional measure would leave an adequate level of accountability mechanisms in place. A major concern is to tackle incrementalism and interaction effects as major features of abusive constitutionalism. The point is to test the constitutional practice in the context of the adjacent legal-political environment with which it would interact. ⁸² This inquiry requires the checking of surrounding constitutional norms and institutions but also circumstantial evidence of the broader context in which the practice takes place.

Varol explained the incremental and cumulative effects increasing the seriousness of harm on the example of Turkey:

Many of Erdoğan’s practices, standing alone, appeared insignificant. But their strength lay in their sustained accumulation. Over time, the pressure on democratic institutions built up, like increased hydraulic pressure on a water pipe. These measures slowly, but surely, corroded the already shaky foundations of Turkey’s democracy—until the pipe eventually burst. ⁸³

To tackle the dangers of treating constitutional practices in an isolated way, it will often be necessary to undertake a systemic assessment of the reformed institution and its links in a complex way. Importantly, it is necessary to anticipate the reform’s effects in interaction with past and parallel developments, with the existing constitutional structure and with the broader legal-political environment. The abovementioned idea of granting the Knesset competence to override the Israeli Supreme Court’s judgments illustrates this. The overall constitutional context makes the risk serious, as Israel’s constitution is (partly) unwritten and very easy to amend. Simultaneously, the unicameral nature of the legislature, lack of vertical checks, fragmented party system, and eroding constitutional culture further reduce the real possibilities of holding the rulers accountable. ⁸⁴ Moreover, as Weill explained, in parallel with the legislative override mechanism, the government announced a broader reform, including changes to the Supreme Court’s procedure and practice of judicial review as well as changes to the appointments mechanism in a way that considerably strengthens the ruling majority’s influence on appointments. ⁸⁵ The combination of all these features considerably increased the seriousness of the potential harm.

To address such issues, the following questions are crucial: what is the place and history of the reformed institution in the system of accountability mechanisms? How does the tested constitutional practice change it? What are the timing and sequencing of the constitutional practice, including past and parallel reforms? How does the constitutional practice interact with surrounding constitutional norms and institutions? How does this change the overall structure of accountability mechanisms in place?

Step 3: harm mitigation

The third step focuses on the existence and adequacy of harm-mitigating measures accompanying the tested constitutional practice. The point of the debate about abusive constitutionalism is not to perpetuate the status quo at any cost. As various scholars have noted, there may be good reasons for reforming accountability mechanisms, including cases of accountability overkill, ⁸⁶ highly unresponsive or dysfunctional accountability actors, ⁸⁷ but also instances of democratic reconstruction after a period of backsliding. ⁸⁸ At the same time, however, scholars of abusive constitutionalism warn against the dangers of abusing such ‘just causes’ as a smokescreen for advancing covert anti-democratic aims. The populist rhetoric which nowadays often accompanies abusive constitutionalism facilitates this. The populist ideology is well equipped to reinterpret fundamental constitutional concepts such as democracy, popular sovereignty, and constitutional identity in ways that make abuse easier. ⁸⁹ Additionally, the intentions motivating a constitutional measure may be mixed, as argued above. Accordingly, preserving a realistic possibility of holding the rulers to account and preventing sliding into one-partyism should be the central consideration here as well. This part of the test is designed to make room for non-abusive adjustments of accountability mechanisms while taking seriously the risk of abusing just causes for a reform. ⁹⁰

The main point is to examine whether harm-mitigating safeguards accompany the tested constitutional practice and analyse their adequacy. Such safeguards can mitigate the probability of harm, seriousness of harm or both aspects. They can be found in the design features of the tested constitutional measure and in the proposed mode of its implementation, but they can also take the form of additional institutional or procedural safeguards in the adjacent constitutional environment. They include, for instance, the introduction of additional accountability and oversight mechanisms, procedural checks, participatory instruments, time limitations (sunset clauses), or staggering personnel changes in time. Such safeguards, however, cannot be taken at face value. They have to be critically evaluated to make sure they are (foreseeably) effective and adequate with respect to the identified harm.

Importantly, Step 3 is not the ultimate, decisive part of the test. The assessment of harm mitigation measures goes hand in hand with the previous aspects of the test. The final verdict on the abusiveness of a constitutional practice is a result of the overall assessment of the three steps together. There is an interactive relationship between assessing the probability and seriousness of harm and the harm-mitigating safeguards. ⁹¹ The previous steps of the test allow us to specify the suspicions about the practice and their gravity. They help us ascertain what the most acute risks and the most dangerous features of the tested measure are, in which contexts and why. Such identified risks are crucial and should lead the analysis of the existence and adequacy of necessary harm-mitigating safeguards accompanying the dubious practice. Combining these considerations leads us to the overarching question of the test: How does the tested constitutional practice, given its design, interaction effects and harm-mitigating features, affect the overall prospect of holding the rulers to account within the constitutional system as a whole?

A good example for illustrating the rationale of Step 3 is the executive empowerment during the Covid-19 pandemic. Since such empowerment took place in many countries, the example well demonstrates how to differentiate between similar measures coming in various designs in various contexts with different harm-mitigation safeguards. The pandemic was a shock that justified unprecedented steps to protect public health. Doing so often required strong and swift executive action. Accordingly, some countries, including those on the track of democratic decay, activated (quasi-)emergency powers, empowered the executive, and simultaneously relaxed some of the legislative and judicial checks. ⁹² This was a logical and often necessary step, since the executive had what protecting core public interests during the pandemic required: access to expertise, and the capacity for swift and decisive protective measures. Still, the executive empowerment increased the probability and the seriousness of harm in backsliding democracies. ⁹³ When assessing the harm, however, the scope of accompanying harm-mitigating measures should be considered too. ⁹⁴ To cite specific examples from the first wave of Covid, New Zealand established an opposition-led parliamentary committee to balance the government’s aggrandised powers in responding to the pandemic. ⁹⁵ In South Africa, the position of Covid-19 designated judge was created to balance the surveillance powers of the executive. ⁹⁶ Other countries used sunset clauses designed to put clear time limits on the extraordinary executive empowerment. ⁹⁷ Hungary, on the other hand, adopted the Enabling Act. According to several commentators, the new legislation gave the Prime Minister almost unlimited powers to govern by executive decree without providing truly effective safeguards. ⁹⁸ This variety demonstrates the interactive logic of Step 3. The measures adopted in New Zealand and South Africa aimed to compensate for the executive strengthening by introducing additional accountability mechanisms. In contrast, several studies warned that the lack of safeguards in Hungary dramatically increased the risks. ⁹⁹

Case study: administrative judiciary in Hungary

Another example from Hungary well serves to demonstrate how the entire test works in practice. While the previous sections have explained the individual steps of the foreseeable effects test, this one applies the test to the reform plan of establishing a new separate branch of administrative courts in Hungary. As an illustrative case study, based on the findings of the Venice Commission and country-specific scholarship, it aims to give the reader a ‘feel’ for the practical application of the test and the related difficulties. ¹⁰⁰ Hungary was chosen as a backsliding regime demonstrating the main challenges of detecting abusive constitutionalism, ¹⁰¹ including legalism, incrementalism, and comparative justifications. In addition, this case well illustrates the possibility of conducting an analysis of foreseeable effects without the benefit of hindsight – the reform plan was not implemented in the end.

Viktor Orbán’s government in 2018 introduced a Bill establishing a new separate system of administrative courts, headed by the new Supreme Administrative Court. Until then, judicial review of administrative acts had been exercised by ordinary courts. The new administrative courts were supposed to have broad jurisdiction over politically highly salient issues such as elections, asylum, assemblies, the media, and tax law. Critics argued that this was a court-packing plan designed to further the government’s grip on checks and balances. ¹⁰² Yet, the government supported its proposal with comparative and historical reasons and used arguments about improving the efficiency of the administrative judiciary. In general, as the Venice Commission argued, there was no principled reason to oppose the idea of having a distinct administrative court system, as it was ‘perfectly compatible with European standards’. ¹⁰³ However, the devil was in the details negatively affecting the new administrative courts’ autonomy, in interaction effects, and in the lack of adequate harm-mitigating measures.

What made the probability of harm high was the combination of the broad powers of the new administrative judiciary with the extensive administrative, budgetary, and personnel competences of the minister of justice. ¹⁰⁴ Most critically, the minister of justice was given far-reaching discretionary powers to appoint new judges of the administrative courts. ¹⁰⁵ Some of the judges were to be recruited from volunteers from the existing courts. These were supposed to be supplemented by new appointees. While the selection process included the involvement of judicial self-governance actors, the final call was taken by the minister, who was allowed to alter the ranking of candidates. ¹⁰⁶ On top of that, the minister could determine the number of administrative judges in the new system. The combination of these two competences would have given the minister the power to adjust the share of new appointees and, thereby, alter the power balance within the new courts, largely free of institutional constraints. ¹⁰⁷

Checking the broader constitutional context and probing into previous constitutional reforms of accountability mechanisms point to a considerable seriousness of the potential harm. The relevant constitutional context did not suggest pressing needs to reform the administrative judiciary. The Venice Commission seemed unpersuaded by the government’s reasons for reform based on historical, comparative and efficiency-increasing grounds: ¹⁰⁸ ‘[t]he various reports drawn up at European level in recent years would suggest that Hungarian administrative justice is not facing any particular problems’. ¹⁰⁹ In contrast, the interaction effects seemed risky. The Hungarian post-2010 party system has been dominated by Orbán’s Fidesz party (and its smaller coalition partner KDNP – Christian Democratic People’s Party), which has won four parliamentary elections in a row with a super-majority necessary to amend the constitution. ¹¹⁰ The ruling majority adopted a new constitution in 2011 and since then has gained considerable control over various horizontal and diagonal accountability mechanisms. ¹¹¹ Concerning the judiciary, the design and composition of the Hungarian Constitutional Court were altered, which led to a considerably more deferential constitutional review. ¹¹² Subsequently, the Supreme Court was reformed and the mandates of its leaders were prematurely terminated. ¹¹³ By lowering the retirement age for judges, the government also removed a number of senior judges in the ordinary courts from their functions. ¹¹⁴ Considering all these past reforms and their combined effect, the significance of an autonomous administrative judiciary for the possibility of holding the rulers to account becomes clear. Várnay and Varju argued that by the time of the reform, judicial review by administrative courts remained ‘almost the only practical opportunity to oppose and control an executive’. ¹¹⁵

As regards harm mitigation, the reform plan largely lacked sufficient guarantees and checks to reduce the aforementioned risks. ¹¹⁶ The government later reacted to the Venice Commission’s criticism by partly modifying the reform plan and adding features to mitigate the harm. ¹¹⁷ However, Várnay and Varju reported that these changes did not seem sufficient to mitigate the harm adequately:

While the modifications did address some of the criticisms raised by the Venice Commission, it would be difficult to argue that risks that may arise from the cumulative effect of the provisions regulating the new system, especially when implemented in the current political setting in Hungary, could be dismissed. ¹¹⁸

As a result, the test suggests that the plan to introduce a new system of Hungarian administrative courts was a high-risk measure in terms of its foreseeable effects on the accountability of the rulers. The analysis of inter-institutional linkages and relevant constitutional context pointed to a substantial risk of erosion of accountability in Hungary that was not accompanied by adequate harm-mitigating safeguards. The episode had a surprising ending. In 2019, the government dropped the reform plan and postponed it indefinitely. Accordingly, the proposed system has not been put into practice. Nonetheless, that makes it an ideal case for demonstrating the feasibility of the early detection of harm through the foreseeable effects analysis.

Strengths and weaknesses of the foreseeable effects test

The detailed explanation and demonstration of the foreseeable effects test aimed to show that the test can work as a practical tool for detecting abusive constitutionalism on the effects side. Effects- and intent-oriented approaches to abusive constitutionalism, however, are not mutually exclusive. The detection exercise is likely to be most successful if the two approaches are combined, as they both have their pros and cons. While I have summarised the drawbacks of bad faith approaches above, it is fair to account for the limitations of the effects test.

First, the test reduces rather than eliminates some of the troubles related to uncovering intent. Admittedly, it does not provide for a clear-cut device that detects abusive constitutionalism in a noncontroversial way completely free of political judgement. I am afraid that no approach is fully capable of that. In addition, the focus on foreseeable effects always includes a speculative element. We inevitably face limits on foresight and predictability, especially due to the abusers’ incentives to disguise their moves. Nonetheless, I believe that one of the strengths of the foreseeable effects test is that it guides us towards asking more productive questions. It provides a clear normative benchmark and makes the reasoning about what makes a constitutional practice abusive more transparent, structured, and open to contestation. The foreseeable effects test directs us to a thorough and contextualised institutional analysis.

Second, to accommodate non-abusive adjustments of accountability mechanisms, the test’s benchmark – substantial diminishment of accountability – is rather indeterminate. However, a certain level of vagueness and flexibility is inescapable. A test of this kind will always include open-ended terms allowing for the consideration of contextual differences. Still, to minimise the indeterminacy, above I have provided a general definition of the benchmark and included more specific guiding questions in each step of the test. Admittedly, some disagreements on whether the threshold has been met are likely to remain. However, the foreseeable effects test at least incentivises transparent and structured reasoning about the core of the disagreement.

Third, the foreseeable effects test detects some types of abusive constitutional practices better than others. It is better designed for identifying abusive reforms of accountability mechanisms and abusive judicial review than it is for abusive utilisation of constitutional rights. ¹¹⁹ I believe that the test’s usefulness is high, nevertheless. After all, incremental attacks on horizontal accountability actors and their subsequent use to advance authoritarian goals represent the most common and most challenging features of democratic decay. ¹²⁰

In sum, I believe that the strengths of the foreseeable effects test outweigh its weaker features, since it responds to the distinctive challenges of the current democratic decay. Most importantly, its focus on foreseeability and on the interplay of various accountability mechanisms facilitates the early detection of harm. In the context of the incremental erosion of democracy, early detection of abusive practices is essential, since preserving multiple accountability mechanisms is crucial for halting democratic decay. ¹²¹ A more objective and structured institutional analysis may be an important addition to intent analysis in tackling this task. In most cases, it takes time for the rulers to build a reputation as constitutional abusers and even more time to gather the contextual evidence of bad-faith intent.

My point, however, is not to eliminate the element of bad faith. That could distort the concept of abusive constitutionalism and bring overinclusive results, as it could become difficult to distinguish abusive from simply sloppy constitutionalism in some cases. This article stressed the substantial accountability reduction as a crucial element of abusive constitutionalism, which can be used to distinguish abusive from merely sloppy constitutionalism. Still, I believe the conclusions about abusiveness of a constitutional practice can be strongest when the intent and effects analyses are combined. The goal of this article has not been to cancel the intent-centred approaches but rather to point to relevant challenges and strengthen the effects-focused prong of the concept of abusive constitutionalism.

Several objections, however, may remain. First, it can be questioned whether a single test is appropriate when abusive constitutionalism is such a context-sensitive phenomenon. The foreseeable effects test aims to provide a framework for a consistent assessment of abusiveness of constitutional practices, but it does not encourage a one-size-fits-all approach. It allows a guided flexibility, designed to take into account relevant contextual factors in a structured and focused manner linked to a specified normative benchmark. Second, some may see the test as too strict, perpetuating the status quo, while others may view it as too permissive. It should be reiterated that the test is not about evaluating reasonableness or proportionality of the constitutional practice. It is simply a framework designed to facilitate calling a practice (non-)abusive, based on its foreseeable effects. ¹²² Yet, calling a practice non-abusive does not necessarily imply a good practice, let alone the best possible one. The test also acknowledges that there may be good reasons for adjusting accountability mechanisms (hence substantial diminishment of accountability as a benchmark) and creates room for doing so in a non-abusive way. Yet, the test makes the probability and seriousness of accountability avoidance its central consideration and emphasises the significance of harm-mitigating measures in such cases. After all, evaluating a constitutional practice as abusive does not make all reforms of the given mechanism forbidden. It points to a high probability of serious harmful effects of the specific way and design of a particular reform in a particular context.

Conclusion

Abusive constitutionalism manifests itself as a governance strategy employed by rulers with an autocratisation agenda. It can contribute to ‘gaslighting’ the external audiences ¹²³ or, at least, reduce the costs of democratic decay. Abusive constitutional practices have, therefore, become the backbone of the autocratic legal infrastructure and they keep on spreading around the world. Proper diagnosis of abusive constitutionalism is thus a crucial task of comparative constitutional studies. As the necessary precondition of a thorough diagnosis, more reliable detection methods are necessary to distinguish abusive constitutional practices from non-abusive ones. The foreseeable effects test introduced in this article provides one such method.

Devising a good detection method is important for analytical purposes – to make clear what exactly is wrong about these practices and why – as well as for the future reconstruction of the constitutional order. As regards the analytical dimension, the foreseeable effects test is centred on the crucial value of accountability for democracy. It builds on the insight that the major pathway of the current democratic decay is based on accountability-avoidance practices leading to consolidation of the incumbents’ positions in power. The test is designed to detect practices characterised by a high probability and seriousness of harm, understood as a substantial diminishment of accountability.

The future ‘healing’ of abusive constitutionalism is a huge, separate topic. This article has focused exclusively on the analytical dimension. Still, looking at abusive constitutionalism through the prism of effects on accountability implies several starting points for the healing or reconstructive endeavours. First, incremental attacks on accountability institutions – the distinctive feature of the current democratic decay – imply incremental effects. That suggests that relative reductions in accountability indicate troubles that may warrant a response even if the situation does not fall below a minimal democratic standard. ¹²⁴ Approaches relying on a minimum threshold may not be ideally designed for a well-timed response capable of maintaining accountability. Second, the healing should focus on restoring the overall accountability in the system as a mechanism facilitating partisan alternation. That, however, can be achieved only by the interaction of several accountability mechanisms in the political system, which will provide voters with reliable, visible, and accessible information, and a fair opportunity to act on the information. Rather than singling out one institution as a savior of democracy (typically an apex court), a focus on a broader range of institutions, their discrete institutional strengths and weaknesses, and their synergy seems necessary. Finally, the foreseeable effects test is based on a mixed analysis of formal and informal (de facto) features of abusive constitutionalism. The reliance of abusive constitutionalism on constitutional tools demands a focus on formal institutions, whereas the reality of abusing such tools urges us to focus on their relationship with the de facto dimension and its political logic. Thus, effective constitutional restoration will likely have to tackle the interaction of the formal and the informal in abusive constitutionalism, which may require one to dig more deeply into the path-dependent origins of such practices.