Parliamentary Supremacy versus Judicial Supremacy How can adversarial judicial, public, and political dialogue be institutionalised?

The battles between proponents and opponents of judicial and parliamentary supremacy have characterized the institutions of supreme constitutional courts right from the beginning, but have intensified in the second half of the 20th century in the US and, more recently, in Europe as well. The aim of this article is not to rehearse the longstanding debates of judicial review in political philosophy, legal theory and comparative constitutionalism, although they will be summarized in Section 2. If it is true that judicial supremacy and parliamentary supremacy are both empirically and normatively indefensible, if there is not one model of liberal–democratic constitutionalism (LDC) but many competing and historically changing institutions and practices of judicial review, then we have to change the terms of the debate. If there is not one, objectively right and context-independent answer, if no one has the last word in LDC, we are not doomed or drowned in scepticism, decisionism, or nihilism. Instead it is a matter of degree, of better or worse (in this context, at this time), and we can learn. Such learning, I submit, can profit from two traditions: firstly, from recent discussions of how to democratize expertise (including judicial expertise) and to expertise democracy in general, particularly under conditions of high complexity, contingency, unpredictability and of contestedness and uncertainty of knowledge in recent societies (the tradition of critical pragmatism and of experimentalist governance); and secondly, from the tradition of democratic institutional pluralism and, in particular, from moderately agonic associative democracy. Democratic institutional pluralism, roughly speaking, refers to an extension of well-known majority-restraining elements of democracies in the field of political/territorial representation – such as executive power sharing, separation of powers, balanced bicameralism, proportional representation, territorial federalism and decentralisation, written constitutions – to social/functional representation (highlighted by associative democracy) and cultural minority representation. These two new approaches have, to my knowledge, not yet been used to deal with problems of judicial review. Both require a shift from abstract discussions of principles to contextualized institutional imagination and learning from developing practices. Before spelling this out in more detail in Section 5, I focus on judicial review in two different contexts. In Section 3, I address judicial review in the traditional context of nation states and ask whether the Canadian notwithstanding clause might provide a rational way forward, if properly amended. In


Introduction
The battles between proponents and opponents of judicial and parliamentary supremacy have characterized the institutions of supreme constitutional courts right from the beginning, but have intensified in the second half of the 20th century in the US and, more recently, in Europe as well. The aim of this article is not to rehearse the longstanding debates of judicial review in political philosophy, legal theory and comparative constitutionalism, although they will be summarized in Section 2. If it is true that judicial supremacy and parliamentary supremacy are both empirically and normatively indefensible, if there is not one model of liberal-democratic constitutionalism (LDC) but many competing and historically changing institutions and practices of judicial review, then we have to change the terms of the debate. If there is not one, objectively right and context-independent answer, if no one has the last word in LDC, we are not doomed or drowned in scepticism, decisionism, or nihilism. Instead it is a matter of degree, of better or worse (in this context, at this time), and we can learn. Such learning, I submit, can profit from two traditions: firstly, from recent discussions of how to democratize expertise (including judicial expertise) and to expertise democracy in general, particularly under conditions of high complexity, contingency, unpredictability and of contestedness and uncertainty of knowledge in recent societies (the tradition of critical pragmatism and of experimentalist governance); and secondly, from the tradition of democratic institutional pluralism and, in particular, from moderately agonic associative democracy. Democratic institutional pluralism, roughly speaking, refers to an extension of well-known majority-restraining elements of democracies in the field of political/territorial representation -such as executive power sharing, separation of powers, balanced bicameralism, proportional representation, territorial federalism and decentralisation, written constitutions -to social/functional representation (highlighted by associative democracy) and cultural minority representation. These two new approaches have, to my knowledge, not yet been used to deal with problems of judicial review. Both require a shift from abstract discussions of principles to contextualized institutional imagination and learning from developing practices. Before spelling this out in more detail in Section 5, I focus on judicial review in two different contexts. In Section 3, I address judicial review in the traditional context of nation states and ask whether the Canadian notwithstanding clause might provide a rational way forward, if properly amended. In

Veit Bader
Utrecht Law Review | Volume 12 | Issue 1, 2016 For our discussion, it seems important to spell out two things. First, strong or weak systems of judicial review are 'not binary alternatives but labels that mark areas along a spectrum' 17 and, secondly, one should clearly distinguish between strong judicial review and judicial supremacy. Obviously, judicial supremacy requires institutions and practices of strong judicial review, but not all of them amount to judicial supremacy: 'Judicial supremacy refers to a situation in which (1) the courts settle important issues for the whole political system, (2) those settlements are treated as absolutely binding on all other actors in the political system, and (3) the courts do not defer to the positions taken on these matters in other branches (not even to the extent to which they defer to their own past decisions under a limited principle of stare decisis).' 18 These distinctions will now be used for three purposes: for conceptual clarification; for some historical remarks; and for some initial normative remarks.

Conceptual clarifications
The concept of 'constitutional supremacy versus parliamentary supremacy' misleadingly suggests that the decisive role of legislators would be ipso facto unconstitutional; 19 the concept of 'popular constitutionalism' or 'political constitutionalism' versus 'legal constitutionalism' 20 suggests that a decisive role of legislative assemblies would not be legal. The relevant focus should be on monopoly claims by constitutional or supreme courts ('judicial monopoly on constitutional interpretation' 21 ) and 'judicial supremacy' seems the most adequate concept in this regard.

'Judicial supremacy' does not exist 22
The distinction between institutions and practices of strong judicial review and judicial supremacy allows us to show that countries such as the US, with institutions of strong judicial review, are also characterized by an ongoing war between defenders and critics of judicial supremacy. Even the briefest summary of Larry Kramer's American constitutional history in a nutshell 23 can show this: the US Constitution originated within a system of 'popular constitutionalism' but a competing model of judicial supremacy was formulated in the mid-1790s by the conservative Federalists (as against the French revolution and the 'baneful influence of Faction' and majoritarian threats to minority rights). It was repudiated in 1800-1802 (repeal of the Judiciary Act) but re-emerged in the early 1830s, in a different guise, emphasizing the 'settlement function' of law and arguing that there would be no end to controversy on slavery, tariffs, banks, and states' rights unless judges had the final say. Yet the Jacksonians challenged judicial authority and won the 1832 presidential election by a landslide. From then on, one finds periodic confrontations with no clear victor or predominant underlying understanding. This pattern began to change only in the second half of the 20th century. From 1958 (Cooper v. Aaron) onwards, the idea of ideal of judicial supremacy found active and widespread acceptance by conservatives (using the traditionally liberal rhetoric of counter-majoritarianism as well as that of judicial restraint developed against the New Deal Court) and, more reluctantly and ambivalently, by liberals as well, particularly from the mid-1960s onwards. The Warren Court's liberal activism encouraged a liberal philosophy of broad judicial authority by adding a new argument to the 'precautions against majoritarianism' and to 'settlement': namely that 'courts are better and more trustworthy than electorally accountable bodies in questions of principle '. 24 Yet older concerns for judicial restraint survived. 25 For the 'in form' (p. 49) as well as the importance of political and judicial 'culture' (p. 50) but it is misleading to mix up legal form and 'fact' conceptually and it is premature to claim that culture 'is far more important than the institutional form' (p. 50). 17 Fallon, supra note 2, p. 1733. 18 Waldron, supra note 5, p. 1354, referring to Barry Friedman. 19 Besselink's criticism of an earlier draft of this paper. questions whether this is a correct description' which it clearly is not. 23 Ibid., all following quotes from Kramer, supra note 3, pp. 962-967; see Kuo, supra note 6, in Section III on 'constitutional departmentalism vs. judicial supremacy'. 24 See Dworkin, supra note 5 and R. Dworkin, Taking Rights Seriously (1977). See C. Eisgruber, Constitutional Self-Government (2001). 25 E.g. Ely, supra note 5.

Parliamentary Supremacy versus Judicial Supremacy
How can adversarial judicial, public, and political dialogue be institutionalised?
Utrecht Law Review | Volume 12 | Issue 1, 2016 first time in American history, conservatives and liberals found themselves in agreement on the principle of judicial supremacy which came to monopolize constitutional theory and discourse, a monopoly that thrived during the tenure of Chief Justice Burger and persisted into the early years of the Rehnquist Court. 26 2.1.4. 'Strong but minimalist' judicial review?
If and as long as the respective defenders of judicial or parliamentary supremacy think that these are logically opposite and mutually exclusive arrangements and that one has to play a zero-sum game, this belief has important effects not only theoretically but also in the 'real world'. It excludes productive forms of contestation/dialogue and institutional co-decision arrangements from the start. Only if strong and weak forms of judicial review are not seen as exclusive alternatives but as 'areas along a spectrum' (see text and note 17 above), can a debate that matters in the real world get off the ground. In this regard, it is important to note that even defenders of a strong and maybe ultimately supreme role of legislation and/or 'the people' such as Jeremy Waldron or Richard Bellamy, are conceding that in some circumstances -not 'healthy' but 'pathological' polities/societies -and in some cases weak review is needed (see note 53 below). If one brackets the construction of ideal or 'nearly ideal' worlds/circumstances and focuses on real worlds and if one recognizes that even under ideal circumstances we seriously disagree, then it seems plausible to develop two connected types of theories: (i) substantive theories of moral minimalism aiming at 'preventing malfare' and 'serious injustices' instead of 'maximizing welfare' or 'justice in the ideal world' (see Section 2.2.4 below), and (ii) proceduralist theories of adversarial communication including deliberation, dialogue/multilogue, contention, arbitration and strategic argument instead of consensus and ideal deliberation (see Section 2.3.3 below). Together, these theories can be used to defend strong but minimalist judicial review (see Fallon and others).
A detailed and necessarily contextualized discussion of the advantages and disadvantages of different varieties of strong judicial review and of weak or absent judicial review is way beyond this article. 27 Instead I present the most well-known advantages and disadvantages of both (see Sections 2.2 and 2.3 below), as claimed by their respective proponents based upon their respective normative tasks or functions under ideal conditions (what supreme courts/judges and legislators ought to do and are in the best position to do). This is followed by a slightly debunking assessment of what both have done and do in the real world and by a sober discussion of their comparative institutional competences in gradational terms of 'better/worse'. I end with a brief discussion of the institutional and cultural conditions for realistic improvement, given the conclusion that both monopoly claims are empirically unsound as well as normatively unwanted.

Advantages of strong judicial review
The core arguments in favour of judicial supremacy or, more modestly, of strong judicial review have already emerged from my summary of Kramer's historical sketch (see above Section 2.1.3): the 'settlement function', the guarantee of basic rights and, in particular of minority rights, and the superiority of supreme courts as a 'forum of principle'.

The settlement function
In general, the normative claim that there has to be some settlement of contested issues is fairly uncontentious. 28 Yet the claim that judicial supremacy is the best way to settle such issues is as contested as the empirical claim The pre-commitment justification for judicial supremacy 29 may be plausible for those constitutional questions having clear answers, yet the distinction between clear and unclear answers is contested and, if anything, this can be and is provided by weak judicial review as well. In addition, most of these questions require choices among plausible alternatives in general, and particularly in a context of uncertainty. 'Subsequent developments may sometimes come to make one of the alternatives appear wrong, even preposterous.' 30 There is not only the 'risk that people (or the politically accountable institutions through which they speak) will make bad decisions' 31 because of irrational and biased impulses, but also judges and supreme courts have done and do so.

The forum of principle
Counter-majoritarianism and who is the 'best interpreter'? 32 The supremacy of judges and supreme courts is justified either by the claim that they guarantee best results in terms of social justice and protection of individual liberties as well as a counterweight against democratic excesses: namely, it may be countermajoritarian 'but this is good because it checks "the tyranny of the majority"'. 33 Or the supremacy is justified by a claim about 'the general nature of legal versus political institutions: the structure of the judiciary -its independence, the setting in which it deliberates, the requirement of drafting an opinion, and so forth -is said to give us confidence that judges can generally be expected to do a better job' 34 compared with other political actors, particularly parliaments. The first empirical claim (better or best results) can be countered by reference to more or less extreme bias and prejudice in historical 35 or actual Supreme Court rulings in different countries. In fair historical and contemporary comparisons -and the comparative perspective is crucialcountries with weaker or even without judicial review (such as the UK, New Zealand, the Netherlands, Sweden or France) demonstrate that this generalized claim is unfounded and untenable: they may be 'at least as free and just as the US'. 36 As I have already indicated above (Section 2.1.1), institutions of 'parliamentary sovereignty' (lacking constitutional review) vary according to the following variables: (i) whether or not parliaments are bound constitutional issues'); see also N. by a written Bill of Rights -whether a national or supra-national one (regional conventions such as the European Convention on Human Rights (ECHR)) or an international one (UN Charta); (ii) whether legislative power is bound (as in the Netherlands or Canada) by a written constitution or not (as in the UK); (iii) whether there are competing, more or less representative political institutions -usually first chambers (as in the UK (the House of Lords) or in the Netherlands (Eerste Kamer)) or additional controlling bodies (such as the Conseil Constitutionnel in France) -checking the legal quality of legislative decisions but not formally or actually having the powers of constitutional review (which is, after all, the decisive distinction between parliamentary and judicial 'supremacy'). Together with more or less vibrant, even activist jurisprudential traditions (as is usual in case-law systems), these variables impact on both the quality of legislation and the effectiveness of rights protection. In addition, the quality of rights protection very much depends on other aspects, such as predominantly legal and cultural traditions, habits and virtues of judges, voluntary or enforced traditions of deference, lively public debate, etc.
The second claim, to start with, is not in itself an argument in favour of the supremacy of constitutional courts but of the judiciary in general (judges as heroes). It has been countered on many empirical grounds such as the kinds of reasons and deliberative settings ('good academic workshop'); the kinds of allowable reasons (judges are not experts in relevant social, political, cultural arguments in changing societal conditions); professionalism, institutional insulation of judges and courts -not in dialogue and hence often uninformed, in the rear-guard. 37 Yet the core arguments by Waldron and others challenge the philosophical defenders of judicial supremacy on their own turf: the 'existence and persistence of pervasive disagreement'. No one is or can be in possession of the 'truth' about rights because there are no 'objectively right answers out there'. 38 Also, there is no shared method or technique for resolving normative disagreements because we disagree as much about what counts as a justification as we do about what rights we ought to have (see Section 2.3.2 below).

Defending 'strong but minimalist' judicial review
These are indeed strong arguments against judicial supremacy but, to repeat, not necessarily against any form of judicial review, as is also clear in Waldron's and Kramer's writings that allow for weak judicial review. Nor do they disqualify all arguments in favour of strong but minimalist judicial review which do not presuppose or depend on an ideal moral consensus or any 'objective truth' about what 'justice requires', but on much less demanding agreement on preventing serious injustice and serious violations of core fundamental or basic rights, even if their interpretation and application may also be contested. On the basis of such a substantive theory of moral minimalism 39 one has good reasons to defend strong but minimalist review. According to Fallon, the crucial question is not whether courts or legislators are better at 'defining rights correctly' or are 'less likely to err' but 'which kinds of errors are most important to avoid'. If the task is to 'minimize the most morally grievous errors' or to prevent the violation of 'fundamental rights', these rights 'deserve to be protected by multiple safeguards', 40

Veit Bader
Utrecht Law Review | Volume 12 | Issue 1, 2016 preferable given the following conditions or premises: (1) 'Even if courts are no better overall at identifying rights violations than are legislatures, courts have a distinctive perspective that makes them more likely than legislatures to apprehend serious risks of rights violations in some kinds of cases.
(2) Legislative action is more likely to violate fundamental rights than legislative inaction. (3) Some rights are more important than others and, accordingly, are more deserving of protections against infringement. (4) A system of judicial review can be so designed that the moral costs of such over-enforcement of rights as judicial review would produce will likely be lower than the moral costs that would result from such under-enforcement of rights.' 41 Such a design has to answer three difficult questions: First, 'which claims of rights'? (minimalism of fundamental rights). Secondly, 'what should be the scope' or 'how searching' and 'stringent'? (deference and self-restraint; see Section 2.3.2 below). Thirdly, 'should judicial review be entrenched against legislative override ("strong") or subject to legislative displacement ("weak")'? 42

Disadvantages of judicial supremacy
Thus, arguments against the philosophical defenders of judicial supremacy are not in themselves arguments against any form of judicial review, nor are they arguments in favour of 'parliamentary supremacy' or even for something like 'popular constitutionalism' (as it seems to be for Kramer) 43 or for 'political constitutionalism' (Bellamy). 44 They are directed against judicial monopoly claims and, at least in my version, against any supremacy or monopoly claims whatsoever. In this section, I will briefly address two well-known core arguments against judicial supremacy, the 'democratic legitimacy' argument (Section 2.3.1) and the 'separation of powers' argument (Section 2.3.2) before discussing the 'reasonable disagreement' argument (Section 2.3.3) and how to deal with it substantively and institutionally (Section 2.3.4).

Weak or non-existent democratic legitimacy
The democratic legitimacy of judges and constitutional courts is non-existent or weak. 45 Following the old distinctions between 'legitima auctoritas, legitima potestas and legitima decisio' one can distinguish institutional, personal, and substantive democratic legitimacy. 46 Democratic legitimacy requires that the institutions of constitutional review and constitutional courts should be in line with the political sovereignty of the people -established either through referenda, or elected constitutional assemblies or by qualified majority voting (QMV) decisions of parliaments in cases of amending constitutions -and that the organization and competences of constitutional courts are regulated in constitutions. Democratic institutional legitimacy is absent or 'weak if constitutional review is not contained in the constitution and further elaborated in law but rather imposed by judiciary power' 47 as, for example, in the US or in the Weimar Republic. Judges ought to be independent and this inherently limits traditional measures to increase personal democratic legitimacy either through direct election of judges of supreme courts by the people (which does not exist anywhere), or dismissal -both measures are inimical and incompatible with independence. Yet there can and should be something like an 'uninterrupted chain of legitimation' 48 with a double focus: firstly on election by parliament (versus selection through co-optation by constitutional courts or nomination/appointment by executive powers (president, government)) and, secondly, for a limited time instead of life-time appointments (e.g. a maximum of 9 years for the ECtHR or 12 years for the 41  German Bundesverfassungsgericht), and by qualified majority voting. 49 Concerning substantive democratic legitimacy of the content or result of decisions, it seems plain that judges ought to be/are bound by laws and the constitution cannot provide more than formal legitimacy, particularly because constitutional rights and principles leave wide margins of discretion/appreciation and because (as I have already stated above and will explain in Section 2.3.3), constitutional rights and principles conflict with each other and need to be weighed and balanced. However, the argument that the democratic legitimacy of constitutional courts/ judges is non-existent or weak does not mean that constitutional judicial review could not or does not contribute to the political legitimacy of liberal-democratic states, because protecting the rule of law and fundamental rights against violations is one crucial element of 'liberal' or 'outcome'/'output' legitimacy. 50

Judicial activism and violation of the separation of powers
The main danger of judicial supremacy and of institutions and traditions of strong non-minimalist judicial review is that courts overstep the formal competences 51 of the judiciary and violate basic principles of the separation of powers. Under the impact of predominant theories and traditions of judicial supremacy, courts are not only tempted to violate legislative political powers but often explicitly do so, doing 'ordinary politics' in the guise of protection of the constitution and, by doing so, close the legitimate open space that LDC should create and leave for important economic (e.g. 'market socialist'), societal (e.g. associative democracy) and cultural alternatives. In this regard, it does not matter whether this judicial activism is motivated by and connected with conservative and/or libertarian 52 or progressive/egalitarian or socialist ideologies/politics. All are equally incompatible with traditions and practices of stronger judicial restraint required by open democratic politics. Under conditions of LDC, judicial review may be strong but it has to be minimalist (see above Section 2.2.4). 53 Obviously, institutions of strong constitutional review can go hand in hand with practices of judicial restraint 54 but, comparatively speaking, the danger of illegitimate judicial activism by constitutional courts is much weaker and more easily countered in countries with weaker constitutional review.

Why persistent reasonable disagreement?
The three most important reasons are as follows. Firstly, constitutional principles and basic rights are, by their very nature, abstract, general and underdetermined -containing not enough constraints to specify a unique solution -and this includes that there is not and cannot be 'one right interpretation' independent of context. Secondly, and in addition, basic rights are not only underdetermined but also conflict with each other (famously: conflicts between different freedom rights and between 'liberty' and 'equality'; pluralism

Veit Bader
Utrecht Law Review | Volume 12 | Issue 1, 2016 of rights) which means that rights are not just trumps -either trumping other rights or outranking all other considerations -but have to be weighed and balanced and this weighing and balancing, again, is context-dependent and reasonably contested. Thirdly, constitutional principles and rights, as well as their underlying moral principles spelled out by LDC, are not only in tension with each other but also explicitly open-ended. Constitutions and human rights conventions (such as the ECHR) are 'living instruments which have to be interpreted in the light of prevailing conditions and ideas in democratic states' (ECtHR) 55 and they are embedded in changing societal, political and legal cultures. This means that there cannot be one ultimate or 'final' interpretation and, also, that 'original intent' approaches and an exclusive focus on written law and 'texts' have to be rejected. For these reasons, no one is in the possession of 'truth', no one 'has the last word' and there is no consensus even under ideal conditions, let alone under more realistic conditions. Institutionally, this means, as already stated, the final flaw of judicial supremacy. 56 Theoretically, this is the common ground and core insight of critical pragmatism and realist deliberation, as well as, more radically, critical legal studies in opposition to both Dworkinian and Habermasian deliberative democracy. 57 Demanding theories of 'deliberative consensus' in ideal worlds should be replaced by proceduralist theories of moderately agonic contestation in the real world under conditions of powerasymmetries and are explicitly critical with regard to the 'rationalist' and 'exclusivist bias'. Adversarial public talk (Barber) or, more broadly, adversarial communication, ranges from actual 'deliberation' and 'talk' via 'dialogue or multilogue', 'contention' and 'arbitration' to 'strategic contestation'. 58

How to deal with reasonable disagreement and to increase substantive democratic legitimacy
The respective mechanisms range from dissenting opinions and from competition/cooperation between constitutional courts (both internal to judicial institutions but less shielded against non-legal institutions and politics) through competition/cooperation between supreme courts and with political institutions -most importantly legislators -to broader public talk and politicization. Following on from the seminal article by Peter Hogg and Allison Bushell in 1997, a broad and sophisticated debate on constitutional dialogue developed, 59 focusing on six issues. (i) The concept itself: opposing monologue by dialogue, multilogue or polylogue. (ii) Its aim, quality or character: a 'process of reaching consensus' -even if 'dynamic' 60 -or an ongoing adversarial compromise and modus vivendi; 'one-way' versus 'two or many ways'; explicit or implicit, formal or informal. 28-50) also accepts that a constitution is a living instrument and 'is to function within time, not beyond it. Its interpretation needs to take account of this.' It is questionable whether 'the term of the justices sitting in a constitutional court, provide a better outlook than that of the members of the legislature' (p. 33). Yet this also works against the democratic legitimacy of judicial supremacy in the 'Kelsenian' model, defended by Besselink, in which it is, ultimately, 'derived' from 'popular sovereignty'. Under conditions of a vibrant liberal-democratic 'political culture' and of'‚deference to the legislature' (p. 33) this 'tension between the legislature and the judiciary' may, indeed, be softened but it should be acknowledged as a matter of principle before one goes into comparative discussions of alternative institutional settings that, indeed, 'require a delicate balance' (p. 35 courts, between courts and legislators (parliaments, sub-committees etc.), government and executive committees, and between 'non-judicial actors' and 'citizens' or broader political, societal and cultural actors (social movement organisations, NGOs, political parties etc.) and networks. 61 (iv) Levels: from state/ Länder via federal to supra-state jurisdictions and polities. (v) Time: short-, medium-, long-term. (vi) Its contested effects: from symbolic window-dressing (either favouring judicial or parliamentary supremacy) to important juridical, political and socio-cultural effects in interpretations and changes of constitutions and constitutional review. Even a short overview of these debates is way beyond the limits of this article. 62 Instead I present some of the promising potential emerging from actual experiences and theoretical discussions of constitutional dialogue.
(1) Democratic habits of judges and the careful argumentation and justification of decisions (e.g. by reasonable balance tests and proportionality tests) 63 can contribute a lot to increasing substantive legitimacy, but the main internal seedbed may be the toleration and flourishing of dissenting opinions (first in the United States, later in Germany, Spain, Portugal and Greece and by the ECtHR). The main advantages of this tradition are: 64 (i) dissenting judges force the majority of judges to argue and justify their position carefully. It serves as an important internal mechanism of control. (ii) It contributes to 'rechtsvorming' (developing law) because dissenting opinions may become dominant or majority opinions. (iii) It may increase the societal and political acceptance of decisions because the arguments of losing parties are spelled out. Far from undermining the authority of constitutional courts, the publication of dissenting opinions and voting ratios may actually help to increase the quality and persuasiveness of the decision and, hence, the substantive legitimacy of constitutional review. 65 It also makes it easier to bring hidden political judgments and biases (e.g. class, gender, ethno-racial, religious) to the fore, making them vulnerable in public and political debate and increasing 'difference-sensitivity'. All in all, it may help to increase the judicial literacy of the (interested) public and contribute to learning about the open, flexible, plural character of Liberal-Democratic Constitutions, to acknowledge the legitimacy of deep diversity of perspectives and in the finding of reasonable compromises. It is one of the most important internal means to fight isolated, overassertive and rigid courts and strategies of caste-like self-isolation of the dominant judiciary or, conversely, to strengthen 'passive virtues'. 66 (2) Competition/cooperation between multiple jurisdictions, 67 particularly under conditions of more 'horizontal coupling', less clear or contested 'hierarchy' or explicit 'heterarchy' 68 actually contributes to the perception of reasonable, deep disagreement amongst judges 69 and amongst legal experts from courts,

Veit Bader
Utrecht Law Review | Volume 12 | Issue 1, 2016 parliamentary committees, the legal profession generally, and human rights NGOs. Under favourable conditions it also may help to increase public and political awareness (see Section 5).
(3) Opening up forms of dialogue/contestation and co-decision with non-judicial, political institutions, particularly with legislators, is crucial to combat judicial supremacy, to increase or even enforce traditions and practices of judicial deference in constitutional courts, and to increase substantive democratic legitimacy (see Section 2.3.1). Particularly under 'non-ideal' conditions they are not meant to replace all forms of constitutional review but to complement them, 70 because there is a clear need to counter the three wellknown dangers of parliamentary sovereignty or supremacy. The first danger is that the quality of laws is not checked for the standards of the rule of law by an independent body. The second is that the rights and freedoms of citizens (and residents) may not be effectively protected by the legislative body. 71 The third is that simple legislative majorities are vulnerable when it comes to overriding minority rights.
(4) Opening up forms of dialogue with and contestation by a broader range of societal, political and cultural actors (such as social movement organisations, civil society organisations, NGOs, political parties and a variety of interested stakeholders in the specific issue-areas at stake, 72 various media and the general public) may help to create and stimulate informed and critical democratic politicization of jurisdiction.

Concluding remarks
Let me finish by summarizing my main general conclusions. Firstly, in constitutional issues there is not and cannot be 'one', best, optimal and ultimate interpretation and decision. Secondly, in LDC no one -neither judiciary, nor politicians, nor other experts -should have the 'last word' and all monopoly claims or supremacy claims should be rejected, whether by supreme courts or parliaments. Thirdly, decisions and institutional arrangements depend on contexts. There is no one best or optimal arrangement, neither for democracy more generally 73 nor for constitutional review and amendment. This also speaks against the export/import of -usually idealized -'models', but it does not prevent comparisons, modest lessons as guidelines, and learning from each other. Fourthly, the most important thing is to counter insulation and closure by opening up, stimulating and institutionally anchoring broad judicial, political and societal dialogue/contestation. Fifth, in this regard competition amongst institutions with similar, overlapping competences -such as overlapping multiple jurisdictions but also multiple legislative institutions -should be reconsidered or re-evaluated. Finally, the main problem is how adversarial dialogue, cooperation and competition can be made more clearly focused, informed and reasonable: how and by whom can it be organized and orchestrated in such a way?

Lessons from the Canadian notwithstanding clause
As we have seen, institutions and traditions of strong and extensive or unlimited judicial review make any substantive dialogue with or control by legislative bodies very difficult and ideologies/practices of judicial supremacy do not allow it. In this sense constitutional courts and parliaments are completely asymmetrical and the decisions by supreme courts are unimpeachable. The legislature is bound by court decisions, not vice versa. In this section I address constitutional review within nation states and discuss the Canadian notwithstanding clause which seems to provide opportunities to avoid a stand-off and for opening up substantive communication between Parliament and Supreme Court. It can be seen as an 'intermediate form' or a 'halfway house' 74 between judicial and parliamentary supremacy or sovereignty, or between strong and extensive judicial review (as in the US or Germany) and weak or even non-existent judicial review. I start with 70 Unfortunately, this is not spelled out clearly enough by Bellamy, supra note 57, nor by Waldron, supra note 5, and Kramer, supra note 3, though all direct their attack against judicial supremacy and make it clear that different forms of weaker judicial review may be necessary.

Parliamentary Supremacy versus Judicial Supremacy
How can adversarial judicial, public, and political dialogue be institutionalised?
Utrecht Law Review | Volume 12 | Issue 1, 2016 minimal introductory remarks (3.1), present arguments against (3.2) and in favour of the clause (3.3) and suggest some proposals for revision and improvement (3.4) before outlining some preliminary lessons.

The Canadian notwithstanding clause
Here is a brief, authoritative summary of the Canadian notwithstanding clause or override clause in Section 33 of the revised Canadian Charter of Rights and Freedoms (1982) by Johansen and Rosen: 75 'Section 33(1) permits Parliament or a provincial legislature to adopt legislation to override section 2 of the Charter (containing such fundamental rights as freedom of expression, freedom of conscience, freedom of association and freedom of assembly) and sections 7-15 of the Charter (containing the right to life, liberty and security of the person, freedom from unreasonable search and seizure, freedom from arbitrary arrest or detention, a number of other legal rights, and the right to equality). Such a use of the notwithstanding power must be contained in an Act, and not subordinate legislation, and must be express rather than implied. Under section 33(2) (…) the overriding legislation renders the relevant Charter right or rights "not entrenched" for the purposes of that legislation. In effect, parliamentary sovereignty is revived by the exercise of the override power in that specific legislative context. Section 33(3) provides that each exercise of the notwithstanding power has a lifespan of five years or less, after which it expires, unless Parliament or the legislature re-enacts it under section 33(4) for a further period of five years or less. A number of rights entrenched in the Charter are not subject to recourse to section 33. These are democratic rights (sections 3-5 of the Charter), mobility rights (section 6), language rights (sections 16-22), minority language education rights (section 23), and the guaranteed equality of men and women (section 28). Also excluded (…) are section 24 (enforcement of the Charter), section 27 (multicultural heritage), and section 29 (denominational schools) -these provisions do not, strictly speaking, guarantee rights. All rights and freedoms set out in the Charter are guaranteed, subject to reasonable limitations under the terms of section 1. This has the effect, in combination with section 32 of the Charter (making the Charter binding on Parliament and the legislatures) and section 52 of the Constitution Act, 1982 (making the Constitution, of which the Charter is a part, the supreme law of Canada), of entrenching the rights and freedoms set out in the Charter. The invocation of section 33, and especially of section 33(2), pierces the wall of constitutional entrenchment and resurrects, in particular circumstances, the sovereignty of Parliament or a legislature. Consequently, the Charter is a unique combination of rights and freedoms, some of which are fully entrenched, others of which are entrenched unless overridden by Parliament or a legislature.' The establishment of the clause has been a 'uniquely Canadian development with no equivalent in either international human rights documents or Western democratic human rights declarations'. 76   Outside the Quebec Language Law issues, the clause has been used only three times. The first such use was in Yukon's Land Planning and Development Act in 1982, the second was in Saskatchewan to protect back-to-work legislation which was declared contrary to the freedom of association in Section 2(d) of the Charter by the Court of Appeal. The third use was in Alberta where the conservative provincial government considered using the notwithstanding clause against a ruling by the Canadian Supreme Court in April 1998 in the case of Delwin Vriend v. Alberta. Vriend had been dismissed by the Christian King's College in 1991 because he was gay and this was taken to appeal by the Alberta Human Rights Commission. The Supreme Court, eventually, decided in favour of Vriend and stated that the right to equality in Alberta's Human Rights Act had to protect against discrimination on the ground of sexual orientation. Because of popular pressure, however, the Alberta government, led by Premier Ralph Klein, did not use the clause. Eventually the province of Alberta introduced a law allowing a veto in the legislative assembly against the notwithstanding clause and requiring a referendum for its use. 79

Arguments against the clause
During the history of the debates on the introduction of the clause and its application in Canada, the old and well-known arguments for and against are used by judicial supremacy absolutists and by parliamentary supremacy absolutists. 80 Critics argued that the clause is inconsistent with the entrenchment of human rights and freedoms, most simply because 'rights are rights'. (i) Rights are, indeed, subject to judicial interpretation 'but must be protected against legislative transgression'. (ii) Generally governments and legislative assemblies do not violate 'rights in defiance of public opinion; rather, it is precisely when the majority of the public is in favour of, or at least not opposed to, the limitation or elimination of the rights of a minority that constitutional constraints are needed'. (iii) 'The Charter does not create absolute rights and freedoms that must be applied literally; section 1 of the Charter provides that the rights and freedoms guaranteed are subject to "such reasonable limits prescribed by law as can be demonstrably justified in a free and democratic society." This (…) should permit the courts enough flexibility to accommodate legislative goals that infringe a guaranteed right or freedom.' (iv) The clause 'creates a hierarchy of rights' 'because the legislative override is applicable to only the fundamental freedoms and legal and equality rights'. (v) The clause raises questions about the nature of the freedom that remains because the 'rights and freedoms that can be overridden are so significant.' 81 (vi) The mere existence of the override power can entice governments to use it. 82 (vii) The clause might be used in cases where rights and freedoms are most in need of protection. 83

Arguments in favour of the clause
The most important arguments in favour of the clause have been the following: 84 Those who argue in favour of Section 33 do not see it as inconsistent with entrenched rights and freedoms and contend that (i) it provides a mechanism whereby, in exceptional circumstances, the elected legislative branch of government may make important policy decisions and isolate them from review by the unelected judicial branch of government. (ii) They argue that the threat to individual rights is not great because there 79  will become evident when we need protection most, referring to the October crisis of 1970, when the federal government set aside the Canadian Bill of Rights by enacting the Public Order (Temporary Measures) Act. Clearly, then, it gives federal and provincial legislators very wide powers to do as they see fit in limiting or denying those rights and freedoms. Perhaps none of our legislatures will use the notwithstanding clause again. But it is there. And if this dagger is flung, the courts will be as powerless to protect our rights as they were before there was a Charter of Rights.' 84 Johansen & Rosen, supra note 75, my numbering and italics.

Parliamentary Supremacy versus Judicial Supremacy
How can adversarial judicial, public, and political dialogue be institutionalised?
Utrecht Law Review | Volume 12 | Issue 1, 2016 is a five-year limit on any use of the notwithstanding power. (iii) Any such legislative override will be subject to public debate at the time of its first enactment and at the moment of any subsequent re-enactment. (iv) They also point out that only some, not all, rights are subject to a possible legislative override. (v) (They) maintain that, while it is useful and, indeed, very valuable for the courts to play a role in the elaboration of the rights and freedoms that Canadians should enjoy, it is not proper for them to act as legislators. Judges may remain in office for many years after their appointment, long after the government that appointed them has left. That they do so now is not questioned; however, if they had a greater 'political' role, their non-accountability to the electorate might well be a source of controversy. (vi) Closely linked to this is the assertion that a policy-making role would compromise the independence and impartiality of the courts and would hasten their politicization. It may thus be argued that a legislative override, by allowing final political decisions to be made by the elected representatives, mitigates the politicization of the courts. (vii) Closely linked to the submission that legislators, and not judges, should have the final word on public policy matters is the 'safety valve' or 'unintended consequences' argument. Simply put, this suggests that the notwithstanding clause is needed where a judicial decision based on Charter guarantees might result in a threat to important societal values or goals. Because the Charter rights and freedoms are generally stated and are susceptible to varying constructions and interpretation, the courts may render judgments that the drafters did not anticipate ('unintended consequences').
In short, Section 33 has been justified on the grounds that it preserves the principle of 'parliamentary sovereignty'. As well, legislators, unlike judges, are electorally accountable. Section 33 also makes it possible for Parliament or a provincial legislature to correct any unfortunate judicial interpretation of the Charter.
Even such a brief oversight of these arguments demonstrates, as is only to be expected, that we have to deal with serious issues and that it is rather difficult to find an institutional arrangement promising better or more 'reasonable balances'.

Some proposals
Some of the most serious objections against Section 33 might be addressed by the following proposals. I start with procedural proposals before dealing with the open, pluralist and changing character of LDC and some remarks on a hierarchy of rights, if any.

Procedural proposals
A serious fault in the construction of the clause is that laws are declared valid even if they violate certain basic rights. 85 This could be remedied, and the protection of basic rights could be considerably strengthened by two changes elaborated by Van Dommelen: (1) by allowing an override clause not before but only after a judgment of the Constitutional Court 86 and, (2) by explicitly stating that parliament is not allowed to suspend certain categories of basic rights (except in cases of obvious emergency) 87  In addition to Van Dommelen's changes, a third change could be that parliament should not be allowed to overrule a court decision by a simple majority but only by a qualified majority, e.g. 60%, and, indeed, for a limited timespan such as three or five years and a limited number of rounds, instead of 'try, try, again'. This would strengthen minority rights 89 against the whims of simple majority decisions.
A fourth change could be that if -after intense public debate stimulated by this procedure 90 -constitutional courts (Grand Chamber, last instance) were still to declare the respective or an appropriately revised law unconstitutional, such a judgment could only be overruled after detailed procedures (second and third readings, timespan) with even more demanding QMV comparable to the constitutional requirements for amending the constitution or, alternatively, by directly amending the constitution.
A fifth change could be that if one were to consider strengthening the democratic legitimacy of such decisions by referenda (as in Alberta) these, and the preceding public debate, could be more focused and detailed 91 compared with encompassing amendments of the constitution. Vetoes, however, would create joint-decision traps and generally work in favour of the status quo.

The open, pluralist and changing character of liberal-democratic constitutions
Critics are right to point out that rights are not absolute rights but 'subject to reasonable limits' (Section 1 of the Charter, quoted above in Section 3.1) but their claims that courts would have enough flexibility for interpreting the inherent conflicts, tensions with other rights and with common interests/goods in accordance with changing societal conditions and evaluations is clearly an overoptimistic petitio principii. The strongest arguments of defenders of the clause result from a combination of the 'unintended consequences and unforeseeable changes' arguments with the lack of democratic legitimacy and implicit politicization arguments. In their weighing and balancing of rights against each other, in their strict scrutiny tests and in their interpretations of politics/policies aimed at the 'common good' (in contextualized 'all things considered' judgments), courts depend, on the one hand, on meta-legal and meta-constitutional second order principles and, on the other hand, they depend on social and cultural values. Both are, more or less rapidly, changing. If there does not exist a vibrant tradition of dissenting opinions and, more broadly of judicial and public political debate, courts tend to be, other things being equal, isolated and rigidly conservative (in cases of lifelong appointment of judges, even more so). An appropriately fine-tuned override clause forces them to listen, to open up to dialogue and to learn. 92 It puts their implicit political biases to the test and demonstrates the impossibility of 'completely independent', 'impartial' and 'purely legal/juridical' judgments. Politicization in the service of 'relational neutrality', 'embedded impartiality' and 'difference sensitivity' 93 is required and parliaments, amongst other democratically accountable institutions, are better able to achieve this.

Hierarchy of rights?
Critics have argued either that the clause allows the override of the wrong categories of rights (Section 2 and Sections 7-15 of the Charter) 94 or more generally that there should be no hierarchy amongst basic 89  rights. From my perspective (LDC as a conflictual historical, developing compromise with two pillars), the old conflicts about primacy of 'liberal' rights (roughly speaking: rule of law, judicial rights, life and liberty) or of 'democratic rights' (equal active and passive voting rights, freedoms of political communication), or the idea of a strict, context-independent hierarchy or lexical ordering of rights, are not promising but rather misleading. As if one could achieve consensus on such an ordering, as if such an ordering would be immune from contexts and changes, as if it would be informative to decide specific cases. Instead we have to live with ongoing dissensus even on basics and we have to raise the standards and develop the arts of reasonable contestation (deliberation cum negotiation), of reasonable balancing, of proportionality and strict scrutiny of all infringements or limitations, and of context-sensitive decisions 'all things considered'. A revised override clause might promise to achieve just this. 95

Some lessons
Let me finish by outlining some preliminary general lessons. If, as already stated, 'consensus' is unachievable, if serious tensions and conflicts have to be acknowledged, if security of the law and effective rights-protection (better guaranteed by courts and constitutional review) are in tension with democratic legitimacy of making and changing the law (better guaranteed by democratically elected and accountable institutions), if the cooperation of the intertwined institutions of constitutional courts and parliaments cannot be productively regulated by simple 'supremacy', a lot depends on smart institutional design. Unfortunately, the very limited application of the clause also means that not many lessons can be learned up to now from its practice. As should be clear by now, not one institutional model fits all circumstances and there are no best or optimal practices that could be exported, but we can compare and learn. Learning can start from one end of the spectrum, as in the UK by adding weak forms of constitutional review or from the other end, as in Canada, by adding elements of 'parliamentary sovereignty' to stronger constitutional review from the American tradition. Yet also in this case, obviously, the notwithstanding clause cannot and should not be exported -not only but particularly also because of its focus on cultural minorities and language rights and weak attention for basic civil rights that has provoked the expected criticism, by defenders of strong constitutional review, as weakening Charter rights and undermining the authority of judges/courts. Still, it opens ways of informed and reasonable, adversarial dialogue through competition and cooperation between constitutional courts and parliaments, particularly if some of the suggested amendments were to be taken seriously. 96 Compared with internal mechanisms to increase substantive democratic legitimacy (such as dissenting opinions and competition/dialogue between state courts and federal courts which are present in the US and Canada) the dialogue, competition and cooperation between the Supreme Court and Parliament as one of the ways to add external democratic legitimacy is fairly new in this context. 97 In my view, this dialogue is very promising because it may also open up constitutional debates and judicial dialogues to a broader political and societal public and the respective interested stake-holders, 98 a claim that is difficult to test empirically because of the limited number and character of cases in Canada. This should, however, not be a reason to underestimate the 'mere existence' 99 of such an institutionalized dialogue because its mere existence, even if not used, makes revitalization much easier. This underestimation is motivated by two arguments: first by sweeping statements on a predominant 'contemporary political-legal culture' and, second, by sweeping statements on a 'general trend toward a judicialization of constitutional politics' and 'law-making' and, hence 'judicial supremacy' (instead of departmentalism or judicial dialogue) all over the world. 100 The competition/cooperation between state constitutional courts and federal constitutional courts in Canada and the US -as, maybe, the most effective way to increase judicial and political dialogue 101 -is however fairly weak compared with the European Union under the condition of dispersed and decentralized constitutional adjudication, which is a laboratory and hotbed of overlapping and competing jurisdictions, as I indicate in the next section.

New forms of cooperation and competition between parliaments and highest or supreme courts in Member States of the EU, the ECJ, and the ECtHR
Even in so-called 'unitary' nation states the unity, finality, comprehensiveness, coherence and consistency of law, as claimed and normatively promoted by traditional legal monism or Rechtsformalismus (legal positivism) is a more or less effective myth masking different degrees and sorts of actual legal pluralism. 102 As we all (should) know: not all law is written law (customary law), not all written law is made by (differentiated) legislative bodies (judges make and change law by applying it) and there is competition between legislative bodies, most obviously in explicitly non-unitarian, federalist states amongst first and second chambers or in presidential democracies (president versus congress, 'la cohabitation' in France). Also from a normative perspective, competition of courts and overlapping, competing and cooperating multiple jurisdictions do not necessarily spell disorder or disaster (stand-off, joint decision-traps, deadlock) but can, under certain conditions, be productive. For the new, highly pluralist polity of the EU, this age-old discussion acquires new and contemporary urgency.

Multi-level constitutionalism in EU demoi-cracy
The EU is in many regards still an enigmatic, new type of polity. It is the most complex, multi-level polity, neither an intergovernmental organization nor a new federal state. In my view, it may be best called a 'complex' or 'compound democracy' or a 'demoi-cracy' 103 characterized by multiple (local, provincial, of this legal power -however rarely it may be exercised -forces politicians to take responsibility for their decisions'. 100 Kuo's analysis of the actual practices and, particularly, his theoretical stress on the importance of cultures, virtues, habits and practices is more than welcome, but does not legitimize downplaying the importance of existing institutional structures, let alone the development of institutional alternatives. 101 However, in this regard as well it is plain that such a productive adversarial judicial dialogue is possible without an institutional import of a notwithstanding clause. Other institutional arrangements may be available. state, European), overlapping, shifting and contested (legislative, executive and judicative) powers and competences. Predominantly, the overlap of all these competences is seen as negative (endless strife, institutional and political deadlock), 104 but it can also work productively and transformatively. 105 This all is way beyond the scope of this article.
Here the focus is on 'Constitutional Pluralism' or 'Multilevel Constitutionalism' as 'an interactive process of establishing, dividing, organizing, and limiting powers, involving national constitutions and the supranational constitutional framework, considered as two interdependent components of a legal system governed by constitutional pluralism instead of hierarchies'; 106 and, more particularly, on the specific role of judicial review by constitutional courts of Member States and of European courts in relation to legislative bodies in the development of European law and politics in general, more specifically under the conditions of the recent financial and Euro-Crisis and strengthening of executive federalism. In general, we can see two opposing but interlinked developments. On the one hand the emergence of new forms of dialogue and cooperation between courts, starting from the 1970s onwards which, on the other hand, is threatened by two recent developments: by competing judicial supremacy claims in response to executive monetary and fiscal federalism, and by judicial supremacy claims of the ECJ in the debate on the accession of the EU to the ECHR.

The emergence of dialogue and cooperation
The complex relationship between courts in the EU generally and particularly regarding constitutional adjudication has been contested right from the start and the 'primacy' or 'supremacy' and 'direct effect' of European law and European courts has been understood differently. 107 Under the impact of globalization and, particularly, the EU, the Kelsenian model of constitutional adjudication by centralized and specialized constitutional courts as 'ultimate arbiters' -predominant in continental Europe -gives way to a more radically dispersed and decentralized constitutional adjudication (compared with the US model). A new dynamics of dialogue and competition developed in three ways (in reverse order) as shown below.
(1) The Simmenthal ruling (1978) 108 'revolutionized national constitutional law and in particular constitutional adjudication by taking away the monopoly of centralized constitutional courts to review the constitutionality of an Act of Parliament. In addition, non-specialized courts and bodies have also acquired new functions of constitutional review in European countries with weak or non-existent constitutional review (the UK, Finland, Denmark, Sweden, the Netherlands) resulting in competition between national constitutional courts and other courts and councils.
(2) Competition between European courts and national constitutional courts. After Simmenthal some kind of equilibrium had developed 'in which national courts allowed the ECJ space for some of its claims to priority while the ECJ allowed space for some of the most fundamental and cherished national constitutional values', yet in the Winner Wetten judgment 109 it is suggested that the BVerfG 'should and could not be trusted to protect and apply EU law'.
(3) Competition between European courts (ECJ versus ECtHR). The ECJ claims and tries to enforce new and unprecedented prerogatives at crucial moments in the negotiation, claiming 'the monopoly to interpret EU law itself by side-lining the ECtHR and imposing the obligation that national constitutional courts must refer the case to the ECJ. The ECJ fears that ECtHR will tread on issues of EU law which it feels it should have unfettered discretion and full autonomy in deciding', 'it fears the ECtHR as a competitor in an area where an overlap of jurisdiction may arise'. This competition acquired a new quality in the debate as to whether the EU should ratify the ECHR. 110 Contested supremacy claims of defenders of national or of European constitutional supremacy may, and actually do stimulate dialogue and reasonability. National constitutional courts not only should but also do talk and listen to each other and may learn from each other, 111 and European courts may do so as well.
At the same time it is important to see that not only is the same old game of competing supremacy claims being played, but also that, even under conditions of weakly formalized supremacy, new forms of dialogue and cooperation have developed between Member State courts themselves and also between the ECJ and the ECtHR. 112 These new forms of horizontal or heterarchical relations, of informal and formalized or institutionalized judicial and non-judicial dialogue among judges in 'European Judicial Networks' but also of Courts of Audit and of Ombudsmen may have two results. On the one hand, they may and do contribute to reasonability in constitutional and judicial dialogue and 'trigger a political debate about legislative activities at the European level'. 113 On the other hand, they also open it up to non-legal experts and a broader public of interested politicians and other relevant stakeholders. Hence, courts should not only listen to and talk with other courts and legal experts without claiming supremacy, but also with legislators, politicians and other experts.

Parliamentary Supremacy versus Judicial Supremacy
How can adversarial judicial, public, and political dialogue be institutionalised?
Utrecht Law Review | Volume 12 | Issue 1, 2016 In sum, multiple and overlapping jurisdictions and competition between all sorts of courts may not only be positive for 'liberal' constitutionalism or principles and practices of the rule of law 114 but also more demandingly for LDC, minimally understood. 115 The most important and until now fairly neglected issue is to specify the social, political and cultural conditions under which these negative or positive effects can be expected (see Section 5, Conclusion).

Executive federalism or judicial supremacy?
Under the conditions of the recent crisis of financialized capitalism and the Euro-Crisis the discussion on courts' supremacy acquired a new quality and urgency both within Member States and in their relation with European courts. Situations of crisis and -declared, perceived, actual -emergency always have worked and work in favour of executive powers, 116 particularly if the legislative powers are weak and if mechanisms of legislative and judicial oversight are underdeveloped or non-existent (as is still the case in the EU even after the introduction of co-decision). As is well known and well documented, the two interlinked crises have led to 'executive federalism' within less than five years, to an 'unprecedented rise and exercise of executive powers', 117 to 'audacious new modes of governance and regulatory powers' in a new regime of financial and economic governance. This 'new constitution of the EMU' 118 contains three components all involving 'novel competences for the EU institutions': (i) budgetary constraints, (ii) financial stabilization, (iii) economic adjustment.
According to Joerges, 119 this new regime is characterized by 'necessarily indeterminate general clauses', it is 'regulatory in its nature, establishing a transnational executive machinery outside the realm of democratic politics and the form of accountability which the rule of law used to guarantee': the 'rule of law and legal protection requirements are being suspended'. This is all seemingly incompatible with the EU's commitment to democracy and the rule of law. Hence it is 'brought to trial at both national and European levels' and this jurisprudence is, indeed 'an acid test of constitutional guardianship', courts trying to counter this new executive federalism in the EU as well as the non-existent or weak control by Member State parliaments and the ECJ. Yet this 'Judicial Scrutiny' 120 eventually resulted in a 'Crisis of Jurisprudence'. 121 Space prevents me from giving a summary of the detailed analysis of the divergent rulings of constitutional courts in five EU Member States (Estonia, France, Germany, Ireland, Portugal), and of the ECJ. 122 According to Fabbrini, 123 the outcome is, on the one hand, a clear 'trend of increasing judicial involvement across Europe' and, on the other hand, that courts have, with the exception of the latest decision of the Portuguese Constitutional Court, validated the legal measures under review, although with two caveats: first, they have expressed 'more discomfort towards measures of financial stabilization and economic adjustment' compared with those of tighter budgetary restraints and, second, 'over the years courts have also revealed

Veit Bader
Utrecht Law Review | Volume 12 | Issue 1, 2016 a greater unwillingness to let political branches have it their way'. The increasing judicial involvement is largely a consequence of the 'intergovernmental approach' because in most state jurisdictions, supreme or constitutional courts are empowered to review a priori international treaties (such as the Fiscal Compact and the European Stability Mechanism (ESM)), but not the legality of EU legislation; only the ECJ can declare an EU act void, the reason why such legislation -the 'Six Pack', 124 the 'Two Pack' 125 and the European Financial Stabilisation Mechanism (EFSM) -'has entirely escaped judicial review'. Hence, the shift from 'legislation to contract' paid a high price in terms of judicialization: 'the choice by the EU member states to respond to the Euro-Crisis through a strategy of intergovernmental governance, and with systematic recourse to international agreements outside the EU legal order, has resulted in increasing judicial involvement in fiscal affairs'. The paradoxical outcome is: 'One of the central tenets of intergovernmentalism in EU governance is that the executive branches (acting within the European council) will dominate decision-making to the detriment of legislatures and courts. 126 Yet the outcome of intergovernmentalism has been an increasing involvement of courts' even far greater and stronger than that which one finds 'in a country such as the United States which is generally credited as having one of the strongest systems of judicial review worldwide'. 127

Consequences of strong judicial involvement
Let me point out some more general consequences of this strong judicial involvement, approaching judicial supremacy, at the level of Member States and at the EU level.
(1) The effort by national constitutional courts, particularly the German court, to protect democracy against European executive powers and judicial supremacy claims by the ECJ, leads them to overstep their competences and tends to weaken their liberal-democratic legitimacy by countering expertocracy with national juristocracy. 128 (2) Substantively, the competing supremacy claims in matters of fiscal and economic policies by the German court and the ECJ are both dominated by a neoliberal ideology that serves as a background to fill the indeterminacy and internal inconsistency of economic and fiscal legislation -only the Portuguese Constitutional Court in its latest ruling objected explicitly. 129 This same 'Merkelantist' 130 paradigm also served as the background for the new EMU constitution: 'the new authorization of "the logic of the market" and its austerity requirements' was an open replacement of the original Maastricht philosophy. Courts, in these cases, however 'should turn to the legislature and respect its messages. In the present case, however, the political emperor seems to have absolutely no clothes. Its promise that crisis law will bring economies in difficulty "back on track" lacks plausibility. The suffering imposed on a large number of European citizens may be merely senseless. The departure from Europe's commitments to the rule of law and democracy comes at a high price.' 131

Parliamentary Supremacy versus Judicial Supremacy
How can adversarial judicial, public, and political dialogue be institutionalised?
Utrecht Law Review | Volume 12 | Issue 1, 2016 take fundamental decisions'. 132 Three main considerations should guide the allocation of competences among alternative institutions in separation-of-power systems. 133 First, with regard to expertise, political institutions (governments, parliaments and central banks) are endowed with greater expertise than courts in the fiscal domain both at Member State and at EU level. They can and do mobilize more specialized, expert knowledge, they are under more political pressure and scrutiny, and they can adopt proactive approaches. Second, with regard to voice, there is undeniably a 'democratic deficit' of intergovernmental decision-making and the democratic legitimacy of the European Parliament -compared with parliaments of Member States -is still weaker, but, compared with courts at both levels, the political branches enjoy greater voice. 134 Third, the major weaknesses of the EU political process from a democratic perspective ought to be addressed by institutional reforms of the EMU, yet it is dubious, to say the least, whether this deficit would be cured by greater oversight by institutions like courts because of weak or non-existent democratic legitimacy. 135 'An intergovernmental system of governance suffers from major legitimacy gaps': 'the political process ought to maintain the lead in the fiscal and economic field'. Hence, we should 'reform the EU political process without the courts'. 136 (4) Both the EMU regime and the extremist judicial review destabilize the still weak constitutional balance 137 in the EU as well as its liberal-democratic legitimacy. Hence, they contribute to a deep and serious crisis not only of the Eurozone but of the EU itself, as is clearly demonstrated by the never-ending Greek tragedy.

Conclusion: institutionalizing adversarial dialogue. Learning from critical pragmatism, democratic experimentalism, democratic institutional pluralism and associative democracy
The outcome of my discussion is that judicial supremacy is indefensible on many grounds and that courts/ judges cannot and should not try to speak legal truth to political powers by analogy with the increasing insight that science cannot and should not speak 'truth to power'. 138 Particularly under conditions of complex societies, it becomes plain that we have to deal with pervasive reasonable disagreement and with perceived contestedness and uncertainty of all cognitive knowledge, 139 yet the criticism of objectivist truth claims does not imply scepticism, decisionism, radical constructivism or 'it's all politics'. Instead, we have to 'democratize science' (or, more broadly, all kinds of expertise) in order to 'expertise democracy'. There are three main mechanisms for democratizing science and, by analogy, 'law and jurisprudence'. 140 (1) Democratic self-organization of science. The ethos of truth-finding needs backing by cultural, legal and institutional safeguards insuring free criticism, pluralism and lively internal debate counteracting power asymmetries. Rivalry and competition between individuals, schools, disciplines, and research institutions help considerably to counter monopolistic claims. No one model of the internal organization or governance of science fits all countries, disciplines and issues, and the main problem is reframing the legal and institutional guarantees of the 'relational autonomy' of science while also recognizing a legitimate role for outsiders (versus absolute autonomy and insulation).
(2) Inclusion of experience-based expertise, stakeholders and their organizations and unorganized public groupings and media.

Veit Bader
Utrecht Law Review | Volume 12 | Issue 1, 2016 (3) The problematic inclusion of lay public groupings that, under appropriate institutional conditions, can beneficially contribute to increase the relational objectivity of science as well as the embedded impartiality of norms and values.
In addition, there is clearly a limited but important and democratically legitimate role for external political institutions (both legislators and government/administrations) in framing the governance of science and for control by courts. Here the problem of finding the right balance of relational autonomy is even tougher.
Obviously, the analogy between science and law/jurisdiction is circumscribed or limited because science flourishes by ongoing, increasing dissensus while law/jurisdiction has to find, at least for the time being and all things considered, the 'right decision'. However, at least in the case of constitutional law, this 'closure', as we have seen, has to go hand in hand with openness and ongoing legitimate dissensus. In this regard, some of the measures for 'expertising democracy' 141 may also be relevant because they help to develop mutual relations and dialogue: attempts to enhance the scientific and judicial literacy of the public by 'scientists and judges going public'; presentation of scientific results and court decisions in generally accessible media; broad societal and political discussions on deeply controversial issues; science hearings -and judicial hearings -organized and conducted by legislative powers, expert witnesses in court cases, and the inclusion of (cognitive and normative) experts of all sorts and relevant stakeholders in public administration, in issue definition, decision-making alternatives and even in decision-making and implementation. 142 My main claim is that the principles and the institutional and policy repertoire of democratic institutional pluralism, particularly of adversarial associative democracy, provide better practical alternatives to democratize expertise as well as to expertise democracy compared with the better known competing democratic theories and practices of 'thin liberal', 'strong republican', 'deliberative' or 'participatory' democracy. 143 Complex or compound, multi-level democracy promises huge advantages compared with more unitary forms of political democracy in terms of higher degrees of information and of relevant knowledge, of reasonable contestation and deliberation, of more adequate problem definitions and solutions, of stakeholder involvement and of ongoing, reiterative learning. Associative democracy, as a specific variety of Democratic Institutional Pluralism, combines government by the people (through political participation and elections), of the people (through representation), for the people (through effective government/governance) and with the people (through interest consultation and stakeholder representation). It combines multi-level 'political' democracy with multi-level 'social' democracy. 144 Its problems -well known from the huge literature on existing governance arrangements -are along the following lines: overwhelming institutional complexity, opacity, highly selective inclusion, lack of legal and democratic accountability and control (even in 'the shadow of hierarchy') and rigidity, joint-decision traps and deadlock, very time-consuming and conflict ridden. 145 The traditional mechanisms to overcome negative competition and endless conflict are, on the one hand, a far-reaching (constitutional, legal, administrative) ex ante specification of the respective powers and competences and their inter-relations and, on the other hand, their strict hierarchical ordering or coupling (in terms of 'supremacy'). Yet, it is well known that these mechanisms have their inherent limitations, which are getting increasingly serious in our complex societies: an exploding complexity of legal and administrative rules; rigidity, unworkability; serious limitations in terms of information, knowledge and capacities to learn and adapt; and the rules intended to overcome conflict Parliamentary Supremacy versus Judicial Supremacy How can adversarial judicial, public, and political dialogue be institutionalised?
Utrecht Law Review | Volume 12 | Issue 1, 2016 themselves become seedbeds of (political, administrative, legal, and constitutional) battles. While multilevel-polities, democratic-institutional-pluralism and associative democracy should not be equated with destructive competence disputes, rigidity and inability to reform, these dangers are not easy to avoid to say the least, and advocates should be quite outspoken in recognizing them which, unfortunately, is often not the case. 146 Fairness requires explicitly stating that these problems are not easily, if at all, avoidable. However, a conclusion is not the place to discuss this. 147 Space also precludes more than a very rough indication of the conditions under which institutionalized competition among courts may have the positive results highlighted by theorists of constitutional dialogue and constitutional pluralism. 148 Cooperation in the spirit of mutual trust is obviously much easier in positive than in zero-or negative-sum games. Apart from such economic and political conditions, social and cultural conditions are crucial: the broader the recognition of reasonable disagreement even on basic constitutional issues is -among justices, among politicians and the wider interested public -the easier it becomes to tolerate and eventually also respect deeply divergent perspectives and contested decisions and to develop the respective cultural practices, virtues and habits of toleration, civility, sensibility and the arts of listening, of moderately agonistic 'dialogue cum negotiation', of searching for and finding reasonable compromises instead of repeating and strengthening monopolistic supremacy claims. 149 There is at least some hope that we might learn from new practices of developing 'Governance-Arrangements' 150 and institutions of 'social' democracy and that the literature on experimentalist governance, 151 associative democracy and realist, deliberative risk governance 152 might show ways in which existing hard trade-offs -e.g. effectiveness/ efficiency versus transparency/accountability and democratic legitimacy -can be transformed into softer ones and how, at the same time, the information (fact base) and knowledge base can be strengthened and the deliberative and inclusive quality of practices can be drastically increased. In this regard, the general core questions concern the development, orchestration or organization of inclusive and adversarial, but focused and determined public, political talk or dialogue. This is the common core of theories and traditions of critical pragmatism, realist deliberation and moderately agonic democracy. In our case, instead of the competing claims of judicial or parliamentary supremacy, the crucial questions are, as I hope to have shown, how to organize and orchestrate iterative, adversarial public, political and judicial dialogue between courts, parliaments and a broader judicial, political and societal public and the respective interested stake-holders, in order to increase the diversity of reasonable perspectives, the information base on all relevant kinds of circumstances, particularly of changing societal conditions and social, political and cultural values and practices. Who is or should be organizing these dialogues? Who is allowed or invited to talk and listen? When and with whom? And how long should such dialogue go on in specific cases and issues?