II. Rights, the trustee court and the proportionality principle

Our project was motivated largely by a straightforward question. Kant forcefully argued that all individuals and state officials were under a moral duty to work to construct a rightful – that is, a stable and morally just – condition under a constitution. Kant treated certain foundational principles as crucial, the most important being the Universal Principle of Right (UPR):

We treat the UPR as a meta-criterion of legal validity and legitimacy. After all, Kant stressed the fundamentally juridical character of the various normative commands and prohibitions derived. Yet he wrote virtually nothing of substance about the structure of such a constitution, how it would operate as a system of justice or how the UPR would be enforced. This begs the question: What types of constitutional arrangements would be ‘most likely to help a community [construct and] govern itself according to the principle of Right?’ (Stone Sweet and Palmer Reference Stone Sweet and Palmer2017: 378).

It turned out that this topic had not attracted much scholarly attention (although see Ripstein Reference Ripstein2009) beyond important explorations of Kant’s general principles of cosmopolitan constitutionalism (especially Brown Reference Brown2006, Reference Brown2009). In this symposium, Brown and Andenas, Benhabib, and Coradetti note the evident challenges our topic generates. Kant’s discussion of Public Right is abstract (at times technical, at times breezy) and often too obscure to be determinant on key issues of institutional design. Nonetheless, we sought to demonstrate the deep connections between Kantian constitutional theory and modern rights-based constitutional law and practice. We argue that Kant provides the materials for elaborating a robust, external theoretical account of ‘modern constitutional law’ (see also Kumm Reference Kumm, Dunhoff and Trachtman2009, Reference Kumm2010; Weinrib Reference Weinrib2014, Reference Weinrib2016).Footnote ³

Those who believe that the question suggested above should not be asked at all will have no use for the book (see Introduction by Brown and Andenas). Those who would reject any of the five interlocking claims elaborated in Chapter 2 (summary at CLO: 32–33) ought to declare our effort a failure. But consider here how implausible it would be to reject these claims, and the consequences of doing so.

A Kantian system of constitutional justice comprises: (1) a charter of rights; (2) judicial review of the public reasons given for restricting individual freedoms; and (3) the enforcement of the Universal Principle of Right (UPR) through the proportionality principle. Kant unambiguously associated Public Right with juridical constraints on individuals and government, leading us and many other neo-Kantians to assess his theory in light of the single most important development in domestic and international law since World War II: the broad diffusion, and gradual standardization, of charters of rights (see the Introduction to this issue by Brown and Andenas, in heuristic defence of this approach). If the UPR – which is designed to safeguard the external freedom of all within a community – applies to acts of government, then positive law must either conform to the UPR or be invalid;Footnote ⁴ modern constitutions contain a judicially enforceable charter of rights, which authorize public officials to make and enforce law, but only insofar as such acts do not violate the charter’s terms.

The two formulations that appear in the last sentence bear more than a passing resemblance. The very notion of the UPR supposes that one’s (external) freedom (as concretized in rights) is to be treated as relative, not absolute: ‘Absolute rights are those that are required to secure one’s place as a juridical person in a system of equal freedom writ large’; they are absolute in that no UPR-compliant reason for their limitation is possible (CLO: 44).Footnote ⁵ Apart from a small handful of rights (e.g. the prohibition of slavery and torture), modern charters authorize government to limit a relative right, but only for a legitimate and sufficiently important public purpose. In the context of constructing Public Right, the UPR commands government officials to elaborate and maintain the conditions under which one person’s freedom does not hinder the freedom of others. If so, Kantian theory anticipates what we now conceptualize as a clash of rights claims, and of tensions between discrete rights claims and legitimate public purposes, which are precisely the types of legal conflicts that dominate constitutional adjudication.

In what we see as the book’s most controversial claim, the UPR may be operationalized through enforcement of the proportionality principle. Proportionality analysis (PA) is tailor-made for the adjudication of such conflicts, and virtually every powerful constitutional court in the world has adopted it (Stone Sweet and Mathews Reference Stone Sweet and Mathews2019). It would seem impossible to deny these deep structural connections. In any event, to do so would consign Kant’s constitutional ideas to an irrelevant and distant past (see the defence by Brown and Andenas in this issue). We argue that Kant provides a distinctive foundation for the moral and political legitimacy of modern, rights-based constitutionalism.

Because charters of rights and the proportionality requirement establish positive criteria of legality, the question necessarily arises of how to guarantee the accountability of public officials. Systemic legitimacy critically hinges on the effectiveness of mechanisms to secure accountability. As Kumm (Reference Kumm2010), Weinberg (2014) and others (Stone Sweet and Mathews Reference Stone Sweet and Mathews2019: ch. 2) have argued, effectiveness depends heavily on (i) conferring on individuals a right to challenge, before a judge, any public act that would infringe upon their rights, (ii) empowering a constitutional court to invalidate acts found to have violated the charter, including statutes, and (iii) constitutionalizing the proportionality principle. We marshal Kantian arguments, echoing his functionalist methods, in support of both points, while highlighting the fact that the philosopher did not discuss rights adjudication in the modern sense (CLO: 33–34). Further, as an empirical matter (CLO: 69), we know that a system of rights protection is far more likely to grow in effectiveness if managed by a trustee court, whose rulings are insulated from reversal by elected politicians (e.g., the legislature), then in a system in which this type of judicial “supremacy” does not exist (e.g., under a legislative sovereignty regime).

Seyla Benhabib, whose influence on the book was seminal,Footnote ⁶ directs her critical attention to the issue of supremacy. She asserts (in this issue) that we do ‘not spell out’ why supremacy is controversial, ‘neglect the republican dimension’ of Kantian theory, ‘in which the people’s constituent power to be a lawmaker is paramount’ and ‘assume without much argument’ that the people have, in our words, ‘legislated judicial supremacy’. In short, we ‘elude crucial questions’, including ‘When and how have the people made such a delegation?’ and ‘What are the limits of such a delegation?’ In truth, Benhabib does not engage our lengthy arguments on precisely these points.

As we stress (CLO: 42–43), in his exposition of ‘the idea of the original contract’, Kant presupposed that human beings had renounced their ‘wild lawless freedom’, deploying their ‘own lawgiving will’ to create a state whose purpose was to produce ‘laws’ capable of instantiating a ‘Rightful condition’. Neo-Kantians have little need to presuppose such a constitution (CLO: 49–54) insofar as, today, the people self-legislate their own. The data we collected and report demonstrate as much: ‘of 106 constitutions written since 1985, and on which there exists reliable information, all contain a catalogue of rights. The last constitution to exclude a charter of rights was the racist 1983 South African constitution. Of these 106 constitutions, all but five established a form of constitutional judicial review to protect rights (the exceptions being authoritarian dictatorships such as North Korea and Saudi Arabia)’ (CLO: 54). The vast majority of new constitutions announce that sovereignty rests with the people; they then go on to establish a system of constitutional justice, with a charter or rights, review and judicial supremacy at its core. It would seem undeniable that the people – through negotiating (indirectly, through representatives) and ratifying (directly) new constitutions – have legislated judicial review and supremacy. If not, one might ask Benhabib, by what other means the people could legislate a new constitutional order.

Judicial supremacy, as we define that term, does not entail an absence of constraints, but only that constitutional rules make it difficult or impossible for those whom the trustee court supervises to override the court’s rulings. In our account, trusteeship is constrained by ‘robust fiduciary duties’ meant to ensure that supremacy will not lead to juristocracy (CLO: 51–52). We think that virtually no judge, on any relatively effective apex court in the world, would deny the existence and validity of these duties, even if they may neglect them. The proportionality framework, too, is constraining: properly used, it will organize dialogues with legislatures and other public officials (CLO: 69, Chs 3, 4, 5 and 6), while requiring the court to justify its decisions in a jurisprudence of public reasons (CLO: 56; Stone Sweet and Mathews Reference Stone Sweet and Mathews2019: Chs 2, 5).

Contrary to Benhabib’s assertions, we directly engage the main controversy that haunts discussions of the legitimacy of judicial review, on which she herself focuses: the ‘counter-majoritarian difficulty’ (CLO: 67–69, 171–72). She suggests that we have ‘slanted’ our exposition of Kant in the direction of ‘courts and judicial supremacy’, which understates our position on both. In fact, we argue that both are positively required. In any case, judicial review and supremacy are at the very core of modern systems of constitutional law and justice, and we believe that Kantian theory can help us to understand why this is so. Benhabib may disagree. In any case, the discussion of judicial review’s legitimacy is fundamentally altered by the massive turn to the ‘new constitutionalism’ (CLO: 67–69), which formally – as a matter of positive, constitutional law – repudiates the dogmas of legislative sovereignty. In modern constitutional law, rights trump statutes; charters of rights comprise binding constraints on the exercise of legislative authority; and constitutions enshrine judicial, not legislative, supremacy.

What we wrote responds directly to Benhabib’s major preoccupation:

Moreover, we discuss the empirical record. There exists a huge literature in law and political science demonstrating that placing legislators under the supervision of constitutional courts alters legislative discourse and enhances rights protection. As Benhabib appreciates, constitutional courts do so through provoking and managing delicate ‘constitutional dialogues’ (Stone Sweet and Mathews Reference Stone Sweet and Mathews2019: Ch 5), which are given pride of place in the book. In contrast, studies of systems in which legislators can overturn rights rulings through legislation have built effectiveness only in a piecemeal manner, if at all (CLO: 69–70).

Benhabib invokes the case against rights review and supremacy put forth by Waldron (Reference Waldron2004), at length, which we dismiss (CLO: 68, 70). Even if one were willing to give credence to Waldron’s views, they are not Kantian, and they have no place in a Kantian account of constitutional justice.Footnote ⁸ As Weinrib (Reference Weinrib2014: 184–85; Reference Weinrib2016: Ch 5) has forcefully argued, rights, review and supremacy are necessary if the right of people to ‘just governance’ and the problem of accountability are to be resolved. The ‘right of rulers to govern’ is ‘accompanied by the duty to respect and defend the rights of the ruled’ (Weinrib Reference Weinrib2014: 172). Yet, in the absence of strong judicial review, legislators are accountable ‘only to a majority or plurality of adult citizens at election time’ (Weinrib Reference Weinrib2016: 152–53). The self-determining individual is not displaced, as Benhabib and Waldron would have it, but is instead situated at the very core of systemic legitimacy. Similarly, in our account of a Kantian system of justice, assuring accountability is an imperative, and cannot be waved away by appeals to the ‘dignity’ of statute (Benhabib, this issue). Bluntly put: there is no dignity to any statute, or any other legal arrangement, that either denies individuals their full juridical status, or is unable to pass a proportionality test. One can do away with supremacy only by diminishing the potential for achieving Public Right.

III. The cosmopolitan legal order in Europe

The book begins not with a theoretical account of a Kantian-congruent system of justice, but with Kant’s essay, Toward Perpetual Peace Among States (1795). Since Doyle (Reference Doyle1986), this text has roiled international security studies, and political scientists have in fact demonstrated the essay’s power to explain the absence of war among liberal states. But Kant understood ‘toward perpetual peace’ as a far deeper process of transformation – the construction and maintenance of a Rightful condition among peoples and states – a fact that political scientists have invariably ignored. Our book gives pride of place to Kant’s priorities. The variables Kant identified (which, as updated, include the spread of liberal democracy; the growth of economic interdependence among states; the building of international organizations within zones of liberalism; and the adoption of charters of rights) and the mechanisms he elucidated explain, in a remarkably prescient and contemporary manner, what in fact has occurred in Europe since the end of World War II (CLO: Chs 1, 3). The fact that none of our commentators takes issue with these claims should not disguise how profound the transformation of European law and politics has been since the ECHR entered into force in 1953, and the Court opened for business in 1959.

What has provoked more discussion is how we conceptualize a CLO, and analyse its components and operation. A CLO is ‘a multi-level, transnational system of constitutional justice, in which (i) justiciable rights are held by individuals; (ii) all public officials bear the obligation to fulfill the fundamental rights of every person within their jurisdiction, without respect to nationality or citizenship; and (iii) both domestic and transnational judges supervise how officials do so’ (CLO: 1). These criteria are highly restrictive, precisely because a Kantian system of constitutional justice demands as much. The ECHR met these criteria only with the combined effects of Protocol no. 11 (1998) and the incorporation of the Convention into domestic law (CLO: Ch 3). From the very beginning of the new regime, the Court used its enhanced authority to bolster the constitutional character of the system and its own status as a trustee court, with the strong support of the Committee of Ministers and the Council of Europe (CLO: Ch 4). We then direct empirical attention to how this multi-level system works (CLO: Ch 5), focusing on the interactions, both cooperative and conflictual, of the European Court and national courts.

The emergence of the European CLO raises complex issues of political legitimacy, both classic and novel. Most importantly, it undermined the foundations of ‘centralized sovereignty’, including the separation of powers notions that have long underwritten the (quasi-official) discourses on state legitimacy (CLO: 82–85). The new system obliterated the traditional dogmas of legislative sovereignty, including the prohibition of judicial review of statute; and it fatally weakened the primacy of constitutional courts in those states that had established ‘centralized’ rights review. In consequence, constitutional pluralism rapidly developed within national legal orders (CLO: 86–103), paving the way for the European Court to develop intensive dialogues on rights protection with domestic judges. All apex courts are now constitutional courts in Europe when it comes to rights protection (not simply those formally designated as such). In practice, the CLO is now governed by a ‘decentralized sovereign’ (CLO: 85), a ‘polyarchy of courts’ (CLO: 234; see Sabel and Gerstenberg Reference Sabel and Gerstenberg2010) that is networked through incorporation, the European Court’s case law and the multi-dimensional ‘constitutional dialogues’ that have evolved (CLO: Ch 5).

Benhabib rightly suggests that a key question for neo-Kantian cosmopolitans is the following: ‘what is the democratic legitimacy of international human courts based upon … and can [such courts] be justified democratically?’ In the case of the European Court, these are not difficult questions to answer. The state members of the Council of Europe freely chose to place their own legal systems under the direct supervision of the Court by giving individual claimants the right to turn to the Court once national remedies have been exhausted. In the domestic realm, officials of every state have incorporated the Convention as justiciable national law, through constitutional amendment, legislation and/or judicial decision (Stone Sweet Reference Stone Sweet, Rosenfeld and Sajó2012: App 1). We can imagine no stronger means of formally legitimizing the Court than through Protocol no 11, and legal acts of domestification. From the standpoint of Kantian constitutional theory, the very definition of a CLO contains the basic criteria that any system of constitutional justice must meet to be considered legitimate in the first place (CLO: Chs 2, 3). The CLO emerged out of decisions taken as if state officials recognized the Kantian obligation to pursue Cosmopolitan Right.

The longest sections of the book (CLO: Chs 3, 5, 6) chart the development of the structural properties of constitutional pluralism, which are a necessary condition for the multi-level ‘constitutional dialogues’ that drive a CLO’s evolution. Constitutional dialogue is constitutional pluralism in action. Benhabib suggests that, under the conditions we have identified, such dialogues can offer their own process-based logics of legitimation, and we agree. As we referenced, and as others have demonstrated in greater detail, incorporation and dialogue have profoundly altered how national systems operate. It is indisputable (Bjorge Reference Bjorge2015; Stone Sweet and Keller Reference Stone Sweet, Keller, Keller and Sweet2008) that the CLO has ratcheted up the effectiveness of relatively high-standard systems (e.g. Germany and Spain), as well as thoroughly dysfunctional ones (e.g. Russia and Ukraine).

PA is the discursive operating system of constitutional dialogues in the world today (Stone Sweet and Mathews Reference Stone Sweet and Mathews2019). In his contribution, Wojciech Sadurski launches a barbed attack on PA and the way it is deployed by the European Court. He finds that there is ‘precious little’ in the book – or in the Court’s approach to proportionality – that concerns the judicial scrutiny of reasons given by officials for the infringement of qualified rights. He argues that the Court’s ‘perfunctory treatment of state motives and purposes’ weakens the claim that it can properly serve a Kantian notion of justice.

We agree with Sadurski that the robust review of the ‘purposes’ of state officials, whenever they act to limit the scope of qualified rights, is a ‘crucial’ component of the CLO. Indeed, this claim infuses the entire book. But, as Von Bogdandy and Venzke (Reference Von Bogdandy and Venzke2012) remind us, Kant rejected the view that correct ‘decisions in concrete situations’ could ‘be deduced from abstract concepts’, and he declined to engage in casuistry in the Doctrine of Right (CLO: 6).Footnote ⁹ Sadurski implies that it is through ‘proper purpose’ inquiry (the first stage of the most developed form of PA) that a court can or should assess ‘legitimate reasons’ for legislative action. We think he is mistaken.

PA obliges judges to consider the reasons given by states at each phase of PA, in light of the concrete case before them. In an opening stage, judges consider whether the government is empowered (by the constitution or treaty) to legislate in the name of a particular public interest. Courts typically treat scrutiny of the ‘legitimacy’ – or ‘proper purpose’ – of the measure under review ‘in the style of a threshold inquiry: if the constitution has not authorized the state to pursue such a purpose, then the rights claimant must prevail’ (CLO: 63–64). A court that finds a proper purpose will then proceed to subsequent stages of PA, which direct judges to examine the harms that have allegedly issued from the measures under review.

Proper purpose is just the first step in scrutinizing state reasons for restricting a qualified right. As the most authoritative and influential treatise on proportionality, Barak (Reference Barak2012: 246–47) explains:

Like us, Barak (Reference Barak2012: 247–49) rejects the view that only one stage – proper purpose inquiry – would suffice. Its aim is to answer the question of which constitutionally recognized public interest ‘considerations’ may justify ‘limitations on constitutional rights’ and which may not. A court only moves on if the ‘nature’ of the government’s purpose has been determined to be proper – that is, once the purpose has been shown to be covered by a public policy heading found in the limitation clause of a rights provision (e.g. public health, security, the rights of others). The sequence enhances transparency to the extent that it tells us precisely why the law fails, and what public reasons are to be excluded, rather than aggregating diverse considerations into an overall assessment of ‘proper purpose’.

Sadurski’s critique raises a host of primordial questions that he makes no attempt to answer. Could a judge adjudicate qualified rights solely through proper purpose inquiry? (We think not.) If yes, could a judge do so without either determining the scope of the right being pleaded, or assessing the harm visited upon rights-holders by the law under review? (We think not.) If the answer is no, then the judge would simply be reproducing PA, within one prong, at the cost of transparency. Sadurski’s worries seem more formal and aesthetic than substantive. It is simply false to suggest that, by treating the proper purpose prong as a threshold requirement, judges forego the possibility of robust scrutiny of public reasons for infringing upon a right. Judges of proportionality simply do the heavy work of such scrutiny in subsequent stages, in the context of the concrete case before them. What has the government actually done, and for what reasons? To what extent are claimants harmed? On what basis do officials justify these harms? In the PA world, proper purpose inquiry is far more abstract: can the right be limited for such a purpose, or not? As Sadurski well knows, judges routinely use least-restrictive means testing (part of necessity analysis) to ‘smoke out’ improper purposes. Would Sadurski banish such testing altogether, or would it migrate into the proper purpose inquiry?

Through subsequent stages of PA, powerful rights-protecting courts routinely produce a list of excluded reasons for restricting rights. The European Court also produces categorical prohibitions (what states cannot do) and commands (what they must do) through conducting PA beyond proper purpose scrutiny. Its case law contains hundreds of important examples, some of which the book documents (CLO: Ch 5). This record is starkly incompatible with Sadurski’s assertions, not least since the Court raises standards of rights protection only by excluding reasons for limiting rights – that is, by pronouncing a state’s justification for a restriction on rights to be insufficient, or in itself revealing of an improper purpose (CLO: 174–84). The list of state measures that can never be justified has steadily expanded, which renders the purposes underlying these measures improper. The progressive development of LGBT rights clearly illustrates how this process works (CLO: 175–80).

The scrutiny of proper purpose by the Court remains important in the overall process of reviewing reasons for limiting qualified rights. When a state seeks to rely on a novel public purpose, or claims an extension of an existing heading, for example, the Court may drop its more ‘succinct’ approach to engage in a more ‘in-depth examination’ (SAS v France [2014: paras. 115–22], discussed at CLO: 189–90). But given the broad and abstract designation of legitimate purposes in the limitations clauses of the Convention, novel public purposes are rare. Rights claimants do sometimes win at the legitimacy stage, most often when the Court finds that the very purpose of the statute is to deny people their rights, but always against the backdrop of how the Court has interpreted these rights, and scrutinized state justifications, in prior cases. The latter stages of PA create the materials that gives more specific content to proper purpose inquiry. Were the Court to limit itself to the legitimate purpose prong, it would produce confused and inscrutable judgments, undermining dialogue between the European Court and its domestic interlocutors.

Consider the case of Open Door and Dublin Well Woman v Ireland [1992], wherein the applicant challenged a ban on the distribution of information concerning the availability of legal abortion in the United Kingdom. Sadurski invokes this case as an example of the Court engaging (at least in part) in the sort of purpose scrutiny he espouses. The sum total of the Plenary Court’s analysis rejecting the asserted purpose is as follows: ‘The Court cannot accept that the restrictions at issue pursued the aim of the prevention of crime since … neither the provision of the information in question nor the obtaining of an abortion [in the UK] involved any criminal offence’ (Open Door, para. 63). This single sentence communicates precious little to the Irish government or its courts about why the state has failed the proper purpose prong, or how it might remedy the violation. Yet even this terse statement smuggles subsequent stages of PA into proper propose inquiry. The prevention of crime – in this case, the crime of abortion in Ireland – is, on its face, a legitimate purpose. The real problem is that state’s means for preventing crime – prohibiting information on abortion services beyond Ireland – is neither suitable nor does it fulfil a pressing social need. Either the Court stops at the proper purpose prong, and the regime would lose this type of analysis, or the public purpose prong would simply become the placeholder for the other prongs of PA. Fortunately for the CLO, the Court held that Open Door (paras. 67–80) presented other legitimate purposes, resulting in a full-throated – and far more transparent – rebuke of the state’s efforts to restrict women’s access to information.

Sadurski’s critique gives the Court no guidance as to how it ought to conduct ‘legislative goal scrutiny’ under analysis limited to proper purpose. This yawning void enables him to sidestep two issues that are crucial in a Kantian system of justice. The first is the accountability question: how would the Court know if the state had acted in good faith? The second is the theoretical/doctrinal question: to what extent can questioning legislative motives, but not concrete means and harms, actually enhance rights protection?

The Court has developed a number of important tools for identifying instances in which state officials act in bad faith, with disregard for rights, including the deployment of least restrictive means testing and comparative (consensus) analysis to ‘smoke out’ bad motives. Further, through ‘proceduralization’, the Court demands that states alter their domestic procedures (legislative, administrative, judicial) in order to ensure that, in future lawmaking and application, officials actually do take human rights into account. These orders require domestic incorporation. This technique, which (spectacularly) has also accompanied findings of non-violation (CLO: 194–96), is designed to ensure that there will be a relevant record to review when a new case arrives. It is hard to see how this nuanced approach – which is embedded in necessity and balancing – would operate under an abstract proper purpose inquiry. Moreover, even if the Court was somehow equipped to fully examine the state’s ‘true’ motives, there is no reason to imagine that outcomes would be any more satisfying to neo-Kantians than those arrived at under current methods.

Finally, Sadurski’s suggestion that our presentation of these issues ‘seems to boil down to judicial deference to the legislature’ seriously misstates our position. We bluntly declared any de jure deference posture to be strictly ‘anathema’ to a Kantian system (CLO: 82), and to any system of justice that enforces the proportionality principle in good faith. In the book, we dwell at length on the gradual destruction of Wednesbury-style deference postures by the European Court’s adoption of PA (CLO: 103–08) and we strongly criticize the Court for abdicating its duties in instances in which it deploys a deferential version of the margin of appreciation doctrine to avoid deciding controversial issues in hard cases (CLO: 167–68, 186–96). We do recognize that judges may build de facto deference to law-makers in later stages of PA, in particular in situations in which there is high epistemic uncertainty. A declaration to the effect that a state has pursued illegitimate, improper purposes constitutes a harsh and stigmatizing type of decision that many judges may seek to avoid. But we neither treat this disposition as Kantian congruent, nor celebrate it.

IV. Towards the constitutionalization of the international legal order

The book does not adequately deal with an issue that bedevilled Kant, a question that has been debated fiercely by cosmopolitans ever since: (1) Is a global state necessary for an international community to achieve Cosmopolitan Right; or (2) can the latter be realized through the construction of a federation of republican states? In Toward Perpetual Peace, Kant made it clear that he had changed his mind, from (1) to (2), while stipulating that the league ‘should not exercise coercive powers within any national order’, and state members should ‘be free to leave it’ (CLO: 17). While acknowledging Kant’s turnabout (CLO: 73, 250; Brown Reference Brown2009; Corradetti Reference Corradetti2016, Reference Corradetti2017a), we do not address it in any serious manner. Instead, we chart the various ways in which the ECHR has meaningfully been constitutionalized, despite the fact that its Court does not possess the power to annul national acts found to have violated Convention rights (CLO: Chs 3, 4). Today, the ECHR ‘appears as a kind of midway point between (i) the type of federation Kant considers in Perpetual Peace, and (ii) a more hierarchically constituted system of rights protection’ (CLO: 250).

Two contributors to this symposium, Garrett Wallace Brown (Reference Brown2006, Reference Brown2009) and Claudio Corradetti (Reference Corradetti2016, Reference Corradetti2017a), have been at the cutting edge of neo-Kantian thinking about the potentials of different forms of cosmopolitan organization. In this forum, Corradetti argues that the ECHR comprises a CLO to the extent that it operates as if ‘the principles of cosmopolitan Right’ grounded the political legitimacy of the Court. One express purpose of the book is to give empirical content to that claim. The Court has laid down guiding principles for a jurisprudence that is, however imperfectly articulated and implemented, strikingly Kantian.Footnote ¹⁰ However, as Corradetti insists, a CLO is not just a set of courts networked through pluralistic arrangements, but rather a legal system in which ‘all public officials [not only judges] bear the obligation to fulfill the fundamental rights of every person within their jurisdiction’. It is this general obligation that makes the judicial supervisory function so important in the first place. There is a great deal of research to be done on the extent to which the incorporation of the Convention and the Court’s case law has impacted the decision-making of non-judicial officials at the domestic level. This variable – the willingness and capacity of domestic officials to implement the Court’s important rulings – is the crucial measure of the relative success (or failure) of the European CLO over time (CLO: 149–53).

We now turn to the multi-level global constitution that we argue (CLO: 82–85, 102–03, 234, 249–50, 256–58) comprises an overarching code of rights that is enforced by a decentralized sovereign. Stone Sweet and Mathews (Reference Stone Sweet and Mathews2019) sum up the view in this way:

How plausible are these claims? We are hardly the only ones to have made them, with Mattias Kumm (Reference Kumm, Dunhoff and Trachtman2009) and Stephen Gardbaum (Reference Gardbaum2008, Reference Gardbaum, Huscroft, Miller and Webber2014) being notable examples. As Gardbaum (Reference Gardbaum2008: 752) puts it, the ‘international human rights system has become one of constitutional law in its own right’, not least, in that ‘the legal status of the protected rights has become similar within each system [domestic and international]’:

Kumm (Reference Kumm2010) and Gardbaum (Reference Gardbaum, Huscroft, Miller and Webber2014) also stress the importance of the proportionality test to the building of effectiveness and dialogic governance.

In his contribution, Po Jen Yap explores the importance of Kant’s first definitive article, focusing on democratization in the context of global, rights-based constitutionalism. Yap demonstrates that proportionality-based constitutional governance is not limited to ‘western states’ but has also taken hold in Asia. In South Korea and Taiwan, the rapid development of competitive party systems in the late 1980s and early 1990s led to the rise of trustee courts, which quickly revealed their potency, both to defend democracy and to organize policy dialogues with the other branches of government (see also Lin Reference Lin2018). By contrast, in Singapore – nominally a democracy that has been dominated by the same political party since its founding in 1965 – PA ‘is rejected in all forms’ and the apex court ‘has abstained from invalidating any legislation [or any administrative act] that comes its way’. As Yap notes, South Korean and Taiwanese constitutional judges (like their peers in Colombia, South Africa, Spain and the states of Central and Eastern Europe) have been heavily influenced by German law, and they regularly consult foreign sources when deliberating on important constitutional cases. More generally, rights and review have been crucial to virtually every successful transition from an authoritarian regime to constitutional democracy since 1950, Japan being an important exception. Indeed, the more successful any transition has been, the more likely one is to find a trustee court working to build systemic effectiveness (Issacharoff Reference Issacharoff2015; Stone Sweet and Mathews Reference Stone Sweet and Mathews2019: 16–17, 23–24).

Wayne Sandholtz directs empirical attention to the constitutional dialogues that take place among trustee courts, within and between regional, treaty-based systems of justice. In a world without a centralized sovereign, inter-judicial dialogues have emerged as a primary mechanism for enforcing the global constitution, and for building the ‘cosmopolitan commons’ (CLO: 249, 258). The fact that the ECHR is the only CLO currently in existence does not mean we are unable to analyse how ‘dialogic cosmopolitanism’ (Benhabib) operates beyond Europe. Dialogue between courts, Sandholtz shows, has not only exploded; it has grounded the most important efforts of the Inter-American and African Courts to build the effectiveness of the regimes they manage. The European Court is the focal point of inter- judicial dialogues on rights. The Inter-American Court systematically surveys relevant ECHR jurisprudence for every important move it makes. That Court adopted PA, declared the Convention to be a ‘constitution’ and a ‘living’ instrument, insisted on the erga omnes effects of its case law and ordered states to incorporate the Inter-American Convention as enforceable domestic law, while discussing and relying upon the European Court’s positions (Stone Sweet and Mathews Reference Stone Sweet and Mathews2019: 178–85). The Court of the African Union, which enforces the African Charter on Human and People’s Rights has followed, adopting PA and declared the erga omnes authority of its interpretation of that Charter, for example, while citing to the European and Inter-American Courts’ jurisprudence (Stone Sweet and Mathews Reference Stone Sweet and Mathews2019: 185–91).Footnote ¹¹ Today, the European Court’s most important rulings are treated as if they were an authoritative source of human rights law.

Eirik Bjorge takes an even broader view of the normative landscape, helping us to see cosmopolitan elements in aspects of global law and institutions that do not meet the criteria of a CLO. He makes a compelling argument that, in certain situations, ‘a non-national, or stranger in the Kantian idiom’ should be treated ‘better than the national’ by the foreign state, precisely because the stranger is relatively ‘more vulnerable’ to the foreign state’s law than the national. As Bjorge details, the European Court began to develop this position in the mid-1980s, joined later by other international judges. This jurisprudence places individuals, not states, at the centre of legal and moral preoccupations – which is not an obvious posture for judges to adopt. Under traditional international law, a state’s treatment of aliens was regulated by state-to-state norms: if state A had unlawfully mistreated a national of state B, then state A could demand that state B cease the mistreatment (if it was ongoing) and pay reparations to state A. In Bjorge’s neo-Kantian construction, states owe duties to individuals as juridical persons in their own right. This development, which has great transformative potential, has taken place despite the fact that extant international law on state responsibility for harms to non-nationals lags far behind.Footnote ¹²

Bjorge concludes by referring to Kant’s views on ‘progress’ and notes that the emergence of the CLO ‘prove[s] that real progress is possible’. Most (perhaps all) of the contributors to this symposium agree on this point. Kant saw progress toward achieving a Rightful condition as one of continuous struggle, given the ‘natural’ and ‘continual antagonism’ among individuals and states (CLO: 27–28).Footnote ¹³ As Williams (Reference Williams, Føllesdal and Maliks2015) reminds us, Kant’s view was that ‘the human race does not progress the easy way. It only adopts rational principles for governing its social relations after a hard and frequently violent struggle with itself and its worst characteristics’ (CLO: 75–76). Indeed, scholars and organizations (Ginsburg and Huq Reference Ginsburg and Huq2019; IDEA 2017; Sadurski Reference Sadurski2018) have begun to document and assess a ‘backlash’ against rights and trustee courts, as well as democratic ‘backsliding’ more generally.Footnote ¹⁴ In Perpetual Peace, Kant (CLO: 80) suggests that:

But Kant took a long view. He believed that progress toward Right would be discernible but non-linear, with deep contestation and political setbacks to be expected along the way.

The European CLO has produced thousands of examples of significant progress, by which we mean enhancement of the effectiveness of national systems of justice. The book (CLO: 152–53) asks readers to consider a counterfactual:

We too notice the storms gathering on the horizons. But, like Kant, we take a long view. However unfashionable, we remain optimistic, because it is Right.Footnote ¹⁶