• 1 Introduction

  The present paper contends that the relationship between democracy and modern constitutionalism is possessed of a degree of complexity that is often underappreciated, and that, unless we address that complexity, we will be unable to come to terms with the nature of the challenge posed to constitutional thinking or the possibilities open to it in today’s globalizing world. The complexity of the relationship between democracy and constitutionalism, I begin by suggesting, is perhaps best approached and appreciated through conceiving of democracy – understood in its most general, classical sense¹xIts classical origins are of course Greek. Democracy is a compound of demos (people) and kratos (power). as ‘the ideal of government by act of the people’²xF.W. Michelman, Brennan and Democracy, Cambridge Mass.: Harvard University Press 1998, p. 4. – as an incomplete ideal. That incompleteness is double-edged. It refers both to the empirical incompleteness of democracy as a notion unable to supply its own terms and conditions of application – call this the internal dimension of incompleteness – as well as to the moral or normative incompleteness of democracy as a guide to good government – call this the external dimension of incompleteness. In turn, this double-edged incompleteness accounts for the contingent necessity of modern constitutionalism. On the one hand, constitutionalism justifies its place as a necessary feature of the modern political configuration – a configuration in which, as we shall see, the values expressed by and through democracy are undoubtedly of central importance – by answering each form of democracy’s incompleteness. It does so both by helping to realize democracy (the internal dimension) and by seeking to supplement and perhaps qualify democracy (the external dimension). On the other hand, if democracy in its double incompleteness requires the accompaniment of constitutionalism, those very same features of incompleteness means that democratic considerations are insufficient to specify in any definitive fashion the content of constitutionalism. Just as democracy cannot ‘complete itself’, so to speak, so too the content of constitutionalism as a vital means to the completion of democracy cannot be supplied solely through the resources of democracy; that content, therefore, remains contingent upon other normative and practical considerations. Democratic incompleteness, in other words, remains both the main justificatory foundation for contemporary constitutionalism and the main reason why that justificatory foundation remains inherently fragile.
  If we turn to examine the practical unfolding of the relationship between democracy and constitutionalism, we can see that some of the ways in which modern constitutionalism treats democracy along the internal and external dimensions of incompleteness are constant or recurrent over time, just as are the types of answers constitutionalism supplies. However, the ways in which democratic incompleteness manifests itself, and in which constitutionalism responds to that incompleteness, also evolve and alter over time. The relationship between constitutionalism and democratic incompleteness is, in short, an iterative one.³xOn the iterative quality of democracy, see S. Benhabib, Another Cosmopolitanism, New York: Oxford University Press 2006, chapter 2. And the iteration with which we are most closely concerned in the present discussion is that which has emerged out of the current wave of globalization. The development of a situation in which states are no longer either the exclusive sites of democratic authority or the only constitutional entities and sources has further compounded and complicated the ways in which democracy is rendered incomplete and how constitutionalism responds to that incompleteness. Nevertheless, I conclude, the historical role of constitutionalism in political modernity as a key means of addressing the double incompleteness of democracy remains intact under conditions of contemporary globalization, just as does the inability of democracy to supply some of the vital terms of constitutionalism. This continuity, crucially, has to do with the way in which the deep moral order of political modernity remains constant even as its institutional architecture evolves. Constitutionalism as a basic orientation and mobile set of techniques remains a necessary support for and supplement to democracy in the global age – and this supportive connection to democracy provides constitutionalism’s abiding justification. Yet the emerging postnational constitutionalism, like state-centred constitutionalism before it, remains contingent upon non-democratic considerations – and indeed does so in a more emphatic fashion than before – so reinforcing constitutionalism’s abiding normative and sociological vulnerability. This conclusion stands as a challenge to the two dominant but opposing understandings of the new forms of postnational constitutionalism of the global age. It sets itself apart from these views that, on the one hand, offer a conclusive indictment of global constitutionalism because of its weakened democratic credentials. It also sets itself apart from those views that, on the other hand, assume that these much weakened democratic credentials pose no problem for postnational constitutionalism, which may thrive through its emphasis on other, non-democratic values.

• 2 Constitutionalism and Democracy: A Contested Relationship

  The idea that I will defend and develop below, that modern constitutionalism stands in a double relationship to democracy, with the former both realizing and qualifying the latter, is not one that has achieved ready acceptance. In the present Section I will try to explain why this is the case, and with what distorting consequences for much of contemporary constitutional theory, including those aspects of constitutional theory that have emerged in the global age. The critique of contemporary constitutional theory that I seek to develop provides a point of comparison and instruction for my own subsequent theory-building.
  Rather than as the kind of double or multi-level relationship that I will propose, the relationship between modern constitutionalism and democracy demonstrates a long historical tendency to be viewed in singular terms, as if it were capable of being captured in just one general relational proposition. Yet there is no single such singular sense, so to speak. Rather, the actual terms of the singular relationship between constitutionalism and democracy are understood quite differently and quite inconsistently between different writers and schools of thought. A broad but significant historical trend is evident here. Crudely, what we have witnessed is a shift from a longstanding tendency for democracy and modern constitutionalism to be viewed in a relationship of mutual opposition, or at least pronounced mutual tension, to one in which, with the exponential spread of a democratically centred political morality over the second half of the twentieth century, they have come to be viewed as standing in an entirely or predominantly mutually supportive relationship – albeit this supportive relationship is itself accounted for on quite different grounds. What is more, with the development of postnational ways of thinking about constitutionalism in the global age, we are entering a new phase – one that recalls but moves beyond the dominant themes of both predecessor phases. In response to a new and starker tension between democracy and constitutionalism there is emerging a sharp division between two opposing singular conceptions – between those who adhere to democracy’s centrality to constitutionalism but doubt its viability in the postnational domain and those who would make a virtue out of constitutionalism’s independence from democracy.
  But why and how has modern constitutional thinking developed along these lines? It is important to pay attention to the adjective ‘modern’ in beginning our account of this series of changes. The idea of a constitution, already extended beyond its original reference to physical organisms, undoubtedly possessed a robust pre-modern tradition. Yet, prior to the age of the modern state, the idea of the constitution tended to be applied in political discourse in a quite different way from its subsequent usage. It was utilized in a historical-descriptive manner, referring – in accordance with its underlying biological metaphor – to the political way of life of a community in quasi-organic terms. This quasi-organic structure was traditionally imagined and portrayed in a highly concrete fashion, as the developed ‘body politic’ – the entirety of ways and means and customs of political life. In late mediaeval times it gradually came to refer, more abstractly, to the institutional form and complex of the political settlement. However, it was not until the advent of the modern state, and in particular the early peak of documentary constitutionalism in the United States and France at the end of the 18th century, that this process of abstraction was completed. The constitution now assumed a doubly normative character. Not only had it begun to refer to the specifically legal mode of articulation and regulation of the body politic (as opposed to the institutional consequences of that articulation), but also, in a more transformative stage of juridification, through the medium of the early written constitutions that legal modality now came to be seen as constitutive or generative of that body politic.⁴xOn these historical trends, see e.g., D. Grimm, The Constitution in the Process of Denationalization, Constellations (2005) 12, 447; M. Loughlin, The Idea of Public Law (Oxford: Oxford University Press 2003, esp. chapters 3-6); C. Thornhill, Towards a Historical Sociology of Constitutional Legitimacy, Theory and Society, 37 (2008), 161-197. The documentary constitution became, in Tom Paine’s words, ‘a thing antecedent to a government’,⁵xT. Paine, The Rights of Man, New York: St. Martin’s 1988, p. 29. a blueprint of how the body politic should be organized rather than a simple reflection of how it was organized.
  It is this sense of law, and in particular constitutional law, as somehow being ‘in charge’⁶xJ. Waldron, Is the Rule of Law an Essentially Contested Concept (in Florida)?, Law and Philosophy 21 (2002) 17-64, 157. – as providing the normative lodestar for the body politic, that provides the key to understanding how modern constitutionalism in its original conception came to be viewed as standing in a relationship of tension with democracy. Crucially, the initial distinction – and the initial distinctiveness – of the modern constitutional state lay in its contrast with its absolutist predecessor of the 16th and 17th centuries. In the early absolutist form of the state there had emerged for the first time, in reaction to the partial and fragmented authority structures of the feudal age, a monopolization and centralization of political power as an indivisible public domain. But, in its incipient form, that indivisible public domain was one in which positive law was the instrument rather than the source of sovereign power. Subsequently, in its new ‘constitutive’ variant, constitutional law claimed to reverse this relationship. Modern constitutionalism, therefore, came first and foremost to be defined in functional opposition to absolutism, as a guarantee of limited (by law) government as opposed to unlimited government. In turn, as this notion of constitutionalism as an ideal of limited government took hold, such a conception did not fundamentally discriminate between the different expressions of ‘sovereign’ authority which should be the due subject of constitutional jurisdictio.⁷xSee G. Sartori, Constitutionalism: A Preliminary Discussion, American Political Science Review 56 (1962) 853-64. The democratic or popular form of sovereignty was as much a form of potentially overweening and abusive authority, and so as much in need of checking, as were the monarchical or aristocratic forms of sovereignty.
  What is more, democracy remained a tenuous and unsettled as well as a suspicious object of constitutional thought, since – as we shall see – the extent to which democratic forms of government actually succeeded in taking root under the early settlements of modern constitutionalism itself remained highly uneven, partial and subject to frequent reversal until the mid 20th century and beyond. Accordingly, for philosophical and empirical reasons – both because democracy was understood to harbour its own threats to the idea of limited government and because democratic institutions were in any case a slow and faltering development of the modern state⁸xOf course, there is also a causal relationship between these two trends. One of the reasons why democracy was such a slow developer was precisely because the constitutionalist argument for ‘limited government’ could be so effectively deployed against it, and vice versa. – constitutionalism in its early modern form tended not to place democracy at its normative centre but rather as something lurking at the margins to be tamed and constrained.⁹xThis is not the place to document that claim in full. However, if we take only the American debate, few would doubt Dworkin’s claim that it has until recently been ‘the near unanimous view’ of American constitutional scholars and lawyers that a strong conception of constitutionalism, especially as articulated through a strong version of judicial review, is one that compromises democracy. See Dworkin, Freedom’s Law: The Moral Reading of the American Constitution, Cambridge Mass.: Harvard University Press 1996, p. 18.
  The semiotics of recent state-centred constitutional theory increasingly suggest an inversion of this historical understanding. Constitutional theorists of the late twentieth century have typically been at pains to reconcile constitutionalism and democracy. This has much to do with the unprecedented prominence of the democratic idea today. The 20th century has been labelled by some as the Democratic Century,¹⁰xFreedom House Report Democracy’s Century: A Survey of Global Political Change in the 20th Century, New York: Freedom House 1999. the high point in the pursuit of ‘the ideal of government by act of the people’.¹¹xMichelman, n. 2 above. It is, according to Amartya Sen, the age in which democracy has ceased to be understood in its various localities as a purely local need and has come instead to be endorsed as a ‘universal commitment’ and so as a ‘normal’ template of government.¹²xA. Sen, Democracy as a Universal Value, Journal of Democracy 10 (1999), 3 -17. Crudely, there are positive and negative reasons for this. Positively, the 20th century, and more precisely the second half of the 20th century, saw unprecedented success for the democratic project. By 1941, in the depths of the Second World War, there were only thirteen countries in the world who still could meet the most basic criteria of democratic self-government. Yet by the end of the century, as many as 119 out of 192 states could be described as electoral democracies.¹³xSee Freedom House Report, n. 10 above; see more generally, J. Keane, The Life and Death of Democracy, London: Simon and Schuster 2009. In numerous cases democracy remained, and remains today, an extremely fragile flower, poorly embedded within the local political culture, threatened and compromised by myriad external pressures in a globalized world. But even if practice often remains deficient, there is no doubting the relentless spread of democracy’s rhetorical endorsement.
  Negatively, moreover, the ideal of democratic self-government has gained traction because of the absence of credible alternative universal ideals of political organization. For the 20th century was also the century of disillusionment, its various and contending totalizing ideologies progressively discredited as culpable agents in an unprecedented ‘age of extremes’.¹⁴xE. Hobsbawm, The Age of Extremes: The Short 20th Century, London: Michael Joseph 1994. Fundamental belief systems, ‘thick’ and comprehensive doctrines of the good life, of course, retain a strong sociological presence today, indeed perhaps more so now than at any time since the mid 20th century.¹⁵xSee e.g., M. Ruthven, Fundamentalism: The Search for Meaning, Oxford: OUP 2005. However, the idea that such belief systems should prevail as global ideals founders both on the equally implacable conviction of opposite fundamentalisms and on the scepticism of countless others. From this perspective, democracy’s attraction is as the lesser evil; its distinction remains, according to Winston Churchill’s famous aphorism, ‘as the worst form of government except all those other forms that have been tried from time to time’.¹⁶xQuoted in J. Keane, n. 13 above, 581.
  But if the increasing centrality of democracy to political discourse and practice produced significant efforts towards the reconciliation of democracy and constitutionalism, the variation we find in the terms of that reconciliation betrays certain older tensions. At one end of the spectrum, there are those who tend to ‘define up’ democracy to meet a thicker sense of constitutional self-government. At the other end of the spectrum there are those for whom constitutionalism is ‘defined down’ to meet a thinly proceduralist conception of democracy. And in between these positions we find a wide range of options where democracy and constitutionalism meet each other ‘half-way’, so to speak. ¹⁷xFor an excellent overview of contemporary constitutional theory’s treatment of democracy, and, in particular, its tendency to view ‘constitutional democracy [as] a tautology rather than an oxymoron’ (90) see R. Bellamy, Political Constitutionalism: A Republican Defence of the Constitutionality of Democracy, Cambridge: Cambridge University Press 2007 esp. chapter 3. Bellamy’s classification of the ways in which constitutionalism is reconciled with democracy is more detailed than the one I deploy in the present article, but the basic distinction between those forms of reconciliation based on a ‘thickened-up’ definition of democracy and those based on a ‘thinned-down’ version of constitutionalism also informs his approach.
  Ronald Dworkin provides a prominent example of those who would ‘define up’ democracy in ambitious terms to satisfy a thicker sense of constitutional self-government. For him, democracy requires ‘that collective decisions be made by political institutions whose structure, composition and practice treat all members of the community as individuals, with equal concern and respect’.¹⁸xDworkin, n. 9 above, 17. This demands, therefore, not just the most basic constitutive rights and structures of democratic participation (voting rights, the right to run for office, a reasonably representative electoral system, legal recognition of political parties, etc.) and other direct preconditions of democracy (most obviously, freedom of expression and association) but a whole additional set of protections against any institutionally authored, facilitated or merely permitted abuses and deprivations that might undermine equal respect and concern broadly conceived.
  At the other extreme, we find positions at or close to the ‘pure procedural’¹⁹xTo use Michelman’s term; n. 2 above, 39. pole, where constitutionalism is ‘defined down’ as a means to serve the ends of a democratic process of government. This kind of stance, associated with writers such as John Hart Ely²⁰xJ.H. Ely, Democracy and Distrust: A Theory of Judicial Review, Cambridge Mass.: Harvard University Press 1981. and Robert Post²¹xR. Post, Constitutional Domains: Democracy, Community, Management, Cambridge Mass.: Harvard University Press 1995. in recent years, understands constitutionalism as lacking any democracy-independent moral foundations. Instead, constitutional thinking and the constitutional process exist only to ensure that political decision-making and the public debate that informs political decision-making is accessible to all interests and preferences on equal terms.
  If these polar positions see democracy as subservient to an idea of the ‘good’ constitutional polity, or conversely, constitutionalism as subservient to democracy conceived of as a purely procedural ideal, a third set of intermediate positions avoids subservience in either direction but nevertheless retains a commitment to the idea of democracy and constitutionalism as fundamentally mutually supportive ideals. Jürgen Habermas, to take a prominent and highly influential example of this middle way, understands constitutional democracy as located in the symbiotic relationship between private and public autonomy in the era of political modernity.²²xSee in particular, J. Habermas, Constitutional Democracy: A Paradoxical Union of Contradictory Principles?, Political Theory 29 (2001) 766-81. The protection of a sphere of individual autonomy and the guarantee of an inclusive public domain of political discourse and participation are seen as twin and indeed co-original virtues of the modern age. And while their common origins suggest some level of moral compatibility or interconnection,²³xThis is further explored in Section V below. in the final analysis each virtue remains irreducible to the other. That is to say, there is no relationship of normative subordination or subsumption in either direction. There is, however, a key relationship of empirical interdependence. Not only is private autonomy deemed to be a factual prerequisite of effective public autonomy and of a healthy domain of political participation, but, reciprocally, the latter also offers the best guarantee of the continued protection of private autonomy. What we have, in short, is a conception of constitutional democracy where, unlike ‘defined up’ democracy or ‘defined down’ constitutionalism, a measure of normative distinction between the values associated with democratic process (public autonomy) and those associated with constitutionalism more broadly (private and public autonomy) is conceded, but where this divergence is deemed not to require any trade-off between them due to the mutually supportive causal relationship which obtains between the operation of the two sets of values.
  These are all powerful and influential positions, but none of them tells the whole story of the relationship between democracy and constitutionalism. Indeed, their very divergence inter se offers some indication of what is missing from each. In their desire to hold constitutionalism compatible with democracy, they tend to paint an unbalanced picture. On the one hand, they succeed in deepening our understanding of the ways in which certain prior constitutional notions are indeed necessary to the articulation of democracy, and to that extent they mark an advance from the earlier orthodoxy of modern constitutional thought as concerned too much with the ‘limitation’ of ‘limited government’ and too little with its basic ‘constitution’. On the other hand, however, in so doing they tend in their different ways to gloss over the continuing and unavoidable tensions between democracy and constitutionalism, and in particular the need for constitutionalism to draw upon resources other than democracy in its vital role of addressing the incompleteness of democracy. Whether by straining democracy beyond the limits of any broadly shared understanding of its terms and towards a thicker conception of constitutionalism which arguably stands beyond and perhaps against democracy, or by reducing constitutionalism to a purely democratic-process supportive role and treating that supportive role as itself fully democratically specified, or by admitting some tensions between democracy and constitutionalism but resolving these by resort to a speculatively serendipitous causal formula, these theories tend to close prematurely the open question of the relationship between constitutionalism and democracy. In so doing, they cannot grasp the iterative and indeed irresoluble quality of the tension between democracy and constitutionalism, and how this may coexist with a relationship of mutual support.
  As we shall specify more fully in due course, if anything, the tendency towards a diversity of singular understandings of the relationship between democracy and constitutionalism is amplified in the global age. However, this development takes place in a context where, in something of a reversal to the earlier modern tradition, democracy and constitutionalism again begin to appear in fundamental tension. On the one hand there are those, heirs to the more recent tradition of a democratically rooted constitutionalism, for whom the very idea of constitutionalism is deeply imperilled by the severing of the umbilical chord connecting it to the democratic state.²⁴xSee e.g., D. Grimm, n. 4 above. On the other hand there are those, heirs to the earlier tradition of modern constitutionalism as a bridle upon and limitation of public power, for whom constitutionalism can flourish anew beyond the declining range of the democratic nation state.²⁵xMuch of the literature on the constitutionalization of international law, for all its internal diversity of approach, takes this basically more optimistic line. Many representative examples of this genre can be found in two recent edited volumes; J. L. Dunoff and J.P. Trachtman (eds.) Ruling the World? Constitutionalism, International Law, and Global Governance, Cambridge: CUP 2009; J. Klabbers, A. Peters and G. Ulfstein (eds.), The Constitutionalization of International Law, Oxford: OUP 2009. Again, in their very singularity these understandings tend not to appreciate the overall picture, either overstating or understating constitutionalism’s dependence on democracy.

• 3 Constitutionalism and Democratic Incompleteness

  Let me now set out the basic structure of my alternative understanding of the double relationship between constitutionalism and (incomplete) democracy. As already noted, in this perspective constitutionalism is necessary both to realize democracy and to supplement and perhaps qualify democracy. In so doing, constitutionalism is responding to the empirical and normative dimensions of the incompleteness of democracy respectively, and in both cases is bound to do so in terms which cannot themselves be fully justified by reference to democratic criteria. In the introductory Section it was noted that certain broad features of this double relationship have remained constant over the period of modern constitutionalism, and it is this continuity which allow us to disaggregate the relationship between constitutionalism and democracy into a number of settled sub-themes or dimensions. An examination of these sub-themes will reveal the general ways in which constitutionalism functions to realize democracy, the general ways in which constitutionalism functions to supplement or qualify democracy, as well as providing a broad indication of how these two relational dimensions themselves interrelate.
  We can identify seven such sub-themes; namely authorship, stakeholding, representation, competence, non-democratic values, implementation and demarcation. The first four sub-themes – authorship, stakeholding, representation and competence – clearly relate to the question of realizing democracy. The fifth sub-theme, non-democratic-values, clearly refers to the supplementation or qualification of democracy. And the last two sub-themes – implementation and demarcation – straddle the divide between realizing and supplementing or qualifying democracy. Let us now say rather more about the way in which modern constitutionalism has traditionally responded to democratic incompleteness across the seven dimensions.

  3.1 Empirical Incompleteness; Constitutionalism as Democracy-Realizing

  We will begin with the four clearly democracy-realizing dimensions of the relationship between constitutionalism and democracy. First, constitutionalism addresses the deep question of authorship of the political community. Under what circumstances should an expression of political voice count as an act of constitution-making or constitution-amending – as a basic initiation or adaptation of a community as a political community and so as the appropriate container of a democratic system of government (the when question)? Secondly, constitutionalism addresses the question of membership, or more broadly and more attuned to the relevant nuances, the question of stakeholding. Whose interests and preferences should be taken into account in the operation of the democratic system (the who question)? Thirdly, constitutionalism addresses the question of representation. Through which forms ought stakeholders, on the authority of the authors, to be democratically represented (the which question)? And fourthly, constitutionalism addresses the question of prerequisite competence. What kinds of capacities are required of the stakeholders and their representatives for them to be competent (re)producers and operators of a democratic system, and what constitutional right and protections are necessary to furnish these capacities? In other words, how do we ensure that the relevant actors possess the wherewithal to operate the political system in a democratic fashion (the wherewith question)?
  Let us now look at the various democracy-realizing dimensions in a little more depth. What are the ingredients that constitutionalism supplies, without which democracy cannot realize itself, and in what sense are these democracy-realizing attributes dependent upon considerations other than purely democratic ones? Take first the question of authorship. This is traditionally viewed in constitutional theory through the concept, first developed by Abbe Sieyes, of pouvoir constituant or ‘constituent power’. In a tract that became the effective manifesto of the French Revolution,²⁶xA. Sieyes, What is the Third Estate? (1789) (Eng. Trans. M. Blondel, New York, Praeger, 1963). Sieyes sought to identify an originary polity-embracing power capable of giving birth to the constitutional settlement – the pouvoir constitué – and standing in some kind of relationship of pre-eminence over it. Constituent power, therefore, speaks to the most basic sense in which constitutionalism displays a ‘constitutive’ quality with regard to the (potentially) democratic polity. It energizes the initial process – the constitutional convention, constituent assembly or other device – through which is produced the initial constitutional document for a polity (or, indeed, through which is generated any successor constitutional document that announces a new founding unauthorized by its predecessor), which in turn provides the necessary framing conditions and any additional norm-generating capacity for the fashioning and operation of democracy within that polity.
  Arguably, too, any amendment of the constitution, involving an alteration of the framing document itself but in terms formally specified by the framing document, is also an act of authorship, in this case authorized but not exercised by the original framers, which will have a bearing on the ongoing fashioning and operation of any democratic settlement. Again here, constitutionalism provides a prerequisite to democracy, one that is necessary not in this latter case to ensure that democracy ‘gets going’, as in a founding document, but rather to ensure that democracy’s basic framing terms do not congeal and become a ‘dead letter’, insensitive to any changes in conditions which might affect their contextual suitability as basic framing terms.²⁷xOf course, some part of the (controversial) argument in favour of a more ‘activist’ conception of judicial review, and against strongly originalist conceptions of the judicial role in constitutional interpretation, depends upon constitutional judges, in the absence of formal amendment, themselves assuming a quasi-amending role in response to post-foundational societal change. See e.g., S.M. Griffin, American Constitutionalism: from theory to politics, Princeton: Princeton University Press, chapter 1.
  Yet constitutionalism’s supply of the terms and conditions of authorship of the polity, although necessary to democracy, is not itself democratically determined. There can be no definitive democratic answer to the question of who is the popular sovereign – of who gets to constitute the polity and under what conditions. As Hans Lindahl puts it, the ‘collective self-constitution’ of a political community ‘means constitution both by and of a collective self’.²⁸xH. Lindahl, Towards an Ontology of Collective Selfhood, in: M. Loughlin and N. Walker (eds.), The Paradox of Constitutionalism, Oxford: OUP 2007, p. 9-24, at p. 10. The collective self duly constituted (the of) is the product of rather than identical with the collective self constituting (the by). It follows that whatever democratic credentials are possessed by the latter duly constituted body cannot be used to provide retrospective warrant for the democratic credentials of the earlier constituting body, which remains a body without any democratic pedigree beyond its own self-assertion. A similar problem of unfounded foundations attaches, at one remove, to formal amendments of the constitution. Here, the immediate credentials of the amending constituency are located in the founding document, which in its amending formula specifies the terms of that constituency. However, that founding document is itself but the product of the self-assertion of an originary and constituent body necessarily lacking its own democratic pedigree. Again, the work of the constitution in enabling democracy possesses no definitive democratic warrant.
  Of course, all this is to say is that collective selfhood is progressively as well as regressively authorized, and so we should be wary of overstating the difficulties the problem of origins poses to democratic theory. Democratic authority is reflexively constructed over time – it is about the continuously self-amending relationship of a collective self to itself rather than a seamless continuity of identity and sameness.²⁹xSee e.g., Lindahl, n. 28 above at14 et seq.; see also B. Van Roermund, Sovereignty: Unpopular and Popular, in: N. Walker (ed.), Sovereignty in Transition, Oxford: Hart 2003, p. 33-54. Such reflexivity, indeed, is an inevitable feature of all collective activity, however organized,³⁰xSee e.g., H. Lindahl, Acquiring a Community, European Law Journal 9 (2003), p. 433-50. – of all bounded contexts of decision-making, whether or not professedly democratic. It does not mean either that the original constituent body lacks a plausible democratic claim, or that we cannot distinguish between more or less plausible such claims.³¹xSee e.g. S. Chambers, Democracy, Popular Sovereignty and Constitutional Legitimacy, Constellations 11 (2004), 153. It is a matter of sociological investigation how much contemporaneous legitimacy a self-proclaimed constituent body enjoys amongst the constituency for whom it professes to speak, and, given the operation of an ongoing reflexive process, later endorsements and re-endorsements mean that the level of any such social legitimacy can vary over time. Yet, these qualifications notwithstanding, there remains something arbitrary – something necessarily democratically unfulfilled and in some measure self-fulfilling about the original constituent act and signature in its own time. The original constitutional imposition of democracy, in other words, is also inevitably an imposition upon democracy.
  Similar considerations apply to questions of stakeholding and representation. The initial constitutional settlement will either specify or will provide the normative basis for specifying who counts in the polity. Categories of citizenship or nationality will set out who are full members of the polity for the purposes of taking part in democratic politics (in particular, voting and standing for elected office). Constitution-level decisions will be taken, or will enable other decisions to be taken, as to who should be ineligible (aliens), or partially eligible (those denizens, for example, who may vote in local or supranational elections but not in national elections), and how to move through the levels of eligibility (through residence, citizenship tests, etc.). The constitutional framework will also specify, or set out authoritative procedures for specifying which otherwise eligible categories may be disqualified (for example, prisoners or minors).³²xFor a wide-ranging study of the varied constitutional treatment of democratic membership at both national and supranational level in the European Union, see J. Shaw, The Transformation of Citizenship in the European Union, Cambridge: CUP 2007.
  In all cases, the identification of those who are deemed to have sufficient stake in the polity to be full participating members of its political system, while providing an important anchor for the democratic process, is again not itself something that can be compellingly derived from democratic principles. Rather, just as with authorship, the initial choice of constituency may create a self-reinforcing bias in addressing the question of who counts. This is not to say that membership of the polity is something that cannot be plausibly addressed and theoretically elaborated from a perspective of democratic principle. Many efforts have been made to do so.³³xFor a recent overview, see D. Miller, Democracy’s Domain, Philosophy and Public Affairs 37 (2009), p. 201-228. On the one hand, membership of the demos may be seen as something that should be restricted to those who enjoy a basic affinity and ‘we-feeling’. Those who share a basic level and continuity of sympathetic identification, of mutual trust and respect, and perhaps of common belief, may be seen to be a democracy’s ‘natural’ constituency. This can be defended in moral philosophical terms, because such an approach to the determination of the ‘collective self’ in the democratic process of ‘collective self-government’ may be animated by just those considerations of collective autonomy as appear to be so closely ethically implicated in the endorsement of the basic idea of procedural democracy itself. Additionally, an affinity-based conception of political community can be defended in pragmatic terms, because such a cohesive or solidaristic demos is advantageously resourced to supply ‘the battery of power’³⁴xM. Canovan, Nationhood and Political Theory, Cheltenham: Edward Elgar 1996. necessary to generate effective self-government. On the other hand, membership may be seen to be properly tied to the impact of the polity upon one’s interests and preferences. From this perspective, democratic theory has produced many versions of the ‘all affected’ principle to argue where the line of impact should be drawn and what the appropriate ‘impacted’ constituency of collective self-government should be in any functioning polity.³⁵xSee e.g., J. Habermas, Moral Consciousness and Communicative Action, Cambridge Mass.: MIT Press 1995; R. Goodin, Enfranchising All Affected Interests and Its Alternatives, Philosophy and Public Affairs 35 (2007), 40-68; T. MacDonald, Global Stakeholder Democracy, Oxford: OUP 2008.
  Far from a dearth of democratic theory on the question of membership of the demos, therefore, what we have in fact is a surplus. These two quite different perspectives, each of which also admits of a variety of different possible internal refinements, are apt to draw conflicting conclusions, with the affinity- and affect-based approach tending to be the more selective and the impact-based approach tending to be the more inclusive.³⁶xMiller (n. 33 above) connects these approaches to what he calls ‘L-Democracy’ (i.e., liberal) and ‘R-Democracy’ (i.e. radical) respectively, p. 204-207. What this means is that even to the extent that constitutional decisions as to the proper stakeholders of the polity are sensitive to democratic considerations and, in the post-constituent moment, remain open to arguments that do not simply reinforce the existing constituency’s sense of the proper constituency, such considerations themselves are controversial inter se and cannot provide definitive answers within the prism of democratic theory and principle.
  For their part, questions of representation suffer from a similar form of democratic underdetermination. Modern constitutionalism has been the constitutionalism of large nation states, and the democracy it has precipitated has been of the representative kind appropriate to large nation states rather than the assembly democracy of Athenian and later small city republics. This has raised myriad questions of institutional design. How is it possible to ensure representative forms best capable of approximating equality of influence amongst members when factors of size and internal differentiation may lead to certain interests and preferences being ignored or systematically outweighed? Various systems of representation vie with each other to answer questions about the optimal allocation of influence. Majoritarian systems of election, oriented towards rewarding parties and constituencies possessing a relative majority throughout the entire population with a clear mandate to govern, vie with proportional representation systems oriented towards rewarding different interests in the community in a statistically accurate manner.³⁷xAn Interparliamentary Union study of 150 democracies in 1993 revealed 83 majoritarian electoral system and 57 PR systems, with the remainder mixed. Electoral Systems Worldwide: A Comparative Study, Geneva: Interparliamentary Union 1993. Unitary systems, oriented towards treating the demos as a cultural and institutional singularity, vie with federal or consociational systems in which the overall system of representation recognizes systematic cleavages in the population in terms of sub-territorial national or regional identification or other forms of cultural, linguistic or religious affiliation.³⁸xSee e.g., S. Choudhry (ed.), Constitutional Design for Divided Societies: integration or Accommodation?, Oxford: OUP 2008. Parliamentary systems oriented towards single or coalition party government vie with presidential systems in which representative influence is channelled separately to different governmental organs (executive and legislative) and executive power is deemed sufficiently distinct to be the subject of the special forms of empowerment and constraint associated with the presidential office.³⁹xSee e.g., R. Elgie, From Linz to Tsebelis: three waves of presidential/parliamentary studies?, Democratization, 12 (2005), p. 106-122.
  In all cases, again the basic constitutional choice is democratically determining rather than democratically determined. There is no one best democratic way how to design a system of democratic representation. Partly, this is a question of the different social environments of democracies demanding different institutional solutions. However, even if we take into account the best evidence of the prior pattern of preferences, interests and allegiances in the relevant environment, there is no objective, situationally specific, democracy-revealing ‘fact of the matter’, but a whole host of contending considerations about how to measure and assess effective influence. Granted, the mature representative system may well in fact closely track the original balance of voice and influence within the collective authorship of the polity – a link which tends to be clearest in federal or consociational systems. Yet this merely refers us back to the contingency of the initial authorial formulation, and so demonstrates how the answers to constitutional questions of representation, like those of authorship and stakeholding, rather than decided in the domain of disinterested democratic principle may be strongly path-dependent.
  If we turn, finally within the category of democracy-realizing dimensions of constitutionalism, to questions of competence, here somewhat different but ultimately similar considerations arise.⁴⁰xSee e.g., J. Waldron, Law and Disagreement, Oxford: OUP 1999, chapter 13. Unlike the other three dimensions, there exists some kind of uncontroversial core of democratically mandated constitutional prerequisites to democracy. Democracy, even in its thinnest proceduralist understanding, requires certain minimal conditions of political and personal freedom on the part of its stakeholders and office-holders. Without freedom of speech and freedom of association, and without the due process rights associated with liberty from arbitrary arrest or interference by the state, it would be impossible to guarantee the basic processes of opinion-formation and dissemination in the absence of which no culture of uncoerced collective decision-making on questions of import to that collectivity can flourish. But even within these core categories, hard questions arise about how far we should go. Do we protect only directly ‘political speech’ or is any expression and communication of opinion on any matter of potentially public interest to be safeguarded? Do we protect only directly political assemblies, or any forms of collective coming together, whether in the economic or the religious sphere, that may indirectly impact upon collective political opinion-formation and dissemination? Do we treat freedom of information – the entitlement to have one views informed and ones preferences influenced by the best available evidence on matters of public importance and by detailed monitoring of the operation of public services and the behaviour of public officials – as every bit as much a prerequisite to democracy as are freedom of speech and association? And what of privacy, property, subsistence and security? As a simple causal thesis, are these indispensable planks of the platform necessary for the performance of democracy, or can we conceive of a thriving democracy in their absence, or, more likely, in their merely heavily qualified presence? And what, too, of these competences, which may overlap considerably with those listed above, but which are defended, not (or not exclusively) on the basis that they are causally prerequisite to democracy, but because they stand in an internal moral relationship to democracy? That is to say, they are treated as democratically-relevant competences because they are deemed to recognize and protect just those values which it is the moral rationale of democracy to recognize and protect – as in Dworkin’s expansive conception of the attitude of equal respect and concern for the moral independence of all members of the community.
  So once we get beyond a consensus on the core of prerequisite competences, or, even more modestly, a consensus on the bare idea of a core which may itself begin to fray immediately as we begin to try to delimit that core, democratic principle and theory becomes unable to determine those competences that are required for democracy’s own meaningful elaboration. Again, the problem is not a lack of democratic theory, but a surfeit – one that marks out this particular question as a key battleground for the standing controversy between ‘thin’ and ‘thick’ conceptions of democracy. In the final analysis, there is no democratically mandated right answer, and again constitutionalism must draw upon resources other than democracy in answering one of the unavoidable questions about the realization of democracy.
  In summary, therefore, we may conclude by noting a distinction between the moral and political purpose and the social epistemology of constitutional thinking across the four dimensions of democracy-realization. On the one hand, when addressing the when, who, which and wherewith questions, constitutionalism thinking and practice is geared to the realization of democracy. On the other hand, in so doing, although there is much scope for drawing upon the resources of democratic thinking in addressing these questions, these democratic resources cannot provide definitive answers. Rather, from a democratic perspective a crucial element of contingency remains. Other factors will inevitably play their part in the making of key constitutional decisions, and, indeed, so intensely reflexive are the democracy-engaging processes set in motion by constitutionalism that there will be a tendency towards the reinforcement of foundational biases.

  3.2 Normative Incompleteness; Constitutionalism as Democracy-Supplementing and Democracy-Qualifying

  Fifthly, and bringing in the second general relational vector connecting constitutionalism to democracy, we may identify a broad and general democracy-supplementing or qualifying dimension. As we saw in Section Two, this is a domain of constitutional theory and practice which attracts a double doze of scepticism. Not only its content, but even its very existence stands as controversial between different positions. In particular, any constitutional reference point outside of democracy is denied by those who would reduce constitutionalism entirely to the service of a procedural version of state democracy, just as, in the recent symbolic pomp of democratic constitutionalism, the possibility of a non-democratic constitutional supplement has been glossed over even by those, such as Dworkin, and to a somewhat lesser extent Habermas, whose much ‘thicker’ constitutional vision is presented as nothing more than the full working out of the premises of democratic theory. Accordingly, this fifth category is one best defined negatively and open-endedly. It addresses the question of what, if any, are the other, non-democratic values, whether understood in terms of individual rights or collective goods that ought to be pursued and upheld by the constitution (the what else question)?
  The list of candidate values here looks very like the open-ended list we perused previously under the democracy-realizing head of prerequisite competences. Such individual rights as property and subsistence, together with other basic welfare or social rights would figure prominently, as would certain public goods less easily reduced to an aggregation of individual rights, such as security and education. So too, at a higher level of abstraction, would some general desiderata which might in some catalogues be seen as part of a democratic conception of the good sensu largo, but which might equally in other moral schemes be seen as equiprimordial or interdependent with rather than reducible to democracy, or indeed as deeper values served by democracy. We are talking here of values as broadly pitched as equality, liberty, dignity or fraternity.⁴¹xSee e.g., Dworkin, n. 9 above, p. 25-6. See also Waldron, n. 40 above, chapter 13.
  This last category of more abstract values neither reducible to nor in conflict with but somehow interwoven with democracy is an intriguing one, and one to which we return in the concluding Section. But for the moment, it is necessary to lay down a broader marker about the knowledge claims associated with all putative members of the category of democracy-supplementing or qualifying values. The point here is not to seek to draw the unequivocal conclusion that the kinds of constitutional claims which might be and frequently are made on behalf of such individual and collective values, however concrete or abstract, are necessarily additional to or inconsistent with democracy. That would be to make an equal and opposite error to those who would declare with similar conviction that any and all such constitutional values are necessarily implied by and so ultimately reducible to democracy (however thinly or thickly conceived).
  Rather, all we can confidently say is that there is no generally agreed understanding of democracy on the basis of which all other candidate constitutional values are necessarily implied by and reducible to its terms. Given the wide range of possible understandings of democracy, indeed, the most we can hope for by way of agreement as to democracy’s core meaning is some kind of overlapping consensus across that wide range. However, since any conception of democracy (or rather, any shared dimension of various and diverse conceptions of democracy) capable of attracting the requisite overlapping consensus could be no more than a thin proceduralist one, there would remain many other candidate constitutional values palpably not reducible to the terms of that thin consensus. In other words, to the extent that there is or could be minimum agreement about democracy as a cornerstone of constitutionalism, that agreement does not and could not extend far enough to cover the many other candidate constitutional values to which at least some parties to any such minimal agreement would subscribe. We have no choice, therefore, when seeking to make sense of modern constitutionalism in a way that does not close off options by (deeply controversial) definitional fiat, but to hold open the putative category of non-democratic values.

  3.3 Empirical and Normative Incompleteness: Two Mixed Constitutional Functions

  We can complete our catalogue of modern constitutional functional domains by referring to two such functions straddling the distinction between democracy-realizing and democracy-qualifying.
  Sixthly, then, constitutionalism also addresses democratic incompleteness at the level of implementation of the overall system of government (the how question). Just because constitutional law, uniquely amongst categories of law within the positive legal order of the modern state, cannot resort to any higher socially sourced and institutionally grounded legal authority in support of its own normative purposes, it must take responsibility for its own effectiveness as law. This work of self-execution can in turn be sub-divided into normative or regulatory design questions, and integrative or cultural questions.⁴²xSee e.g., D. Grimm, Integration by Constitution, International Journal of Constitutional Law 3 (2005), p. 193-208.
  Normatively, what institutional framework – that is to say, what particular combination and interconnection of legislative, executive and judicial functions and institutions, will best reflect and achieve the normative values and purposes of the constitution? Importantly, just because these deep normative purposes may be mixed, relating to democracy and (arguably) non-democratic values alike, so too their implementation may reflect this mixity of democracy-realizing and democracy-supplementing or -qualifying functions. For example, an independent emphasis on those presumptively non-democratic values based on individual rights might be thought to be well served by a strong regime of judicial review of legislation and of administrative discretion.⁴³xAlthough as Waldron argues, this does not necessarily follow, and a robust conception of rights protection can be defended as compatible with a constitutionally unchecked system of majoritarian legislation. See n. 40 above, chapter 10 ‘Between Rights and Bills of Rights’. Or a strong emphasis on the goods of internal and external security might be thought to be well served by a strong emergency powers regime against internal and external threats, or by protecting certain types of executive action from close contemporaneous involvement or scrutiny by the other branches of government.⁴⁴xFor a critical appraisal of these types of arguments in the post 9/11 United States environment, see S. Holmes, The Matador’s Cape: America’s Reckless Response to Terror, Cambridge: Cambridge University Press 2007. Equally, even to the extent that the normative design of the constitution is only concerned with the implementation of the value of democracy, this will by no means always require democracy ‘all the way down’. Rather, even the fullest and ‘purest’ commitment to the collective procedures of self-rule will accept that the very objective of collective self-rule will be frustrated if it is extended to very nook and cranny of decision-making relevant to that collectivity. Rather, in order to be effective as self-rule democratic procedures in a large and complex modern polity inevitably required be supplemented in some measure (the extent and forms of which, however, are themselves a matter of significant controversy and dispute) by certain ‘depoliticizing’ mechanisms that instead emphasize expertise, disinterestedness, dispassionate analysis or deliberative rationality.⁴⁵xSee e.g., P. Pettit, Depoliticizing Democracy, Ratio Juris 17 (2004), 52-65.
  Culturally, to what extent can the constitution assist in promoting and sustaining the cultural conditions most conducive to its own effective implementation? This requires us to look at the symbolic rather than the normative performance of the constitution. We noted in our discussion of stakeholding above that democratic theory will often favour constitutional conceptions of membership that stress the affinity and mutual sympathy of the members. When we come to consider the role of constitutionalism in the implementation of its own normative programme, however, the constitution itself no longer simply reflects existing solidarities. Rather, it seeks through its own symbolic resonance to cultivate existing or new solidarities.⁴⁶xFor a comparative analysis of the different trajectories of cultivation of communal solidarities in different constitutional models, see M. Rosenfeld, The Identity of the Constitutional Subject: Selfhood, Citizenship, Culture and Community, London and New York: Routledge 2010, esp. chapters 5-7. Typically, it does so by means of encouraging an attachment to itself or to its key institutions (e.g. a Parliament, a presidential office, a Supreme Court, a monarchy), whether through the inclusive procedure or successful fact of its generation or – more significantly in the long run – in the case of a mature settlement, due to ‘the people attributing to [it] a metalegal meaning’ as the durable standard bearer of the ‘achievements and aspirations of society’.⁴⁷xD. Grimm, n. 42 above, at p. 204. Again, as with the normative dimension of implementation, this symbolic dimension has to do with the effectiveness of the whole, and so it cuts across such democratic and other values as may be contained in the constitutional settlement.
  Seventhly, and finally, constitutionalism has also always asked and answered the boundary questions of democracy (the where question). It has been required to specify the territorial and/or functional ‘scope’⁴⁸xR. Bellamy and D. Castiglione, Legitimizing the Euro-‘Polity’ and its ‘Regime: The Normative Turn in EU Studies’, European Journal of Political Theory 2 (2003), 7-34. of the polity. And like questions of implementation, questions of demarcation have traversed the divide between democracy-realizing and democracy-qualifying functions. On the one hand, territorial questions are typically democracy-realizing. A specification of the territorial boundaries of the polity must be made prior to the realization of a democratic system, but in so doing it need not engage values which directly rival or qualify democracy itself. Rather, what will be germane will be a mix of external geopolitical considerations of the relative strength and overall alignment of powers within the region and the kind of internal associative considerations we have already encountered in our discussion of constitutionalism’s democratic stakeholding criteria. A specification of the functional boundaries of the polity, on the other hand, by defining certain spheres as in or out of democratic bounds, whether this be the system of private property or the organization of religion, may well do so in direct consequence of, and in furtherance of its engagement with other, presumptively non-democratic, fundamental values.

• 4 The Global Age of Constitutionalism

  It is this seventh constitutional function of demarcation, as we shall see, that has been radically transformed under contemporary conditions of globalization. But why and how has this happened?
  If we think of globalization as speaking to the various intersecting ways in which different circuits of power and influence cut across national boundaries with increasing intensity,⁴⁹xFor a representative range of contemporary thinking on globalization, see A. Jones, Globalization: Key Thinkers, Cambridge: Polity 2010. then we can begin to appreciate how this leads to changes in our understanding of constitutionalism in general and constitutionalism’s relationship to democracy in particular. In a nutshell the constitutionally relevant effects of the growth of various transnational circuits of power and influence to which the contemporary wave of globalization refers, from the increasing flow of capital, goods, people and services, to the growth of new communications media, to new intercultural influences and institutional forms, can be categorized under the heads of new challenges, new opportunities and new preferences respectively.
  As regards challenges, present day state democracies increasingly find themselves impotent to deal with issues that take the form either of externalities – how decisions taken in one state have a serious impact on people living in other states – or, relatedly, of new collective action problems that can only be resolved by co-ordination or co-regulation at a transnational level. Whether we are dealing with global banking and investment markets, the flow of refugees, transnational criminality, the spread of infectious diseases or global climate change, individual states increasingly lack the rules and resources to address the relevant questions on their own. Alongside these challenges and constraints, and acting both as response and reinforcing cause, there are new opportunities for co-ordination. Since the middle of the 20th century, with the initial impetus provided by the post-war internationalism of the United Nations, there has developed a range of global and regional institutions and regulatory forms possessing a mixture of territorial and functional jurisdiction. These institutions and forms are empowered to address new transnational regulatory problems, but they also contribute to them in two ways. They do so first, by further disempowering state institutions – for example, the EU by introducing a single currency and by limiting state aids to national economic concerns in the name of economic union, severely restricts the way in which individual member states can respond to a domestic economic crisis.⁵⁰xAs was spectacularly illustrated in the Greek debt crisis of the early Summer of 2010. They do so secondly, by providing the means not just to respond to existing problems but to produce new forms of transnational interconnectedness, for example, less restricted cross-border movement of persons, corporations, goods and services, which in the scale and intensity of activity and interconnection they enable (e.g. more multinational corporations with a stake in many states, more interdependent transport, more interdependent money markets, larger dual national or denizen communities) in turn generate new externalities and new collective action problems. At the level of preference structures, too, globalization is a double-edged sword. The development of transnational interests and value structures can create new global or universal normative standards and vernaculars – for example as regards human rights.⁵¹xSee e.g., C. Gearty, Can Human Rights Survive?, Cambridge: CUP 2005. Yet by bringing different cultures into overlap, it can also create new or heightened oppositions and enmities, as with the rise of new forms of regionally sourced fundamentalism.⁵²xSee e.g., Ruthven, n. 15 above.
  The regulatory circuits that develop within this complex set of movements affect the demarcation dimension of constitutionalism in two apparently opposite but connected ways, with various ramifications for other dimensions of constitutionalism. In some cases, globalization engenders more and more heavily overlapping boundaries relevant to constitutionalism, while in others, by contrast, it dispenses with the logic of boundaries altogether.
  Globalization engenders more boundaries and more heavily overlapping boundaries to the extent that it encourages the creation of new post-state polities, paradigmatically the EU, but increasingly too the other regional economic unions, which no longer stand in a relationship of mutual exclusivity with each other or with state polities.⁵³xSee e.g., my Beyond Boundary Disputes and Basic Grids: Mapping the global disorder of normative orders, International Journal of Constitutional Law 7 (2008), 373-396. That is to say, the more basic integrity of boundaries under the modern state system, the sense that the constitutional limits, wherever marked out, would be coincidental with the limits of other constitutional polities, and that in this way internal sovereignty would be aligned with external sovereignty, no longer holds. This in turn creates two sets of difficulties, which we may again illustrate through the example of the EU as an entity notorious both for its ‘democratic deficit’ and for its institutional complexity. On the one hand, by dividing up democratic functions into a greater number of polity ‘slices’, the EU threatens vital resources of democracy-enabling political culture at both state and post-state level. On the other hand, as the political life of individuals and communities is increasingly dispersed across a variety of polities, the complex and obscure technology of the coordination and co-articulation of these different regimes becomes key to life-chances.
  On both of these counts, each of constitutionalism’s democracy-realizing functions become more urgent, more stretched and more likely to encounter problems. As the aborted constitutional settlement of the EU in 2003-5 demonstrated, authorship problems are compounded in a postnational unit whose constituent power – perhaps the European people, perhaps the European peoples, or perhaps a mixture of the two – is a matter of deep dispute, not least because it overlaps the sources of constituent power of the pre-existing states.⁵⁴xSee my Post-Constituent Constitutionalism in the EU, in: Loughlin and Walker (eds.), n. 29 above, p. 247-268. Membership criteria, too, become more problematic in the European Union when one is dealing with a secondary political community in which every citizen is always already a citizen of a primary state community, and where the presence of both intra-EU second country nationals and non-EU third country nationals provide a more graduated and less clearly demarcated sense of distance from the citizenship core than under the traditional insider/outsider dichotomies of one-dimensional state citizenship.⁵⁵xSee e.g., Shaw, n. 32 above: N. Walker, Denizenship and Deterritorialization in the EU, in: H. Lindahl (ed.), A Right to Inclusion and Exclusion?, Oxford: Hart 2009, p. 261-72. For their part, constitutional systems of democratic representation have been predicated on the exclusivity of the democratic arenas they inhabit, and it becomes difficult to make constitutional sense of representative prospects and possibilities when this logic of exclusivity is displaced. Arguments about federalism take on a new dimension in political configurations that are multi-level rather than just two-level; arguments about consociationalism become more complex when the number of overlapping minorities expands; arguments about the relative merits of strong parliamentarianism or presidentialism are confounded when there are a multiplicity of parliaments and presidents located at different levels.⁵⁶xSee e.g., V. Bader, Complex Citizenship and Legitimacy in compound polities (MLPs and MLG): The EU as example, EUROSPHERE, WP 1 (2008). What is more, if we turn to the democracy-realizing dimension of constitutional implementation, the new split and multi-level system produces not only problems of complexity at the normative level but also motivational problems at the cultural level. Much of the criticism of the EU constitution as a false constitution, or as a ‘low intensity constitution’,⁵⁷xSee M. Maduro, The Importance of Being Called a Constitution; Constitutional Authority and the Authority of Constitutionalism, International Journal of Constitutional Law 3 (2005), 332-356. or as a treaty masquerading as a constitution, and much of the popular endorsement of that criticism in the backlash against the constitution culminating in the referendum defeats of 2005 in France and Holland, had to do with the perceived symbolic overkill of political elites seeking to attach the rhetoric of constitutionalism to such an apparently deracinated political form as a post-state polity.⁵⁸xSee e.g., my Not the European Constitution, Maastricht Journal of European and Comparative Law 15 (2008), 15-21.
  This multiplication of boundaries stands in stark contrast to the other self-styled postnational constitutional development. For a second form of constitutional initiative for the global age concerns the attribution of constitutional credentials to normative phenomena that are not in any strong sense polity-specific or polity-demarcated. Elsewhere, I have discussed many of these trends under the heading of ‘beyond the holistic constitution’.⁵⁹xSee my Beyond the Holistic Constitution?, in: P. Dobner and M. Loughlin (eds.), The Twilight of Constitutionalism?, Oxford: OUP 2010, p. 291-308. The holistic constitution, again with the state as its paradigm form, is a constitution that seeks to embrace and contain the polity as whole and at a number of different levels simultaneously – in terms of formal legal sovereignty and comprehensive institutional design as much as in terms of the exclusivity of the underlying claim to popular political sovereignty or the reference to a distinct political and cultural ‘society’. Even where these latter thick elements are in jeopardy, there may, as in the case of the EU, still be a holistic element at the level of ‘own’ legal system and institutional design. In the new post-holistic constitutionalism, in contrast, constitutional claims are far more partial, constitutional traces far more fragmentary. They concern regional or global human rights treaties, the specific peace-keeping dimensions of the UN Charter, or developed single issue ‘sectoral’⁶⁰xA. Peters, Membership in the Global Constitutional Community, in: J. Klabbers, A. Peters and G. Ulfstein (eds.), The Constitutionalization of International Law, Oxford: OUP 2009. treaty regimes in areas such as climate change or international crime. They concern the thin universalism of certain core principles of customary international law. They concern private or hybrid public/private self-regulatory regimes for transnational functions such as sport (e.g. IOC, FIFA) or the internet (e.g. ICANN). What these highly diverse forms have in common is their lack of holistic attributes, their absence of claims to comprehensive authority in any register, whether legal, institutional, political or societal. The multi-layered regulatory intensity and self-sufficiency of the holistic polity is substituted by the open-ended scope and connectivity of a more networked pattern of regulation. What is more, to the extent that these initiatives are proclaimed to be constitutional – and it is striking how widespread the language of constitutionalism has become in certain old corners of international law as well as in some of the new functional regimes⁶¹xFor critical perspectives on the spread of international constitutional discourse, see M. Koskenniemi, The Fate of Public International Law: Between Technique and Politics, Modern Law Review 70 (2007), 1-30; D. Kennedy, The Mystery of Global Governance, in Dunoff and Trachtman, Ruling the World?, p. 37-68. – they tend to concentrate on values other than democracy, and so focus on what we traditionally understood as the democracy-qualifying and supplementing functions of constitutionalism when applied in state-democratic setting. For the mixture of individual rights and public goods with which the new forms of transnational post-holistic constitutionalism are typically concerned, from peace and environmental security to consumer sovereignty and the protection of ‘first generation’ freedom and ‘second generation’ social rights, are those that are quite widely understood as distinctive from and perhaps challenging to democracy, or at least as in keeping with an expansive sense of democracy’s meaning and purposes.

• 5 The Resilience of Democratic Incompleteness in Constitutionalism’s Global Age

  Earlier I suggested that the onset of the global age of constitutionalism has generated a new opposition. There is an increasingly sharp division of views between those who see the constitutional paradigm as approaching exhaustion in the global age and those who see it as undergoing a transformation and reinvigoration. The former tend to concentre on the hollowing out of state polities as the traditional hosts of democratic constitutionalism, the depleted democratic credentials of post-state polity alternatives, and the one sidedness of a post-holistic constitutionalism seemingly remote from any idea or practice of democracy.⁶²xSee Grimm, n. 4 above. See also essays by Loughlin (What is Constitutionalization?, p. 47-72) and (Grimm The Achievement of Constitutionalism and its Prospects in a Changed World, 3-22) in: Dobner and Loughlin (eds.), The Twilight of Constitutionalism? The latter, while by no means necessarily unconcerned about the fate of national or postnational democracy,⁶³xOne important sub-distinction here is between those like Anne Peters (see her Compensatory Constitutionalism; The Function and Potential of Fundamental International Norms and Structures, Leiden Journal of International Law 19 (2006), 579-610, who talk about the ‘compensatory’ quality of constitutionalism at the postnational level, so acknowledging the basic loss to constitutionalism (as well as to democracy) as a whole occasioned by the diminution of democratic constitutionalism at the national level; and those who believe that the trend in favour of the redirection of constitutionalism towards non-democratic transnational settings, while undoubtedly transformative of the role of constitutionalism, does not necessarily entail any depletion in the value of constitutionalism as a regulatory ethic (see. e.g. essays by G. Teubner, Fragmented Foundations: Societal Constitutionalism Beyond the Nation State, and by U.K. Preuss, Disconnecting Constitutions from Statehood: Is Global Constitutionalism a Viable Concept?, in: Dobner and Loughlin (eds.), The Twilight of Constitutionalism? concentrate instead on the added value of the new transnational forms.
  What both of these positions deny or downplay, in their very different ways, is the continuing significance of constitutionalism’s double relationship with democracy. On the first view, there is no or limited value in looking to the non-democratic virtues of constitutionalism in and across settings where these are not balanced by democratic capacity and by the forms of constitutionalism associated with realizing that capacity. Constitutionalism as a means of realizing democracy, from this perspective, is so central to the constitutional function that any constitutionalism that professes itself as such in the absence of realized or realizable democratic forms is wholly or largely dismissed. On the second view, by contrast, the forms of constitutionalism at the postnational level are reckoned to be in no sense compromised by their lack of democratic credentials.
  In conclusion, I want to argue that each position dismisses the relevance of constitutionalism’s other relationship to democracy too quickly. As regards the defenders of postnational constitutionalism without democracy, ideological and meta-democratic considerations counsel against any complacent acceptance of their position. As for those who hold postnational constitutionalism to be valueless or essentially diminished due to the declining resonance of state democracy, broader structural and moral consideration demand that this position likewise be not too readily affirmed. In either case, it is at root the resilience of constitutionalism’s double relationship to democracy which prompts this conclusion.
  Let us take first the two objections that might be levelled against those who would defend postnational constitutionalism without democracy. Ideologically, constitutionalism, for all the fashionable claims of today’s postnational adherents concerning the global appeal of the values it represents, has traditionally been a situated discourse. It has nurtured and preserved its authority as being embedded within a particular political setting. Constitutional design in some respect inspired by general constitutional thinking may frame the particular polity, but the particular polity also provides a vital context for the authoritative interpretation of constitutionalism’s framing values. When, for example, Habermas talks of constitutional patriotism,⁶⁴xSee e.g., the essays collected in his The Postnational Constellation, Cambridge: Polity 2001, M. Pensky trans.). his pivotal point is that the general values of constitutionalism have to be adopted by a community as their own before they can achieve any ideological resonance. Their patriotism is not centred on the ideas of constitutionalism in the abstract, but on general constitutional values as mediated and concretized through their own particular political and cultural history. The increased and increasingly effective mobility of constitutional ideas,⁶⁵xSee e.g., S. Choudhry (ed.), The Migration of Constitutional Ideas, Cambridge: CUP 2006. it follows, is not based upon the infinite scope of an abstract universalism, but upon the context-variant replicability and adaptability of a particular model of political thought across different times and places. And so it is that constitutional thought finds it hard to gain traction in non-polity embedded connects. It further follows that we should be suspicious of the impact and value of a purely ‘external’ and polity-disembodied academic discourse of constitutional thought – of those claims that find no or only faint echo and those labels that find little endorsement, say, in the actual insider self-understanding and practice of this self-regulating transnational entity, or of that global functional regime, or of these general regimes of international law.
  Secondly, and closely related to the question of the presence or absence of a situated polity, as a matter of historical practice, even where constitutionalism involves the qualification of democracy, the terms of such qualification tend to possess a meta-democratic pedigree. The idea of a popular sovereignty and constituent power, as we have seen, is an early and central theme of political modernity. It is one that predates effective operational democracy, and one, consequently, that does not necessarily imply a democratic form of government. Yet its performative meaning is itself as a form of higher order- or meta-democracy – of a democracy about democracy (or about whatever other form of quotidian political organization is chosen). An apparent difficulty with most postnational forms of constitutionalism, therefore, is that as well as not being grounded in quotidian democratic arrangements, they typically lack even this kind of meta-democratic founding.
  Where the absence of such foundations is not simply ignored or passed over as an unavoidable deficit by the defenders of postnational constitutionalism, democratic authority is either claimed to be supplied indirectly – a pooled delegation of national democratic authority, or dismissed as somehow irrelevant or even inappropriate to the non-democratic values and forms of political organization with which it is concerned.⁶⁶xThe delegation argument is more often implicit than explicit. Even Habermas, with his strong commitment to directly engaged forms of democracy, seems to accept a kind of delegated legitimacy for the global (if not the regional) level in his defence of the United Nations as one player in a multi-level transnational constitutional settlement; see J. Habermas, Does the Constitutionalization of International Law still have a chance?, in: J. Habermas, The Divided West, Cambridge: Polity 2006, C. Cronin trans.), p. 115-193. But such responses are not convincing. As regards the argument from delegation, this raises structural questions, duly considered below, to which as we shall see no fully satisfactory answer can be given. As regards the redundancy argument, the notion that non-democratic values and forms do not in any case require a democratic pedigree can be quite simply turned on its head. As we saw in our discussion of the traditional dimensions of the relationship between constitutionalism and democracy, both the question of the existence and content of non-democratic constitutional values and that of the nature and extent of non-democratic forms of implementation of even the most procedurally unimpeachable general systems of collective self-rule, are themselves controversial. All the more reason, then, that a democratically aware transnational constitutionalism endeavour to ensure that such non-democratic values and forms retain some kind of meta-democratic imprimatur and backstop accountability.
  Yet while these factors pose difficult challenges to postnational constitutionalism, and ensure a sceptical reception for it beyond the narrow academic or specialist confines where it is sponsored, we should not push this too far. The ideas of postnational constitutionalism should not be dismissed out of hand just because they lack a polity-specific embedding or a meta-democratic pedigree, and so just because they may perform no democracy-realizing function within a discrete polity. Again, there are two reasons for this.

  First, in structural terms, we have already noted that in the global age of overlapping polities and of individual and community life-chances becoming spread across different polity sites, the relational dimension is key. No less important in the assessment of overall constitutional legitimacy than the discrete democratic credentials of a particular constitutional setting, therefore, are the ways in which particular constitutional points and vectors, themselves lacking in democratic credentials, may connect to other more strongly democratic settings in the relevant constitutional constellations of particular actors or communities; for example, the ways in which a regional human rights catalogue such as the European Convention of Human Rights of the (democratically attenuated) Council of Europe informs judicial review of legislative and executive action in the procedurally democratic (or, at least, aspirationally democratic) environment of 47 European states. This is not to say, of course, that every non-democratic site is somehow fully democratically homologated through the option of the adoption of its norms by the traditional democratic site of the state. Rather, it is to suggest that if any integrated assessment of constitutional legitimacy has always depended upon how constitutionalism responds to the two forms of democratic incompleteness – addressing the ways in which democracy requires to be realized at the same time as addressing the ways in which democracy requires to be supplemented or qualified – then the new forms of constitutional connections available between constitutional sites means that this is an assessment which may now plausibly be made across sites and not merely within one.
  Such a response, however, tends to accept much of the force of the ideological and meta-democratic critique. Its aim is somehow, through an idea of democratic pedigree and delegation or transfer, to reconnect a constitutionalism of non-democratic values with a democratic source. To that extent it will always remain vulnerable to the objection that in the global age the connection becomes an attenuated one, essentially second best to a single and holistically conceived polity solution. Indeed, the trend within the global regulatory configuration towards an ever more complex range of private and hybrid transnational forms is exacerbating this problem, rendering national democratic pedigree an ever more remote or even irrelevant source of endorsement without necessarily revealing new and alternative sites and forms of democratic pedigree.⁶⁷xOne obvious alternative source of democratic pedigree is the regional polity, and this has been a central theme in the debate over the constitutional legitimacy of the EU (see references at n. 54-58 above). Less obviously, but just as significant if not more significant in the long term, a case may be made for viewing certain functionally specialist self-constituting transnational private or hybrid regimes as democratic, or at least as its functional equivalent. They may be argued to be so to the extent that they seek to give voice to those interests most affected in their narrow focal policy field against a backdrop of the proliferation of other such specialist self-constituting sites and associated ‘democratic’ opportunities and a corresponding decline in influence of holistic state polities; see e.g., Teubner, n. 63. The kind of deep transformation of the very context of democracy contemplated here takes us to the heart of the question of the relationship between the resilient moral order of modernity and its changing architecture considered in the text below.

  However, there is a deeper response to postnational constitutional scepticism. This response turns not on a structural connection but instead on an as yet insufficiently acknowledged internal moral connection between constitutionalism in its democracy-realizing and democracy-qualifying modes.
  We can approach this by asking the broad question of how constitutionalism fits into what Charles Taylor calls the overall ‘moral order’ of modernity – referring to our most basic ‘natural’, unremarked and so often invisible understanding of the world and of our place and purpose in it.⁶⁸xC. Taylor, Modern Social Imaginaries, Durham: Duke University Press 2004, p. 3. From that perspective, the age of modern constitutionalism was not just about an innovative political architecture centred on the modern state. It was also the age at which a new cluster of mutually reinforcing values gradually came to define the deeper moral order; in particular individualism, egalitarianism, constructivism and progressivism. The idea of the individual as the basic moral unit in the world, and the consequential notion of the presumptive equality of all such basic morel units, replaced the medieval idea of human existence as part of an integrated whole in which different strata of society were accorded distinct stations and roles. Additionally, the notion that the world is constructed from our own evolved knowledge and practice, and that, over secular time, we seek to improve the world on our own terms on the basis of that knowledge and experience, is new to the modern age. It replaced the idea of human understanding, conduct and striving being oriented towards or judged against a pregiven order of things, situated not in secular time but in some sacred or otherwise ‘higher’ metaphysical time. In other words, individualism succeeded holism, equality succeeded status, constructivism succeeded essentialism and finite progress succeeded infinite conformity. The modern world was now, for the first time, something to be made over in their own terms by presumptively free and equal individuals.⁶⁹xSee further, R. Geuss, Public Goods, Private Goods, Princeton: Princeton University Press 2001; R. Dahl, On Political Equality, New Haven: Yale University Press 2006.
  Arguably, it is these elements of the moral order and its associated social imaginary that provide the deep context out of which our modern understanding of constitutionalism and democracy alike emerge. Crucially, although the relationship between the two purposes of constitutionalism – the articulation of democracy as the motor of a constructive and progressive approach to the world by free and equal individuals and associations of individuals and the qualification of democracy in term of other individual rights and collectively accomplishable and beneficial goods on the basis of the very same cluster of deep commitments to individualism, equality and designed progress – remains problematic and controversial, we can begin to see how these twin purposes belong together notwithstanding such continuing conflict. For constitutionalism and democracy are joined not in the sense that one conceptually reduces to or is assimilated by the other, or that they are otherwise fully mutually presupposed, or because they are bound in a relationship of necessary and sufficient mutual causality. Rather, they are joined in the sense that they emerge from and refer back to the same underlying moral order – they are woven out of the same moral fabric – and so are inextricable even as they remain in mutual tension. What is more, this very combination of mutual inextricability and mutual tension helps to account for how – as trailed in Section 2 above – our contemporary political and constitutional theory expends so much effort in the infinitely suggestive but ceaselessly controversial exercise of working out the precise relationship between them.
  Two important implications follow from this understanding of the character of the underlying moral order. First, the grand and remorseless project of working out the relationship between constitutionalism and democracy cannot be resolved by reference to some definitive and singular formula embracing both. That is why, indeed, the exercise is infinitely suggestive and ceaselessly controversial. For if the underlying moral order could be reduced to any such deeper foundational terms, then this would simply be another – in this case hierarchical – route, to the perfect compatibility of democracy and constitutionalism. But what we have, to repeat, is inextricability without perfect compatibility – hence the double aspect of constitutionalism as a means both to realize and to qualify democracy.⁷⁰xA similar double-edged relationship of interdependence and tension is set out in a series of highly suggestive recent works by Gianluigi Palombella. For him, the meta-regulatory dilemma, at once both productive and irresoluble, that is confronted and addressed by our deepest sense of national and transnational legality, lies between the particular collective ‘good’ and the universal ‘right’. According to Palombella, the Rule of Law – or the basic constitutional integrity of our overall political order – depends upon the guarantee that collective will and universal reason be held in dynamic balance, with neither ultimately prevailing over or subsuming the other. See e.g. The Rule of Law and its Core, in: G. Palombella and N. Walker (eds.), Relocating the Rule of Law, Oxford: Hart 2009, p. 17-42; The Rule of Law as an Institutional Ideal, in: L. Morlino and G. Palombella, The Rule of Law and Democracy: Internal and External Issues, Leiden: Brill 2010.
  Secondly, on this view of the crucial and resilient importance of the underlying moral fabric of modernity, the long familiar pattern of state-centred democracy and state-centred constitutionalism as a response to the double incompleteness of democracy can no longer be elevated to the ‘one best way’ of collective self-design. Rather, this supplies no more than one particular and ultimately contingent architectural formation for the modern moral order.⁷¹xI discuss the relationship between the moral-imaginary and architectonic aspects of political modernity from the perspective of constitutionalism in greater detail in The Global Age in the Longue Durée of Constitutionalism, International Journal of Constitutional Law (forthcoming). And if the environment which supports that architecture alters, as it has gradually done in the globalizing age, and it becomes no longer possible to arrange the political world as a series of mutually exclusive state polities, then we need to find, and, crucially, we may be capable of finding an adjusted architecture to accommodate the underlying moral order rather than assume that there was only ever one political architecture fit for modernity.⁷²xSee discussion at n. 67 above.
  From this perspective, constitutionalism’s ambivalent, double-edged relationship to democracy remains vital to the political understanding and regulation of the global age. For that ambivalence continues to capture something key about the nested centrality, rather than the singular pre-eminence, of democracy to the moral order of modernity. It reminds us that constitutionalism must continue to seek to articulate and realize democracy at the same time as it endeavours to qualify and challenge democracy in ways that are nevertheless justifiable from democracy’s broader moral horizon – even if this drama will no longer be played out in full across each and every mutually exclusive and neatly replicable state form, but distributed unevenly over the networked space of global society.

Noten

• 1 Its classical origins are of course Greek. Democracy is a compound of demos (people) and kratos (power).

• 2 F.W. Michelman, Brennan and Democracy, Cambridge Mass.: Harvard University Press 1998, p. 4.

• 3 On the iterative quality of democracy, see S. Benhabib, Another Cosmopolitanism, New York: Oxford University Press 2006, chapter 2.

• 4 On these historical trends, see e.g., D. Grimm, The Constitution in the Process of Denationalization, Constellations (2005) 12, 447; M. Loughlin, The Idea of Public Law (Oxford: Oxford University Press 2003, esp. chapters 3-6); C. Thornhill, Towards a Historical Sociology of Constitutional Legitimacy, Theory and Society, 37 (2008), 161-197.

• 5 T. Paine, The Rights of Man, New York: St. Martin’s 1988, p. 29.

• 6 J. Waldron, Is the Rule of Law an Essentially Contested Concept (in Florida)?, Law and Philosophy 21 (2002) 17-64, 157.

• 7 See G. Sartori, Constitutionalism: A Preliminary Discussion, American Political Science Review 56 (1962) 853-64.

• 8 Of course, there is also a causal relationship between these two trends. One of the reasons why democracy was such a slow developer was precisely because the constitutionalist argument for ‘limited government’ could be so effectively deployed against it, and vice versa.

• 9 This is not the place to document that claim in full. However, if we take only the American debate, few would doubt Dworkin’s claim that it has until recently been ‘the near unanimous view’ of American constitutional scholars and lawyers that a strong conception of constitutionalism, especially as articulated through a strong version of judicial review, is one that compromises democracy. See Dworkin, Freedom’s Law: The Moral Reading of the American Constitution, Cambridge Mass.: Harvard University Press 1996, p. 18.

• 10 Freedom House Report Democracy’s Century: A Survey of Global Political Change in the 20th Century, New York: Freedom House 1999.

• 11 Michelman, n. 2 above.

• 12 A. Sen, Democracy as a Universal Value, Journal of Democracy 10 (1999), 3 -17.

• 13 See Freedom House Report, n. 10 above; see more generally, J. Keane, The Life and Death of Democracy, London: Simon and Schuster 2009.

• 14 E. Hobsbawm, The Age of Extremes: The Short 20th Century, London: Michael Joseph 1994.

• 15 See e.g., M. Ruthven, Fundamentalism: The Search for Meaning, Oxford: OUP 2005.

• 16 Quoted in J. Keane, n. 13 above, 581.

• 17 For an excellent overview of contemporary constitutional theory’s treatment of democracy, and, in particular, its tendency to view ‘constitutional democracy [as] a tautology rather than an oxymoron’ (90) see R. Bellamy, Political Constitutionalism: A Republican Defence of the Constitutionality of Democracy, Cambridge: Cambridge University Press 2007 esp. chapter 3. Bellamy’s classification of the ways in which constitutionalism is reconciled with democracy is more detailed than the one I deploy in the present article, but the basic distinction between those forms of reconciliation based on a ‘thickened-up’ definition of democracy and those based on a ‘thinned-down’ version of constitutionalism also informs his approach.

• 18 Dworkin, n. 9 above, 17.

• 19 To use Michelman’s term; n. 2 above, 39.

• 20 J.H. Ely, Democracy and Distrust: A Theory of Judicial Review, Cambridge Mass.: Harvard University Press 1981.

• 21 R. Post, Constitutional Domains: Democracy, Community, Management, Cambridge Mass.: Harvard University Press 1995.

• 22 See in particular, J. Habermas, Constitutional Democracy: A Paradoxical Union of Contradictory Principles?, Political Theory 29 (2001) 766-81.

• 23 This is further explored in Section V below.

• 24 See e.g., D. Grimm, n. 4 above.

• 25 Much of the literature on the constitutionalization of international law, for all its internal diversity of approach, takes this basically more optimistic line. Many representative examples of this genre can be found in two recent edited volumes; J. L. Dunoff and J.P. Trachtman (eds.) Ruling the World? Constitutionalism, International Law, and Global Governance, Cambridge: CUP 2009; J. Klabbers, A. Peters and G. Ulfstein (eds.), The Constitutionalization of International Law, Oxford: OUP 2009.

• 26 A. Sieyes, What is the Third Estate? (1789) (Eng. Trans. M. Blondel, New York, Praeger, 1963).

• 27 Of course, some part of the (controversial) argument in favour of a more ‘activist’ conception of judicial review, and against strongly originalist conceptions of the judicial role in constitutional interpretation, depends upon constitutional judges, in the absence of formal amendment, themselves assuming a quasi-amending role in response to post-foundational societal change. See e.g., S.M. Griffin, American Constitutionalism: from theory to politics, Princeton: Princeton University Press, chapter 1.

• 28 H. Lindahl, Towards an Ontology of Collective Selfhood, in: M. Loughlin and N. Walker (eds.), The Paradox of Constitutionalism, Oxford: OUP 2007, p. 9-24, at p. 10.

• 29 See e.g., Lindahl, n. 28 above at14 et seq.; see also B. Van Roermund, Sovereignty: Unpopular and Popular, in: N. Walker (ed.), Sovereignty in Transition, Oxford: Hart 2003, p. 33-54.

• 30 See e.g., H. Lindahl, Acquiring a Community, European Law Journal 9 (2003), p. 433-50.

• 31 See e.g. S. Chambers, Democracy, Popular Sovereignty and Constitutional Legitimacy, Constellations 11 (2004), 153.

• 32 For a wide-ranging study of the varied constitutional treatment of democratic membership at both national and supranational level in the European Union, see J. Shaw, The Transformation of Citizenship in the European Union, Cambridge: CUP 2007.

• 33 For a recent overview, see D. Miller, Democracy’s Domain, Philosophy and Public Affairs 37 (2009), p. 201-228.

• 34 M. Canovan, Nationhood and Political Theory, Cheltenham: Edward Elgar 1996.

• 35 See e.g., J. Habermas, Moral Consciousness and Communicative Action, Cambridge Mass.: MIT Press 1995; R. Goodin, Enfranchising All Affected Interests and Its Alternatives, Philosophy and Public Affairs 35 (2007), 40-68; T. MacDonald, Global Stakeholder Democracy, Oxford: OUP 2008.

• 36 Miller (n. 33 above) connects these approaches to what he calls ‘L-Democracy’ (i.e., liberal) and ‘R-Democracy’ (i.e. radical) respectively, p. 204-207.

• 37 An Interparliamentary Union study of 150 democracies in 1993 revealed 83 majoritarian electoral system and 57 PR systems, with the remainder mixed. Electoral Systems Worldwide: A Comparative Study, Geneva: Interparliamentary Union 1993.

• 38 See e.g., S. Choudhry (ed.), Constitutional Design for Divided Societies: integration or Accommodation?, Oxford: OUP 2008.

• 39 See e.g., R. Elgie, From Linz to Tsebelis: three waves of presidential/parliamentary studies?, Democratization, 12 (2005), p. 106-122.

• 40 See e.g., J. Waldron, Law and Disagreement, Oxford: OUP 1999, chapter 13.

• 41 See e.g., Dworkin, n. 9 above, p. 25-6. See also Waldron, n. 40 above, chapter 13.

• 42 See e.g., D. Grimm, Integration by Constitution, International Journal of Constitutional Law 3 (2005), p. 193-208.

• 43 Although as Waldron argues, this does not necessarily follow, and a robust conception of rights protection can be defended as compatible with a constitutionally unchecked system of majoritarian legislation. See n. 40 above, chapter 10 ‘Between Rights and Bills of Rights’.

• 44 For a critical appraisal of these types of arguments in the post 9/11 United States environment, see S. Holmes, The Matador’s Cape: America’s Reckless Response to Terror, Cambridge: Cambridge University Press 2007.

• 45 See e.g., P. Pettit, Depoliticizing Democracy, Ratio Juris 17 (2004), 52-65.

• 46 For a comparative analysis of the different trajectories of cultivation of communal solidarities in different constitutional models, see M. Rosenfeld, The Identity of the Constitutional Subject: Selfhood, Citizenship, Culture and Community, London and New York: Routledge 2010, esp. chapters 5-7.

• 47 D. Grimm, n. 42 above, at p. 204.

• 48 R. Bellamy and D. Castiglione, Legitimizing the Euro-‘Polity’ and its ‘Regime: The Normative Turn in EU Studies’, European Journal of Political Theory 2 (2003), 7-34.

• 49 For a representative range of contemporary thinking on globalization, see A. Jones, Globalization: Key Thinkers, Cambridge: Polity 2010.

• 50 As was spectacularly illustrated in the Greek debt crisis of the early Summer of 2010.

• 51 See e.g., C. Gearty, Can Human Rights Survive?, Cambridge: CUP 2005.

• 52 See e.g., Ruthven, n. 15 above.

• 53 See e.g., my Beyond Boundary Disputes and Basic Grids: Mapping the global disorder of normative orders, International Journal of Constitutional Law 7 (2008), 373-396.

• 54 See my Post-Constituent Constitutionalism in the EU, in: Loughlin and Walker (eds.), n. 29 above, p. 247-268.

• 55 See e.g., Shaw, n. 32 above: N. Walker, Denizenship and Deterritorialization in the EU, in: H. Lindahl (ed.), A Right to Inclusion and Exclusion?, Oxford: Hart 2009, p. 261-72.

• 56 See e.g., V. Bader, Complex Citizenship and Legitimacy in compound polities (MLPs and MLG): The EU as example, EUROSPHERE, WP 1 (2008).

• 57 See M. Maduro, The Importance of Being Called a Constitution; Constitutional Authority and the Authority of Constitutionalism, International Journal of Constitutional Law 3 (2005), 332-356.

• 58 See e.g., my Not the European Constitution, Maastricht Journal of European and Comparative Law 15 (2008), 15-21.

• 59 See my Beyond the Holistic Constitution?, in: P. Dobner and M. Loughlin (eds.), The Twilight of Constitutionalism?, Oxford: OUP 2010, p. 291-308.

• 60 A. Peters, Membership in the Global Constitutional Community, in: J. Klabbers, A. Peters and G. Ulfstein (eds.), The Constitutionalization of International Law, Oxford: OUP 2009.

• 61 For critical perspectives on the spread of international constitutional discourse, see M. Koskenniemi, The Fate of Public International Law: Between Technique and Politics, Modern Law Review 70 (2007), 1-30; D. Kennedy, The Mystery of Global Governance, in Dunoff and Trachtman, Ruling the World?, p. 37-68.

• 62 See Grimm, n. 4 above. See also essays by Loughlin (What is Constitutionalization?, p. 47-72) and (Grimm The Achievement of Constitutionalism and its Prospects in a Changed World, 3-22) in: Dobner and Loughlin (eds.), The Twilight of Constitutionalism?

• 63 One important sub-distinction here is between those like Anne Peters (see her Compensatory Constitutionalism; The Function and Potential of Fundamental International Norms and Structures, Leiden Journal of International Law 19 (2006), 579-610, who talk about the ‘compensatory’ quality of constitutionalism at the postnational level, so acknowledging the basic loss to constitutionalism (as well as to democracy) as a whole occasioned by the diminution of democratic constitutionalism at the national level; and those who believe that the trend in favour of the redirection of constitutionalism towards non-democratic transnational settings, while undoubtedly transformative of the role of constitutionalism, does not necessarily entail any depletion in the value of constitutionalism as a regulatory ethic (see. e.g. essays by G. Teubner, Fragmented Foundations: Societal Constitutionalism Beyond the Nation State, and by U.K. Preuss, Disconnecting Constitutions from Statehood: Is Global Constitutionalism a Viable Concept?, in: Dobner and Loughlin (eds.), The Twilight of Constitutionalism?

• 64 See e.g., the essays collected in his The Postnational Constellation, Cambridge: Polity 2001, M. Pensky trans.).

• 65 See e.g., S. Choudhry (ed.), The Migration of Constitutional Ideas, Cambridge: CUP 2006.

• 66 The delegation argument is more often implicit than explicit. Even Habermas, with his strong commitment to directly engaged forms of democracy, seems to accept a kind of delegated legitimacy for the global (if not the regional) level in his defence of the United Nations as one player in a multi-level transnational constitutional settlement; see J. Habermas, Does the Constitutionalization of International Law still have a chance?, in: J. Habermas, The Divided West, Cambridge: Polity 2006, C. Cronin trans.), p. 115-193.

• 67 One obvious alternative source of democratic pedigree is the regional polity, and this has been a central theme in the debate over the constitutional legitimacy of the EU (see references at n. 54-58 above). Less obviously, but just as significant if not more significant in the long term, a case may be made for viewing certain functionally specialist self-constituting transnational private or hybrid regimes as democratic, or at least as its functional equivalent. They may be argued to be so to the extent that they seek to give voice to those interests most affected in their narrow focal policy field against a backdrop of the proliferation of other such specialist self-constituting sites and associated ‘democratic’ opportunities and a corresponding decline in influence of holistic state polities; see e.g., Teubner, n. 63. The kind of deep transformation of the very context of democracy contemplated here takes us to the heart of the question of the relationship between the resilient moral order of modernity and its changing architecture considered in the text below.

• 68 C. Taylor, Modern Social Imaginaries, Durham: Duke University Press 2004, p. 3.

• 69 See further, R. Geuss, Public Goods, Private Goods, Princeton: Princeton University Press 2001; R. Dahl, On Political Equality, New Haven: Yale University Press 2006.

• 70 A similar double-edged relationship of interdependence and tension is set out in a series of highly suggestive recent works by Gianluigi Palombella. For him, the meta-regulatory dilemma, at once both productive and irresoluble, that is confronted and addressed by our deepest sense of national and transnational legality, lies between the particular collective ‘good’ and the universal ‘right’. According to Palombella, the Rule of Law – or the basic constitutional integrity of our overall political order – depends upon the guarantee that collective will and universal reason be held in dynamic balance, with neither ultimately prevailing over or subsuming the other. See e.g. The Rule of Law and its Core, in: G. Palombella and N. Walker (eds.), Relocating the Rule of Law, Oxford: Hart 2009, p. 17-42; The Rule of Law as an Institutional Ideal, in: L. Morlino and G. Palombella, The Rule of Law and Democracy: Internal and External Issues, Leiden: Brill 2010.

• 71 I discuss the relationship between the moral-imaginary and architectonic aspects of political modernity from the perspective of constitutionalism in greater detail in The Global Age in the Longue Durée of Constitutionalism, International Journal of Constitutional Law (forthcoming).

• 72 See discussion at n. 67 above.