With the euro crisis creating pressures for a fiscal union within the eurozone, the compatibility of a free market with political freedom in the EU has become an especially pressing issue. The four market freedoms establishing the free movement of capital, labour, services and goods define the EU’s core purposes. In many respects, these four freedoms are the archetypal ‘modern freedoms’ praised by the French political theorist Benjamin Constant in his lecture of 1819 on ‘The Liberty of the Ancients Compared with that of the Moderns’. In this article, I shall argue that Constant’s analysis of the passage from ancient to modern liberty proves instructive for understanding the contemporary dilemma of how to combine market with political freedom in the EU.Footnote ¹ Constant argued that commerce had undermined the ancient form of political liberty, that of direct participation. Certain interpreters have seen this argument as suggesting an inherent tension between ‘liberalism’ and ‘republicanism’, markets and democracy, which is exemplified by the EU (Scharpf Reference Scharpf2009: 174–8). They contend that – unchecked – the EU’s four market freedoms might undermine democratic citizenship within the member states, without being able to create anything as substantial at the European level. In their view, the result will be the erosion of the democratically constituted social rights and policies typical of these states by EU-led pressures for an unrestricted free market (Scharpf Reference Scharpf2009: 192–8). Yet, Constant maintained that modern liberty was compatible with a new kind of political liberty, representative democracy (Constant Reference Constant1819: 325–6). Likewise, some analysts have argued that the EU can adopt similar political arrangements to those found in the member states (Hix Reference Hix2008), and even that the freedoms associated with modern liberty offer a basis for establishing the political rights typical of representative democracy on post-national grounds (Habermas Reference Habermas1998: 116–18). Indeed, Article 8A of the Lisbon Treaty declares ‘The functioning of the Union shall be founded on representative democracy’. However, others believe the EU presages new forms of transnational (Benhabib Reference Benhabib2008; Kostakopoulou Reference Kostakopoulou2008), or supranational citizenship (Majone Reference Majone2001; Sabel and Zeitlin Reference Sabel and Jonathan2007) that can detach social from political rights, and the democratic ‘outputs’ of economic equity and efficiency from any democratic ‘input’. This article explores all these possibilities.

The first section outlines Constant’s classic diagnosis of the shift from ‘ancient’ to ‘modern’ liberty. It examines why he thought political liberty remained ‘indispensable’ in a commercial age, and regarded representative democracy as compatible with modern liberty. As we shall see, though, Constant believed representative democratic institutions could only function in cultural and social conditions that are not themselves the products of the civil liberties he associated with modern liberty – indeed, such liberties promoted practices that potentially undermined these conditions. The second section describes how these conditions facilitated the development of representative democracy within European nation states. These states combined modern civil and commercial liberties with the representative form of political liberty in the manner imagined by Constant. However, this combination involved three factors – national identity, a social contract among citizens, and political parties – that all modify the individualistic and private character of modern liberty. These factors are shown to be largely absent from the EU, raising the question of whether the social and cultural conditions exist for the institutions of representative democracy at the EU level to be effective. Some commentators have contended that a new, post-national basis can be found for them, others argue that new forms of transnational and supranational citizenship are emerging that go beyond representative democracy. The third section investigates these possibilities. I examine three accounts, each of which seeks to do away with one of the three factors highlighted above as necessary for representative democracy at the national level. I start with the ‘post-national’ argument that public autonomy can be seen as logically entailed by the modern liberty of private autonomy, with a constitutional patriotism flowing from individual rights replacing nationalism as a civic bond (e.g. Habermas Reference Habermas1996: Appendix 2; Eriksen Reference Eriksen2009: 125–6). I then turn to the ‘transnational’ view that ‘freedom of movement’ has de-territorialised certain social rights, and decoupled them from the reciprocal bonds of economic and political participation that ground them in the member states (e.g. Benhabib Reference Benhabib2004: ch. 4; 2008; Kostakopoulou Reference Kostakopoulou2008). Finally, I explore the technocratic defence of ‘supranational’ forms of non-majoritarian governance for certain key EU functions, with party competition replaced by selective consultation with experts and civil society groups (Majone Reference Majone and Majone1996, Reference Majone2001; Sabel and Zeitlin Reference Sabel and Jonathan2007). I argue all these schemes overlook certain key features of modern liberty that Constant identified as making political liberty difficult to achieve, yet as necessary as ever.

From ancient to modern liberty

Constant’s account distinguishes between different conceptions of liberty, the formal entitlements and practices that are associated with them, and the social, cultural and economic conditions needed to render these entitlements and practices plausible and effective. He equated the ancient conception of liberty with collective autonomy and the modern conception with individual autonomy (Constant Reference Constant1819: 310–12) Ancient liberty consisted of the political freedom provided by the collective participation of all citizens in ruling the polity. Although citizens enjoyed a superior civil status to non-citizens, such as slaves or women, their public duties largely extinguished their opportunities for freedom in the private sphere. Indeed, private interests were regarded as inherently partial and detracting from an attachment to the public good. The political freedom of being ruled by oneself rather than by others could only be obtained collectively and through the sacrifice of all personal freedom. Only such total involvement could ensure politics was not captured by particular interests and all citizens devoted themselves to the public good.

By contrast, modern liberty was predominantly civil rather than political, and mainly exercised in the private and social sphere. Instead of civic duty, modern liberty encouraged individuals ‘to occupy their days or hours in a way which is most compatible with their inclinations or whims’ (Constant Reference Constant1819: 311). It was fostered by individual rights to freedom, such as the civil freedoms of conscience, association, speech and movement, and above all by the freedoms of contract and property ownership. These last went hand in hand with commerce, which had undermined the small-scale, slave economies supporting ancient liberty and provided the socio-economic conditions favourable to modern liberty.

Constant welcomed this development as having expanded both the types of freedom open to people and the range of social classes who could enjoy them. He also thought the spread of modern liberty through commerce had made political oppression less likely. People had become more jealous of their private liberty and suspicious of all government rules and regulations that might inhibit it. They also looked to trade rather than war to enrich themselves, thereby reducing the capacity of rulers to embark on military adventures that increased their own wealth and power. Instead, rulers became dependent on private banks and taxpayers for their revenues, with their income likewise relying on trade and industry. Nonetheless, if ‘individual liberty’ was ‘the true modern liberty’, ‘political liberty is its guarantee’ and remained ‘indispensable’ (Constant Reference Constant1819: 323). Constant worried that in their enjoyment of their private liberties, citizens might neglect and even subvert these political guarantees (Constant Reference Constant1819: 323–4). The difficulty was that the very factors that made these guarantees necessary also encouraged their neglect and subversion.

Though Constant thought all individuals had an interest in the rights associated with modern liberty, he appreciated that not everyone necessarily had an equal interest in every one of them or in upholding them on an equal basis for all. Nor did he think a free market would inevitably harmonise each person’s pursuit of their own interest with a similar pursuit by every one else in ways that promoted the best interests of all. The particular interests of different individuals could and did clash. Following Adam Smith (Viner Reference Viner1958; Winch Reference Winch1978: 97–8), Constant saw commerce as potentially corrupting from a civic point of view, reducing sympathy for others and encouraging cupidity – dangers all too evident in the ‘monopolizing spirit’ of the mercantile system (Constant Reference Constant1815: 217–18). Certain political structures, not least an impartial legal system, were needed to secure the civil rights related to individual liberty and ensure all respected them. Political mechanisms were also necessary to resolve conflicts and solve coordination problems – such as the supply of public goods. One solution might be to charge independent administrators with the task of providing these guarantees, leaving individuals free to engage in their private pursuits. As he caustically observed, it was an offer those in authority were all too happy to make, being ‘so ready to spare us all sort of troubles, except those of obeying and paying!’ However, the Napoleonic regime had revealed the error of trusting to self-declared ‘enlightened’ despots. Tempting though this solution might appear to individuals who felt they had better things to do than engage in politics, it would be ‘folly’ to hand over political power to any group of people without being able to ensure they served the interests of the ruled rather than their own (Constant Reference Constant1819: 326).

Therefore, a modern form of political liberty had to address the same two key political tasks of ancient liberty: namely, to guard against the uncertain virtuousness of the guardians of liberty, and to gain the support of the citizenry for certain common rules and regulations. Moreover, it had to do so for the selfsame reasons as those that had motivated the ancients – the concern that politics might be captured by ‘factions’ and employed for personal gain. Even modern liberty required a degree of civic virtue to induce citizens to guard the guardians who provided the political guarantee of their freedom, and to see the necessity for such political guarantees in the first place (Constant Reference Constant1819: 327–8). Yet, this civic virtue could not be of the kind associated with ancient liberty. That version had involved the dropping of private interests for the public interest. By contrast, to be compatible with modern liberty, civic virtue and politics more generally had to be consistent with citizens regarding the furthering of their private interests as the main purpose of their freedom.

Constant contended that representative democracy provided a form of political liberty of the requisite kind. It embodied the ‘eternal rights to assent to the laws, to deliberate on our interests, to be an integral part of the social body of which we are members’ (Constant Reference Constant1819: 324) in a way attuned to the liberties of the moderns. To assent to the laws, citizens did not have to be directly involved in decision making themselves. They elected ‘hired stewards’ to do it for them, leaving them plenty of time for their private affairs. Constant saw these representatives very much as delegates and stressed the importance of their being directly elected to ensure their accountability and responsiveness to the electorate (Constant Reference Constant1815: 202, 206). It was as delegates rather than trustees that they were authorised ‘to deliberate on our interests’ on our behalf. Representative democracy involved a very different conception of the public interest to that associated with ancient liberty. Instead of being distinct from all private interests, it was the product of their aggregation and negotiation. The role of representatives was to represent the diverse private interests of their electors, and produce legislation that responded to their concerns. The general interest could only arise from ‘the negotiation that takes place between particular interests’. This required ‘the representation of all partial interests which must reach a compromise on the objects they have in common’ (Constant Reference Constant1815: 205). A representative must remain ‘partial towards his own electors’, therefore, because the ‘impartiality of all’ only resulted when ‘the partiality of each of them’ was ‘united and reconciled’ (Constant Reference Constant1815: 206). The upshot of this system was to make all citizens feel ‘an integral part of the social body’. For, each citizen could claim an equal responsibility for and stake in the laws given they had been made by their elected representatives so as to reflect their interests. In this way, ‘political liberty, by submitting to all the citizens, without exception, the care and assessment of their most sacred interests, enlarges their spirit, ennobles their thoughts, and establishes among them a kind of intellectual equality which forms the glory and power of a people’ (Constant Reference Constant1819: 327).

On Constant’s account, therefore, representative democracy retains the links found in ancient liberty between self-government, the public interest, and civic virtue, on the one side, and liberty, on the other. However, it reworks their rationale and functioning to coincide with the modern liberty of individual autonomy, with its focus on social and personal life, rather than the essentially political collective autonomy characteristic of ancient liberty. It makes politics less onerous and reconceives the public interest in terms of private interests. Yet, neither of these features will of themselves overcome the key problem of factionalism, which is likely to be an even greater danger under modern as opposed to ancient liberty. Other aspects of Constant’s institutional design address this issue to a degree. For example, he recommended a clear separation of powers, with an independent judiciary to police infringements of basic rights and a constitutional monarch with the power to dissolve parliament. But the ‘bastion’ of individual freedom remained ‘the existence of a large and independent representation’ (Constant Reference Constant1815: 289). It alone enables citizens to guard the guardians and legitimise the reciprocal modifications to their private interests needed to construct the public interest.

Constant hoped political participation itself might generate some of the civic virtue needed for these tasks. However, it is doubtful it can do so if citizens lack any disposition towards public spiritedness in the first place. If individuals are simply concerned to promote their own interests as much as possible, then they will remain tempted to free ride on the civic virtue of others, and to devote their own political energies to rent seeking. Fostering mutual dependence through federal arrangements that allowed each local faction to block the self-interested behaviour of other factions, as he also recommended, will be insufficient to promote their common interests efficiently and equitably. It simply invites the universalising of factionalism. It will either produce deadlock or generalised rent-seeking by representatives attempting to buy the support of their followers, neither of which is likely to be optimal or just unless the parties are equal to start with. Even then prisoner’s dilemmas and other quandaries of rational choice will arise. To avoid these difficulties, individuals need to be disposed to view the exercise of their own private rights in ways that take into account and accommodate their similar exercise by others.

The dilemma had been appreciated as endemic to commercial republics by Montesquieu and Smith, on whom Constant drew (Winch Reference Winch1978: 97–9), as well as the authors of The Federalist (Elkin Reference Elkin2006). They too had sought to devise political institutions that economised on virtue, while recognising that in politics as opposed to markets private vice rarely generates public benefits. However, like other French liberals, Constant was more attentive to the role cultural and social conditions play in fostering appropriate political attitudes than the Anglo-American tradition has tended to be (Siedentop Reference Siedentop and Ryan1979). He had witnessed the failure not just of the French Revolution’s attempt to reinvigorate ancient liberty, but also of the Napoleonic attempt to establish an Empire of modern liberty through the Code Napoléon against the rise of nationalism.

Two such conditions underpin Constant’s analysis of representative democracy. First, he noted how a shared political culture fosters allegiance both to political institutions and one’s fellow citizens. As he remarked, ‘the natural source of patriotism’ was found in ‘a vivid attachment to the interests, the ways of life, the customs of some locality’ (Constant Reference Constant1813: 74). As Mill later argued (Mill Reference Mill and Acton1861: ch. 16), drawing on Constant and other French liberals (Siedentop Reference Siedentop and Ryan1979: 172–4), such national sentiments make representative democracy possible by facilitating public debate in ways that reduce factionalism (Miller Reference Miller2009: 208–13). For a start, it will be easier to have a discussion among the public as a whole if there are shared cultural instruments, such as a common media – newspapers and, nowadays, television and radio programmes – that address and are accessible by all, not least because they are in a common language all can understand. Such instruments help different sections of society to inform and respond to each other. It becomes harder for governments to play them off against one another and to pander to sectional rather than public opinions commanding broad support. A political culture also supplies shared values that provide citizens with a basis for debating matters of collective concern. For citizens and their representatives to feel obliged to respond to and accommodate the views and interests of others, they must consider the arguments their interlocutors raise are more than private opinions and concerns. A common stock of values, that all recognise as important for the political community, aids discussion and compromise. It provides agreement on the sorts of points that can be raised and need to be addressed and responded to, even if there is disagreement about their relative importance and the most appropriate response. In their absence, people will just talk past each other. The sharing of values and language can also help build trust among the electorate by highlighting common sympathies and priorities. Politics can be less about gaining advantages for one’s own group or oppressing rival groups, and more about what set of policies best balances the different concerns of individuals so as to achieve the most satisfactory outcome for all.

Second, similar issues arise if there is a lack of interdependence of interests and individuals do not possess a roughly equal stake in the collective decisions affecting them (Christiano Reference Christiano2006). These conditions supply both an equal right, as a matter of fairness, for all members of the political community to have an equal say in how it is run, and stimulate the sort of cross-cutting cleavages whereby people who find themselves opposed on one issue are on the same side on others. Both conditions reduce the prospect of majority tyranny. Though people will have a greater interest in some issues than others, everyone involved in politics will have a more or less equal interest in the totality of collective decisions. They will be able to compromise by trading votes between the issues that matter to them and those that they regard to be less important, and will be less likely to be consistent winners or losers as a result. This reasoning underpinned the traditional limitation of the vote to property owners, a view Constant fully endorsed. He distinguished landed from ‘industrial’ property in this respect. Only the former ‘binds man to the country where he lives, surrounds his departure from it with obstacles, creates patriotism through interest’ (Constant Reference Constant1815: 218). Land alone gave each member of the association ‘a common interest with the other members of the association’ (Constant Reference Constant1815: 214). It signified involvement in common cultural, social and economic structures, and a commitment to their future efficient and equitable functioning. The moveable ‘industrial’ property of labour and trade was less affected by such structures, and its possessors had less of a common interest in supporting them and ensuring they operated in a fair way.

Given the topic of this article, it is significant that Constant regarded the absence of these two conditions as a chief failing of the Napoleonic Empire. (Fontana Reference Fontana and Pagden2002: 126–7). This project had endeavoured to unite Europe around the benefits of a uniform model of good governance encapsulated in the Code Napoléon. ‘The same code of law, the same measures, the same regulations, and if they could contrive it gradually, the same language, this is what is proclaimed to be the perfect form of social organisation’ (Constant Reference Constant1813: 73). However, ‘a fictitious passion for an abstract being, a general idea stripped of all that can engage the imagination and speak to the memory’ could not replace the ‘genuine patriotism’ that springs from ‘local customs’ (Constant Reference Constant1813: 73–4). Uniform laws also overlooked local diversity. In this respect ‘large states have great disadvantages. Laws proceed from a place so remote from those places where they must be applied, that frequent and serious errors are the inevitable result’ (Constant Reference Constant1813: 77). Good governance needed more than a uniform imposition of the rules and rights of private interest. If laws were to be sensitively and impartially applied and adhered to, in ways that reinforced public goods while being sensitive to local differences, then institutions of political liberty that could draw on shared cultural norms and common interests were necessary.

Though Constant never mentions The Federalist, one can assume that he would have regarded its proposals as incompatible with these two preconditions for representative politics as the Napoleonic Empire. A large size might aid the checks and balances needed for the negative task of blocking factions, as Federalist 10 famously maintained, but – at least within Europe – he would have regarded such a solution as unavailable. Europe was too diverse for the common culture and interests he believed were needed for the positive task of constructing and generating allegiance to a shared public interest. In these respects, his conception of modern political liberty is more ‘republican’, more concerned to preserve those qualities of ancient liberty that ‘achieve the moral education of citizens’ (Constant Reference Constant1819: 328), than The Federalist – hence his preference for an association of European nation states.Footnote ² Yet, he feared this possibility was endangered not just by Napoleon but also by the corrosive effects of commerce, which was turning the European peoples into ‘a great mass of human beings, that … despite the different names under which they live and their different forms of social organization, are essentially homogeneous in their nature’ (Constant Reference Constant1813: 52–3). If the ‘natural’ source of civic pride was local, the danger was such a mass would be little more than an agglomeration of self-seeking private individuals.

The various dilemmas Constant diagnosed as bedevilling the combination of modern liberty and democracy are all too evident in the current euro crisis. On the one hand, the financial crisis and the high levels of sovereign debt that have resulted from it reveal both the need for market regulation and the tendency not to do so when a merchant class with highly moveable property and little stake in the polity capture governments (Mair Reference Mair2011). If economies are not socially and politically embedded, they are prone to various forms of market failure (Polanyi Reference Polanyi1944). On the other hand, the difficulties of getting agreement on a suitable rescue package among the eurozone members follows from the type of democratic politics that Constant associated with modern liberty, in which agreement requires Pareto improvements for all the parties involved. This reasoning will only provide a rationale for Germany and the other solvent states to guarantee the sovereign debt of those states likely to default so long as it is compatible with enlightened national self-interest as the solution to an assurance game. However, if the appeal begins to shift so that it is less to enlightened national self-interest and instead to the collective self-interest of the Union as a whole – that is, if the sacrifices called for from either the debtor or the creditor states rise beyond a certain threshold and look to be uncompensated in the medium or even the long term – then cooperation will weaken (Scharpf Reference Scharpf1997: ch. 6) In the view of many national politicians and their electorates, it would appear that we are perilously close to this situation – hence the tentativeness of the proposals being made to resolve the situation. The EU has traditionally sought to overcome such problems by imposing a non-political solution via the Commission or the European Court of Justice, or in this case the European Central Bank. However, such solutions are also only likely to prove stable and acceptable in the case of a symmetrical Prisoner’s Dilemma, in which uniform rules and cooperation will benefit all and the problem is to avoid free riding (Scharpf Reference Scharpf2009: 183–5, 189–90). A key difficulty with the euro, though, has been that the underlying constellation of interests in this case does not conform to this pattern. It is not just that certain countries failed to abide by the conditions of the Stability and Growth Pact and keep public spending under control, but also that the economies of the participating states have proven too diverse. In other words, Constant’s critique of the Napoleonic Empire’s technocratic imposition of uniform norms and regulations apply with particular force here. The problems currently confronting the EU are increasingly characterised by asymmetric conflicts of interest. In these cases, an appeal to mutual national self-interest ceases to be credible and must be to the collective self-interest of the Union as a whole – as has occurred in current calls to ‘save the euro’ and with it ‘Europe’.Footnote ³ Yet, to construct a European public interest in these circumstances requires a pan-European democratic politics capable of overcoming national self-interest. At this point, the issues raised by Constant’s analysis of modern liberty become highly relevant. As I noted, the EU upholds the civil liberties necessary for a free market. What remains to be seen is if sufficient cultural and social conditions exist at the European level to imbue those who enjoy this pan-European modern liberty with sufficient elements of the civic spirit of the ancients for a political union based on the principle of representative democracy to be possible.

The modern union of political and commercial liberty

Contrary to Constant’s expectations, commerce did not end military conflict (Wenar and Milanovic Reference Wenar and Milanovic2009). Representative democracy was only firmly established in Europe after two world wars and a third, cold, war. Nevertheless, the liberal democratic states that gradually emerged from the nineteenth century onwards could be said to involve precisely the marriage he anticipated between political and market freedom. Moreover, he has been proved justified in believing that there could be no liberal (civil and commercial liberty) regimes that were not also in some meaningful sense democratic (political liberty) regimes too, and vice versa. Yet, this modern form of political liberty depended on the two factors Constant had identified as necessary, a common culture and common interests, being preserved in a form he had thought potentially incompatible with commercial liberty: namely, nationalism and a social contract between unpropertied labourers and those possessing industrial property, plus a third factor, political parties, that he had not anticipated. These respectively generated the common values and idiom, the alleviation of conflicts of interest, and the creation of ideological rather than factional political associations, that were needed to engender enough elements of the public spiritedness and social solidarity typical of the citizenry of the ancient polity for the less demanding, yet more inclusive, representative democracies to work in ways that promoted rather than subverted the public goods appropriate to a modern system of liberty.

As we saw, Constant feared modern liberty was dissolving cultural differences and creating common tastes as well as economic bonds that were transnational in character (Constant Reference Constant1813: 52–3). Yet, a desire for similar consumer products does not seem to have eroded national sentiments. Indeed, to some degree the development of unified markets for production and exchange promoted nationalism as local practices were replaced by a common economic and cultural system tied to a single state (Gellner Reference Gellner1983). These economically driven processes of nation building fed into the development of a national political demos (Rokkan Reference Rokkan and Tilly1974). The resulting national cultures possess many of the hallmarks of ancient liberty in being created by the state, often through compulsory education programmes, and stressing a common public bond of a civic kind that is superior to, and partly shapes, an individual’s private preferences. They also ensure a shared language or languages necessary for a common public sphere. Meanwhile, as he noted, size remains important. The representative system can only be stretched so far. Citizens not only rightly feel their vote counts for less if a political community gets too big (Dahl and Tufte Reference Dahl and Tufte1973: 13ff), but also larger communities tend to be more diverse – culturally, socially and economically – making it less likely all have an equal stake in the issues on which they have an equal say, and increasing the chances of persistent minorities and hence of majority tyranny (Christiano Reference Christiano2006).

Second, in addition to cultural bonds, a community of interest among citizens has been buttressed by a social contract promoting a degree of reciprocity in economic relations (Offe Reference Offe, Greven and Pauly2000: 67–8). Market rights have been supplemented and constrained by social rights in exchange for a willingness to work and pay taxes. Though Constant followed Smith in regarding commerce as socially beneficial, like Smith he appreciated its operations might be attended by conflicts generated by the large inequalities it promotes in patterns of ownership and income (Winch Reference Winch1978: 98–9; Marshall Reference Marshall1950; Barbalet Reference Barbalet1988). By basing the franchise on ‘landed’ property alone, he had attempted to exclude the conflict between capital and labour from politics, and prevent commercial interests capturing the political process (Constant Reference Constant1815: 217). In a commercial world, however, insistence on a landed property qualification became increasingly untenable. Meanwhile, mass mobilization in two world wars undermined his contention that the ‘patriotism’ required ‘to die for one’s country’ involved a lesser commitment to, and understanding of, its interests than possession of land (Constant Reference Constant1815: 214). Instead, they prompted the enfranchisement of the unpropertied, with public systems of social assurance and education offering an alternative to propertied wealth for ensuring a citizen could act and think independently without being dependent on particular private interests (Mann Reference Mann1987). In the process, class conflict was attenuated sufficiently to be containable within the democratic system. However, in return, participation in the economy, at the very least through being available for work if able to do so, became both a legal requirement and an expectation of one’s fellow citizens for those seeking the full social and political benefits of citizenship. This expectation that a universal entitlement to social welfare will be reciprocated by everyone’s doing their bit to contribute to the welfare of others when they can, obtains support in its turn from citizens feeling they belong to a national political community (Miller Reference Miller1995: 83, 93). Again a degree of commonality has helped both to create a social bond and increase confidence that one’s fellows will act justly by you if you act justly towards them (Galston Reference Galston1991: 215–24).

Finally, Constant shared his contemporaries’ distrust of parties (e.g. Hume 1741: Essay 8). However, for a mass electorate, the process of combining the disparate views of millions of citizens and bringing them to bear on representatives, while avoiding both factionalism and clientalism, has depended on the development of political parties defined by ideology or programmes rather than patronage networks or the narrow interests of their members, and obliged regularly to compete for power in free and open elections (Lipset and Rokkan Reference Lipset, Rokkan, Lipset and Rokkan1967). As early analysts of modern democracies noted (Bryce Reference Bryce1921: 119; Schattschneider Reference Schattschneider1942: 1), parties play an ineliminable role in the ‘modern’ form of political liberty by channelling the pursuit of private interests in a more public direction, and making political participation cost-effective in terms of time and effort. Electoral competition forces parties to construct coalitions of different interests and unite them behind a programme of government to obtain a majority. As a result, different private interests are brought to accommodate each other and seek common ground, and so come, in part at least, to shape their demands in terms of a broader and more public interest. Parties also economise on the time citizens have to give to informing themselves about the merits and failings of their potential and actual rulers and the views of their representatives and fellow citizens. Mutual criticism by rival parties highlights electorally salient information, while party discipline controls and vets representatives.

However, party competition only tends to work well when those involved are not additionally divided by ethnic, religious, linguistic and cultural divisions or overly polarised by class conflicts: conditions provided by a common nationality and social rights. For these factors prevent politics becoming zero-sum and allow ideologically or programme based parties to unite very diverse groups around a number of mutually intersecting concerns that cut across cultural and class cleavages. At the same time, as Constant noted, political participation helps give citizens a sense of responsibility for and control over these policies, with electoral pressures serving to shape national political culture and the social system. Our confidence that the laws treat those subject to them in an equitable manner is strengthened through their being open to contestation through fair political processes in which each citizen’s vote is treated with a reasonable degree of equal concern and respect.

Therefore, though contemporary representative democracies are liberal-democratic, with the ancient liberty of direct collective political participation transformed by the modern liberties of freedom of choice, the exercise of autonomous judgement about how to lead one’s life, and the separation of public and private, their practices are shaped by cultural bonds Constant feared might be destined to disappear, social bonds he had not imagined as possible, and political mechanisms of a kind he had thought pernicious. All three factors serve to constrain the operation of modern liberty in various ways so as to render it compatible with a commitment to collective decision-making. Part of the difficulties with establishing any degree of political liberty within the EU arises from the fact that the continued unfolding of modern liberty appears to have done little to abate the importance of the first factor, while potentially making the second and third factors increasingly problematic.

Far from national and cultural differences decreasing, they have become ever more significant. Thus, multinational states, such as Britain, Belgium and Spain, have begun to fragment along linguistic, religious and ethnic lines, and been subject to increasing calls for self-government on the part of territorially concentrated minority groups and, in certain cases, even secession (Kymlicka Reference Kymlicka2001: 212–13). Cultural criteria have if anything increased in importance for those seeking access to citizenship from outside, with many states enfranchising non-resident co-nationals while remaining reluctant to grant full citizen rights to resident aliens (Joppke Reference Joppke, Aleinikoff and Klusmeyer2001). By contrast, welfare settlements have been under pressure since the 1980s from governments influenced by New Right thinking, with global markets often invoked as having helped promote such neoliberal policies. Despite some modest retrenchments, though, social rights have remained remarkably robust in the face of such onslaughts and there is little evidence that globalisation has forced a reduction in welfare spending in order to maintain international competitiveness, though some restructuring has taken place (Swank Reference Swank2002: 276). However, what remains strong is the view that welfare forms part of a contract between citizens which involves duties as well as rights. Whether justified or not, citizens have demanded governments pursue policies that guard against putative welfare ‘scroungers’ and have been sensitive to ‘economic’ immigration if that is felt to detract from the employment opportunities available to existing citizens or to place additional burdens on social services such as housing, hospitals and schools, without any compensating gain in tax revenue towards their maintenance and improvement (White Reference White2003: ch. 1). Meanwhile, all advanced democracies are witnessing a slow but steady decline in electoral turnout, along with a shift towards more focussed – and in certain respects more privatised and factional – forms of political participation, as party membership has declined even more rapidly than voting (Hay Reference Hay2007: 12–16). Citizens appear to see politics increasingly through the lens of commercial and civil liberty. It is the mechanism through which private interests are pursued and individual rights upheld. There has been a commensurate rise in consumer groups and single issue movements, particularly in areas such as consumer rights, and the increased resort to the law by those with the resources to do so (Pattie, Seyd, and Whiteley Reference Pattie, Seyd and Whiteley2004: 275–80). Contemporaneously, there has been a move towards the sort of depoliticisation Constant feared as citizens have come to distrust politicians and the political process. Ever more areas of public life have been handed over to ‘expert’ regulatory bodies of one kind or another that claim to govern on the basis of the ‘public interest’, yet with few if any mechanisms for ensuring accountability to the public (Hay Reference Hay2007: 91–5).

All three of these developments pose a challenge for the development of a system of representative democracy at the level of the EU. If national sentiments remain strong for defining political membership and the boundaries of the political community, then how can the EU compete with such allegiances? Likewise, if social rights are rooted within national systems of welfare and solidarity, how can they be disembedded without potentially further weakening the bonds of reciprocity among citizens that sustain them? Finally, without the support of pan-European cultural or social bonds, how can a European party system develop and politics avoid becoming the preserve of myriad pressure groups and depoliticised administrative bodies?

Unsurprisingly, the EU performs poorly on all three of the factors that have made a modern form of political liberty possible in the member states (Bellamy Reference Bellamy2008). With regard to the first factor of cultural identification, Eurobarometer surveys consistently indicate that less than 10% of EU citizens have a strong sense of EU identity, with only 50% feeling even a weak attachment – and that strongly secondary to their local and national ties.Footnote ⁴ Likewise, so far as the second factor of social rights goes, opinion polls also show little support for the EU taking responsibility for welfare. Issues relating to socio-economic rights, in so far as they involve health, welfare and education, all have a low level of legitimacy as EU competences, with 65% or more of European citizens regarding these as exclusively national responsibilities. Finally, on the third factor of political parties, these exist at the European level purely as groupings of national parties within the European Parliament, with the take-up of EU-level political rights at a lower level and declining faster than within the member states. Average turnout in elections to the European Parliament runs at below 50% and in many countries is as low as 25%, with each increase in parliamentary power being accompanied by a decline in turnout in European elections, which continue to be fought largely on domestic issues (Hix and Marsh Reference Hix and Marsh2011). Among the 12 million EU citizens resident in another member state turnout is even lower, with the proportion of non-national EU citizens even registering to vote ranging from a mere 9% in Greece and Portugal to 54.2% in Austria.

It might be argued that just because these factors do not exist now does not mean they could never exist in the future. However, their emergence within the EU is pre-empted by their presence in the member states, while the processes that promoted them there, not least war, are unlikely to be repeated (Offe Reference Offe, Katenhusen and Laming2003: 73-4; Miller Reference Miller, Laborde and Maynor2008: 145-6). For example, there is considerably more diversity on Constant’s two elements of a shared political culture and common interests in Europe than in the United States (Baldwin Reference Baldwin2009). Views on abortion policy are often taken as a proxy for religious and moral values more generally, with the United States notoriously divided on the issue. However, if one compares Swedish policy, where abortion is available on request and on average there are over 17 abortions a year per 1000 women, with the far more restricted Irish policy, which allows abortion only to save the life of the mother and has on average only 6 abortions a year per 1000 women, then the division is as great if not greater. Moreover, despite the spread of English as the lingua franca of the educated European classes, a European media has failed to develop even among this class. Possibly the only newspapers that enjoy a pan-European readership are the mildly Euro-sceptic Financial Times and the US Herald Tribune. Meanwhile, social differences are similarly wide, with the gap between the EU country with the lowest per capita income (Lithuania) and that with the highest (Netherlands) two to three times that between the poorest US state (Mississippi) and the richest (Delaware).Footnote ⁵ Thus, in the short to medium term at least, it appears doubtful that within the EU modern liberty can be linked to political liberty on the same basis as in the member states, as has often been assumed would be necessary for the EU itself to become a ‘representative democracy’ as Lisbon directs (Hoffman Reference Hoffman1966: 868). The differences on the crucial dimensions are simply too great and have deepened rather than diminished over the past 15 years (Shore Reference Shore2004), despite the dramatic increase in EU competencies over this same period. However, others have argued that the novelty of the EU lies in promoting new types of citizenship that do not rely on these three factors but are based on the civil liberties of the moderns alone. It is to the plausibility of such schemes that I now turn.

EU citizenship as the liberty of the moderns

EU citizenship can be seen as the archetype of a purely modern conception of political liberty. Though only citizenship of a member state gives access to EU citizenship, it does not itself relate directly to any of the three factors that led to the establishment of liberal democracy within nation states. Instead, citizenship of the Union stems from the four quintessentially modern commercial liberties that lie at the heart of the EU – namely, the free movement of labour, capital, goods and services. At an early stage, these four commercial freedoms became associated with a broader set of civil rights linked to a right to equal treatment and the absence of discrimination on grounds of nationality (Article 12 EC) or ‘sex, racial or ethnic origin, religion or belief, disability, age, or sexual orientation’ (Article 13 EC). Although initially tied to economic issues such as employment and pay, successive judgments of the European Court of Justice (ECJ) came to interpret the commitment to equality as an implicit component of any legal system that seeks to take individual rights seriously. Following the establishment of the status of Union citizenship in 1992 with the Maastricht Treaty, the Court has gradually come to read the four freedoms through the lens of Article 18 EC, giving every Union citizen the right to move and reside freely in the territory of other member states. By 2001, the Court felt bold enough in one such case – Grzelcyk, involving students studying in a different member state to their own – to adopt a rhetorical formula it has regularly employed ever since: namely, that ‘Union citizenship is destined to be the fundamental status of nationals of the member states, enabling those who find themselves in the same situation to enjoy the same treatment in law irrespective of their nationality, subject to the exceptions as are expressly provided for’.Footnote ⁶ A number of commentators have argued in similar terms, welcoming this new status as signalling the move towards a new form of citizenship beyond the nation state (e.g. Kostakopoulou Reference Kostakopoulou2008).

In what follows, I shall explore the plausibility of this attempt to build the political and social liberties associated with citizenship on modern commercial and civil liberty alone. I shall examine three contrasting accounts, each of which seeks to minimise the need for one of the three factors that made the modern form of political liberty possible within the member states. Each will be shown to suffer from one or more of the pathologies of modern liberty feared by Constant. The first account, provided by Habermas, attempts to create a post-national identification with the EU through a constitutional patriotism to EU-wide rights. Yet, his argument ignores both the local roots of civic patriotism noted by Constant and the tensions between private and public autonomy that Constant showed lay behind the undermining of ancient by modern liberty. The second, transnational account, offered by Seyla Benhabib, reflects on how EU citizenship has come to ‘disaggregate’ citizenship rights and to ‘deterritorialise’ certain social rights (Benhabib Reference Benhabib2004: 153–5; 2008: 46–7). However, her enthusiasm for this development overlooks how it generates a commercialisation of citizenship that undermines the social contract at the national level that gives rise to these rights in the first place. The third, supranational technocratic account of Majone (Reference Majone2001), suggests the benefits of political liberty can be provided by proxy and without parties – through expert regulators and selective consultation with civil society. Yet, this solution offers an updated version of the Napoleonic version of good governance that Constant recognised as more likely to subvert than strengthen liberty.

Habermas contends EU legitimacy requires the development of a post-national form of Union citizenship that derives from identification with European-level rights (1996: Appendix 2), and supported the proposed Constitutional Treaty as a necessary means to achieve this result (Habermas Reference Habermas2001a). To be valid, he believes international law must conform to the ‘democratic principle’ by incorporating the preconditions for political accountability within it. He sees this incorporation arising through the civil and commercial liberties of the moderns providing the new foundations for the liberties of the ancients. Indeed, his general theory can be seen as an attempt to unite ancient and modern liberty, republicanism and liberalism (see Habermas Reference Habermas1998: 68–9; 2001b: 116–18, where he uses these terms, and 1996: 99–104). On the one hand, he argues that private autonomy requires social as well as the standard civil rights for its exercise. On the other hand, he contends that these self same rights are the basis for, and can only be legitimised through, democratic processes. In this way, civil and civic liberties go hand in hand as mutually entailing each other. At the same time, rights can thereby offer an alternative basis for democratic citizenship to membership of a national political community, making possible its extension to the European and potentially the global level (Habermas Reference Habermas2001b: 98–103).

There are a number of problems with this thesis, some prefigured in Constant’s analysis of modern liberty. Both logically and empirically his linking of private and public autonomy is too neat (Christiano Reference Christiano1996: ch. 1; Weale Reference Weale2007: 106–15; Bellamy Reference Bellamy2007: 210–12). As Constant noted, there is at the very least a tension between the time and effort that has to be devoted to politics and the pursuit of one’s private activities (Constant Reference Constant1819: 316–17; Weale Reference Weale2007: 108). True, it might with some justification be argued that in the circumstances of social life private autonomy depends on public regulation. However, this raises the problem that the private autonomy of different citizens may often clash, as may their views as to the appropriate public rules and goods needed to uphold and foster it. Such conflicts reflect their differing preferences and moral values, and the general difficulties that attend identifying and agreeing on what count as good outcomes and the best ways to secure them given the complexities and openness of most social processes. R. P. Wolff (Reference Wolff1970) showed many years ago how to portray public policies and laws as expressions of the autonomous will of the people as a whole requires either an implausible account of collective agency or an ethical naturalist account of ‘real freedom’ of the kind objected to by liberal critics of ‘positive’ freedom (Berlin, Reference Berlin1969: 129–31). Indeed, many theorists who adopt this approach have a tendency to compile such extensive lists of the rights and policies needed to secure the preconditions of democratic autonomy that one wonders what would remain for citizens actually to decide democratically (e.g. Held Reference Held1995: 153–6, 190–201). Such accounts seem entirely circular – they obtain a spurious democratic legitimacy for their preferred list of rights by so defining democracy that it inherently involves them, so that any democratic consideration of their normative importance and practical implications becomes at best unnecessary at worst self-contradictory.

These conceptual problems become all the more manifest when one considers how the functional and cultural diversity of modern societies multiplies the various spheres of life, each with their different guiding values and priorities, and the plurality of moral codes and valuations of different individuals and groups of people operating within and between them. These processes are themselves the result of modern liberty, yet they increase the potential for tensions and conflicts between the diverse activities of citizens and make convergence on the preconditions for private and public autonomy even less likely. To these difficulties need to be added those linked with the very territorial extent of the proposed post-national political communities. As we saw, Constant noted how size matters (Constant Reference Constant1813: 76–7), diminishing both the impact any citizen feels he or she may make on collective decision-making and the identification they may have with their fellow citizens (see too Miller Reference Miller2009: 212–13).

What leads a given group of people to coalesce around a particular constitutional settlement would appear to be less its intrinsic merits and more a pre-existing belonging to the polity and people to which it applies. To quote Constant again (1813: 73–4), patriotism does not issue from ‘a fictitious passion for an abstract being’ but ‘a vivid attachment to the interests, way of life, the customs of some locality’ (see too Shore Reference Shore2004). After all, every member state already has its own constitutional settlement which embodies liberal democratic values. However, that has not diminished the demand for enhanced self-government or even secession by national minorities within these states. Why, then, should one expect the existence of an EU-level constitution to enhance support for the EU? In fact, only 27% of citizens say an EU constitution per se would strengthen European citizenship. Indeed, only 32% of Europeans know their rights as citizens and only 43% claim to understand what the term ‘citizen of the EU’ means.Footnote ⁷ All the peoples of Europe might value rights, but they have different valuations of them. Within the member states their disagreements in these respects are settled through democratic and judicial decisions that reflect a national political culture. However, we saw that take-up of EU political rights are limited, as is identification with EU-level institutions – a circumstance that has led the German Federal Constitutional Court for one to question the democratic legitimacy of the ECJ and EU law to claim constitutional competence over domestic understandings of constitutional rights.Footnote ⁸

Nevertheless, Habermas is right to fear the disembedding of rights from a functioning democratic system that can mesh their private exercise with the public goods that provides their rationale (see Raz Reference Raz1994: ch 3 for this collective dimension of rights). The dangers attending such a development emerge clearly from a consideration of the ‘disaggregation’ and ‘deterriorialisation’ of rights Benhabib celebrates. As initially designed, Union citizenship was supposed to be proof against this arising. Not only is it restricted to citizens of a member state but also, the ECJ’s rhetoric notwithstanding , the Treaty insists it must ‘complement and not replace national citizenship’ (Article 17 (1) EC). Meanwhile, with the exception of the right to vote in elections to the European Parliament, the EU itself does not provide citizens with goods or services through EU funds or agencies. Rather, what EU citizenship offers is access on a par with national citizens to engage in economic activity with, and enjoy the services and benefits provided by, another member state. It is only activated through a citizen moving to, or trading with, another member state through the exercise of the four freedoms. So, Union citizenship does not offer a form of dual citizenship with the EU per se. Rather, it allows EU citizens to pursue their commercial liberties on a par with nationals of another member state to their own. Moreover, certain ‘limitations and conditions’ were instituted to protect various non-market liberties associated with national citizenship. Thus, the 1990 Residence Directives, later repealed and incorporated into Article 7 (1)b and c of Directive 2004/38, together with certain provisos of what are now Articles 39, 43 and 49 EC, restrict the right of residence to those engaging in economic activity or possessing adequate funds not to become a burden on the national system of social assurance and covered by sickness insurance. The four freedoms also do not apply in areas that are ‘wholly internal’ or involve restrictions based on public policy, security and health (Uecker Footnote ⁹). Recently, though, these limitations have been implicitly and explicitly challenged by ECJ rulings to the effect that Union citizenship offers a Treaty-based, directly effective right of its own. It is these decisions that have fleshed out a form of market citizenship at the EU level that potentially conflicts with political and social citizenship at the member state level (Everson Reference Everson, Shaw and More1995).

In a series of cases, the Court has increasingly argued that the restrictions protecting national citizenship have to be applied in a ‘proportional’ manner (Baumbast Footnote ¹⁰) that do not deprive Union citizens of a right to move and reside that exists independently of their pursuit of any economic activity (Chen Footnote ¹¹), thereby creating new rights for non-workers (Sala,Footnote ¹²Trojani Footnote ¹³), students (Grzelczk) and job-seekers (Collins Footnote ¹⁴), weakening public interest derogations that excluded non-nationals from certain public service jobs (Marina Mercante Espanola Footnote ¹⁵), and altering what could be considered a ‘wholly internal’ matter (Avello,Footnote ¹⁶Chen, Rottmann, Footnote ¹⁷Zambrano,Footnote ¹⁸ although see McCarthy Footnote ¹⁹ which arguably reasserts the internal rule). In a parallel move, the Court has also questioned the previous understanding that the state provision of healthcare and education are not ‘services’ in the commercial sense of Articles 49, 50 EC, but legitimately correspond to the democratically decided collective preferences of the citizens of each of the member states, reflecting national financial priorities and other public interest considerations (e.g. Commission v. Austria, Footnote ²⁰Humbel Footnote ²¹). As such, these services had not been subject to the prohibition on restrictions of the freedom to provide services. However, decisions such as Schwarz,Footnote ²²Kohll,Footnote ²³Geraets-Smits,Footnote ²⁴Müller-Fauré,Footnote ²⁵ and Watts Footnote ²⁶ have undermined this reasoning by allowing individuals to escape national processes of rationing these goods by shopping for alternatives elsewhere in the Union. Grzelczk held that ‘a certain degree of financial solidarity’ now existed between the member states. Yet, though the euro crisis suggests that such solidarity is decidedly limited, the Court has consistently refused to treat national fiscal concerns as posing restrictions on the exercise of European liberties – even treating national rules against tax avoidance as violations of free capital movement. Finally, there have been a series of judgments that have prioritised EU-level economic freedoms over member state-level social rights (Viking,Footnote ²⁷Laval,Footnote ²⁸Rüffert Footnote ²⁹ and Luxembourg Footnote ³⁰). In these cases, the Court has attempted to impose a uniform, minimum standard of wage legislation that overrides local collective bargaining agreements, thereby hindering the exercise of union rights.

In various ways, these decisions uncouple the rights of individuals freely to pursue their personal goals and interests on an equal basis to others either from economic participation within and a contribution to, or membership of and identification with, the polity in which one resides (Scharpf Reference Scharpf2009: 191–8). Consequently, many citizenship rights, including access to important social and economic benefits, have been disassociated not just from political citizenship, but also from what we have seen have become the standard prerequisites for obtaining the same: namely, an economic stake in the fortunes of the state, membership and a degree of identification with it, and political participation in shaping and sustaining the goods that it provides its citizens. It is this process that has produced what Benhabib calls the disaggregation of citizenship (Benhabib Reference Benhabib2008: 46–7), whereby the synthesis of civic with commercial and civil liberties achieved within the nation state has been pulled apart as the latter have become detached from the three factors we identified as linking them to the former. Instead, modern commercial liberties have become the trigger of themselves for access to certain civic liberties: notably, the ability to vote and stand in local and European elections when residing in another member state, and admission to social benefits that hitherto have been both privileges of political citizenship and part of their foundation.

Although an advocate of ‘another cosmopolitanism’, Benhabib’s transnational position shares the general cosmopolitan endorsement of such moves as following their critique of the moral arbitrariness of borders, and the exclusionary nature of state-centred citizenship (Carens Reference Carens1987; Nussbaum Reference Nussbaum and Cohen1996). Even though many, if not all, these rights apply only to EU nationals rather than all non-citizens resident within a member state, and to that extent are unsatisfactory, cosmopolitans are apt to regard any deterritorializing and denationalizing of citizenship as a step in the right direction. However, there is a split within the cosmopolitan camp over what universal obligations we owe to all humans, and the mechanisms that might be necessary to uphold them. Libertarians see the liberties of the moderns in largely negative terms, as merely necessitating the removal of barriers that interfere with free exchanges between individuals (Kukathas Reference Kukathas and LaFollette2003: 572). On this view, there was little need for Union citizenship as a social or political status – it was sufficient to uphold the four freedoms as inherent aspects of a ‘common market’, avoiding welfare and political rights as creating potential distortions with its free operation while supporting the possibility of economic migration from poor to rich countries as consistent with a genuinely free market in labour. By contrast, more socially minded cosmopolitans have argued that rich countries also have more positive obligations towards the poor (e.g. Pogge Reference Pogge2008). Theorists differ as to how far these extend, but most contend some redistribution is warranted given that the wealth of the rich depends in part on their having exploited the resources of the poor and deployed their superior bargaining position to gain favourable terms of trade.

The quandary confronting social liberals, though, is that the institutional capacity for securing the libertarian, market-reinforcing view of the liberties of the moderns is far greater than that for implementing the market-correcting view they favour. As Fritz Scharpf has noted (1999: 54–8), ever since Cassis de Dijon Footnote ³¹ the ECJ has effectively constitutionalised free competition within the EU, overriding the political judgement of national legislatures on the reasonableness of their environmental, health and safety, and other regulations whenever it felt they lacked an adequate public interest defence. The opening up of the full range of public services to competition, so that Union citizens may choose from a range of providers, is simply an extension of this logic. Yet, this possibility potentially undermines the social contract within each of the member states without establishing any at the EU level (Scharpf Reference Scharpf2009: 198). For example, the decision in Watts simply enables those citizens who are sufficiently mobile and proactive to seek a given health treatment in another member state to jump the waiting lists and other restrictions that national services employ to prioritise the spending of limited resources among different kinds of health care (Newdick Reference Newdick2006). As such, it certainly enhances the ‘modern liberty’ of those citizens able to take advantage of this option. But, given that national budgets are not infinitely elastic; their doing so may be at the expense of the health or other social needs of many of their fellow citizens. Moreover, these other individuals are not in a position to contest such Court decisions through the political system. Instead, their collective civic liberty has been undercut by this extension of an essentially commercial liberty.

Benhabib appears to acknowledge this dilemma in seeking to distinguish the ‘human rights’ claims made by refugees, asylum seekers and migrant workers from the deregulatory legal framework promoted by global capitalism – what she calls the lex mercatoria (Benhabib Reference Benhabib2007: 22, 33). Yet, this theoretical argument overlooks how in practice the language of the first has often been deployed to legitimise the second. Like others (Caporoso and Tarrow, Reference Caporoso and Tarrow2008; Kostakopoulou Reference Kostakopoulou2008; Kochenov 2011) she has seen the Court’s extension of rights to free movement and to non-discrimination on the basis of nationality to those outside the labour market as marking a move from its market bias to one based in rights. She argues they reflect a cosmopolitan duty of ‘hospitality’ that, in time, ought to enable migrants from poor countries to gain access to the social rights of wealthier states (Benhabib Reference Benhabib2008: 22–3, 36). However, in many respects the Court has simply deployed the language of human rights to further extend its market logic. By portraying the negative rights associated with market-reinforcing liberties as extensions of humanitarian duties not to unduly interfere or exploit others and to uphold basic rights, it has been able to overcome all democratic objections on the part of the member states. These have not been examples of ‘democratic iterations’ as Benhabib claims (2007: 33), but rather a means to trump national exercises of self-determination (Scharpf Reference Scharpf2009: 193).

The difficulties of extending positive rights on the basis of free movement are both normative and practical. Normatively we incur such obligations to our fellow citizens through being associated with them within a given political system that possesses the capacity to determine and compel obedience to the rules governing our social and economic interactions with each other. Through the exercise of our political liberty we are co-responsible for these rules, and so have a mutual obligation to ensure they operate in as equitable and impartial manner as possible. We also help sustain them through our economic activity and taxes. However, if we can claim these rights without incurring the related obligations, say by forcing my fellow citizens to pay for a service in another country that as a result of collective decisions in which I could and probably did participate is unavailable or less available to me in my country of full citizenship, then this social and political compact is undone. Meanwhile, citizens do not exist in such relations with non-citizens. International organisations – even highly developed ones such as the EU – have limited powers and are authorised not by citizens directly but by their states. They are voluntary agreements to pursue certain circumscribed common purposes for the mutual benefit of the parties concerned. Cosmopolitans are sometimes inclined to suggest that we should create global institutions capable of treating all individuals equally, given that membership of any given state is a mere accident of birth – and even, somewhat contentiously given its continued intergovernmental character and manifest democratic failings, to view the EU as demonstrating the possibility of such schemes (e.g. Held Reference Held1995: 111–13, 254–5; Cohen and Sabel Reference Cohen and Sabel2006). However, this proposal confronts the normative problem of already existing states. As a matter of consistency, our enjoyment of the rights of citizenship may imply a duty to ensure that everyone can also enjoy this right. Yet, the right of everyone to be a citizen of some state does not entail that we all must be citizens of a global state or federation of states. Indeed, the continued role played by national political cultures in defining the boundaries of citizenship, and the problems of establishing effective political mechanisms within large-scale, multinational political systems, suggests the attempt to do so would be ill-fated (Sangiovanni, Reference Sangiovanni2007).

Certain analysts have suggested that these concerns with EU democracy and its impact on welfare policies are misplaced. The legal regulation required to uphold the market rights that form the EU’s core business is both uncontroversial, given that it is Pareto-efficient and reflects common interests, and best administered by expert, technocratic bodies that are immunised from potentially distorting political interferences (e.g. Majone Reference Majone2001; Moravcsik Reference Moravcsik2002). Such matters are of low electoral salience and often depoliticised even within the member states. However, this argument raises Constant’s main worry regarding the ‘liberty of the moderns’: namely, that individuals will be tempted to delegate their safe-keeping to ‘Enlightened’ rulers promising to act on behalf of public utility (Bellamy Reference Bellamy2010).

As he noted, such schemes have three main weaknesses: they offer no safeguard against factionalism; they fail to encourage a patriotic identification with public policies; and offer no means for ‘guarding the guardians’. All three apply here. First, regulators have no incentive to respond to the concerns of the public, and no effective mechanism for gathering information on what those concerns may be. Constraining the access of the general public raises the risk of regulators being ‘captured’, or unduly influenced, by certain sectoral groups, and producing policies that are inequitably and possibly inefficiently partial to those interests (Coen and Thatcher Reference Coen and Thatcher2005).

Second, there is a danger that citizens will lack a sense of ownership of these regulations, disputing their point even when they are in their own interests, and become alienated from those who uphold them. The fact that less than 50% of citizens think ‘EU membership is a good thing’ – not just in Euro-sceptic Britain (where only 33% think it good and 25% regard it as ‘bad’) or Finland (36% ‘good’, 25% bad), but also in traditionally Europhile nations such as France (44% good) – sits ill with the view that civic engagement is unnecessary for the EU’s legitimacy so long as the ‘outputs’ provided by the independent technocracy are themselves legitimate.Footnote ³² Moreover, it is disputable precisely how uncontentiously win-win market-making regulatory policies are – even taking into account the compensation offered by the Social Fund and other mechanisms to overcome short-term costs for particular groups. As Majone acknowledges, redistributive policies do require more democratic legitimacy than purely regulative ones (Majone 1996: 294–6). For in these cases it is likely to prove even harder to get citizens to buy into the provision of public goods that may appear prima facie to conflict with their personal exercise of their civil and commercial liberties.

Finally, this argument overplays the domestic analogy, underestimating the ways elected politicians control non-majoritarian regulatory bodies in the member states. The autonomy of domestic regulatory bodies is generally limited by various screening and sanctioning mechanisms that allow the political principals to control their technocratic agents. Though many formal instruments appear too costly and arduous to employ with any regularity, potentially impugning the neutrality of the agency and thereby undermining its chief asset, or risking associating the political principals with any failure, a range of less overt and informal measures prove as effective. By selecting friendly yet independent experts, with no direct party or other link to government, and managing the effectiveness of the body through their hold on information or role in implementing its recommendations, politicians can shape the institutional incentives in such ways that regulators propose congenial policies (Thatcher Reference Thatcher2005: 347). At the EU level, the plurality of principals and the ability of the Commission to develop a complex network of overlapping agencies, all reduce this influence while introducing the dangers of conflicting forms of accountability. Moreover, the possibilities for regulatory capture are increased by the closeness of EU regulation to various ‘stakeholders’ – notably business and unions (Coen and Thatcher, Reference Coen and Thatcher2005: 341–2). Domestic regulators also come under diffuse public pressure from the media and other organs of the national public sphere – a pressure that is far harder to exert at the EU level given the virtual absence of a pan-EU public sphere. One can hardly regard monetary policy as of low electoral salience, yet the European Central Bank is far more independent from public and political opinion than any of its member state equivalents.

Some analysts argue that these difficulties can be overcome by selective consultation with an emerging European public formed of transnational civil society groups (Sabel and Zeitlin 2007). Yet, in the absence of shared identities, reciprocal social relations and proper parties, the political liberties pursued by Union citizens are of a markedly individualistic rather than a collective kind (Scharpf Reference Scharpf2009: 176–8). They consist of special interest and single issue groups and court actions that typically seek benefits for the individuals themselves and their supporters, while transferring the costs onto others (Warleigh Reference Warleigh, Bellamy, Castiglione and Shaw2006; Kröger Reference Kröger2008; Harding Reference Harding1992) These channels invariably promote rather than counterbalance factionalism and rent seeking (Olson Reference Olson1974).