I. Introduction
Control over nationality was long considered to be one of the few remaining bastions of national sovereignty within the European Union. Member States jealously guard their prerogatives in this area and the relationship between EU citizenship and nationality is still defined by strict linkage and derivation: national citizenship enjoys primacy over, and conditions access to, EU citizenship. The relevant Treaty provisions stipulate this clearly: ‘every person holding the nationality of a Member State shall be a citizen of the Union. Citizenship of the Union shall be additional to and not replace national citizenship’.Footnote 1 This bastion of sovereignty has come under siege, however, precisely because of the EU having created its own citizenship.Footnote 2 Commentators have questioned the derivative nature of EU citizenship and its firm connection to Member State nationality and have argued for a radical restructuring of EU citizenship. This article challenges these proposals and offers a qualified defence of the primacy of nationality over EU citizenship.
Commentators on EU citizenship increasingly take the position that the power of Member States to define who constitutes an EU citizen should not be unbounded and absolute; a position informed, broadly speaking, by two arguments. As the Vice-President of the European Commission, Viviane Reding put it: ‘naturalisation decisions taken by one Member State are not neutral with regard to other Member States and to the EU as a whole’.Footnote 3 First, the strict linkage between nationality and EU citizenship gives a European dimension to every domestic decision concerning nationality; one acquires (or loses) EU citizenship upon the acquisition (or loss) of the nationality of a Member State. The belief is that the EU should bear greater responsibility for protecting EU citizenship by having rules concerning the possession of it. This position has been widely espoused in the context of deprivation of EU citizenship, occurring in individual cases when persons are deprived of their Member State nationality,Footnote 4 or affecting an entire national people as a result of a country's political decision to withdraw from the EU.Footnote 5 Secondly, because the rights to move and reside freely in other Member States and non-discrimination on grounds of nationality are among the substantive rights enjoyed by EU citizens, all domestic decisions in the field of nationality potentially have a cross-border dimension, affecting other Member States. This explains why it is felt necessary for there to be common rules that constrain the powers of national decision-makers to combat national practices that disregard the interests of other states. This debate emerged against the backdrop of what many perceive as the excesses of nationality laws, including the creation and expansion of preferential naturalisation regimes for wealthy investors and other individuals with particular abilities (athletic, scientific, or cultural),Footnote 6 and practices allowing for the mass-naturalisation of specific groups of individuals with particular historical, though tenuous, contemporary connections to the country. One such instance is the regime established by Spain and Portugal that allows for the naturalisation of the descendants of the Sephardic Jews that were forced into exile in the 15th and 16th century.Footnote 7 Another example is the access to citizenship offered by Italy to descendants living in Latin America.Footnote 8
The purpose of this article is to reassess the relationship between national and EU citizenship and examine the normative desirability of important proposals for changing that relationship which have been advanced by various scholars in recent years. The normative implications of these proposals have not been properly assessed and on closer inspection appear to be problematic. This article will argue that some of the concerns regarding the acquisition and loss of national and EU citizenship are valid and there are good reasons for placing some restrictions on the Member States’ power in relation to citizenship, but challenging the primacy of nationality over EU citizenship is not the answer. In response to some of the problems, two alternative models have been proposed which do challenge national primacy: a federal model that maintains the linkage between national and EU citizenship but reverses the derivative relationship, giving primacy to EU citizenship; and an autonomous model, which proposes a (partial) decoupling of EU citizenship from nationality. In contrast to these views, this article defends the current model that recognises the primacy of national over EU citizenship, though in a qualified form. It argues that linkage must define the relationship between EU citizenship and nationality, and the acquisition and loss of EU citizenship be conditioned by national rules on the acquisition and loss of Member State nationality. However, to the extent that the above examples demonstrate the limitations of the current structure of EU citizenship, the answer is not to reconsider the criteria of ‘linkage’ and ‘derivation’, but to create some common EU rules on ‘access’ to national and EU citizenship.Footnote 9
This article is structured as follows: Section II offers conceptual clarity concerning the alternative proposals for changing the relationship between nationality and EU citizenship, distinguishing three ideal-type models: a federal, autonomous, and national model of EU citizenship. Section III dismisses the federal model for failing to respect the associative ties among those who have contributed to building and maintaining the collective goods offered by State institutions. Section IV dismisses the autonomous model of citizenship on the ground that it places EU decision-making beyond the control of the different national bodies of citizens. It follows that the primacy of nationality over EU citizenship must be preserved to avoid these problems. Section V argues that the problem of offering the Member States full power in setting the rules on the acquisition and loss of national and EU citizenship is that it allows for practices that disregard the interests of other States’ and third-country nationals. Certain shared minimum rules on access are needed to prevent this. Introducing the idea of a ‘genuine link’ in EU nationality law is an attractive solution. Section VI questions whether the EU can introduce common European rules that provide for minimum standards on access to national, and thus EU, citizenship.
II. Three models of EU citizenship
Before defending the case for recognising the primacy of national citizenship within the EU, it is necessary to clarify the different ways in which we can conceptualise the relationship between nationality and EU citizenship. Three dimensions define this relationship: linkage, derivation, and access. Linkage concerns the question of whether a connection should exist between EU citizenship and nationality. Derivation is about the causal direction between the two; about which enjoys primacy over the other. Access refers to who decides on the criteria of acquisition and loss of citizenship.Footnote 10 Based on these dimensions, three ideal-type models can be identified: a national model that treats EU citizenship as deriving from the nationality of a Member State; an autonomous model that aspires to remove the link between EU citizenship and nationality; and a federal model in which EU citizenship has primacy over nationality.Footnote 11 There are variants of all these models and the boundaries between them are not as clear as this section may suggest. Furthermore, as this article demonstrates, these models do not simply assert a different connection between both nationality and EU citizenship; they redraw the boundaries of democratic and social inclusion within the EU. Hence, an assessment of these models of citizenship depends on an understanding of the relevant boundaries of inclusion. This section only draws the contours of these ideal-type models, while subsequent sections connect these with the concrete proposals that currently exist.
The national model recognises the different national peoples and is sympathetic towards the status quo. It respects the linkage between national and EU citizenship and treats the latter as the derivative status. One understanding of the national model is that the primacy of nationality is absolute and no interference with national decisions regarding citizenship is permissible. However, advocates of the national model need not be completely opposed to European constraints. As argued below, there is reason to believe that within a supranational Union, where national citizenship decisions have greater cross-border implications, national decision-makers should take into consideration the interests of other parties affected by their decisions. From that perspective, the Member States remain primarily responsible for setting the standards on the acquisition and loss of nationality and EU citizenship, but those standards must conform to European standards of mutual respect. In other words, there must be a strong presumption in favour of the primacy of nationality, but this can be rebutted in certain circumstances. What is important for now is that this national model proposes that the authority to decide on who is a citizen of a Member State and of the European Union remains principally with the Member States.
Of the two alternatives, the autonomous model is the more widely accepted. Those who seek to establish the autonomy of EU citizenship from national citizenship challenge, either in part or in full, the linkage that currently exists. They propose (partially) autonomous conditions governing the acquisition and/or loss of EU citizenship. That is, EU citizenship should not be determined (solely) by national rules, but by the EU collectively. Theoretically, one can envisage EU citizenship as a fully autonomous status, completely disconnected from nationality. Member States would retain the power to determine the conditions for the acquisition and loss of national citizenship, whilst the EU would determine who its citizens are. Usually, however, the preferred alternative to the status quo is one of partial autonomy. This means that the current linkage between EU citizenship and nationality is retained—Member State nationals are EU citizens by definition—while non-nationals are offered an additional European path to EU citizenship that is independent of naturalisation. The various proposals will be discussed in more detail below. It suffices to note that proponents of the autonomous model prefer EU policies not to be determined by national governments. Domestic matters can be decided by national bodies but EU matters should be determined by the people of Europe collectively.
The federal model is furthest removed from the current approach and seeks to reverse the existing derivative connection between EU citizenship and nationality.Footnote 12 Rather than preserving the primacy of national citizenship, advocates of the federal model want EU citizenship to have a primacy akin to that in federal states.Footnote 13 Like the autonomous model, the federal model reflects the unity of a European people and subordinates national citizenship. Under such a model, acquisition and loss of EU citizenship would be subject to a set of uniform European rules: national citizenship would be derived from EU citizenship, and rather than being based on the Member States’ citizenship and naturalisation policies, it would be based on the country of residence.Footnote 14 An EU citizen with residence in the Netherlands would be a Dutch citizen, but would acquire German citizenship following a change of residence to Germany. It seems plausible to presume that if national citizenship were to become conditional upon one's State of residence, access to political and social rights would also become residence-based. Rather than have claims of social and democratic inclusion towards their former State of residence, they would be in a relationship of social and political belonging with their current State of residence.Footnote 15
These different models do not just offer different understandings of EU citizenship but a restructuring of political relations within the EU. That matters, because proposals to alter the structure and meaning of EU citizenship are sometimes dismissed on the ground that the current model is most compatible with the EU's current organisation, as an institution that sits somewhere in between a classical international organisation and a federal union. The Member States remain the principal political institutions within the EU and a federal or autonomous model of citizenship is thus incompatible. These remarks are accurate, but do not fully address the arguments made by proponents of a different model of citizenship within the EU. A full reply should explain not just that the EU's current structure presupposes the primacy of national citizenship, but why it is desirable that no radical restructuring takes place and the Member States remain the prime loci of political and social belonging.
The EU's current legal framework is thus important when the discussion concerns the possibility of realising different relations between EU and national citizenship with or without Treaty change, but of limited relevance when discussing the normative desirability of alternative models of EU citizenship. However, it happens that arguments for a different model of EU citizenship draw explicitly on the case law of the Court of Justice of the European Union (CJEU or Court). The Court has acknowledged that ‘it is for each Member State … to lay down the conditions for the acquisition and loss of nationality’, but added that the Member States must carry out that task, ‘having due regard to EU law’.Footnote 16 In addition, it decided that EU citizenship ‘is destined to be the fundamental status of nationals of the Member States’.Footnote 17 Such statements have been used to justify a different relationship between EU and national citizenship. In their defence for an autonomous status, for example, Dawson and Augenstein wonder ‘how fundamental European citizenship really is’ if all UK citizens will be stripped of their EU citizenship following Brexit.Footnote 18 The answer is that EU citizenship is evidently not fundamental in that sense and still contingent upon Member State nationality. The Court's understanding of EU citizenship has always been hard to square with the ‘text, teleology and legislative history’ of the Treaties,Footnote 19 given that the Treaties so clearly designate EU citizenship as the derivative status. It appears difficult, in other words, to challenge the primacy of nationality over EU citizenship, and it is improbable that the qualifier ‘having due regard to EU law’ will acquire much more than symbolic significance under the EU's current legal framework.Footnote 20
More challenging are the considerations of those who aspire to realise a different form of citizenship of the EU in the future, if only because their arguments seem highly attractive at face value. It is difficult not to sympathise with those who want to protect all EU citizens against the involuntary loss of their status, or with those who envisage the cosmopolitanisation of EU citizenship which, according to its supporters, would generate an inclusive European people committed to mutual solidarity and capable of deciding collectively and democratically the Union's future. However, there are significant downsides to the federal and autonomous models of citizenship for the EU, to which the following two sections draw attention.
III. A critique of the federal model
The federal model constitutes such a radical departure from the present construction of EU citizenship that, probably for that reason alone, no one embraces it fully. No one appears to argue for the full harmonisation of the rules on the acquisition and loss of EU citizenship and the replacement of national rules. On the other hand, the model is frequently embraced in part. Recall that it presumes that national citizenship is derived from a common EU citizenship, with residence being the sole factor conditioning the boundaries of national membership. Several commentators want EU citizenship to have such inclusionary implications and to entitle mobile citizens to full democratic and social citizenship within their State of residence.Footnote 21 This section offers two grounds for dismissing such calls for redrawing the boundaries of social and democratic membership within the EU. First, contrary to what these commentators claim, there is no obligation of social and democratic justice. Second, they fail to respect the associative ties among those who have contributed to building and maintaining the collective goods offered by State institutions. It also follows that the federal model of EU citizenship more generally ought to be rejected.
That economically active EU citizens should be entitled to claim social assistance in the Member State where they pursue their employment as a fair return for their contribution and participation seems relatively uncontroversial.Footnote 22 Far more contentious is whether the unemployed should have an entitlement to social assistance, certainly in their initial period of residence within the host State. Several EU lawyers have taken the position that principles of social justice require full and equal access to social benefits for all mobile citizens. Hence, when the CJEU permitted Germany in Dano to deny benefits to a Romanian national who had been resident in Germany for a limited period of time, during which she had not been employed, nor searched for work or tried to integrate,Footnote 23 they argued that the Court failed to deliver justice.Footnote 24 That position is subject to two objections. First, as this author has argued elsewhere, it rests upon a problematic conflation of justice with political legitimacy.Footnote 25 In addition, it takes for granted that domestic principles of egalitarian justice extend to the EU,Footnote 26 an argument which should be rejected.
At first sight, this argument appears to resemble the cosmopolitan argument that, given the equal moral worth of human beings, egalitarian principles that apply domestically also extend to the global level.Footnote 27 However, different accounts of social justice within the EU argue for the application of egalitarian principles to the EU not on cosmopolitan, but on relational grounds.Footnote 28 On the relational view, ‘principles of distributive justice cannot be formulated or justified independently of the practices they are intended to regulate’.Footnote 29 Taking a relational position, one could argue that the implications of political integration within the EU are so pervasive and the interactions between EU citizens so profound that no meaningful difference exists from the national level. Domestic principles of social justice must then extend to the EU. More frequently, however, relational accounts assume that the existence of EU citizenship in and of itself offers sufficient support for the argument that mobile EU citizens should be entitled to claim social assistance on an equal footing with the nationals of the Member States where they reside.Footnote 30
While the argument that principles of social justice have a relational basis is a plausible one, the existence of EU citizenship alone offers insufficient ground for the extension of domestic egalitarian principles of social justice to the European domain. This misses the complementary nature of EU citizenship and its dependence on national membership. It seems difficult, more generally, to argue on relational grounds that the EU and Member States are bound by the same duties of social justice. EU citizens have very different relationships to one another when compared to national citizens, this being a direct consequence of the EU serving very different purposes and having modest institutional capacities when compared to the Member States. The Member States seek to provide the goods needed for citizens to flourish and live autonomous lives;Footnote 31 the purpose of the EU is to strengthen the Member States’ capacity to provide those goods.Footnote 32 Whether the EU always succeeds in realising these ambitions is another matter. Indeed, there is no denying that the EU has fallen short of this ideal and occasionally stood in the way of Member States realising the conditions that allow citizens to flourish. Relational accounts recognise that due to the interconnectedness between the EU and its Member States, the EU is bound by duties of justice, but they can also accept that the scope of principles of justice are more demanding at the national than at the EU level.Footnote 33
Despite the pressures of globalisation, the institutional capacity of States remains unprecedented. Well-functioning States manage to create ‘a condition of civic equality’ among citizens,Footnote 34 offering protection against social deprivation and physical threats, upholding their legal rights and entitlements and allowing for their participation in decision-making on the collective system of rules under which legal disagreements can be resolved. States can succeed in generating such civic equality only when citizens are willing to cooperate and incur the obligations necessary to generate and uphold the system of collective rules and rights.Footnote 35 Such willingness is greater when there is a sense of solidarity among citizens as well as a degree of generalised trust, not just ‘in the willingness of others to reciprocate benefits when the need arises’,Footnote 36 but also in the willingness of those in power to act in the general interest, even when moral views diverge.Footnote 37 Trust ‘is more likely among citizens who come together within a stable infrastructure of state institutions and who share a sense of solidarity, rooted in a shared political culture’.Footnote 38 Hence, a collective identity and the willingness to participate in the production of collective goods contribute in important ways to the feeling of solidarity needed to generate such trust among individuals.Footnote 39
Relational accounts of justice can accept this and maintain that States are justified in privileging those who have contributed to building and maintaining the collective goods they offer. This position does not question the moral equality of EU citizens (and other human beings) but assumes that our societies are ‘systems of cooperation over time, from one generation to the next’,Footnote 40 whose successful functioning depends on the cooperation and participation of their citizens. Those who have helped produce those public goods also have a legitimate claim to their enjoyment.Footnote 41 Hence, even though a common nationality may contribute to shared feelings of identity and a greater sense of solidarity,Footnote 42 nationality per se is not a justifiable ground for exclusion. All EU citizens who have contributed to the production of these public goods should be entitled to a fair return and claim social inclusion.Footnote 43 Without going into detail here, the EU has long incorporated in its legislation the principle that Member States owe mobile citizens who have made a significant contribution a fair return, conferring a right to claim social benefits to workers,Footnote 44 jobless persons who were economically active previouslyFootnote 45 and long-term residents.Footnote 46 Even if they have not directly contributed though economic participation, the latter will at least have contributed over time on the basis of their societal participation and payment of taxes. However, on the principles of social justice defended here, Member States are under no obligation to provide comprehensive social assistance to all newcomers from other Member States.
By allowing individuals to move across the arbitrary borders that separate national regimes and overcome arbitrary forms of exclusion in their places of residence, the rights of free movement and non-discrimination make a powerful contribution to enhancing the life opportunities of EU citizens. However, these rights also have the potential to undermine the social bonds that support national social practices, by allowing individuals to claim rights and entitlements without incurring the necessary obligations and demonstrating some commitment to their maintenance.Footnote 47 Studies consistently show that by ensuring that access to social benefits is conditional, the EU has largely managed to avoid possible tensions between citizen mobility and national welfare states.Footnote 48 However, such studies also show that free movement can adversely affect the sustainability of social benefits in cases where welfare entitlements are unconditioned.Footnote 49 While claims that free movement undermines the stability of national welfare regimes are generally exaggerated, guaranteeing that a degree of reciprocity exists between mobile citizens and their State of residence prevents mobile citizens putting unwanted strains on national welfare entitlements. Their mere presence for brief periods does not establish sufficiently strong reciprocal ties. Newcomers thus should not be entitled to full social citizenship within their State of residence.
Newcomers need not be included in the national demos immediately either. Currently, EU citizens resident in Member States other than that of their nationality can stand as candidates and vote in elections to the European Parliament and municipal councils, but are excluded from the national franchise.Footnote 50 Moreover, because certain Member States exclude nationals who have left the territory and are resident abroad (for a certain duration) from the franchise,Footnote 51 mobile Union citizens risk being deprived of the right to vote in national elections altogether and thus also of a say over who represents them in the Council of Ministers. Such outcomes constitute clear democratic wrongs, but whether the solution lies in giving mobile EU citizens the national franchise in their country of nationality or of residence remains an open question.
Those who favour redrawing the boundaries of democratic inclusion to include all lawfully resident EU citizens in the national demos usually offer two justifications for their argument. First, they argue that such inclusion is demanded by the ideal of equality embedded within citizenship.Footnote 52 Second, they insist that democratic inclusion in the State of residence is a requirement of democratic justice, following the democratic principle that all those affected by democratic decisions should have a say in the process that led to their adoption.Footnote 53 As regards the latter argument, there are well-known problems in conditioning the boundaries of the demos by the criterion of affected interests. First, one cannot know in advance of a political decision being taken whose interests will actually be affected, meaning that the appropriate boundaries of the demos cannot be decided until after the adoption of the measure. The principle of affected interests thus presents logical impossibilities.Footnote 54 In addition, decisions of foreign entities constantly affect us, so that the principle of affected interests provides ‘good grounds for thinking that (at least in principle) we should give virtually everyone a vote on virtually everything virtually everywhere in the world’.Footnote 55 The EU offers an instructive example. Decisions of different Member States affect EU citizens simultaneously. Mobile Union citizens, for example, are not only affected by the political decisions taken by their country of residence but also by the actions of their country of nationality. In addition, EU citizens’ lives are also affected by political decisions taken by third countries. If it is affected interests that matter, EU citizens would have to have a vote in all these elections, which—if only for practical reasons—will obviously not be appropriate.Footnote 56
More importantly, the impact of the outcomes of political decision-making alone cannot define the boundaries of democratic membership in the EU, as political communities are ongoing (re)arrangements of cooperation among citizens over time. Visitors and short-term residents, such as exchange students and jobseekers, have not built the associative ties and produced the long-term commitments necessary to be included within the national demos.Footnote 57 No injustice is done to them if they are included within the demos of their State of nationality rather than their State of residence. It is important, therefore, that mobile citizens resident in Member States other than that of their nationality are not deprived of their political rights in their country of nationality. Only after a certain period of residence in another State should EU citizens be able to acquire voting rights (and lose their political rights in their State of nationality in order that they do not have a double say in EU affairs).Footnote 58 Among the possible options are (1) the facilitated naturalisation of those EU citizens who have been living in the host Member State for a reasonable period, or (2) the immediate enfranchisement of those with permanent residence.Footnote 59 Such reforms, however, need not include the democratic inclusion of relative newcomers lacking genuine associative ties. There is reason, therefore, to reject the federal model of EU citizenship and to believe, as the Treaties say, that EU citizenship must ‘complement and not replace national citizenship’.Footnote 60
IV. A critique of the autonomous model
If the problem of the federal model is that it removes from the Member States the power to delimit the boundaries of national citizenship, the autonomous model of citizenship within the Union does not seem to present such problems at first sight. By (partly) decoupling EU citizenship from nationality, granting the EU the prerogative to decide on the possession of EU citizenship, while leaving decisions on national citizenship to the Member States, it appears to escape these shortcomings. This section, however, argues that decoupling EU citizenship from nationality risks diminished control of national citizens over EU political decision-making.
Proposals for turning EU citizenship into an autonomous status come in different forms. Theoretically, one could imagine a complete decoupling of EU citizenship from nationality, but no one seems to favour a fully autonomous EU citizenship status.Footnote 61 It would be odd if Member State nationals were not EU citizens by default; the EU's entire political system presupposes that national citizens are involved within the EU's political processes and bound by the political outcomes produced by EU institutions. It is difficult to imagine the full separation of the national and EU dimensions of governance. This probably explains why most proponents of an autonomous EU citizenship status favour its partial decoupling from nationality. What follows focuses on two different proposals for partial autonomy. If these do not withstand scrutiny, models of full autonomy certainly will not.
EU citizenship can be disconnected from nationality through both the loss or acquisition of that status. In the aftermath of the UK referendum on Brexit facing the prospect of British citizens’ losing EU citizenship, both options were considered. Dawson and Augenstein argue that there should be a decoupling of EU citizenship from nationality with respect to the loss of that status. If their idea was implemented, ‘the decision to grant Union citizenship may still rest with the Member States, via Member State nationality, [but] the decision to withdraw it would rest with the individual EU citizen’.Footnote 62 Kostakopoulou has focused on the acquisition of EU citizenship, and her proposal is limited to those individuals who do not have EU citizenship because they do not possess the nationality of a Member State. For third-country nationals, EU citizenship would be conditioned by domicile, available to those ‘who have been residing on a lawful and permanent basis in the territories of the EU for five years’.Footnote 63 There are other possibilities in-between and beyond these two proposals, including the different options explored by Garner in his recent ‘argument for an autonomous status’,Footnote 64 but an examination of the schemes suggested by Kostakopoulou and by Dawson and Augenstein should be sufficient to illustrate the downsides of an autonomous Union citizenship.
All proposals for decoupling EU citizenship from nationality, irrespective of their form, run into one immediate problem, which we can call the problem of partial political representation. EU citizens are currently represented twice in the EU's decision-making processes, once directly by the European Parliament and once indirectly by national Council representatives, who are accountable to national parliaments. Assume that, in accordance with Dawson and Augenstein's proposals, it would be possible for UK citizens to retain EU citizenship following the UK's departure from the EU, or, as Kostakopoulou proposes, that individuals could acquire EU citizenship without simultaneously acquiring Member State nationality. Such Union citizens would lack the indirect representative connection with the EU decision-making process through national citizenship and domestic elections. Hence, the decoupling of both statuses alone hardly seems suitable for what proponents of an autonomous form of citizenship within the EU wish to realise: EU citizens becoming the collective authors and owners of EU politics.
There are two possible ways in which that shortcoming might be addressed. One is the full inclusion of EU citizens within their State of residence and conferring them with the right to vote in national elections.Footnote 65 The previous section dismissed this option. A second answer to the problem of partial political representation is to get rid of national citizens’ indirect channels of representation by removing the control of national political representatives and parliaments over EU decision-making. The European Parliament could become the sole body representing the EU citizen. Such an institutional innovation would give all Union citizens full political rights in European elections, even if not all of them would be allowed to vote in national elections.
On the face of it, taking national processes out of European processes of decision-making appears to offer a number of advantages. It would solve the problem of partial representation, but also seems to reflect the idea of a European people; a united body of citizens that self-regards one another as equals and is capable of collective self-determination. It would, as Dawson and Augenstein put it, premise the EU ‘upon the unity of the peoples of Europe’, which for them is ‘another way of saying the people of Europe’.Footnote 66 That is, they seek to realise ‘a European polity based on mutual commitment of its peoples to construct a new form of civic and political allegiance on a European scale’.Footnote 67 A ‘more cosmopolitan political community’, in Kostakopoulou's words, would be realised.Footnote 68 Finally, it may appear that an autonomous form of citizenship within the EU would ‘liberate individuals from the preferences of their states’.Footnote 69 In such a Union, ‘the future EU citizenship … is not a domestic matter but an issue … for the Union as a whole to determine’.Footnote 70 That is, decoupling the connection between nationality and EU citizenship is inspired largely by the aspiration to let a European people self-determine its own future collectively, independent of the normative ambitions of the various States.
The likely outcome is very different from that predicted by these accounts. By detaching EU decision-making from the control of domestic political processes, EU citizens will enjoy less control over the political decisions they are subject to. This risk emerges from there still being a highly underdeveloped European demos and public space. Democracy offers a fair mechanism for deciding on political action when there is disagreement among citizens, a solution proponents of an autonomous EU citizenship seek to implement at EU level. However, for political procedures to confer democratic legitimacy, it is not sufficient that they ensure the formal political equality of citizens and guarantee equal voting rights. Citizens must also self-regard themselves as a collective of equals and be in a position to exercise equal influence over the allocation of political authority. Decision-making among citizens who do not belong to a common demos, a collective that sees themselves as equals in processes of collective governance, is less likely to be geared towards the general interest and more towards the particular interests of those in positions of power. The existence of a public sphere enables citizens to understand what is at stake in the political arena, enabling them to respond to and hold accountable decision-makers.
The time available to ordinary citizens to follow political debates and developments is limited. Yet, they are expected to have an adequate understanding of what is at stake in political debates in order to exercise their democratic rights adequately and responsibly. A public sphere is a necessary ‘response’ to the scarcity of time. To have a reasonable understanding of what is at stake and debated, citizens must and do rely on the comprehensive set of channels and institutions—political parties, mass media, and a public discourse—that are present in functioning democratic States.Footnote 71 The European public space, by contrast, remains very underdeveloped. A European mass media and public discourse is almost non-existent and European political parties may function like national parties in internal legislative politics, but play only an insignificant role in external electoral politics of mobilising voters and providing citizens with relatively accessible information regarding the political preferences of particular candidates.Footnote 72 It also seems unlikely that a rapid growth of the institutions necessary to mediate between citizens and political actors at the European level will occur any time soon. Language remains the most prominent barrier to the realisation of a European public sphere. Most citizens speak only their mother tongue and even those who have a reasonable understanding of a second or third language rarely grasp it sufficiently to participate in transnational political discourse. Democratic politics being ‘politics in the vernacular’,Footnote 73 most citizens will only be able to grasp domestic political debates and inform themselves about European politics through national channels of communication and information dissemination. Without national political institutions exercising some control over EU decision-making and national media outlets providing the communication between national and EU political institutions and the national citizens, control over EU politics risks being even further removed from the majority of EU citizens and will become more dominated by the elite than it already is. Contrary to what proponents of an autonomous form of citizenship assume, therefore, the aspiration to separate national from EU decision-making will not enable EU citizens to collectively decide the EU's future course of action. Rather, it will further remove control over EU political institutions from them.Footnote 74
Premising decision-making within the EU upon the unity of citizenship also risks removing control over EU decision-making from smaller and weaker States. If citizens are unable to identify with each other and treat those they disagree with as equals, there is a greater likelihood of dominant majorities persistently disregarding the interests of insulated minorities. Such minorities can have a vote, but their lives will be dominated by strangers. There are still strong indications that nationality is the dominant cleavage cutting across the European political spectrum, and there is no sense of a European people that overcomes this divide and which is able to act in the collective European interest. The political response to the Eurocrisis offers one important example. It has been difficult to detect some sense of solidarity among Union citizens and the burden of the crisis was placed largely on a smaller group of weaker Member States, who were obliged to implement far-reaching institutional and economic reforms, which greatly impacted upon their standards of national welfare. The different national governments sought primarily to promote their own particular interests, occasionally blatantly disregarding the interests of the others. This is no defence of the political decisions taken in the wake of the crisis, but it is important not to ignore the fact that the crisis was ‘resolved’ by removing decision-making from processes that aspired to realise the shared and equal control by all Member States over EU decision-making. Instead, decisions were taken by unaccountable executives (the ECB, Commission, and IMF) or by newly constituted institutions that accorded Member States degrees of influence in accordance with their financial position.Footnote 75 Citizens of weaker Member States were affected by political decisions over which they lacked any real influence. This shows that without a greater sense of commonality among EU citizens, they ‘will not and should not accept to be bound by a majority of Europeans’.Footnote 76 Nicolaïdis was correct to observe that the risk of realising procedural equality between EU citizens is producing ‘de facto domination between peoples’.Footnote 77 National democratic institutions should be involved in EU political decision-making. Removing control from them in order to overcome the problem of partial political representation will not enable a European people to collectively decide the EU's course of action; on the contrary, it will diminish citizens’ opportunity to control the power to which they are subjected.
V. The primacy of national citizenship
This article has argued against reversing or severing the connection between EU citizenship and nationality. First, EU citizenship should not acquire primacy over nationality because this risks undermining the social bonds that support national democratic and social practices. The previous section also dismissed the idea of turning EU citizenship into an autonomous status. The exclusion of national political institutions from EU decision-making processes, a necessity if EU citizens are to enjoy full rights of political participation under the autonomous model, would weaken the control EU citizens can exercise over decision-making by EU institutions. From this, it follows that the primacy of nationality over EU citizenship remains desirable. That is, linkage must define the relationship between nationality and EU citizenship, and the latter is derivative.
It may seem that this article has also suggested that the EU should not constrain the domain of nationality, over which Member States should enjoy full autonomy. That is, it may appear to defend a model of self-determination that requires non-interference with the choices taken by national institutions. As Iris Marion Young has demonstrated, however, the non-interference model is not the only understanding of self-determination.Footnote 78 In an increasingly interconnected world, various social and economic interdependencies exist between different self-governing entities and the decisions taken by one may negatively affect the interests of outsiders. This may produce relationships of domination which are fully compatible with a model of self-determination understood in terms of non-interference. The limitation of that model is that its insistence on the full autonomy of self-determining entities can in fact undermine the autonomy of smaller and weaker entities. Therefore, Young offers a different model of self-determination, ‘one that puts the objective of mutual respect and avoidance of domination more at the center’.Footnote 79
Even in the domain of nationality, the non-interference model of self-determination no longer exists. International law restricts States in matters of nationality.Footnote 80 Within the EU, however, there are additional reasons for limiting the ability of Member States to determine who can enjoy national citizenship. As has been said in the Introduction, because of the strict linkage between nationality and EU citizenship, national decisions in the field of citizenship also acquire a European and cross-border dimension. That is, such decisions define the personal scope of EU citizenship and they also affect other Member States, because the right to free movement and non-discrimination on grounds of nationality are among the substantive rights enjoyed by EU citizens. Of course, by virtue of the current structure of EU citizenship, each and every decision in the field of nationality has European and (potentially) cross-border implications. However, it should not be thought that a full harmonisation of rules on the grant of citizenship at EU level is preferable. National citizenship must retain its primacy over EU citizenship for the reasons set out in the previous sections. At the same time, Member States should not ignore the interests of citizens of other Member States and third-country nationals. The model of self-determination proposed by Young, and which is adopted here, involves a presumption of non-interference which can be rebutted if national authorities pay insufficient consideration to the interests of others in the decisions they take.
Consider the views offered by Advocate General Maduro in his Opinion in the Rottmann case. He said that ‘Union citizenship assumes nationality of a Member State but it is also a legal and political concept independent of that of nationality’.Footnote 81 Those favouring the autonomous model have emphasised the AG's use of the word ‘independent’ and have suggested that he is advocating the decoupling of nationality and EU citizenship,Footnote 82 but that is not what he proposed. He added that ‘it is not that the acquisition and loss of nationality … are in themselves governed by [EU] law, but the conditions for the acquisition and loss of nationality must be compatible with the [EU] rules and respect the rights of the European citizen’.Footnote 83 He assumes that the link between nationality and EU citizenship will be retained, but proposes introducing certain fetters over the powers of Member States. The national model of EU citizenship would leave the current linkage between national and EU citizenship intact and retain its derivative nature, but the primacy of national citizenship is not incompatible with there being certain common European principles concerning access to national and EU citizenship.
Whether the introduction of such additional constraints is possible under the Treaties or requires an amendment is addressed in the next section. First, it is necessary to examine whether there are actual reasons for adopting EU standards on access to national citizenship. According to commentators there are three different reasons for doing so. These are: the interests of the EU in protecting the status of EU citizenship; the potential negative consequences for other Member States by the grant of national citizenship; the position of third-country nationals (i.e., individuals without EU citizenship). The second and third grounds, in particular, justify the adoption of supranational principles which would limit the power of Member States to decide unilaterally on matters of nationality within the EU.
If it were left for them to decide, some European institutions would constrain Member States in order to protect the integrity of EU citizenship. For example, the European Parliament called for the Commission to intervene in the sale of citizenship by some Member States because such practices undermined ‘the very concept of European citizenship’.Footnote 84 The Commission recently issued a report that expressed similar concerns about similar investment schemes.Footnote 85 Moreover, reasoning that EU citizenship ‘is destined to be the fundamental status of nationals of the Member States’, the Court has ruled that the deprivation of EU citizenship falls within the scope of EU law.Footnote 86 As has been seen, debates concerning EU citizenship following the UK referendum were fuelled by similar beliefs. The need to protect the status of EU citizenship, it is argued, justifies limiting the power of Member States in the sphere of nationality.Footnote 87
These arguments are misguided. EU citizenship is not of such significance as to outweigh national claims to non-interference. EU citizenship is still merely additional to national citizenship and the rights to free movement and non-discrimination offer EU citizens access to rights which flow from national citizenship rather than from a set of independent European rights.Footnote 88 In practice, little would change if EU citizenship was abolished tomorrow. If EU citizenship were abolished, nationals of Member States would still retain their rights and entitlements; Malta would no longer be selling EU citizenship, but investors could still acquire the full set of free movement rights based on the acquisition of Maltese nationality alone; and rendering someone stateless would no longer result in the loss of EU citizenship, but it would still deprive that person of the rights enjoyed as a result of having the nationality of a Member State. Without dismissing the symbolic value of EU citizenship and the relevance of individuals deriving a set of core minimum rights from their status as EU citizens directly,Footnote 89 EU citizenship adds little which is independent of Member State nationality. Protecting the integrity of EU citizenship is therefore an insufficient reason to rebut the presumption of non-interference.
Stronger grounds for restraining the power of Member States to determine the conditions for a grant of nationality are the possible negative consequences of their decision-making and the protection of the position of third-country nationals. It is precisely because EU citizenship is parasitic upon national citizenship that the practice of Member States relating to the grant of nationality can violate the spirit of EU integration. The benefits conferred on EU citizens include the right to travel throughout the EU, to reside in a Member State of which the citizen is not a national and to claim—with limitations—the benefits of national citizenship. Decisions concerning the acquisition and loss of nationality are indeed of significance to other Member States. Investment schemes which grant citizenship to third-country nationals are popular precisely because they allow the purchase of EU citizenship. The adoption of such schemes with little to no consideration of the interests of other Member States’ shows disregard for those with whom they stand in an important and close relationship. As the Commission and various NGOs have warned, such citizenship-by-investment schemes create very specific risks, including the circumvention of security checks introduced by EU migration law, money laundering, or tax evasion.Footnote 90 That individuals enjoying the most dubious of reputations, including persons accused of misappropriating financial assets and of having ties to international criminal organisations, have managed to acquire national and EU citizenship demonstrates the reality of these problems.Footnote 91 Moreover, these risks extend beyond the states of naturalisation to all EU Member States. As Transparency International and Global Witness aptly put it, ‘a minority of Member States are reaping profit from jointly shared EU assets by hawking internal free movement and external visa-waiver agreements, and they are enjoying the spoils whilst exposing their neighbours to risk’.Footnote 92
On the one hand, Member States are giving a small category of individuals without a genuine link to their country privileged access to national citizenship, with all the problems this entails. On the other hand, they also (and some more than others) deny national citizenship to individuals whose genuine link cannot be in dispute, in particular third-country nationals with lengthy periods of residence within their territories. A European Union committed to democratic fairness has an interest in ensuring that foreigners are not excluded permanently from national and EU citizenship. It is not just the case that the permanent exclusion of foreigners from citizenship constitutes a democratic wrong at national level; since national representative processes are key to the legitimisation of EU policies, this failure of democratic inclusion affects the EU's own legitimacy. Denying long term third-country national residents the ability to acquire national citizenship and the corresponding political rights undermines their ability to live freely within those States and to participate in the exercise of political control over those whose power they are subjected to.
The introduction of common European rules providing minimum standards on access to national and thus EU citizenship could solve these issues. Given the problems highlighted above, the obvious solution would be for the acquisition of nationality to be guided by the idea of a ‘genuine link’. This reflects the notion that only those individuals who enjoy a proper connection with a particular State should enjoy its citizenship. Individuals can have a genuine link with multiple States and such rules should be neither over- nor under-inclusive. They are over-inclusive when individuals lacking meaningful social ties can naturalise, as in the case of investor citizenship regimes. They are under-inclusive when full membership is unavailable to individuals that are social members of particular State, such as individuals with a period of genuine residence, first- or second-generation immigrants, and those born and residing in the country for most of their childhood.Footnote 93 In addition, such rules should presume that naturalisation is an entitlement based on objective criteria, and not subject to discretionary assessments by national authorities.
VI. Towards common EU rules on access to national and EU citizenship?
The European Commission recently published a report that draws attention to the risks associated with citizenship and residence by investment regimes within the EU, in which it defended the notion of a genuine link. It argues that Member States should provide ‘that nationality is not awarded absent any genuine link to the country or its citizens’.Footnote 94 This proposal has been subject to some intense criticism, with detractors offering two objections. First, that the principle of genuine links is an objectionable as well as an unrealisable ideal. Second, that even if a solution is available in theory, it is hard to realise in practice as the EU has not been accorded the necessary competences in the field of nationality. This section will reflect further on the desirability and feasibility of a genuine link requirement in EU law. It will argue that arguments questioning its desirability should be rejected, but that its feasibility is indeed questionable.
When making its proposals, the Commission drew inspiration from the notorious Nottebohm case decided by the International Court of Justice in 1955, in which the Court determined that Guatemala was not required to recognise the grant of nationality made by Liechtenstein to Mr Nottebohm on the grounds that he did not enjoy a genuine link with Liechtenstein at the time it was made.Footnote 95 According to Kochenov, leading authorities in the field of international law consider Nottebohm to be a remnant from an overly ‘romantic period of international law’.Footnote 96 The EU's own Court of Justice rejected such an approach in the Micheletti case.Footnote 97 Spiro has offered a similar though more lengthy criticism of Nottebohm and the genuine link test, highlighting the shortcomings in the ICJ's reasoning and demonstrating that the decision never rose to ‘the level of customary international law or of a general principle of law relating to nationality’.Footnote 98 His verdict of the genuine link test is harsh, dismissing it as ‘a nostalgic and romanticised view of the relationship between the individual and the state’.Footnote 99
Such criticism is valid in part. Kochenov is right to note that the CJEU rejected Nottebohm in Micheletti, which precluded Member States from using a genuine link as a requirement for the recognition of Member State nationality. Moreover, in repudiating Nottebohm, he and Spiro joined a large chorus of international lawyers.Footnote 100 At the same time, their criticism confuses the recognition of nationality with the grant of nationality. Although not always clear in its reasoning, Nottebohm concerned the former, rather than the latter—and it is the latter which has been argued requires a genuine link.Footnote 101 The lack of a genuine link may be insufficient reason to refuse to recognise a grant of nationality, but that does not undermine the argument in favour of there being a genuine link for the purposes of its being granted. The Commission's reference to Nottebohm was perhaps unfortunate and a consequence of it not fully realising the importance of maintaining a distinction between the grant of nationality and recognition of a grant of nationality, but this it is by no means fatal to the report's suggestion that a genuine link should be one of the conditions for a grant of nationality by EU Member States.
This distinction also demonstrates why another of Spiro's arguments against the genuine link test does not withstand scrutiny in the EU context. The genuine link requirement does not make much sense, he argues, in the face of globalisation and ‘proliferating multiple nationalities and the dramatically enhanced capacity for individuals to maintain social connections with multiple countries’.Footnote 102 Multiple nationality being a ‘fact of globalisation’,Footnote 103 a genuine link test would not just be increasingly difficult to administer, it would also radically undermine ‘the reliance value that millions of individuals place in their nationality to facilitate their movement through the globalised world’.Footnote 104 However, while Spiro is right to highlight that the fact of globalisation must caution against conditioning the recognition of nationality on the existence of a genuine link, it is precisely this fact that supports the case for conditioning the granting of nationality on the existence of a genuine link. This is certainly so in highly integrated regions such as the EU, where the unwillingness of States to fashion naturalisation policies with an eye to a genuine link creates the problems of citizenship by investment regimes and also risks depriving individuals with a genuine link of the right to enjoy the full benefits offered by European integration. Thus, although globalisation may have undermined the justification for Nottebohm, it has increased the need for a genuine link requirement as a condition for the grant of nationality.
The third argument Spiro offers against a genuine link requirement concerns its lack of precision and the impossibility of translating a genuine link ‘into a practical standard for the allocation of nationality’.Footnote 105 Quoting Bauböck, who says that such links ‘cannot be measured in a uniform way either as a subjective sense of belonging or through objective indicators such as duration of residence’,Footnote 106 he suggests that even the most ardent proponents for the introduction of genuine link tests admit this. Spiro's position, however, rests upon a misunderstanding. Bauböck is correct to note that residence requirements, while objectively measurable, do not always reflect accurately individuals’ ties to a country. It does not follow, however, that a genuine link cannot be translated into practical standards. The duration of residence is a practical standard for determining whether there is a genuine link, even if it is one that may be over- and under-inclusive at times.Footnote 107
Arguments against introducing a genuine link requirement turn out to be unpersuasive. A genuine link is a normatively desirable condition for the grant of nationality and the introduction of this standard could remedy several of the shortcomings that currently mark the relationship between EU and national citizenship. Unfortunately, implementing such a solution at the EU level is hard to achieve. Not only is there likely to be limited support for it among the Member States, but the EU's division of competences prevent its realisation. Article 5(2) TEU stipulates that ‘[u]nder the principle of conferral, the Union shall act only within the limits of the competences conferred upon it by the Member States in the Treaties to attain the objectives set out therein. Competences not conferred upon the Union in the Treaties remain with the Member States.’ Nationality is not listed among the competences transferred to the EU.Footnote 108 In addition, Article 20 TFEU provides that EU citizenship ‘shall not replace national citizenship’. The conferment and the loss of nationality being a national competence, the EU legislature will not be able to legislate on nationality and introduce a genuine link requirement.Footnote 109 According to established case law, however, ‘that a matter falls within the competence of the Member States does not alter the fact that, in situations covered by European Union law, the national rules concerned must have due regard to the latter’.Footnote 110 Moreover, given that the CJEU has already reshaped the legal norms of EU citizenship, including by placing some fetters on the capacity of Member States to deprive individuals of national citizenship, some believe the Court should expand on these decisions to further restrict national citizenship practices.Footnote 111 It is hard to imagine that it will. The Member States have made it explicit on numerous occasions that it remains for them to decide who their nationals are. It will be difficult for the Court to justify broadening the reach of its case law in order to challenge national rules on nationality under the current Treaty framework.
However, the EU is not entirely impotent. By putting pressure on the Member States and highlighting problems with the current arrangements, national authorities might be persuaded to reflect upon and possibly change their current practices. Whilst not a satisfactory solution, a continuing conversation between the EU and the Member States regarding the design of nationality laws can go some way towards addressing existing problems. This might at least reduce some of the negative side-effects of current investor regimes. The 5th Anti-Money Laundering Directive already defines third-country nationals who apply ‘for residence rights or citizenship in the Member State in exchange of capital transfers, purchase of property or government bonds, or investment in corporate entities in that Member State’ as higher-risk situations.Footnote 112 When transposed by January 2020, Member States are expected to have mechanisms in place to ensure that investor citizenship and residence programmes do not undermine European standards against money-laundering. Yet, ultimately, if it wants to remedy existing shortcomings, the EU should be conferred with competence to adopt minimum norms on access to national and EU citizenship. This need not result in a radical reform of the relationship between EU and national citizenship. This is a proposal for a modest change, even if it is unlikely to be perceived as such by many national actors. Until then, the ability of the EU to guarantee that third-country nationals who have clearly demonstrated their genuine link with their State of residence can enjoy full national and EU citizenship and to prevent Member States from conferring citizenship on individuals lacking such meaningful connections will remain limited.
VII. Conclusion
The current relationship between national and EU citizenship is not without problems. It allows Member States to design their nationality laws with disregard for the interests of other States and the democratic rights of third-country nationals. However, these problems should not be addressed by abandoning or reversing the connection that currently exists between EU and national citizenship. EU citizenship should not acquire primacy over nationality since this would ignore the associative ties between those who have contributed to building and maintaining the collective goods offered by State institutions. EU citizenship should also not become autonomous from national citizenship, because the resulting exclusion of national political institutions from EU decision-making processes would undermine the control EU citizens can exercise over EU decision-making. However, while the criteria of linkage and derivation should remain unaltered, the current rules leave too much discretion to the Member States. To address these concerns, shared minimum European rules concerning access to national and EU citizenship are desirable. The idea of a genuine link is a particularly attractive solution but for the time being this cannot be done as the EU lacks the necessary competences. Member States should therefore consider granting the EU the competence it needs to adopt minimum norms on access to national and EU citizenship.
