Conceptualizing and evaluating (new) forms of citizenship between nationalism and cosmopolitanism

In an age of transnational flows and interdependencies, democratic citizenship can no longer be conceptualized exclusively within national boundaries. This paper presents a conceptual map that allows tracing emerging and proposed forms of citizenship within and beyond the nation state in a comprehensive and differentiated way. We disentangle two anchor points: membership in a political community as the fundament of citizenship and the arena of political decision-making as the focal point of citizenship rights, identities and practices. For the former we differentiate between a single national community, the universal community of humankind and multiple (national) communities. For the latter we distinguish the national, supra-national and transnational arena. Our typology thus consists of nine different forms of democratic citizenship. It is used to provide a brief overview over normative proposals and empirical findings leading to the following insights: while membership in a particular national community still dominates the reality in all three political arenas, in the normative discourse it is perceived as deficient. Membership in the universal community of humankind is widely endorsed in the normative discourse, but almost nonexistent. In contrast, membership in multiple communities is not only a growing reality but also normatively promising for democratizing a transnationalizing world.


Introduction and overview
One of the most widespread phenomena of the last 20 years has been that the increasing volume and speed of flows of capital, goods, information, and people leads to increasing interdependency among nation-states across the globe. 1 These flows and interdependencies endanger the self-determination of national peoples within territorial states. Attempts to regain political control over a globalized economy by supra-national institution-building compromise democratic selfdetermination if the new institutions are not connected to citizens through individual rights, shared identities, and participatory practices. They undermine national democracy if they are not accompanied by new forms of citizenship. 2 Those who believe that it is possible to develop citizenship rights, identities, and participation on a supra-national level often use the term "cosmopolitan democracy". They argue that people belong to a range of polities of which the nation-state is only one. 3 In this vein, many scholars are convinced that "(s)tate citizenship and world citizenship form a continuum whose contours, at least, are already becoming visible". 4 However, a widespread criticism contends that, so far, political cosmopolitanism misses an account of "how social solidarity and public discourse might develop enough in these wider networks to become the basis for active citizenship". 5 Cosmopolitanism might be nothing more than an attitude of "frequent travellers" 6 who profit from the increase in international connections in a very individualistic manner.
However, the normative debate in political philosophy about the principles and adequacy of new forms of citizenship is often disconnected from the empirical research in social sciences about actually emerging forms of citizenship. 7 We want to make a contribution to this discourse by bringing together the normative and the empirical literature and by pointing to promising forms of citizenship that lie between nationalism and cosmopolitanism -the two conceptual poles that dominate the discourse. We structure our contribution with respect to two questions: (a) which forms of citizenship are complementing national citizenship in a transnationalizing world? (b) Are these forms just empty shells without much impact on democratic self-determination or are they filled with citizenship rights, collective identities, and political practices that lay the groundwork for the democratization of political arenas within and beyond the sovereign nation-state?
To provide a systematic overview of current answers to these questions, we develop a two-dimensional typology with three categories in each dimension. The first dimension focusses on the political spaces or arenas to which citizenship rights, practices, and identities are directed. We distinguish three political arenas: (a) the domestic arena within territorially demarcated nation-states, (b) the supranational arena above nation-states; and (c) the transnational arena that emerges as the result of cross-border flows and (inter)dependencies. 8 The second dimension refers to the origins of citizenship and thus to three conceptions of membership: (a) membership in a particular national community; (b) membership in the universal community of humanity; and (c) membership in multiple communities. This leads to a typology with nine distinct forms of citizenship. The typology helps to structure our overview of the conceptualizations and evaluations of old and new forms of citizenship in the literature that forms the main part of the essay. For each form, we try to summarize the literature and to evaluate (a) the extent to which this form is seen as existent or emerging and (b) the extent of its acceptance as normatively adequate.
Overall, we conclude with the insights that forms of citizenship based on membership in a single particular nation still dominate the current reality, but they are seen as deficient with respect to normative standards. In contrast, the concepts based on membership in the universal community of humanity are weak in their actual realization. They often dominate the normative discourse but are challenged there as well. Most importantly, the conceptual map reveals that the forms of citizenship that lie between and are based on membership in multiple communities are not only growing rapidly but also widely accepted in the normative discourse.
Directions, dimensions, and categories for conceptualizing (new) forms of citizenship in a post-Westphalian era In this section, we want to lay the groundwork for the typology that will help us to trace, systematize, and evaluate old and new forms of citizenship. First, we specify our understanding of citizenship and clarify the direction of our discussion of new forms of citizenship. Second, we justify the introduction of two dimensions and three categories in each dimension as conceptual cornerstones of our typology. Finally, we provide a first brief overview of nine types of citizenship and justify the naming of these types.
In the literature on democratic citizenship, it has become common use to distinguish four elements of citizenship: (a) status as a citizen via formal membership in a political community; (b) civil, political, and social rights; (c) participatory practices; and (d) belonging/collective identity. 9 These elements are linked to different political philosophies and theories of democracy. The legal status and the civil and political rights of individuals are emphasized foremost by liberal conceptions of democracy; participation in political decision-making is crucial for republican understandings; and the identification of citizens with collective entities is seen as a necessary precondition of democracy by communitarians. Many citizenship debates have focussed on how much emphasis we should give to the different elements and how rights, practices, and identities relate to each other in terms of legitimacy and causality. 10 Furthermore, feminist and post-colonial critics have challenged the universality claim embodied in Western citizenship theories and stressed the relevance of gender and cultural differences not only for the normatively adequate assignment of formal (minority) rights but also for substantive citizenship, identification, and participation. 11 At the same time, the established understanding of citizenship in the Western world was challenged by the tremendous growth of socio-economic flows across the boundaries of nation-states and the attempts to regain political control through international cooperation and supra-national institution-building. This triggered new conceptualizations of citizenship in which new political arenas as well as new kinds of memberships take centre stage. We will focus on the membership aspect and the questions of who is and who should be included since this element has regained priority in the discursive and practical struggles to redefine citizenship in a post-Westphalian order. This does not mean that we ignore rights, participation, and identity; on the contrary, we will systematically trace these elements when we look at the normative and empirical standing of old and new forms of citizenship. Furthermore, we take two important insights into account: first, the notion of membership is not restricted to formal or legal membership; some concepts of citizenship are based on informal or de facto membership. Second, from feminist and post-colonial approaches, we take up a sensitivity to cultural pluralism, diversity, and hybridity -the latter aspect in particular is looming large in our account since we highlight the recognition of individuals who are members of multiple political communities. 12 Finally, we admit that the following overview has a clear Western bias since we start from the Westphalian concept of citizenship and trace primarily the discourses and realities as they emerged in the Western world and thus in relatively stable democracies. Nevertheless, we will see that new forms of citizenship in the transnational arena link the Western and the non-Western world, and we are convinced that they represent promising avenues for connecting citizens and their understandings and practices of citizenship across the globe.

Dimensions and categories
In the twentieth century, the predominant perception of the relationship between democratic governance and citizenship was that all adult people who live on the territory over which the nation-state rules should be included in the citizenry having individual (civil, political, and social) rights, participating in political decision-making, and fostering the national identity. This Westphalian orderand nationalistic fiction -emphasized two principles: exclusivity and congruence. Both the nation-state as the dominant form of governance and the nation as the dominant form of political community and individual membership claimed exclusivity. 13 The nation-states claimed a monopoly on the legitimate use of violence internally and sovereignty externally. Political membership was based on the principle in international law that "every person should have a nationality and should have one nationality only". 14 Furthermore, a clearly demarcated territory provided a shared basis for delineating the scope of each state's monopoly in rule-making and for defining the boundaries of the national community that forms the corresponding citizenry or demos. In other words, the democratic legitimacy of the Westphalian world order rested on the territorial congruence between the citizenry as the authors of democratic rule-making, the state as the arena of political decision-making and agent for the implementation of democratic regulations, and the inhabitants as the addressees or subjects of democratic rules.
Today, both principles -exclusivity and congruence -are giving way to new principles: multiplicity and contingency. The congruence between authors, arenas, agents, and addressees of democratic decision-making is becoming contingent. Immigration and emigration lead to a growing incongruence between inhabitants and nationals both in migrant-sending and migrant-receiving countries. Also, the emergence of supra-national and trans-national arenas of decision-making is not automatically accompanied by corresponding forms and boundaries of citizenship. Normative concepts such as post-national 15 and cosmopolitan citizenship 16 as well as the "all affected principle" 17 try to regain congruence in a world characterized by migration and multi-layered systems of governance, but both their realization and their normative merits remain doubtful. To capture conceptions and realities of citizenship that are based on the assumption of a congruence between authors, arenas, and addressees and those that are not, we propose to distinguish two fundamental dimensions of citizenship: (a) the "origin of citizenship" dimension, which refers to the (imagined) political community on which the assignment of membership and corresponding rights, practices, and identities rests and (b) the "direction of citizenship" dimension, which refers to the political arenas (and agents) to which citizenship rights, practices, and identities are directed.
The exclusivity of the nation (state) with respect to political membership and decision-making is challenged by two distinct normative principles and empirical transformations: (a) the complementation of particularism by universalism and (b) the complementation of monism by pluralism. These two innovations show up in both dimensions: the domestic arena of decision-making within nationstates is complemented not only by supra-national arenas but also by transnational arenas. Whereas supra-national arenas are defined by formal or de facto delegation of decision-making competencies from particular polities to more universal polities, transnational arenas are characterized by the recognition of the (inter)dependencies among multiple national polities. The same processes can be observed with respect to membership: formal membership in a particular national community is no longer the only form of membership on which citizenship rights, identities, and practices can be based. Human rights are derived directly from the status of being a member of the universal community of humankind. Furthermore, dual citizenship and European citizenship undermine the classic assumption. In consequence, nationality is losing its exclusivity for political membership, being complemented by (more) universal forms as well as by membership in multiple political communities.
The resulting typology Based on these foundations, we can draw a typology for mapping forms of citizenship within, above, and across nation-states. Vertically, we differentiate between three kinds of membership from which citizenship rights, practices, and identities emerge: membership in a particular national community, membership in multiple national communities, and membership in the universal community of humankind. In the horizontal dimension, we distinguish between three political arenas to which citizenship rights, practices, and identities are directed: the domestic arena in the country of (primary) residence, the transnational arena between various (particular) nation-states, and supra-national political arenas.
In each dimension, we have as a first category the classic form that prevailed in the Westphalian order of sovereign nation-states and bounded national communities. Then there are the more radical alternatives to the sovereign nation-state and the particular nation: global arenas of political decision-making and a universal community of humankind. In between, there are further, less radical, alternatives: relevant (since they are interdependent) national arenas outside the domestic political arena and a multiplicity of (particular) memberships and communities. The combination of the two dimensions with three categories in each dimension leads to a typology that includes nine different forms of democratic citizenship ( Figure 1).
The field in the bottom left-hand corner represents the (idealized) Westphalian form of bounded citizenship based on a single national community and directed toward decision-making within the nation-state. However, the concept of membership in a particular national community is not limited to the domestic realm. The ongoing recognition of emigrants as citizens leads to a growing phenomenon that has been called "external citizenship". 18 Both liberal philosophers and realist social scientists conceptualize citizenship in supra-national arenas in a way that we call "mediated citizenship": nationals are seen as the major principals within bounded nation-states; in the inter-or supra-national realm, the nation (state) is the only recognized principal.
The upper row contains three proposed forms of citizenship that imply that membership originates in the universal community of humankind. Whereas the notion of "post-national citizenship" introduced by Soysal 19 indicates the growing rights of immigrants within the domestic realm of liberal nation-states, the term "cosmopolitan citizenship" refers to proposals for individual rights, practices, and identities on supra-national levels. 20 For the transnational realm, we use Koenig-Archibugi's term "fuzzy citizenship" 21 as a promising attempt to conceptualize the notion of "all affected interests" as the basis for delineating the boundaries of the demos and democratic citizenship. 22 In between these alternatives, there are various forms of citizenship, the membership attribution of which is based neither on a (single) national community nor on the universal community of humankind but on membership in multiple communities. In the domestic arena, multiple memberships draw our attention to the fact that resident aliens are members of their country of descent (enjoying external citizenship rights) and simultaneously granted quasi-citizenship in the country of residence. This means that they have a plurality of "partial citizenships". In the supra-national arena, multiple membership implies being a member of national and supra-national political communities simultaneously in the sense of "multilevel citizenship", such as EU citizens who enjoy a supra-national status dependent on citizenship in a member state. In the transnational arena, multiple membership refers to those who are full members in more than one national community, a phenomenon that has been intensively studied and debated under the heading of "dual citizenship".

Normative debates and empirical findings on the divergent forms of citizenship
The divergent forms of citizenship are, to different degrees, normatively recognized and empirically existent. In the following, we provide a brief overview of the debates. We cannot do justice to the much more differentiated discourses in each field, but we try to catch the most important positions and insights. The core aim is to show the similarities and differences across all three kinds of arenas.

Forms of citizenship in domestic national arenas
Particular membership: Westphalian citizenship of all mono-national residents Citizenship based on membership in a particular national community and directed toward the domestic political arena captures the traditional form of citizenship in the Westphalian order with corresponding rights, identities, and practices. This conception is particularly challenged by a growing number of immigrants. At least the rights dimension of this conception is secured when immigrants naturalize while giving up their former citizenship. Migration is often conceived as a directed movement with a point of departure and a point of arrival. Migrants should either assimilate into the new society by abandoning their heritage or try to preserve their ethnic identity to return to the country of origin; but in the end, they are supposed to belong to either one of them. Naturalized immigrants who give up former rights in the country of descent are examples of the classic ideal of political integration into a new homeland.
This ideal-typical conception of integration, of course, still exists and is important. However, a change in status does not necessarily bring about a similarly clear transformation of identities and activities, as research on immigrant transnationalism has shown. 23 This research underlines that migration is nowadays increasingly turning into an ongoing movement between two or more social spaces. The recognition that some migrants maintain strong, enduring ties to their homelands even as they are incorporated into their countries of residence calls into question the conventional assumptions. We witness instead the passage of many people from a national to a transnational condition, "a growing number of persons who live dual lives: speaking two languages, having homes in two countries, and making a living through continuous regular contacts across national borders". 24 This development is not uniformly welcomed. Some scholars -based on communitarian, civic republican, or liberal nationalist reasoning -fear the devaluation and decline of citizenship as an institution and practice because migration and especially on-going transnational ties might undermine the cultural distinctiveness of and solidarity within national communities. 25 They argue that rights are developed "from shared conceptions of social goods; they are local and particular in character". 26 Exclusive membership within and attachment to a specific bounded community are regarded as necessary for democracy, a crucial aspect of a community's self-understanding and self-determination. Therefore, in practice, boundary drawing toward immigrants is judged as legitimate. 27 However, this stance is very contested by those proposing a universal kind of citizenship.

Universal membership: post-national citizenship of all inhabitants
In contrast to the traditional approach to citizenship, several political philosophers forcefully argue that rules for immigration and naturalization have to be seen as a matter of global justice and individual rights. This is based on the conviction that "the status of ultimate unit of concern attaches to every living human being equally", 28 not to any kind of community. Therefore, boundary regimes should be radically revised toward more openness to concur with the universalistic principles of liberalism. 29 This would involve not only open borders but also the equal and easily attained right to citizenship for all inhabitants.
There is also a lively debate about the empirical question whether and, if so, the extent to which we are witnessing the emergence of a "post-national citizenship" within national arenas. Since the 1990s, this idea has become fashionable among scholars who argue that rights that were once derived from formal membership in a particular national community have become increasingly abstract, defined, and legitimated with reference to universal membership in the community of humankind. 30 According to Soysal's influential analysis, the increasing codification and elaboration of human rights led to a situation in which the status of permanent resident immigrants in Europe today is not easily distinguishable from formal citizenship status: "Permanent residents of European host polities are entitled to full civil rights and have access to a set of social services and economic rights almost identical to those available to national citizens". 31 This post-national citizenship would be a new model of membership anchored in deterritorialized notions of personal rights and universal personhood.
However, recently, the conception of post-national citizenship has been strongly criticized as misrepresenting reality. First, from a normative point of view, the forms of quasi-citizenship that are granted to permanent residents are seen as deficient. Whereas socio-economic rights are granted, political rights usually are not. Only a few states provide electoral rights in local elections 32 and even fewer in national ones. 33 Empirical evidence additionally suggests that non-citizens use much less the limited political rights that all residents have. 34 Therefore, critics speak about "denizenship" to indicate the reduced set of citizenship rights of permanent residents. 35 This status may even perpetuate inequalities in political influence by satisfying socio-economic concerns while at the same time depriving non-citizen residents of political self-determination and effective participation. 36 Furthermore, as soon as immigrants leave their country of residence, they do not have the right to claim diplomatic protection, and even within the country of residence they are never free of the threat of expulsion as long as they are not full citizens.
Second, with respect to causal analysis, the term "post-national" seems misleading. Hansen argues that the force of human rights and treaties would depend largely on domestic courts. 37 They would be effective only when incorporated into domestic legislation and into domestic courts' jurisprudential frame of reference. He therefore claims that post-nationalists would confuse a transfer among the institutions of the state, for example from the executive to the judiciary, with a transfer from the national to the supra-national arena.
A final and important argument against the notion of post-national citizenship is the fact that not all immigrants are treated equally. Instead, many forms of differentiation or discrimination based on nationality prevail; in other words, the rights of immigrants are not only reversible but also stratified. 38 Certain groups of migrants are more privileged than others, such as legal permanent residents, political refugees, and nationals from common-market countries or from countries with special ties (such as the EU or the British Commonwealth). The differences between groups of migrants reveal that the basic principle for granting citizenship rights is not universal personhood. Instead, the scope of citizenship rights of migrants depends on a plurality of memberships: the de facto membership in the country of residence and the de jure membership in the country of descent.
Multiple memberships: diverse partial citizenships of migrants In light of the empirical shortcomings of the concept of post-national citizenship and because the traditional exclusive conception of national citizenship is normatively unacceptable in an "age of migration", Bauböck argues that, especially for migrants, we should consider "citizenship constellations" as structures "in which individuals are simultaneously linked to several political entities, so that their legal rights and duties are determined not only by one political authority, but by several". 39 In our normative and causal analysis of emerging forms of citizenship, we should take into account that the citizenship package of migrants resembles a patchwork influenced by the quasi-citizenship that the country of (primary) residence grants and by the external citizenship provided by the country of descent. While quasi-citizenship is empirically widespread but normatively contested as just described -and the same is true for external citizenship, as shown belowthe recognition that the citizenship package of migrants comprises two partial citizenships influences their normative assessment. 40 Limited forms of citizenship that immigrants receive in their country of residence might be judged depending on the extent of external citizenship granted by their home country. 41 It seems furthermore necessary to add a dynamic perspective to the judgement when a citizenship constellation is deficient. Again, Bauböck offers a promising solution: based on his conception of "stakeholder democracy", he underlines that "biographical subjection" should be the guiding criterion for the attribution of membership and citizenship rights. "(O)nly those persons have a claim to citizenship in a particular political community who (. . .) have been subjected to that community's political authorities for a significant period over the course of their lives." 42 Thus, the extent of quasi-citizenship in the country of residence should grow with time of residence (and subjection to law), while external citizenship granted by the country of descent should decline over time if the migrant does not move back. We think that this proposal leads the way to a better understanding and a normatively adequate conceptualization of emerging forms of citizenship between a monistic conception of membership and the universalistic ideals that are implied by the notion of post-national citizenship. This way is characterized by pluralism instead of particularism or universalism.
Overall, we can summarize that, even if the Westphalian conception of citizenship based on membership in one particular national community and directed toward the domestic national political arena is still dominant for most citizens, there are empirically growing deviations to this norm in an "age of migration" as well as normative concerns about its consequences for immigrants. In contrast, the idea of post-national citizenship of all inhabitants based on personhood instead of nationality dominates normative aspirations, but its existence and extent is disputed in light of its deficiencies compared to full citizenship as well as its variations depending on the specific countries involved. In consequence, we have a growing empirical situation of various constellations of partial citizenships of migrants based on different de jure and de facto memberships.

Particular membership: citizenship mediated by nation-states
The traditional approach to politics in supra-national arenas is that nation-states do politics amongst themselves. In particular, the realist school in International Relations emphasizes the interests and power of states in world politics while denying any decisive influence of individual citizens or their organizations. 43 All rights, practices, identities and interests of individuals are aggregated and mediated by nation-states; we thus speak of "mediated citizenship" in this arena.
However, the theoretical exclusivity of state-to-state relations seems inadequate in the twenty-first century. Alternative approaches to international relations are on the rise, emphasizing trans-state institutions and a growing international community. 44 The major thrust of the debate today is the question of whether in this "new world order" 45 there is an international society that sets limits on the behaviour of states and sets the agendas of international institutions 46 and whether a cosmopolitan democracy should be established in which individuals are assigned a major role. 47 Indeed, there is evidence that domestic political choices are becoming more and more restricted by international agreements and international law, as had already been highlighted in the 1990s in the literature on international regimes. 48 Today, the world operates according to understandings that have severely modified previous norms. This is mainly due to the ideal and practice of human rights claiming universal validity. These rights have become more and more relevant since their declaration on a global level in 1948. 49 Based on the recognition of the equal dignity of every individual, they only recently led to the establishment of the International Court on Crimes against Humanity. The same logic of universal norms superseding national sovereignty is also guiding humanitarian interventions as answers to the behaviour of nation-states toward their citizens and residents. 50 In addition, the United Nations and a panoply of international treaties are the first signs of a global rule of law and a constitutional framework that establishes step by step civic, socio-economic, and political rights of a global citizenry. 51 States might still be the major players on the supra-national level, but they are no longer the only ones. 52

Universal membership: cosmopolitan citizenship of all humans
In contrast to the traditional approach, cosmopolitans envision that individuals as members of the community of humankind are the principal agents in supra-national governance. This is founded in the normative philosophy of cosmopolitanism urging us all to be "citizens of the world", committed to universal values -even if they are often portrayed from a Western European perspective. 53 The corresponding political programme aims to establish a cosmopolitan democracy: a global layer of governance that limits the sovereignty of states. 54 However, it is widely disputed whether democratic citizenship can ever operate in a meaningful way above the national level. 55 Dahl, for example, is convinced that an international organization "cannot be a democracy" 56 because of its scale, remoteness from people's lives, the complexity of issues, and the diverse contexts that would make common interests elusive at best. Sceptic voices are particularly pronounced when emphasis is laid on political participation and identification. Communitarians insist that "humanity" is too thin an identity to motivate mobilization, participation, or solidarity on its behalf. 57 Cohen, for example, points to "the abstractness of cosmopolitan individualism, its failure to take particular identities, (political and cultural) contexts, and traditions into account". 58 Another counterargument refers to the republican ideal of collective autonomy and the citizen who actively takes part in politics and political debates. 59 The enactment of this ideal could be best achieved within national communities in which citizens share a common nationality, solidarity, and trust sustaining deliberative procedures and enabling the expression of collective autonomy. (Neo-)republicans fear that any kind of global hegemony quickly becomes imperial domination. 60 Empirical information about the current realities provides a differentiated picture. On the one hand, we can meanwhile detect various instances of supranational rights in the cosmopolitan sense, as mentioned above. On the other hand, we have seen only a very limited transfer of formal governance competencies to the supra-national level. Whereas supra-national governance is mostly restricted to agenda-setting and state-binding decision-making, citizen-binding law-making as well as implementation and enforcement remain squarely under the helm of the nation-state. With the sole exception of the International Criminal Court, no major institutional reform has occurred since the end of the Cold War. 61 Furthermore, until now, individuals have not been granted a formal role as world citizens with individual political or even electoral rights on a global scale. Also, other political freedoms such as the right to demonstrate or to associate have to be secured by national authorities.
There is even less and sometimes contradictory empirical evidence concerning the question of whether the phenomenon of cosmopolitanism is also endowed with cosmopolitan identity and participation. Norris, for example, found that local and national identities remain far stronger than any cosmopolitan orientation. Only one-sixth of the public (15%) feels close to their continent or "the world as a whole" in their primary identity. 62 In the European Union, considerably more people, namely almost 30%, feel intensely that they are world citizens, but there is ambiguous evidence on broader cosmopolitan attitudes. 63 The empirical evidence on political participation in supra-national arenas is also ambiguous. On the one hand, scholars have noted impressive growth in the number and size of global social movements, networks, and nongovernmental organizations (NGOs). 64 On the other hand, so far, their participation in global governance has been limited to a mere advocacy role, deprived of any decision-making power. Those who are actually active are better educated than most of their compatriots, better connected, speak more languages, and travel more often 65 -they are those elitist cosmopolitans that others criticize. 66 Thus, overall, a sceptical account of cosmopolitan rights, identity, and participation remains.

Multiple memberships: multilevel citizenship of all citizens of member states
The empirical shortcomings of cosmopolitan citizenship and the normative and empirical deficiencies of mediated citizenship again highlight the need to look for alternatives between them. We find considerable potential in "multilevel citizenship" based on membership in multiple (bounded) communities and directed toward supra-national arenas. Outstanding examples are the rights, practices, and identities associated with membership in the European Union. The EU is a site in which a bounded supra-national form of citizenship is in the making. Legally, the Treaty of Maastricht introduced European citizenship, an explicit supra-national form of citizenship derived from membership in a member state. 67 EU citizens enjoy today more supra-national citizenship rights than citizens in any other world region or third country nationals within the EU. In addition to the right to free movement and residence and to the right to consular protection abroad by any member state, EU citizens also enjoy political rights: the right to vote and stand in local and European elections in any member state and the right to petition the European Parliament and Ombudsman. EU citizens are, as individuals, directly represented at the Union level in the European Parliament, which in recent decades has considerably gained influence. These rights are thus a specific combination of national and supra-national membership.
Whether supra-national rights in a specific regional integration project are also connected to the other dimensions of democratic citizenship, namely identity and participation, is a largely open empirical question. In 2009, three-quarters of citizens in the 27 member states felt at least somewhat European, in addition to their national identities. 68 However, only a third feel strongly European, and it is disputed whether the existing sense of community is a sound basis for solidarity, legitimacy, and democracy in the EU. 69 Furthermore, there is a lively discourse on widespread Euroscepticism 70 and low participation rates in EU politics. 71 European elections are often even portrayed as second-order national elections. 72 Overall, we can summarize again that the traditional conception of citizenship in supra-national arenas, which we called "mediated citizenship", is still dominant in world politics, but there are growing empirical deviations, and its dominance is normatively challenged. The opposite "cosmopolitan citizenship" has gained a growing number of adherents, but evidence for its empirical existence is inconclusive and rather minimal. The conception in between, "multilevel citizenship", might (still) be exceptional within the EU, where it is furthest advanced. Still, its existence cannot be denied, and since it is normatively widely accepted, it seems especially promising to further investigate the legitimacy and developments of this new form of citizenship "in between".

Forms of citizenship in transnational arenas Particular membership: external citizenship for all national non-residents
The phenomenon of "external citizenship" is coherent with the logic of membership in a particular national community, but it transcends the boundaries of nation-states. Most countries allow emigrants and their descendants to keep the nationality and related rights. In contrast to earlier times (and still present in many non-democratic countries), the current trend is not the result of the authoritarian concept of "perpetual allegiance". 73 However, the acceptance of external citizenship still follows an ethnic-nationalistic understanding of citizenship based on "natural affiliations". 74 In addition, the new appreciation of emigrants is driven by an attempt to secure their economic contribution, their financial remittances, or investments. Migrants are seen as highly valuable socio-economic agents and as political lobbies working in the sending countries' interests. 75 This appreciation led to the expansion of external citizenship either by enlarging the group of individuals eligible for receiving or keeping the citizenship status, by providing some citizenship rights to specific external kin-groups, or by strengthening the practices of external citizenship. 76 In 2007, more than half of all countries and almost all democracies allowed their nationals abroad to cast their vote. Additionally, many countries have facilitated the practice of external voting; some even grant their external nationals special representation in parliament. 77 The debate on the consequences of migration for citizenship has been preoccupied by the challenges of immigration (at least in the Western world). The consequences of emigration and the expansion of "external citizenship" have been politically much less controversial. 78 This is, however, different in the normative theoretical literature. In democratic theory, such rights are widely contested since the right-holders are rarely directly concerned by the effects of their votes. From the point of view of liberal democracy, there seems to be no justification for granting citizenship status and rights to non-residents who are not subjected to the government and the laws of their country of descent; even worse, external citizenship seems to tilt citizenship toward claiming rights without fulfilling obligations. 79 There is, however, another form of external citizenship that is highly appreciated and prominently discussed in democratic theory, to which we turn in the following.
Universal membership: fuzzy citizenship of all affected Based on a universal conception of membership and directed toward one or several specific political arenas outside the country of residence is the idea that all affected individuals should have the right to participate in countries that produce external effects. Following Koenig-Archibugi, we call this concept "fuzzy citizenship". 80 There is a lively debate in democratic theory on whether the liberal principle of inclusion has to be adjusted in a transnationalizing world of cross-border (inter)dependencies and spill-overs by exchanging the concept of "subject to political rule" by "affected by political rule". 81 Robert Dahl proposed that "[e]veryone who is affected by the decision of a government should have the right to participate in that government" 82 and famously asked "whether there is not some wisdom in the half serious comment of a friend in Latin America who said that his people should be allowed to participate in our elections, for what happens in the politics of the United States is bound to have profound consequences for his country". 83 In recent years, many theorists have started to take this comment more seriously and discussed and proposed divergent ways for the inclusion or representation of affected externals in national politics. However, there are quite different propositions to define who is concretely entitled to political rights. Held suggests all those "significantly affected" 84 by political decisions; Warren even proposes including every individual "potentially affected" 85 by a collective decision. These different quotes show the difficulty of specifying exactly the "all affected" principle. It leads to either an infinite regress since we only know after a decision is made who is actually affected by it, 86 or to a ballooning franchise, including "(v)irtually (maybe literally) everyone in the world". 87 The logical conclusion that all decisions should be made on a global level based on a global demos seems neither feasible nor desirable.
We do not find any evidence that this principle has been implemented in reality. The principled and practical problems of the "all affected principle" again highlight the need to search for promising solutions between particularistic and universalistic forms of citizenship that address the challenge of increasing transnational (inter)dependencies in a realistic and normatively adequate way.

Multiple memberships: dual citizenship of all people with multiple affiliations
Empirically, we witness a strong proliferation of dual citizenship. A century ago, most countries viewed dual citizenship as bigamy; today, an increasing number of states permit and even promote it. 88 Pressure comes at times from international conventions and at other times from domestic policy, economic interests, or emigrant mobilization. 89 For evidence on the question of whether the status of dual citizenship and the corresponding rights in two countries also go hand in hand with multiple identities and participation in more than one country, we have to refer to the broader research on immigrant transnationalism. However, in spite of the recent surge in interest in these phenomena, generalizable empirical findings are rare. Most research has centred on the United States as the country of reception. 90 There are only a few studies in Europe. 91 Most scholars conclude that a stable and significant transnational field of political action connecting immigrants with their countries of origin indeed exists. The extent and forms of transnational loyalties and activities vary with contexts of exit -conflictive or peaceful, urban or rural -as well as with the conditions of reception -incorporation into the host society, the extent of discrimination, and settlement patterns. 92 Dual citizenship is not a necessary condition for transnational activities and identities, but they are facilitated in important ways by formal membership in both national communities. Empirical studies have to further advance our knowledge regarding the scope, conditions, and consequences of such multiple citizenship rights.
In normative debates, this development is contested, but we can discern a growing acceptance. While some diagnose or fear the devaluation of citizenship, 93 others welcome dual citizenship as a practical first step to reduce the gap between those who are affected by policy-and law-making in powerful states and those who are entitled to vote in these countries. 94 Dual citizenship can be conceived as a promising element for democratizing transnational arenas in a world that de facto resembles a system of "imperial" relationships among nation-states. This system is characterized by asymmetric flows of (military, political, economic, and cultural) influence from powerful "core states" to less powerful "peripheral states" and corresponding "counter flows" of migrants from peripheries to core countries largely caused by the uneven development resulting from the asymmetries of influence. While Koenig-Archibugi therefore proposes that powerful states should invite representatives of those who live abroad and are affected, 95 it can also be argued that, with migrants, there are already individuals within the territorial confines of influential nation-states who might represent the affected externals. 96 However, to do so effectively, they need the full set of citizenship rights of the core and the peripheral state because this signifies and strengthens on-going ties to the external community. As long as migrant-receiving countries do not grant non-national residents the right to vote, naturalization is an important condition for the effective representation of affected externals by migrants. In addition, the recognition of "also-others" as equal members in the country of residence might signal and stimulate an awareness of transnational (inter)dependencies and external effects. 97 Involving two particular and bounded territorial sites of solidarity and identity formation, this phenomenon is also less vulnerable to critiques of abstractness, hollowness, and instability that are often advanced against universalistic propositions. 98 Dual or multiple citizenship is not the ideal and non-contradictory new form of citizenship but a pragmatic tool that addresses one of the most salient democratic deficits in a world of cross-border flows. In addition, it potentially has positive effects in the domestic and in the supra-national arena as well. First, removing a major motivational hurdle for immigrants to naturalize helps to reduce in national arenas the current incongruence between those subjected to law and those entitled to participate in making these laws. Permitting multiple citizenships thus probably enhances political integration in the country of residence, 99 expands the "training ground" in which people learn transferable political skills, 100 and forms the basis of (trans-)national solidarity. Second, it is likely (but not yet proven empirically) that multiple citizenship provides its holders with the kind of cosmopolitan attitude that the proponents of supra-national citizenship are hoping for. The recognition and practice of dual citizenship might help in overcoming exclusive forms of identity, solidarity, and interest formation. Those who realize that memberships, identities, and practices can be multiplied on a horizontal level might be more open to similar multiplications in a vertical dimension.
Summing up our discussion on transnational arenas, we find that external citizenship is empirically widespread but normatively disputed, while the opposite is true for the citizenship of "all affected". At the same time, we witness a strongly growing extent and acceptance of dual citizenship with considerable potential. To get into a more thorough debate about these potentials, we need further empirical information. Nevertheless, the scrutinized lines of argumentation indicate that dual citizenship potentially plays an important role in all political arenas. Our conceptual map visualizes this centrality of dual citizenship in the post-Westphalian world, situating it right in the middle. Figure 2 takes up the systematized conceptualization from Figure 1 while adding the results of our brief overview based on the empirical findings and normative debates on all nine forms of citizenship.

Summary and conclusion
Globalization, transnational flows, and (inter)dependencies present modern democratic theory with the task of reconciling the principle of "rule by the people" with a world in which power is exercised increasingly in supra-national or transnational arenas and where there is growing incongruence between inhabitants and citizens as well as between those affected by law and those subjected to it. Today, the fate of democratic communities is increasingly interwoven. This paper therefore concentrated on the question how we can conceive of the demos beyond the nation-state, as it is ingrained in citizenship rights, identities, and practices.
Our conceptual map allows for a systematic and nuanced understanding of emerging and proposed forms of democratic citizenship within and beyond the nation-state. We argued that it is important to disentangle two anchor points: the origin of citizenship based on membership in various political communities as its fundament on the one hand and the direction of citizenship, thus the diverse arenas of political decision-making as the focal point of citizenship rights, identities, and practices, on the other hand. Furthermore, we emphasized two distinct alternatives to the exclusivity of the Westphalian system of sovereign states and membership in a single political community: the exclusive community of the nation cannot only be overcome -or complemented -by the universal community of humankind but also by a multiplicity of overlapping national communities; in parallel, the national political arena is complemented not only by supra-national arenas but also by transnational ones. Consequently, combining three categories in each of the two dimensions leads to a typology of nine spaces of citizenship within, across, and beyond the nation-state.
This typology has been used to connect philosophical-normative discourses with the empirical-analytic literature. Brief overviews of both kinds of literature revealed that understandings of citizenship that are based on membership in a (single/particular) national polity are strongly questioned in the normative literature; proposals that are based on membership in the universal community of humankind, in contrast, remain severely limited and seem unrealistic for the foreseeable future. For each political arena, we therefore pointed to empirically realistic and normatively promising alternatives in between: taking quasi-and external citizenship adequately into account in national arenas, combining national and supranational citizenship rights, identities, and practices in supra-national arenas, and accepting and acknowledging dual citizenship in transnational arenas. For those who adhere to universalistic principles, these forms of citizenship might be promising stop-overs; for those who are rather sceptical in this respect, they might represent valuable destinations in themselves. 101 Furthermore, the conceptual map highlights the role of the transnational arena as a space "in-between" the domestic and the supra-national realm. The theoretical and normative debates so far have focussed foremost on the latter, with major disputes about the extent to which cosmopolitan and post-national citizenship exist and whether they are feasible and desirable. Transnational arenas have been discovered rather recently. In light of its spread and growing acceptance, dual citizenship might be a promising pathway for the democratization of transnational relations as long as more tailor-made proposals seem unrealistic.
Finally, our differentiated map lays the conceptual groundwork for more systematic discussions and empirical research about the relationships between different forms of citizenship. This concerns in particular a more thorough understanding of the relationship between transnational and cosmopolitan forms of citizenship. For those who are searching for "thick" citizenship beyond the single nationstate, it might be promising to take a closer look at membership in multiple communities. Dual or multilevel citizens who have political rights in as well as personal ties to more than one country or level might increase the number of "cosmopolitan patriots" 102 through celebrating different ways of being and expanding solidarity with several particular communities. If we want to evaluate the potential of new forms of citizenship for a vitalization of democracy transcending national borders, we need to be clear about the concepts we apply. Our typology proposes to capture systematically the various forms of citizenship that complement national citizenship. However, further empirical knowledge on these forms is necessary to judge their role in an emerging post-Westphalian world. We hope that many colleagues will join us in conducting comparative empirical studies to understand how multiple citizenship statuses influence political identification and practices and how transnationalism influences cosmopolitanism and vice versa. 103