Our new GLOBALCIT website is under construction. In the meantime, please use the current website as before.

Citizenship Forum

Forum Debates

Freedom of movement under attack: Is it worth defending as the core of EU citizenship?


Freedom of movement under attack: Is it worth defending as the core of EU citizenship?

By Floris de Witte (London School of Economics and Political Science)


Freedom of movement is under attack from different sides. It is under attack politically in different Member States due to its alleged effect on the sustainability of the welfare state; it is under attack legally by the CJEU’s retrenchment of the rights of the poorest of Europe’s citizens; and it is under attack conceptually by those scholars and politicians who wish to understand EU citizenship to be primarily about the connection between all Member State nationals and the EU rather than focusing on the rights of mobile citizens alone. In all these accounts, the main fault line that seems to be emerging is that between mobile and immobile citizens in the EU – a fault line that the EU struggles to internalise politically and that can be traced back directly to the right to free movement. 

Is there any reason to defend free movement as the core of EU citizenship? I think that there is more than one. Below, I will argue that EU citizenship should be primarily about free movement as a) it emancipates the individual from the nation state; b) it serves to recalibrate questions of justice and democracy in a more appropriate manner; and c) it lacks the ties to a homogenous political ‘community of fate’ that perpetuate significant exclusionary practices. For these reasons, free movement is the central thing that EU citizenship should be about: it is what makes EU citizenship distinctive from, and genuinely supplementary to, national citizenship.   

Free movement as emancipation

Free movement is often understood in terms of its economic costs and benefits to the Member States of the EU. But we see something very different when we change the lens through which we look at free movement from one that is preoccupied with its effect on states to one that looks at its effect on the individual. From the latter perspective, freedom of movement is primarily about exactly that: the freedom to move out of one’s own state and to choose a different type of life in a different type of place. Thus understood, free movement is an emancipatory force. It allows individuals to live their lives unencumbered by the limits that their place of birth imposes on them, and freedom of movement allows them to understand themselves (and the possible realisations of that self) in much more authentic terms. 

This freedom of movement allows an LGBT+ couple that lives in a country in which the legal, political, cultural or social conditions do not allow for meaningful recognition of their love to move to a more permissive environment. It allows a retired teacher from Middlesbrough to enjoy her pension in sunny Lanzarote, and it allows a Romanian IT-consultant to move to Lille to live with his Hungarian girlfriend who works as a nurse in Belgium. Freedom of movement allows Europe’s citizens to move for love, work, family, language, social or cultural reasons, or simply to be somewhere ‘else’. It is about liberating the individual from the possibilities, opportunities, prejudices, cultural and social norms or convention (or even weather) that exists in their ‘own’ country, and about making available realisations of life in other states that might much more closely fit with the individual’s own preferences. To turn this around, it also means limiting the capacity of states to force the individual to live her life in a particular fashion.

This emancipatory potential of free movement is not only realized through actual movement. It also has a reflexive virtue: it orients the individual’s visions of self-realisation and self-understanding outwards. The possibility of free movement allows for many different realisations and understandings of the self that may have been unavailable but for free movement. Freedom of movement, in other words, liberates not only the body but also the mind from the normative structures of the state. 

Free movement, as such, is to be defended normatively as it problematizes the domination that the nation state exerts over our choices, self-understanding and images of self-realisation. To put it as bluntly as possible, the nation state’s mode of social integration reduces the incredibly complex individual to a one-dimensional being: a national. We all have many meaningful relationships and ties of identification with different groups in society, based on our profession, sexual orientation, ethnicity, religion, residence, language group, hobbies, or sharing of certain social or cultural preferences (a football team, a mode of transport, a type of music, cuisine or mode of living). The nation state, however, essentially tells us that while those relationships and patterns of identification may matter to us privately, the only one that matters for us as public individuals is that of nationality. It is with nationals, after all, that we have to share our resources and that we have to discuss what is allowed or not in society. And it is the nation-state that can coerce us into (not) taking particular actions, that can criminalise certain behaviours, that can trivialise certain needs or that can prevent certain aspirations. As Amartya Sen explains, this “increasing tendency towards seeing people in terms of one dominant ‘identity’ (…) is not only an imposition of an external and arbitrary priority, but also the denial of an important liberty of a person who can decide on their respective loyalties to different groups [1]”.  

The first reason why freedom of movement ought to be defended as the core of EU citizenship, then, is that it enhances our capacity to understand ourselves and realise ourselves in a more authentic and genuine fashion. 

Free movement as a recalibration of justice and democracy

The second reason why free movement ought to be defended as the core of the relationship between the individual and the EU is because it makes us sensitive to practices of exclusion. The construction of EU citizenship, in particularly within the context of the rights to free movement and non-discrimination, has the potential to lead to more inclusive ways of thinking about what freedom, justice, equality and participation should mean in the EU. It also has, however, the potential to lead to more practices of exclusion. The fact that EU citizenship and free movement are not embedded in a sufficiently sophisticated, responsive and democratic institutional structure makes it very difficult for the EU to mediate the social conflict that practices of inclusion and exclusion produce, and to legitimise the choices made [2].

There are many different ways to approach and address these issues. In very general terms, the right to free movement and non-discrimination attached to EU citizenship can be understood to correct instances of injustice and promote the inclusion of outsiders: it makes national distributive systems sensitive to the need to incorporate EU migrants who contribute to the host state in an economic and social way. The Court’s case law, and its criteria of ‘a certain degree of integration’ or ‘real link to the host state society’ can be understood as mechanisms that serve to identify which migrants should have a right to access redistributive practices in the host state on account of the fact that they meet the conditions of reciprocity the sustain those welfare benefits. 

I will not here discuss precisely how EU law attempts to balance the incorporation of outsiders in domestic practices of sharing with the need to sustain the reciprocal or solidaristic nature of those practices (which presume that access is bounded). The wider point that I am trying to make is that free movement makes us sensitive to the structural processes of exclusion that the nation state perpetuates, and serves as an instrument to problematize these processes. Here, instead, I will touch very briefly on two of the most topical ways in which contemporary understandings of free movement and EU citizenship can be understood to produce instances of exclusion – which suggest that there is a need to defend free movement as the heart of EU citizenship. 

The first example is the ‘emergency brake’ that the UK has managed to secure in its renegotiation on the terms of its EU membership [3].  This should eventually allow for the exclusion of EU migrant workers from in-work benefits for (at most) the first four years of their presence in the UK. In the UK, this has been presented as an exercise in justice: it ought to create more opportunities for nationals on the job market, and to prevent payments from the public purse to individuals who have not sufficiently contributed to that same public purse. This argument has been accepted by the heads of state of the other Member States and the Commission despite the absence of empirical corroboration. In fact, the most elaborate studies suggest that the fiscal effects of free movement on the UK are probably positive, and certainly neutral at worst [4].  What we see here, then, is the problem if we understand freedom of movement as a luxury rather than an individual right at the heart of EU citizenship: it is prone to scapegoating and politicking, which are the exact forces that it is meant to combat. This is not to say that free movement cannot create pressures that produce exclusionary effects for national citizens (and which EU law ought to be sensitive to). It seems to me, however, first, that those pressures are primarily infrastructural (which cannot be scaled up sufficiently quick to accommodate access for all) and not of a financial nature, and second, that EU law’s understanding of the limits to free movement and non-discrimination offer sufficient guarantees to prevent such practices. The compatibility of the ‘emergency brake’ with the right of free movement is likely to be tested if the UK votes to remain in the EU, and we could place our fate in the Court to protect free movement and non-discrimination as being at the heart of the relationship between the individual and the EU. 

Unfortunately, it appears that the Court itself is not convinced of this. The recent Dano case offers a good example of how the Court is increasingly turning its back on understanding free movement to be a right attached to the ‘fundamental status’ of every EU citizen. In that case, the Court suggested that the right to basic social assistance mechanisms (as a corollary of the right to equal treatment tied to residence in a host state) is unavailable for those citizens who do not have ‘sufficient resources’ to take care of themselves. In a ruling that comes quite close to depicting Ms. Dano in racist terms as a citizen whose presence in Germany is of no functional use to German society, the Court changes the category of EU citizens that can realistically make use of the promise of free movement. In simple terms, Dano suggests that free movement is not for all Europeans. It is not a right attached to the ‘fundamental status of all EU citizens’, but rather a privilege that European playboys are allowed to make use of. Again, this judgment was celebrated throughout Europe as bringing about justice; as defending the welfare systems against the parasite that is the poor (or poorly-educated) fellow European. Instead, I would argue that it is about the perpetuation of exclusion of vulnerable citizens from the processes that serve to remedy those very vulnerabilities. It is a judgment that legally mandates the creation of a European underclass of vulnerable citizens who, because of their exercise of free movement, are neither politically represented nor materially protected from the most egregious forms of exclusion. This case shows why we need to defend free movement as a right at the core of EU citizenship: something that ought to be available under similar conditions for all nationals of the Member States, and not only for the privileged ones. 

Free movement as separating ‘the nation’ from ‘the state’

The third and final reason why we ought to defend free movement at the core of EU citizenship is because of the latter’s idiosyncratic structure. Unlike national forms of citizenship, EU citizenship is not linked to a ‘community of fate’ that reflects certain ethno-cultural ideas of a homogenous community, forged on the basis of a shared language, history, myths and ethnicity, and solidified through boundary closure, narrowly-defined membership groups and exclusion of outsiders. EU citizenship, instead, is a ‘stateless’ or ‘anchorless’ idea of belonging and community: it suggests that its subjects are part of something that is incipient, ill-defined, and diverse. Often, this is understood as the main weakness or source of illegitimacy of EU citizenship. I would argue that it is exactly its strength. 

The absence of a link between the institutional idea of EU citizenship and a specified ‘ethnos’ or the idea of a ‘nation’ is exactly what makes EU citizenship normatively appealing. Accounts of the ‘long history’ of European integration suggest that the inter-war experience and the Second World War identified the problems with parliamentary or national sovereignty. Very simply put: democracies premised on these ideals appeared not to be very good at remaining democratic. On this account, the creation of the EU was deliberately meant to constrain democratic externalities [5],  and particularly the capacity of states to enforce practices of internal exclusion or external aggression. In other words, EU law serves to foreclose the capacity of domestic democratic actors to commit democratic suicide. Usefully, this narrative proved appealing for Member States that acceded to the EU in the aftermath of periods of totalitarianism. This project of depoliticisation was massively helped by the role of law in the integration process. The scholarship on ‘integration through law’ suggests that law is both the agent and object of integration, and is used to push through the objectives of integration even in the presence of political objection on the national or supranational level. 

What has all of this to do with free movement and EU citizenship, though? Free movement is at the core of the objective of constrained democracy. The legally enforceable right to enter and exit spaces of state authority and the legally enforceable right to equal consideration in whichever space an individual finds him or herself, go a long way towards limiting the power of the state to internally exclude certain groups or antagonise their neighbours. It is free movement, in a sense, which disciplines the nation state, and ensures that its civic institutional structure does not fall in the traps of the ethnos within which it historically grew. In that sense, our ‘anchorless’ EU citizenship is the perfect institutional container for a new – less ethnic – way of thinking about the role of the individual in the EU [6].  And free movement is how this virtue is implemented. The third and final argument in defence of understanding free movement to be at the conceptual heart of EU citizenship, then, is that free movement is the perfect instrument for the implementation of the core normative promise of EU citizenship. 


The Treaty suggests that EU citizenship is to be ‘additional to’ national citizenship. This contribution has argued that the added value that EU citizenship can offer primarily lies in its connection to freedom of movement. Freedom of movement, on this view, is an instrument that liberates the individual’s mind and body from the domination that the nation state exerts over it; that reorients domestic processes of justice and democracy towards more inclusive practices; and that institutionalises an idea of civic belonging on a continent that has been plagued for a century by the consequences of ethnic ideas of belonging. For these reasons, free movement must be celebrated and defended as the core of EU citizenship, as a right that is available for all 500 million EU citizens, and as an idea that benefits all those citizens – whether they make use of it or not. 

[1] A. Sen, The Idea of Justice (Penguin 2010) 247.

[2] See, generally, F. De Witte, Justice in the EU: The Emergence of Transnational Solidarity (OUP 2015) 22-37.

[3] See European Council Conclusions (EUCO 1/16) 19-24, 34.

[4]C. Dustmann and T. Frattini, ‘The Fiscal Effects of Immigration to the UK’ (2014) 124 Economic Journal 563.

[5] The most recent contribution is J.W. Muller Contesting Democracy (Yale 2011).

[6] See, generally, L. Azoulai, E. Pataut and S. Barbou des Places (eds.) Ideas of the Person and Personhood in European Union Law (Hart 2016).



The Failure of Union Citizenship beyond the Single Market

By Daniel Thym (University of Konstanz)


Floris de Witte’s defence of free movement presents us with a decidedly non-economic vision of cross-border mobility. It is this normative dimension which connects his argument to broader debates on Union citizenship whose ‘core’ he considers to be free movement. His thinking builds upon the rich tradition of institutional practices and academic reconstruction that has highlighted the non-economic value of the original market freedoms ever since the late 1960s – the period when the EU legislature opted for generous implementing legislation on the basis of which the ECJ later advanced citizens’ rights in cases with purely corollary economic aspects. 

I accept this normative starting point and yet will highlight its limited reach nonetheless. De Witte concentrates on the potential of free movement in correcting outcomes at national level without connecting the evolution of citizens’ rights to constitutional trends at European level. However, such a broader outlook could help explain the volatile state of Union citizenship at this juncture. I will argue that restrictive tendencies appear as epitaphs of a Union losing self-confidence as a supranational polity, emphasising instead the continued significance of solidary political communities at national level. If we want Union citizenship to thrive, we have to move beyond a minimalist reading.

Correcting the Nation-State

I subscribe to De Witte’s defence of free movement as emancipation without hesitation, but want to ask: is that all? Much of the liberty he associates with intra-European mobility is guaranteed as a matter of domestic or international human rights law anyway, which, together with changing self-perceptions of Western societies, considerably extended the degree of private and public choice in recent decades. Gays and lesbians may move to the big cities in their home state to find (relative) freedom – and German pensioners can settle in my current hometown of Konstanz or other domestic cities known for their quality of life instead of relocating to Spain. To be sure, European rules extend our freedom geographically and in substance, but the surplus remains gradual instead of categorical.

The same can be said about his third contention on separating ‘the nation’ from ‘the state’. Here he subscribes to an essentially corrective vision of supranational citizenship. Again, I do not take issue with his analysis as a matter of principle, but wonder about the degree of normative value involved. Arguably, the separation between the nation and the state defended by De Witte is no longer a novelty for most (Western) European societies. Nationality law is a perfect prism to highlight changing self-perceptions: two decades ago, ethno-cultural foundations of national identity were pertinent in many Member States. Immigrants obtained certain rights, but their status could be described as a form of ‘denizenship’ which stopped short of full membership through the formal acquisition of nationality and equal participation in the public realm (1). Today, the picture looks different: some Member States moved towards ius soli and essentialist definitions of national identity are being supplanted by various degrees of civic-pluralistic identities (2). 

To be sure, European integration may have been instrumental in bringing about this adjustment through more than its rules on free movement. Such change also remains an ongoing challenge characterised by ups and downs and occasional backlashes. While it is well advanced across Western Europe (notwithstanding the surge of populist movements whose success can be rationalised, in part at least, as a reaction to social change), some countries in Central and Eastern Europe are still in need of similar metamorphoses, in which the corrective potential of European rules described by De Witte may play a beneficial role (as recent developments in Poland and Hungary illustrate). But this does not unmake the move towards inclusionary nationality laws and civic-pluralistic identities. If that is correct, the emancipatory dimension of transnational mobility remains limited. It may reinforce a trend whose dynamism, however, is not intrinsically linked to Union citizenship. 

Moreover, broader societal debates on the impact of immigration across Europe illustrate that the corrective reading of transnational mobility defended by De Witte remains mostly negative. It invites European societies to abandon essentialist self-perceptions, but does not contribute much to how the normative foundations of social cohesion should be construed instead. The EU’s vision of ‘a society in which pluralism, non-discrimination, tolerance, justice, solidarity and equality between women and men prevail’ (Article 2 TEU) or the ECHR’s standard invocation of ‘pluralism, tolerance and broadmindedness’ as hallmarks of a democratic society which are supposed to structure the proportionality assessment often remain hollow. That is why De Witte’s vision of free movement reinforcing emancipation at national level remains a thin normative account.

Access to Social Benefits as a Test Case

Equal access to social benefits has received much attention in scholarly treatises on Union citizenship over the years, but only recently has it caused widespread political frictions. One reason for the limited impact of the original equal treatment guarantee may have been that it concerned those who were engaged in some sort of economic activity. For such scenarios, the Court extended the range and vigour of equal treatment against restrictive national laws, but the principle itself was uncontroversial, since most Member States had embraced territoriality as the door-opener for work-related social benefits anyway. Moreover, free movement did not substitute national policy preferences with a supranational vision of social justice. Britain and Sweden had to treat equally Union citizens who were economically active, but this did not affect the distinct structure of their welfare state (3). Again, free movement rules reinforced a trend which took place anyway.  

Against this background, the central novelty of the original free movement provisions was not equal treatment of those engaged in economic activities, but access to the labour market. To this date, the central difference between a Polish and a Ukrainian national who wants to work in Amsterdam is not equal treatment once they have taken up work. The added value of Union citizenship is the right to be admitted to the labour market – a distinction fortified by Article 15 of the Charter of Fundamental Rights which guarantees equal working conditions to everyone, but reserves the right to seek employment in any Member State to Union citizens. 

That right to be economically active across the Union is firmly inscribed into the DNA of the European project, since it presents itself as one of the pillars of the single market. We may question the outer limits of corresponding equal treatment, such as in-work benefits for part-time workers or the level of child benefits for children living in another Member State, which feature prominently in the new deal the British government promotes in the run-up to the Brexit referendum. But such disputes about the fringes should not distract from the essentially economic rationale of equal treatment for those who are economically active, which De Witte himself proposed to reconstructed as an expression of a Durkheimian organic solidarity (4). The internal market provides the frame for this arrangement. Its pan-European reach remains largely uncontested.

In relation to citizens like Ms Dano the picture looks different. Their status transcends the single market and emanates directly from the rights attached to Union citizenship. Their reach had never been subject to a principled political consensus – as the emphasis on ‘limitations and conditions’ (Article 21.1 TFEU) in primary law illustrates in the same way as the compromise formulae enshrined in the Citizenship Directive 2004/38/EC (5). That is not to say that the Court was right to flatly deny equal treatment to citizens like Ms Dano: a different position could have been defended (6). All I say is that we cannot expect the single market case law to be extended to non-economic activities indefinitely, since the constitutional frame of reference differs. It builds upon the (vague) idea of political union of which generic free movement rights for the economically inactive were always an integral part (7). Arguably, it is this connection to political union which explains the failure of citizens’ rights beyond the single market.

Connecting to the Union as a Whole

Twenty years ago, the European Union could reasonably be considered a political union in the making. Union citizenship could be perceived, like direct elections to the European Parliament or the ill-fated Constitutional Treaty, as a building block of the EU constituting itself as a supranational political community based upon meaningful public discourse and a functioning ‘representative democracy’ (Article 10.1 TEU). The famous dictum of the Court that citizenship was ‘destined’ to be a fundamental status arguably hinted at this forward-looking aspiration. A vision of social justice embracing the fight against social exclusion, whose absence in the Dano judgment De Witte criticises, would undoubtedly have been an integral part of such supranational polity (see Article 34 Charter of Fundamental Rights). 

We all know that the state of the EU is a different one at this juncture. The confidence that some sort of political union would be forthcoming was seriously damaged after the failure of the Constitutional Treaty, as a result of the eurocrisis and regarding the surge of anti-European populism. That is why a continuation of the trend towards ever more citizens’ rights was no forgone conclusion. Indeed, the Dano judgment is not the only example in which the ECJ refrained from developing its vision of social justice: not assessing austerity measures in light of the Charter is another example (8). By deciding not to engage in such debates, the Court signalled that it would not develop a thick reading of citizens’ rights. This hands the initiative back into the domestic arena. National constitutional courts or the ECHR will ultimately have to decide the fate of Ms Dano (9). The ECJ abdicated responsibility in the same way as it handed questions of family unity in purely domestic situations back to national courts and the ECHR (10).

This leaves us with the overall conclusion that any fortification of citizens’ rights beyond the single market remains linked to broader constitutional trends. If we want the Court to employ citizens’ rights to foster a supranational vision of social justice, we arguably have to move beyond a minimalist reading of free movement as correcting unwelcome outcomes at national levels. What would be required, instead, is a vision of social justice for the Union as a whole, not only for those moving to other Member States.



(1) See Tomas Hammar, Democracy and the Nation-State. Aliens, Denizens and Citizens in a World of International Migration (Avebury, 1990).

(2) See Christian Joppke, Citizenship and Immigration (Polity Press, 2010), chs 2, 4.

(3) See Daniel Thym, ‘Towards “Real” Citizenship? The Judicial Construction of Union Citizenship and its Limits’, in: Maurice Adams et al. (eds), Judging Europe’s Judges. The Legitimacy of the Case Law of the European Court of Justice (Hart, 2013) 155-174.

(4) Cf. Floris de Witte, Justice in the EU. The Emergence of Transnational Solidarity (OUP, 2015). 

5) See Kay Hailbronner, ‘Union Citizenship and Access to Social Benefits’, CML Rev. 42 (2005), 1245 (1258-1264).

(6) See Daniel Thym, ‘The Elusive Limits of Solidarity. Residence Rights of and Social Benefits for Economically Inactive Union Citizens’, CML Rev. 52 (2015), 17-50.

(7) See Antje Wiener, Building Institutions. The Developing Practice of European Citizenship (Westview Press, 1998)

(8) See Augustín José Menéndez, ‘Which Citizenship? Whose Europe?—The Many Paradoxes of European Citizenship’, GLJ 15 (2014), 928-931.

(9) After the German Federal Social Court had granted Ms Dano (and some other Union citizens who are economically inactive) a right to social benefits on the basis of statutory rules, the Merkel government announced a change in the law, which would ultimately require the German Constitutional Court to decide whether Union citizens can be expected to return to their home state to obtain social benefits.

(10) Cf. ECJ, Dereci u.a., C-256/11, EU:C:2011:734, paras 70-74.



State Citizenship, EU Citizenship and Freedom of Movement

By Richard Bellamy (European University Institute and University College, London)



I agree with the two key premises of Floris de Witte’s ‘kick off’: namely, that 1) freedom of movement lies at the core of EU citizenship and is worth defending as such, and 2) that many of the attacks on it are at best misinformed, misguided and mistaken, at worst malign, mendacious, and motivated by prejudice and xenophobia.

However, I disagree with much of what he says in support of these positions. I think he confuses the moral case for some form of cosmopolitanism and the empirical reinforcement this gets in an interdependent world, on the one hand, with an argument for a fully fledged political and legal cosmopolitanism that looks to the ultimate demise of nation states as a necessary condition for justice, on the other. The first may offer normative and empirical support for a supranational Union of states along the lines of the EU, in which there is a status such as Union citizenship that offers free movement between the component polities. However, that need not imply a version of the second involving a teleological account of the EU’s development, such as de Witte offers, whereby individuals must cease to be members of nation states; democracy becomes in some way constrained by, or even substitutable by, a given conception of justice; and we need no longer conceive ourselves as members of a community of fate. What I want to suggest in this comment is that one can accept a broadly cosmopolitan moral and empirical case for free movement within the EU as both normatively compelling and of practical benefit, while disputing all three of his arguments for this position and maintaining the very statist perspective on each of the three issues that he seeks to challenge.

Let me start by briefly setting out (space constraints mean I cannot here defend, though I have attempted to do so elsewhere (1)) what might be called a cosmopolitan statist perspective on the EU. I shall then deploy this perspective to comment on Floris de Witte’s three arguments, noting in each case how free movement can be defended while stopping short of the view he advocates. 

Cosmopolitan Statism, EU Citizenship and Freedom of Movement

On the account I adopt, the most normatively appealing and empirically plausible way of conceiving the EU is as a republic of democratic nation states. The argument is broadly Kantian, tweaked to accommodate contemporary concerns and conditions. It is both statist and cosmopolitan, and orientated around the value of non-domination. It is statist in arguing that to institute justice among individuals who reasonably disagree about its nature and application requires the establishment of a sovereign authority to govern the relations of those who share a social space. Yet if that authority is to be non-dominating and not itself a source of injustice, it must be under the equal influence and control of those to whom it applies. Therefore, justice implies the establishment of both a state and a democratic regime within it. Just relations can only pertain among citizens. However, in an interconnected world it becomes possible not only for states and their citizens to dominate other states and their citizens, both directly and indirectly, but also for various non-state agents and agencies, such as corporations and terrorist groups, to do so. That possibility increases when not all states operate democratically, with such non-democratic states not only dominating their own citizens, but also more likely to seek to dominate the citizens of other states too and to provide a haven for non-state agents and agencies to do so as well. Meanwhile, citizens of all states have various reasons to move freely between states– some to escape dominating or failing regimes, others to trade, find employment and for leisure, among other motives. As a result, states have good cause to cooperate and establish supranational legal and political structures to prevent their mutual domination, help them support non-dominating regimes in states where they do not as yet exist or are under threat, tackle domination from other non-state sources, and to facilitate free movement of citizens between these states in ways that avoid discrimination or domination, either of or by them. My claim is that the EU can be regarded as the closest we have to such a republican system of states at present.

Of course, I am not suggesting either that the EU perfectly meets the criteria of such a structure or that all the actors involved by any means conceive it in these terms (2). I merely contend that it is a plausible way of conceiving it and one that has normative appeal as a guide to how it should and could develop. On this view, a commitment to the role of democratic states as offering a context for non-dominating relations among citizens requires as a matter of consistency that states act towards other states and their citizens on the basis of certain cosmopolitan norms, not least through establishing structures such as the Council of Europe and the EU that seek to reduce non-domination between, within and across states in the four ways mentioned above. In this regard, the current call among certain Conservative politicians in Britain for Brexit and/or withdrawal from the ECHR must be regarded as either incoherent – at odds with their professed desire to defend the very idea of democratic statehood, or insincere – either done for political advantage or because they are not that attached to democracy in the first place.

I make these points to indicate how one can be opposed to the populist nationalist rhetoric of those critical of the very idea of the EU and of free movement within it, without necessarily being opposed to the idea of democratic statehood. On the contrary, one can regard the EU as existing to support democratic statehood in a variety of ways rather than as supplanting and substituting for it. From this perspective, the linkage of Union citizenship to member state citizenship is not a transitionary feature destined to whither away but inherent to its very nature. Its purpose is not to supplant but to supplement member state citizenship in two main ways: first, it allows free movement between states in ways that involve showing equal concern and respect to the citizenship regimes of both the host state and that state of origin; second, it gives citizens a direct say in the supranational structures to ensure they show them equal concern and respect as citizens of distinct member states. As we shall see, this is very different to the characterisation that Floris de Witte offers.

de Witte’s Three Arguments

de Witte”s  first argument favouring free movement is that it emancipates the individual from the nation state. He offers rather different instances of this emancipation. One of his examples, that of an LGBT couple denied recognition in their country of birth, concerns a denial of human rights within a given state. The others, such as the retired teacher from Middlesbrough seeking to enjoy her pension in Lanzarote, relate to various personal choices that will be facilitated through freedom of movement between states, some involving more significant interests than others. He claims that emphasis on nationality only provides public recognition to individuals on the basis of a single dimension that ignores or even suppresses the other dimensions of people’s lives – as he puts it, in a statement I find extraordinary, ‘the nation state’s mode of social integration reduces the incredibly complex individual to a one-dimensional being.’ This hyperbole grossly mischaracterises the role of nationality within the public cultures of the member states, all of which are constitutional democracies. It is not as if the retired teacher is obliged by UK law to only spend her pension on holidaying in an approved British seaside resort with suitably grey weather and wearing a hat displaying the Union Jack. The legal systems of most member states uphold rights to as diverse a range of life style choices as are to be found across the EU, even if all states fall short in certain respects, some more egregiously than others. Yet all these rights require a political infrastructure to determine and enforce them. This infrastructure involves citizens of any polity in a complex set of mutual obligations, that in the case of securing many rights – such as pensions – require a degree of solidarity among them. Emancipation from these sorts of bonds constitutes a form of free-riding that is ultimately self-defeating for all but a privileged few. For these very bonds make the rights individuals claim possible in the first place (3). The retired teacher would not wish to go to Lanzarote if such an infrastructure was not in place that ensured a system of property rights sufficient to allow her holiday home to be built and uphold her civil rights to personal security once there, and would not have a pension enabling her to do so in the first place if she had not worked under a similar regime in the UK. Any system of free movement, therefore, has to be such that it respects and upholds the two systems of citizenship rights that make her movement from one to the other possible in the first place. 

His second argument for free movement, as a recalibration of justice and democracy, is in this respect more nuanced in that it appears, initially at least, to recognize that there is a need for reciprocity both between citizenship regimes and among those who participate within any one of them. As it happens, I agree with him that there is no evidence that the UK would be justified in applying an ‘emergency brake’. But none of what he says here seems to justify the statement that freedom of movement serves ‘to recalibrate questions of justice and democracy in a more appropriate manner’, a position that is hardly addressed at all. At best, it suggests that appropriate mechanisms do not exist for a constructive democratic dialogue that allows for a clear discussion of how we might balance reciprocity between citizenship regimes and reciprocity within them in an equitable and sustainable way. So far that has been a matter for the CJEU looking at particular cases, on the one hand, and intergovernmental agreements, on the other. Yet both seem somewhat ad hoc and insufficiently connected to citizens as a body, which perhaps explains the general alienation from the decisions of both bodies (4).

His third argument restates the first in a neo-Habermasian manner as separating ‘nation’ and ‘state’, because EU citizenship ‘lacks the ties to a homogenous political ‘community of fate’ that perpetuate significant exclusionary practices’. Again the element of truth in this statement gets lost through exaggeration. Floris de Witte suggests that national citizenship within the member states ‘is linked to a ‘community of fate’ that reflects certain ethno-cultural ideas of a homogenous community’. As I observed above, though, what Habermas called ‘constitutional patriotism’ forms the norm across the EU. All the member states have citizenship regimes involving elements of ‘ius soli’ as well as ‘ius sanguinis’ and most have citizenries with considerable cultural diversity and mixed blood. Sadly, and worryingly, there are parties of the extreme right everywhere that are motivated by ‘ethno-cultural ideas of a homogenous community’, and in a very few countries these parties are in government. But such sentiments are not intrinsic to the very idea of a nation state. EU citizenship has no tie to any notion of nationality because that is not its function. It exists to facilitate inter-nationality and to some degree multi-nationality, but not the demise of any sense of nationality whatsoever. As I noted, a sense of political solidarity is important for the upholding of rights that we can only possess as members of a political community. The role that EU citizenship and free movement should play is in heightening our awareness of and respect for such solidarity within all the states of the Union.


As I said at the beginning, I fully agree with Floris de Witte’s concern at the attacks on the EU currently coming from the populist right, a challenge epitomized by, but unfortunately not restricted to, the Brexit campaign in the UK. However, I doubt that the best way to answer their misleading rhetoric is to make rhetorical counter-claims that are the mirror image of theirs. Rather, it is to show that their views are largely without foundation and that far from undermining national citizenship, EU citizenship and free movement defend it in the context of the normative and empirical challenges of an inter-dependent world.



(1) Among other pieces, see Richard Bellamy, `An Ever Closer Union of Peoples: Republican Intergovernmentalism, Demoi-cracy and Representation in the EU’, Journal of European Integration 35: 5 (2013), 499-516, ‘Between Cosmopolis and Community: Justice and Legitimacy in A European Union of Peoples’ in S. Tierney (ed), Nationalism and Globalisation: New Settings, New Challenges (Hart, 2015), Ch 10, 207-232, and (with Sandra Kröger), ‘Beyond a Constraining Dissensus: The Role of National Parliaments in Domesticating and Normalising the Politicization of European Integration’, Comparative European Politics 14.2 (2016) 131-53

(2) For a critique of current EU economic and monetary policies from this perspective, see Richard Bellamy and Albert Weale, ‘Political Legitimacy and European Monetary Union: Contracts, Constitutionalism and the Normative Logic of Two-Level Games’, Journal of European Public Policy 22.2 (2015) 257-74.

(3) I’ve criticised a somewhat similar argument to de Witte’s by Dimitry  Kochenov in Richard Bellamy, ‘A Duty Free Europe? What’s Wrong with Kochenov’s Account of EU Citizenship Rights’, European Law Review 21.4 (2015) 558-65

(4) On the ‘democratic disconnect’ in EU policy making, see Richard Bellamy and Sandra Kröger, ' The Politicization of European Integration: National Parliaments and the Democratic Disconnect ', Comparative European Politics 14.2 (2016) 125-30


Free Movement as a Means of Subject-Formation: Defending a More Relational Approach to EU Citizenship

by Päivi Johanna Neuvonen (University of Leicester)


Should EU citizenship ‘be primarily about free movement’? According to Floris de Witte, free movement as the core of EU citizenship can contribute to emancipation, justice, and the distinction between the ‘nation’ and the ‘state’ within the EU. I share his view that these objectives ought to be important to European integration in general and to EU citizenship in particular. But I am not fully convinced that free movement as ‘the central thing that EU citizenship should be about’ will automatically result in more just and emancipated relations between EU citizens. I would therefore like to advance a more relational understanding of subjectivity in this context. 

Floris de Witte suggests that free movement as an ‘emancipatory force’ can make the citizens of the European Union more ‘sensitive to the structural process of exclusion that the nation state perpetuates’. For him, free movement can be seen as ‘an instrument to problematize these processes’. It nevertheless seems important to consider on what basis free movement would problematize the potentially exclusionary practices within the nation state. Although I am positive about the suggestion that free movement ‘orients the individual’s visions of self-realisation and self-understanding outwards’, I have some reservations about the scope and nature of this emancipatory re-orientation through free movement. 

The interesting question is what the term ‘outwards’ means in the context of EU citizens’ free movement.  Does the idea of transnational ‘self-realisation’ recognise citizenship as an inherently relational form of human interaction and agency, or does it primarily advance an ‘atomistic’ or ‘unencumbered’ view of the self? In so far as the normative ideal of free movement is based on the mere objective of individual self-realisation, the danger is that it will foster a narrow and individualistic view of subjectivity for the purposes of European integration. The ‘subject’ that emerges from the exercise of free movement may easily appear as self-centered, rather than as capable of encountering the ‘Other’ as part of its own emancipation (1). 

According to Floris de Witte, free movement can advance a ‘more inclusive way of thinking about what freedom, justice, equality and participation should mean in the EU’. He also writes that free movement as the core of EU citizenship ‘benefits all those citizens – whether they make use of it or not’. Here we encounter the question of whether all EU citizens are able to enjoy the right to free movement. I agree with Daniel Thym’s point that, if we take seriously the argument that free movement ‘ought to be available under similar conditions for all nationals of the Member States’, a more comprehensive account of social justice is still needed for the EU (2). The idea of free movement may indeed be central to actualising the principles of transnational justice. But it will then be seen as a tool for justice, rather than as an end of EU citizenship.

It seems difficult to justify the non-economic right to free movement and residence without first accepting a more independent equality objective for EU citizenship. Any reference to EU citizenship as an equal status nevertheless raises a set of difficult questions about belonging and solidarity. According to Richard Bellamy, EU citizenship must not bring about the demise of the ‘political infrastructure’ that advances the degree of solidarity that is arguably required for securing many rights within the Member States. Bellamy’s account holds that just relations between citizens can be understood as ‘relations of those who share a social space’. But his statist conclusion becomes less self-evident if we assume that ‘a social space’ can also be constructed transnationally. If equality is understood as a ‘normative ideal of human relations’ (3), the important question for EU citizenship, as well as for the existence of the EU as an ‘emergent polity’ (4), is whether it is possible to construct meaningful relations for the purposes of equal treatment outside the context of ex ante belonging.

It seems to me that the argument of self-realisation through free movement does not yet adequately address the relational potential of EU citizenship. In his contribution, Floris de Witte refers to ‘a “stateless” or “anchorless” idea of belonging and community’ as the ‘strength’ of EU citizenship. Although his argument of free movement as an ‘emancipatory force’ seeks to challenge ‘communitarian ties’, I hope it would also say more about whether those relationships that constitute a meaningful ‘social space’ can be transformed and redefined through European integration – without just replicating the exclusionary ‘community of fate’ transnationally? If this question can only be answered in the negative, those authors who are concerned about the harmful implications of EU citizenship for political and social emancipation may have an important point to make (5). However, I have argued elsewhere that EU citizens’ equal treatment is closely connected to the gradual process of transnational subject-formation, the outcome of which may ultimately justify a more positive answer to the above question of meaningful relations (6). 

In sum, free movement can have a central role in constructing transnational political and legal subjectivity. But I would see it as one method of advancing more just and equal relations between EU citizens, rather than as the only objective of EU citizenship. Floris de Witte suggests that ‘the added value that EU citizenship can offer primarily lies in its connection to free movement’. However, if the added value of EU citizenship is ultimately connected to how we respond to otherness within the EU, free movement is not the only context in which EU citizens can exercise their subjectivity as EU citizens in a meaningful way. At the end of his forum post, de Witte, too, seems to come close to this view when he writes that free movement is how the ‘virtue’ of ‘a new – less ethnic – way of thinking about the role of the individual in the EU’ is ‘implemented’.



(1) For further discussion, see Päivi J. Neuvonen, Equal Citizenship and Its Limits in EU Law: We the Burden (Hart 2016).

(2) For further discussion see e.g. Dimitry Kochenov, Graínne de Búrca, and Andrew Williams (eds), Europe’s Justice Deficit (Hart 2015). 

(3) E.g. Samuel Scheffler, Equality and Tradition: Questions of Value in Moral and Political Theory (OUP 2007) 234.

(4) E.g. Antje Wiener, ‘Constitution-making and Citizenship Practice – Bridging the Democracy Gap in the EU?’ (1997) 35 Journal of Common Market Studies 595, 596. 

(5) E.g. Alexander Somek, The Cosmopolitan Constitution (OUP 2014) 160-61. 

(6) Päivi J. Neuvonen, Equal Citizenship and Its Limits in EU Law: We the Burden (Hart 2016).


Free Movement Emancipates, but what a Freedom this is?

By Vesco Paskalev (University of Hull)


I must start my response to Floris de Witte with a personal note – I am a Bulgarian national, living and working in Britain. As such, I am strongly attracted by his argument that sees free movement as the core of EU citizenship aimed at extending individual liberties. Indeed, my moving away from Bulgaria was an act of emancipation from the perennially corrupt and increasingly fascist country where I was born. Contrary to what Daniel Thym and Richard Bellamy argue, the Member States, while nominally democratic, do differ in their respect for fundamental rights of their citizens, and the professed ambitions of the current Hungarian prime minister to build an illiberal state does not seem to suggest that convergence towards the highest democratic standards is forthcoming.

Indeed, freedom of movement is emancipatory in a number of senses. On a conceptual level, EU citizenship liberates everyone: for centuries contractarian theories have claimed that people who do not leave their country of residence can be seen as consenting to its authority. While until recently the exit option has been only putative, now EU citizenship allows us to conceive those who stay as accepting state authority voluntarily. Certainly, EU citizenship should be the dream of libertarians – in a marketplace of governments you can shop around and chose the one which is freer, or perhaps the one which is best tailored to your personal taste.  EU citizenship is emancipatory also in pragmatic terms (one may call this argument neoliberal) – the fear of possible mass exit of citizens (a.k.a. workforce, taxpayers, electorate) may deter governments from abusing them. All in all, if we equate freedom with individual pursuit of happiness in a social context that is taken for granted, it is difficult to argue against De Witte. However, it is not so on a more robust, Arendtian understanding of freedom as equal participation in a self-governing community, which free movement tends to erode. 

Certainly, De Witte (and all of the previous contributors) do not understand freedom negatively. Indeed free movement may promote certain positive aspects of freedom. For example, De Witte correctly argues that free movement ‘liberates not only the body but also the mind from the normative structures of the state.’ The Brexit referendum provides a wonderful empirical confirmation of this point. Opinion polls suggest that while older Britons are clinging on antiquated ideas about sovereignty, the younger generation – born as EU citizens and in conditions of widespread mobility – are very much at ease with joint decision-making and are more likely to see the Union as empowering rather than crippling their own country (1). There is no similar evidence for the attitudes of older Britons living in Europe, but it is plausible to expect some similarity between the views of the people who actually move and of those who are born with the right to (2). There is ‘reflexive virtue’ to be gained from free movement indeed. 

Such collateral benefits of free movement notwithstanding, civic virtue is ill served by free movement and it is hardly surprising that Richard Bellamy disagrees with De Witte. On republican accounts citizenship is relational and European mobility by definition loosens the link between citizens and their state. Even in the age of Ryanair and Skype the opportunities of the external citizens to participate in the democratic life of their home state are significantly reduced. Indeed, in most cases they retain the right to vote, and its exercise abroad is often – but not always – facilitated by postal, proxy and e-voting. But democracy is so much more than the ballot box! Citizens who do not move can go on rallies, volunteer for various causes, join political organisations, speak in public or engage in community initiatives. One need not subscribe to Pierre Rosanvallon’s concept of counter-democracy (3) to agree that all this is part and parcel of any democracy. Thus, on the more robust understanding of freedom, which encompasses equal opportunity for participation in the collective system of governance, free movement inevitably reduces freedom. The fact that the mobile citizens have moved out freely may satisfy contractarians but not civic republicans. As long as the link with the home state is not broken completely – which may happen eventually – the freedom of the external citizens is limited in this sense. 

Now, this attenuated freedom might be normatively satisfactory as the stake of the external citizens in their country of origin is decreased, too (4). And of course, along with the freedom to move, the EU citizens now have – uniquely - extensive rights to participate in the democratic governance of the EU itself, which remain unaffected by their movement. Further to this, from day one EU citizens have enjoyed significant rights to participate in the political process of the host state. Apart from the electoral franchise, the rights of participation they have there, albeit limited, roughly correspond to the rights which are difficult to exercise from a distance in the home state. It might appear that freedom – even republican freedom – lost equals freedom gained. The problem is that in practice the external citizens are far less likely to exercise the rights they have in the host state than they would exercise equivalent rights in their state of origin. While your ability to attend a rally ceases on the day you have left the country, it is highly unlikely that you would participate in another rally on the day you arrived in your new country of residence. Notwithstanding the legal rights the Treaties will give, there is an inevitable lag before a mobile citizen integrates in the political process of the host state to the degree he or she was integrated in the home state. For this period – and it can be very long – the mobile citizens are losing a significant aspect of their freedom due to their movement.

This may all appear trivial. Indeed, reality rarely conforms fully to our normative expectations; even in the simplest case of national voting not every single citizen has effective and equal opportunity to vote and we are still satisfied when the overwhelming majority does. As long as only about 15 (out of 508) million EU citizens (5) have actually moved one may be right not to lose much sleep over the impact on democracy in the EU. The problem is one of aggregation. Republican freedom, and democracy in general, depend on a critical number of citizens who do participate actively in the political process. When fewer people participate – in voting and in the informal modes of contestation – the robustness of freedom decreases for all. Indeed, one of the main reasons for the democratic deficit of the EU is alleged to be the low turnout in elections for the European Parliament. This is the darker side of free movement. Notwithstanding its apparent emancipatory effect for the individual citizens – which may well outweigh what is lost in terms of non-participation – it tends to decrease, rather than increase republican freedom in Europe. 



(1) A YouGov poll found that the intergenerational gap is immense “73 per cent of those aged between 18-29 want to remain in the EU, while 63 per cent of those aged over 60 want to leave”, The Telegraph, 12 May 2016, available at http://www.telegraph.co.uk/news/2016/03/21/eu-referendum-who-in-britain-wants-to-leave-and-who-wants-to-rem/.

(2) Some anecdotal evidence available in Paxman goes to Brussels: Who really rules us? BBC Documentary, first shown 19 May 2016, available at http://www.bbc.co.uk/iplayer/episode/b07c6n58/paxman-in-brussels-who-really-rules-us

(3) Pierre Rosanvallon, Counter-Democracy: Politics in an Age of Distrust (Cambridge University Press 2008).

(4) For a discussion see Rainer Bauböck, Stakeholder Citizenship and Transnational Political Participation: A Normative Evaluation of External Voting, 75 Fordham L. Rev. 2393 (2007), available at http://ir.lawnet.fordham.edu/cgi/viewcontent.cgi?article=4257&context=flr 

(5) Eurostat, see data available at http://ec.europa.eu/eurostat/statistics-explained/index.php/Migration_and_migrant_population_statistics



Free Movement and EU Citizenship from the Perspective of Intra-European Mobility

By Saara Koikkalainen (University of Lapland)


In his kickoff text, Floris de Witte argues that the value of free movement lies in its capacity to emancipate the individual from the nation state, to recalibrate questions of justice and democracy, and to sever ties to a homogenous political ‘community of fate’. My contribution builds on empirical research on intra-European mobility and elaborates on his first claim on emancipation. I offer two factors to support my interpretation of the strong link between free movement and EU citizenship: 1) the development of the very concept of European citizenship is at least partly the result of a longer history of free movement and 2) the concrete advantages of EU citizenship are strongly linked to free movement. I finish with the conclusion that free movement makes EU real also for those Europeans who have not exercised their right to move. As de Witte says: “Freedom of movement, in other words, liberates not only the body but also the mind from the normative structures of the state.“ 

The history of free movement and EU citizenship 

The foundations of free movement date back to the 1950s and the Treaty establishing the European Coal and Steel Community (ECSC), where the cross-border movement of coal and steel industry workers was to be eased to aid the growing post-war economies. The EEC-Treaty extended free movement rights to workers in other industries, with the exception of the public sector, and these rights were codified in 1968 for the workers from the six original Member States. Since the 1970s, the European Court of Justice has played a fundamental role in widening the scope of free movement, as ordinary Europeans have been active in testing its boundaries in court, thus gradually extending the right of free movement to persons. The process culminated with the introduction of European citizenship in the Maastricht Treaty in 1992 where the right was extended to citizens (1).

While free movement was originally based on an economic rationale and the desire to provide a flexible workforce for the industry, it has developed into a civic right that might have been impossible to envision without the preceding decades of mobility. The right is also highly valued by the Europeans themselves: in the Eurobarometer surveys (2), freedom of movement consistently ranks high among the things that Europeans value in the EU. In the autumn of 2015, 78 % of the respondents supported free movement, even though differences among countries were significant (94 % support in Latvia and 92% in Estonia, while only 64 % in the UK and 66 % in Austria). Free movement is also routinely listed as the most positive or the second most positive result of the EU along with “peace among the EU member states”. Therefore, while issues such as access to social security and transferability of pensions across borders are still problematic, it is clear that free movement is, according to the Europeans themselves, at the core of European citizenship. 

The value of EU citizenship is linked with free movement

Europeans take advantage of free movement as students, trainees, professionals, family members, retirees, and workers of different skills and educational backgrounds. Not all are moving for life, as many choose to live abroad temporarily or seasonally or engage in various cross-border activities (3). In response to de Witte, Daniel Thym writes that while he understands the value of free movement for the individual, he also sees the limitations of de Witte’s argument: “European rules extend our freedom geographically and in substance, but the surplus remains gradual instead of categorical.” Thym downplays the exceptionality of a situation where a German pensioner, for example, is free to settle in Spain, instead of just relocating to a more pleasant environment within Germany. However, along with others engaged in research in intra-European mobility, I argue that those who exercise their right to free movement are pioneers of European integration, whose lives and actions impact both the countries of origin and destination as well as the socio-cultural construction of Europe in a multitude of different ways (4).

In Richard Bellamy’s view, EU citizenship does not undermine national citizenship but rather defends it in the “context of the normative and empirical challenges of an inter-dependent world.” Yet when examining EU citizenship from the viewpoint of the intra-European migrant, I am tempted to agree with de Witte that it is “distinct from, and genuinely supplementary to, national citizenship”. Namely, the extensive rights granted by EU citizenship have made adopting the (legal) citizenship of the country of destination largely unnecessary, and for Europeans the value of citizenship acquisition is clearly lower than for third-country nationals wishing to legally settle within the EU. In 2013, for example, in twelve EU member states at least nine out of ten persons who were granted citizenship were non-EU citizens while only in Hungary and Luxembourg EU-migrants were in the majority (5). There is hardly any other circumstance where EU citizenship would have such a manifest impact on the lived experience of an individual than the possibility of being a legal, long-term resident of a country with minimal pressures to naturalisation.

Imaginary horizons and cognitive migration

Free movement is at the core of EU citizenship also because it opens horizons for Europeans who have not moved abroad, but may have seriously considered the matter, plan to do so in the future or see mobility as an option for their children. The imagination of a potential future involving international migration is a way of making Europe or the EU seem real in the mind of an individual (6). It relies on a process we have called cognitive migration where the mind may travel multiple times before the actual bodily move takes place (7). In the Flash Eurobarometer of spring 2016 four in five respondents were aware of their mobility rights as European citizens (8), so the option is widely known among ordinary Europeans. The impact of such a high share of individuals potentially imagining futures that transcend national borders should not be underestimated as a factor influencing what EU citizenship currently is and what it will be in the future. 



(1) E.g. Saara Koikkalainen, 'Free movement in Europe: Past and Present'. Migration Information Source, April 2011. Washington, DC: Migration Policy Institute. www.migrationinformation.org/Feature/display.cfm?ID=836.

(2) Standard Eurobarometer 84, 2015, EC, DG for Communication, Autumn 2015. First results. http://ec.europa.eu/COMMFrontOffice/PublicOpinion/index.cfm/Survey/index#p=1&instruments=STANDARD

(3) E.g. Adrian Favell, Eurostars and Eurocities: Free Movement and Mobility in an Integrating Europe (Oxford: Blackwell, 2008); R.King,  'Towards a new map of European migration', International Journal of Population Geography 8 (2002), 89-106; EUCROSS, 'The Europeanisation of Everyday Life: Cross-Border Practices and Transnational Identifications Among EU and Third-Country Citizens', Final Report, 2014. www.eucross.eu/

(4) Ettore Recchi and Adrian Favell, Pioneers of European integration. Citizenship and mobility in the EU (Cheltenham: Edward Elgar, 2009). 

(5) Eurostat, Acquisition of citizenship statistics. Eurostat Statics Explained (July 2015). http://ec.europa.eu/eurostat/statistics-explained/index.php/Acquisition_of_citizenship_statistics

(6) E.Castano, 'European identity: A social-psychological perspective', in R.K. Herrmann, T. Risse & M.B. Brewer (eds.), Transnational identities: Becoming European in the EU (Oxford: Rowman & Littlefield, 2004) 40-58.

(7) Saara Koikkalainen & D. Kyle, 'Imagining Mobility: The Prospective Cognition Question in Migration Research', Journal of Ethnic and Migration Studies (2015). Taylor & Francis Online (October 7th) doi: 10.1080/1369183X.2015.1111133

(8) Flash Eurobarometer 430 2016, EC, DG for Justice and Consumers & DG for Communication. http://ec.europa.eu/COMMFrontOffice/PublicOpinion/index.cfm/Survey/index#p=1&instruments=FLASH



The New Cleavage Between Mobile and Immobile Europeans

By Rainer Bauböck (European University Institute)


The Brexit vote on 23 June 2016 has cast a long and dark shadow over our debate on free movement and the future of EU citizenship. At several points in the past, the European project has experienced periods of crisis or stagnation. But now is the first time that it seems to be going into reverse gear with two possible outcomes: the EU losing one of its largest member states or a process of disintegration that could affect the Union as a whole. 

The Brexit referendum was not inevitable. It was a political gamble by David Cameron to overcome a split in the Tory Party. After the vote the attitude of political irresponsibility that caused this mess in the first place has been spreading like a contagious disease across the political spectrum, with the most prominent Leave campaigners refusing to take responsibility for the disastrous consequences of their victory and the lukewarm Remainers like Jeremy Corbyn incapable of realizing the historic dimensions of their failure. Brexit was not thus British destiny but a contingent outcome triggered by an extraordinary lack of responsible political leadership. Yet this does not mean that there is no need for grasping the deeper forces that made this result possible and that are in no way uniquely British. 

Floris de Witte’s spirited defence of free movement focuses on its contribution to individual liberty, to cosmopolitan conceptions of justice and democracy and to overcoming exclusionary national communities of fate. I broadly agree. But there is something important missing in his story. What he does not speak about is the reactionary backlash against intra-EU mobility that threatens now to determine the outcome of votes not only in Britain and could sweep right wing populist parties into power in several continental member states. While the Remain campaign focused on the economic folly of Brexit, the Leavers won the battle by mobilising popular resistance against free movement rights of EU citizens.

Many post-referendum analyses agree that there is a new political cleavage in Europe that can no longer be reduced to the traditional divide between left and right and that is most strongly articulated through citizens’ attitudes towards European integration. The social characteristics of populations on either side of this divide are everywhere the same: young versus old, high versus low education, urban versus rural, and – less universally so – female versus male (1). Yet there is one further characteristic that tends to be overlooked and that is causally connected with political stances on free movement. Mobile citizens tend to vote for pro-European parties or policies and immobile ones for anti-European ones. One of the most striking charts published by the Financial Times after the Brexit vote shows a very strong positive correlation between the percentage of local residents who did not hold a passport in 2014 – and thus were unlikely to have travelled abroad – and the share of the Brexit vote (2). It seems we are witnessing a political revolt of immobile against mobile Europeans. 

This may seem an odd diagnosis given that EUROSTAT data show less than 4% of EU citizens currently residing in another member state for more than 12 months. But, as Saara Koikkalainen argues in her contribution and as Ettore Recchi and Justyna Salamonska show in a recent survey in seven EU countries, the numbers of mobile populations are much larger if one counts those with some lifetime experience of intra-EU mobility and includes transnational cognitive and network mobility. The EUCROSS study finds 13% who have lived for more than 3 months in another European state and slightly more than 50% who communicate regularly with family and friends across European borders, who have visited another European country in the last 24 months or who watch TV in a language other than their native one or the official one of their country of residence (3). Theresa Kuhn has shown that such individual experiences of transnationalism shape positive attitudes towards European integration but that this effect is social stratified. Conversely, the absence of transnational activities is likely to lead to perceptions of negative externalities of intra-EU mobility and negative attitudes towards European integration (4).  

Traditional cleavages along class, religious or ethno-linguistic fault-lines divided the political spaces of nation-states into distinct segments who lived either in separate parts of the state territory or in separate life-worlds. These divides could be either bridged through consociational power-sharing between parties representing the different sections or eroded through fostering geographic and social mobility across the divides. The new European cleavage is different because of divergent political spaces and time horizons. Mobile citizens regard Europe as their emerging space of opportunity and increasingly also of identity, whereas the immobile ones look back to the time when closed nation-states provided comprehensive social protection. 

Floris de Witte shares the diagnosis: ‘The main fault line that seems to be emerging is that between mobile and immobile citizens in the EU’. But he is not interested in bridging the cleavage. Instead he criticises ‘those scholars and politicians who wish to understand EU citizenship to be primarily about the connection between all Member State nationals and the EU rather than focusing on the rights of mobile citizens alone’. This is the wrong response to the crisis. As long as European citizenship is nearly exclusively about free movement, immobile Europeans will not perceive it as a value and as an important aspect of their identity. I agree with Daniel Thym that what is needed to win this battle is ‘a vision of social justice for the Union as a whole, not only for those moving to other Member States’.

For de Witte, ‘[t]he scholarship on ‘integration through law’ suggests that law is both the agent and object of integration, and is used to push through the objectives of integration even in the presence of political objection on the national or supranational level’. But today, this seems like the strategy of generals who always fight the last war. The battle for freedom of movement and European integration is no longer fought primarily in courts where individual rights can trump majority preferences; it is increasingly fought in polling stations, parliaments and the mass media. In order to survive, European integration through law will have to be complemented with integration through democracy, by winning the hearts and minds not only of mobile Europeans, but of immobile ones as well. 



(1) The gender gap was especially dramatic in the recent Austrian presidential elections: On 22 May, 60% of female voters cast their ballot for the left-liberal green candidate, while 60% of males voted for the right wing populist one. 

(2) John Burn-Murdoch in FT, 27 June 2016 (based on data from 382 voting areas).

(3) Ettore Recchiand J. Salámonska. Europe between mobility and sedentarism: Patterns of micro-transnationalism and their consequences for European integration. Unpublished working paper (2014).

(4) Theresa Kuhn, Experiencing European Integration: Transnational Lives and European Identity (OUP 2015)



Whose Freedom of Movement is Worth Defending?

By Sarah Fine (King’s College London)


‘Should the UK remain a member of the European Union or leave the European Union?’, the British government asked the electorate in a referendum on 23 June 2016. On 24 June, we awoke to the momentous news that a majority of voters had opted for ‘LEAVE’ (Brexit). Against this backdrop, the informative EUDO Citizenship debate on the relationship between EU citizenship and freedom of movement could hardly be more timely, and obviously has even greater poignancy following the historic Brexit decision.

Since the referendum question did not directly ask voters about migration, the Leave result itself cannot be interpreted straightforwardly as a rejection of EU freedom of movement. However, long before the votes were counted, commentators were connecting Brexit’s popularity with widespread negative attitudes towards the free movement of EU citizens and net immigration figures (among many other factors, of course). The dominant view was that Leave voters tended to be particularly swayed by concerns about immigration control, as distinct from Remain voters who tended to prioritise economic arguments. As the Economist wrote several months ago, ‘immigration is one subject on which Leave campaigners have a clear lead. The correlation between hostility to immigration and support for Brexit is high, so if they can turn the vote into one about migration, they will win’. Though analysts are still collecting and examining important data about voting patterns, and digesting the results, there is no doubt that migration was a pivotal issue in the national debate (1). The Brexit side clearly considered free movement to be a central concern for the electorate. With their appealing tagline of ‘take back control’, the Leave campaign put migration at the heart of the argument in favour of withdrawing from the EU. On their official website, for example, they explained ‘what would happen if we vote to leave the EU’. The second and third points (after the claim that ‘we will be able to save £350 million a week’), were that ‘we will be in charge of our own borders’ and that ‘we can be in charge of immigration’. They described immigration from the EU as ‘out of control’ and as a ‘big strain on public services’. They also linked it with security concerns, stating that the ‘EU Court’ prevents Britain both from stopping ‘violent convicted criminals coming here from Europe’ and from ‘deporting dangerous terror suspects’.  In short, the Leave side presented freedom of movement as one of the core features—if not the core feature—of EU membership, and they clearly considered that highlighting this connection would be a vote-winner for the Brexit camp. On the other side, the Remain campaign’s website and materials made little or no mention of free movement as a feature of EU membership, a silence which itself speaks volumes about general perceptions of this topic’s selling power.   

One of the most noteworthy issues discussed in the EUDO citizenship debate is the growing evidence that there is, as Rainer Bauböck explains, a ‘new political cleavage in Europe’, which is ‘most strongly articulated through citizens’ attitudes towards European integration’. I think the emphasis on the differences between ‘mobile’ citizens, for whom freedom of movement represents exciting opportunities, and ‘immobile’ citizens, who associate free movement with serious costs to themselves and their communities, is especially illuminating for trying to interpret the various anti-EU developments across the continent, including the factors which have contributed to the Brexit result.

However, I want to focus on a crucial, related issue that was striking for its absence from Floris de Witte’s kickoff contribution and the subsequent debate about the relationship between freedom of movement and EU citizenship—an issue which demands attention in any attempt to conceptualise the place of free movement in the European project. This issue is brought into stark relief by the ongoing refugee crisis, the thousands of avoidable deaths so far this year in the Mediterranean, and the growing number of people stuck in makeshift camps in European countries. It is the fact that, beyond the political, cultural and socio-economic cleavages between citizens within Europe, there is a far greater and growing divide between European citizens and the people they want to keep out. The European Parliament’s website identifies the refugee crisis and the ‘heightened terrorist threat’ as central challenges to the European free movement zone. The European Parliament states that these ‘ongoing challenges have served to underline the inextricable link between robust external border management and free movement inside those external borders and persuaded the Commission to come forward with proposals both to enhance security checks on persons entering the Schengen area and to improve external border management’. It is not news that the free movement of European citizens is widely understood to rely on hard external European borders, and now also enhanced monitoring of movement between European states. But surely this has to factor into any response to Floris de Witte’s question of whether there is ‘any reason to defend free movement as the core of EU citizenship.’

Most importantly, the conspicuous refusal of the EU to respond humanely to refugees and migrants seeking entry, as well as its collective failure to show solidarity with its own member states at the forefront of the crisis, cannot be neglected from this discussion. How can we try to defend free movement as the core of EU citizenship without considering what is happening right now at (and indeed within) the EU’s own borders?

Returning to the Brexit case, the refugee crisis and the EU’s response to it featured prominently in the public debate about the costs and benefits of EU membership. The Brexiteers were accused of trying to stoke up anti-refugee and anti-migrant fears, particularly with Nigel Farage’s now infamous ‘Breaking Point’ poster, which pictured people crossing the Croatia-Slovenia border. But it is crucial to note that the Remain side raised the issue of refugees, too—we must not forget, for example, David Cameron’s warnings that leaving the EU could mean that Calais-style camps move from France to the UK. In effect, both sides were arguing, either explicitly or implicitly, that their position offered the best prospects for keeping refugees and unwanted migrants out of the UK. As long as the EU itself continues to present refugees as a problem to be kept at bay, with repeated promises to strengthen its borders against unwanted arrivals, those of us who wish to defend freedom of movement as a core component of EU citizenship have to ask ourselves not just about Europe’s ‘immobile’ citizens who associate free movement with unpalatable costs, but about the people on the wrong side of Europe’s territorial and civic borders who are paying the ultimate price. 



(1) See e.g. this interesting GQR poll conducted for the TUC: https://gqrr.app.box.com/s/xb5sfzo19btsn74vawnmu7mn033p1ary



The Court and the Legislators: Who Should Define the Scope of Free Movement in the EU?

By Martijn van den Brink (European University Institute)



Floris de Witte makes the case for free movement as the core of EU citizenship and offers three reasons in support. I agree with these principles, at least in the abstract. De Witte’s vision certainly is normatively more appealing than the one of scholars who have pushed for a decoupling of EU citizenship from free movement. In fact, it is hard to see how EU citizenship cannot revolve to a very large extent around the right to move freely within the EU and to choose the preferred Member State of settlement. But if that is the case, de Witte seems to be asking the wrong question. What he seems to address is not the question of whether free movement should be defended, but how that should be done; through which procedures free movement is to be given shape.

Free movement as the core of EU Citizenship

De Witte is concerned about the Dano decision and sees it as an attack on free movement. No doubt the decision is a departure from earlier case law and signifies a move away from the very extensive interpretation of the free movement principles present in certain earlier decisions. Still, de Witte’s opinion of the case as well as the way he uses the decision to support his claim is remarkable and not fully persuasive. 

First, let’s for a moment think about the difference EU citizenship has made. If one would have claimed in the mid-1990s, shortly after the introduction of EU citizenship, that in 2016 many EU lawyers have serious misgivings about a decision that denies social assistance benefits to an economically inactive EU citizen with very weak links to the Member State of residence, many would have been quite surprised about such a claim. The transformation brought about by EU citizenship has in that sense been remarkable. But was free movement not the core of EU citizenship before the Court started developing this concept in its case law? Of course it was. In fact, that so many lawyers thought EU citizenship to be a meaningless addition to the Treaties was precisely because it was largely premised on free movement (1).

In other words, also post-Dano free movement remains the core of EU citizenship; it is simply that the precise contours of this right have changed somewhat. The real discussion thus is about the precise scope of the freedom to move and, relatedly, the principle of non-discrimination on grounds of nationality. The Treaty provisions on free movement are of course indeterminate and their meaning far from evident. Indeed, as de Witte suggests, the decision demonstrates that free movement is not unlimited, but whether this is a problem is something people will reasonably disagree about. This is also recognized by de Witte, who acknowledges that not everyone shares his belief that Dano is unjust. 

Justice, free movement, disagreement, and authority

But it is at this point that the real issue arises, namely, how, in the face of disagreement about justice, such contestation is to be settled? Through which political procedures do we want to resolve such disagreements? What is necessary, in other words, is to ‘complement one's theory of rights with a theory of authority’ (2). This issue has been largely ignored by most discussions of the recent social assistance case law.

While not addressed explicitly, for de Witte the authority to settle such disputes is clearly to be given to the Court. This might be, as Bellamy in his reply submits, because de Witte’s argument ‘suggests that appropriate mechanisms do not exist for a constructive democratic dialogue that allows for a clear discussion of how we might balance reciprocity between citizenship regimes and reciprocity within them in an equitable and sustainable way’. However, while far from perfect, the EU has in fact adopted decision-making procedures that to the extent possible allow for such a dialogue. This dialogue, of course, takes place when the different institutions involved in the EU’s legislative process, in which the EU citizen is represented by the national governments as well as by the European Parliament, deliberate and decide. These institutions have also spoken on many of the questions underlying the social assistance case law. They did so when the Citizenship Directive was adopted, in which the eligibility criteria for social assistance benefits for the economically inactive are laid down. The basic rule is that the economically inactive, such as students and jobseekers, are not entitled under EU law to benefits before they have acquired permanent residence. In Dano, but also a number of subsequent decisions, the Court deferred to these criteria.

I am uncertain on the basis of which grounds precisely de Witte objects to Dano, but it appears as if he suggests that the Court should have ignored the Citizenship Directive. After all, would Member States be obliged to give EU citizens like those in the position of Ms Dano social assistance benefits, it becomes increasingly difficult to see in what situations benefits can be denied to mobile Union citizens. Of course, de Witte might think this is what principles of justice require, but why the Court is the preferred institution to settle these issues, in particular when the legislator has spoken, he does not address. It simply does not suffice to claim that Dano is unjust, because it is precisely because there is disagreement about principles of justice that we need to decide who is to be given the authority to decide on these matters. The argument, which one often finds in the literature, that the Court should ensure that secondary law complies with primary law is not persuasive either (3). After all, the Treaty provisions are indeterminate, which raises the question why the Court’s interpretation of them should be preferred over that of the legislator (also the Citizenship Directive is an interpretation of the relevant Treaty provisions). For de Witte’s argument to work he would thus need to explain on what grounds he would want to leave those matters to the Court and not the legislator. In other words, if there is no obviously correct answer to the question of substance, to how the free movement provisions are to be interpreted, why should we, if we care about the law’s democratic legitimacy, not answer the question of authority in favour of the legislator?

To put it differently, I am struggling to see how de Witte’s Court-oriented perspective is compatible with his emphasis on the need ‘to calibrate questions of justice and democracy in a more appropriate manner’, because what he seems to suggest is that his preferred conception of justice is to be adopted by the Court against the wishes of the EU’s legislator.

How to defend free movement

It is, for this reason, also that I think his suggestions might be counter-productive. To understand why, let’s consider de Witte’s objections against the ‘emergency brake’. As a matter of principle I agree that this emergency brake is unnecessary and unjust. Whether the Court should also strike it down or interpret it away if it were ever adopted is a different matter. If de Witte believes that the boundaries of free movement set by the Citizenship Directive should have been ignored by the Court in Dano, he must also believe that the ‘emergency brake’ should be annulled. If, after all, the Member States should not be allowed to deny benefits to the economically inactive, he certainly must think that the Court should prevent benefits to be withheld from the economically active. Now, let’s assume that the majority of the UK electorate had voted to remain part of the EU. Following the referendum, the Citizenship Directive would have been amended so as to include an emergency brake to give effect to the UK renegotiation. If the Court would decide to strike down these amendments large parts of the UK electorate would predictably be outraged and support for free movement would likely further erode.

The question thus also is how to defend free movement. If it is left to the Court alone to decide on the scope of the mobility rights of EU citizens, and certainly if that means disregarding legislative decisions, those who are hostile towards free movement are even less likely to support free movement. Problematically, absent support for free movement principles among EU citizens, this right will be difficult to sustain. I agree, therefore, with Rainer Bauböck’s argument that the aim must be also to convince immobile Union citizens of EU citizenship’s value. 

Contrary to Bauböck, however, I am uncertain how this is to be achieved by working towards what Daniel Thym calls ‘a vision of social justice for the Union as a whole’. Thym explains, correctly in my view, that ‘free movement did not substitute national policy preferences with a supranational vision of social justice’, but thinks that the Court should foster such a uniform supranational vision. With all respect, I think it would be highly problematic for the Court to do so, not only because such judicial behaviour is likely only to reinforce the backlash against the EU, but also because as Seyla Benhabib once explained with admirable clarity

‘[s]ocioeconomic justice and criteria by which to examine it cannot be identified independently of democratic freedom and self-determination … Precisely because there is no certainty on these matters even among experts, judgments as to who constitutes the “worst off” in society or in the world at large require complex democratic processes of opinion and will-formation’ (4).

On this issue Floris de Witte seems to agree (5). But it is precisely because of his emphasis on the importance of deciding issues of great normative salience through democratic processes that I struggle to understand his Court-centred perspective when what is at stake is the question from what moment in time mobile EU citizens are to be given full equal treatment. The EU’s legislative process might be far from perfect in this regard, but it is comparatively superior, democratically speaking, to the judicial process. I think, therefore, that a plausible case can be made for the Dano decision from this angle. Furthermore, if our concern is to persuade those who are hostile towards free movement – if not of its value, then at least of the reasons why it should be respected – then defining the limits of free movement through the legislative process seems preferable. This at least allows us to explain to those who are sceptical of free movement that the rules in place were adopted on the basis of procedures in which their national governments were involved.


All of this does not change that I agree with de Witte that EU citizenship scholars should value free movement more than they tend to do. EU citizenship is not about the centralisation of rights and about replacing the democratically legitimated substance of national laws by uniform European ones. Instead, the value of EU citizenship lies in the opportunity it offers to EU citizens to take up residence in another Member State to pursue their dreams and ambitions. But while this is so, we should not forget that its value is not uniformly accepted by all Union citizens. Neither should we ignore that free movement never was meant to be unlimited. One may deplore this and criticise the status quo for being unjust, but that alone is insufficient to claim that the Court should change the scope of the free movement rights. To the contrary, if we want to defend the right to free movement and enlarge its support, respecting the legislative limitations might be the better way to go.  



(1) Michelle Everson, ‘The Legacy of the Market Citizen’ in Jo Shaw and Gillian More (eds), New legal dynamics of European Union (Clarendon Press 1995).

(2) Jeremy Waldron, ‘Participation: The Right of Rights’ (1998) 98 New Series 307, 322.

(3) Síofra O’Leary, ‘Equal Treatment and EU Citizens: A New Chapter on Cross-Border Educational Mobility and Access to Student Financial Assistance’ (2009) 34 European Law Review 612, 622.

(4) Seyla Benhabib, The Rights of Others: Aliens, Residents and Citizens (CUP 2004) 110-111.

(5) See: Floris de Witte, Justice in the EU: The Emergence of Transnational Solidarity (OUP 2015) 54.



Reading Too Much and Too Little into the Matter? Latent Limits and Potentials of EU Freedom of Movement

By Julija Sardelić (University of Liverpool) 


This EUDO Citizenship debate has shown that EU freedom of movement is under attack. As Rainer Bauböck highlighted in his contribution, the outcome of the EU membership referendum in the UK cast a ‘long and dark shadow’ over any debate on EU freedom of movement. However, EU freedom of movement still has as many defenders as attackers on the legal, political and academic fronts, as well as in practical everyday contexts. As did Vesco Paskalev, I want to start this contribution with my own personal experience: I am a Slovenian citizen, who studied and worked in EU Member States other than my own, such as Hungary, Italy and the United Kingdom, where I currently reside. I have personally benefited from the rights granted to me under the EU Freedom of Movement Directive (2004/38/EC). This made it a lot easier to be mobile across the EU than it has been for my non-EU colleagues. And with the Brexit vote, I contemplate what my own position will be as a non-UK EU national living and working in the UK in the long run.  

In this contribution I aim to show that both attackers and defenders of free movement share some presumptions in their arguments. In the debate on free movement both often read too much into the potential of freedom of movement and underestimate its limits. At the same time, they are not sufficiently aware of the potential injustice freedom of movement produces as a side effect. I will illustrate both of my claims by focusing on the position of marginalized minorities, who have often been in at the centre of public anxieties about EU freedom of movement. 

Floris de Witte argues that EU freedom of movement is important not only from a state, but also from an individual perspective because it offers opportunities for emancipation as well as a ‘recalibration of justice and democracy’ beyond the state level. Both of his points are well illustrated by his example of an LGBT couple, who move from one EU Member State to another in order to get their union recognised and to lead a life with reduced risk of discrimination. In my view, this example shows how we can read too much into the potential of free movement and underestimate its limits to deliver justice. First, the question is how many EU citizens have genuine access to this right. To give a banal example, it will be a lot easier for a middle-class educated and employed LGBT individual from Zagreb to access it than an impoverished lower-class lesbian or gay man from a rural area in southeast Poland. This is not simply a question of economic means, which De Witte already indicated, but also of the social and cultural capital individuals possess according to Pierre Bourdieu (1). Having a right does not necessarily mean that you have a possibility to access it and that you will indeed do so. Second, I wonder whether freedom of movement can be the main instrument for overcoming inequalities and discrimination marginalized minorities face in the EU, if they are not tackled at the state level first. Is it in fact emancipation and recalibration of justice, when the only option for an LGBT couple is to ‘flee’ their own country to avoid discrimination?  

Saara Koikkalainen investigates the development of EU freedom of movement from its inception within the European Coal and Steal Community, where it followed a strictly economic reasoning. It has only later developed as a fundamental right of all EU citizens. Still this fundamental right is not without restrictions, which are laid down in Directive 2004/38/EC. Article 14 in the Directive states that EU migrants should not represent ‘an unreasonable burden on a social assistance system of the host Member State’. Strictly legally speaking, Martijn Van Den Brink is correct, that CJEU case C-333/13 Elisabeta Dano v Jobcenter Leipzig was in accordance with this article and did deliver justice. But the question is whether the consequences of such decision are just. De Witte and Bellamy both claimed the UK-EU ‘emergency brake’ has very little to do with justice but a lot to do with economic reasoning. I would add that it contributed to the prevalent belief that many EU migrants in the UK are in fact ‘benefit tourists’. This belief goes beyond mere economic arguments and is based on sometimes latent and sometimes open xenophobia. 

The debates on ‘benefit tourism’ are manifestly closely related to the question whether socio-economic disparities within the EU should be primarily addressed through freedom of movement. Are the member states responsible for flattening them or should they be targeted also at the EU level? Many researchers (2) showed how the debate about benefit tourism particularly highlighted the position of another marginalized minority in the EU: Romani migrants. Before 2014 British newspapers were implying that once the work restrictions for Romanian and Bulgarian citizens were been lifted, the UK would face an ‘invasion of Roma’. Newspapers such as Daily Mail adopted a xenophobic stance toward Romani migrants with headlines such as “Roma already ‘defecating at our doorsteps’”(3).

Such reporting reinforced a common misconception that Roma have a propensity to migrate from post-socialist EU Member States to the more prosperous ones, where they would become a burden on the social welfare systems. In accordance with the 1993 Copenhagen Criteria, the EU paid particular attention to the minority protection of Roma in the post-socialist countries, which had a candidate country status before joining the EU in 2004 and 2007 respectively. As some scholars argued (4), this was not merely out of humanitarian concern for this marginalized population. Such emphasis on the position of Roma was also present because of the fear that once the EU free movement policy was coupled with perceived Romani nomadism and discrimination in their own states it would become a conglomerate of push and pull factors for ‘Romani mass migration’. 

However, this is another clear example of reading too much into the potential of EU freedom of movement. According to the very small number of available studies, such as the one by Elspeth Guild and Claude Cahn (5), Romani migrants represent a miniscule proportion of all EU migrants. In addition, their migration cannot be explained simply by the theory of push and pull factors. It also does not clearly fit the presumption that EU freedom of movement is flattening socio-economic disparities between EU Member States especially for the poorest individuals. According to the study by Maria Pantea, Romani individuals who belong to the “poorest of the poor” are among the most immobile EU citizens. Maria Pantea argued that Romani EU migrants have certain economic resources, but even more importantly ‘social ties at work’ (6) and networks that make their mobilities possible. This is something immobile Romani EU citizens lack.  

Considering the position of EU Romani migrants, we can see that in practice EU free movement does not necessarily address injustices produced by nation states. In fact, it can also result in new injustice that is not present on the nation-state level, as De Witte argued. The French L’affaire des Roms showed what measures the French authorities have taken to deport ‘unwanted’ EU Romani migrants. Among these was the collective demolition of their settlements. We read too little into the potential of EU freedom of movement if we only think of it as a source of ‘recalibration of justice’. The official stance of EU freedom of movement is still to a certain extent connected to the economy, but Romani migrants can be deported in case they are also labelled as being a threat to public security and order. Here both opponents and proponents of freedom of movement are reading too little into the matter, if they think this is only a question of economy and if they downplay the sentiments based on xenophobia towards Roma. Although many Romani migrants are labelled as economically inactive, they are only inactive on the official labour market. According to the 2015 Eurobarometer on Discrimination (7) the stigmatization of Roma is so strong in virtually all EU Member States that most of them are not able to get work in the official economy and therefore find employment in unrecognized alternative economic niches. Some studies have shown that many EU Romani migrants end up as irregular workers without employment contracts or even in forced labour (8) as victims of human trafficking. While they are able to migrate because of the EU Directive on Freedom of Movement, they do not benefit from the EU Framework Directive (1989/391/EEC) on safety and health at work and face additional layers of inequality. 

Despite the many objections listed above, I still concur with those who claim that EU free movement should be defended on a normative as well as practical level. But it is only so much that EU freedom of movement can deliver. We cannot expect that as a standalone policy it would ‘recalibrate justice’ for marginalized minorities in the EU, on the one hand. On the other hand, we should take into account that the EU free movement policy does not only belong within a strictly legal domain but has broader societal implications for questions of justice. The Brexit vote showed that EU freedom of movement should be constantly debated and renegotiated not only as a core of EU citizenship, but also beyond that core. This would not imply limiting it, but thinking about it from a global justice perspective. As Sarah Fine’s contribution suggests, this perspective would ask us to consider whether EU citizens should be given a privilege of free movement over all other residents who did not draw the most favourable ticket in the citizenship birthright lottery (9).



(1) Bourdieu, P. (1984),  Distinction. Cambridge, Mass.: Harvard University Press. 

(2) Parker, O. (2012) ‘Roma And The Politics Of EU Citizenship In France: Everyday Security And Resistance’, JCMS: Journal of Common Market Studies 50.3: 475-491; Sardelic, J. (2016) ‘The Position And Agency Of The 'Irregularized’: Romani Migrants As European Semi-Citizens’, Politics.

(3) Available at: http://www.dailymail.co.uk/news/article-2531793/Roma-Britain-defecating-peoples-doorsteps-says-Tory-council-leader-warns-burden-Romanian-Bulgarian-immigrants-place-public-services.html (Accessed: 29th September 2016). 

(4) Kymlicka, W. (2007)  Multicultural Odysseys. Oxford: Oxford University Press.  

(5) Available at:  https://www.gfmd.org/recent-migration-roma-europe-claude-cahn-and-elspeth-guild (Accessed 29th September 2016).

(6) Pantea, M. ‘Social Ties At Work : Roma Migrants And The Community Dynamics’, Ethnic and Racial Studies 36 (11): 1726-1744. 

(7) Available at: http://ec.europa.eu/justice/fundamental-rights/files/factsheet_eurobarometer_fundamental_rights_2015.pdf  (Accessed: 29th September 2016) 

(8) Available at: https://ec.europa.eu/anti-trafficking/sites/antitrafficking/files/forced_labour_and_uk_immigration_policy._status_matters_1.pdf (Accessed: 29th September 2016).

(9) Shachar, A. (2009) The Birthright Lottery. Cambridge, Mass.: Harvard University Press.



What to Say to Those Who Stay? Free Movement is a Human Right of Universal Value

By Kieran Oberman (University of Edinburgh) 


Free movement requires defence, both within the Europe and at the frontier. Within Europe, we are witnessing Brexit, the Swiss vote to limit EU migration and the electoral rise of the far right. At the frontier, free movement has never fared well. The EU has always been something of a gated community, allowing insiders to move while keeping outsiders out. The only difference now, with wars in Syria, Afghanistan and elsewhere, is the higher numbers seeking entry and the higher numbers dying in the attempt. How has Europe responded? The current drive is to reinforce the borders, while calling on “safe” third countries, such as a Turkey, to house refugees. Expect more deaths and more misery in the years to come. 

It is a good time then to be raising Floris de Witte’s question: is free movement worth defending? Like De Witte, I think the answer is definitely yes but I offer a different line of argument. For De Witte, free movement is important in encouraging Europeans to change their values: to move away from a narrow concern with nations, membership and exclusion and towards a cosmopolitan regard for multiple identities and “anchorless” belonging. While this a fascinating and original take on free movement, it seems unnecessarily complex and controversial. Not everyone will accept the cosmopolitan ideological stance it assumes and even those who do might question whether free movement is either necessary or effective at promoting this ideology. The argument I offer is simpler and, in one important sense, less controversial. It defends free movement not as means to change values but rather as an extension of the values we already hold. It also offers reasons for why those who stay in their country of origin should nevertheless value their freedom of movement.

The Human Right to Immigrate

In democratic societies, great emphasis is placed on basic liberties. These basic liberties are protected in international law. The Universal Declaration of Human Rights (UDHR) and the International Covenant on Civil and Political Rights (ICCPR) lists rights to freedom of religion, expression, association, marriage and occupation. These rights are essential to protecting people’s personal and political liberty. In terms of personal liberty, they entitle people to make basic life decisions such as whom they marry, which (if any) religion they practice, with whom they associate, how they communicate and to whom and which job they do. In terms of political liberty, they make it possible for people to engage in crucial political activities such as investigating the effects of government policies, debating solutions to social problems and campaigning in support of a cause. 

Free movement is important because it is prerequisite to the exercise of these other basic liberties. People cannot worship, communicate, associate, marry and work freely unless they are able to move freely. Recognising this fact, international law declares a right to free movement. Article 13 of the UDHR and Article 11 of the ICCPR proclaim a right to free movement within any country and a right to leave any country to go elsewhere.

There is one right, however, that is conspicuously absent: the right to immigrate. This is a problem since immigrations restrictions, no less than emigration restrictions and internal restrictions, curtail personal and political liberty. When foreigners are prevented from entering a country, they are prevented from worshiping, communicating, associating, marrying and working within that country. Their freedom, as well as the freedom of consenting citizens, is constrained. Individual autonomy suffers but also democracy. In an age in which so many problems are international problems and the effects of government policies are felt globally, it is crucial that citizens of different countries are permitted to interact. The power of governments and corporations transcend borders; ordinary people must not be trapped behind them.

If personal and political liberty are to be sufficiently protected, immigration must be recognised as a human right. In recognising immigration as a human right, we discover the full value of EU free movement.  Not only does free movement allow EU citizens to freely interact, it also provides a model for the rest of the world. In time, the world can and should follow Europe’s example.

There are other implications that are less flattering to the EU project, however. Human rights are universal. If EU citizens have a human right to immigrate to other EU states, non-EU citizens do likewise. The Syrians, Afghans and others at the frontier must be free to enter. Refugees have a right to live where they choose, not just the “safe” third countries to which the EU seeks to confine them. Indeed, all migrants have a right to migrate not just refugees.

Floris de Witte’s article thus starts out exactly right. Retired teachers, Romanian IT consultants, Hungarian nurses and everyone else should be able to make basic decisions regarding their lives free from state interference. But that is what is crucial. There is no need to add that free movement encourages people to achieve a new “self-understanding” as “anchorless” EU citizens. The value of free movement is both more basic and more important than that.

The Freedom to Stay

The discussion following Floris De Witte’s article has been fascinating and I have learned a great deal. Let me address two points. First, it is striking that, in the midst of a refugee and migration crisis, the discussion should have focussed so narrowly on free movement within the EU. Sarah Fine is certainly right to pick up on this and ask “whose freedom of movement is worth defending?" If the EU is not to forever remain a gated community, we should at least take note of the gates. 

Second, a number of contributors have raised an important problem: how can supporters of free movement demonstrate its value to those citizens who do not move? Floris de Witte distinguishes between “mobile” and “immobile” citizens; a distinction that Rainer Bauböck picks up on. The question becomes how can the “immobile” be won over? The problem is an important one but, in its general form, is far from new. A central theme of what remains the greatest book on the subject of liberty – JS Mills’ “On Liberty” – is the problem of justifying the freedom to pursue minority options to the disinterested majority. The answer Mill gave then still holds true today. There is an enormous difference between choosing not to pursue an option and being prevented from pursuing it. In the former case, one retains control and, with it, the opportunity to assess how one lives in comparison to alternative possibilities. In the latter case, one never makes a choice; one’s life is dictated by others. Consider the point in relation to freedom of religion. One does not have to be a religious Jew (say) to regard a state ban on Judaism as a violation of one’s freedom of religion. One’s freedom of religion entitles one to have the option to practicing Judaism, even if one never chooses to pursue it. The option is important because in having it one has a source of control over one’s life that is rightfully one’s own. 

There is a further point to be made, however. It is not only that people who stay have an interest the option of moving. It is also that the people who stay are exercising the same basic liberty as the people who move: their freedom of movement. This point is too easily missed.  People tend to assume that freedom of movement is all about movement, when in fact freedom of movement includes the freedom to stay. Freedom of movement entitles one to control over one’s movements. One does not have control of one’s movements if one is forced to move. The point is not purely conceptual. Freedom of movement encompasses the freedom to move and to stay because the same set of interests are at stake in each case. People’s personal and political liberty depends as much on the freedom to stay as the freedom to move. We cannot make our own life decisions and engage in free political activity unless we are free to stay and to move as we wish. 

It is easy to picture the free movement debate in terms of stereotypes: the “young Euro jetsetters” vs. the “resentful go-nowhere locals”. It is also easy to assume that freedom of movement is all about the movers and offers nothing to the stayers. But we need to think again. From a normative perspective, there is no sharp contrast. When we move or when we stay we are engaged in the same core activity: deciding how we spend our lives and with whom. Whatever choices we make, and wherever our choices take us, we should all be able to see how important it is that our choices are our own.

Having come this far, we can now discern something misleading in De Witte and Bauböck’s terminology. The “immobile” category is much too broad. It lumps together people who, due to poverty and other restrictions, cannot move, with people who simply choose not to. In the case of the former, the correct response is to make free movement an effective rather than merely formal freedom by tackling poverty and other barriers to movement. In the case of the latter, the correct response is to remind these people that they are not, in fact, immobile. They have made a choice about where they live and, thereby, exercised their freedom of movement. They should now allow others to do likewise.

Of course, providing a philosophical argument for why everyone should value freedom of movement is not the same as actually convincing them. After Brexit, the task appears daunting. But the problem we face is, in at least one sense, easier than the problem Mill faced in his day. Mill had to convince people of the value of basic liberties. In our day, most people accept the value of basic liberties; they just fail to realise their full implications. 


Union Citizenship for UK Citizens

By Glyn Morgan (Maxwell School, Syracuse University)


In the wake of the Brexit vote, Floris de Witte’s defense of citizenship-based freedom of movement is as important as it is timely. In linking movement to citizenship, as de Witte notices, those who move have a secure status in their new country. In any member state, the new arrival is not a foreigner, not a guest, not someone who has to apologize for being there, but a citizen whose rights are guaranteed by the EU. No one can say: “you don’t belong here.” And if they did; the response would be: “I have the same rights as you to live and work anywhere in the EU.”

Now with Brexit, UK Citizens will lose freedom of movement, and Europeans resident in Britain will lose the protection afforded by Union Citizenship. More worrying still, Brexit threatens to unravel the postwar achievements of European integration. If the UK prospers in the immediate aftermath of Brexit, other countries might follow. A Europe of nation-states will be the outcome. The idea of a unified European polity powerful enough to defend itself and project its values abroad will be lost. 

The EU must act to ensure that Brexit is a failure. It can do this by crafty deployment of a carrot and stick strategy. The stick should come in the form of refusing the UK any privileged access to the Single Market without accepting freedom of movement. No “passporting” for the UK financial services industry—a key component of the British economy—should be allowed. US and other foreign banks should be forced to relocate their headquarters to an EU financial center. The EU should make crafty use of non-trade barriers to hinder the exports of British manufacturers. If the UK wants out of the Customs Union, then the EU should monitor in fine-grained detail, a slow and cumbersome process, the foreign component of UK exports. British visitors to the Continent should be required to attain expensive visas.

The carrot comes in the form of citizenship-based freedom of movement. One step in the right direction would be for the EU to move towards a form of Union citizenship unmediated by any prior national citizenship. At the moment, people in Europe are offered only the status of being hyphenated Europeans (French-European; German-European, Italian-European etc.) rather than Europeans as such. Brexit provides an opportunity here. Sixteen million UK electors voted to remain in the EU. These people will now lose even their meagre hyphenated status and become, for the most part, reluctant national citizens of a country in the grip of populist nativism. The EU can rescue pro-Europeans from their fallen state by offering them Union citizenship –European passports unmediated by national citizenship, which will provide them with the right to live and work anywhere in Europe. Many UK citizens will jump at the opportunity. 

One difficulty with this proposal is that it offers UK citizens an advantage not currently extended to other Europeans, including, most worryingly, those now living in Britain who are threatened with losing their right to live and work there. To address this problem, the offer of unmediated European citizenship for Brits could be made conditional on Britain offering current EU citizens full national citizenship in Britain. Doubtless, the current Conservative Government will reject this suggestion. Alternatively, the offer of EU citizenship for Brits could be made contingent on certain forms of equitable treatment for current EU citizens resident in Britain. Such conditional offers from the EU will further encourage the pro-European British citizens to fight for the rights of current EU citizens in Britain. Any future British government that might wish to play fast and loose with such people will face the ire of the pro-European British eagerly awaiting the opportunity to acquire EU citizenship.

More generally, it might be objected that this citizenship proposal rewards secessionists like Britain by offering UK citizens a desirable form of unmediated citizenship. Surely, this might simply encourage other European member states to follow Britain out of Europe. This objection can be met by charging UK citizens a fee, say €10,000, to acquire European Citizenship. This policy will not only provide the funds to finance the Citizenship Office, which will have to be created de novo, but will discourage countries from thinking that they can secede from Europe while enjoying the full benefits of membership. If €10,000 is too much for some people, they could be offered European citizenship for free in return for working on pro-EU projects, which could be arranged and overseen by the new Citizenship Office. Alternatively, UK citizens could acquire Union Citizenship only if they agree to pay a small tax indexed to their salary.  Needless to say, these are bold and radical proposals, which those familiar only with mediated member-state based citizenship will take time to accept.   

Admittedly, this proposal does nothing to address the concerns of Rainer Bauböck who worries about the divide between the mobile and the immobile citizens. Indeed, in some ways this proposal further exacerbates the division between these two groups.  Nor does it solve Sarah Fine’s concerns about “the growing divide between European citizens and the people they want to keep out.” The proposal does, however, connect with Kieran Oberman’s appeal to John Stuart Mill. Many of Mill’s political writings are the works of a partisan. They are written to and for progressives who find themselves in a society where they are a minority. Mill was forever coming up with clever institutional wheezes and innovative policies that would move the cause of progress along. Knowing how to overcome setbacks is a necessary part of this project. Citizenship for pro-European Brits does not solve all of the problems that now plague post-Brexit Europe. But it offers rewards for people who need encouragement and are the most likely to become the agents of change needed to address the more serious global problems raised by Fine, Oberman, and others.  



UK Citizens as Former EU Citizens: Predicament and Remedies 

By Reuven (Ruvi) Ziegler (University of Reading)


This contribution, like those immediately preceding it, is written in the aftermath of the 23rd June 2016 referendum on the UK continued membership of the EU. At the time of writing, there are precious few known knowns (‘Brexit means Brexit’), critical known unknowns (notably, the nature of future relations between the UK and the EU-27 and ensuing free movement arrangements), and doubtlessly many unknown unknowns. Nevertheless, the premise for this contribution will be that, following negotiations (pursuant to Article 50 of the Treaty on European Union) the UK will cease to be a member of the EU, and the EU treaties will cease to apply to it on exit day (E-day).

Before turning to the legal predicament of UK citizens, and its potential remedies, it would be helpful to consider the effect on the estimated 3.1 million citizens of the EU-27, resident in the UK (unofficial data: Migration Observatory).

EU-27 citizens resident in the UK

The status of citizens of other EU Member States qua EU citizens will not be affected by the UK’s departure. Their ability to continue to exercise their acquired rights in post-Brexit UK would largely depends on decisions which can be made by the UK Parliament irrespective of the outcome of negotiations with the EU-27. 

Hence, the UK could ‘take back control’ over (future) immigration and nevertheless maintain, mutatis mutandis, the arrangements in the Immigration (European Economic Area) Regulations 2006 which implement the Citizens Directive. Nevertheless, while the UK government (July 2016 statement) ‘fully expects’ that the legal status of EU-27 citizens ‘will be properly protected’, it refuses to give assurances before the commencement of the withdrawal negotiations. At the time of writing, parliamentary initiatives (Early Day motion 259, EU Citizens Resident in the UK (Right to Stay) Bill, and a Rights of EU Nationals motion) as well as public campaigning (see here and here) have not yet come to fruition (but one remains hopeful).

Similarly, there is nothing preventing the UK from continuing to enfranchise EU-27 citizens in local government elections (pursuant to section 2 of the Representation of the People Act 1983) even when it is no longer bound to do so in order to implement its treaty obligations. Notably, the UK has broadly interpreted the phrase ‘municipal elections’ in Article 20(2)(b) of the Treaty on the Functioning of the European Union so that EU-27 citizens can currently vote in elections for devolved administrations (Scottish Parliament, Welsh Assembly, Northern Ireland Assembly), for the London Assembly, and for mayors (where they are directly elected). Most notably, EU-27 citizens, resident in Scotland, were enfranchised in the 2014 Scottish independence referendum (see EUDO forum discussion).

In contradistinction, their right to ‘participate in the democratic life of the Union’ (Article 10(3) of the Treaty on European Union) is inevitably going to be affected, as the UK will no longer send MEPs to the European Parliament whom they could elect. Nevertheless, citizens of the EU-27 need not thereby fully lose their right to vote for MEPs. A 2015 European Parliament report notes that 22 Member States of the EU-27 allow their citizens to vote for the European Parliament when they reside in a non-EU state. There are sound reasons for the EU institutions to pressure the remaining five to change their policy, especially in light of the Court of Justice’s Zambrano ratio that Member States should not hinder [43] ‘the genuine enjoyment of the substance of rights conferred by virtue of their status as citizens of the Union’. Moreover, it is not implausible that the Court of Justice’s Delvigne ratio (concerning disenfranchisement of prisoners, discussed here) will prompt a legal challenge to disenfranchisement of Union citizens residing in non-EU Member States. In addition to the legal framework, it is submitted that the policy reasons for enfranchisement apply a fortiori when Member State citizens reside in a former EU Member State.

UK Citizens as former EU citizens

Under the current treaty arrangements, all UK citizens who do not hold the citizenship of another EU Member State will cease to be citizens of the Union, whether or not they have exercised their EU mobility rights: citizenship of an EU Member State is a necessary and sufficient condition for Union citizenship (Article 9 TEU). 

Bauböck lucidly describes a ‘cleavage’ between mobile and immobile EU citizens, characterising the Brexit vote as ‘a political revolt of immobile against mobile Europeans’. But, as Oberman notes, ‘the right to immigrate’ means that today’s immobile EU citizens can choose whether to become mobile in future: indeed, while for citizens of other EU Member States, the choice may come early in life (see Paskalev’s and Sardelic's contributions) many Britons (cue Harry Shindler, excluded from voting in the EU referendum pursuant to the 15 year of non-residence bar) made use of mobility rights in their later years. To borrow Joni Mitchell’s exhortation, for many Britons, the vote to leave may yet turn into a case-in-point of ‘you don’t know what you’ve got till it’s gone’.

The Court of Justice held in Rottmann that, when a Member State strips its citizen of her or his citizenship, the situation [42] ‘falls by reason of its nature and its consequences, within the ambit of EU law’. Consequently [55-56] ‘each individual decision…must be in line with the European principle of proportionality’ and ‘take into consideration…the loss of the rights enjoyed by every citizen of the Union’. Hence, it is perhaps ironic that the prospective en masse stripping of EU citizenship from UK citizens (save for those who are also citizens of another EU Member State) will likely occur without (Court of Justice) judicial scrutiny as and when the treaties cease to apply to the UK. It casts a realistic light on the judgment in Chen where the Court of Justice pronounced that [25] ‘Union citizenship is destined to be the fundamental status of nationals of the Member States.’

Nevertheless, as Advocate General Maduro noted in his opinion in Rottmann [23] ‘Union citizenship assumes nationality of a Member State but it is also a legal and political concept independent of that of nationality. Nationality of a Member State not only provides access to enjoyment of the rights conferred by Community law; it also makes us citizens of the Union. European citizenship is more than a body of rights which, in themselves, could be granted even to those who do not possess it.’ Coupled with the Zambrano ratio (above), the question is: how can the predicament arising from UK citizens’ loss of EU citizenship status and rights be addressed? 

Automatic/accelerated naturalisation of UK citizens (residing) in other Member States 

Politicians in Germany and Italy were reported to have suggested naturalisation of ‘young’ Britons residing in their respective states, and many Britons have (individually) started searching for a nationality link, most notably to Ireland. Now, it is within the gift of Member States to determine their naturalisation criteria, and requires no treaty change; the UK tolerates acquisition of other nationalities. 

However, notwithstanding their well-intentioned premises, such proposals pose substantive challenges: First, they will inevitably lead to divergent treatment of UK citizens across the Union, as some states will not relax their naturalisation requirements to accommodate UK citizens. Second, relaxation of naturalisation requirements (which may include citizenship tests) on an ad hoc basis for one national group may be deemed unjustified. In this context, it is noteworthy that the Union generally encourages states that do not permit dual nationality to relax their objections in relation to the acquisition of the nationality of another Member State (Germany, for instance, generally requires its nationals to obtain permission before acquiring another nationality, save in the case of another EU Member State or Switzerland). In contradistinction, the Maltese ‘citizenship for sale’ (see EUDO debate) caused a degree of discomfort; hence, doubts could be raised as to the propriety of e.g. en masse waver of residence requirements for naturalisation. Third, naturalisation is an inexact remedy: Britons have an effective nationality, albeit one whose instrumental value for free movement will have (potentially, depending on the UK-EU future relations) been adversely affected.

(Partial) decoupling of Union citizenship from Member State citizenship

Morgan’s proposition to create ‘a form of Union citizenship unmediated by any prior national citizenship’ appears to offer a more direct link between the predicament (loss of EU citizenship rights) and the remedy. However, by making the offer ‘conditional on Britain offering current EU citizens full national citizenship in Britain’ or ‘contingent on certain forms of equitable treatment for current EU citizens resident in Britain’, and by making the re-attainment (viz. retention) of EU citizenship financially contingent (€10,000) so as to ‘discourage countries from thinking that they can secede from Europe while enjoying the full benefits of membership’, he significantly weakens the normative basis of his proposition. If the Union ought to be concerned about the individual loss of EU citizenship status and rights, why should its retention be made contingent on policy choices of a former Member States? This seems like the UK government’s ‘bargaining chips’ or ‘cards’ strategy in reverse (it is also noteworthy, for the reasons noted above, that the predicament of EU-27 citizens residing in the UK does not arise from an ineffective nationality). 

Moreover, it is not only morally contestable, but also puzzling how making the re-attainment of Union citizenship costly for individuals will discourage states from ‘seceding’ (withdrawing) from the Union: it seems plausible to assume that (most if not all of) those UK citizens who would wish to retain their Union citizenship had preferred that the UK remain in the Union, but were outvoted. If the Union is concerned about preservation of their individual status and rights, it is due to the disjuncture between their own preferences and the aggregate preferences of their polity. Finally, it is rather unclear whether Morgan proposes to open the Union citizenship route to all UK citizens who do not hold the citizenship of another EU Member State, or just to those who have already exercised mobility rights. 

Dawson and Augenstein argued elsewhere that the decision to withdraw Union citizenship (viz. obtaining Union citizenship through citizenship of a Member State) should rest not with the withdrawing state but with the individual EU citizen, who may either retain or renounce his or her citizenship. This proposition appears more comprehensive and normatively consistent: it would apply to all Britons (whether or not they have exercised mobility rights), recognising their unique predicament as citizens of a former Member State, and thus distinguishing them from citizens of other states, for whom the route to Union citizenship will remain via acquisition of Member State citizenship. It also does not tie their fate to that of the EU-27 citizens. Indeed, there is arguably a qualitative difference (for individuals) between exclusion from club membership and benefits, on the one hand, and non-inclusion therein, on the other hand: many organisations (think universities) retain a special relationship with their alumni. 

However, the decoupling of Union citizenship from citizenship of a Member State would require a fundamental treaty change. What will the creation of two categories of Union citizens do to the self-perception of the EU as an ‘[ever closer] union among the peoples of Europe’ (Article 1 TEU), a ‘demoicracy’ a-la Nicolaïdis? To draw on the university alumni analogy, their privileged status (compared with members of the public) is manifested by retention of access to entitlements generally restricted to members – but they are nonetheless former members. 

UK citizens as Third Country Nationals

Absent treaty change that will address the predicament of UK citizens qua former Union citizens, UK citizens who do not hold the citizenship of another Member State on E-day will become Third Country Nationals (TCN). Mobile UK citizens may qualify as Long Term Residents (LTR). As such, they will benefit from the LTR Directive (as amended in 2011 to extend its scope to beneficiaries of international protection) and from the Right to Family Unification Directive. Both Directives apply in all EU-27 Member States except Ireland (where UK citizens have a free-standing right to reside that preceded Union membership) and Denmark. 

Pertinently for this debate, after five years of continuous residence (Article 4 LTRD) and subject to satisfying additional criteria, LTRs acquire the right to reside in the territory of Member States other than the one which granted them the long-term residence status (Article 14(1) LTRD); following the acquisition of LTR status, they enjoy substantive entitlements under EU law wherever the reside in the Union, including equality of treatment with Union citizens in a wide range of economic and social matters (Article 11 LTRD) and enhanced protection against expulsion (Article 12 LTRD). A fairly modest legislative change to the LTRD that would mitigate the predicament of UK citizens could be the granting of LTR status to mobile UK citizens, irrespective of whether they have met the continuous residence and/or other LTRD requirements.

One of the substantive differences between mobile EU citizens and TCNs concerns political participation. TCNs are not entitled to participate in the ‘democratic life of the Union’; nor are Member States required to enfranchise them in local elections. However, nothing prevents Member States from so doing: indeed, the (limited) number of signatories to the Council of Europe Convention on the Participation of Foreigners in Public Life at Local Level are committed to residence-based enfranchisement on the local level and among the EU-27, eleven states permit TCNs to vote in local elections. The Union could amend the LTRD by disaggregating certain citizenship rights, such as national treatment in respect of local voting rights, from Union citizenship. 

Ruptures in the legal terrain

The Brexit vote came about, in part, due to anxieties surrounding the (perceived) absence of suitable controls on the exercise of the right to free movement by Union citizens. To borrow an earthquake metaphor, the extent to which the aftershocks of 23rd June 2016 will rupture the Union’s legal terrain remains to be seen.  


“Migrants”, “Mobile Citizens” and the Borders of Exclusion in the European Union

By Martin Ruhs (Oxford University)


In his opening contribution to this debate about the future of free movement in the European Union, Floris de Witte concludes that “free movement must be celebrated and defended as the core of EU citizenship, as a right that is available for all 500 million EU citizens, and as an idea that benefits all those citizens – whether they make use of it or not.” [emphasis mine] One of the key reasons de Witte provides for his defence of free movement is that “it makes us sensitive to practices of exclusion”. He argues that “the right to free movement and non-discrimination attached to EU citizenship can be understood to correct instances of injustice and promote the inclusion of outsiders: it makes national distributive systems sensitive to the need to incorporate EU migrants who contribute to the host state in an economic and social way.” 

De Witte is right that free movement serves the important goal of promoting the inclusion of EU migrants (the “outsiders” in de Witte’s analysis) in the economies and societies of EU member states. This is clearly an important achievement of free movement. However, what about the inclusion and exclusion of the much larger group of “outsiders”, namely, people from outside the European Union? If part of our evaluation of free movement is based on its effects on the exclusion of outsiders –and I agree that this should be a fundamental concern – don’t we need to consider all outsiders, not just those from within the EU? 

My main critique of de Witte’s discussion is that it considers free movement in complete isolation of EU countries’ immigration policies and exclusionary practises toward non-EU nationals. This focus on free movement without consideration of wider immigration policies is striking, especially in the context of many EU member states’ highly restrictive policies towards the large number of refugees and other migrants seeking protection in Europe over the past few years. As Sarah Fine points out in her contribution, “how can we try to defend free movement as the core of EU citizenship without considering what is happening right now at (and indeed within) the EU’s own borders?” 

I argue that we need to connect debates about the free movement of EU citizens with discussions about EU member states’ “immigration policies” toward people from outside Europe. This is exactly the opposite approach to the one traditionally taken and advocated by the European Commission and many other European policy-makers who have insisted on a clear distinction between “mobility” of EU citizens on the one hand, and “immigration” of third-country nationals on the other. 

To develop my argument, I first outline some of the key differences between how “migrants” and “mobile EU citizens” are debated and regulated in the European Union. This is followed by a brief explanation of why I think the current distinctions may be considered problematic from both a moral and political perspective. 


There are very large differences between how EU member states currently treat “migrants” from outside Europe and “mobile EU citizens” from within Europe, in terms of both regulating their admission and rights after entry. In all countries, immigration is restricted through an often complex range of national admission policies that regulate the scale and selection of migrants. National immigration policies typically distinguish between high-skilled migrants (who face fewer restrictions on admission), lower-skilled migrants (relatively more restrictions) as well as different rules for admitting family migrants, students, asylum seekers and refugees. 

National immigration policies also place considerable restrictions on the rights of migrants after admission, including their access to the labour market, welfare state, family reunion, permanent residence and citizenship. As it is the case with admission policies, rights restrictions typically vary between high- and low-skilled migrant workers (with the rights of lower-skilled migrant workers significantly more restricted) and across family migrants, students, asylum seekers and refugees. As I have shown in my recent book that focuses on international labour migration; European and other high-income countries’ immigration policies are often characterised by trade-offs between “openness” and some “migrant rights”, that is, labour immigration programmes that are more open to admitting migrant workers are also more restrictive with regard to specific rights (especially social rights). 

Public debates and policy-making on immigration vary across countries but they are typically framed in highly consequentialist terms, i.e. based on the (perceived and/or real) costs and benefits of particular admission policies and restrictions of migrants’ rights for the host economy and society. This cost-benefit approach to policy-making has been a long-standing feature of labour immigration policies.  Arguably, it is also becoming an important factor, and in some European countries the most important consideration, when it comes to polices towards asylum seekers and refugees. Some European countries’ recent policies toward refugees and migrants fleeing conflicts and violence in Syria and other places are primarily shaped by the perceived impacts on the national interest of the host country rather than by humanitarian considerations, protection needs or respect for international refugee conventions.  

A central feature of national migration policy debates in European and other high-income countries is the idea of “control” i.e. the idea that immigration and the rights of migrants can be controlled and regulated, at least to a considerable degree, based on the perceived costs and benefits for the existing residents of the host country. Of course, states’ control over immigration is never complete and subject to a number of constraints but the idea of control is still at the heart of national immigration debates and policy-making.  The perceived “loss of control” over immigration has been a major driver of the rise of Donald Trump in the United States, Britain’s referendum vote to leave the European Union, and the growing support for right-wing parties across various European countries. 

Mobile citizens

The policy framework for regulating the movement of EU citizens across member states, and their rights when residing in a member state other than their own, is very different from the restrictions imposed on people from outside the EU (or the EEA, to be exact). The current rules for free movement give citizens of EU countries the right to move freely and take up employment in any other EU country and – as long as they are “workers” – the right to full and equal access to the host country’s welfare state. This combination of unrestricted intra-EU mobility and equal access to national welfare states for EU workers is an important exception to the trade-off between immigration and access to social rights that characterises the labour immigration policies of high-income countries. Critically, while the idea of “control” is a central feature of debates and policies on the immigration of people from outside the EU, EU member states have effectively no direct control over the scale and characteristics of the inflows of EU workers. From the perspective of the EU, the overall aim has been to encourage rather than limit and control the mobility of EU citizens between different member states.  

In terms of the European institutional framework, free movement is kept completely separate from the immigration of third-country nationals. While free movement is part of the remit of DG Employment, Social Affairs and Inclusion and DG Justice, policies for regulating immigration from outside Europe are dealt with by the DG Migration and Home Affairs. One of the consequences of this division has been that EU debates and policy aimed at the integration of migrants have been heavily focused on migrants from outside the EU. 

A third distinction relates to the terminology used to describe and discuss the cross-border movement of  EU citizens and non-EU citizens. European policy-makers typically insist that EU citizens moving from one member state to another are not “migrants” but “mobile EU citizens”.  (Although I am critical of this distinction, for the sake of clarity I have stuck with this terminology in this contribution.) This distinction is not just a reflection of differences in policy approaches but also serves the purpose of framing public debates in a way that suggests that mobile EU citizens are very different from the (non-EU) outsiders whose migration needs to be carefully regulated and controlled.   

Linking migration and mobility 

The distinctions made in the public debates and policies on immigration and “mobile EU citizens” raise a number of important ethical and political questions. First,  insisting on near-equality of rights for mobile EU citizens while at the same time tolerating what are sometimes severe restrictions of the rights of migrants from outside the EU is, in my view, morally problematic. On the one hand, current policy insists on equality of rights for EU workers including, for example, equal access to non-contributory welfare benefits, i.e. benefits that are paid regardless of whether the beneficiary has made prior contributions or not. On the other hand, many EU member states are unwilling to admit and protect large numbers of refugees who are fleeing violence and conflict and/or grant them full access to the national welfare state. While a preference for protecting the interests and rights of insiders can of course be defended on moral grounds, I suggest that the magnitude of the discrepancy between how EU member states treat each other’s ‘citizens compared to most migrants from outside the EU should give us pause for critical reflection.           

The disconnect between “mobile EU citizens” and “migrants” may also be politically problematic, in the sense that it potentially endangers (rather than protects, as is commonly argued) the future sustainability of the free movement of EU workers within the European Union as well as public support for immigration more generally.  The inflow of “mobile EU citizens” in a particular member state has very similar types of effects, and raises very similar economic issues and tensions, as the immigration of migrants from outside the EU. As it is the case with “migrants”, “mobile EU citizens” affect the labour markets and welfare states of host countries in one way or another, creating costs and benefits for different groups. Insisting that “mobile citizens” are not “migrants” runs the danger of obscuring these impacts that mobile EU citizens have on the economies and societies of their host countries.  This may, in turn, prevent important debates at European level about the consequences of free movement for EU citizens who do not move, and ultimately result in a decline in political support for the free movement of labour within the EU and perhaps also for immigration more generally.  

A third question concerns the potential inter-relationships between EU member states’ policies on immigration and mobility.  How are our policies for the inclusion/exclusion of EU citizens related to our policies for the  inclusion/exclusion of people from outside the EU? We know  that past EU enlargements have in many member states led to more restrictive labour immigration policies for non EU-nationals, especially lower-skilled workers. This may be a perfectly justifiable response within the sphere of labour immigration. The picture gets more complicated and problematic, however, if we consider the potential relationships between the free movement and equal treatment of EU workers and the highly regulated admission and restricted rights of asylum seekers and refugees from outside Europe. How, if at all, do the current policies for the inclusion of mobile EU citizens affect our policies for excluding/excluding asylum seekers, refugees and other migrants from outside Europe – and vice versa? These are open and important issues for empirical research.  Any defence of the current rules for free movement should, in my view, consider and engage with these wider questions and inter-relationships. 



EU Citizenship, Free Movement and Emancipation: a rejoinder

By Floris de Witte (London School of Economics and Political Science)

As was to be expected on a topic such as the relationship between free movement and Union citizenship, the discussion above has been both fruitful and wide-ranging, not in the least due to the decision of the British electorate to leave the EU that took place half-way through this EUDO forum debate (and that throws in doubt the rights of residence of Union citizens in the UK, as well as that of UK citizens in the EU). Rather than replying to the many interesting and insightful contributions individually, I will aim to address some of the themes that transcend the various points of view expressed. These are, in my view, three. 

First, many commentators have suggested that my defense of Union citizenship as being primarily about free movement is insufficiently sensitive to its exclusionary potential. This exclusion may take place at Europe’s borders (think of the refugee crisis), but also within the Member States, where free movement has been understood as being available primarily – and actually – for the young, urban, educated elite. For those left behind – be it at Europe’s borders or at home – free movement, on this account, is not a promise but a problem at the core of Union citizenship. The second theme that can be traced in the discussion is the effect that my understanding of free movement has on the state and its structures. On this view, construing Union citizenship as being centered on its capacity to discipline the nation state and its political processes through free movement creates a number of problems of its own – ranging from the reconstruction of political participation to the destabilization of internal processes of solidarity and will-formation. The third and final theme that many commentators have picked up on, in different ways, is that understanding Union citizenship as being primarily about free movement offers (at best) a partial, inaccurate and normatively unattractive vision of what the individual European subject is. Union citizenship, on this account, ought to be about more than allowing individuals to escape their nation state and its political or normative preferences.

The exclusionary potential of free movement and Union citizenship

In my initial contribution, I have defended the close link between free movement and Union citizenship on account that it liberates the individual from the shackles of majoritarian views ‘at home’, that it recalibrates ideas of justice in a more appropriate fashion, and that it is not premised on a vision of community that is exclusionary. Several commentators have emphasized that this vision is, however, also one that is selective. Sarah Fine highlights that while free movement might be a right that is available for EU citizens, it is also a reason for the creation of ‘hard’ external borders that attempt to keep out non-Europeans, who often find themselves stranded in terrible circumstances outside (or even inside) the EU. Kieran Oberman, likewise, points out that such closure on the EU-level simply recreates the problem that I am trying to overcome: it excludes outsiders, limits their emancipation and capacity to enjoy a range of rights. 

Others, such as Rainer Bauböck, have emphasized that free movement (and therefore Union citizenship) prioritises the needs and aspirations of certain Europeans (let’s say the well-educated, young, urban, and middle-classes) over those of other Europeans (in the traditional account, this group comprises of the elderly, the rural, working classes). This division, as is well documented, also appeared to lie at the core of the electoral split in the UK on Brexit. Julija Sardelić adds to this account that mobility requires certain social and cultural resources that are unavailable for whole swathes of the population. The differentiation that is implicit in free movement, on these accounts, jars with the basic premise of equality that is central to our understanding of citizenship. 

Let me take these arguments in turn, starting with the exclusionary potential of free movement and Union citizenship in the most dramatic – territorial – sense. Sarah Fine is certainly correct to highlight that the process of integration and its manifestation as Fortress Europe has dramatic consequences for those that cannot claim a right to free movement or Union citizenship. And certainly the institutions of the EU have placed a dodgy understanding of vulnerability at the core of its external border policy: one that understands the EU and its Member States to be vulnerable from intrusion by ‘the other’ as an object, rather than one that understands the refugees as the subject of vulnerability. This process can, possibly, indeed by reduced to the creation of a category of Union citizen that remains the main subject of the EU legal order, and main preoccupation of its institutional order, as Martin Ruhs claims. But the argument that reduces the drama unfolding at the Union’s borders (and within the Union’s borders) to a problem that is created by free movement and Union citizenship conflates two – admittedly intertwined – processes. 

Borders matter. Borders construct polities and engender certain relationships. And those relationships matter for the way in which we attribute rights and obligations. Many of us would not feel particularly inclined to share our resources with, say, a Russian oligarch. Most of us, conversely, would give up just about anything for the sake of our children or partner. Between these two extreme poles, our commitment to others depends on a range of factors – from the agapistic reflex that extreme suffering generates, to ideas of historical, ethnic, linguistic communities and from those revolving around sexual orientation, gender or political preferences to those created by shared institutional frameworks. On this relational account, the creation of the EU is, in its simplest form, the institutionalization of new relationships between citizens across borders. Something links the Polish national to the Belgian national that does not include their Ukrainian (or Australian, Ecuadorian, Senegalese) neighbour. The relationships generated by Union citizenship, in other words, must mean something – in my account primarily a shared commitment towards opening up national communities. That does not, of course, mean that non-EU citizens ought to be treated poorly, that they are somehow undeserving of protection, admission to the territory of the EU or help. What it does mean is that these are two conceptually separate discussions. The extent to which we defend or limit free movement of Union citizens, and the matter in which we construe internally the rights and obligations of those EU citizens based on their inter-personal relationship as institutionalized by EU citizenship, has little to do with how we treat those outside our borders of membership. Conditions, obligations, and rights of membership are bounded: they include members, and exclude non-members. This applies to book clubs, terrorist organisations, and transnational political communities. What we owe refugees fleeing war in atrocious and hazardous circumstances is a question that is distinct from the question what we owe fellow European citizens. The difference – which also explains why the former is so difficult to answer legitimately and authoritatively – is that one of these questions has been institutionalized, and the other has not: we have legal norms and institutional structures that translate the ill-defined bond between Europeans into Union citizenship. And that institutionalization, in turn, is only possible once we accept that borders matter. Not as instruments to keep people out, but as instruments to solidify the relationships between those inside the borders. This also means that, contrary to Glyn Morgan’s suggestion, if the UK were to leave the EU its citizens cannot remain Union citizens, and cannot derive rights and obligations under that heading. 

But the close link between free movement and Union citizenship has a second exclusionary effect – one that is internal to the EU. On this view, Union citizenship is an advantage for those willing and able to make use of it; but a disadvantage for those who cannot. On this narrative, the ‘immobile’ citizen faces increased competition for jobs and welfare resources from the ‘mobile’ citizen. Citizenship and free movement, then, create winners and losers – a process that became blatantly obvious in the Brexit-vote. I would contest this argument both normatively and institutionally. The fact that only a portion of citizens actually make use of their right to free movement does not make the right any less important or relevant. A small proportion of the population makes use of their rights to free association or freedom of expression. Increasingly fewer people use their right to vote. Does that make these rights increasingly less relevant or important? Of course not. The same applies to free movement. The mere possibility of movement, legally guaranteed by free movement and Union citizenship, moreover, also has a reflexive virtue, as picked up by Vesco Paskalev (who highlights that younger Brits – regardless of their exercise of free movement – understand it to be a public good) and Saara Koikkalainen (who shows that 78% of EU-wide respondents support free movement). The possibility of free movement liberates the self-understanding of the individual from the collective self-understanding of the polity they happen to be born in. 

This does not mean that Union citizenship and free movement offer an equal opportunity of exercise to all Union citizens. As I will discuss below, the EU does not dispose of the institutional framework that can articulate and sustain such an understanding of substantive justice. To be sure, free movement presupposes certain social and cultural resources (if not necessarily those associated with the ‘transnational elite’ – it is hardly against these groups that the Brexit-vitriol was directed). What matters for our argument in this section, however, is not that free movement has high conditions of entry, but that it creates negative effects for those that do not make use of it (for whatever reason). This can be explained in one of two ways. First, and in the most important narrative in the Brexit campaign, it was suggested the EU migrants not only take jobs away from UK citizens, but also welfare resources, and create increased pressure on schools and hospitals. At the same time, every single piece of research that has taken place suggests that the UK benefits fiscally from free movement. How can these two positions be squared? It’s very simple: by the decision of successive UK governments not to invest the fiscal windfall from free movement in additional welfare resources such as schools and hospitals. 

The other way of explaining the cleavage between pro-mobile and anti-mobile citizens is not to focus on the economics, but on identity politics. Rainer Bauböck, for example, refers to Theresa Kuhn’s book in highlighting that individual experiences of transnationalism (‘lived experience’) makes them more pro-European and pro-free movement. On this view, it is to be expected that, say, Gibraltar, would vote in favour of staying in the EU with 96%, while, say, Stoke-on-Trent would vote with 69.4% against it. But that argument understands the idea of ‘transnational experiences’ much too narrowly. The ‘Leave’ voters in Stoke eat pizza and drink Stella Artois. The local pride – Stoke City football club – combines English players with Dutch, Austrian, Spanish, Italian and Irish stars. These kind of less obvious transnational experiences matter as they locate elements of ‘the other’ in something we consider our ‘own’. The social capital built up through free movement extends much beyond instances of Erasmus or holidays abroad. It pervades our world. 

But if both the economic and the sociological argument explaining the cleavage between pro-mobile and anti-mobile are at best oversold, how can we explain its indisputable emergence? The starting point in this analysis was intimated above – and that is that many of the citizens that have rejected the EU have indeed been ‘left behind’. But the main culprit is their own government, not the EU. This is different, of course, in instances where the EU mandates welfare reform through its austerity drive. But blaming welfare scarcity in the UK on the EU seems a bit rich. Yet why do voters blame the EU and not the UK? This is where Brexit reveals a more structural problem for the EU, as I have discussed elsewhere. It is because the EU cannot institutionalize contestation appropriately. In a well-functioning democracy, discontent that is so widely spread as the Brexit-cleavage is internalized in the system, and mediated through the politics of contestation. In such a system, political conflict feeds into the process of decision-making and stabilizes the overall project. Democracy, on this view, is a safety valve for emergent discontent. In the EU, on the other hand, the nature, conditions, and consequences of free movement cannot be contested. The only possible way to contest the normative orientation of the European market is to leave the EU. And so we see ‘hard left’ parties claiming that if we want to resist the neo-liberal nature of the EU, the only thing we can do is leave it. And so we have parties on the extreme right claiming that if we want to resist immigration (for whatever reason), the only thing that we can do is leave the EU. If the EU doesn’t start to think about how it can internalize and institutionalize contestation of its values, Union citizenship and free movement will indeed be seen as something that divides, that creates winners and losers. Not necessarily because it does create winners and losers, but because its contours and effects are not mediated through a process that can legitimize and explain the outcome.

What Union citizenship and free movement do to the state

A second theme that has emerged in the discussion is that my account of Union citizenship as ‘anchorless’ and as getting round the ‘ethno-centric’ vision of the nation state, on the one hand, underestimates the virtues of national institutional structures in solidifying inter-personal commitments and general will-formation and, on the other hand, underestimates the power of free movement as an instrument for cosmopolitan justice. These two points – made by Richard Bellamy and Kieran Oberman, respectively – suggest that my argument takes seriously neither the nation state nor cosmopolitanism. Instead, it is precariously perched between these two poles: free movement and non-discriminatory access to welfare benefits for EU migrants, on this account, are parasitical on domestic political commitments and extend them across borders without succeeding in fully disentangling them from the nation state and its structures. In simple terms, it destabilizes the nation state without replacing it with the promise of egalitarian cosmopolitanism. More is lost than gained in the exercise.  

Richard Bellamy highlights that my argument is premised on the quest to create “a fully fledged political and legal cosmopolitanism that looks to the ultimate demise of nation states as a necessary condition for justice”. He suggests that in doing so I misrepresent the state. The modern-day nation state is less exclusionary and more pluralist than I have presented it to be; and it is, crucially, indispensible for the determination, enforcement and stability of sharing practices. On the first point Bellamy is partially right: it is empirically demonstrable that membership to national political communities has never been more inclusive and pluralist. But this has clearly not affected the capacity for exclusionary and ethno-centric political narratives to control the political process – across the EU (and the US). It seems that the more diverse and inclusive our societies have become in terms of their membership, the less sensitive their politics become to diversity and inclusion of those that are not members. More inclusive membership thus does not equate to a pluralist society. On the second point Bellamy is completely right: the nation state remains absolutely indispensible in the determination, enforcement and stability of sharing practices and the processes of collective will-formation. My argument is not based on the rejection of the state – and on the slow process towards a political cosmopolitanism. Instead, it is based on the realization that, given that the state and its institutional structures are indispensible for structuring authority in a legitimate fashion, we must be sensitive to its externalities. On this view, EU citizenship and free movement are not meant to obliterate the state; they are meant to limit the externalities of the institutional structure of the nation state. The first externality has been discussed at length in my initial post: that the nation state limits choices available for the citizen. The trade-off in liberal democracies is that in return for your chance to vote, you accept the majority position. Free movement cuts across this limit and offers Union citizens the choice of different visions of ‘the good’. 

The second and more structural externality, which Bellamy has picked up, is most easily explained if we pick up an element discussed above – that demands of justice are relational. As Païvi Johanna Neuvonen also reminds us, relationships between individuals that generate claims of justice or solidarity might take place within borders, but also across borders. The point that my initial contribution makes is essentially that while we have institutions that can make sense of the former type of relationships (namely, the domestic political process, that can mediate between competing claims of justice by insiders), it cannot possibly make sense of the relationships across borders. Imagine an Irish woman living in France and working in Belgium. What does the social relationship between her and French society mean in terms of justice? What does Belgium ‘owe’ her because of her economic participation in the Belgian market? Allowing the French or Belgian political system to answer these questions creates a democratic problem, as our Irish women (and therefore her relational position vis-à-vis nationals) are not included in its determination. This is where and how free movement and Union citizenship, and particularly through the principle of non-discrimination based on nationality serve to more appropriately settle questions of justice and democracy. Union citizenship and free movement respect the institutional structures of the nation state in the determination and enforcement of questions of justice. As far as EU law is concerned, the Austrian decision to have free tertiary education is as ‘just’ as the UK’s decision to charge £9000 per year. This respect for national choices is indeed, as Bellamy highlights, necessary for its stability. What free movement law does is simply extend that decision to include those Union citizens who have a sufficiently strong relationship with the host state – be it through economic or social interactions. As such, free movement and Union citizenship do not serve to substitute for the nation state and its structures, they serve, instead, to recalibrate questions of justice in a fashion that is more sensitive to the relationships across border that the EU has engendered, to which the nation state is structurally blind.

This idea of Union citizenship as supplementing national citizenship can, of course, be deduced from the text of the Treaty. It also means that Union citizenship and free movement are not codifications of a sort of cosmopolitanism as suggested by Kieran Oberman. Free movement, on my account, is not a good per se. It is a good because of the way in which it recalibrates domestic processes of will-formation and sharing – not because it seeks to replace or subvert them. It creates emancipatory potential for individuals exactly because they can navigate between existing institutional articulations of justice and ‘the good life’. Without those institutional structures of the nation state, free movement and equal treatment is pointless. Being equally entitled to nothing, after all, entitles you to nothing. This leads to me to the final point that I made in my initial contribution. The way in which Union citizenship and free movement attempt to internalize relationships across borders within domestic institutional structures is normatively appealing exactly because it piggy-backs on those domestic structures. There is no need for the articulation of a European form of ‘being’ that can integrate and structure its own idea of community and political form. EU citizenship is, in a sense, agnostic. 

A normative vision for Union citizenship

The final theme that has come through the discussion is that this link between free movement and citizenship that is central in my account makes for a normatively and institutionally impoverished vision of justice. This critique comes in many flavours. Daniel Thym highlights that the Court’s ruling in Dano, for example, demonstrates the absence of a robust or ‘thick’ principle of justice in the Court’s understanding of what free movement and Union citizenship mean. More broadly, he argues that even if free movement once constituted the core of a normatively ambitious idea of what it means to be a European citizen; the Union has now lost constitutional confidence, and has become more deferential to domestic ideas of belonging, sharing and justice. What is needed, then, is a thicker vision of social justice that engages all Europeans, whether they move or not. Criticism of this lack of a more coherent and nuanced idea of justice can also be traced in other contributions. Johanna Païvi Neuvonen suggests that the normative centrality of free movement understands the subject as atomistic and unencumbered – which makes for a narrow, individualistic, and, ultimately, not particularly emancipated, self. Emancipation, after all, requires the capacity to encounter the ‘other’ as part of the construction of the self. Vesco Paskalev argues that the ease of movement across borders destabilizes another element that is crucial to justice and citizenship: that of (equal) political participation and engagement. On these accounts, free movement does not suffice. Union citizenship should be about more than simply free movement if it is to be normatively appealing. I broadly agree on these points, in so far as they indicate that Union citizenship offers a partial vision of justice at best. It does not set out a vision of substantive justice for 500 million Europeans. It does not deal well with instances of discrimination that transcend borders, as Julija Sardelić reminds us.

The problem is that Union citizenship can never offer more than a partial vision of justice unless we recreate on the European level the institutional preconditions that we find on the national level (and that I have argued in favour of above). To use the example mentioned above: is it more ‘just’ to fund tertiary education through general taxation or make the student pay himself? If we have a fiscal windfall of €400 million, should we spend it on healthcare, pensions, or education? These questions can only legitimately be answered if they are discussed and mediated through a ‘thick’ democratic institutional structure – of which the EU is a very very distant cousin. The EU simply does not possess the institutional structure required to answer the question: ‘what is it to be European?’ or ‘what do we Europeans owe each other’. And this institutional incapacity leaves us in a double bind: national political processes are sufficiently ‘thick’, but structurally exclude relationships across borders from consideration. European political processes are too thin to articulate a substantive vision of justice for all 500 million Europeans. What we are left with is the legal norms of free movement and EU citizenship that seek (and not always succeed) to figure out what these relationships across borders mean in terms of justice. The idea of justice in the EU, in a sense, is tiered: it depends on both national institutional structures and transnational legal norms. Free movement and Union citizenship, then, as Païvi Johanna Neuvonen forcefully argues in her comment and her recent book, may not be sufficient to achieve justice in the EU – but they are necessary. Unless and until the EU develops its institutional structure in a way that is more sensitive to the views of its citizens, this is as good as it gets. 

The institutional implication of this argument is that the scope and limits of free movement and Union citizenship cannot be decided through political structures as they currently are. Martijn van den Brink has argued that the Union legislator offers a democratic – if not particularly ambitious – vision on what free movement means, and the Court in Dano was right in accepting this vision. Union citizenship, in his view, is not a normative commitment towards emancipation and the limitation of state power, but its opposite: an expression of state power. Member States, acting together in the Council, have the right to decide when and under which conditions free movement is possible, and to decide what the limits of EU citizenship ought to be. This argument underestimates the level of institutional sophistication that is required for a system to be able to ‘translate’ inter-personal relational claims of justice and solidarity into a legitimate and enforceable system of rights and obligations. As Richard Bellamy has highlighted as well, the institutional presuppositions for this task cannot be found beyond the nation state. The EU lacks thick representative, deliberative and participatory elements, it lacks the support cast of integrated political parties, civil society, grassroots movements, transnational media that allow for inter-personal communication about the question: ‘how do we want to live together in this social space?’ ‘What do we owe each other by virtue of our shared participation in the EU?’. The Union legislator cannot possibly get the answer to these questions ‘right’. So we are left in an institutional situation where these incipient, partial, and ill-defined bonds and relationships between Europeans, created by transnational economic, social, political and cultural links cannot be articulated by either national legislatures or their European counterparts. Ill-equipped as the Court may be to makes sense of these new relationships, it is the only institution that can make good on its premise: that it must mean ‘something’ to be a Union citizen beyond what it means to be a national of a Member State. 

Bloodlines and Belonging: Time to Abandon Ius Sanguinis?

Bloodlines and belonging: Time to abandon ius sanguinis?

By Costica Dumbrava (Maastricht University)


The transmission of citizenship status from parents to children is a widespread modern practice that offers certain practical and normative advantages. It is relatively easy to distribute legal status to children according to parents’ citizenship, especially in the context of high mobility where the links between persons and their birthplace are becoming increasingly strained. Granting citizenship status to children of citizens may also be desirable as a way of avoiding statelessness, acknowledging special family links and fostering political links between children and the political community of their parents. These apparent advantages of ius sanguinis citizenship are, however, outweighed by a series of problems. In what follows I argue that ius sanguinis citizenship is (1) historically tainted, (2) increasingly inadequate and (3) normatively unnecessary. Ius sanguinis citizenship is historically tainted because it is rooted in practices and conceptions that rely on ethno-nationalist ideas about political membership. It is inadequate because it becomes increasingly unfit to deal with contemporary issues such as advances in assisted reproduction technologies and changes in family practices and norms. Lastly, ius sanguinis citizenship is normatively unnecessary because its alleged advantages are illusory and can be delivered by other means.


As a key instrument of the modern state, the institution of citizenship has been closely linked to nationalism. Ius sanguinis citizenship was reintroduced in Europe by post-revolutionary France, which sought to modernise French citizenship by discarding feudal practices such as ius soli.[1] Whereas in modern France the adoption of ius sanguinis was premised on the idea of a homogenous French nation, in countries with contested borders, such as Germany, ius sanguinis played a key role in maintaining ties with co-ethnics living outside borders and thus in nurturing claims to territorial changes. Although ius sanguinis citizenship is not conceptually “ethnic” (in the same sense in which ius soli citizenship is not necessarily “civic”), there are a number of ways in which the application of the ius sanguinis principle has been used in order to promote ethno-nationalist conceptions of membership.

Firstly, the application of unconditional ius sanguinis in the context of a long history of emigration means that emigrants can pass citizenship automatically to their descendants regardless of the strength of their links with the political community. No less than twenty countries in Europe maintain such provisions.[2] Whereas one can find several non-nationalist arguments for justifying emigrants’ citizenship, these weaken considerably when applied to successive generations of non-residents.

Secondly, there are cases in which countries rely on the principle of descent in order to confirm or restore citizenship to certain categories of people whom they consider to be linked with through ethno-cultural ties. Apart from cases where ethnic descent is an explicit criterion of admission (e.g. in Bulgaria, Greece), there are countries where ethnicity is camouflaged in the language of legal restitution or special duties of justice (e.g. in Latvia, Romania). In this way, persons can have their citizenship status “restored” on the basis of descent from ancestors who had been citizens or residents in a territory that once belonged, even if briefly, to a predecessor state with different borders.

Thirdly, the combination of unconditional ius sanguinis citizenship with the reluctance to accept alternative ways of incorporating children of residents (such as ius soli) is also a strong indicator of an ethnic conception of citizenship, especially in the context of a long history of immigration. Convoluted attempts to adopt and expand ius soli provisions in Germany and Greece illustrate this point. In 2000 Germany adopted ius soli provisions[3] but maintained that, unlike persons who acquire German citizenship through ius sanguinis, those who acquire citizenship via ius soli could retain it only if they relinquish any other citizenship before their 23rd birthday.[4] In 2011 the Greek Council of State halted an attempt to introduce ius soli citizenship in Greece[5] by claiming that ius sanguinis is a superior constitutional principle whose transgression would lead to the “decay of the nation”.[6]


Consider the following two real cases.

Samuel was born in November 2008 in Kiev by a Ukrainian surrogate mother hired by Laurent and Peter, a married gay couple of Belgian and French citizenship respectively.[7] Samuel was conceived through in vitro fertilisation of an egg from an anonymous donor with Laurent’s sperm. Upon his birth and according to practice, the surrogate mother refused to assume parental responsibility and thus transferred full parentage rights to Samuel’s biological father. When Laurent requested a Belgian passport for Samuel, the Belgian consular authorities refused on grounds that Samuel was born through a commercial surrogacy arrangement, which was unlawful according to Belgian law. After more than two years of battles in court, during which Laurent and Peter also attempted and failed to smuggle Samuel out of Ukraine through the help of a friend pretending to be Samuel’s mother, a Brussels court recognised Laurent’s parentage rights and ordered authorities to deliver Samuel a Belgian passport. With it, Samuel was able to leave Ukraine and settle with Laurent and Peter in France.

In 2007 Ikufumi and Yuki, a married Japanese couple, travelled to India and hired Mehta as surrogate mother for their planned child.[8] Using Ikufumi’s sperm and an egg from an anonymous donor, the Indian doctors obtained an embryo, which they then implanted in Mehta’s womb. Only one month before the birth of Manji, the resulting child, Ikufumi and Yuki divorced. When Ikufumi attempted to procure a Japanese passport for Manji, the Japanese authorities refused on grounds that Manji was not Japanese. According to the Japanese Civil Code, the mother is always the woman who gives birth to the child. Despite having three “mothers” – a genetic mother, who contributed with the egg, an intended mother who later declined involvement, and a surrogate mother, who did not plan to take up parental responsibilities – Manji had no obvious legal mother. Indeed, Manji’s Indian birth certificate mentioned Ikufumi as the father but left the rubric concerning “the mother” blank. After much legal wrangling Manji was issued a certificate of identity stating that she was stateless, with which Ikufumi managed to take her to Japan.

These are just two of a growing number of cases that test the legal and normative linkage between human reproduction, legal parentage and citizenship. Not only do they question conventional assumptions about the biological and cultural basis of citizenship, but they also show the limits of the principle of ius sanguinis in ensuring the adequate determination of citizenship status.

The incongruity between reproduction, legal parentage and citizenship is not an issue triggered solely by advances in reproductive technologies. Traditionally, children born out of wedlock could not acquire the father’s citizenship through descent. Many countries still maintain special procedures for the acquisition of citizenship by children born out of wedlock to a foreign mother and a citizen father. In most cases this implies submitting a request for citizenship after parentage is legally established, although in Denmark these children can acquire citizenship only if the parents marry. In the Czech Republic and the Netherlands (for children older than 7), the determination of parentage for the purpose of citizenship attribution requires showing evidence of a genetic relationship between the father and the child. As argued by the European Court of Human Rights in its 2010 judgment on Genovese v Malta[9], the differential treatment of children born within and out of wedlock with respect to access to citizenship amounts to discrimination on arbitrary grounds. This practice is also at odds with contemporary trends that indicate an impressive surge in births out of wedlock; the share of such births in the EU27 rose from 17% of total births in 1990 to 40% in 2013.[10]

One of the biggest challenges to ius sanguinis citizenship comes from the spread of assisted reproduction technologies (ART). About 5 million babies worldwide have been born through ART since the birth of Louise Brown, the first “test-tube baby”, in 1978.[11] ART have developed rapidly generating a multi-billion dollar market in assisted reproduction. A significant share of this market involves the international movement of doctors, donors, parents, children and gametes. In order to avoid legal restrictions or to cut costs, a growing number of infertile men and women, usually from high-income countries, travel to destinations such as India, Thailand or Ukraine in order to have “their” babies conceived through in vitro fertilisation procedures using sperm or eggs (or both) donated by people from places such as Spain or Romania.

Many problems arise because the international market for assisted reproduction is not properly regulated, which means that national regulations often conflict with one another. Countries that oppose surrogacy consider the surrogate mother as the legal mother even if they are not genetically related to the child. According to this reasoning, the husband of the surrogate mother is the presumed father of the child. However, countries that encourage surrogacy usually recognise the intended mother and father as the legal parents, regardless of whether they are genetically related to the child. As the stories on Samuel and Manji show, when these two approaches collide the children risk becoming, as Justice Hedley put it, “marooned, stateless and parentless”.[12]

In some cases intended parents have the possibility to establish parentage and citizenship for their children born through surrogacy. However, such special arrangements often discriminate between (intended) mothers and fathers. For example, in the US children born to surrogate mothers outside the country are treated as children born out of wedlock, so fathers can be recognised as legal parents and therefore extend citizenship to children if they provide proof of a genetic relationship with the child (through a DNA test). Intended mothers, however, cannot be recognised as mothers even if the child was conceived using their eggs and even if they are married to the intended father.[13] It follows that, in cases where another woman’s womb is involved, paternity and citizenship can still follow the sperm but not the eggs.

The practice of gamete donation has become increasingly accepted and regulated, so donors are in principle discharged of parental responsibilities with regard to children they help to conceive. However, it is not always clear what counts as donation. In a recent US case, a man successfully claimed parentage with regard to a child who was born after an informal agreement in which he agreed to “donate” sperm to a friend. The Court decided in the man’s favor arguing that his act did not count as donation because the procedure used in the insemination did not involve “medical technology” (they used a turkey baster). The ultimate test of paternity in this case relied on a mere technicality, which can hardly be seen as a morally relevant fact for establishing fundamental ties of filiation and citizenship.[14]  

The development of ART is likely to further complicate questions about parentage and citizenship. The new techniques of embryo manipulation, for example, make now possible the transfer of a cell nucleus from one woman’s egg to the egg of another, which means that the resulting child will have three genetic parents. Advances in technologies for freezing gametes and embryos raise questions about the rights and responsibilities over future births and about the status of future children. There have already been a number of cases of posthumous conception in which the sperm or eggs of a deceased person were used by the spouse or another relative in order to conceive children. For example, it was recently reported that a 59 years old woman from the UK gave birth to “her” daughter’s child.[15] These practices raise obvious questions as to whom these children belong to and they may as well trigger issues of citizenship. Lastly, progress has been made on the creation or “artificial” gametes through the modification of other types of human cells. Apart from opening possibilities for bypassing the heterosexual model of procreation[16], these techniques raise concerns about abuse or reproductive “crime”. Imagine a world in which it would be possible to create a child from a tissue sample collected from somebody’s cup of coffee. Those famous actors and footballers would probably think twice before shaking their fans’ hands.


One could argue that the main problems do not lie with ius sanguinis citizenship but with the determination of legal parentage. Once we solve issues related to legal parentage, then the ius sanguinis principle will effectively address citizenship matters. However, this view ignores that dilemmas regarding the attribution of parentage are often triggered or complicated by citizenship (and migration) issues. It can also be argued that relying solely on legal parentage to settle citizenship issues disregards fundamental normative questions about who should be a citizen in a political community.

Despite much liberal-democratic talk about social contract, democratic inclusion and active citizenship, the overwhelming majority of people in the world acquire citizenship by virtue of contingent facts about birth (descent or place of birth). While ius soli citizenship has received considerable political and academic attention recently due to pressing concerns about the inclusion of children of immigrants, ius sanguinis continues to be taken for granted. In the remainder of this essay, I briefly challenge two main theoretical defences of ius sanguinis: (a) that ius sanguinis citizenship recognises and cements the special relationship between the parent and child; (b) that ius sanguinis citizenship ensures the intergenerational stability of the political community.

The main problem of ius sanguinis citizenship is that it is parasitic on external factors concerning the legal determination of parentage. As one of the examples presented above shows, it may only take a choice between a petri dish and a turkey baster to make somebody a parent and hence a supplier of citizenship status. The relevance of horizontal family ties between spouses in citizenship matters has largely diminished, as a flipside of the spread of gender equality norms, since in liberal states wives no longer automatically acquire their husbands' citizenship. By contrast, parental ties continue to remain paramount for the regulation of citizenship. Even if there are good reasons for seeking to ensure the swift transfer of citizenship from parents to children (e.g. to prevent statelessness), this approach is questionable because it renders children vulnerable. Ius sanguinis citizenship makes access to citizenship for children dependent on parents’ legal status, actions or reproductive choices.

As in the case of spouses, joint citizenship adds little to the legal and normative character of the parent-child relationship.  There is little doubt that the law should treat children and the parent-child relationship with special attention. However, this could and should be achieved regardless of the citizenship status of children and parents. One could, for example, extend the legal rights associated with parentage and filiation (e.g. conferring full migration rights to children of citizens) or seek to establish a universal status of (legal) childhood that confers fundamental right and protection to children regardless of their or their parents’ citizenship or migration status.

The second argument for ius sanguinis citizenship is that the automatic transition of membership status from parents to children ensures the smooth reproduction of the political community. As children of citizens grow, they become socialised in the political community of their parents and develop political skills necessary for furthering their parents’ project of democratic self-government, skills that they will eventually pass on to their own children. An easy objection to this view is that it is empirically naïve, especially in the context of increased migration and diversification of family practices. Citizenship is thus based on a contested expectation. Instead of granting citizenship ex-ante to persons who are likely to develop desirable citizenship attitudes and skills, we could delay the attribution of citizenship until such attitudes and skills are confirmed. Alternatively, there may be other normative considerations for turning children into citizens. For example, being born in the country and/or living there at a young age makes children not only subject to the law of the country but also highly dependent on the state, which, for example, is required to provide regular and reliable access to medical care such as vaccinations. These considerations could justify granting children at least provisional citizenship.

The intergenerational dimension of democratic membership can hardly be achieved by relying on legal fictions or on biological contingencies. Our efforts should rather be channelled towards consolidating democratic institutions and promoting citizenship attitudes and skills among all those who find themselves, by whatever ways and for whatever reasons, in our political community. As for the children who happen to be born here, we should treat them as political foundlings and give them all the care and support they need to become full political members.



[1] Weil, P. (2002). Qu' est-ce qu'un Français? Paris: Grasset.

[2] Dumbrava, C. (2015). Super-Foreigners and Sub-Citizens. Mapping Ethno-National Hierarchies of Foreignness and Citizenship in Europe. Ethnopolitics 14(3), 296-310. http://dx.doi.org/10.1080/17449057.2014.994883.

[3] Hailbronner, K. and A. Farahat (2015). Country Report: Germany. EUDO Citizenship Observatory. http://cadmus.eui.eu/bitstream/handle/1814/34478/EUDO_CIT_2015_02-Germany.pdf?sequence=1.

[4] Bock, L. (2015). Germany: As ‘Option Duty; reform comes into force, first figures on individuals having lost their citizenship emerge. Citizenship News, EUDO Citizenship Observatory. http://eudo-citizenship.eu/news/citizenship-news/1342-germany-as-option-duty-reform-comes-into-force-first-figures-on-individuals-having-lost-their-citizenship-emerge.

[5] Christopoulos, D. (2011). Greek State Council strikes down ius soli and local voting rights for third country nationals. An Alarming Postscript to the Greek Citizenship Reform. Citizenship News, EUDO Citizenship Observatory. http://eudo-citizenship.eu/news/citizenship-news/444-greek-state-council-strikes-down-ius-soli-and-local-voting-rights-for-third-country-nationals-an-alarming-postscript-to-the-greek-citizenship-reform

[6] The Greek parliament has recently pushed forward another proposal regarding ius soli in an attempt to overcome the deadlock. See (2015). Greece: New citizenship bill now opened for public consultation. Citizenship News, EUDO Citizenship Observatory. http://eudo-citizenship.eu/news/citizenship-news/1385-greece-new-citizenship-bill-now-opened-for-public-consultation.

[7] The European Parliament (2013). A comparative study on the regime of surrogacy in EU Member States. Study for the Directorate-General of Internal Affairs, Policy Department: Citizen’s Rights and Constitutional Affairs, pp. 90-1. http://www.europarl.europa.eu/RegData/etudes/etudes/join/2013/474403/IPOL-JURI_ET(2013)474403_EN.pdf.

[8] Points, K. (2009). Commercial surrogacy and fertility tourism in India: The case of Baby Manji. Case Study, Kenan Institute for Ethics, Duke University. https://web.duke.edu/kenanethics/CaseStudies/BabyManji.pdf.

[9] Genovese v. Malta, Application no. 53124/09, European Court of Human Rights, 11 October 2011. http://hudoc.echr.coe.int/sites/eng/pages/search.aspx?i=001-106785#

[10] BBC News (2013). Two in five EU babies born out of wedlockhttp://www.bbc.com/news/world-europe-21940895.

[11] The European Society of Human Reproduction and Embryology (2014). ARTs fact sheethttp://www.eshre.eu/Guidelines-and-Legal/ART-fact-sheet.aspx.

[12] Re: X & Y (Foreign Surrogacy), [2008] EWHC (Fam) 3030 (U.K.). http://www.familylawweek.co.uk/site.aspx?i=ed28706.

[13] Deomampo, D. (2014). Defining Parents, Making Citizens: Nationality and Citizenship in Transnational Surrogacy. Review of. Medical anthropology. Published online: 25 September 2014. http://dx.doi.org/10.1080/01459740.2014.890195.

[14] Brandt, R. (2015). Medical intervention should not define legal parenthood. Bionews. http://www.bionews.org.uk/page.asp?obj_id=523229&PPID=523190&sid=282.

[15] Smajdor, A. (2015). Can I be my grandchild's mother? BioNews. http://www.bionews.org.uk/page_504476.asp.

[16] Shanks, P. (2015). Babies from Two Bio-Dads. Biopolitical Times, Center for Genetics and Society. http://www.biopoliticaltimes.org/article.php?id=8418.



Ius filiationis: a defence of citizenship by descent

By Rainer Bauböck (EUDO CITIZENSHIP co-director, EUI)

Aristoteles famously defined a citizen as someone “giving judgment and holding office” in the polity.[1] Yet, this does not settle the issue since we first need to know who qualifies for holding office. And so he continues: “For practical purposes a citizen is defined as one of citizen birth on both his father’s and his mother’s side”.[2] Times have changed. From the French Revolution, which revived ius sanguinis, until the second half of the 20th century, citizenship was mostly transmitted only from the father to the child. Today, largely as a result of international conventions against the discrimination of women, all democratic states define a citizen as one of citizen birth on either the father’s or the mother’s side. Yet ius sanguinis remains the dominant rule for acquisition of citizenship worldwide. True, in the Americas the stronger principle is ius soli, the acquisition of citizenship through birth in the territory. But even there those born abroad to citizen parents who were themselves born in the country are recognized as nationals by birth.

Given this overwhelming presence of ius sanguinis in nationality law, Costica Dumbrava’s call for abandoning it is bold. Some might even say, it is quixotic, but I disagree. It is indeed time to reflect on the future of ius sanguinis and to abandon it as a doctrine linking citizenship to biological descent. Yet there are good practical and normative reasons why the principle of citizenship transmission from parents to children will remain alive and ought to be retained.

Dumbrava runs three main attacks against ius sanguinis: It is tainted by its association with ethno-nationalism; it is inadequate because, in an age of artificial reproduction technologies, same sex marriage and patchwork families, biological descent no longer traces social parenthood; and it is unnecessary since its protective effects can be achieved by other means. I will accept the first and second argument with some modifications but reject the third.

Not the only one tainted

As Dumbrava points out, modern ius sanguinis was seen as a democratic and revolutionary principle in contrast with ius soli that had its origins in the feudal idea that any person (or animal) born on the territory was subject to the ruler of the land. Deriving citizenship from citizen descent rather than territorial birth made it possible to imagine a self-governing people reproducing itself. Dumbrava is of course right that seeing the nation as a community of shared descent across generations made it also easier to justify the exclusion of foreigners as well as the inclusion of co-nationals across the border. Yet this is not a sufficient reason for abandoning ius sanguinis.

First, an ethnonationalist disposition can be overcome while maintaining ius sanguinis if this principle is supplemented with ius soli and residence-based naturalisation. The latter has created an ethnically highly diverse citizenry in continental European immigration countries even in the absence of the additional inclusionary effects of ius soli. The reason for this ethnically inclusive effect of ius sanguinis is simple: If first generation immigrants have access to citizenship and take it up, then ius soli and ius sanguinis does not make much difference: the children of immigrants will be citizens under either rule.

Secondly, a pure ius soli regime is also tainted and not only because of the feudal origins of the principle. Territorial nationalism can be just as nasty as ethnonationalism and may be fanned by thinking of ius soli as the right of the “sons of the soil.” Even the case of Romania that Dumbrava lists among the problematic ones is ambiguous in this regard. If Romania awards citizenship to the descendants of those born in its lost territories, is this an instance of ius soli or ius sanguinis and an illustration of ethnic or of territorial nationalism? The answer is probably: both. Ius soli and ius sanguinis are therefore not alternatives, but can be combined in benign ways that neutralise the potentially illiberal effects of either principle, as well as malign ways that enhance their nationalist potential for ethnic exclusion and territorial expansion.

Thirdly, pure ius soli also has vicious exclusionary effects for migrants. In most American states, the immigrant generation 1.5 – those who have entered the country as minor children – cannot acquire citizenship before the age of majority. President Obama’s Dream Act is an attempt to mitigate some of the worst consequences for the children of irregular immigrants. Even more problematic is the common distinction between nationals and citizens in many Latin American states (see the new American country profiles on EUDO CITIZENSHIP). Only those born in the territory are considered nationals (they are sometimes also called “naturals”). They turn into citizens with full voting rights at the age of majority. Immigrants who naturalise become citizens, but not nationals. They remain excluded from many public offices (also the US president still has to be a “natural born citizens”) and they can be deprived of their citizenship status, whereas nationality can often never be lost. In Uruguay even the concept of “naturalisation” does not exist because those who are not born in the territory can never become “naturals”.  Similar exclusionary effects of ius soli traditions apply to those born abroad to citizen parents. They often do not acquire citizenship unless they are registered in time by their parents and they may lose it unless they “return” before the age of majority.

If both ius sanguinis and ius soli are tainted in these ways, should we consider an even more radical alternative of abandoning citizenship by birth altogether? Why not replace it with ius domicilii so that citizenship is acquired automatically with taking up residence and lost with outmigration? Or should we maybe replace it with ius pecuniae,[3] i.e. a global market for citizenships in which individuals can bid for membership status anywhere and states can set the admission price?  Neither of these alternatives is morally attractive and something important is lost when we give up birthright citizenship.

Why not ius filiationis?

Dumbrava’s second argument is that developments in reproduction technologies and in the social and legal recognition of new family patterns make ius sanguinis increasingly unworkable and obsolete.

This problem is not entirely new and a solution is already available. International law has long abandoned the idea that children should acquire only one citizenship at birth. Since they can inherit two different citizenships from the mother’s and the father’s side (maybe in addition to a third one acquired iure soli), why should they not receive the citizenship of both an intended and a surrogate mother or an intended father and a sperm donor? Asking the question makes it already clear that the problem is not the multiplicity of citizenships per se, but the mismatch between biologically determined citizenship and parental care arrangements that would also open the door to abusive claims. The traditional solution that is already available in most nationality laws for cases where the biological parent is not the social parent is transmission of citizenship through adoption.[4] Why should it not be possible to generalise this model from the marginal case of adoption so that a modified ius sanguinis refers to social rather than biological parenthood (as it already does in several jurisdictions)?

The main issue with such a new ius filiationis might be that determination of citizenship is less automatic than it used to be for children born in wedlock to their biological mother and father. Yet states that are committed to the welfare of children have to figure out anyhow how to determine legal parenthood in the more complex family arrangements of contemporary societies. In order to avoid statelessness it is important that every child obtains at least one citizenship immediately at birth. And in order to make sure that children are not caught between conflicting legal norms and can develop stable relations to their countries of citizenship, it is important that their citizenship status does not change automatically when they become part of a new family. If these concerns are taken into account through a combination of ius soli with legally determined initial parenthood, what objections can be raised against recognizing primary caregivers as well as persons with additional custodial rights as legal parents who can transmit their citizenship to the child?

Don’t abandon the children!

Dumbrava’s third argument is that ius sanguinis is not necessary because children’s rights can be protected through other means. He claims that ius sanguinis renders children vulnerable by making their “access to citizenship … dependent on parents’ legal status, actions or reproductive choices.” This is indeed a reason why the children of immigrants need ius soli as an independent right to citizenship in their country of birth. Unfortunately, in the US, their birthright citizenship does not prevent them from being deported together with their undocumented parents, whereas immigrant minors who are EU citizens have a right to stay that protects also their primary caregivers from deportation.[5]

Yet small children are in any case dependent on their parents’ migration decisions. This is a an equally strong reason why they also have a claim to share their parents’ citizenship, since they risk otherwise to remain stranded in their country of birth or be treated as foreigners in their parents’ country of nationality. Dumbrava suggests preventing this by “conferring full migration rights to children of citizens”. But would migration rights become more secure if they are disconnected from the legal status of citizenship that is the only one obliging states to unconditionally admit them? Alternatively, he suggests to “establish a universal status of (legal) childhood that confers fundamental rights regardless of their or their parents’ citizenship or migration status.” This is what the Children’s Rights Convention, which is one of the mostly widely signed and ratified human rights documents, aims to do. The question is not only whether states are willing to respect these rights, but whether they can be held responsible for protecting them. For this, children need not only human rights, they also need their parents’ citizenship.

Delayed citizenship for all?

Dumbrava has, however, a much more fundamental objection that targets both ius sanguinis and ius soli: Citizenship as membership in a political community should not depend on “contingent facts of birth (descent or place of birth)”. This is a common critique that always leaves me puzzled.[6] My very existence depends on these contingent facts. Humans cannot will themselves into being but are thrown into the world without choosing where to be born and to which parents. What is morally arbitrary is not that states use these fundamental features of personal identity to determine membership in political communities, but that in our world citizenship provides individuals with hugely unequal sets of opportunities. This is not an inherent feature of birthright citizenship but of the global economic and political (dis)order. If we want to overcome it, we have to address the causes of global inequality directly instead of attributing them to those rules that make individuals equal in status and rights as citizens of a particular state.

Dumbrava’s critique focuses, however, on another birthright puzzle that has bothered republican theorists. Shouldn’t membership in a self-governing political community be based on consent? And does it not presuppose certain attitudes and skills that first need to be developed?[7] We may expect that children who are born and grow up in the state territory or who are raised by citizen parents will eventually want to join the political community and will also acquire the skills required for political participation. Yet these are expectations rather than certainties. Dumbrava suggests therefore that “we could delay the attribution of citizenship until such attitudes and skills are confirmed.” However, since children also depend on the state for their health and education, he adds that they could at least be granted provisional citizenship. The Latin American distinction between nationality acquired at birth and citizenship acquired at majority seems to approximate this idea.

One reading of Dumbrava’s proposal is that this is just a terminological distinction harking back to Aristotle’s two definitions of citizenship. If we consider as citizens those who “give judgments and can hold office”, i.e. the members of the demos, then children are indeed only provisional citizens but will automatically become full citizens at the age of majority. The other interpretation draws, however, a line between the two statuses that can only be crossed by demonstrating the right attitude and skills. Instead of naturalising immigrants into a birthright community, this community itself would be denaturalised and reconstituted through a citizenship test imposed on all provisional native citizens. It may seem a form of poetic justice to treat natives like immigrants. Yet there is a big difference between expecting and promoting citizenship attitudes and skills and making them a requirement for access to citizenship rights. The only reason why immigrants can be expected to spend a few years as residents before becoming citizens, which gives them time to develop citizenship skills, and to apply for naturalisation, which demonstrates a civic attitude, is that they are birthright citizens of another state who have grown up there.

Citizenship across generations

Dumbrava concludes by suggesting that the intergenerational continuity of democratic membership should be achieved through consolidating institutions and educating citizens rather than the legal fictions and biological contingencies of birthright citizenship. One might ask why democracies need intergenerational continuity. The answer leads us back to the original justification for ius sanguinis after the French Revolution. It should not be the rulers who determine who the citizens are, nor the citizens themselves through some democratic procedure in which they decide whom to admit or reject, nor the mere fact of subjection to the laws due to temporary presence in the territory. All of these rules lead to too much contingency and discontinuity with regard to the composition of the citizenry. Promoting civic attitudes and skills among those who are citizens is important, but it cannot resolve the puzzle who has a claim to be a citizen in the first place. Automatic acquisition of membership at birth and for life sets this question aside. It makes citizenship a part of citizens’ personal identities that they are like to accept. And it allows democracies to tap into resources of solidarity and to promote a sense of responsibility towards the common good and future generations.

In a nutshell, these are my two arguments why a modified version of ius sanguinis should be accepted as necessary for democratic states:

In a world of territorial states that control immigration, ius sanguinis (or ius filiationis) is as indispensible as ius soli for protecting the children of migrants. It provides them with the right to stay and to be admitted in their country of birth as well as their parents’ country of origin. No other legal status can secure these rights as well as a birthright to dual nationality.

Deriving citizenship from unchosen and permanent features of personal identity – where and to whom one is born – sets aside the politically divisive membership question for the vast majority of citizens, creates a quasi-natural equality of status among them and signals that membership is linked to responsibilities for the common good and for future generations. No citizenship education programme can fully substitute for these signalling effects of birthright citizenship.  

[1] Aristotle. 1962. The Politics. Transl: T.A.Sinclair, revised and commentary: J. Saunders ed. London Penguin, III.i: 169.

[2] ibid, III.ii: 171-2, original emphasis.

[3] Joachim Stern, Ius Pecuniae – Staatsbürgerschaft zwischen ausreichendem Lebensunterhalt, Mindestsicherung und Menschenwürde, in: Dahlvik/Fassmann/Sievers (eds.), Migration und Integration – wissenschaftliche Perspektiven aus Österreich, Jahrbuch 1/2011. See also the EUDO CITIZENSHIP forum „Should citizenship be for sale?“ and Jelena Dzankic (2015) Investment-based citizenship and residence programmes in the EU, EUI Working Papers RSCAS 2015/08, Florence.

[4] See the EUDO CITIZENSHIP modes of acquisition database.

[5] Case C-200/02 Zhu and Chen v Secretary of State for the Home Department, 2004; Case C-34/09 Ruiz Zambrano v Office National de L'emploi, 2011.

[6] For nuanced critiques of birthright citizenship based on this idea see Carens, Joseph H. 2013. The Ethics of Immigration. Oxford: Oxford University Press; Shachar, Ayelet. 2009. The Birthright Lottery. Citizenship and Global Inequality. Cambridge, MA: Harvard University Press.

[7] See Dumbrava, C. (2014). Nationality, Citizenship and Ethno-Cultural Belonging, Preferential Membership Policies in Europe. Houndmills Basingstoke, Palgrave Macmillan, chapter 8, 9.



Tainted law? Why history cannot provide the justification for abandoning ius sanguinis

By Jannis Panagiotidis (IMIS Osnabrück)


In his thought provoking piece, Costica Dumbrava rejects ius sanguinis as 1) historically tainted, 2) increasingly inadequate and 3) normatively unnecessary. In my response, I will mainly focus on the first, historical dimension. Drawing on examples from the case of Germany, often used as the prime example to show what is wrong with ius sanguinis, I will contest the idea that ius sanguinis as such has been discredited by history. 

Regarding the second and third points, I will restrict myself to the following brief observations, which are broadly in line with Rainer Bauböck’s comments: while the issue of ART and citizenship raised by Dumbrava is indeed intriguing, I would go along with his own observation that this is more about the determination of legal parentage than about ius sanguinis, and with Bauböck’s emphasis on social rather than biological parenthood. Discarding the ius sanguinis principle due to certain specific cases it might not adequately cover would mean throwing the baby out with the bathwater. 

I am also simultaneously intrigued and skeptical regarding the suggestion to introduce a sort of “a-national,” universal status for children. Against the backdrop of recent historical research into children as the object of nationalist contestation and agitation during the first half of the twentieth century, a scenario in which “children belonged more rightfully to national communities than to their own parents,” this idea appears intuitively attractive.(1) Having said that, one can turn the argument around and see the suggested disconnection of parent and child citizenship as another attempt to claim children from their parents, this time on behalf on an imaginary inter- or supranational community. Yet in a world still (and for the foreseeable future) structured by nation states, where most so-called human rights are in fact citizens’ rights, one may indeed wonder about the benefits of such a status “above” or perhaps “beyond” the nation if the parents cannot enjoy similar rights. 

Tainted by history?

As to the argument of ius sanguinis being historically tainted, Dumbrava first of all needs to be commended for recognising that “ius sanguinis citizenship is not conceptually ‘ethnic’” (my emphasis, J.P.). Nevertheless, he argues that “there are a number of ways in which the application of the ius sanguinis principle has been used in order to promote ethno-nationalist conceptions of membership.” These include 1) the maintaining of emigrant citizenship beyond the first generation of emigrants; 2) the use of “the principle of descent in order to confirm or restore citizenship to certain categories of people whom [states] consider to be linked with through ethnocultural ties;” and 3) the exclusion of immigrant children from citizenship by an exclusive use of ius sanguinis with no ius soli elements. 

Regarding the third point, I fully agree with Bauböck that it can be remedied quite easily by combining these two principles of citizenship allocation and simultaneously allow for residence-based naturalisation. The first issue is similarly unproblematic: extra-territorial transmission can simply be interrupted at a certain generational stopping point, much like the rule Germany introduced in section 4, paragraph 4 of its reformed 1999 citizenship law regarding the non-acquisition of German citizenship by the offspring of German citizens who themselves were born abroad after 31 December 1999.(2)  There is no rule that says that the transmission of citizenship to descendants has to be possible ad infinitum.

Not all types of “descent” are the same

I would like to discuss the second point in more detail, which touches upon the topics of preferential membership policies and co-ethnic citizenship and migration.(3) Here we are dealing with a terminological confusion quite typical for much of the literature in this field. The “ethnic descent” that Dumbrava mentions as a criterion of admission to citizenship in some cases and the descent implied in the ius sanguinis principle are not the same and should not be conflated. In fact, they are mutually exclusive: “descent” in ius sanguinis is about descent from a citizen, whatever his or her “ethnicity.” The “ethnic descent” used as a criterion in some cases of co-ethnic inclusion is precisely about people who are not citizens. 

The supposed historical taintedness of the ius sanguinis principle results from the conflation of these different types of “descent,” and of the related unhappy connotations of the term “blood,” which invokes associations of “race.” A lot of this confusion was created in the Brubaker-inspired debates of the 1990s about German citizenship. In a telling example, political scientist Patricia Hogwood claimed that “the concept and law of citizenship in Germany were originally formulated in the context of nation-state development based on cultural or ‘völkisch’ nationalism. … The fact that the German legal framework for citizenship and naturalisation remains firmly rooted in the jus sanguinis principle has meant that citizenship policy in Germany is inextricably entangled in concepts of ethnicity and race. … The principle of legal privilege [for ethnic Germans] on the basis of racial origins smacks of the racial policies of the Nazi period …” (my emphasis, J.P.).(4)

Yet ius sanguinis per se has nothing to do with “ethnicity” and “race.” As Dieter Gosewinkel pointed out in his important book on German citizenship, the “blood” here is a “formal and instrumental” notion, not to be confused with “substantial” blood conceptions of racial biology.(5) Those only entered German citizenship law through the Nazi Nuremberg laws. Before, a German Jew, whom the Nazis would later construe to be of a different “race” for having the wrong “blood,” would transmit his German citizenship to his children iure sanguinis, just like other Germans whom the Nazis would construe as “Aryans.” Ius sanguinis is ethnicity-blind. In fact, when young Israelis nowadays claim German citizenship with reference to an ancestor who fled from Germany, they also do so iure sanguinis. I would find it hard to interpret this as an objectionable völkisch practice. This example shows that the problem is not with ius sanguinis itself, but with the respective contexts in which it is embedded. 

Co-Ethnic citizenship is a different story

Nor is ius sanguinis particularly useful (or even necessary) for the conveying of citizenship upon “co-ethnics” in other countries. This is a whole different discussion in my opinion which cannot be used to make a case against the ius sanguinis principle. Taking again the example of Germany, the main European supplier of co-ethnic citizenship in past decades, it needs to be stressed that “ethnic Germans” from Eastern Europe did not receive German citizenship by means of the ius sanguinis of the 1913 citizenship law. This was not possible, as in most cases they had no ancestor with German citizenship to refer to. Their claim to citizenship rested on special provisions in the constitution and expellee law, which equalised the status of German Volkszugehörige with that of German citizens. 

At this point we leave the solid ground of formal citizenship and enter into the murky territory of “ethnicity.” But even here, it is not all about “descent.” While the peculiar notion of Volkszugehörigkeit is often identified with “ethnic descent,” it was much more complex than that: it was actually very much a political-plebiscitary notion predicated on self-avowal (Bekenntnis) as German to be confirmed by an “objective” criterion, which could be language, descent, upbringing, or culture (section 6 of the 1953 Federal Expellee Law).(6)  “Descent” (Abstammung) – notoriously hard to define in administrative practice – was thus neither a necessary nor a sufficient condition for recognition as a German.(7)


In sum, I would argue that the supposed “taintedness” of ius sanguinis has to do with issues not intrinsic to this principle of transmitting citizenship, namely restrictive admission practices and racially based exclusion. The issue of co-ethnic citizenship should be kept apart from this discussion altogether. History cannot provide the justification for abandoning ius sanguinis, as its use in certain problematic ways and contexts in the past does not mean it necessarily has to be used like that in the future. If complemented by other, inclusionary mechanisms of allocating citizenship in conjunction with increased tolerance for multiple citizenship it certainly remains a useful – and necessary – method of transmitting citizenship in the day and age of multiple transnational migrations.




1) The quote is from Tara Zahra, Kidnapped Souls: National Indifference and the Battle for Children in the Bohemian Lands, 1900-1948, Ithaca: Cornell UP, 2008, 3. See also Idem, The Lost Children: Reconstructing Europe's Families after World War II, Cambridge, MA: Harvard UP, 2011, 20.

2) Christian Joppke, Citizenship Between De- and Re-Ethnicization, Russell Sage Foundation, Working Paper No. 204, March 2003, 12-13. The full text of the law can be found under: http://www.gesetze-im-internet.de/rustag/BJNR005830913.html. 

3) Costica Dumbrava, Nationality, Citizenship and Ethno-National Belonging: Preferential Membership Policies in Europe, Basingstoke: Palgrave MacMillan, 2014; Jannis Panagiotidis, Laws of Return? Co-Ethnic Immigration to West Germany and Israel (1948-1992), PhD Diss., European University Institute, 2012.

4)Patricia Hogwood, Citizenship Controversies in Germany: the twin legacy of Völkisch nationalism and the Alleinvertretungsanspruch, in: German Politics 9, 3 (2000), 125-144, here 127, 132-133.

5) Dieter Gosewinkel, Einbürgern und Ausschließen: Die Nationalisierung der Staatsangehörigkeit vom Deutschen Bund bis zur Bundesrepublik Deutschland, Göttingen: Vandenhoeck & Rupprecht, 2001, 327.

6) http://www.bgbl.de/xaver/bgbl/start.xav?startbk=Bundesanzeiger_BGBl&jumpTo=bgbl153022.pdf

7) I elaborate on the plebiscitary, quasi-“Renanian” nature of the German conception of Volkszugehörigkeit in: Jannis Panagiotidis, “The Oberkreisdirektor Decides Who Is a German”: Jewish Immigration, German Bureaucracy, and the Negotiation of National Belonging, 1953–1990, in: Geschichte und Gesellschaft 38 (2012), 503-533, esp. 511.



Family matters: Modernise, don’t abandon, jus sanguinis

By Scott Titshaw (Mercer University)


I appreciate the ideas that Costica Dumbrava and others have introduced into this debate. States’ concerns about the quality and political consequences of their citizenship are important. But citizenship is a two-way street. Our discussion of ius sanguinis laws should extend beyond the concerns of states to also consider the serious practical consequences of citizenship laws on citizens, including the long-term unity and security of their families. Families facing instability or separation because children are denied their parents’ citizenship are unlikely to be satisfied with the explanation that ius sanguinis is inadequate or historically tainted; the resulting individual sense of injustice might even discourage the loyalty and identification states seek in citizens. 

This debate about citizenship transmission is necessary because of two modern changes in the facts of life: (1) increased international mobility based on cheap and easy transportation and communication; and (2) the advent and diffusion of assisted reproductive technology (ART) and new legal family forms (e.g., same-sex marriage and different-sex registered partnership). I will address each in turn. First, I’ll explain why Dumbrava’s proposal to abandon the ius sanguinis principle is an undesirable response to increased international mobility. Second, I’ll build on Dumbrava’s and Bauböck’s recognition of the inadequacy of unlimited and exclusive ius sanguinis rules for today’s families by suggesting that ius sanguinis be modernised rather than abandoned altogether. I’ll also illustrate how citizenship in federal states can add an additional layer of complexity to any universal proposal regarding citizenship. 


Dumbrava’s proposal to eliminate the ius sanguinis principle would increase, rather than decrease, problems based on greater international mobility. It would eliminate one tool parents currently use for transmitting citizenship to children conceived through ART.  While current versions of ius sanguinis are inadequate to deal with other ART issues, that problem can be corrected. And, as Jannis Panagiotidis points out, abandoning ius sanguinis because of this inadequacy would be like “throwing the baby out with the bathwater.” Most children are still conceived through sexual reproduction rather than ART, and many of their families would be worse off without ius sanguinis. 

An example is easy to imagine. Let’s say an Indian couple moves every seven years for employment reasons. They obtain residence permits, but not citizenship, in South Africa, the United Kingdom, and the United States, in turn. They also have a child in each country. Under ius soli regimes with no ius sanguinis rules, the children of these Indian parents would each have different passports (from South Africa, the UK, and the US). This might pose no problem in the short term. But what happens if a parent dies or loses his job? 

Under a ius sanguinis regime, the surviving family members would be able to enter India and remain there together permanently as citizens.(1)

Dumbrava argues that such common citizenship is unnecessary to recognise and cement parent-child relationships if children of citizens have “full migration rights.” But “migration rights” or benefits are substantially less stable than citizenship rights. What if a non-citizen family member becomes deportable because he or she commits a crime?(2) What if both Indian parents die while the children are minors? Without ius sanguinis, the children with their different nationalities might not be allowed to remain together anywhere, let alone in India where their extended family members (grandparents, aunts and uncles) most likely live. 

Dumbrava’s proposal of a universal legal status for all children would ameliorate some of these problems, but only until each child reaches the age of majority.  At that time they might be separated from their parents and siblings. 


I agree with Dumbrava’s and Bauböck’s rejection of exclusive, unconditional ius sanguinis rules as inadequate in dealing with the consequences of ART and modern family law. 

I disagree, however, with the conclusion Dumbrava draws from his argument that “joint citizenship adds little to the legal and normative character of the parent-child relationship.” In fact, the permanence and stability stemming from common citizenship among close family members can have profound consequences for the unity required to develop and maintain family relationships.  

I also disagree with Dumbrava’s argument that “the main problem” is that ius sanguinis “is parasitic on external factors concerning the legal determination of parentage.” In fact, some federal States already delink federal citizenship determination and state or provincial family law,3) creating greater problems than do citizenship laws that reflect legal parentage. In the United States, for example, legal parentage is generally a matter of state law. Yet, the US Constitution defines citizenship as an exclusively federal matter,4) and Congress has established and revised a complex, autonomous algorithm for determining when a citizen parent transmits US citizenship to a child born abroad.(5) The problematic example Dumbrava points out regarding parents’ inability to transmit US citizenship to children conceived through ART was created by a misguided autonomous federal policy, not parentage determinations under family law.(6) It could, and should, be corrected by federal reinterpretation of its rules to rely on family law parentage determinations.(7)

Rather than misplaced reliance on family law, the problems Dumbrava and Bauböck describe regarding the application of ius sanguinis following ART are consequences of outdated family law or of international conflict-of-law issues where relevant jurisdictions define parentage differently. 

To the extent that the problems stem from conflict-of-law issues, it is worth noting that the Hague Conference on Private International Law is currently exploring whether to draft a multilateral instrument on international parentage and surrogacy, which might resolve some issues.(8)

To the extent that the problems stem from outdated family law, the best solution is to fix the family law. Family law generally reflects a more individualized, in-depth understanding of parent-child relationships than do citizenship or migration laws. Based on long experience and empirical data, family law tends to favor the stability of permanent family relationships with commensurate duties and benefits in the best interests of children. By tending to ensure the same citizenship for children and their parent(s), ius sanguinis rules also generally promote stable solutions in the best interests of children in a way that less permanent migration rules do not.   

I agree with Bauböck that multiplicity of citizenships for children is generally not a problem, and I support his call for a more generous understanding of parenthood for purposes of citizenship transmission. But I would not opt for a ius filiationis proposal if it requires an entirely independent determination of social parenthood for citizenship transmission purposes.  Officials dealing with citizenship issues are not as well suited to determine these issues as those administering family law. Also, too much generosity in this area might instigate cross-border mischief in familial disputes by “social parents.” 

Instead, I would suggest replacing all outmoded rules that fail to consider parental intent and the best interests of the child in the context of children conceived through ART, whether these are family laws determining parentage or autonomous federal citizenship laws reading ius sanguinis as a literal “right of blood.” 



(1) India would have automatically recognised these children as Indian citizens through 2004; it still recognises a greatly eased path to apply for citizenship in this context.  http://www.loc.gov/law/help/citizenship-pathways/india.php 

(2) While hardship of citizen relatives is sometimes considered, US immigration law generally requires removal of non-citizens who commit any of a long list of criminal infractions. 8 USC §1227(a)(2). https://www.law.cornell.edu/uscode/text/8/1227 

(3) A Study of Legal Parentage and the Issues Arising from International Surrogacy Arrangements, Prel. Doc. No 3C, Hague Conference on Private International Law [March 2014] 66-68 (listing Australia, Canada and the United States as examples). http://www.hcch.net/upload/wop/gap2015pd03c_en.pdf 

(4) The Fourteenth Amendment guarantees that “[a]ll persons born or naturalized in the United State and subject to the jurisdiction thereof are citizens of the United States and the state in which they reside.” Not only does this Amendment adopt a nearly absolute ius soli rule, but it clarifies that citizenship is a purely federal matter, with no meaningful state role beyond establishment of its own standards for recognising state residence.  

(5) 8 USC §§1401 - 1409. http://www.uscis.gov/sites/default/files/ilink/docView/SLB/HTML/SLB/0-0-0-1/0-0-0-29/0-0-0-9696.html   

(6) Under current US law, a genetic and legal father and/or one or more legal and “biological” mothers (i.e., genetic and gestational mother(s)) transmit birthright citizenship to children conceived through ART, but non-biological parents do not.  Scott Titshaw, A Transatlantic Rainbow Comparison: “Federalism” and Family-Based Immigration for Rainbow Families in the U.S. and the E.U., in Rights on the Move: Rainbow Families in Europe 189, 194-95 (Carlo Casonato & Alexander Schuster, eds. 2014). http://eprints.biblio.unitn.it/4448/

(7) Scott Titshaw, Revisiting the Meaning of Marriage: Immigration for Same-Sex Spouses in a Post-Windsor World, 66 Vand. L. Rev. en banc 167, 174-75 (2013).  http://www.vanderbiltlawreview.org/2013/10/revisiting-the-meaning-of-marriage-immigration-for-same-sex-spouses-in-a-post-windsor-world/ 

(8) http://www.hcch.net/upload/wop/gap2015pd03a_en.pdf 



Abolishing ius sanguinis citizenship: A proposal too restrained and too radical

By Kristin Collins (Boston University School of Law)


Costica Dumbrava maintains that ius sanguinis citizenship is a historically tainted, outmoded, and unnecessary means of designating political membership.  He argues that it is time to abandon it.  His proposal is bold, and it has significant implications for an array of policies and practices. The parent-child relationship not only serves as a basis for citizenship transmission; it also entitles individuals to immigration preferences, and – in some countries – it facilitates automatic or “derivative” naturalisation of the children of naturalised parents. In many countries that recognize ius soli citizenship, the parent-child relationship serves as an added requirement: one must be born in the sovereign territory and be the child of a citizen or a long-term legal resident. Dumbrava limits his challenge to ius sanguinis citizenship per se, and even suggests that family-based migration rights could be used to minimise the disruptive effect of abolishing citizenship-by-descent.  But his core complaints about ius sanguinis citizenship – the mismatch of biological parentage and political affinity, the difficulties of determining legal parentage – can be, and have been, levied against these various family-based preferences and statuses, which are likely found in every nation’s nationality laws. It is therefore important to consider his proposal in light of the role that the parent-child relationship plays in the regulation of migration, naturalisation, and citizenship more generally. With this broader context in mind, I concur with Rainer Bauböck and Jannis Panagiotidis that Dumbrava’s proposal rests on an under-informed assessment of the historical record. I also argue that that, as a remedy for the problems that he has identified, Dumbrava’s proposal is at once too restrained and too radical. 

The Complex History of Ius Sanguinis Citizenship

Dumbrava first argues that ius sanguinis citizenship should be abolished because, historically, it has been associated with ethno-nationalist conceptions of citizenship. I appreciate Panagiotidis’ insistence that “the problem is not with ius sanguinis itself, but with the respective contexts in which it is embedded” (1). Panagiotidis also reminds us that ius sanguinis citizenship has sometimes functioned to create political communities that draw from different ethnic and religious groups, as in the case of German Jews whose membership in the German polity was secured by the country’s ius sanguinis laws prior to the Nazi era.  I want to elaborate and underscore the importance of this point with an additional example from United States history: During seventy years of Chinese exclusionary laws, ius sanguinis citizenship provided one of the very few routes to entry, and to American citizenship, for ethnic Chinese individuals born outside the U.S.  For precisely that reason, exclusionists sought to limit or repeal the ius sanguinis statute, which recognised the foreign-born children of American fathers as citizens (2). If one expands the historical frame to include parent-child immigration preferences and derivative naturalisation, the story becomes even more complex. By 1965, the race-based exclusions and national-origins quotas had been abolished, and previously excluded Asian families began immigrating to the U.S. in unprecedented numbers (3). They were able to do so by relying on the generous family-based preferences in American nationality law, which facilitated entry, settlement, and – especially significant to this discussion – derivative naturalisation for children (4).

Even a cursory review of the historical record thus counsels a cautionary assessment of the contention that ius sanguinis citizenship’s tainted past justifies its abolition. First, calls to end ius sanguinis citizenship have their own ugly history. Second, although one cannot gainsay that, in certain circumstances, ius sanguinis citizenship has been used to maintain ethnic homogeneity, the notion that parents and children do and should share the same political affiliation has also facilitated racial, ethnic, and religious diversification of some political communities. Rather than abolish ius sanguinis citizenship wholesale, we should be alert to the ways that it can operate as a tool of ethnic exclusion and degradation in particular socio-legal contexts, and work to minimise those effects (5).

A Proposal Too Restrained and Too Radical

To be fair, Dumbrava does not extend his proposal to migration and naturalisation policies that enlist the parent-child relationship; indeed, he would preserve such migration policies. He speaks only of traditional ius sanguinis citizenship, and argues that it often fails to map on to the reality of modern family formation, making it inadequate to “deal with contemporary issues such as advances in assisted reproduction technologies” (ART), same-sex coupling and marriage, and the steady rise of nonmarital procreation. The problems Dumbrava identifies in this regard are important and difficult. But as a remedy for these problems, abolishing parent-child citizenship transmission is simultaneously too restrained and too radical. It is too restrained because, after abandoning ius sanguinis citizenship we would still be confronted with the difficulty of determining which parent-child relationships should count for purposes of regulating migration, derivative naturalisation, and (in many ius soli countries) birthright citizenship. Moreover, in all of these contexts, the “fundamental normative questions about who should be a citizen in a political community” – and about the role that the parent-child relationship should play in that determination – would persist.

At the same time, Dumbrava’s proposal is too radical. He argues that ius sanguinis citizenship is not necessary to protect children from statelessness and “adds little to the legal and normative character of the parent-child relationship.” On this point I agree entirely with Bauböck and Scott Titshaw that Dumbrava underestimates the disruptive potential of his proposal. If all countries recognised unrestricted ius soli citizenship, Dumbrava’s assertion that ius sanguinis citizenship is unnecessary to prevent statelessness would be basically correct. But, in fact, very few ius soli countries go that far. Instead, as noted, they use ius sanguinis concepts to restrict the operation of ius soli birthright citizenship, thus leaving some children at a risk of statelessness if traditional ius sanguinis citizenship were abolished. And it is not just formal statelessness that would increase in a world without ius sanguinis citizenship. Children whose citizenship does not align with that of their parents can find themselves divided by nationality from the individuals who are charged, ethically and legally, with their care. As Bauböck and Titshaw observe, in an era of voluntary and compelled migration, ius sanguinis is the most effective method of protecting against such destabilising and precarious circumstances. 

How to Modernise?

I agree with Titshaw and Bauböck that the modernisation of ius sanguinis citizenship, rather than its complete repudiation, offers a better way to address the problems Dumbrava identifies. The difficult question is how? I am hesitant to embrace Titshaw’s proposed method of modernisation, and I offer a friendly but important amendment to Bauböck’s proposal.  

Titshaw argues that the officials who administer citizenship law should adhere to the parentage determinations made by officials who generally administer family law. In the U.S., these are state-level family law judges applying state law. But domestic family law, in the U.S. and elsewhere, does not necessarily generate ideal or even tolerable outcomes on questions of citizenship. Titshaw holds up a particularly poorly drawn U.S. federal policy that regulates ius sanguinis citizenship as it applies to foreign-born children conceived using ART, but there are many examples of how the use of state family law to regulate citizenship transmission has generated equally objectionable outcomes (6).

Alternatively, Bauböck would have us adopt a “ius filiationis” standard that recognises the “social parent” or the “primary caregiver” as the parent for purposes of ius sanguinis citizenship.  He urges that this would help remedy the “mismatch between biologically determined citizenship and parental care arrangements that would also open the door to abusive claims” (7). He is correct. My concern, however, is that his emphasis on “social parenting” and “primary caregiving” is insufficient and has its own perils. First, it could increase the likelihood of abusive denials of citizenship by officials who, at least in the U.S., are often all too eager to find reasons to reject claims to citizenship (8). In the case of nonmarital children – who make up a far greater portion of the global population than children conceived through ART  – the restriction of parent-child citizenship transmission to “primary caregivers” could lead to circumspect treatment, or outright rejection, of the father-child relationship as a basis for citizenship transmission. Indeed, the primary caregiver standard could stymie the caregiving efforts of unwed fathers who are divided by nationality from their children, and hence may never be able to establish themselves as the “primary caregiver.” The emphasis on caregiving as a prerequisite could also aid unwed fathers who prefer to avoid parental responsibility by distancing themselves geographically from their children. The result: a ius sanguinis citizenship regime that would buttress gender inequality by undermining men’s parental rights and helping them to avoid their parental responsibilities (9). Moreover, and regardless of one’s view of the equities as between parents, it is ultimately the nonmarital child’s citizenship and migration rights that could be destabilised, depending on how officials understood the concept of “social parent.” Dumbrava recognises the inequities associated with “the differential treatment of children born within and out of wedlock with respect to access to citizenship,” but his solution – to abolish parent-child citizenship transmission altogether – would give cold comfort to nonmarital children and marital children alike. 

This is not an endorsement for a purely genetic model of citizenship transmission. Despite the references to “blood,” ius sanguinis citizenship has never rested on purely biological conceptions of citizenship. Traditionally, marriage was fundamental to the ability of fathers to secure citizenship for their children, and – at least in the development of U.S. law – the presumption that the mother is the sole caregiver of the nonmarital child led to the recognition of the mother-child relationship as a source of citizenship for foreign-born nonmarital children (10). Rather, I suggest that – unless and until we move beyond citizenship as the enforcement mechanism for basic human rights, and beyond the family as a foundational source of material and psychological support for children, we cannot overstate the importance of the generous recognition of the parent-child relationship for citizenship transmission. The modernisation of ius sanguinis citizenship should thus include the recognition of “social parents” and parents with “custodial rights”– as Bauböck rightly asserts – and also recognition of all who can be held legally responsible for a child’s care or support. Dumbrava may be unhappy that the whims of parents, people’s reproductive choices, and factors beyond the control of the individual would continue to determine membership in a political community. But it is precisely because citizenship designations rest on factors such as these that I wholly agree with his admonition that we channel our efforts “towards consolidating democratic institution and promoting citizenship attitudes and skills among all those who find themselves, by whatever ways and for whatever reason, in our political community.”



1) See Jannis Panagiotidis, ‘Tainted law? Why History Cannot Provide the Justification for Abandoning Ius Sanguinis’, in EUDO Citizenship Forum.  

2) For a discussion of these laws and efforts to restrict the recognition of ethnic Chinese individuals under the ius sanguinis citizenship statute, see Kristin A Collins, ‘Illegitimate Borders: Jus Sanguinis Citizenship and the Legal Construction of Family, Race, and Nation’ (2014) 123 Yale Law Journal 2134, 2170-2182.  Starting in 1934, the ius sanguinis statute also allowed American mothers to transmit citizenship to their foreign-born children.  See id. at 2157. 

3) See David Reimers, ‘An Unintended Reform: The 1965 Immigration Act and Third World Immigration to the United States’ (1983) 3 Journal of American Ethnic History 9, 23-24; Bill Ong Hing, Making and Remaking Asian America Through Immigration Policy, 1850-1900 (Stanford UP,1999), 81-120.  

4) See, for example, Immigration and Nationality Act of 1952, 66 Stat. 163, 245, § 323. 

5) A particularly notable example of how ius sanguinis principles can operate as tools of ethno-racial exclusion is the 2013 ruling of the Constitutional Tribunal of the Dominican Republic, TC/0168/13, which effectively expatriated ethnic-Haitian individuals born and residing in the D.R., leaving hundreds of thousands of people stateless.  

6) For example, in 1940 the federal ius sanguinis citizenship statute was amended to include the nonmarital children of U.S. citizen fathers under certain circumstances, such as when the father had “legitimated” the child. Federal officials turned to the law of the father’s domiciliary state to determine whether legitimation had, in fact, occurred.  In the 1940s and 50s, marriage to the child’s mother was a very common mode of legitimation, but federal officials making citizenship determinations would not recognise an interracial marriage as the basis of a child’s citizenship claim if the father’s home state banned such marriages – and many did. See Collins, ‘Illegitimate Borders’, above n 2, at 2210. 

7) Rainer Bauböck, ‘Ius Filiationis: A Defence of Citizenship by Descent’, EUDO Citizenship Forum.  

8) See, for example, Saldana Iracheta v. Holder, 730 F.3d 419 (5th Cir. 2013). 

9) I develop this argument in ‘When Fathers’ Rights Were Mothers’ Duties: The Failure of Equal Protection in Miller v. Albright’ (2000) 109 Yale Law Journal 1669, 1699-1705, and in ‘Illegitimate Borders’, above n 2, at 2230-34. 

10) See Collins, ‘Illegitimate Borders’, above n 2, at 2199-2205. 



Citizenship without magic

By Lois HarderUniversity of Alberta


I share Costica Dumbrava’s critique of ius sanguinis citizenship, and ultimately what is, I think, his rejection of birth as the basis for political membership generally. Of course, there are issues of practicality - of the world as we find it - that might limit whether and how one would advance the abolishment of birthright citizenship in light of specific political dynamics. But it is precisely those practicalities, and the near unthinkability of alternatives to birth-based citizenship that demand our interrogation of birthright in the first instance. As Joseph Carens has argued with respect to his advocacy of open borders, “even if we must take deeply rooted social arrangements as givens for purposes of immediate action in a particular context, we should never forget about our assessment of their fundamental character. Otherwise we wind up legitimating what should only be endured.”(1)

In his contribution to this forum, Rainer Bauböck defends birthright citizenship and argues that in both of its iterations (ius sanguinis and ius soli) it avoids political division and “creates a quasi-natural equality of status” among citizens who are entitled to claim it. But what about the inequality that divides the entitled from the unentitled? Political communities may be unavoidably bounded, but if a normative commitment to human rights is our guiding frame, it seems incumbent upon us to advance methods or prospects for membership that reduce the barriers to belonging as much as possible. Moreover, as Jacqueline Stevens trenchantly observes, in defining the bounds of equal citizenship, borders also form the boundaries of our non-emergency expressions of compassion.(2) To the extent that birthright entitlement advances a seemingly unassailable claim to exclusionary membership, its advocacy runs counter to a broader commitment to humanitarianism. 

Bauböck’s description of birthright citizenship evades the fact that establishing citizenship through birth, as with any other basis for membership, is an inherently political decision. One of the central appeals of birthright is that it involves innocent, vulnerable babies, infants who are not (yet) marked by misdeeds, criminality, inadequate knowledge or commitment, or the wrong ideological proclivities. It is this innocence that helps to obscure the profoundly political basis of birthright; that makes it possible to describe birthright citizenship as avoiding political division and establishing a quasi-natural equality. However, the use of criteria of birth to determine political membership – whether it is birth to a citizen parent (variously defined) or birth in the territory (variously defined) – is not innocent. Prevailing views about

•   wedlock and patriarchal forms of social organisation (e.g. unwed mothers having responsibility for their children and conferring citizenship, but unwed fathers having no such responsibility or capacity);

•   the relative significance of biological and social parenting as well as gender and sexuality (can a lesbian co-mother confer citizenship on her genetic progeny to whom she did not give birth – just as fathers do?);

•   national attachment (is this child born abroad as second or subsequent generation?); and 

•   how generous territorial definitions should be (is a child born to a Ugandan mother on an American airline flying in Canadian airspace from Amsterdam to Boston a Canadian? Answer = yes)(3)

all play out in the rules that determine birthright entitlement. The magical power of birthright citizenship is that it makes it possible for us to know and rehearse these rules while simultaneously making birthright seem straightforward, static and apolitical.  In contrast to citizenship debates that engage migration, legal and illegal status and naturalisation, birthright citizenship makes these political choices disappear with a wave of a wand. 

I am currently researching a book on the lost Canadians. These are people who thought they had a birthright claim to Canadian citizenship, but subsequently learned that they were mistaken. Their difficulties arose for various reasons, and have now largely been resolved through statutory amendment (a rule change). In making their case to Parliament, the courts and the media, their primary, and highly successful, strategy, was to denigrate the rule-boundedness of “mean-minded bureaucrats” and advance the merits of their claims through appeals to lineage and blood-based belonging.(4) Despite being born in the UK, residing in Canada for five years as a small child, and having subsequently lived in the UK for six decades, one such lost Canadian insisted, “I, sir, am a Canadian. To the roots of me, to the spirit of me, to the soul of me, I'm Canadian.”(5) This impassioned claim to Canada – not exactly your ‘go to’ example of an ethnic nation – nonetheless succeeds as a rhetorical strategy because it re-enchants the nation,(6) underscoring the country’s desirability to the Canadian public, and insisting that this connection is an essential feature of her identity. This is a logic that only works in a world of birthright citizenship. And it is a strategy that eventually succeeded in securing legislative amendments, because the birth-based claims of the lost Canadians (and not necessarily residency or connection) carried an overwhelming political potency. 

To the extent that birthright citizenship enables progressive people to cordon off a substantial portion of membership determination from a potentially nasty political debate, one can certainly understand its attractions. But the occlusion of politics with an unsupportable appeal to nature is ethically dubious. If we are committed to democratic equality, we need principles to manage how we live together that refuse the privilege of birth over naturalisation, and that require us to come to terms with our mortality.(7) Political membership should be a lively, on-going process of negotiation in which everyone has a stake. Some critics might argue that abandoning birthright citizenship and its intergenerational character will create the conditions for decision making in which we are no longer future-oriented, or indeed, that we will neglect the lessons and obligations of our past. If our children do not have a stake in the polity to come, why should we commit ourselves to making it better? This kind of argument is morally bereft. We can continue to care about the future and attend to the damages we, and our ancestors, have wrought, even if, or precisely because, our political membership is limited by our mortality. It was, of course, ever thus.



1. Joseph Carens, The Ethics of Immigration (New York: Oxford University Press, 2013), 229.

2. Jacqueline Stevens, States without Nations: Citizenship for Mortals (New York: Columbia University Press, 2010), 9.

3. Michael Levenson, “Birth and joy midflight” Boston Globe 1 January 2009. Online: http://www.boston.com/news/local/massachusetts/articles/2009/01/01/birth_and_joy_midflight/?page=full

4. Canada, 26 February 2007, House of Commons Standing Committee on Citizenship and Immigration 39th Parliament 1st Session no. 38. (at 11:50)

5. Canada, 26 February 2007, House of Commons Standing Committee on Citizenship and Immigration 39th Parliament 1st Session no. 38. (at 11:45)

6. Bonnie Honig, Democracy and the Foreigner. Princeton: Princeton University Press, 2001, 74.

7. For a full elaboration on the dangers of intergenerational citizenship, see Stevens, 2010.




The Janus-face of ius sanguinis: protecting migrant children and expanding ethnic nations

By Francesca Decimo (University of Trento)


 Costica Dumbrava’s proposal for abandoning ius sanguinis is timely and bold. My intuition is to reject his suggestion that children’s citizenship might be disconnected from that of their parents, but to join his advocacy for a radical rethinking of the ius sanguinis principle with a view towards eliminating it once and for all. These are rather contrasting stances in relation to the same principle. Let us see if the apparent contradiction can be resolved.

To begin, let us consider the element of Costica Dumbrava’s proposal that has elicited most attention and controversy among the respondents, but was picked up and expanded by Lois Harder, namely the assertion that granting citizenship at birth is unnecessary and, above all, that making children dependent on the legal status of their parents exposes them to a form of vulnerability. The idea of postponing the acquisition of citizenship until adulthood, taking into account birthplace and residence or possession of the appropriate attitudes and skills, derives from the classic opposition between ius sanguinis and ius soli according to which the former is considered ethnic and exclusive while the latter is considered civic and inclusive. Yet Rainer Bauböck’s comments on this point explain how, in the absence of parental transmission of citizenship to children, ius soli and ius domicilii can generate individual and familial conditions that are both legally paradoxical and morally unfair.

I share the doubts and critiques raised by Rainer Bauböck, Scott Titshaw and Kristin Collins regarding the alleged emancipatory value of a citizenship system that disconnects children from their parents. Particularly, I consider any legal system that fails to specifically protect the relationship between parents and children to be highly risky. Indeed, who should children depend on if not their parents? Dumbrava’s proposal that children might instead be subject to, and protected by, a kind of international law faces the problem of subordinating the individual and familial reproductive spheres to institutional logics. 

As Luc Boltanski (1) has noted, the event of birth is inextricably linked to the definition of belonging and social descent – and therefore legal, political, cultural, national, etc. descent as well. Historically, devices for legitimating the procreative event were provided by religion, ancestry, the nation-state and, in more recent times, a long-term relationship among a couple. In a scenario in which parentage and citizenship are not tightly connected from the beginning, the risk is not only that of generating stateless children but also an excess of state power. Even after World War Two, the Catholic Church in Ireland took children considered illegitimate away from their unmarried mothers. It was nationalist demographic policies, both in Europe and overseas, that shaped the reproductive choices of individuals and families during the 20th century with a view to producing children for the fatherland. We might recall these policies when interpreting some recent nationally-oriented arguments encouraging the children of immigrants to rid themselves of the burden of their cultures of origin in which their inadequately assimilated mothers and fathers remain stuck(2). With this in mind, do we really want to define children’s citizenship irrespective of their parents’? Do we really want to shift the task of determining the legitimate membership of our offspring from relationships to institutions?

The considerations made thus far therefore lead me to agree with those who have argued that, as long as the system of nation-states regulates our rule of law, children’s citizenship must be linked from birth to that of their parents.

At the same time, it seems to me that ius sanguinis is a legal instrument which, especially in a global context of increased geographical mobility, opens the way to policies of attributing nationality that go far beyond protecting the parent-child relationship. This point relates to Dumbrava’s observation that ius sanguinis is historically tainted that was critically addressed by Jannis Panagiotidis but has not yet been decisively refuted. 

As scholars have noted, ius sanguinis makes it possible to recognise a community of descendants as legitimate members of the nation regardless of its territorial limits, but that is not all. This principle has been used to grant the status of co-national to individuals dispersed not only across space but also across time, leading to the construction of virtually inexhaustible intergenerational chains(3). This principle is based on blood, identified as the essential and primordial element of descent, belonging and identification. It is true that this potential for unlimited intergenerational transmissibility is effectively defused by the fact that many countries interpret ius sanguinis narrowly, applying it generally only up to the second generation born abroad. And yet, is this limit enough to bind and delimit the potential of ius sanguinis? In national rhetoric the image of a community of descendants continues to exert a powerful appeal that goes beyond the attribution of birthright citizenship. In historical emigration countries - but also others(4) -, ius sanguinis as a legal practice is used to grant preferential conditions and benefits to descendants as part of the direct transmission or ‘recovery’ of ancestral citizenship well beyond the second generation(5). Generational limits in the granting of citizenship to descendants can thus be bypassed because, in principle, ius sanguinis itself poses no particular restrictions in this regard.

The most controversial aspects of ius sanguinis emerge when this principle ends up competing with ius soli or ius domicilii, that is, when individuals born and raised elsewhere enjoy a right to citizenship in the name of lineage and an assertion of national affiliation while immigrants who participate fully in the economic, social and cultural development of the country are denied this same right or face serious obstacles in accessing it. In such context  — Germany in the past and Italy today – the right to citizenship effectively becomes a resource which, like economic, human and social capital, is distributed in a highly unequal way, benefitting certain categories of people – ‘descendants’ – at the expense of others – ‘foreigners’.

In view of its unlimited intergenerational potential, I conclude that, if its purpose is merely to bind children’s citizenship to that of their parents, ius sanguinis as a legal instrument suffers from ambiguity and disproportionality. All of these critical points seem to be implicitly overcome in Bauböck’s proposal of a ius filiationis principle, which would focus entirely on linking children’s citizenship to that of their parents, especially for migrants and non-biological offspring. Under a different name and with distinct content, does this move not suggest that, rather than modifying or modernising ius sanguinis as advocated by Rainer Bauböck and Scott Titshaw, it is time to abandon it once and for all, adopting in its place a principle that explicitly protects parentage and citizenship in contexts of geographical mobility instead of linking it to genealogical lineage and nationhood?



(1) See Boltanski L. 2004 La condition foetale. Paris: Gallimard.

(2) See Hungtinton S. 2004 Who are we? New York: Simon and Schuster

(3) See Brubaker R. 1992 Citizenship and Nationhood in France and Germany, Cambridge, Harvard University Press

(4) Joppke’s comparison of three highly divergent countries, France, Italy and Hungary, is quite effective in shedding light on this issue in Joppke C. 2005 Selecting by Origin, Cambridge, Harvard University Press, pp. 240-250

(5) For an in-depth analysis of the Italian case, see Tintori G. 2013 Naturalisation Procedures for Immigrants – Italy, Fabruary, EUI, EUDO Citizenship Observatory



The prior question: What do we need state citizenship for?

By David Owen (University of Southampton)

In his kick-off contribution, Costica Dumbrava offers a threefold critique of ius sanguinis as a norm of citizenship acquisition. In reflecting on this critique, I share the scepticism expressed by Rainer BauböckJannis PanagioditisScott Titshaw and Kristin Collins. In particular I would note, along the lines of Titshaw’s Indian family example, that the abolition of ius sanguinis would have led in my own family context to four siblings, of whom I am one, being split among three different nationalities: Nigerian, British and Malaysian). However rather than address Dumbrava’s critique head on, I want to suggest that the kind of critique of ius sanguinis that he offers – and the same point would apply to the critique or defense of any of the classic membership rules taken singly as free-standing norms – gets things moving askew from the start. To see this, one needs to take a step back and situate this debate within a slightly different context. When asking what citizenship rules we ought to endorse or reject, we ought to begin with a prior question: ‘what do we need state citizenship rules for?’ 

In a world of plural autonomous states, there are two basic functions that such rules are to play: 

(1) to ensure that each and every human being is a citizen of a state and hence that everyone has, at least formally, equal standing in a global society organised as a system of states;

(2) to allocate persons to states in ways that best serve the common interest, that is, where this allocation supports protection of the fundamental interests of individuals, the realization of the common good within states and the conditions of cooperation between states.

A plausible response to these requirements is a general principle that Ayelet Shachar (2009) calls ‘ius nexi’ which highlights the importance of a genuine connection between persons and the state of which they are citizens. The notion of ‘genuine connection’ can be glossed in terms of Bauböck’s “stakeholder” view which proposes that those and only those individuals have a claim to membership of a polity whose individual autonomy and wellbeing is linked to the collective self-government and flourishing of that polity (Baubock 2015). It seems to me that we should see ius soli, ius sanguinis and ius domicilii under the general principle of ius nexi as denoting different routes through which a genuine connection is presumptively established: through parental citizenship, through place of birth and through residence. 

Seeing each of these rules under this more general principle, rather than seeing each as a single free-standing norm, makes clear two points that are salient to this discussion. First, that in adopting any of these rules we are not reifying ‘blood’ or ‘territory’ or ‘residence’. We regard them instead simply as acknowledgments of the diverse ways in which ius nexi may be given expression – and we need each of them if we are to do justice to the relations of persons to states. Second, that each of the ius soli, ius sanguinis and ius domicilii rules should be qualified by the general principle of ius nexi that they serve. So, for example, an unlimited ius sanguinis rule or a ius soli rule that included a child born to visiting tourists or a ius domicilii rule that granted citizenship after three months residence would be incompatible with the overarching ius nexi principle.

Still it would be in line with Dumbrava’s argument for him to object that the ‘birthright’ rules of ius soli and ius sanguinis can only operate on the basis of the general presumption that parental citizenship and place of birth establish a genuine connection, so why not wait until the children reach their majority? Here I concur with the view advanced by Bauböck that the adequate protection of children’s rights implies that ‘children need not only human rights, they also need their parent’s citizenship’ (p.9) Titshaw’s example of the serially mobile Indian family and my own peripatetic family history suffice to make this point. Contra Harder, I don’t think that ‘birthright’ rules disguise the political character of membership norms, rather they acknowledge important interests of children, parents and states. Harder’s stress on the relationship of those entitled to citizenship of a given state and those not so entitled doesn’t provide reasons to drop either ius soli or ius sanguinis, what it does is provide reasons for relatively generous ius domicilii rules, of rights to dual/plural nationality and of a more equitable distribution of transnational mobility rights. 

And perhaps there may be a clue here to an unstated background commitment of Dumbrava’s critique. If we ask under what, if any, circumstances in a world of plural states, it could make sense to abolish ius sanguinis rules, then I think that the only answer that has any plausibility is a world of open borders characterised by rapid access to citizenship through ius domicilii rules.  It may even be plausible that the abolition of ius sanguinis rules would generate political support for more open borders given the problems liable to be posed for sustaining the human right to a family life after the removal of such rules. Whether this is a prudent way of seeking to realise such a world and whether such a world is desirable are, of course, further questions. 



Bauböck, R. (2015) ‘Morphing the Demos into the Right Shape. Normative Principles for Enfranchising Resident Aliens and Expatriate Citizens’ Democratization, Volume 22, Issue 5, 2015: 820-839, DOI:http://dx.doi.org/10.1080/13510347.2014.988146.

Shachar, A. (2009) The Birthright Lottery, Cambridge, Mass., Harvard University Press.


No more blood

By Kerry Abrams (University of Virginia School of Law)


Problems have plagued the ius sanguinis principle—the transmission of citizenship from parent to child— for as long as it has existed. Costica Dumbrava is surely correct that the time has come to ask whether ius sanguinis is still necessary. But the core problem with ius sanguinis, I would argue, is not that it uses the parent-child relationship to determine membership but that it overemphasizes the importance of the genetic tie to this relationship. 

The very term ius sanguinis —“right of blood”—makes the genetic tie the sine qua non of belonging. It is this obsession with genetic purity that has linked ius sanguinis to tribalism, xenophobia, and even genocide. This problem, I believe, is distinct from the very real need to ensure children’s access to the same geographic territory and legal system as that of their parents. Rainer Bauböck’s proposal for a “ius filiationis” based on family association rather than genetic ties would excise many of the problems caused by a focus on blood while protecting the parent-child relationship and the stability for children that flows from it.

Let me explain in more detail why I think that retaining recognition of parent-child relationships while abandoning the other features of ius sanguinis is sensible. At first glance, protecting the tying of children’s citizenship to that of their parents may appear problematic because of that relationship’s historical ties to property ownership. But a closer look shows that children really do deserve different legal treatment than adults, and ius filiationis is one critical way the law can recognise that difference. 

Ius sanguinis feels retrograde today because it developed during a time in which relationships between parents and children, as well as relationships between husbands and wives and masters and servants, were much more akin to property-chattel relationships than we understand them to be today. Today’s family law was yesterday’s law of the household, which set forth entitlements and obligations based on reciprocal legal statuses – parent and child, husband and wife, master and servant, master and apprentice (and sometimes master and slave).  Each of these relationships was hierarchical, involving responsibilities on the part of the superior party in the hierarchy (father, husband, or master) and obligations of service on the part of the inferior party (child, wife, servant, apprentice, or slave).(1) The inferior party derived identity from the superior: a wife or a child’s nominal citizenship often followed that of the husband or father, but this identity did not confer the same rights enjoyed by the superior party. In early America, for example, male citizens were often entitled to the right to vote, right to contract, and right to own property (in fact, ownership of property was often a prerequisite for voting) but their wives – also technically citizens – were not entitled to any of these rights. Their political participation took the form of providing moral guidance to their husbands and raising virtuous sons who could themselves exercise political power.(2) 

Today, we no longer think of citizenship in this way. The rights conferred by citizenship are understood in Western democracies as universal. If, for example, I become a naturalised U.S. citizen, the same neutral voting laws apply to me that apply to any other citizen, regardless of my gender, marital status, race, or national origin. Likewise, laws that imposed derivative citizenship on wives, and even laws that expatriated women upon marriage – both of which used to be widespread – are no longer the norm. In many parts of the world, women are no longer understood as intellectually and financially dependent on their husbands but instead as autonomous adults, capable of making their own economic, moral, and legal decisions, including the decision to consent to citizenship or renounce it. And even more dramatically, we no longer think of servants as deriving legal identity from their masters; instead, workers are free to participate in free, if regulated, labour markets and their citizenship status is independent of their employee status.(3)

The one legal distinction, however, that all countries still maintain in determining the capacity to exercise the rights associated with citizenship is age. Children are generally considered to be incapable of giving legal consent and in need of legal protection. The particular age at which they become capable of reasoning is contested, but it is uncontestable that a newborn cannot care for himself nor meaningfully choose a nationality. In many circumstances, the law provides the protection children need by requiring children’s parents to provide for them, care of them, and make decisions for them; in some instances, the state takes on this responsibility (foster care and universal public education are both examples). Children occupy a very different legal space than women or workers, one that makes them vulnerable when their ties to their parents are weakened. Providing children with a citizenship that they can exercise simultaneously with that of at least one of their parents is a critical protection for their wellbeing. We can believe this to be so while simultaneously rejecting the traditional hierarchies of parent-child, husband-wife, and master-slave. The United Kingdom’s move away from conceptualizing parent-child relationships as “custodial,” property-like relationships and instead describing them as involving “parental responsibility” is a good example of this shift. The emphasis has changed from ownership and control to care and protection.

If, then, we still need a form of parent-to-child citizenship transmission, is ius sanguinis as traditionally understood what we need? Scholars, courts, and government agencies often take ius sanguinis literally, as the “rule of blood.” But I think that rigidity is misplaced. Even centuries ago the notion of ius sanguinis meant something distinct from a pure genetic tie. For men, who could never be certain of their child’s paternity, transmission “through blood” often really meant transmission through choice. A man chose to acknowledge his children by marrying, or already being married to, their mother. Children born to unmarried mothers generally took on the citizenship of their mothers, not their fathers, regardless of whether the father was known. The notion of “blood,” then, was complicated by the requirement of marriage for citizenship transmission through the father and the man’s unique ability to embrace or repudiate his offspring based on his marital relationship to their mother. Presumably, many children, prior to blood and DNA testing, acquired citizenship iure sanguinis when there was actually no blood tie, sometimes in circumstances where the father was ignorant of this fact and sometimes where he knew full well no blood relationship existed.(4) Ius sanguinis has always been about more and less than simply blood.

Thus, Bauböck’s notion of ius filiationis seems to me both the most appropriate form today for citizenship transmission from parent to child to take, and a more accurate description of what really occurred historically. As I see it, the most challenging obstacle to implementing a ius filiationis system is that birthright citizenship is fixed in time. Courts are not in a position to predict on the date of a child’s birth the adult who will ultimately assume parental responsibility for a child, but they can determine who the genetic or marital parent is. Shifting to a ius filiationis system, then, requires a multifaceted response. First, statutes outlining the requirements for citizenship transmission at birth should be amended to identify the intended parents. In most circumstances, the intended parents will be the genetic parents, but in some instances they might be someone else – for example, a non-genetic parent who contracts with a gestational surrogate or the spouse or partner of a genetic parent. Various pieces of evidence, from birth certificates to contracts to court judgments, would be necessary to determine parentage. In cases involving ART, this solution would solve many of the current problems. A genetic tie would be but one piece of evidence in determining citizenship at birth.

In addition to reforming ius sanguinis statutes, however, I believe we must also broaden the other available pathways to citizenship outside of birthright citizenship and traditional forms of naturalisation. There could be a deadline – perhaps by a specified birthday – by when a functional parent could request a declaration of citizenship for the child he or she has parented since birth. This alternative means of citizenship transmission should not substitute for birthright citizenship; as Kristin Collins points out, making citizenship determinations using only functional tests would put children at the mercy of officials seeking to deny citizenship and could disadvantage genetic or intentional fathers who may be unable to demonstrate that their care has been substantial enough to be “functional.” But combined with a robust recognition of genetic and intentional parentage at birth, recognition of functional parentage later on could serve a supplemental purpose, ensuring that children will ultimately have access to citizenship rights in the country in which their functional parents reside. Full recognition of parent-child relationships requires going beyond the moment of birth so that we can recognise the individuals who actually take on parental responsibility.

It is premature to forsake the recognition of parent-child relationships in citizenship law, not when citizenship is still the mechanism for ensuring that every human being has membership in at least one state and providing access to basic human rights. But it is time that we abandoned the idea that “blood” is the sole basis of these relationships. 



(1) Janet Halley, What is Family Law?: A Genealogy, Part I, 23 Yale Journal of Law & Humanities 1, 2 (2011).

(2) Linda Kerber, Women of the Republic: Intellect & Ideology in Revolutionary America (1980).

(3) In contrast to the independent citizenship status of workers, employer-sponsored immigration provisions may represent the vestiges of the ancient master-servant status relationship. See Raja Raghunath, A Founding Failure of Enforcement: Freedmen, Day Laborers, and the Perils of an Ineffectual State, 18 C.U.N.Y. L. Rev. 47 (2014).

(4) Kerry Abrams & Kent Piacenti, Immigration’s Family Values, 100 Va. L. Rev. 629, 660, 663, 692 (2014). 



Law by blood or blood by law?

By David de Groot (NCCR on the move, Universität Bern)


I agree to certain extent with Costica Dumbrava that ius sanguinis encompasses certain problematic issues, especially where it concerns newer forms of procreation, like IVF for lesbian couples and surrogacy. However, the origin of the problem cannot be attributed to ius sanguinis, but to non-solidarity of states that overuse the ordre public exemption for the denial of the recognition of parentage. But before delving into family relations and private international law conflicts, I would like to first argue that ius sanguinis is still the most suitable option for the main purposes of nationality law where it concerns children.


The main purposes of nationality

The commonly accepted main purposes of nationality are, first of all, that there is a territory to which an individual can always return and from which he cannot be deported, as was already pointed out by Bauböck and Titshaw; secondly, diplomatic and consular protection while being abroad; thirdly, national political participation in the state of nationality; and lastly, for EU citizens, free movement rights within the EU. 

An abandonment of ius sanguinis in favour of ius soli might lead to the situation described by Titshaw, where within the same family the children might have different nationalities, which could, for example, lead to the situation that they would have to move to different countries in case of their parents’ death while they are minor or that they might need to seek diplomatic protection from different foreign representations. Such a break-up of the family unit due to differing nationalities would certainly conflict with the right to family life. Therefore, for the purpose of preserving the unity and protection of the family, ius sanguinis is the most suitable option. If, when having attained majority the children feel that they have a closer bond with another nationality, they could still apply for naturalisation in that state. 

This bond of attachment brings me to the national political participation purpose of nationality which is connected to Dumbrava’s argument concerning the reproduction of the political community. Having the nationality of a certain state does not automatically mean integration into its society. This problem, depending on the mobility of the persons involved, does, however, not only occur with ius sanguinis and ius soli, but also with every other form of nationality transmission that one could think of. It should therefore be decided whom national political participation concerns most. If the definition of a ‘state’ refers primarily to a permanent population within its borders, long-term (non-national) residents should have national political participation rights and long-term absent nationals should not (except if they are working abroad in service of the state). National political participation rights should then be detached from nationality and therefore actually not be seen as a purpose of nationality (but that is a different discussion).  

It should however be noted that for purpose of inclusion of long-term resident families, who for some reason have not acquired the nationality by naturalisation, into the national population, a third generation ius soli or even a second generation ius soli, in cases where the first generation migrant has entered the country at a young age, would be appropriate. However, this should not come with an option requirement for dual nationals at reaching majority, as in Germany, in order to avoid a conflict of identity if one is forced to make a choice between the nationality acquired iure soli and another nationality acquired iure sanguinis. 


Non-Solidarity of States

The problems that arise when a state does not grant its nationality to a child due to non-recognition of parentage can only occur in cases where parentage has been established by another state in accordance with its national family law. In surrogacy cases this means a non-recognition of a foreign judgement or birth certificate and in cases of dual motherhood of married or registered lesbian couples a non-recognition of the extended pater est quem nuptiae demonstrant principle. The pater est principle means that the husband of the woman that gives birth to the child is automatically considered to be the father and therefore directly at birth has a parentage link to the child. Increasingly, states have extended this principle to stable non-marital relationships and to same-sex marriages.

If the child is born in the state of the discussed nationality the national family law (mostly) applies to the establishment of parentage. It would therefore not make any sense that parentage ties to a national could be established at birth by the state in question, without also granting the nationality (if ius sanguinis is applied). The problems that arise are thus nearly always recognition issues between states.

There is a general principle of recognition of a civil status which was legally established abroad. Recognition can only be refused in cases of overriding reasons of ordre public. This ordre public principle should be limited by the best interest of the child and the right to family life. It can never be considered to be in the best interest of the child to have no parents at all instead of having parents with whom (s)he has no blood ties who want to care for her or him. This has also been stated by the European Court of Human Rights in the Paradiso and Campanelli v. Italy case. In that case an Italian couple had gotten a child through a surrogacy arrangement in Russia. When they brought the child to Italy the state refused to recognise the parentage ties, took the child away and placed him under guardianship. The Court stated that Italy had failed to take the best interest of the child sufficiently into consideration when weighting it against ordre public. It had especially failed to recognise the de facto family ties and imposed a measure reserved only for circumstances where the child is in danger. Another example where the best interest of the child should prevail is when the child from a second (polygamist) marriage is put in a worse position than a child born out of wedlock. 

The problem is thus a lack of solidarity between states who do not recognise family ties legally established in another state. The parentage for the purpose of acquisition of nationality should thus be based on family law, including a more lenient approach in the private international law rules to recognition of a civil status acquired abroad.   

I therefore like Bauböck’s proposal of a ius filiationis. I see it, however, more as a change from ‘law by blood’, meaning parentage ties based on blood relationship, to a ‘blood by law’ relationship, meaning that parentage ties are seen to be established by the law. This thus means only an extension of the ‘blood’ definition. Bauböck’s fear that this could create a situation where the child could not acquire a nationality at birth, due to the complex determination of parenthood, could technically be avoided by a pre-birth determination of parentage.  


Limiting the transmission of family advantage: ius sanguinis with an expiration date

By Iseult Honohan (University College Dublin)


Costica Dumbrava has done a great service in stimulating us to reconsider the justification of ius sanguinis and to disaggregate its different forms.

I am sympathetic to critiques of ius sanguinis as a dominant mode of citizenship acquisition. Yet I acknowledge that the significance of family life for parents and children seems to offer some grounds for ius sanguinis citizenship – at least in a world of migration controls where citizenship is the only firm guarantee of right of entry to a country. I will argue here that to limit the extension of inherited privilege in this domain, however, this form of citizenship should be awarded provisionally.  

Others here have shown convincingly that there is nothing inherently ethnically exclusive about ius sanguinis. Furthermore, it does not have to be understood in terms of bare genetic descent; so sorting out the deficiencies of current ius sanguinis provision does not depend on resolving all the issues of biological parenthood raised by the new reproductive technologies.  If ius sanguinis can be detached from the strict genetic interpretation, it no longer provides a warrant for indefinite transmission across successive generations on the basis of biological descent.  Thus two of the sharpest criticisms of ius sanguinis seem to have been defused. 

It remains to consider in what way ius sanguins might be necessary. On the one hand, various forms of ius soli can be seen as giving continuity of membership for the state and security for children born in the country. For those born in the country of their parents’ citizenship there is little material difference between ius soli and ius sanguinis. But ius sanguinis citizenship may be seen as necessary when a child is born to parents living outside the state of their citizenship. Even if the child gains ius soli citizenship in the country in which she is born, this does not guarantee the security of the family. Focusing on what have been termed ‘social parenthood’, or functional parenting relationships of care, rather than simply biological descent, others here (Bauböck, Owen and Collins) have pointed to the way in which common citizenship best secures family life in allowing parents and children to stay together or move back to the country of their parent’s citizenship.  

What I want to address here is the further question: what forms or extent of ius sanguinis citizenship are warranted on the basis of this account?   

Protecting families but not privilege

We may start from the consideration that those in the position of parents have an interest in and a particular responsibility to care for their children when young, implying a clear and fundamental interest in living together and being able to move together. These can be seen as necessary conditions for realising many of the intrinsic and non-substitutable goods of family life, or what have been called ‘familial relationship goods’,  which include child-rearing and asymmetric intimacy (Brighouse and Swift, 2014). These involve agent-specific obligations that can be realised only within family relationships of care and throughout childhood (Honohan, 2009).  Thus this fundamental interest should be protected. Brighouse and Swift emphasise however, that we should not, in protecting these intrinsic goods, fail to distinguish them from other advantages external to familial relationship goods that parents can confer on their children, such as private education or concentrated wealth, which do not warrant protection (Brighouse and Swift, 2014).

Can ius sanguinis citizenship, even if not based solely on genetic descent, support such unwarranted transmission of privilege?  Citizenship grants more than the opportunity to live with and be cared for by your parents when you are a child. It provides membership of a political community and the benefits at least of entry and residence in that state, the right to participate in national elections and sometimes access to other rights. Under a regime of ius sanguinis, even understood as grounded in the rights of parents and children to share citizenship, the transmission of citizenship to children born to citizens abroad can mean that people with no connection to the country retain the benefits of citizenship, and, at the very least, can lead to a mismatch between the citizen body and the community of those who live in, and are particularly subject to, the state. Thus, life-long citizenship in the absence of real connections could well be seen as falling into the category of advantages that parents should not necessarily be able to convey to their children. 

This is not to suggest that ius sanguinis citizenship is just a form of property or unearned privilege. (cf. Shachar, 2010). But there are still concerns about how to secure the legitimate interests of parents to care for their children, and of children to be protected, without justifying the transmission of privilege.  My focus here is on considering how to guarantee the security of children to live and move with their parents through shared citizenship without supporting the unwarranted extension of privilege in the domain of citizenship. 

This suggests the following limited justification for birthright ius sanguinis citizenship - rather than the universal child status and deferred, or provisional, ius soli citizenship that Dumbrava recommends. 

Provisional ius sanguinis 

First, birthright citizenship per se is justified because people need the protection of citizenship from birth (1). Note that this is not mainly because they are children and thus innocent or particularly vulnerable (pace Harder), nor despite the fact that they are children and thus (arguably) not capable of consenting or participating politically, but while they are children, and like others, are both subject to the power of a state and in need of protection by a state. Dumbrava’s proposal that children might gain a universal status of childhood and that citizenship should depend on their being able to choose, have established a connection, and developed capacities and virtues of citizenship overlooks the centrality of the legal status of citizenship to security, and the fact that this security should not be conditional on the qualities or practices of citizens.

The specific justification of ius sanguinis citizenship then derives from the way in which common citizenship between parents and children is the most secure way of guaranteeing their ability to live and move together. This can be in addition to the citizenship the child may acquire by ius soli; dual citizenship of the state of birth and that of parents’ is not in itself problematic if a person has connections in both countries.

Because children need citizenship from birth, there is an argument for birthright citizenship; because young children need to be able to live with (and be cared for by) their parents, there is an argument for ius sanguinis citizenship at the time where this is most needed.  Both of these concerns support an award of citizenship that is not deferred, but that is also not always retained indefinitely.

It may be objected that the withdrawal of citizenship should not be lightly recommended.  Indeed this is true. But the strongest ground for withdrawal is the absence of any genuine link between a person and the state of citizenship. Thus, writing on birthright citizenship, Vink and De Groot (2010) offer a similar suggestion: ‘an alternative to limiting the transmission of citizenship at birth is the provision for the loss of citizenship if a citizen habitually resides abroad and no longer has a sufficient genuine link with the state involved’(2). Indeed they go on to say that ‘[f]rom our perspective, a provision on the loss of citizenship due to the lack of a sufficient link is to be preferred to limiting the transmission of citizenship in case of birth abroad’, on the grounds that this gives the child herself the opportunity to decide whether to establish that link, which thus should remain available until after majority, at the point when the child is better placed to make an independent decision (Vink and De Groot 2010: 12).  

Thus, the parsimonious account of ius sanguinis defended here suggests that it should be awarded only provisionally – held through childhood, but requiring the establishment of connections of certain kinds, most clearly by a period of residence in the country of that citizenship by, or soon after, majority (3). Confirmation would not depend on abjuring any other citizenship, as the aim would not be to avoid or reduce dual citizenship, but rather to reduce the numbers of citizens whose connections to a country are minimal or non-existent.  

Such a conditional citizenship could take seriously the justifiable claims of families without leading to the unwarranted extension of family advantage.



Brighouse, H., and Swift, A. (2014) Family Values: The Ethics of Parent-Child Relationships, Princeton University Press 2014.

Honohan, I. (2009) ‘Rethinking the claim to family re-unification’ Political Studies 57 4: 765-87

Shachar, A. (2010) The  Birthright Lottery,Cambridge MA: Harvard University Press.

Vink, M. P, and De Groot, G.R. (2010) Birthright citizenship: trends and regulations in Europe, EUDO Citizenship Observatory, Robert Schuman Centre for Advanced Studies, RSCAS/EUDO-CIT-Comp. 2010/8, Badia Fiesolana, San Domenico di Fiesole (FI), Italy



 (1)  Of course, not all birthright provisions apply from birth, rather than on the basis of birth,  but they generally apply from the establishment of the fact of birth, whether in the country or to a citizen.

 (2) Such provisions already exist in Belgium, Denmark, Finland, France, Iceland, the Netherlands, Norway, Sweden and Switzerland (see Vink and de Groot (2010)). In many of these cases, however, loss of citizenship can be pre-empted by submitting a request to retain it.

 (3) This would not necessarily be the only basis for retaining citizenship. If, for example, the parent(s) had returned to the country of their citizenship, this also could create a connection of their potential care in old age by adult children, which might justify their retaining citizenship.


Retain ius sanguinis, but don’t take it literally! 

By Eva Ersbøll (EUDO citizenship expert, Danish Institute for Human Rights)


There is no doubt that Costica Dumbrava has raised an important question about whether to abandon ius sanguinis citizenship. His arguments are that ius sanguinis is historically tainted and unfit to deal with contemporary issues such as developments in reproductive technologies and changes in family practices and norms; he also claims that ius sanguinis is normatively unnecessary, as it is possible to deliver its advantages by other means.

In my opinion, it is not time to abandon ius sanguinis, mainly because it is impossible to secure its advantages by other means. Admittedly, ius sanguinis, if taken literally, is unfit to deal with contemporary issues such as complex family arrangements involving, among other things, assisted reproduction technologies (ART). However, it seems possible to solve many problems by applying a modified principle of ius sanguinis translated into ius filiationis, as suggested by Rainer Bauböck and supported by most of the participants in this debate. 

What matters is, as also expressed by many authors, that children from a human rights perspective need their parents’ citizenship - or rather, the citizenship of their primary caretakers, be they biological parents or not. 

A solution to many of the problems related to reproductive technologies has been advanced by the Council of Europe, the Committee of Ministers, in Recommendation CM/Rec(2009)13 on the nationality of children: 

Member states should apply to children their provisions on acquisition of nationality by right of blood if, as a result of a birth conceived through medically assisted reproductive techniques, a child-parent family relationship is established or recognised by law (1). 

Still, it is of course necessary to examine more closely the arguments against ius sanguinis and the practical solutions to its shortcomings.

History is not an argument 

As Jannis Panagiotidis writes, history cannot justify abandoning ius sanguinis. The use of the principle may have been problematic in the past, and still, it may be all right today. Besides, as argued by Rainer Bauböck and others, it is possible to overcome ethno-nationalist dispositions by modifying a ius sanguinis principle, supplemented with ius soli and residence-based modes of acquisition. 

As things stand, ius sanguinis citizenship is in my opinion irreplaceable. It provides, in accordance with the Convention on the Rights of the Child (article 7) for automatic acquisition of citizenship by birth. In addition, it seems to be one of the most simple and secure acquisition modes when it comes to protection against statelessness, as it has the ability to protect children against statelessness from the very beginning of their life. 

What is more, it is a central international law principle. For instance, state parties to the European Convention on Nationality are obliged to grant citizenship automatically at birth to children of (one of) their citizens (if born on their territory, cf. article 6(1)).

To me, it seems risky to jettison such an effective principle anchored in binding human rights standards.

Unity of the family

Ius sanguinis is not the only relevant principle. Others, like the unity of the family, safeguard the same interests and may be applied in a broader perspective. To mention a few situations, take acquisition by adoption and acquisition by filial transfer based on the fact that the target person is a natural, adopted or foster child of a citizen. 

In addition, new automatic modes of acquisition by birth are developing. Denmark, for instance, has amended its law in 2014 to provide for automatic acquisition of citizenship by birth by children with ‘a Danish father, mother or co-mother’(2). This is an example of citizenship acquisition based on ius filiationis as advanced by Rainer Bauböck.

As Costica Dumbrava rightly anticipated, a reasonable reservation in this debate has been that the main problems connected with the development of ART do not lie with ius sanguinis citizenship but with the determination of legal parentage. Such determination may take long time and involve a number of legal uncertainties and ethical dilemmas. Still, as argued by among others Rainer Bauböck and Scott Titshaw, states have in any case to fix their family law and figure out how to determine legal parenthood. Subsequently, children’s right to their legal parents’ citizenship may not raise major problems.

Ius filiationis benefits

Developing a a ius filiationis principle may entail even more advantages. Among others, it may solve some of the problems originating from loss or so-called quasi-loss of citizenship following the disappearance of a family relationship (3). Disappearance or annulment of a family relationship may have consequences for a person’s citizenship based on that family relationship. Many states assume that if a person has acquired his or her citizenship through a child-parent family relationship that citizenship will be lost or even nullified if the family relationship disappears (4). If, however, states recognise citizenship based on social rather than biological parenthood, the threat of loss or quasi-loss may not arise in the case of disappearance of a biological family relationship. 

Human rights protection at this stage

According to the Council of Europe recommendations on the nationality of the child, quoted in the introduction, member states should apply the ius sanguinis principle in ART-cases where the child-parent family relationship is established or recognised by law. The crucial question is of course under which conditions the intended parents’ country must recognise such a family relationship if it has been legally established abroad.

David de Groot points out that states can only refuse recognition in case of overriding reasons of ordre public, and he criticises states’ overuse of the ordre public exemption for the denial of parentage. As he rightly argues, it cannot be in the best interest of the child to have no parents at all, instead of caring parents without blood ties. David de Groot refers to the 2015 judgment of European Court of Human Rights (ECtHR) in Paradiso and Campanelli v. Italy (5). Here, the Court ruled that the removal of a child born to a surrogate mother and his placement in care amounted to a violation of the European Convention on Human Rights article 8 on respect for private and family life.

In 2014, the ECtHR dealt with another case concerning the effects of non-recognition of a legal parent-child relationship between children conceived through assisted reproduction, Mennesson v. France (6). A French married couple had decided to undergo in vitro fertilisation using the gametes of the husband and an egg from a donor with the intention to enter into a gestational surrogacy agreement with a Californian woman. The surrogacy mother gave birth to twins, and the Californian Supreme Court ruled that the French father was their genetic father and the French mother their legal mother. France, however, refused on grounds of ordre public to recognise the legal parent-child relationship that was lawfully established in California as a result of the surrogacy agreement.

The ECtHR ruled that the children’s right to respect for their private life – which implies that they must be able to establish the substance of their identity – was substantially affected by the non-recognition of the legal parent-child relationship between the children and the intended parents. Having regard to the consequence of the serious restriction on their identity and right to respect for their family life, the Court found that France had overstepped the permissible limits of its margin of appreciation by preventing both recognition and establishment under domestic law of the children’s relationship  with their biological father. Considering the importance of having regard to the child’s best interest, the Court concluded that the children’s right to respect for their private life had been infringed.

The Court also dealt with the children’s access to citizenship as an element of their identity (see also Genovese v Malta) (7). Although the children’s biological father was French, they faced a worrying uncertainty as to their possibilities to be recognised as French citizens. According to the Court, that uncertainty was liable to have negative repercussions on their definition of their personal identity.

In Mennesson, the ECtHR’s analysis took on the special dimension where one of the parents was the children’s biological parent; it is, however, in my opinion difficult to imagine that the Court should reach a different conclusion in a similar case where both gametes and egg were from a donor. Paradiso and Campanelli may underpin this position that also appears to be supported by the fact that the Court has explicitly recognised that respect for the child’s best interest must guide any decision in cases involving children’s right to respect for their private life. In this context the Court has made it clear that respect for children’s private life implies that they must be able to establish the substance of their identity, including the legal parent-child relationship.  

Other ways to protect parent-child relationship

Costica Dumbrava argues that there are other and better ways to protect the parent-child relationship than through the same citizenship status, for instance by conferring full migration rights to children of citizens or establishing a universal status of legal childhood that protects children regardless of their or their parents’ status.

I find it hard to believe that any of these means can afford children a similarly effective protection of their right to a family life with their parents in their country.

Children need their parents’ citizenship¸ as pointed out by Rainer Bauböck and many others, because citizenship is a part of a person’s identity. Where and to whom one is born are facts that feed into developing a sense of belonging.  Moreover, the unity of the family in relation to citizenship secures that children can stay with their parents in their country. 

The course of events that followed the independence of women in citizenship matters seems illustrative. In Denmark for instance, when married women gained independence in citizenship matters in 1950, it was a major concern that in mixed marriages, where the spouses had different citizenship, the woman might lose her unconditional right to stay in her husband’s country. The legislator assumed that the aliens law would be administered in such a way that a wife would not be separated from her husband unless a pressing social need necessitated the separation (8). Things have, however, developed differently. Nowadays, foreigners married to Danish citizens are subject to the same requirements for family reunification as foreign couples. Thus, a foreign spouse may be expelled if for instance her Danish husband has received cash benefits within the last three years before a residence permit could be granted; notably, this may apply regardless of whether the couple has a child with Danish citizenship.

 A need for international guidelines on legal recognition of parenthood

As already mentioned, there is no doubt that Costica Dumbrava has raised an important discussion about continuous application of ius sanguinis citizenship. While there seems to be little support for abandoning the ius sanguinis principle, there seems to be almost unanimous support for modifying and modernising it. As recommended by the Council of Europe, states should apply to children conceived through medically assisted reproductive techniques their provisions on ius sanguinis acquisition of citizenship. 

The problem remains that states must establish or recognise the child-parent family relationship by law, and often, two states with different approaches are involved in the recognition procedure. Therefore, ordre public considerations may arise as demonstrated in many of the concrete cases mentioned in this Citizenship Forum. In order to achieve consensus about the recognition of a parent-child family relationship in the best interest of the child, states should engage in international cooperation with a view to adopting common guidelines – as they have done in adoption matters.



(1) See the recommendation at https://wcd.coe.int/ViewDoc.jsp?id=1563529 

(2) Costica Dumbrava gives an inadequate Danish example regarding the acquisition possibilities for children born out of wedlock. For long, such children have been entitled to naturalise regardless of residence in Denmark, although until 2013, it was a requirement that the father had (shared) custody over the child. This requirement is now repealed.

(3) See more about quasi-loss of citizenship at http://www.ceps.eu/publications/reflections-quasi-loss-nationality-comparative-international-and-european-perspective 

(4) See more about quasi-loss etc. at http://www.ceps.eu/publications/how-deal-quasi-loss-nationality-situations-learning-promising-practices 

(5) Case of Paradiso and Campanelli v. Italy, judgment of 27 January 2015 http://hudoc.echr.coe.int/eng#{"itemid":["001-150770"]} 

(6) Case of Mennesson v. France, judgment of 26 September 2014 (Final) http://hudoc.echr.coe.int/eng#{"fulltext":["mennesson"],"documentcollectionid2":["GRANDCHAMBER","CHAMBER"],"itemid":["001-145389"]} 

 (7) Case of Genovese v. Malta, judgment of 11 October 2011 http://hudoc.echr.coe.int/eng#{"fulltext":["genovese"],"documentcollectionid2":["GRANDCHAMBER","CHAMBER"],"itemid":["001-106785"]} 

(8) See the Danish citizenship report at http://cadmus.eui.eu/bitstream/handle/1814/36504/EUDO_CIT_CR_2015_14_Denmark.pdf?sequence=1 


Distributing Some, but Not All, Rights of Citizenship According to Ius Sanguinis

By Ana Tanasoca (University of Canberra)


In an article published in 1987 Joseph Carens famously remarked that “[c]itizenship in Western liberal democracies is the modern equivalent of feudal privilege – an inherited status that greatly enhances one’s life chances. Like feudal birthright privileges, restrictive citizenship is hard to justify when one thinks about it closely.” (1) Some 30 years after, he himself offers a justification of birthright citizenship, a change of heart and mind that he partly explains by the following: “I thought that my open borders arguments was getting at an important truth. At the same time, I recognized that it was not a practical proposal and that it did not provide much guidance for actual policy issues…”; “In thinking about what to do in a particular situation, we have to consider questions of priority and questions of political feasibility, among other factors. One cannot move always from principles to a plan of action.” (2) Yet succumbing too much to such feasibility constraints, to use a popular term in the field, is dangerous. Moral (political) theorizing should not be too tightly hemmed in by empirical facts. Rather it should be the other way around, insofar as our moral and political theory aims to tell us what existing empirical facts we should strive to change or overcome.

That is why Costica Dumbrava’s critique of the ius sanguinis principle of citizenship ascription is, in a way, a much-needed intervention.(3) While I overall agree with Dumbrava’s argument that ius sanguinis is unable to cope with the diversification of family structures and not that morally appealing to begin with, I disagree with him on the details. I disagree especially with his background assumption that family ties (although not exclusively genetic, as it is presently the case) must play a salient role in the distribution of citizenship – although in the second part of this contribution I do offer a potential defence of his view against what is probably the strongest objection to his argument, which is that the abolishment of ius sanguinis would split families apart.

The main question is: Why should we insist on ius sanguinis except because it would ensure that nobody is stateless, that is, that everyone’s human right to citizenship is satisfied? And insofar as statelessness can be equally avoided via ius soli, why should blood ties create an entitlement to citizenship?

The problem of making citizenship dependent on family ties

Dumbrava notices that ius sanguinis is unable to cope with the increased diversification of family structures made possible by the assisted reproduction technologies (ART). Yet there are solutions to that problem. 

One would be, as Scott Titshaw notices, to reform family laws as to recognise diverse forms of parentage. Another one would be to replace ius sanguinis with ius filiationis, as Rainer Bauböck proposes. If the purpose of upholding ius sanguinis citizenship is to recognise and protect the family, we should replace it with more reliable indicator(s) of parenthood in the case where parenthood is no longer uniquely a matter of biology. As Kerry Abrams argues, the recognition of parenthood now requires “going beyond the moment of birth.”

Notice, however, were multiple indicators of parenthood to be accepted, those individuals born via ART might be entitled to multiple citizenships. They might, for example, be entitled to the citizenship of the egg donor or the sperm donor or the surrogate mother, as well as to the citizenship of those who intend to raise the child. Such a situation may be deemed problematic in various respects: first because it would create great inequalities; second, because it would end up trivializing citizenship if all types of parenthood (e.g., the relationships the surrogate mother, the egg donor or the sperm donor, and the intended parents have with the child) would be treated as equally morally relevant and therefore worthy of state recognition. 

Dumbrava also bemoans ius sanguinis as failing to capture the political function of citizenship. If we grant citizenship to the children of citizens because we expect such children to develop the attitudes and skills required for political participation in their parents’ state, why not wait to confer citizenship until these attitudes and skill are actually confirmed? And what would happen if they never develop these skills and attitudes? Should people be deprived of their birthright citizenship altogether, or perhaps only of their political rights? Besides, while we might have a clear idea of what skills (e.g., reading and writing to enable voting) citizenship requires, what can we say of the attitudes citizens should display? Should apathetic voters be stripped of their political rights for failing to display the right attitude towards their right to vote? According to Dumbrava’s reasoning, perhaps we should.

But the main problem both with Dumbrava’s critique and the other contributors’ accounts is that they conceive of citizenship as primarily reflecting a bond (genetic or affective or intentional) between two individuals—the parent and the child – and not as a bond between an individual and a state, or an individual and a community. As such, it overlooks the political nature and function of citizenship.  It is also likely to leave us with a very limited, rigid, and exclusionary conception of the demos, one that is at the same time unjust and inefficient. As Rainer Bauböck put it elsewhere, “[n]ormative principles for membership must instead lead to boundaries that avoid both under- und over- inclusiveness,” (4) particularly in the context of increased global mobility.

In his contribution to this forum, however, Bauböck argues that birthright citizenship creates a “quasi-natural equality of status” among those entitled to it. He represents it as avoiding divisions, by making citizenship part of people’s unchosen and permanent personal features, namely, where and to whom one is born.(5) Yet as such it creates exclusion and inequality between those entitled and those unentitled that can be hard to justify or overcome, as Lois Harder rightly notices. Why should the son of a citizen of state A be entitled to citizenship in that state, but not a regular immigrant residing for years in state A, paying taxes there and having virtually all of his interests deeply affected by the institutions of state A? While the first has unconditional and automatic access to citizenship – a right to citizenship in virtue of his blood ties to another citizen – the second has to apply for naturalisation, which is subject to the state’s discretionary powers. That is, his residence in that state, contributions to the community or his interests being affected by that state’s institutions, do not automatically ground any right to citizenship for him in the same way blood ties do for the citizens’ progeny.

Why should the boundaries of the demos be defined by family ties, rather than social or political kinship? By ascribing citizenship on the basis of blood ties we conceive of political communities as big extended families rather than communities gathered around common interests, values, and goals. Such a conception of the demos is disrespectful of individual consent (no one consents to being born, to having these parents rather than others or to the colour of their passport).  It attaches too much value to contingencies and too little value to individual choices. A political community based on ancestry is, after all, just an overinflated dynasty. 

Limiting the scope of ius sanguinis

While abolishing ius sanguinis might be a good idea, we could nonetheless be worried that the transition costs would outweigh potential benefits. After all, most families today are still founded on blood ties. Abolishing ius sanguinis altogether could create situations where parents and children are not citizens of the same state. Such policy, it is argued by several contributors, would have the disruptive effect of potentially separating families, preventing parents from discharging their parental duties and leaving children deprived of the care they are entitled to. (Of course, nothing prevents parents from applying for a visa or for citizenship if they wish to reside or share a citizenship with their progeny; but let us assume that the parents do not have the means to do that, or that even doing that would not guarantee that they can be reunited with their child immediately as we would wish.) This is, I think, the strongest argument against Dumbrava’s proposal. 

One solution would be, of course, to replace ius sanguinis with another principle for citizenship allocation, perhaps affected interests or perhaps ius domicilii. As children’s and parents’ interests are interdependent, the affected interests principle would ensure that children and parents are members of the same state. So would ius domicilii, at least in cases where parents and children are currently domiciled in the same state (although it would provide no citizenship-based grounds for family reunion, in cases where they are not).

My proposal, however, takes a different tack.  Notice that in a world with genuinely open borders we need not be worried that parents and children would be separated if they are citizens of different states.  The solution I propose would therefore be to limit the scope of ius sanguinis – that is distribute some, but not all rights traditionally associated with citizenship, on the basis of ius sanguinis. This would be an appealing compromise, insofar as some of us may think citizenship should not be distributed on the basis of blood ties, while nonetheless accepting that blood ties are one (albeit not the only) relevant ground for the distribution of some categories of rights. 

As Bauböck notices in his contribution, immigrant minors who are EU citizens have a “right to stay” that protects their primary caregivers from deportation. Yet, most likely, this policy is a recognition of an entitlement to care that the child has – not a recognition of a right the parents have to stay strictly in virtue of their blood ties to the child.  Blood ties may simply serve as the operational indicator of the primary caregivers. 

My preferred solution, however, would entitle a person to the limited enjoyment of some rights in a state, on the basis of having blood ties to someone who is already a citizen of that state. I primarily have in view, among that limited subset of rights, the right to enter and leave the state and the right of residence. By “limited” I also mean that the enjoyment of these rights, purely on the basis of ius sanguinis, should be time-constrained. (6) 

Take the case of minors having a different citizenship from their parents. My proposal would be:  either the parents should be granted extensive residence rights, until the minor reaches adulthood as in the case above; or else the minor should be granted these rights, provided the parents wish to remain in their country of citizenship. Consider the case of a couple, both citizens of state A, who move to state B and give birth there to a child, who becomes via ius soli citizen of B. Under my proposal, the parents would be automatically entitled to residence in state B until the child is 18, provided the family decides to reside in state B; equally, the child would be automatically entitled to reside in state A until 18 if the family decides to reside there. 

Things would be different in the case of adults. Say my mother and I are citizens of different countries, she of state A and I of state B. Under my proposal, I as an adult would not be entitled to all the current rights of citizenship in state A on the basis of ius sanguinis. Still, I may nonetheless be automatically entitled on the same ground to a right to freely enter state A and reside there for a limited period of time (for example, 1 month). That would allow me to visit and spend time with my mother, preserving my family ties intact and allowing me to discharge whatever ordinary duties I have towards family members. But what if my mother becomes frail or ill, and I become her caregiver and need to spend more than one month in state A? If the circumstances require it, I should be able to petition for my right to remain to be extended, and that petition should be automatically granted so long as authorities are satisfied that the requisite circumstances really do prevail. The period for which one can enjoy such rights, and the categories of rights one enjoys, might be extendable in this way. Alternatively, of course, I could bring my mother to reside with me in state B on a (elderly) dependent visa. 

Under my proposal, there would thus be a limit to what one is entitled to under ius sanguinis alone. We should not think of the distribution of citizenship rights as an all-or-nothing affair. Among the many component rights currently associated with citizenship, different rights can and should be distributed separately according to different criteria. By the same token, many different criteria can serve as a legitimate ground for the distribution of any one of those constituent rights. 


1. Joseph Carens, “Aliens and citizens: the case for open borders,” Review of Politics, 49 (1987), 251–73.

2. Joseph Carens, The Ethics of Immigration (Oxford: Oxford University Press, 2013), pp. x, 3.

3. I say “in a way” because he also relies heavily on empirical facts when arguing against ius sanguinis.

4. Rainer Bauböck, “Morphing the demos into its right shape. Normative principles for enfranchising resident aliens and expatriate citizens,” Democratization, 22 (2015), 820–39. 

5. This last bit is problematic in itself. Tying citizenship – that has an immense influence on individuals’ life opportunities and welfare – to underserved and permanent personal features like ancestry is after all morally problematic even if practically convenient for states.

6. In the same vein, Iseult Honohan proposes in her contribution to this debate that minors born in another states other than that of their parents should be also entitled to their parents’ citizenship but only until they reach adulthood; from then on, they can lose this citizenship if they do not continue residing in the country of parental citizenship. This would be another way of limiting ius sanguinis entitlements.


Learning from naturalisation debates: the right to an appropriate citizenship at birth

By Katja Swider (University of Amsterdam) and Caia Vlieks (Tilburg University)


Citizenship has a political and a legal dimension. In his opening contribution, Costica Dumbrava only marginally addresses the legal dimension of citizenship, acknowledging its importance, but suggesting that it is replaceable with alternative arrangements, such as a universal status for children. Maybe he is right in his priorities; maybe citizenship status should primarily be reserved for the purpose of fostering a political community. But in reality much legal baggage is attached to citizenship, and one cannot simply shake it off, even if this appears normatively attractive. In a way, the whole human rights movement can be seen as an effort to separate access to legal rights from possessing a status of political membership, and this attempt has not reached its goal (yet). As Jannis Panagiotidis points out, “most so-called human rights are in fact citizens’ rights”. Citizenship is still the ‘right to have rights’. Avoidance of statelessness is therefore not just a legal whim; it is a human rights failsafe mechanism. 

In our contribution we start from the assumption that leaving anyone, including (and especially) children, without a citizenship for any significant period of time is not an option due to the essential legal rights that are attached to the status of national citizenship. The question therefore is not whether children should acquire a citizenship at birth, but which citizenship they should acquire at birth. Should it be the citizenship of their parents? And if not, what alternatives to birthright citizenship arrangements are adequate? 

While we consider attribution of citizenship at birth to be necessary, we also maintain that it is inherently unfair, regardless of what mechanisms of attribution are relied upon. There is nothing fair about attaching the fate of a child to one state, when states differ so tremendously in their ability (and willingness) to provide access to basic rights, such as education, healthcare, physical safety and pursuit of happiness for their minor citizens. Rainer Bauböck shifts attention from this unfairness by suggesting that “we have to address the causes of global inequality directly” instead of criticising the contingencies of birthright citizenship. However, we should not forget that this discussion takes place largely among the privileged ‘winners’ of the ‘birthright lottery’ (1). There is no doubt that global inequalities need to be addressed, but is it morally justifiable to suggest to the ‘losers’ of the birthright lottery to wait for global equality? 

If fairness in birthright citizenship cannot be achieved and leaving children without any citizenship is unacceptable, what is the normative ideal that we could strive towards in attributing citizenship at birth? As Lois Harder correctly argues here, rules about birthright attribution of citizenship are as politically charged as rules about acquiring and losing a nationality during adulthood, even though the former are not as much part of the public debate. According to Harder, “[t]he magical power of birthright citizenship is that it makes it possible for us to know and rehearse [politically charged] rules while simultaneously making birthright seem straightforward, static and apolitical”. Can we reverse this logic, and perhaps also learn from the extensively politicised discourse on migrants’ rights to naturalisation in order to improve birthright citizenship rules? 

In particular, we suggest applying the concept of appropriate citizenship to strengthen the normative foundation of birthright citizenship attribution. This notion is based on the ideas of Ernst Hirsch Ballin, who advocates “a citizenship that is appropriate to everyone’s life situation, where he or she is at home – which can change during the course of a person’s life: a natural right to be recognized as a citizen, born free”(2). He believes that this type of citizenship and citizens' rights can overcome the existing gap between “the universality of human rights” and “the changing political and social settings of people’s lives”(3). Drawing on that, we feel that appropriate citizenship, even when acquired at birth, could do the same. Appropriate citizenship is of course a highly subjective concept, the interpretation of which would be dependent on numerous cultural and specific national legal factors. Ensuring that birthright citizenship is appropriate would imply a case-by-case evaluation of the individual situation of each newborn, a process which in most cases would be as simple as the registration of birth, but in some cases would require a complex investigation to be conducted in a very brief period of time.  

While perhaps logistically counter-intuitive, introducing the normative standard of appropriateness into the attribution of citizenship at birth is not more complex than trying to solve ad hoc ‘hard cases’ of citizenship within the traditional logic of ius sanguinis versus ius soli. This complexity of some birthright citizenship cases has been extensively discussed in the contributions by Dumbrava and Scott Titshaw. Requiring that birthright citizenship is appropriate emphasises the importance of (meaningful) ties (4) of a person (including a child) to a country, and thus incorporates the idea of ius nexi discussed by David Owen. With the criterion of appropriateness we accept that birthright citizenship is a political issue, not a contingent biological fact of life, and therefore should be based in a reasoned decision-making process and subjected to normative criticism. 

The requirement that citizenship acquired at birth needs to be appropriate is far from being precise. However, we believe that a certain amount of flexibility is necessary in order to ensure that attribution of citizenship at birth has a normative foundation in each individual case. The exact modes of implementation of the criterion of appropriateness would need to be developed within the individual legal systems, but important factors to be considered include the ones that have been discussed elaborately in this forum discussion:  

-the nationalities of the persons that are expected to care for the child (biological, social or functional parents or otherwise, thus including and reinforcing the ius filiationis proposal put forward by Bauböck);

-the country where the child is born;

-the country where the child is expected to build his or her future, receive education and effectuate his or her rights as a citizen;

-the necessity of ensuring that at least one nationality is acquired and that the best interests of the child are safeguarded (in line with the almost universally ratified Convention on the Rights of the Child) (5).

It is not always easy to determine all the relevant criteria for establishing appropriateness of citizenship with a high degree of certainty. Kerry Abrams, for example, identifies some possible obstacles when discussing Bauböck’s ius filiationis proposal, namely that courts sometimes cannot determine who will ultimately be the parent that is truly (legally) responsible for the child. However, since the proposal of appropriate nationality is based on multiple relevant factors rather than a single one, the risks associated with the inability to assess some of the factors are ameliorated by the availability of other factors that can compensate for uncertainties.

Finally, we would like to emphasize that Hirsch Ballin’s ideas and the concept of appropriate nationality that we have introduced are compatible with having multiple nationalities, as well as changing one’s nationality over the course of one’s life. It is appropriate to enable children, as well as adults, to acquire a new nationality to reflect the changes in their personal circumstances. When attributing an appropriate nationality at birth to a child, states therefore do not need to embark on the impossible task of predicting the future.



(1) Ayelet Shachar, The Birthright Lottery. Citizenship and Global Inequality (Harvard University Press 2009). 

(2)  Ernst Hirsch Ballin, Citizens’ Rights and the Right to Be a Citizen (Developments in International Law, Vol. 66, Brill Nijhoff 2014) 145.

(3)  ibid. 144.

(4) Or ‘genuine connection(s)’, see also Nottebohm (Liechtenstein v Guatemala) ICJ Reports 1955, p 4; General List, No 18.

(5) See Articles 3(1) and 7 of the Convention on the Rights of the Child.


Don’t put the baby in the dirty bathwater! A Rejoinder

By Costica Dumbrava (Maastricht University)


This has been a fascinating debate that succeeded in unravelling some of the major issues about the past, present and future of ius sanguinis citizenship. I was delighted to see that many of the contributors shared my concerns about the failings of the current system of transmission of citizenship from parent to child. I learned a great deal from reading the various reactions to my deliberately provocative propositions. With these concluding remarks, I use the privilege of the last word to engage with several key points emerging from the debate and to clarify and, as much as possible, elaborate my position. However, I am hopeful that this debate does not finish here and I look forward to continuing through other ventures.

How ethnic is ius sanguinis and why does it matter?

I think we are in agreement that ius sanguinis is not inherently ethnic and that it can take on ethnic connotations depending on particular historical and policy contexts. The apple of discord is whether the gravity of such occurrences recommends the abolishment of ius sanguinis. I concede that empirical evidence is not conclusive for dismissing the principle of ius sanguinis. However, I caution that we should not underestimate the dangers of ethnonationalist instrumental uses of ius sanguinis. 

Panagiotidis explains clearly the difference between legal descent (descent from a citizen) and ethnic descent (descent from a non-citizen of a particular ethnicity) and shows that the objection about the ethnic character of ius sanguinis is founded on a big conceptual confusion. While I agree that ius sanguinis is conceptually distinct from ethnic or racial descent, I would hesitate to say that the two have “nothing to do” with one another. Unfortunately, it is not only distracted scholars that make this confusion. The ambiguity between legal and ethnic descent is often present in legal practices and political discourses about birthright citizenship. In my initial contribution I mentioned co-ethnic citizenship because these policies frequently rely on the ambivalence between legal and ethnic descent. For example, legal criteria of descent from citizens (or from former citizens or from former citizens of a former part of a country, etc.) are often used as a smoke screen for selecting future citizens according to (perceived) ethnic descent. It matters less that these policies rarely achieve the goal of ethnic selectivity as long as the very statement of the commitment to include co-ethnics is likely to bring significant political and ideological gains. As Decimo and Harder argue, despite being a technical and legalistic principle, ius sanguinis carries significant ideological connotations, among which the myth of commonality of blood or ethnic descent is often prevalent. 

I also doubt that the ethnonationalist uses of ius sanguinis are only a matter of the past and I am not convinced that they are unlikely to be “used like that in the future” (Panagiotidis). What else if not the fear of ethno-national extinction drove Latvia and Estonia in 1990 to reinstate their pre-war citizenship laws and to apply ius sanguinis retrospectively back to pre-1940 citizens? It is besides the point that not all newly recognised citizens were ethnic Latvians or Estonians (as not all of the pre-war citizens were). The political-nationalist gains obtained from the perception that the overwhelming majority of them were co-ethnics and from the symbolic reinstatement of the original national citizenry were significant. The same can be said about the Romanian policy to restore citizenship to all those who lost Romanian citizenship independently of their will. In this case, ius sanguinis has been used to trace descendants of citizens several generations back in view of recovering the “national stock” lost with the territorial changes during WWII.

It is true, as Bauböck and Collins rightly point out, that both ius sanguinis and ius soli (and combinations thereof) can have either emancipatory or exclusionary implications, depending on the context. Since empirical facts do not translate well into normative arguments (Tanasoca), I think that wrestling over empirical evidence about the positive or negative effects of ius sanguinis is not going to help us settle the normative questions about the justification of the principle of ius sanguinis. If we have strong moral reasons for maintaining ius sanguinis, we should endorse it regardless of how wrong it is applied in practice and how often this happens. Of course, we should adjust the ways in which to implement a morally justified principle to match changing empirical circumstances. Yet, the prior question is whether ius sanguinis can be morally justified as a principle of admission to citizenship. 

Why bother fixing ius sanguinis?

Many contributors to this debate grant that ius sanguinis is a morally justified principle and propose ways to reform the ways in which we implement it. Bauböck, Ersbøll and Abrams argue that the ethno-nationalist disposition of ius sanguinis can be counterbalanced through adopting supplementary ius soli and residence-based naturalisation. Bauböck, Titshaw, Abrams and De Groot discuss possibilities of rethinking legal parentage in order to accommodate complex cases of citizenship determination in the context of ART birth. 

There is a broad consensus that ius sanguinis should be reformed, albeit disagreements prevail as to how and by whom. Bauböck’s proposals of ius filiationis, which reinterprets legal parenthood as a combination of genetic and social parenthood, is cheered by some but welcomed with scepticism by others. Titshaw and Collins, for example, worry that ius filiationis will not eliminate the uncertainty related to the determination of legal parentage and that it may also encourage abuse. Another contention is about the administrative level at which decisions about ius sanguinis should be taken. Writing in the context of the US federal system, Titshaw argues that fixing the family law will solve many problems related to legal parentage and therefore to ius sanguinis citizenship. Yet, Collins fears that leaving citizenship determination to those applying the family law will unwarrantedly expose citizenship to parochial concerns (e.g. immigration control). I think this is an important point, which we should consider beyond the level of administrative decision-making. I argue that the recognition of legal parentage and the determination of citizenship should not only be implemented through two separate procedures, but also regarded as two normative processes driven by distinct principles. While I appreciate the practical importance of the proposals for reforming ius sanguinis, I am not convinced that the strategy of fixing legal parentage addresses the prior and more fundamental question about the moral justification of ius sanguinis as a principle of admission to citizenship.

It is surprising to me that in a debate about ius sanguinis citizenship so little is being said about citizenship. Most contributors seem to take for granted the normative link between parentage and citizenship and to give priority to instrumental arguments over normative ones. Let me explain this point by discussing three key arguments in support of ius sanguinis: (1) ius sanguinis protects children against statelessness; (2) ius sanguinis enables and protects family life; and (3) ius sanguinis expresses the social identity of the child.

Preventing statelessness

There is a wide consensus in the debate that children need (at least one) citizenship from birth and that ius sanguinis provides the “most simple and secure” means (Ersbøll) to prevent statelessness. This view is accepted even by those who argue that birthright citizenship is ultimately an unfair arrangement (Swider and Vlieks). It is true that in today’ world the possession of the legal status of citizenship (aka nationality) predetermines access to a set of important rights and privileges, in the absence of which a person’s life is significantly constrained. It is also true that, despite a number of complications caused by changing family patterns and the spread of assisted reproductive technologies, ius sanguinis still provides a relatively simple solution to tackle statelessness at birth. However, one can think of other ways to prevent statelessness that are equally convenient, as well as better justified normatively.

The problem of statelessness could be arguably solved by a system of generalised unconditional ius soli or by a citizenship lottery in which new-borns are assigned randomly the citizenship of a state. These alternatives remove the uncertainties associated with the determination of legal parenthood for the purpose of ius sanguinis. However, convenience alone does not count for normative justification. Against the citizenship lottery suggestion, defenders of ius sanguinis would probably insist that new-borns should receive the citizenship of “their” parents. Notice that this is not an argument about convenience anymore but one about the importance of a shared citizenship between parents and children. But nothing in the argument about avoiding statelessness requires shared citizenship between parents and their children.  To avoid statelessness at birth (in the absence of ius soli), it is sufficient that a child receives one citizenship from either of the parents. This means that in international families only one parent needs to transmit citizenship to the child and, if a parent has multiple citizenships, he or she needs to transfer only one these citizenships to the child. The argument about avoiding statelessness does not offer any guidance as to which citizenship should be shared between parents and children and why.

Alternative solutions based on ius soli elements may offer better normative justifications. I argued elsewhere that states have a collective duty to grant access to a fundamental status of legal protection (nationality) to those born and living in their jurisdiction due to states’ joint participation in an international system that leaves individuals no real possibility of opting out, i.e. to establish a new citizenship or to remain stateless. My point here is not that the parent-child relationship has no normative implications for citizenship; it is merely that the argument about avoiding statelessness is unable to bring such normative concerns to the surface. 

Protecting family life

The second major argument in defence of ius sanguinis is that the (automatic and immediate) transmission of citizenship from parent to child enables and protects family life. In the absence of a shared citizenship between parents and children, it is feared, family life would be severely disrupted as family members risk being separated from one another by borders and immigration restrictions. I do not contest that family life deserves special protection and that the legal recognition of parent-child relationship provides “critical protection for their [children’s] wellbeing” (Abrams). However, I am not convinced that the automatic and immediate transfer of citizenship from parent to child is a major normative prerequisite of family life. 

It appears to me that the overwhelming majority of contributors subscribe to an indirect and instrumental defence of ius sanguinis. The biggest concern is about securing joint migration rights for family members, which are instrumental for family life. De Groot mentions two other important citizenship privileges, i.e. diplomatic and consular protection and political participation, but surrenders quickly to the concern about migration rights. The prevailing argument in these interventions is not so much a defence of ius sanguinis citizenship but a defence of ius migrationis sanguine – the right to migrate in virtue of a blood relationship. The downside of linking too tightly ius sanguinis to family migration rights is that the argument only holds as long as migration rights are strictly determined by citizenship status and as long as there are no other ways to secure migration rights for family members apart from ius sanguinis. Hence in a world of (more) open borders, where children would not be separated from their parents or siblings by migration restrictions, ius sanguinis citizenship loses its importance. However, a system of generalised family migration policies, such as the one suggested by Tanasoca, could provide the “permanence and stability” (Titshaw) required for achieving meaningful family life in the absence of ius sanguinis citizenship.

Expressing social identity

Another intriguing argument in defence of ius sanguinis rests on the idea that (birthright) citizenship is an important part of a child’s social identity. According to the judgement of the European Court of Human Right in the case Genovese v Malta, the failure to acquire a particular citizenship at birth is likely to affect negatively the identity of the child. I distinguish two versions of this argument: a softer/instrumental version, according to which the ius sanguinis principle “makes citizenship a part of citizens’ personal identities that they are like to accept” (Bauböck); and a harder/essentialist version, for which the ius sanguinis principle recognises and confirms the (inherited) identity of the child. 

The essentialist version of the argument about a child’s social identity can be easily dismissed by pointing at the fact that citizenship is a contingent social and legal convention rather than a mechanism that confirms prior genetic, ethnic or cultural identities. Recall that in the Genovese case the Court used this argument in connection with the principle of non-discrimination. The failure to acquire citizenship via ius sanguinis by a child born out of wedlock will affect negatively his or her social identity because children born in wedlock do not face similar restrictions of ius sanguinis as children born out of wedlock. The situation can be remedied not only by removing the discriminatory treatment in the application of ius sanguinis but also by abolishing ius sanguinis altogether. The instrumental version of the identity argument is more interesting, not least because it supports our intuition that (birthright) citizens are likely to feel attached to their country of birth. However, this is valid for both ius sanguinis and ius soli, so the instrumental argument cannot show why we should preserve ius sanguinis or why we should chose one form of birthright citizenship over another. 

Long-lasting institutions usually shape people’s attitudes and generate attachments and identities. They acquire the kind of “quasi-naturalness” that Bauböck ascribes to birthright citizenship. However, the test of time and familiarity is not a valid moral test because bad institutions can also acquire that kind of “magical power” (Harder). We ought to question the moral foundations of deeply rooted institutions such as birthright citizenship especially because they are so popular and because they shape our identity. 

Opportunities for intergenerational membership

There are several arguments in the debate that deal more seriously with normative aspects of ius sanguinis citizenship. I agree with Owen that the principle of ius nexi or genuine connection is the best we have for determining access to citizenship and that this general principle can be served by different policy arrangements, including some form of qualified ius sanguinis. I assume that the principle of “appropriate citizenship” defended by Swider and Vlieks goes along the same path. My concern with their proposal is that allowing for “a case-by-case evaluation of the individual situation of each newborn” (Swider and Vlieks) might not serve well the commitment to avoid statelessness, which seems essential to the principle of appropriate citizenship. 

Honohan endorses the principle of genuine connection and defends a limited version of ius sanguinis by arguing for imposing restrictions to the intergenerational transmission of citizenship. She endorses ius sanguinis but proposes that citizenship be withdrawn from (adult) citizens who fail to develop a genuine link with the country. I am sympathetic to this proposal but I am not fully convinced about its underpinning justification. Honohan’s main objection to ius sanguinis, which is shared by Decimo and Harder, is that the unconditional acquisition of citizenship by children from their parents can amount to an unfair privilege. Although I acknowledge the implications of citizenship policies in today’s world characterised by sharp economic inequalities, I think that the concern with economic privilege should be disconnected from the concern about admission to citizenship. I agree with Bauböck that there are more appropriate means to fight global inequality and injustice than redistributing citizenship (e.g. economic redistribution, fairer migration policies).

Honohan rightly argues that citizenship “provides membership of a political community” but she does not explain why children should be admitted in the political community of their parents rather than in another (e.g. the best political community). My answer is that both parents and children have an interest in the continued participation to a particular intergenerational political project. This interest can be served through providing opportunities for intergenerational membership in the form of provisional ius sanguinis. The citizenship acquired provisionally at birth should be withdrawn upon majority from those (provisional) citizens who do not have a genuine link with the country. However, if a person fails to prove a genuine link with at least one country, his or her provisional citizenship should still be extended but only in the form of formal legal membership, i.e. without political rights.

Notice that the argument for intergenerational provisional citizenship stands even after we solve the problems related to the recognition of parenthood and to migration restriction for family members. Bauböck points at this when talking about the “signalling effects of birthright citizenship” but his argument slides into an instrumental and collectivist defence of birthright citizenship. My argument for intergenerational citizenship puts emphasis on the individual interests in continued political membership. Incidentally, this solution is also likely to have positive implications for the political community as a whole, e.g. by fostering “a sense of responsibility towards the common good and future generations” (Bauböck). I am sympathetic to Harder’s idea of political membership as a “lively on-going process of negotiation in which everyone has a stake”. However, I disagree that admission to political membership should be entirely up to negotiation, as I maintain that there are certain concerns that demand inclusion regardless of people’s preferences and abilities. I also no not think that political membership should be “limited by our mortality” (Harder). While I reject continuation based on genetic, ethnic and racial traits or simply convenience, I argue that there should be opportunities for intergenerational political continuity, which can be provided through provisional ius sanguinis.

It is beyond dispute that any attempt to dislodge a deeply rooted and widespread institution such as ius sanguinis is bound to pose serious practical challenges. However, if one has compelling moral reasons for dismantling such an institution, one ought to work towards this end. Babies are born into a physical world and from actual bodies but they are not naturally born into families and citizenship. The latter are social conventions that demand our acceptance when they are justified and our courage to change and replace them when they are not. To my critics worried that abolishing ius sanguinis amounts to throwing out the baby with the dirty bathwater I reply that we should not put the baby in the dirty bathwater in the first place.

The Return of Banishment: Do the New Denationalisation Policies Weaken Citizenship?


Kick-Off contribution


By Audrey Macklin (University of Toronto)


After decades in exile, banishment is back. Britain resuscitated the practice as part of its counter-terrorism strategy in the wake of the terrorist attacks 9/11 and 7/7 in New York, Washington and London, and Canada followed suit with the 2014 Strengthening Canadian Citizenship Act. As of late 2014, assorted legislators in Austria, Australia, Netherlands, and the United States expressed interest in enacting (or reviving) citizenship stripping laws.*

From antiquity to the late 20th century, denationalisation was a tool used by states to rid themselves of political dissidents, convicted criminals and ethnic, religious or racial minorities. The latest target of denationalisation is the convicted terrorist, or the suspected terrorist, or the potential terrorist, or maybe the associate of a terrorist. He is virtually always Muslim and male.

Citizenship-stripping is sometimes defended in the name of strengthening citizenship, but it does precisely the opposite. The defining feature of contemporary legal citizenship is that it is secure. Making legal citizenship contingent on performance demotes citizenship to another category of permanent residence. Citizenship revocation thus weakens citizenship itself. It is an illegitimate form of punishment and it serves no practical purpose.

Denationalisation refers to involuntary loss of citizenship. [1] Denaturalisation is a subset of denationalisation, and applies selectively to those not born into citizenship via ius soli or ius sanguinis. The most common basis for denaturalisation is fraud or misrepresentation in the acquisition of citizenship. The operative premise is that had the material facts been known at the relevant time, the state would not have conferred citizenship in the first place. Denaturalisation for fraud simply annuls the erroneously conferred citizenship and restores the status quo ante. [2]

My remarks focus exclusively on denationalisation for allegedly disloyal conduct by a citizen, while a citizen. In its present incarnation, citizenship revocation is best understood as a technique for extending the functionality of immigration law in counter-terrorism. Since 2001, states have turned to deportation to resolve threats to national security by displacing the embodied threat to the country of nationality. But deporting one’s own citizens is exile, and exile extinguishes a singular right of citizenship, namely the right to enter and to remain. Citizenship revocation circumvents that problem by introducing the two-step exile: first, strip citizenship; second, deport the newly minted alien.

The British Nationality Act authorises the Secretary of State for Home Affairs (Home Secretary) to deprive a person of British citizenship where she “is satisfied that deprivation is conducive to the public good.” That happens to be the same low and vague standard for depriving a person of permanent resident status (indefinite leave to remain), which provides one illustration of the downgrading of citizenship to permanent residence. In Canada, the executive power to revoke citizenship depends on a criminal conviction for a listed offence and a minimum sentence of either five years or life imprisonment. The offences include treason, spying, any terrorism offence defined under the Criminal Code and a variety of offences applicable to members of the military. In the case of terrorism offences, the conviction may be by a foreign court for an offence committed outside Canada, if it would also constitute a terrorism offence under Canadian law. [3] The UK law authorises citizenship stripping of naturalised citizens (but not birthright citizens) even if it renders them stateless, while the Canadian law prohibits the creation of statelessness but puts the onus on the individual to satisfy the Minister that statelessness would ensue from revocation. The UK declines to publicly disclose the exact number, identities or circumstances of those deprived of UK citizenship, but investigatory journalists estimate that at least 53 Britons have lost citizenship since 2002, over half on national security grounds. In 2013, the Home Secretary deprived 20 UK nationals of citizenship, more than all other years since 2002 combined. [4] The Canadian legislation has yet to be declared in force.

Citizenship revocation raises an array of practical, legal and normative questions: Does it advance a valid objective? Does it comply with domestic, constitutional and/or transnational law? Is it normatively defensible? The answers to these questions turn, in part, on one’s underlying conception of citizenship as legal status. Defenders of citizenship revocation liturgically intone that “citizenship is a privilege, not a right”. The rhetoric of citizenship-as-privilege trades on a popular and laudable sentiment that is sometimes expressed as follows: ‘I feel privileged to be a citizen of Canada, or the UK, or Italy, etc, and I consider it my duty to demonstrate my commitment through actively participating in civic life, or joining the armed forces, and standing up for my country as a good and loyal citizen should do.’ But a privilege in law is something different: A privilege emanates from the patron (here a government minister) and can be rescinded from an undeserving beneficiary (here the citizen) at the former’s discretion.

In two US Supreme Court cases in the 1950s, Chief Justice Warren rejected the classification of citizenship as privilege, proclaiming that “citizenship is not a licence that expires on misbehaviour”. Instead, he invoked Hannah Arendt’s famous depiction of citizenship as “no less than the right to have rights.” [5] Framing citizenship as a right vests citizenship in the rights-bearer. Depicting it as a meta-right dramatically increases the justificatory burden for any curtailment, because it places all rights in the balance.

Yet the force of Arendt’s ‘right to have rights’ aphorism may seem attenuated, at least with respect to liberal democratic states of the twenty first century. After all, permanent residents enjoy almost all the same rights as citizens, and even foreigners without status can, in principle, claim a long menu of basic human rights under international law and many domestic legal orders. But this rejoinder overlooks one crucial fact. The exercise of virtually all rights depends on territorial presence within the state, [6] and only citizens have an unqualified right to enter and remain on state territory. So once stripped of the right to enter and remain in the state, enforcement means that one is effectively deprived of all the other rights that depend (de jure or de facto) on territorial presence. This fact has not been lost on the present UK government: With two exceptions, all her targets were abroad when the Home Secretary chose to exercise her discretion to strip them of citizenship. This meant they were absent and unable to respond when the notice of intention to deprive was delivered, and therefore barred from entry qua alien in order to appeal the decision.

Another strand of citizenship discourse describes citizenship as a contract in which the citizen pledges allegiance to the sovereign in exchange for the sovereign’s protection. Acts of disloyalty amount to fundamental breach of contract, and so citizenship revocation simply actualises in law the citizen’s voluntary severance of the relationship. [7] This was, more or less, the logic of constructive expatriation under US law. But neither the rhetoric of contract nor privilege can mask the flagrantly punitive rationale for the citizenship revocation regimes currently in play in the UK and Canada: baldly stated, some citizens are very bad citizens, and therefore do not deserve to be citizens. The move from ‘bad citizen’ to ‘not citizen’ is explicit in the Canadian law, where conviction for a criminal offence is a condition precedent to revocation and eventual deportation. Citizenship revocation in the UK arguably turns on prevention of future risk rather than punishment for past wrong, but statements by UK politicians to the effect – ‘We think that deprivation is a way of expressing extreme displeasure at the way in which someone has behaved’ – reveal that the difference is more apparent than real. [8]

Banishment as criminal penalty has a long pedigree, and dates to a time before the rise of penal systems that enabled states to segregate, punish, rehabilitate and reintegrate wrongdoers within the state. In other words, modern states have criminal justice systems and an infrastructure that obviates the utility of banishment. These systems can, and are, deployed in response to the range of conduct encompassed under the rubric of terrorism. Banishment is both superfluous and anachronistic.

One might counter that offences threatening national security are qualitatively distinct from other offences. For these putative ‘crimes against citizenship’, incarceration is insufficient and withdrawal of citizenship is uniquely appropriate as supplement or substitute. It bears noting, however, that none of the offences precipitating loss of citizenship on grounds of national security – including treason – apply exclusively to citizens. Moreover, the idea that ‘national security’ misconduct is an affront to the state and so warrants a distinctive punishment fails to take proper account of the fact that all crime is regarded as an affront to the state’s maintenance of public order (the ‘King’s Peace’ in common law systems) and its monopoly on the legitimate use of violence. It is this public dimension of criminal law that differentiates it from private law, and confers on the state the authority to investigate, prosecute and punish wrongdoers, in addition to and apart from any private remedy that an individual victim might seek in tort, contract or property.

The purported symmetry between ‘crimes against citizenship’ and denationalisation echoes the defence of the sovereign’s other technique for permanent elimination of wrongdoers, namely the death penalty. Banishment fits the crime of disloyalty the way capital punishment fits the crime of murder. When tethered to expulsion, citizenship revocation effects a kind of ‘political death’. A citizen stripped of nationality and banished from the territory is, for all intents and purposes, dead to the state. Once outside the territory, the state has neither legal claim nor legal duty in respect of the former citizen, and is relieved of any obligation to object if another state tortures, renders or kills one of its nationals. [9] Indeed, denationalisation is not only a political analogue to death, it may also be a prelude to it. [10] At least two former UK citizens were executed by US drone strikes after the Home Secretary deprived them of citizenship, and another was rendered to the United States for trial on terrorism charges.

As with the death penalty, denationalisation extinguishes the prospect of rehabilitation or reintegration. The paradigmatic subject of citizenship revocation – the terrorist – is excluded from the ambit of human dignity that underwrites contemporary penal philosophy and affirms capacity for autonomy, rational self-reflection and reform. He is, in that sense, not fully human and thus incapable of rehabilitation. Banishment operates as pure and permanent retribution. There is no re-entry into the political community, no life after political death. Even creative and sophisticated attempts to classify and isolate those crimes that merit denationalisation from those that do not still founder on the instability of the distinction and the legitimacy of the punishment. [11]

One might object that that this parallel neglects the statelessness constraint. To the extent that a prerequisite of denationalisation is actual or potential possession of another citizenship, the individual has another political life to live somewhere else. This is also an answer to the complaint that stripping citizenship from dual nationals but not mono-nationals violates the principle of equality of citizenship. [12] The dual national is not similarly situated to the mono-national precisely because the former has another citizenship and the latter does not, so differential treatment does not constitute invidious discrimination. (Of course, the counter-intuitive consequence of this reasoning is that dual citizenship becomes a liability. Multiple citizenship becomes less than the sum of its parts: the mono-citizen is secure from revocation, while the dual or multiple citizen is not).

The cogency of this argument depends on how one characterises the impact of citizenship revocation. From an external, statist perspective, the function of nationality is to catalogue the world’s population and to file each person under at least one state. Nationality provides states with a return address they can stick on non-citizens for purposes of deportation, and is one reason why statelessness is an inconvenient anomaly for states. And just as all sovereign states are formally equal under international law, so too are all citizenships. Within this framework, citizenship becomes fungible. Statelessness is the problem, and nationality the solution. So, it may not actually matter what nationality a person possesses – Canadian or Somali, Brazilian or North Korean – as long as he or she possesses at least one. All nationalities are equal for purposes of averting statelessness. [13] This formal equality of nationality may partly explain international law’s diffidence, or at least ambiguity, on whether citizenship deprivation that does not induce statelessness may nevertheless be arbitrary and contrary to international law. [14] In any event, as long as an individual retains a nationality somewhere, denationalisation poses no human rights problem.

From an internal, individual perspective, however, citizenship is not fungible. [15] The revocation of citizenship severs a unique relationship between the individual and a specific state. It is unique in two respects: First, the formal equality of nationality suppresses the substantive inequality of citizenship. The bundle of social, political, economic, cultural and legal opportunities and entitlements to which citizenship provides access varies radically between countries. Canadian or Brazilian citizenship is dramatically and indisputably heftier than that of present-day North Korea or Somalia.

Secondly, the subjective experience of that legal bond, what the International Court of Justice in Nottebohm v. Guatemala calls ‘the social fact of attachment’ [16] is as infinitely diverse as the people who make up the citizenry. It may range from the ‘nominal citizen’ whose social attachment is highly attenuated, to the individual whose existence is, and has always been, wholly and exclusively embedded in the country of residence. Citizenships are not substantively equal in comparison to one another and the nature of the individual citizen-state relationship is not invariant. But my point is not to propose a metric capable of measuring the quantitative, qualitative, experiential, emotional, personal, familial, cultural, social, financial, linguistic and political impacts of exile on any individual, in order that some state official could determine precisely when citizenship revocation inflicts an appropriate versus excessive degree of punishment. Citizenship as legal status obviates both the need and the legitimacy of an ongoing or comparative evaluation by state authorities of how much or how well a citizen performs as a citizen. [17] The very act of subjecting a subsisting citizenship to this kind of normative scrutiny subverts the security that distinguishes legal citizenship from other statuses that define the relationship between state and individual.

The history of banishment generates only cautionary tales about the inevitably arbitrary and prejudicial abuse of a discretionary power to identify the ‘bad’ citizen for purposes of relegating him or her to the non-status of non-citizen. The violence of rupturing the link between citizens and state is not negated by possession of citizenship status in another polity, if one conceives of the relationship (whatever its intensity, depth, etc.) between a state and a citizen as singular and unique. On this view, citizenship revocation inflicts an intrinsically grave harm that is separate from (though exacerbated by) the harm of statelessness. [18]

I leave to one side an account of the myriad procedural and substantive deficiencies of the UK and Canadian denationalisation regimes that make them ripe for legal challenge. Nor do I dwell here on the dubious practical value of denationalisation in preventing terrorism or protecting national security. Suffice to say that if the aim of citizenship revocation is deterrence, there is no evidence that stripping citizenship will deter a potential terrorist any more or better than the prospect of a criminal conviction and lengthy imprisonment or, for that matter, the risk of blowing oneself up, getting killed or executed, or being detained indefinitely, rendered, or tortured. To the extent that exile supposedly makes a country more secure by removing dangerous people, the justification knows no limits: it is not obvious why Canadians or Britons would not also be made safer by exiling all citizens who commit violent offences. From the other side, expelling convicted or alleged terrorists is an oddly parochial response that transfers rather than reduces risk. Depending on the destination country, deportation may actually make it easier for the individual to engage in activities that pose a threat to global security. [19]

And, finally, the sheer absurdity of banishment as a response to the terrorist qua global outlaw is best illustrated by speculating on what would happen if all states behaved like the UK and Canada: Imagine a dual UK-Canada citizen who is convicted of a terrorism offence in the UK. Since terrorism is a global menace, Canada can treat a terrorism conviction in another state as proof of being a bad Canadian citizen. Both Canada and the UK can lawfully denationalise him. But both states are also somewhat constrained in law not to create statelessness, and both are constrained in fact by the need to find another state to take the expelled person. And the only country that has a legal obligation to do is a state of nationality. So, now it becomes a race to see which country can strip citizenship first. To the loser goes the citizen.

Modern exile, as imagined under UK and Canadian law, is erected upon unsustainable and incoherent propositions about the nature of legal citizenship. If citizenship is irrevocable only where withdrawal causes statelessness, then citizenship is a right for mono-citizens but a privilege for dual or multiple citizens. Legal citizenship can be contingent on normative criteria for one state if and only if it is not similarly contingent for another state. State A can deprive a national of citizenship and banish him because he is a bad citizen. But State A can do so lawfully if and only if State B is compelled to admit the individual simply because he is a citizen of State B, irrespective of whether he is a good or bad citizen of State B. One state’s authority to deem the bad citizen a non-citizen presupposes another state lacking that same authority.

To contend that punitive denationalisation in the twenty-first century is an illegitimate and futile exercise of sovereign power does not refute or deny that social solidarity, belonging and allegiance have a place in conceptions of citizenship and deserve to be promoted. It is rather that these goals will not and cannot be advanced by citizenship revocation. Nor will citizenship revocation make any state, or the global community, more secure. Citizenship revocation only enhances the discretionary and arbitrary power of the executive, at the expense of all citizens, and of citizenship itself. Banishment deserves to be banished again. Permanently.



* For a more elaborate comparative analysis of recent legislative developments in the United Kingdom, Canada and the US, see Audrey Macklin, Citizenship Revocation and the Privilege to Have Rights, EUI/RSCAS Working Paper, forthcoming

[1] Before the widespread acceptance of dual citizenship, acquisition of a second citizenship or marriage to a foreign man commonly triggered denaturalisation. In a world where states tolerated only one legal bond between individual and state at a time, acquisition of a second nationality denoted a transfer of membership from one state to another.

[2] The United States law combines renunciation of citizenship and denationalisation for birthright citizens into a category labelled expatriation. The US Constitution guarantees the citizenship of ius soli citizens as a constitutional right. The doctrine of expatriation operated on the legal fiction that certain acts by a citizen denoted an intention to renounce citizenship. In a series of judgments culminating in 1967 in Afroyim v. Rusk, the US Supreme Court progressively restricted the government’s ability to deem conduct short of explicit renunciation as conclusive proof of an intention to expatriate, and the executive effectively abandoned attempts to pursue constructive expatriation in the 1980s.

[3] The law also permits revocation of a citizen who ‘served as a member of an armed force of a country or as a member of an organised armed group and that country or group was engaged in an armed conflict with Canada.’ This is not a criminal offence, though it is almost identical to the existing offence of treason, except that it includes non-state armed groups, whereas the offence of treason only includes armed forces of a state.

[4] Id.

[5] The unattributed quote comes from Hannah Arendt, The Origins of Totalitarianism, (New York: Harcourt & Brace, 1951), at 294. It was picked up by US Supreme Court Justice Warren in Perez v. Brownell, 356 US 54 (1958) at 64 and again in Trop v. Dulles, 356 U.S. 86 (1958) at 102. See discussion in Patrick Weil, The Sovereign Citizen (Philadelphia: University of Pennsylvania Press, 2013).

[6] Expatriate voting is one exception. Many people suppose that diplomatic or consular assistance is also a right available outside the territory of the state, except that states tend to deny that they owe a legal duty to extend assistance to their citizens abroad. See, e.g. R (Abbasi) v Foreign Secretary [2002] EWCA Civ 1598.

[7] The US model of expatriation implicitly relied on this metaphor to characterise a series of acts, from desertion, to voting in a foreign election, as acts signifying an intention to renounce citizenship.

[8] See See United Kingdom, Parliamentary Debates, HC Standing Committee E, 30 April 2002, col 54 (Angela Eagle), quoted in Thwaites, supra * at note 94.

[9] Since the United States’ lethal drone strike on US citizen Anwar al Awlaki (and his son), the United States’ position is that it may lawfully execute its own citizens without trial when they are abroad. This, of course, obviates the necessity to strip citizenship prior to execution. See “US cited controversial law in decision to kill American citizen by drone”, The Guardian, 23 June 2014; online at http://www.theguardian.com/world/2014/jun/23/us-justification-drone-killing-american-citizen-awlaki . See also Peter Spiro, “Expatriating Terrorists”, (2014) 82 Fordham Law Review.

[10] This was the case with the Nazi extermination of German Jewry, as Hannah Arendt recounted. First, the Nazi government stripped Jews of German nationality and then, when no country would take them in, proceeded to murder them.

[11] For a recent example, see Shai Lavi, “Citizenship Revocation as Punishment: On the Modern Duties of Citizens and Their Criminal Breach”, (2011), 61 UTLJ 783-810, at 806.

[12] It does not, of course, answer the charge of discrimination against naturalised mono-citizens under UK law. They are exposed to the risk of statelessness whereas birthright citizens are not.

[13] One could even imagine how a creative government wedded to this view might venture that protecting mono-citizens from statelessness is really an affirmative action initiative under s. 15(2) of the Charter.

[14] See Peter Spiro, “The New International Law of Citizenship”, (2011), Am J. Int’l Law, 694-746, at 711-12.

[15] Thwaites makes a similar argument, supra note * at 263.

[16] Nottebohm (Liechtenstein v. Guatemala), ICJ 4 (1955) at 23.

[17] This does not preclude an argument that the depth and duration of a resident non-citizen’s relationship to a state could and should generate an entitlement to remain and to be put on a path to citizenship. See, e.g. Joseph Carens, The Ethics of Immigration (Oxford: OUP 2013).

[18] For a similar argument, see Rayner Thwaites, supra note *

[19] Audrey Macklin, “Borderline Security”, in R.Daniels et al. (eds.), The Security of Freedom: Essays on Canada’s Anti-Terrorism Bill (Toronto: U of T Press, 2001), 383-405; “Still Stuck at the Border”, in Craig Forcese and François Crépeau, eds., Terrorism, Law and Democracy: 10 Years After 9/11 (Montreal: Canadian Institute for the Administration of Justice, 2012), 261-306.



Terrorist expatriation: All show, No bite, no future

By Peter Spiro (Temple University)


I agree with the bottom line of Audrey Macklin’s excellent kick-off for the forum. New expatriation measures adopted by the United Kingdom and Canada are ill-advised and possibly unlawful. The UK and Canada moves make for a kind of trendlet, and other states (even human rights-pure Norway) are considering similar measures as the “foreign fighter” phenomenon captures global attention. Denationalization of terror suspects clearly merits the attention of scholars and activists; after decades of disuse, states are now stepping back into the practice of forced expatriation. Macklin sets the scene with a primer on recent developments and a powerful critique of the UK and Canadian measures.

But I would get to the destination along another path. I see denationalisation as anachronistic and toothless in the face of diminished conceptions of citizenship as an institution and the changed locations of allegiance. The expatriation measures are empty gestures, a kind of counter-terror bravado to make up for the deficiency of more important material responses. Government officials must be seen to be doing something, and so they may (for appearances sake) throw expatriation into the counter-terror toolbox. But expatriation won’t advance the counter-terror agenda in any real way. Given the lack of policy advantage, I expect that the human rights critique will suffice to suppress the broad use of denationalisation in this context. 

In theory, expatriation could help shore up the boundaries of membership and national solidarity. Terrorist expatriation might be consistent with the historical practice of terminating nationality upon formal transfer of allegiance. This was once the near-universal practice; original nationality was lost automatically upon naturalisation in another state. Military and government service in another country would also typically result in expatriation, even when the other state was a friendly one. This practice helped police the boundaries of community. One could be a member of one or another polity, but not both. States that continue to prohibit dual citizenship still operate on this principle. A Japanese citizen who naturalises as an American, for example, automatically forfeits her Japanese nationality.  

One might situate security-related expatriation in this tradition. To the extent that fighting for the Islamic State represents a shift of loyalty incompatible with loyalty to the United Kingdom, expatriation merely reflects social conditions on the ground. Membership in the United Kingdom would be exclusive of membership in forces associated with the Islamic State. Expatriation clarifies the “us” and “them” in a way that clarifies social solidarities and the special obligations that come with co-nationality. (Ayelet Shachar makes a similar argument with respect to “hollow” citizens acquiring citizenship on the attenuated basis of descent.)

But this logic doesn’t map out onto denationalisation in the current security context. There is no citizenship in the Islamic State (ISIL not being a state, the label notwithstanding). One cannot naturalise or be born into ISIL; there is no formal evidence of loyalty or membership. Expatriation doesn’t work without the symmetry. To the extent that only dual nationals are subject to security-related expatriation, the criterion no longer makes any sense: the other citizenship is random, unrelated to the motivation for expatriation. (As Macklin points out, it could lead to a strange dynamic in which states allied against groups such as ISIL could race to expatriate foreign fighters in an effort to offshore putative threats.) The condition then arbitrarily discriminates against individuals on the basis of their dual-citizen status.  

That takes care of the only normatively tenable rationale for the expatriation measures. The punitive basis is more easily dispatched. Punitive uses of expatriation have long been condemned. As early as 1958, the U.S. Supreme Court was able to observe that “[t]he civilized nations of the world are in virtual unanimity that statelessness is not to be imposed as punishment for crime.” The Canadian measure marks a return to the practice of exile. As Macklin argues, non-application to cases in which statelessness would result does not save it from this rap. A person may well feel a deep social attachment to one country while holding alternative nationalities (which themselves may be nominal). The denationalisation of a Canadian citizen long-resident in Canada will feel like banishment even as he holds another nationality, especially to the extent the latter is attenuated. 

Finally, the protective rationale for terrorist expatriation makes little sense as a practical matter.  The “foreign fighter” problem is largely framed as a problem of return. Citizens radicalised by their experience in Iraq and Syria with brutal ISIL forces will return to their home countries in the West to undertake terror attacks. It’s a potent narrative of weaponised citizens. Without citizenship, these individuals would have no right of re-entry, thus defusing their utility as ISIL operatives. 

Or so our politicians would have it. In practice, denationalisation adds little counter-terror value. You can’t take away someone’s citizenship for being associated with ISIL before you know that he’s associated with ISIL. But once the security apparatus is aware of the connection, it will have other, standard counter-terror tools to protect against the threat. There will be the possibility of criminal prosecution in many states on material support charges, with incarceration on conviction. (Canada’s punitive scheme can hardly sustain even the pretense of a protective rationale.) Short of prosecution, watch lists and well-practiced surveillance techniques should prevent returning foreign fighters from undertaking terror attacks. Passport revocation and travel bans will help prevent citizens from becoming foreign fighters in the first place. 

So terrorist expatriation advances counter-terror efforts not at all. It supplies yet another example of security-related theater, a feel-good move that will be popular with some voters. (The features are shared with some Western responses to the vastly exaggerated Ebola threat, where politicians must be seen to respond dramatically even if dramatic moves make no sense in policy terms.) Terrorist expatriation is unlikely to have staying power against a powerful human rights critique. The UK and Canadian measures may well fall to legal challenges, domestic or international. Even if they are sustained in court, they are unlikely to be put to broad use. Few other states will follow suit (it is interesting that terrorist expatriation has almost no political traction in the United States, its aggressive counter-terror posturing notwithstanding). The failure will evidence an emerging norm against involuntary expatriation. If states can’t make expatriation stick here, they won’t be able to make it stick anywhere. 




Should those who attack the nation have an absolute right to remain its citizens?

By Peter H. Schuck (Yale Law School)


Audrey Macklin’s call for the banishment of banishment is eloquent and persuasive on many points.  She is surely right that particular denationalisation regimes may suffer from a variety of fatal defects. The standards for revocation may be too vague to constrain official discretion or to provide adequate notice to the citizen concerning what conduct will risk revocation. Most important, the grounds for revocation must be limited to only the most extreme, unmitigated attacks on the nation’s security, attacks that are consistent only with a desire to bring the nation to ruin. This conduct must be scrupulously-defined and highly specific conduct; mere malignant thoughts will not suffice. Revocation cannot be permitted to lead to statelessness and thus a loss of the “meta-right” (as Macklin puts it) to have rights, especially the right to the territorial presence that in turn confers a broad panoply of liberal rights.  The procedures for revocation must be robust in all respects, including of course the right to be actually or virtually present rather than having to contest the government’s action from exile. The government’s burden and standard of proof must be exceedingly demanding, perhaps even the proof beyond a reasonable doubt required for criminal convictions.  

But even these extraordinarily demanding and rare preconditions are irrelevant to Macklin; she is utterly categorical in her rejection of the very notion of denationalisation.  She would preclude denationalisation even if these (and other) strict conditions were met; indeed, no protections for the individual citizen – or for the threatened nation – would suffice.  Here is where we disagree.  I see no reason in logic or justice why a state should be powerless to protect itself and its people from imminent, existential threats (suitably defined) from an individual who has launched a dangerous attack (suitably defined and rigorously proved) on itself and its people.  And I see no reason in logic or justice why that state cannot defend itself and its people against such an attack by, among other things, severing the attacker’s connection to a state with which he is manifestly at war, thereby making it much more difficult for him to succeed in that war. Should the individual’s interest in maintaining that connection, which (by my definition, embedded in the preconditions listed above) can only be tactical and cynical, utterly and categorically outweigh the nation’s interest in protecting those for whom it bears a sacred trust?  This question, I submit, answers itself – and the answer is grounded not merely in a utilitarian balancing but in a deontological principle: the nation’s fundamental duty to protect its people.

I also have some reservations about a few of Macklin’s other, less fundamental arguments. First, she claims that denationalisation weakens citizenship by eliminating its security and thus rendering it a form of mere legal residence. I don’t understand her logic.  Am I less secure in my citizenship if I know that the state may execute me or imprison me for life if I murder a fellow citizen?  I suppose that I am less secure, but that insecurity is warranted and I can easily avoid it. Moreover, there is a sense in which denationalising one who has demonstrably satisfied the exceedingly demanding conditions for revocation that I have specified does, contrary to Macklin’s claim, strengthen citizenship by reaffirming the conditions on which it is based.  

Second, she categorically condemns revocation in part because it categorically denies the individual the opportunity to rehabilitate himself.  We should and ordinarily do protect a wrongdoer’s opportunity to rehabilitate himself, but there are many situations in which we don’t.  An employer who catches an employee embezzling from the company may fire him without giving him an opportunity to rehabilitate himself there; if he wishes to rehabilitate himself, he will have to do so elsewhere, on his own time.  When we sentence a murderer to life imprisonment without parole, we are denying him the right to regain his freedom through rehabilitation.  

Third, it is true that denationalising a dual citizen would still leave him with a state while denationalising a mono citizen would not.  But so long as we do not allow revocations that would render one stateless, this particular inequality between categories of citizens is hardly one that should trouble us – any more than we should be troubled that a dual citizen has an additional passport and can vote in an additional polity.   

Finally, Macklin states that there is no evidence that denationalisation will deter a would-be terrorist if other, more conventional counter-terrorism measures fail to do so. I agree, but so what?  Deterrence may be an important reason to punish wrongdoers but it is by no means the only reason to do so.  If we are justified in punishing them, that justification is not nullified by a claim that the punishment will not deter others. And if more conventional measures are indeed effective in eliminating threats, they should of course be our first and perhaps final resort.  In such situations, denationalisation may well be a superfluous, unnecessary remedy. But this is a question of policy and prudence, not moral principle.          

Macklin is certainly right to worry about the possible abuses of denationalisation. The history of political banishment is hardly reassuring on this point. But a liberal constitutional regime can control such abuses by scrupulously controlling the state’s exercise of this power through a variety of familiar institutions and practices. These include a careful definition and exacting limitation of the grounds for revocation; demanding procedural and evidentiary requirements before such a power can be exercised; and an independent judiciary accustomed to challenging state power in the name of protecting individual rights. We have entrusted our precious liberties to the faithful working through of these institutions and practices. Some of these liberties are even more precious than our right to retain our citizenship when we have knowingly acted in horrendous ways that make it justifiable, under the safeguards I have described, for the state to declare that status forfeited. 


Terrorists repudiate their own citizenship

By Christian Joppke (University of Bern)


The recent trend to strip international terrorists of their citizenship raises general questions about the changing nature of terror and of citizenship. Let us start with “terror”. In the era of Marxist-inspired violence against the state (or rather “capitalism”, of which the state was suspected to be merely a servant), terror was a purely domestic affair, committed by the flower children of the elite, particularly its most educated and morally minded. No one would have fathomed stripping an Ulrike Meinhof or Andreas Baader, leaders of the 1970s' German Red Army Faction (RAF), of their German citizenship. The current “return of banishment” is a response to an altogether different type of terror, one that transcends borders and is committed by people who explicitly posit themselves outside the political community of the nation-state—allegiance to the community of believers (ummah) cancels out the secular community of citizens, it is even deliberately mobilised against the latter. Only notice the cynical ritual of the Islamic State`s henchmen to have a fellow-national do the mediatised head-chopping. By the same token, RAF limited its murderous acts to high-ranking representatives of the “system” (of which ordinary citizens were seen as merely victims who thus stood to be recruited as fellow-fighters). Al Quaeda and its Islamic State sequel seek death for ordinary citizens, whose humanity is denied through being demoted to “unbelievers”. Paul Kahn (2011: 138) took the ubiquitous threat of terror to be today`s ultimate moment of citizenship, the “moment of conscription”. Indeed, Islamic terror is meant to be “war”, while RAF aspired to “revolution”—two very different things, with obvious implications for citizenship in the former but not the latter. That terror against citizens should lead to reconsidering the citizen status of its culprits, who proved the ties to their state of citizenship to be at best “tactical and cynical”, as Peter Schuck writes in his contribution, seems logical. One is therefore astounded about the measured response by Western states, which have mostly respected the international norm of avoiding statelessness (only lately, in response to the unspeakable atrocities committed by the fighters of the Islamic State, have there been cracks in this commitment, most notably in Britain). But academics cry out that “banishment weakens citizenship”, as Audrey Macklin does. They draw an idyllic and reality-resistant picture of “singular and unique” ties between terrorists and the citizenship they despise; “intrinsically grave harm” is said to be inflicted here, separate even from “the harm of statelessness” (ibid.). Evidently, more sympathy is invested on the culprits than on their victims.

Make no mistake. One should hold no illusion about populist, spin-doctored politicians, from Britain to America, Norway to Italy, who hide their chronic incapacity to lead in our contemporary “audience democracies” (Manin 1997: ch.6) behind the sable-rattling “security” and “War on Terror” rhetoric that the people wish to hear. Macklin has a point when she finds that under the guise of “security” only “the discretionary and arbitrary power of the executive” is increased. Particularly the recent experience in Britain under Tory Home Minister Theresa May, with a rather capricious practice of citizenship stripping for the loosely defined reason of being “conducive to the public good”, with sometimes lethal and conspiratorially concocted consequences for the targeted individuals, lends itself to this interpretation. And Peter Spiro is on target that conducting the fight against terrorism on the citizenship front is “empty gestures” and not likely to have much effect—though his proposal of “passport revocation and travel bans” in lieu of denationalisation reads eerily off the mark after the recent tragedy of a would-be jihadist, who had been grounded by the Canadian government exactly in these terms, turning his rage about the passport denial against an innocent guardsman in Ottawa.

The practical question of effectiveness is secondary to the principled question whether citizenship for proven (naturally not just suspected or potential) terrorists who conduct war (in the literal sense) against Western states and their citizens should be unassailable. At heart, the issue is one of “loyalty and allegiance”, as the Canadian Immigration Minister, Chris Alexander, defended the 2014 Strengthening Canadian Citizenship Act in parliament. This act, representative of similar bills currently being considered in a number of European states, Australia, and the United States, allows the stripping of citizenship in the cases of treason, spying, taking up arms against the Canadian Forces, and terrorism, even if the latter is committed outside Canada and sentenced by foreign courts, should the action in question constitute a terrorism offence also under Canadian law. The expanded geographic scope for terrorism, which stirred controversy, was clearly dictated by heightened security concerns. But it also recognises the global nature of the new terror and its affront to the secular state and citizenship at large, wherever it may occur; one might read it as a comity of nations response to a global challenge. In any case, it is not just bizarre but self-destructive to measure the “strength” of citizenship in terrorists' unencumbered possibility to make tactical use of it in their war against the godless state and its unbelieving median citizen.

For calibrating banishment, next to taking into account the changing nature of terror, one also needs to recognise the changing nature of citizenship in a globalizing world. Whoever has reflected for a second on the colossal injustice inflicted on the vast majority of mankind by being born into the “wrong” kind of state that cannot guarantee its “citizens” physical safety and the elementary means of survival (see Shachar 2009), must be irritated to see citizenship depicted as something that an individual should never be able to lose, however randomly it had been assigned to her in the first place, and however much a particular individual has done to undermine or even destroy this very citizenship (and the state that guarantees it). Audrey Macklin sees the danger of banishment in “making legal citizenship contingent on performance”. “Performance” strikes me as a rather vague and anodyne term for the behavior in question. It is one thing to make citizenship acquisition contingent on virtuous behaviour, which could never be exacted on born citizens (as Britain entertained for a while in its “probationary” or “earned” citizenship scheme that was never implemented); it is quite another to make a declared war against the secular state and its citizens a ground for renunciation. As much as one should eschew virtuous citizenship from a liberal perspective, one should welcome, even require the withdrawing of citizenship from someone to whom it is at best a tactical weapon.

It may warm the heart to elevate citizenship to a “right to have rights”, enunciated by US Supreme Court Chief Justice Earl Warren in a different time and context (voting in foreign elections (1) and desertion during World II (2), in both cases without any third-party harm inflicted and at best a vague and constructed violation of allegiance). The gospel of citizenship stripping as “cruel punishment”, pronounced in Trop v. Dulles (1958), needs reconsideration in the age of global terror. And the accompanying formula of citizenship as a “right to have rights” obscures that persons without states or citizenship are no longer the “scum of the earth” they may have been in the late 1940s, when Hannah Arendt wrote the Origins of Totalitarianism. But most importantly, the formula “rights to have rights” dodges the fact that, indeed, citizenship in a globalising world is increasingly “privilege” and “contract”. It is a privilege if one considers the mentioned exclusion from a lucrative OECD-state citizenship of most of mankind (that has to make do with less than US$ 2 per day). And it is a contract by definition for the ever growing number of immigrants who are not born with it but seek it out for their own benefit. In the post-feudal world, most states allow the possibility to renounce one`s citizenship—this was the point of departure of “democratic” America from “monarchical” Britain. But then it is not outlandish (or illiberal) to concede the converse capacity to states to rid themselves even of born citizens who have despised or patently abused their citizenship through their actions (and why stop at the threshold of statelessness?).

Macklin claims that banishment is “both superfluous and anachronistic” because states now have “criminal justice systems” at their disposal to “rehabilitate and reintegrate wrongdoers within the state”. This claim is misleading and paternalistic. International terrorists are not criminals but warriors—they don`t want to be “reintegrated”. The liberal state should acknowledge their claim, eye to eye, by taking away from them what they have factually renounced and even wish to destroy. Canadian minister Chris Alexander is right: “They (terrorists) will have, in effect, withdrawn their allegiance to Canada by their very actions.” Peter Spiro lawyerly ups the ante by arguing that there could not be a “shift of loyalty” on the part of Islamic terrorists because “there is no citizenship in the Islamic State”. Does he want to wait until they have acquired a seat in the United Nations?



Kahn, Paul. 2011. Political Theology. Ithaca, N.Y.: Cornell University Press.

Manin, Bernard. 1997. The Principles of Representative Government. New York: Cambridge University Press

Shachar, Ayelet. 2009. The Birthright Lottery. Cambridge, Mass.: Harvard University Press. 



(1) Perez v. Brownell, 356 U.S. 54 (1958)

(2) Trop v. Dulles, 356 U.S. 86 (1958)



It’s not about their citizenship, it’s about ours

By Vesco Paskalev (University of Hull)


The very passion and fury pouring from Christian Joppke’s contribution should prompt both the lawyer and the political philosopher that he is wrong. I too am outraged by what ISIS fighters are doing, but it is well known that the function of constitutional rights, and of the constitutions themselves, is precisely to assure that the legislator is not driven by the passion of the day. One decade after 9/11 we know that the actions taken both by the President and the Congress of the U.S. based on the rationale that it is a new world that we have woken up into were not all reasonable, to put it mildly. So may be today’s rush to strip terrorist suspects of their citizenship. When watching the daily news on TV, one is easily tempted to think that we are living in extraordinarily dangerous times, which warrant a return to what the US Supreme Court considered to be ‘cruel punishment’ half a century ago. Yet as a matter of statistics, and despite our contrary impressions, violence of all kinds in the world is actually declining (Pinker 2011). On the other hand, the capacity of law enforcement agencies for surveillance and control, especially in the OECD countries, have increased dramatically, so the return to practices which have long been abandoned is difficult to justify. This is not to say that that citizenship is a sacred cow and any return to abandoned practices is excluded by some historic laws of human progress. Nothing can be further from the truth. But it does follow that the proponents of banishment must provide a more subtle justification than we have seen so far.

Joppke has a point when distinguishing the old school revolutionaries from the contemporary jihadists, who conceive of themselves as members of the global ummah, and not of any state. (Do we know that for sure? ISIS aims to create an Islamic state after all). He also has a point that waging war against a country is a good reason to strip the warrior of the citizenship of that country. I can accept even stretching this argument to apply to all those who take up arms against any allies of that country, or even to those who have taken arms against the international system of states. This would bring me already quite close to the position of the ‘deprivationists’.

What I find difficult to accept is the unquestioned assumption that this gesture would serve any of the goals Joppke, and the politicians favouring banishment, may have. If the jihadists were as cosmopolitan as he takes them to be, deprivation would not have any meaning, neither for the actual fighters, nor for any like-minded followers. It might be the case that taking their passport will have the practical effect of preventing them from travelling to Syria or back, but as a person who is genuinely outraged by their deeds, I would rather see them locked up in prison rather than left at large in a legal limbo in the Middle East out of all places. For Joppke the practical side is only of secondary concern, but I am afraid his theoretical argument is self-defeating.

Now, if we accept that the jihadists just do not care if they are deprived of their western citizenships, let us consider whether this would still matter for anyone else. On the one hand, there are the ’normal’ citizens of the same country who may wish to see the extremists publicly excommunicated. This is a legitimate concern. However, it is in no way different from the desire of many law-abiding citizens to see murderers and rapists sent to the electric chair. So the usual objections to the latter punishments apply here too. More importantly, while there is some commensurability between a murder and a death sentence, the very gravity of the offences of the jihadists make citizenship deprivation superfluous. Ironically, not the cruelty of citizenship deprivation, but its softness make it appear quite inappropriate for the case of terrorists. If we take into account also the practical difficulties arising in the prosecution of a foreigner, on balance it might be better to keep him as a citizen. On the other hand, the possibility or impossibility of revocation defines and redefines the meaning of the concept of citizenship itself – of our citizenship, not of theirs. That is why many academics, whose professional duty is to care for precisely such nuance, are so uneasy about the recent trend.  I would be glad if this concern remains confined to the ivory towers of the academia, but I suspect that the conditionality of citizenship is more than a theoretical concern for those citizens who are not white, Anglo-Saxon and Christian and have only recently arrived from the wrong side of the OECD border. 

One may argue, as Peter Schuck does, for deprivation administered under narrowly circumscribed conditions. Indeed, due process can alleviate some of the anxieties the conditionality of citizenship would create, but he does not provide much of a justification for this conditionality in the first place. He also relies on the intuitive, yet questionable assumption that citizenship deprivation serves to protect the state and its people. But all grounds for deprivation he suggests already constitute a serious crime, and if the perpetrator must be convicted to be denationalised as he suggests, then again, what difference would it make if he is a citizen or not? If deprivation were administered properly – for grave crimes and with due process, it becomes redundant.  

Beyond these conceptual concerns, and paying due consideration to the all too present terrorist threats, I want the Islamic State bombed out of existence, and I want all jihadists punished for what they do. But as a citizen I also want my tax money spent on police to put the bad apples in jail, not on border patrols to keep them out. 



Pinker, S. (2011). The Better Angels of our Nature. New York: Viking.



You can't lose what you haven't got: citizenship acquisition and loss in Africa

By Bronwen Manby


The heading for this discussion makes a person focused on sub-Saharan Africa scratch her head somewhat. Which ‘new’ denationalisation policies are we talking about? In Africa, we have continued to see the same old denationalisation policies that have been in place since the 1960s. The context of national security has changed in some countries, especially the threat of 21st century terrorism methods in places such as Kenya or Nigeria, but the methods used by the governments in response have not changed.

The legal provisions

If we start from a survey of the laws, most African countries allow for deprivation of nationality acquired by naturalisation, some of them on quite vague and arbitrary grounds. The former British colonies borrow language from the British precedents and provide for deprivation on the grounds of “disloyalty” or the “public good”; while the francophone countries talk about behaviour “incompatible with the status of a national” or “prejudicial to the interests of the country”. However, more than half of Africa’s 54 states forbid deprivation of nationality from a national from birth (of origin, in the civil law terminology), whether or not the person would become stateless.[1] And although a large number of the remaining countries have a provision framed along the lines provided in the 1961 Convention on the Reduction of Statelessness for a person who works for a foreign state in defiance of an express prohibition to lose their nationality,[2] only a small handful provide for deprivation of a birthright citizen in case of a crime against the state — Egypt, Eritrea and Mali.[3] None of the sub-Saharan countries come close to the extremes of Egypt, where citizenship can be deprived from anyone (citizen from birth or by naturalisation) if, among other things, “at any time he has been qualified as Zionist”.[4] 

On the positive side, the South African and Ethiopian constitutions provide blanket prohibitions on deprivation of nationality, whether from birth or naturalised (though South Africa then goes on to violate this prohibition in its legislation).[5] Several constitutions and laws create serious due process hurdles for governments seeking to revoke citizenship. In Kenya for example, the 2010 constitution requires a naturalised citizen (citizenship by birth cannot be revoked) to have been actually convicted of a serious crime, including treason;[6] less specifically, Burundi, Malawi, and Rwanda have constitutional provisions forbidding arbitrary deprivation of nationality.[7] Meanwhile, Gambia, Ghana, Liberia and Rwanda all provide that deprivation can only be done by a court, on the government’s application;[8] and a majority, though not all, others provide for judicial review of administrative decisions to deprive.[9] A few countries provide for protection against statelessness in deprivation cases: just Lesotho, Mauritius, and Zimbabwe (since 2013) provide in principle for protection from statelessness in all cases where nationality is revoked by act of the government; and Namibia, Rwanda, Senegal and South Africa provide partial protection, allowing statelessness to result in some circumstances.[10]

On the negative side, Botswana, Lesotho, Malawi, Mauritius, Seychelles, Tanzania, Zambia and Zimbabwe — notably, all with a British legal inheritance — explicitly state in their legislation that the decision of the minister on any matter under the nationality law cannot be reviewed in court.[11] These are all countries which do not allow for deprivation of birthright citizenship (though some provide for loss in case of acquisition of another nationality); but it’s questionable what the protection against statelessness in deprivation cases provided by Mauritius means, if the decision of the minister cannot be challenged. In Swaziland, where a certificate of nationality “shall” be issued by the minister to a person who is qualified to be a citizen, it is also provided that the minister “may revoke” a certificate and no grounds are specified.[12] Namibia allows deprivation of nationality on the grounds that a person was already deprived in another country, increasing the likelihood of rendering them stateless.[13] In 2013, the Seychelles inserted a new article to its citizenship law expanding the grounds for deprivation of citizenship if the minister “is satisfied” that the person has been involved in terrorism, piracy, drugs offences, treason, and other offences, or has acted with disloyalty.[14] In 2010, the South African Citizenship Act was amended, providing for automatic loss of citizenship by a naturalised citizen “if he or she engages, under the flag of another country, in a war that the Republic does not support”, leaving lawyers wondering how you would know whether or not the Republic “supported” a particular war (and would it matter which side the person was on?).[15]

The practice

But this review of deprivation provisions has a slightly unreal feel. These procedures are hardly used, so far as one can tell. Only South Africa publishes any statistics — or at least it used to do so — revealing that at least 17 people have been deprived of citizenship since 2001-02 (despite the constitutional ban on deprivation), though no details are given.[16] Countries such as Kenya and Nigeria, both facing well-publicised and serious security threats from the Al-Qaeda-affiliated Al-Shabaab and Boko Haram are not known to have deprived any individual of citizenship through the formal procedures of the law on deprivation.[17]

The legal provisions on deprivation of citizenship are, in fact, more or less irrelevant in countries where (a) as described above, citizens from birth cannot be deprived of citizenship under law except in the rather rare circumstance of working for another state despite a formal request not to do so; (b) naturalisation is very difficult to obtain; and (c) the government has easily accessible other means of achieving the same result in relation to (people who believed they were) birthright citizens, obviating any need to amend the law on withdrawal of nationality.

As regards (b), statistics on naturalisation are hard to come by, but it seems that only a handful of people a year may be naturalised in most countries – even in Nigeria, with more than 150 million people, only around a hundred people acquire nationality by naturalisation or marriage annually – and those who are naturalised are mostly non-Africans operating in the formal economy, with all the panoply of lawyers and documents to support their claim.[18] So few people are involved, and the procedures for obtaining naturalisation are so highly discretionary, that it seems unlikely that anyone who has the slightest possibility of becoming a threat to the security of the state could pass that barrier — and therefore be at risk of subsequent deprivation. It’s not impossible of course; but very unlikely. South Africa has had much more accessible naturalisation procedures, rendering it perhaps more vulnerable in this regard; but the numbers have dropped dramatically in recent years, without explanation.[19]

Therefore, (c) comes into play. The methods traditionally used in Africa to “denationalise” a person are simply to deny that he or she ever had nationality to start off with; to argue that the nationality documentation previously held was issued in error, or to fail to issue or renew a document providing proof of nationality (not even requiring an allegation of fraud). The key amendments to nationality laws in Africa have not been to increase government powers to deprive, but to restrict access to nationality based on birth and residence and to exploit any ambiguity in the rules applied on succession of states at independence.[20] These are the methods used against some high profile individuals: Kenneth Kaunda of Zambia and Alassane Ouattara of Côte d’Ivoire most famously; but also John Modise of Botswana, who found himself no longer considered a national by birth when he set up a political party in order to run for president. These cases reached the African Commission on Human and Peoples’ Rights, but there are many others litigated only at national level involving politicians, journalists or activists.[21]

UNHCR’s clear guidance is that a retrospective finding that a person was not a national and was issued nationality documents in error is just as subject to rules on arbitrariness as any procedure under formal provisions on deprivation.[22] However, under national law, why bother with deprivation proceedings if you can manage matters so much more easily by other methods? And this applies especially when whole categories of people are seen as problematic, or potentially so.

It is, in fact, not the individual difficult cases that raise the greatest concerns in the African context, but the tendency to attribute collective responsibility to whole groups of citizens when a country is faced with a (real or perceived) security threat – or simply an organised opposition with support from a particular ethnic group. Faced with the challenges of “nation-building” in states created by colonial fiat, the question of who belongs is not necessarily an obvious one to answer. African states have a history of mass expulsions based on ethnic grounds — there is even a style of bag known in Nigeria as a “Ghana Must Go” bag, dating to one such episode in the 1980s when (actual or alleged) Ghanaians had to pack up and leave — and it remains the case that the usual approach is to assert that someone is a non-national, and then expel them.[23] The prevalence of such practices led to the inclusion of a specific provision banning mass expulsions, not found among similar treaties, in the African Charter on Human and Peoples’ Rights.[24] Even where those who have been expelled fail to find recognition in their alleged country of origin, they may be unable to reclaim their status in the former country of residence: among those persons of Eritrean origin who were expelled by Ethiopia to Eritrea during the 1998 war between the two countries, a number subsequently became refugees from the highly repressive Eritrean regime. Even in their case, when some applied for reacquisition of Ethiopian nationality, they were reportedly told that they were security risks, so could not get papers.[25] 

In Kenya, discriminatory measures in relation to documentation and identity have been sharply stepped up against Kenyan Somalis and coastal Muslims, tarred with the brush of the Westgate Mall siege and other outrages. In addition to a general round up and detention of suspected youth, the issue of national ID cards has been suspended in the three counties that are located in the former North Eastern Province bordering Somalia (Garissa, Wajir and Mandera Counties, created by the 2010 Constitution), meaning that those without IDs cannot travel out of that zone, and effectively lose the reality of citizenship rights — without the need for the government to undertake any bothersome legal proceedings.[26] In Nigeria, the peculiar features of the country’s federal system have led to the possibility of “denationalisation” from a particular part of the country, even though such measures haven’t been taken at national level. In the context of the threat from Boko Haram, governors of states in the south-east of the country in 2014 stepped up long-standing discrimination based on the idea of “indigeneity” to adopt controversial measures to register and possibly deport “non-indigenes”, leading to an emergency meeting of the National Council of State in July 2014 to condemn these practices (but no action beyond establishing a committee to make recommendations).[27] Ghana’s consul-general in Nigeria indeed recently blamed the Boko Haram insurgency on “stateless people” excluded from the benefits of citizenship, and urged efforts to strengthen documentation across the sub-region.[28]

There are the beginnings of recognition that stronger guarantees around the right to a nationality may be part of the solution to some of the security challenges in the continent. The African Commission on Human and Peoples’ Rights is working with the AU Commission in Addis Ababa to draft a protocol to the African Charter on the right to a nationality.[29] The African Committee of Experts on the Rights and Welfare of the Child recently adopted a General Comment on the rights of children to a name, birth registration and a nationality.[30] In parallel, there is a major push to improve documentation through the initiation or strengthening of requirements to hold a national identity card, for civil registration in general, and for the use of biometric data in these documents. But this push on information technology carries significant risks that governments will seek only to police the boundaries of their systems, excluding anyone of “doubtful” nationality, while failing to reform legal provisions and administrative practices that restrict access to nationality for those who constitute no security threat at all. To date, the international agencies responsible on these issues — especially UNHCR, UNICEF and IOM — are also failing to join up the dots with a coherent approach on nationality and documentation in their interventions with national governments. Given the very real security threats they face, it remains an open question whether governments such as Nigeria’s and Kenya’s will commit to more secure rights to citizenship, rather than only more secure documentation.



[1] Botswana, Burkina Faso, Burundi, Chad, Comoros, Ethiopia, Gabon, Gambia, Ghana, Kenya, Lesotho, Libya, Mauritius, Namibia, Nigeria, Rwanda, Seychelles, Somalia, South Africa, Swaziland, Tanzania, Uganda, Zambia and Zimbabwe. In the case of Botswana, Ethiopia, Libya, Tanzania and Zambia, dual nationality is not permitted, and voluntary acquisition of another nationality results in automatic loss. Lists from Bronwen Manby, Citizenship Laws in Africa: A Comparative Study, Open Society Foundations, 2nd edition 2010; updated information for a forthcoming 3rd edition. On the number of states in Africa: Morocco is not a member of the African Union, while the Sahrawi Arab Democratic Republic is: if both are counted, there are 55 states.

[2] Angola, Cameroon, CAR, Congo Republic, Côte d’Ivoire, Djibouti, Egypt, Equatorial Guinea, Eritrea, Guinea, Guinea Bissau, Liberia, Madagascar, Morocco, Mozambique, Sao Tomé & Príncipe, South Sudan, Sudan, Togo and Tunisia.

[3] Egypt Law No. 26 of 1975 Concerning Egyptian Nationality, Article 16(7); Eritrea Nationality Proclamation 1992 Article 8; Mali Nationality Code 1962, Article 43 (amended 1995).

[4] Law No. 26 of 1975 Concerning Egyptian Nationality, Article 16(7), translation from UNHCR website, http://www.refworld.org/docid/3ae6b4e218.html. Libya had similar rules until they were changed in 2010.

[5] South Africa Constitution 1996, Article 20; Ethiopia Constitution 1993, Article 33.

[6] Kenya Constitution, 2010, Section 17.

[7] Burundi Constitution 2005, Article 34; Malawi Constitution 1994 (as amended to 1998), Article 47 ; Rwanda Constitution 2003, Article 7.

[8] Gambia Constitution 1996, article 13; Ghana Constitution 1992 Article 9, Citizenship Act 2000, Article 18; Liberia Aliens and Nationality Law 1973, Articles 21.53; Rwanda Nationality Law No.30 of 2008, Article 20.

[9] Most of the civil law countries provide quite detailed procedures for nationality litigation through the courts; the Commonwealth countries tend to have weaker protections, and do not have the same tradition of providing procedures in the substantive law itself, but South Africa for example, provides for all decisions of the minister to be reviewable by the courts, as do Gambia and Kenya.

[10] Lesotho Constitution 1993, as amended to 2001, Article 42 (however, this provision is not respected in the Citizenship Order 1971 Article 23); Mauritius Citizenship Act 1968, as amended to 1995, Article 11(3)(b); Namibia Constitution 1990, as amended to 2010, Article 9(4); Rwanda Nationality Law 2003, Article 19; Senegal Nationality Code 1961 as amended 2013, Article 21; South African Citizenship Act 1996, as amended 2013, Article 8; Zimbabwe Constitution 2013, Article 39(3) (but this is not respected in the Citizenship Act, 1984, as amended to 2003, Article 11(3), which provides in principle prohibition of rendering a person stateless, but allows the minister to override if it is in the “public good” to do so).

[11] Botswana Citizenship Act 1998, Article 22; Lesotho Citizenship Order 1971, Article 26; Malawi Citizenship Act 1966, Article 29; Mauritius Citizenship Act 1968 Article 17; Seychelles Citizenship Act 1994, Article 14; Tanzania Citizenship Act 1995 23; Zambia Citizenship Act Article 9; Zimbabwe Citizenship Act 1984 Article 16.

[12] Swaziland Citizenship and Immigration Act 1992, Article 20.

[13] Namibia Citizenship Act 1990, Article 9(3)(e).

[14] Section 11A of the Citizenship Act, No. 18 of 1994, inserted by Act 11 of 2013.

[15] South African Citizenship Act 1996, as amended 2013, Article 6(3). This amendment came into force on 1 January 2013. The 1996 Constitution provides in Article 20 that “No citizen may be deprived of citizenship.” It is possible that the phrasing of the revised Article 6(3) is designed to avoid this prohibition by providing for automatic loss. See further Submission on the South African Citizenship Amendment Bill, B 17 – 2010, Citizenship Rights in Africa Initiative, 6 August 2010.

[16] The numbers of deprivations given in annual reports of the Department of Home Affairs are:





















See http://www.dha.gov.za/files/Annual%20Reports/, last accessed 17 November 2014. The 2002/03 Annual Report is not on the website (nor are earlier years) and the 2012/13 and 2013/14 Annual Reports have ceased publishing any statistics on citizenship procedures; so it is impossible to know if the 2010 amending act has had any effect on deprivation figures.

[17] Email correspondence, November 2014, with Chidi Odinkalu of the Nigeria National Human Rights Commission and Adam Hussein Adam of the Open Society Initiative for East Africa, both following these issues closely.

[18] Bronwen Manby, Nationality, Migration and Statelessness in West Africa, UNHCR and IOM, forthcoming, 2015.

[19] According to South Africa’s Department of Home Affairs Annual Reports, the numbers of people naturalised each year up to 2011-2012 are as below:





















No explanation is given for the dramatic fall in numbers in the last two years of statistics.

[20] Bronwen Manby, “Trends in citizenship law and politics in Africa since the colonial era”, chapter in Engin F. Isin, and Peter Nyers (eds.), Handbook of Global Citizenship Studies, Routledge, 2014.

[21] See Bronwen Manby, Struggles for Citizenship in Africa Zed Books, 2009, chapter 7, for an account of these and other cases where denationalisation has been a tool of censorship; and Manby, Citizenship Laws in Africa for a discussion of the jurisprudence of the African Commission.

[22] Expert Meeting - Interpreting the 1961 Statelessness Convention and Avoiding Statelessness resulting from Loss and Deprivation of Nationality ("Tunis Conclusions"), UNHCR, March 2014, especially paragraph 9.

[23] See Manby, Struggles for Citizenship in Africa, Chapter 4.

[24] Article 12(5) of the African Charter.

[25] Amsale Getnet Aberra, “Ethiopians in Limbo: from statelessness to being a refugee in one’s own country”, ECADF Ethiopian News and Opinions, 14 February 2014.

[26] Email communication, Adam Hussein Adam, OSIEA, November 2014.

[27] “Council of State moves to ‎stop citizens’ registration, deportation” The Citizen, Abuja, 1 August 2014. On the history of discrimination in relation to nationality in Kenya, see Manby, Struggles for Citizenship in Africa, Chapter 6; on Nigeria and “indigeneity”, see Chapter 5.

[28] “Envoy Blames Insecurity in Nigeria, Others on Stateless People”, Premium Times, 29 April 2014.

[29] See ACHPR Resolution 234 on the Right to Nationality, 53rd Ordinary Session, 9- 23 April 2013, Banjul, The Gambia; Resolution 277 on the drafting of a Protocol to the African Charter on Human and Peoples’ Rights on the Right to Nationality in Africa, 55th Ordinary Session, 28 April to 12 May 2014, Luanda, Angola.

[30] Available at the Committee of Experts website: http://acerwc.org/the-committees-work/general-comments/.



Revocation of citizenship of terrorists – a matter of political expediency

By Kay Hailbronner (University of Konstanz and member of the EUDO CITIZENSHIP advisory board)


Let’s be clear: We are not in a dispute about the use of denationalisation policies to get rid of unwanted citizens who do not comply with a code of conduct how to behave as a “good” or “loyal” citizen. Nor are we talking about deprivation or revocation of citizenship on account of race, political opinion, religion, descent etc. There are clear rules of public international law prohibiting discriminatory citizenship policies and none of the policies discussed here call these into question. What we are discussing is the different and by no means absolutely novel issue of revoking the citizenship of persons who have given up or are irrefutably considered as having renounced that basic attachment which distinguishes citizenship from a residence permit. A recent report of de Groot and Vink for the European Commission lists voluntary military service and non military public service in nine, and eight EU countries respectively as a ground for revocation of citizenship, subject of course to some restrictions (prevention of statelessness) and exceptions.

In around half of the 28 countries included in the study, seriously prejudicial behaviour is considered as a ground for revocation of citizenship. The European Convention on Nationality of 6 November1997 provides for  revocation of citizenship for conduct seriously prejudicial to the vital interests of the State party ( Art.7 para 1 lit.d). Very similar provisions on revocation are laid down in Art. 8 para 3 of the Convention on the Reduction of Statelessness of 1961.

What is new is the inclusion of a specific type of seriously prejudicial behaviour which is considered as endangering the safety of the population of a state and its security into this catalogue. The actors are not totalitarian or authoritarian regimes but democratic states with well established institutions to protect human rights and to ensure the rule of law. Not that the democratic character of the states in question would dispense us from closely watching the transfer and exercise of powers to the executive branch, particularly I  such a rights-sensitive area as denationalisation policies. Safeguards against arbitrary actions and abuse of power, conditions and procedures must be predominant on the watch list, as Peter Schuck rightly emphasizes. But why should revocation of citizenship of terrorists result  inevitably in arbitrary and abusive exercise of power, as Audrey Macklin assumes?

What makes international terrorism so distinctive is not only its criminal and administrative relevance, but also its relevance for discontinuance of that special relationship established by citizenship. In order to answer this question it  is not sufficient to conjure up emphatically  the uniqueness of the ties between a citizen and a state. It is true that citizenship establishes a special relationship based upon security and stability. Security and stability on the side of the individual citizen require that denationalisation remains a rare exception.  Citizenship implies rights, whether it is designated as a privilege, as a right to have rights or as a contract. For that reason deprivation of citizenship requires  an overriding public interest and is subject to proportionality. 

Ordinary crimes, even of a serious nature, have not been considered as sufficient under Art. 7 of the European Convention to destroy the bond of citizenship. Yet, fundamental and persistent alienation from the nation as a political community   has – in spite of divergent interpretations and applications – frequently  been considered as a justification for revocation of citizenship. Democratic states in the defence of their constitutional order and protection of the safety of their population and the security of the state are not restricted to  a regime of criminal and administrative sanctions if  their own nationals turn against them.

Legal comparison shows that there is no uniformity. States, according to their particular political conditions, and history that is sometimes reflected in constitutional provisions, have largely prohibited involuntary revocation of citizenship.  Germany is one example, though it provides for loss of citizenship for voluntary service in foreign military services or in case of voluntary acquisition of a foreign citizenship. Other states, like Britain, have applied the concept to high treason, espionage, etc.  International treaties, like the European Convention on Nationality of 1997 or the Convention on the Reduction of Statelessness provide further barriers. States may not provide for the loss of nationality if the person concerned would become stateless (except in case of fraud). One could discuss what this means if a state’s national joins a group or organisation, such as the “Islamic State”, which is dominating a state-like territory and exercises state-like authority. 

Discussion of the international and constitutional law prerequisites of revocation of citizenship is not the concern of Audrey Macklin. She argues primarily with illegitimacy.  As a lawyer I have some difficulty with this term. If it is not illegal, what are the criteria for illegitimacy or immorality?  Her personal idea of how democratic states should behave? That of course may be an acceptable political reasoning, provided I learn more about its ideological premises which I may share or not.  I do understand  Peter Spiro’s objection about the revocation of citizenship as a” security theatre” although I feel not confident on the basis of the  facts to judge whether  it is true that  revocation of citizenship for international terrorists is  impractical and irrelevant. The arguments of illegitimacy, in my view are hardly convincing. Assuming that revocation of citizenship is a (prohibited) form of punishment simply ignores the legal nature of revocation of citizenship. It is not destined to sanction acts of international terrorism, in addition to a potential criminal or administrative sanction. By untying the bond of citizenship, the former citizen can no longer rely upon his/her citizenship for unlimited entry and residence and free international travel. The further argument that there is no chance of rehabilitation is based on the same misunderstanding of revocation of citizenship as a special form of punishment. Citizenship of such persons is revoked because they have given up their attachment to a community by attacking the very fundament of that community, not by merely violating its internal rules of public order. To talk in this context of an inalienable right of rehabilitation, distorts the purpose of citizenship revocation.

The hard questions arise with the formulation of a precise and judicially reviewable provision authorising the executive to revoke citizenship. International terrorism as such is a term open to divergent interpretations. We do, however, have quite some experience, based upon the jurisprudence of national and international courts and Security Council Resolutions, in defining international terrorism. In order to be effective, a provision must include such actions as joining extremist organisations for training in order to use such training for participation in terrorist activities, as well as a membership in an organisation destined to fight against the state whose citizenship the person concerned possesses.

A further question is whether the introduction of a new provision on revocation of citizenship serves a useful purpose.   Utility cannot be denied by reference to criminal law. It goes without saying that acts of international terrorism should be punished and that administrative action should, where possible, be taken to limit the use of passports for international travel for the purpose of preparing or assisting international terrorism – the technical and cynical use of citizenship rights, as Peter Schuck has phrased it. Criminal and administrative sanctions are always attached to specific activities. They do not cover the aspect of using citizenship in a general and in principle unforeseeable manner for acts destined to endanger the security of the state of which the perpetrators are citizens.

The cosmopolitan nature of this type of terrorism, as Christian Joppke has aptly described it, is misunderstood by Vesco Paskalev when he argues that the jihadists do not care about their citizenship. They might indeed not care about their attachment to the state whose citizenship they posses  but they do care about the possibilities that a Canadian, US, British or German passport conveys with visa-free international travel, free entry and residence  in their  “home” country  and diplomatic protection if something does not go quite as smoothly as expected.

Revocation of citizenship means a substantial interference with individual rights. It can only be justified if tightly defined material conditions in accordance with the constitutional law of each country and its international commitments are fulfilled. Risk assessment and proof of an affiliation, assistance or membership in an international terrorist organisation will be essential elements in this procedure. Whether there is a practical value in revocation of citizenship for citizens engaged in international terrorism in addition to criminal and administrative sanctions is within the framework of law a matter of political expediency which may well lead to different results in different countries.   



Whose bad guys are terrorists?

By Rainer Bauböck (EUDO CITIZENSHIP co-director)


Peter Schuck, Christian Joppke and Kay Hailbronner have provided strong arguments why liberal democracies should have the power to strip terrorist suspects of their citizenship. As good lawyers, Schuck and Hailbronner add that such power must be exercised with restraint and hedged in by the rule of law. 

Everybody in this debate agrees that terrorists ought to be punished. Most would also agree that liberal states need exceptional powers in order to prevent terrorism and that this justifies some constraints on freedom of speech and association, for example by making incitement to terrorist violence or joining a terrorist organisation punishable crimes.  

Terrorists commit particularly evil crimes. Yet denationalisation does not look like punishment for these crimes. First, it is normally based on executive order rather than court judgment. Second, it does not meet the standard purposes for criminal punishment. It cannot be justified as retribution, since it is not proportionate to the monstrosity of the crime. It does not promote rehabilitation, since the effect is to remove the criminal from the jurisdiction. And it is not effective in deterring or preventing terrorist crimes, since – as Vesco Paskalev has argued global jihadists hardly care about losing citizenship status in a Western democracy that they detest. 

Hailbronner points out that terrorists care about losing their right to travel, but restricting their freedom to move can also be achieved by other means, e.g. by invalidating their passports without denationalising them. Banishing jihadists to exactly those states where they want to go anyway to commit their atrocities can hardly count as an effective strategy against global terrorism. As a political scientist I suspect that governments have other motives apart from policy effectiveness when they seek denationalisation powers. They do not only want to do something against terrorism, they also want to be seen by voters as doing something. Stripping terrorist suspects of their citizenship is a strongly visible policy and for that reason possibly also a strongly symbolic one, as suggested by Peter Spiro. 

This is not yet a conclusive refutation, since on some views it is exactly the symbolic nature of the sanction that justifies the denationalisation of terrorists. This argument starts from the assertion that liberal democracies are value-based political communities. Their basic values include freedom of conscience and religious practice, of speech and association and democratic self-government. Since these states are liberal, they cannot force their citizens to share their basic values. These are instead enshrined in their constitutions and their political institutions are designed to protect these values. Terrorists do not merely reject liberal values, they act to destroy the very institutions that protect these values. So why should liberal states not take away citizenship from those who attack the very foundations of liberal citizenship? Wouldn’t this serve to defend these states’ core values?

The answer is that the norms guiding the acquisition and loss of citizenship status have little to do with either the promotion or the defence of liberal values. In all states, including liberal ones, citizenship is acquired automatically at birth and normally retained over a whole life. Native citizens are never asked to show their commitment to liberal values as a condition for retaining their citizenship, nor are they stripped of their status when they commit crimes. Serious criminals are locked up in prison and thereby stripped of many citizenship rights, most importantly that of free movement. In some countries they also lose – and in my view much more questionably – voting rights. But they do not lose their citizenship status. Citizenship in our world has an extremely sticky quality. It does not have an expiry date, it can be passed on to subsequent generations and it can be carried abroad and increasingly also exercised from outside the state territory. 

Yet many liberal states have introduced citizenship tests or naturalisation oaths in which immigrants are asked to affirm their commitment to the polity and its constitution.(1)  Doesn’t this show that acquisition of citizenship status and therefore also its loss may depend on a commitment to liberal values? No, it doesn’t. Leaving aside the tricky question whether such commitments can be tested by filling in a questionnaire or taking an oath, naturalisation integrates newcomers into a political community that is based on birthright membership and equal citizenship. No matter how they have been selected and how they have acquired their citizenship, all citizens have equal membership status and those who have got it through naturalisation can retain it in the same way as if they had got it by birth. 

This statement needs two minor qualifications. First, if citizenship has been acquired unlawfully, for example through concealing a criminal record, then it may be withdrawn. This reasoning cannot be applied to citizens who assert their commitment to a liberal constitution in a citizenship test or loyalty oath that they subsequently violate. Because liberal states cannot force ordinary citizens to support their core values, they also cannot claim that citizenship status has been acquired unlawfully if a naturalisation applicant was not sincere when swearing loyalty or was sincere and subsequently changed his views. 

Second, the norm of equal treatment of native and naturalised citizens is not accepted by all liberal states – as we all know, the American President must be a native citizen. It is, however, enshrined in Art. 5 of the 1997 European Convention on Nationality and it is not difficult to see why unequal treatment of citizens based on their circumstances of birth is discriminatory and undermines the core value of equality. Faced with terrorism that is now no longer just imported but also home-grown, Western governments may anyhow be reluctant to limit the application of their denationalisation powers to naturalised immigrants. 

There are two closely connected reasons why citizenship status is sticky and why it should not be taken away even for acts that attack the foundations of the polity. The first reason has to do with the function of nationality in the international state system. Citizenship is a mechanism for assigning responsibility for individuals to states. In its 1955 Nottebohm judgment the International Court of Justice asserted that citizenship should be based on a genuine connection in order to prevent  states from abusively bestowing their citizenship on individuals residing abroad who want to escape a legal duty towards their host country. The same genuine link argument has been invoked by the European Parliament and Commission against Malta in January this year as an objection against the sale of EU citizenship to wealthy foreigners without a residence requirement.(2)  If states can abuse their powers to confer citizenship by naturalising foreigners who lack a genuine connection, they can also do so by denationalising their citizens in order to shift responsibility for them to another state. This is exactly what happens when Western countries deprive terrorist suspects of their citizenship. As Audrey Macklin has already explained, the effects can be particularly perverse for dual citizens. Since deprivation does not make them stateless, each of the two states involved has an incentive to act first so that the other state becomes responsible. 

International law can thus not provide a full answer to our question. We must also consider what depriving terrorist suspects does to the citizenship bond as an internal relation between an individual and a state. Joppke points out that Germany did not expatriate the left wing terrorists of the Red Army Faction. They wanted to transform the German state whereas the global jihadists de facto renounce their membership by affiliating themselves with an Islamic pseudo-state. But the RAF was certainly as effective in shaking the foundations of a liberal Rechtsstaat by triggering illiberal responses as was Al Qaeda when it fell the twin towers in New York – and much more so than IS, who primarily wants to scare Western powers out of Iraq and Syria. In any case, the question here is not whether Ulrike Meinhof and Andreas Baader had a moral claim to German citizenship that jihadist terrorist suspects do not have. The question is whether Western democracies can shed responsibility for their home-grown citizen terrorists and shoulder it upon other states. This is what the new denationalisation policies are about. 

Imagine for a moment that after 1945 Germany or Austria had posthumously denationalised Adolf Hitler. Would this symbolic act have strengthened their post-war liberal orders by demonstrating their abhorrence of Hitler’s destruction of their liberal constitutions and his genocidal elimination of Jews and Roma from the political community? The answer is clearly no, because Hitler’s denationalisation would have entailed a denial of responsibility for his crimes and their consequences and would thus have achieved the very opposite of the intended defence of liberal values. Moreover, if either Germany or Austria had taken such a decision, it would have signalled that they merely wanted to pass on the buck to the other state. Recognising that Hitler was “our bad guy” was therefore crucial for building a liberal democratic consensus in both countries and good relations with other states that were the victims of Nazi aggression. 

Why should this be different today with the jihadist terrorists? Joppke’s answer involves an attempt to distinguish domestic from global terrorists. This may be often difficult, since Hitler turned out to be a global terrorist too. But the crucial point is that citizenship is by its very nature a domestic relation between an individual and a state. By cutting the bond, states deny their responsibility, including that towards the rest of the world upon whom they inflict the terrorist threat.

If denationalisation were a necessary and effective tool to prevent terrorism, it might be justifiable on such utilitarian grounds. But as a symbolic defence of the liberal values that terrorists attack it is entirely unconvincing.



(1) see EUDO CITIZENSHIP Forum Debate 'How Liberal are Citizenship Tests?' 

(2) See the press release of EU Justice Commissioner Vivian Reding (15 January 2014), the European Parliament resolution of 16 January 2014 on EU Citizenship for Sale, and the EUDO CITIZENSHIP Forum Debate 'Should citizenship be for sale?'



Denationalisation, assassination, territory: Some (U.S.-prompted) reflections                                    

By Linda Bosniak, Rutgers University School of Law    


Unlike the several liberal states Macklin cites which have already, or will soon, deploy citizenship revocation as an anti-terrorism mechanism, the United States is unlikely to implement similar policies. The U.S. Constitution has been interpreted to prohibit unilateral citizenship-stripping as a tool of governance. Instead, denationalisation via expatriation in the U.S. requires the individual to specifically consent to relinquish the status, and such consent cannot be inferred from acts alone – even from acts which some (including some commentators in this symposium) would like to characterise as intrinsically antithetical to citizenship identity. The vigorous safeguarding of individual citizenship in US law is borne of the nation’s history of race-based slavery and its aftermath. Today, courts  quite stringently interpret the Fourteenth Amendment’s guarantee of  citizenship status for “all persons born or naturalised in the U.S.” I realise the matter of slavery will seem remote from the concerns of contemporary transnational debates over citizenship-stripping in Europe and Canada (although it might be worth wondering, another day, if “slavery” could ever serve – along with “political death” – as a fruitful analytic metaphor here. Think, for example, of the recent mass denationalisation of Dominican-born Haitians in the Dominican Republic). Nevertheless, we know that national citizenship law and policy look inward as well as outward. In the U.S., the legacy of slavery forms a part of a deep conversational grammar about citizenship in a way that will almost certainly stay the hand of congressional advocates of the “Enemy Expatriation Act” and similar proposed measures.  

That the US is not about to join Britain and Canada and other states in a politics of forcible expatriation, however, by no means implies that the US does not wish to “permanently eliminate” suspected or confirmed terrorists, nor that it is unable to do so. Indeed, we have recently seen deployment by the U.S. of what Macklin calls  “the sovereign’s other technique for permanent elimination” of such persons: namely: state-inflicted death. The 2013 assassination of U.S. citizen Anwar al-Awlaki in Yemen was a widely noted recent example of this policy (with the apparently accidental assassination Anwar’s 16-year-old U.S. citizen son, Abdulrahman, a notorious follow-up.)  For some commentators, state acts of this kind may appear more “proportional” to the claimed offenses than expatriation is. Personally, I would not endorse any policy of assassination, much less when visited upon its target without application of due process. But my comments don’t concern the policy’s defensibility. Instead, I raise the al-Awlaki case to frame a few brief observations about the relationship between citizenship-stripping, targeted assassination and territoriality in the United States and beyond.  

First, as Macklin points out, states strip citizenship not merely in order to territorially banish the affected going forward but sometimes perhaps, as a “prelude to assassination,” whether by themselves or others. In particular, Macklin cites the cases of Britons who were denationalised and subsequently killed by US drone strike in Somalia.[1] Denationalisation here can be understood to have strategically relieved Britain of the imperative of protecting its own nationals from harm, including assassination, by another state party.  In this scenario, denationalisation is not merely a form of political death; as Macklin argues, it may facilitate bodily death as well.  

Nevertheless, we have also seen that since United States law makes it “easier to kill than expatriate,” in Peter Spiro’s succinct phrasing,[2] the U.S. government does not await denationalisation to assassinate its own citizens.  We could, indeed, view assassination of al Awlaki senior as the nation’s only route to denationalise him, with assassination serving as the actual mechanism for stripping his citizenship.  

On the other hand, al Awlaki’s assassination precipitated a fascinating debate in the United States about territoriality and citizenship which perhaps bears on our transnational conversation here. In the wake of the killing, a segment of the US political class erupted in concerted anxiety about whether the government actually claimed authority not only to assassinate US citizens abroad but to do the same “on US soil.”  Senator Rand Paul led a filibuster against the confirmation of proposed CIA Director John Brennan, promising to “speak as long as it takes until the alarm is sounded from coast to coast that our Constitution is important, that your rights to trial by jury are precious, [and] that no American should be killed by a drone on American soil without first being charged with a crime, [and] found…guilty by a court.” Much media fan-flaming followed, and eventually, Attorney General Eric Holder conceded that targeting any U.S. citizen for assassination within national territory—in the absence of imminent threat – is unacceptable. [3]  

What was striking in this episode was the normative distinction taken up in popular discourse between in-country and out-of-country citizen assassination. The implied claim was that death of a citizen by its own government was somehow uniquely intolerable when accomplished inside national territorial bounds. For that moment, at least, the American political imaginary seemed to coalesce more around fear of tyrannical government than of the foreign terrorist within.  

Of course, if government were in fact bound by this normative logic – i.e., that territorially present citizens are uniquely out of bounds for targeted killing – then the target would need to be denationalised and/or territorially expelled first and only executed thereafter. Yet since the US state is constrained in denationalising citizens, and since, like all states, it is precluded from expelling citizens, it would seem to have to await such person’s travel outside the country in order to strike.   This seems odd, yet it notably parallels the form denationalisation practices take in many countries – where, according to Macklin, governments tend to strip citizenship from those citizens who are already located abroad. In both settings, we see not only that territorially-present citizens are regarded as possessing more fundamental protections against government power than those territorially absent, but that governments make opportunistic use of citizen absence to act against them.  Among other things, this amounts to a kind of penalty on citizen mobility, and seems to rest on an arbitrary locational distinction. This, at least, is what the US Supreme Court itself concluded in 1957 in a related context when it wrote that a citizen’s constitutional rights may not “be stripped away just because he happens to be in another land.”[4]  

Of course, territoriality’s relationship with citizenship sometimes reaches back well beyond any possible denationalisation and assassination to the moment of the citizen’s birth. For some, the Awlaki affair itself evoked longstanding debates about assignment of citizenship based on territorial presence at birth, with Awlaki an exemplar of the “nominal citizen” whose extraterritoriality for most of his post-natal life rendered his social attachment to the nation “highly attenuated” (to use Macklin’s phrase).  Yet in this setting as well, the United States will remain robustly-citizenship protective. The country’s inclusive birthright citizenship rules are another stanchion of its post-slavery, post-Civil War, constitutionalism.  Consequently, and much as some “anti-birthers” wish it were otherwise, citizenship cannot be easily eliminated on the front end here, except by way more stringent immigration and border control policies to prevent, ex ante, potential parents’ territorial presence. Broadly drawn and often selectively-applied grounds of inadmissibility and deportability based on “terrorist activity” arguably go some of the distance in accomplishing that end.[5] 

In short, citizenship status, especially for those in national territory, still remains more secure in the U.S. than it is in some other national settings.  Our government works to counter the alleged “bad guys”  (Baubock’s shorthand) by different means.      

[1] See “British terror suspects quietly stripped of citizenship… then killed by drones,” The Independent, 28 February 2013 and “Britain Increasingly Invokes Power to Disown Its Citizens”, New York Times, April 9, 2014
[2] Spiro, Expatriating Terrorists, 82 Fordham L. Rev. 2169, 2177 (2014).
[3] For more extensive discussion and citations, see Bosniak, Soil and Citizenship, 82 Fordham L. Rev. 2069 (2013).
[4] Reid v. Covert, 354 U.S. 1 (1957).
[5] E.g., Stephen H. Legomsky, The Ethnic and Religious Profiling of Noncitizens: National Security and International Human Rights, 25 Boston College Third World Law Journal 161 (2005)


Human rights for all is better than citizenship rights for some

By Daniel Kanstroom (Boston College Law School)    


This is an exceptionally rich and challenging discussion in which I am honored to participate, though time and space limitations will inspire brevity. Audrey Macklin’s essay reaches two major conclusions with which I heartily agree:  

  1. Citizenship-stripping weakens the concept of citizenship;
  2. It is of highly-questionable efficacy and legitimacy as punishment;[1]

Despite my deep agreement with Macklin about the dangers of denationalisation trends in the UK, Canada, and elsewhere, I am not convinced that she has chosen the best way to counter them. In brief, I fear that Macklin may have missed some of the forest for the trees.  

My view of the forest is this: Denationalisation should be situated against a broader backdrop in which pervasive rights deprivations against noncitizens – and even such extraterritorial rights deprivations against citizens as drone strikes – are central components. Macklin points us in this direction when she distinguishes the aspirational safe harbor of citizenship from a functional methodology:  

“But my point is not to propose a metric capable of measuring the quantitative, qualitative, experiential, emotional, personal, familial, cultural, social, financial, linguistic and political impacts of exile on any individual, in order that some state official could determine precisely when citizenship revocation inflicts an appropriate versus excessive degree of punishment.”  

I fully support Macklin’s desire to enhance, “the security that distinguishes legal citizenship.” I worry, though, about what certain approaches to such security might mean for, “other statuses that define the relationship between state and individual.” The challenge is to protect citizenship rights without relegating those “other statuses” unduly tenuous and marginal.  

Put simply, I suggest that the best way to do this is less (formally) citizenship–centered and more (functionally) rights-centered. By “rights-centered,” I mean, essentially, a critical examination of state practices (including the government’s intentions and justifications, and the practices’ mechanisms, and effects) measured against the norms of a fully-developed human rights protection system.[2] More specifically, the important legal and policy questions raised by Macklin may be best answered by viewing denationalisation along a continuum of state practices that use citizenship status and territorial formalism to achieve policy goals with weakened (and in some cases no) rule of law encumbrances. This is one of the great human rights legal challenges of our times. It must be engaged fully – in all of its manifestations – in order to be properly understood and effectively engaged.  

Macklin rightly notes that, “…citizenship revocation is best understood as a technique for extending the functionality of immigration law in counter-terrorism.” Moreover, “[s]ince 2001, states have turned to deportation to resolve threats to national security by displacing the embodied threat to the country of nationality.” However, the deep significance of these insights may be lost by too formalistic and narrow an examination of the particular practice of denationalisation. A basic reason for this is the powerful attraction—symbolic and practical--of citizenship as a safe harbor. That, in and of itself, is unobjectionable. But it risks denigration of the rights claims of noncitizens. Let me emphasize that I do not think that Macklin intends this at all. Still, her method may take us there.  

Here is an example. Macklin writes, “Banishment fits the crime of disloyalty the way capital punishment fits the crime of murder.” This works for me passably well as analogy (though, of course, the “crime” of disloyalty is a much more complex proposition than murder). But the analogy prompts a question: How does banishment (of citizens) differ from what I have termed “post-entry social control deportation,” which in the U.S. has resulted in lifetime exclusion of many thousands of long-term legal residents from their families and communities due to minor criminal offenses?[3] Does their lack of citizenship status render the death penalty analogy less apt? In another passage, Macklin correctly worries about “arbitrary and prejudicial abuse of a discretionary power.” What do we make of the fact that such abuses are rare against citizens and troublingly common against noncitizens? Macklin is thus right, but perhaps insufficiently expansive when she asserts that the particular practice of denationalisation “is exile.” Is denationalisation categorically different from expulsion and removal of long-term legal residents because, as Macklin argues, it “extinguishes a singular right of citizenship, namely the right to enter and to remain”?  This seems formalistic and perhaps a bit circular. A fuller exploration might consider the actual effects of deportation and denationalisation on people of various statuses, various levels of assimilation, and various fears of harm. This would help explain why the “right” to enter and remain is so crucial to protect against disproportionate or arbitrary state action against all people.  

My main concern is about the potential implications of Macklin’s methodology. The formalistic reification of citizenship may justify the relegation of noncitizens to a nether world of inferior balancing tests.[4] This is especially the case if that reification is connected to an implicitly exclusive set of rights claims to enter and remain. Noncitizens have such rights, too, at least under certain circumstances. Insufficient attention to such rights – though they are concededly still works–in-progress – is especially dangerous where the rights claims at issue include the right to life, to proportional punishment, to family unity, against arbitrary detention, and to procedural fairness.  

Let us also consider the etiology and evolution of denationalisation. Harsh expulsion and exclusion practices against noncitizens can provide a conceptual matrix that facilitates similar practices against citizens. As Thomas Jefferson – writing to oppose the Federalists’ Alien Friends Act, Alien Enemies Act, and Sedition Act – warned in 1798: “The friendless alien has indeed been selected as the safest subject of a first experiment, but the citizen will soon follow...”[5] The best response to this concern, however, is not a regime of exclusive protections only for citizens. Rather, we should strengthen reasonable (procedural and substantive) human rights protections for all people, regardless of status or location. I expect that Macklin would not strongly disagree with this. Still, insufficient attention to such experiments against noncitizens have had metastatic tendencies in the past.  

Denationalisation should not be viewed as an anomalous practice that requires a unique normative critique grounded on a strong, formalistic conception of citizenship as the (supposed) Arendtian “right to have rights.” Rather, it should be viewed as the apotheosis of an evolving array of exclusion and removal practices, as well as the episodic search by governments for what some termed Guantánamo Bay: “a legal black hole.” [6] A more capacious analysis would thus not only critique the British, “conducive to the public good” standard as relegating citizens to the status of permanent residents. It would equally question the standard’s legitimacy and propriety for the latter group.[7] (Indeed, its attempted application to citizens might be ironically salutary, as political opposition will be more readily mobilised if it is practiced widely.)  

Easy denationalisation deserves normative and practical critique, to be sure. As Rainer Bauböck properly highlights, citizenship is (and should be ) “sticky” and thus denationalisation must be justified as punishment. This practice is ill advised, problematic, and especially difficult to justify in liberal democracies for the reasons he highlights. However, critique should be primarily grounded in a broader set of human rights norms that apply whenever a state seeks to use its power disproportionately or arbitrarily against anyone anywhere. This is especially important for those who are strongly assimilated, who would be rendered juridically or functionally stateless or who would face severe harm, persecution, or torture.  

In a similar vein, I would not recapitulate the rather formalistic and ultimately sterile debate between a “right” and a “privilege,” nor rely too readily on Justice Earl Warren’s channeling of Hannah Arendt. When Warren asserted that citizenship is “the right to have rights,” he was tactically using this phrase to justify a particular position in a dissent in a 1958 case.[8] The case involved a U.S. citizen (by birth) who had lived much of his life in Texas but had been raised in Mexico. He later returned there, apparently to avoid US military service. He had also voted in Mexico.[9] The court narrowly upheld the denationalisation (also called “expatriation”). Justice Warren wrote a somewhat rambling dissent built around the (unattributed) reference to Arendt.[10] He concluded with two apparently contradictory propositions. The first was seemingly absolute, if a bit puzzling: “The Government is without power to take citizenship away from a native-born or lawfully naturalized American.”[11] The second conclusion focused on the intention of the citizen: “The citizen may elect to renounce his citizenship, and under some circumstances he may be found to have abandoned his status by voluntarily performing acts that compromise his undivided allegiance to his country.” Thus, even Justice Warren accepted that certain conduct could justify expatriation, so long as the conduct was voluntary.[12] But this fits poorly with the absolutist reading of the “right to have rights.” Who would voluntarily relinquish the right to have all rights?  

Later U.S. cases elaborated on the criterion of voluntariness, ultimately elevating it to the dominant principle.[13] However, as Justice Harlan once noted, the historical evidence limiting government power to voluntary expatriation was questionable, to say the least.[14] Harlan highlighted a more functional, less formalistic defense of citizenship: “Once obtained, citizenship is of course protected from arbitrary withdrawal by the constraints placed around Congress’ powers by the Constitution….” This model seems to dovetail with Peter Schuck’s proposal in this debate.[15] It has the powerful virtue of situating denationalisation within the rubric of well-accepted protections of the rule of law.  

Finally, one should also note something obvious but worth highlighting: Hannah Arendt’s position was not that citizenship should be the “right to have rights.” Rather, as she expressly put it: “The Rights of Man, supposedly inalienable, proved to be unenforceable…whenever people appeared who were no longer citizens of any sovereign state.”[16] (Arendt 1966: 293) Her concerns were practical: Such people lacked any real protection. When she explored the subject substantively her argument was much more nuanced: “…recent attempts to frame a new bill of human rights, which seem to have demonstrated that no one seems able to define with any assurance what these general human rights, as distinguished from the rights of citizen, really are.” (Ibid.)[17] But Arendt published The Origins of Totalitarianism in 1951. It hardly needs to be said that—despite its evident challenges and deficiencies--the corpus of human rights protections is today more specific, more robust, and more widely enforced than was the case during the times she considered.  

Arendt also poignantly described the “calamity of the rightless” as “that they no longer belong to any community whatsoever.” The main reason this was a calamity was that “no law exists for them.” (Ibid: 295) The best way to avoid such calamities is not only to strengthen citizenship protections. That may well have the perverse consequences of, on the one hand, rendering citizenship ever harder to achieve, and on the other, relegating noncitizens to an increasingly rightless realm. We must do the harder, more basic work of defining and instantiating meaningful human rights protections for all people, regardless of status, or location. Focusing too specifically on the problem of deprivation of citizenship must not blind us “to the numerous small and not so small evils with which the road to hell is paved.”[18]

[1] Though I agree with Kay Hailbronner that legitimacy is an elusive concept in need of further definition. I believe that one can do this relatively easily in this context.
[3] See e.g., Daniel Kanstroom, Daniel. 2007. Deportation Nation: Outsiders in American History. Cambridge: Harvard University Press; Kanstroom, Daniel. 2012. Aftermath: Deportation Law and New American Diaspora. New York: Oxford University Press.
[4] I suppose that the opposite might also be true in certain circumstances. Rights gains won by citizens could form models that protect long term residents, albeit in depreciated form. But this pathway works best if citizenship is viewed on a continuum.
[5] The Kentucky Resolution, Documents of American History 181(Henry Steele Commager ed., 6th ed. 1958).
[6] See Hamdan v. Rumsfeld, 548 U.S. 557 (2006) (concluding: “We have assumed … that Hamdan is a dangerous individual whose beliefs, if acted upon, would cause great harm and even death to innocent civilians, and who would act upon those beliefs if given the opportunity. … But in undertaking to try Hamdan and subject him to criminal punishment, the Executive is bound to comply with the Rule of Law that prevails in this jurisdiction.”) See also, Johan Steyn, Guantanamo Bay: A Legal Black Hole, 1 International and Comparative Law Quarterly 53 (2004).
[7] By which I mean conformity to the best understanding of the “rule of law” in all its aspects, including procedural and substantive protections of basic rights.
[8] Perez v. Brownell, 356 U.S. 44 (1958).
[9] The 1940 law at issue had been passed largely in response (ironically for our purposes) to voting by American citizens in a 1935 plebiscite relating to Hitler’s annexation of the Saar region. As one member of congress put it. The legislation “would “relieve this country of the responsibility of those who reside in foreign lands and only claim citizenship when it serves their purposes.” Id. at 55 (in opinion of Justice Frankfurter).
[10] (joined by Justices Douglas and Black)
[11] Puzzling because the latter practice (denaturalisation) was well accepted in a wide variety of situations, such as where naturalisation had been illegally procured. The term, "lawfully,” thus meant that one could not be denaturalised absent a finding that the naturalisation (viewed retrospectively, had been in some way unlawful).
[12] Cf. Trop v. Dulles, 356 U.S. 86 (1958)(in which Justice Warren, writing for a plurality, found denationalisation of a military deserter to be invalid for similar reasons, and also invalid as cruel and unusual punishment, because it resulted in “the total destruction of an individual’s status in organized society.”)
[13] See e.g., Aforyim v. Rusk, 387 U.S. 253 (1967)(“First we reject the idea…that…Congress has any general power, express or implied, to take away an American citizen’s citizenship without his assent.”); Vance v. Terrazas, 444 U.S. 252 (1980)(“[T]rier of fact must in the end conclude that the citizen not only voluntarily committed the expatriating act prescribed in the statute, but also intended to relinquish his citizenship.” Proof may be by a “preponderance of the evidence” standard.) See also, Pub.L.99-653 (1986) (adopting this approach).
[14] Senator Howard, who had sponsored the Citizenship Clause of the Fourteenth Amendment, had conceded that citizenship could be “forfeited” due to “the commission of some crime.”
[15] It should also be noted that US law has long provide for such denationalisation for a wide variety of actions, including:  “committing any act of treason against, or attempting by force to overthrow, or bearing arms against, the United States, violating or conspiring to violate any of the provisions of section 2383 of title 18, United States Code, or willfully performing any act in violation of section 2385 of title 18, United States Code, or violating section 2384 of said title by engaging in a conspiracy to overthrow, put down, or to destroy by force the Government of the United States, or to levy war against them, if and when he is convicted thereof by a court martial or by a court of competent jurisdiction.” Immigration and Nationality Act Sec. 349. [8 U.S.C. 1481].   The operative standard, as noted, is the following: “A person who is a national of the United States whether by birth or naturalisation, shall lose his nationality by voluntarily performing any of the following acts with the intention of relinquishing United States nationality- …”  Kay Hailbronner correctly highlights the prevalence of such standards elsewhere though I am less optimistic than he about the ability of states to define terrorism with sufficient precision to justify denationalisation.
[16] Arendt, Hannah. 1966. The Origins of Totalitarianism. New York: Harcourt, Brace & World, Inc.
[17] Indeed, Arendt herself defined the “right to have rights” not as formal citizenship status as such, but as the right “to live in a framework where one is judged by one’s actions and opinions…” She distinguished this from the related “right to belong to some kind of organized community.” (Ibid: 296-7)
[18] Arendt, Hannah. 1994:271 Essays in Understanding 1930-1954, Formation, Exile, and Totalitarianism. (New York: Schocken Books).

Beware states piercing holes into citizenship
By Matthew J. Gibney, University of Oxford
I find a great deal to agree with in Audrey Macklin’s trenchant and wide-ranging argument against denationalisation power’s recent revival in Western countries. Yet I also understand where her critics are, somewhat abrasively, coming from. It is of course possible to imagine carefully fashioned cases where denationalisation seems a morally appropriate response as long as a range of guarantees are met (for example, when an individual represents a clear threat to the state, where there are no doubts about his guilt or intentions, and where he could be stripped of citizenship without being made stateless.) However, while this realisation might help us identify the terms on which the denationalisation of a particular individual is permissible, it tells us little about the broader consequences of piercing the norm of unconditional citizenship for punitive reasons.(1) I think that once we are realistic about the political dangers of conceding to the state powers to withdraw citizenship, we’re brought back to a position compatible with Audrey Macklin’s ban on denationalisation.
Before explaining why I think an absolute bar might be justified let make a couple of comments on the previous discussion. The first of these is on what one might call the statelessness constraint. All of the critics of Audrey Macklin’s position start (with the possible exception of Christian Joppke) by accepting that individuals, even those who commit terrorist acts, should not be made stateless. This constraint against statelessness is not simply a matter of international or domestic law; it is also a normative constraint that stems from basic liberal commitments. The problem with statelessness is that it leaves individuals subject to state power without citizenship’s basic protections against that power, including security of residence, political rights, and potentially a host of other entitlements. If we accept this normative rationale for guarding against statelessness, as I think we should, we will also want to ensure that those denationalised are not made de facto stateless, that is, forced to rely on a state that is unable and unwilling to protect them or otherwise to deliver the fundamental rights citizenship (or nationality) is supposed to guarantee.(2)
Yet taking this additional constraint seriously is going to be very consequential. The secondary citizenships of the individuals Western states most want to strip of citizenship tend to be those of countries with dubious human rights records and histories of civil war and conflict (Somalia, Iraq, Eritrea, Sudan, to name a few).(3) If de facto statelessness is a bar, most of the prime targets are going to be out of denationalisation’s reach. Of course, de facto statelessness does not establish a case for an absolute rejection of the state’s power to denationalise. But it does show why the power’s scope may be very narrow indeed, at least for liberals.
Second, I find myself attracted to the position of Rainer Bauböck that one reason denationalisation is unacceptable is because it involves states “passing the buck” of their own responsibilities on to other states, a point that adds a different dimension to Audrey Macklin’s claim that citizenship is, in important respects, not fungible. This view that banishment is unfair to other states is a very old one. None other than Voltaire argued against the practice of banishment on the grounds that it involves throwing into our neighbour’s field the stones that incommode us in our own. 
Powerful as it is, however, the consideration that there’s something wrong with denationalising “home grown” terrorists, wouldn’t mean that denationalisation was always inappropriate. States might still claim the moral right to denationalise individuals who had held citizenship only for a short period of time or had spent most of their lives living in the other country in which they held citizenship. Germany certainly should not have posthumously denationalised Hitler. But Hitler was the leader of the German state and celebrated in this role by a significant proportion of the German people during the 1930s and 1940s. Putting aside the question of what should be done posthumously, some citizens have a much more tenuous, even a merely nominal, relationship to the state. Not all are even grown at home.
These considerations help to clarify some of the constraints necessary for a liberal denationalisation power. Even from the short discussion here, we can identify plenty of others. Peter Schuck suggests that an individual’s threat to the state needs to be “rigorously proven” and Kay Hailbronner argues that citizenship deprivation must be “subject to proportionality”. It’s clear that satisfying all of these different requirements will make the construction of denationalisation law consistent with liberal principles a Herculean task. However, where I part company with the denationalisers is not so much over whether it’s possible to identify a liberal starting point for the practice.(4) Rather, my concern is over the illiberal direction denationalisation seems likely to take once it returns to the political repertoire. Here my position has been greatly influenced by the recent experience of the UK.(5) 
When denationalisation was first revived after over thirty years of desuetude by the Blair government in 2002, the power was tightly constrained: the definition incorporated was taken from the European Convention on Nationality, only dual nationals were targeted, and an automatic judicial appeal was to follow any decision by the Home Secretary. The government promised to use the power rarely. This modest beginning for denationalisation did not last. After the London bombings in July 2005, a new act passed by the Blair government in 2006 lowered the standard required for denationalisation. While previously the Home Secretary had to be satisfied that an individual had engaged in actions that threatened the "vital interests of the UK" state, now he or she had only to be satisfied that taking away someone’s citizenship was “conducive to the public good”. The standard for continuing to hold British citizenship had now become the same as the one used to judge whether a non-citizen should be deported. Even after this radical change, it was possible to convince oneself that the government would use the power sparingly. Only a handful of people lost their citizenship under the Labour government’s watch.
But with the coming of the Conservative/Liberal Democrat coalition government things have gone seriously awry. In the Cameron government’s first year of office in 2010–11, no fewer than six people were stripped of their citizenship. This was more people than the Blair and Brown governments had denaturalised in the previous nine years (in the immediate aftermath of the terrorist events of September 11, 2001 and July 7, 2005). The enthusiastic use of deprivation power has continued apace in the years since, though almost always in secret. By May 2014, it was evident that Cameron’s government had some 23 people stripped of citizenship on ‘not conducive’ grounds in the last three years. Almost all of these individuals were stripped of citizenship while outside the UK, undermining real access to appeal procedures. In January this year the government presented a bill to parliament requesting the power to strip citizenship from naturalised citizens even if they would be made stateless. The amendment passed, albeit, in a modified form. Under current law in the UK a naturalised citizen can be made stateless if the Home Secretary deems there are reasonable grounds for believing they have access to another citizenship. 
Now it might be said – and Christian Joppke would probably be the one to say it – that the UK is an outlier. The unravelling of constraints on denationalisation evident in Britain is unlikely to be repeated elsewhere because other Western countries are less insouciant about protecting rights. But note that the circumstances that have geed along transformation in UK law are generally applicable: terrorist events (the 2005 Tube bombing) and a change of government (the coming of the Conservatives to power). Moreover, I’m not confident that other countries are as legally protected against creep of denationalisation power as they might seem. Australia has fewer rights based protections even than the UK; Canada has some alarming inclusions in its recent denationalisation legislation, including the state’s ability to rely on a conviction for terrorist offences in another country; and, as I write, a large number of prominent US politicians (buoyed by public opinion polls) have effectively endorsed torture as a practice for dealing with terrorists past and future. 
I thus find myself agreeing with Audrey Macklin’s embrace of unconditional citizenship, albeit because I fear where we will end up if we try to pierce even a small – liberal size – hole into citizenship to punish terrorists. Liberalism is not simply a set of principles, it’s also a political stance – one that encourages a healthy scepticism of state attempts to encroach upon established rights and protections. In these fraught times, it is wise to adopt the stance as well as to protect the principles.
(1) My focus in this short piece is exclusively on the punitive withdrawal of citizenship. There are, of course, other reasons why states have sought to “pierce” citizenship, for example, to address fraudulent acquisition of citizenship or to prevent dual nationality.
(2) Cf. C. Barry and L. Ferracioli, “Withdrawing Citizenship”, paper delivered at Australian National University, Canberra, 16 July 2013. I accept that specifying exactly what is included in the concept of “de facto statelessness” is not necessarily clear, as is the relationship between de facto statelessness and simple human rights abuses. A good starting point for further consideration of this issue is C. Sawyer and B. K. Blitz, eds. Statelessness in the European Union: displaced, undocumented, unwanted. Cambridge University Press, 2011.
(3) Note, for example, the second nationalities of the denationalised individuals that the Bureau of Investigation Journalism has been able to track.
(4) I discuss the normative complexities of denationalisation in M. J. Gibney (2013), “Should citizenship be conditional? The ethics of denationalization”. The Journal of Politics, 75(03), 646-658.
(5) I give a fuller account of the history of UK denaturalisation power in M. J. Gibney (2013) “‘A Very Transcendental Power’: Denaturalisation and the Liberalisation of Citizenship in the United Kingdom”. Political Studies, 61(3), 637-655.




Disowning citizens

By Reuven (Ruvi) Ziegler, University of Reading


Macklin’s kick-off focused ‘exclusively on denationalisation for allegedly disloyal conduct by a citizen, while a citizen’. Most contributions to this debate weighed the predicament of the former citizen against state interests. In my contribution, I offer a typology of cases in which revocation could be sought according to some of the contributors. I contend that disowning of citizens by their states is incoherent, tenuous, or disingenuous.  

The first type of case involves acts which, according to Hailbronner, undermine the constitutional order by seriously threatening public safety and state security. Hailbronner contends that individuals performing such acts ‘have given up their attachment to a community by attacking the very fundament of that community, not by merely violating its internal rules of public order’. However, this line-drawing exercise seems to be quite difficult: every crime may cause insecurity, threaten public order, and prevent democratic societies from functioning properly; citizens (and decision-makers, including those entrusted with citizenship revocation) will diverge, based on their ideological biases, as to whether particular crimes cross Hailbronner’s threshold. For instance, did the perpetrators of the Brighton hotel bombing cross the threshold in light of the potential ramifications of Thatcher’s assassination for the stability of the United Kingdom? If so, would a person financing such an attack qualify, too? 

Nevertheless, perhaps a ‘core’ case can be identified, such as a criminal conviction for treason. One of the constituent elements of such acts is often that they are committed by citizens qua citizens. For instance, Lord Haw Haw (William Joyce) could be convicted of espionage for Germany in the Second World War because he possessed British nationality; he unsuccessfully argued that he did not owe loyalty to the Crown. If the basis for Joyce’s conviction was that his crimes against the state were committed as a British national, then disowning Joyce ex post facto seems incoherent: the state must reject the claim that treasonous acts amount to renunciation of citizenship, because that would disable the state from prosecuting the perpetrator for treason. (for an analogous argument concerning the legitimacy of disenfranchisement of convicted adult citizens, see my article).  

The second type of case involves crimes (including crimes defined as ‘terrorism’ under international treaties or domestic law) committed by a citizen of state A against individuals or institutions in state B. The fact that the person who has committed such crimes holds the citizenship of state A seems incidental. Consider the attack on the Jewish museum of Belgium in Brussels on 24 May 2014, which is likely to have been carried out by a French national affiliated with ISIL. ISIL has been designated as a terrorist organisation by the EU, of which France is a member, as well as by the UN. Were France to revoke the citizenship of this member of an internationally designated terrorist organisation, it would be severing legal relations with a citizen even though the citizen’s actions were not directed specifically towards the French state, its institutions, or its population. This seems rather tenuous.  

Joppke argued that ‘international terrorists are not criminals but warriors’. But the state exercises its sovereign powers vis-à-vis ‘international terrorists’ qua citizens. The fact that such persons commit acts that are of an international character does not make it more plausible for their state of nationality to legally disown them as a result. Hailbronner argues that ‘[w]hat makes international terrorism so distinctive is … also its relevance for discontinuance of that special relationship established by citizenship.’ I am not quite sure why engagement in international terrorism (such as the ISIL-sponsored attack on the Jewish museum) necessarily or even plausibly indicate that a citizenship bond has been severed by the terrorist. This seems to conflate the fact that their state of nationality perceives (and rightly so) the terrorist’s act as heinous with a direct effect on that state.

The third type of case concerns acts which are committed by a citizen in the name of the Ancien Régime. Following political transformation, the state wishes to disassociate itself from such past acts by dissociating itself from the perpetrators. As Bauböck rightly notes, Hitler’s posthumous denationalisation by either Germany or Austria would have been considered ‘a denial of responsibility for his crimes and their consequences’. In addition to the revocation’s outward-looking dimension (towards the international community), it has an inward-looking dimension too. When Augusto Pinochet stood trial in in 2004, he was charged with crimes committed by him as head of the military junta which ruled Chile after the 1973 coup. He died in 2006 before the conclusion of his trial. Let’s imagine that Pinochet had another (nominal) citizenship, and that his conviction would have resulted in his denationalisation. This would have seemed, rightly, as an attempt to undermine the fact that these acts were committed in the name of the Chilean state.

Paskalev asserted that, ironically, the ‘softness of citizenship revocation makes it appear quite inappropriate for the case of terrorists’. However, even if (some) terrorists may be blasé about losing their citizenship, we ought to be concerned about states’ eagerness to wash their hands of them.



Our epoch's little banishments

By Saskia Sassen, Columbia University


I arrive late to this discussion, to these excellent pieces that cover much ground... not much left to cover. For the sake of debate and commentary, rather than scholarly analysis, let me throw into the discussion what is no more than a little wrench. 

Denationalisation is an ambiguous concept. This discussion has given it one sharp meaning: being stripped of one's nationality and thrown out of one's country. In my own work I have used it to capture more ambiguous meanings, thereby giving it the status of a variable that can be applied to a range of domains, not only citizenship.(1)

Thus, I see denationalisation at work when, beginning in the 1980s, global firms pushed for and got most national governments to institute deregulations and privatisations so as to maximise their access into any national economy. It meant that states had to denationalise key elements of the legal framing (i.e. protections) they had long offered their own firms, markets, investors. One might say that in doing so, these states instituted a partial 'banishment' of their own national firms from a legal framing that granted these national firms exclusive privileges/rights. This is a form of banishment that does not entail a physical departure from a country's territory. It only entails a loss of particular exclusive rights and protections. We can conceive of it as a kind of micro-banishment.

Similarly, I would argue that such internal micro-banishments are also present in the decisions of many national states, beginning in the 1980s and onwards, to eliminate a few rights here and there that their citizens may long have had. Examples for the U.S. are, among several others, Clinton's 1996 Illegal Immigration Reform and Immigrant Responsibility Act which took away the rights of citizens to bring legal action against the INS in lower courts; or when credit card companies obtained the right to pursue payment even if a household had declared bankruptcy – a right so abusive it eventually got cancelled. We might argue that in these cases, citizens experienced a partial banishment from specific rights (even as some new rights were also attained, notably gay marriage). The better language to describe these losses may be what Audrey Macklin refers to elsewhere as civil death.(2)

Current examples for the gains of rights for global firms and the loss of protections for national firms and workers can be found in some of the clauses of both the Transpacific and the Transatlantic Trade Partnerships.

Long before we get to the dramatic figure of the terrorist, where the debate about banishment turns clearly pro or contra, I see a range of micro-banishments that take place deep inside national territory. If I wanted to give this image an extreme character, I would say that in today's interaction prone world (see, for instance, the earlier behind closed-doors negotiations between Iran and the U.S., or, for a period, between the U.S. and the Taliban) there is no more terra nullius for banishment.

If I were to use the term "banishment," I would want to use its conceptual power to get at the multiple little banishments that happen inside our countries and that often entail a move into systemic invisibility – the loss of rights as an event that produces its own partial, or specialised, erasure. I refer to these micro-banishments as expulsions, a term I intend as radically different from the more common term "exclusion," which refers to a condition internal to a system, such as discrimination.(3) I conceive of such expulsions as a systemic capability, clearly a use of the term capability that diverges from the common use which marks it as a positive. Thus micro-banishments can be seen as a profoundly negative systemic capability that is far more widespread than our current categories of analysis allow us to see.(4) 

To conclude I would like to return to Audrey Macklin's argument. 

I agree with Audrey Macklin's proposition that citizens should not be banished even when they engage in terrorist attacks on their own country. I share her concern with the importance of protecting a robust form of  citizenship. But I do so partly also from a transversal and dystopian perspective that may have little to do with the rationale put forth by Macklin. Let me clarify. It is not only terrorists that are destructive and attack the innocent; it is also predatory actors of all sorts –corporate firms that exploit workers worldwide, financial speculators, abusive prison systems. Further I agree with Macklin that a country should develop the needed internal instruments to deal with terrorists rather than banish them. But again, I would take this beyond terrorists who are citizens, and include the types of predatory actors I refer to above.

Beyond all of this, I am above all concerned with the larger history in the making that I refer to earlier in this short text. This larger history is shaping an epochal condition that takes me away from prioritising banishment as loss of citizenship and of the right to live in one's country as discussed in this forum. 

Briefly put, I would argue that the conceptual locus of the category banishment in today's world is not banishment in the historical sense of the term, but a new kind of banishment. It is one predicated on the formation of geographies of privilege and disadvantage that cut across the divides of our modernity – East-West, North-South. The formation of such geographies includes a partial disassembling of the modern national territorial project, one aspiring (and dependent on) national unity, whether actual or idealised. This then also means that there is a weakening of the explanatory power of the nation-based encasements of membership (for citizens, for firms, for political systems) that have marked our modernity. The micro-banishments I refer to are part of emergent (and proliferating!) geographies of disadvantage (for citizens, firms, districts) internal to a country.  



(1) See chs 4,5, and 6 in Saskia Sassen. Territory, Authority, Rights: From Medieval to Global Assemblages, Princeton University Press 2006; 2nd ed. 2008.

(2) Audrey Macklin, Citizenship Revocation, the Privilege to Have Rights and the Production of the Alien, (2014) 40 Queen’s LJ 1-54 at 8.

(3) Saskia Sassen. Expulsions. Brutality and Complexity in the Global Economy, Harvard University Press, Cambridge, Mass. 2014.


(4) This also raises the possibility of an obverse condition: that the tissue constructed via the recurrence of micro-banishments inside a nation-state could, with time, become the tissue for a claim to transnational citizenship. Could it be that as citizens experience the limits of national citizenship, transversal notions of membership become more plausible? I am thinking here of substantive conditions for transnational citizenship, not just ideational one.  


Deprivation of citizenship: is there an issue of EU law?, by Jo Shaw

By Jo Shaw (University of Edinburgh and EUDO-CITIZENSHIP co-director)


The purpose of this short intervention in the debate on The Return of Banishment initiated by Audrey Macklin, where the pros and cons of various forms of deprivation policies pursued by, or sought by, liberal states have been fully debated, is to add an element of EU law. Specifically, in the light of the judgments of the European Court of Justice in Rottmann and Ruiz Zambrano, how – if at all – are Member States’ law and procedures on involuntary loss of citizenship affected by EU law, given that the primary competence to determine the rules on the acquisition and loss of citizenship remains with Member States? We have yet to hear conclusively, but well informedobservers who followed the UK Supreme Court hearing in the case of B2 (Pham) v SSHD concerned with the UK’s rather extensive deprivation powers and the issue of statelessness have indicated that they think it likely that the Supreme Court will now make a reference to the Court of Justice. It seems that the judges will ask the CJEU if it really meant what it said when it decided the case of Rottmann. B2 (Pham), like the earlier cases of G1 (discussed below) and as well as the case of Al Jedda, a former Iraqi citizen who has twice been stripped of his UK citizenship as well as spending time in military detention in Iraq, all concern naturalised citizens who are suspected of some form of terrorist involvement, but against none of whom criminal proceedings have been brought in the UK.

We are likely, therefore, to be in a phase of further legal development – initially in iteration between the UK courts and the Court of Justice, but with implications for all of the Member States as quite a number of states have started to look closely at using expatriation measures in order to combat radicalisation and terrorist threats, even if many judge this approach to be ill-advised and inappropriate.

I will explain briefly what the issues are. The Rottmann case was the subject of an earlier forum debate on the EUDO Citizenship website. Rottmann was a case of loss of citizenship conferred by naturalisation, after it came to light that the naturalisation had been obtained by fraud. In this case, Rottmann, an Austrian citizen, had failed to reveal that he had been the subject of unconcluded criminal proceedings in Austria when seeking naturalisation in Germany. Rottmann raised issues of EU law in his appeal against the deprivation decision before the German administrative courts, which led to a reference to the Court of Justice. He pointed out that having obtained German citizenship he lost Austrian citizenship, by operation of law. Thus, if he were deprived of German citizenship he would be stateless, and – furthermore – he would have lost his EU citizenship. One issue that had been raised – and which caught the attention of Advocate General Maduro in his Opinion – was whether this was a ‘wholly internal situation’ – i.e. a German court reviewing a decision of a German public authority regarding a German citizen. In that sense, it could be said, EU citizenship was not engaged at all. In response, the Court repeated its standard formulation when dealing with matters which fall outside the competence of the EU and its legislature. It reminded us that EU cannot adopt measures with regard to national citizenship, but none the less while national competence remains intact, it must be exercised ‘with due regard’ to the requirements of EU law in situations covered by EU law. Specifically, in this case, said the Court:

‘It is clear that the situation of a citizen of the Union who, like the applicant in the main proceedings, is faced with a decision withdrawing his naturalisation, adopted by the authorities of one Member State, and placing him, after he has lost the nationality of another Member State that he originally possessed, in a position capable of causing him to lose the status conferred by Article 17 EC [i.e. Union citizenship] and the rights attaching thereto falls, by reason of its nature and its consequences, within the ambit of European Union law’ (para. 42 of the judgment).

The Court went on to recognise that states may have legitimate reasons to withdraw citizenship, but it is worth noting that the Court of Justice does not, in this paragraph, focus on statelessness, but rather on the loss of the rights specific to EU law. In other words, this can be seen as an EU-specific reason for requiring the testing of any decision to withdraw citizenship against – as the Court went on to hold – a standard of proportionality. Factors to be borne in mind in assessing the proportionality of the withdrawal decision included the gravity of the original offence or deception, lapse of time, the impact on the subject of the decision and their family, the possibility of recovering the original citizenship lost at the time of naturalisation, and the availability of other less severe measures than withdrawal.

While some have suggested that the essence of Rottmann lay in the way that the claimant is strung across between the national citizenship laws of two EU Member States, one at least of which claims exclusivity and thus operates an automatic rule of withdrawal in the event that a citizen acquires the citizenship of another state, the point about loss of the benefits of EU citizenship as a freestanding principle of EU law without regard to prior movement from one Member State to another was given a further boost in the case of Ruiz Zambrano. In that case, the EU citizens threatened with losing their rights of citizenship were the children of the claimant, who were born in Belgium and who had acquired Belgian, and thus EU, citizenship at birth. Meanwhile, through a combination of circumstances their Colombian citizen father had not regularised his situation in Belgium (or had perhaps been prevented from doing so by a series of delays perpetrated by the Belgian authorities in relation to his case). Because the refusal of a residence permit for Ruiz Zambrano and his wife would, in effect, have meant that the EU citizen children would have been obliged to leave, with their parents, the territory of the EU and thus would not have been able to avail themselves of their rights as EU citizens (notably the right of free movement which they had not yet exercised, but which they might exercise in the future), the Court concluded that a Member State could not refuse to grant either a residence permit or indeed a work permit. The test that the Court articulated was whether the measure taken in relation to a third country national upon whom the EU citizen children were dependent was whether it would make them unable to exercise ‘the substance of their rights’ as citizens of the EU.

Neither Rottmann nor – in particular – Ruiz Zambrano have been met with unalloyed enthusiasm at the national level. It goes beyond the scope of this short comment to discuss how and why Member States and indeed their courts might react to challenging judgments of the Court of Justice that appear to extend the scope of EU law and, in particular, the scope of EU citizenship.(1) That said, there is no evidence to suggest that, thus far, Rottmann has had a significant or disruptive effect on national citizenship laws.(2)

The UK is one of the few states where Rottmann has thus far been discussed in national cases, but – until the case of B2 (Pham) which is before the Supreme Court – the limit of consideration had been a rather dismissive swipe at the Court of Justice taken by Lord Justice Laws in the Court of Appeal in the case of G1 v SSHD.(3)  Laws LJ sceptically asked ‘[u]pon what principled basis, therefore, should the grant or withdrawal of State citizenship be qualified by an obligation to "have due regard" to the law of the European Union?’ (para. 38), given that the grant and withdrawal of citizenship remains a matter of Member State competence.

The Supreme Court refused to give leave to appeal to the applicant in G1, but perhaps it was only a matter of time, given the salience of deprivation of citizenship in the UK at the present time, before it had to grasp the nettle of considering not only the meaning of statelessness in the context of the then applicable UK law (this having moved on somewhat since that time, as Gibney’s contribution to the forum highlights) but also the possible applicability of EU law as a restraint upon executive freedom, and as a standard which UK courts, in exercising their review function, would need to uphold. Hence the appellant in B2 has been given leave to appeal, with perhaps a reference to the Court of Justice still to come. 

As the discussion by Simon Cox, a lawyer working with the Open Society Institute which intervened in this case, has made clear, it seems quite likely that if the applicability of EU law as a frame of reference against which UK deprivation legislation needs to be judged is duly established by the Court of Justice and accepted by the Supreme Court, then the proportionality standards which need to be applied by UK courts exercising their review function may differ from those otherwise applicable within UK public law. The key issue seems likely to surround the putative autonomy of EU citizenship: is there a freestanding EU law related concern with citizenship stripping, namely the loss of EU citizenship rights, which goes beyond the issue of statelessness? Rottmann seemed to suggest there was, but this is the issue on which the Supreme Court may probe the CJEU further. It should be noted that there may also be higher standards of disclosure of otherwise secret evidence, following the judgment of the Court of Justice in the ZZ case, if the applicability of EU law is accepted. 

Finally, it should be pointed out that the OSI interest in the case is not directly with the Rottmann point, but concerns the definition of statelessness, which, they argue also has an EU element and should have a common EU level definition to which Member States are obliged to adhere. This call stems from the fear that in its earlier judgment in B2 (Pham) the Court of Appeal created significant difficulties when it resolved that B2 was not to be judged as de iure stateless, once deprived of UK citizenship, because although the Vietnamese government indicated they did not recognise him as a citizen, it was clear that this was unlawful under Vietnamese law. The UK courts, said the Court of Appeal, were bound by the rule of law. Therefore, they could not recognise an unlawful act of the Vietnamese government. This seems to be peculiarly Kafka-esque reasoning and the OSI, given its investment in the campaign against statelessness ongoing under the leadership of the UNHCR, would be concerned if this reasoning were to take hold in the UK, which is bound to have further cases coming before the courts, given the remarkable rate at which the state is now expatriating its citizens on grounds that this is conducive to the public good.



(1) M. Blauberger, ‘With Luxembourg in mind ... the remaking of national policies in the face of ECJ jurisprudence’, (2012) 19 Journal of European Public Policy 109-126, M. Blauberger, ‘National Responses to European Court Jurisprudence’, (2014) 37 West European Politics 457-474, S. Schmidt, ‘Judicial Europeanisation: The Case of Zambrano in Ireland’, (2014) 37 West European Politics 769-785.

(2) See N. Nic Shuibhne and J. Shaw, ‘General Report’, in U. Neergaard, C. Jacqueson and N. Holst-Christensen (eds.), Union Citizenship: Development, Impact and Challenges, The XXVI FIDE Congress in Copenhagen, 2014, Congress Publications Vol. 2, DJØF Publishing, Copenhagen) at p154-155.

 (3) G1 v Secretary of State [2012] EWCA Civ 867.



On producing the alien within: A reply

by Audrey Macklin  

Shortly after the last contributor posted a comment on this forum, reports of the Charlie Hebdo attacks erupted in the media. The assailants were two French brothers (Cherif and Siad Kouachi) who claimed affiliation to Al Qaeda in Yemen.  Hours later, an associate (Amiday Coulibaly) killed a police officer, then rampaged through a kosher Hyper Cacher supermarket and murdered four hostages. All three men were slain two days later in confrontations with French police and security. That same day, the notorious ‘Finsbury Mosque cleric’, British national Abu Hamza, was sentenced to life in prison by a US court for terrorism related crimes. Most recently, the French Conseil Constitutionnel upheld a law permitting denaturalisation of dual-national French citizens convicted of terrorist offences.[1] One cannot but wonder whether the Charlie Hebdo and Hyper Cacher attacks cast a long shadow over the Conseil Constitutionnel’s deliberations, even though all three men were French by birth and therefore outside the purview of the denaturalisation law.  

The horrific deeds of the French perpetrators struck at the heart of liberal democratic values: freedom of expression and religious tolerance. States understandably seek new and better tools to prevent future atrocities; the impulse toward retribution at such moments seems hard to resist. Do these attacks make the case for citizenship revocation? I remain skeptical that citizenship revocation advances the objective of protecting liberal democracies, or that pursuit of unalloyed retribution is an objective worthy of liberal democracies.  

Defenders of citizenship stripping offer a mix of instrumental and non-instrumental justifications, but Kay Hailbronner, Christian Joppke and Peter Schuck lean toward the latter more than the former. Despite its rejection by the US Supreme Court over fifty years ago, both Hailbronner and Joppke revert to the legal fiction of constructive renunciation and insist that certain conduct communicates an irrefutable intention of terrorists to renounce their own citizenship. Schuck revises the fiction by acknowledging that perpetrators may not actually wish to renounce citizenship, but then discounts an intention to maintain citizenship for ‘tactical and cynical’ purposes. But however attractive the fiction of constructive renunciation, it does not become truer with repetition, or with the passage of time, or by writing new characters into the narrative. Citizenship revocation for misconduct while a citizen is not chosen by the citizen; it is inflicted by the state.  

Joppke explains that Germany would have been wrong to regard members of the RAF as menacing enough to warrant denationalisation, and I suspect he would also condemn the United States denaturalisation of Communist citizens in the twentieth century as hysterical overreaction. But he remains confident that one can transcend historic patterns of panic-induced political myopia and he arrives at the conclusion that Islamic terrorists are uniquely suitable for citizenship revocation. Peter Schuck contends that citizenship revocation, when employed judiciously against terrorists, strengthens the value of citizenship itself. Kay Hailbronner adds that my arguments do not address the illegality of citizenship revocation under international or constitutional law, but rather proceed from unarticulated notions of legitimacy and morality. Space does not permit a proper reply to the last criticism. Readers are invited to read my published article on citizenship revocation in the Queen’s Law Journal,[2] which addresses citizenship revocation for misconduct under international and constitutional law.  

Consider citizenship revocation in relation to the goal of bringing perpetrators to justice. As I mentioned in my kickoff text, fear of citizenship revocation is unlikely to deter those bent on martyrdom, and the deaths of the Kaouchi brothers and Coulibaly seem to demonstrate that. As for Abu Hamza, it is worth noting that the UK did attempt to strip him of citizenship. It was thwarted because deprivation would have rendered the Egyptian-born cleric stateless. But the fact that Abu Hamza remained in the UK as a UK citizen made him available for extradition to face charges in the United States, where he was tried, convicted and sentenced to life imprisonment for terrorism offences after an open and fair trial. Had he been stripped of UK citizenship and expelled to Egypt, he would never have faced justice in a US court, or anywhere for that matter.[3]  I take the view that prosecution, trial and conviction are a better outcome than the alternatives. Various states have begun revoking passports of citizens allegedly bound for IS camps in Syria and Iraq. Restricting exit in this manner is only available in relation to citizens. Stripping citizenship permits states to shed their duty and responsibility toward nationals; it also deprives them of the authority to subject them to criminal prosecution and thereby make a tangible contribution to bringing terrorists to justice under the rule of law.  

Schuck, along with Hailbronner and Joppke, concede that existing practices of citizenship revocation breach basic norms of fairness. They regard these flaws as contingent defects that are severable from the abstract question of whether citizenship revocation for misconduct can be justified. I find the attempt to segregate theory from practice unconvincing in this context, and Matthew Gibney’s intervention highlights the way in which attempts by the judiciary to hold the state to requirements of legality simply breed more tactics of evasion. A chronic failure of a state practice to comply with fundamental norms of legality across time and space invites the inference that there is something about what the state is endeavouring to do that ineluctably and incorrigibly perverts the process of how it does it.[4] A fair process leading to banishment, like a fair process culminating in the death penalty, can only ever operate as a mirage that legitimates ongoing practices that will – inevitably and necessarily – fail to meet basic norms associated with the rule of law.  

This leaves a defence of citizenship revocation that does not depend on practicality or utility, but instead rests on the insistence that revocation is just and fitting punishment of those who abuse the privilege of citizenship. I argue that when citizenship becomes revocable for misconduct, citizenship as legal status is demoted from right to privilege. This is a specifically legal argument about the juridical fragility of a privilege compared to a right. Joppke’s comment that citizenship in western states is a privilege because citizenship delivers so little to citizens of most states is a non-sequitur. I may feel privileged to be a Canadian citizen and to benefit from the rights, entitlements and security of Canadian citizenship, but that does not make citizenship as such a privilege. And it would be peculiar indeed if only liberal democratic states that guarantee robust citizenship were entitled to revoke citizenship qua privilege, while poor and dysfunctional states that deliver only a feeble and impotent citizenship, were not. Schuck maintains that citizenship revocation, properly wielded, does not weaken citizenship, but can actually ”strengthen citizenship by reaffirming the conditions on which it is based.” I am not sure exactly what this means but his subsequent invocation of capital punishment does alert one to the way in which the rhetorical form of his claim echoes similar assertions by death-penalty advocates. Indeed, if one is convinced that the value of life is strengthened when the state executes a murderer, perhaps one will also be persuaded that citizenship is strengthened when the state denationalises a terrorist. The corollary also applies: If one is not attracted by the first proposition, perhaps one should resist being seduced by the second.  

Jo Shaw’s insightful intervention about the implications of denationalisation for EU citizenship brings to the discussion the important issue of proportionality, a matter Hailbronner also addresses briefly. Stepping back from the specificities of EU citizenship, a proportionality inquiry into citizenship deprivation directs us to the question of whether the state can achieve its objectives through less rights-infringing means than the impugned law. If one takes seriously the injunction against statelessness, the answer must surely be yes. However one frames the goals and purposes of citizenship deprivation, it remains true that states can and do deploy other means to address, contain and denounce threats to national security from mono-nationals.[5] They must do so because denationalisation is not a legal option, yet no state will be heard to say that it is disabled from protecting the nation adequately because it cannot denationalise mono-citizens.  

Schuck proclaims that a state is “powerless to protect itself and its people from imminent, existential threats”, if denied access to denationalisation as a weapon. Not only does this ignore the resources currently available to states, it dramatically overestimates what citizenship revocation would add to the arsenal. Unless a state could mount evidence showing that dual citizens pose a qualitatively different and graver threat to national security than mono-nationals, I doubt that citizenship revocation for some citizens (but not others) could survive a rigorous proportionality analysis.  

Rainer Bauböck correctly and helpfully reminds us that what is at issue is citizenship as legal status. Citizenship signifies membership, but beyond that general descriptor, citizenship inhabits multiple registers across many disciplines. Among liberal states, equality of status and security of that status are two defining features of legal citizenship. The former speaks to citizenship’s internal dimension by ensuring that all citizens of a state are recognised and treated as equal to one another. The latter speaks to citizenship’s external dimension. In functional terms, nationality not only protects individuals from what Michael Walzer calls the ‘infinite precarity’ of statelessness, it also serves an international system of sovereign states in ensuring that at least one mailing address is affixed to every individual for purposes of state responsibility and deportation.  

Apart from Joppke, all contributors accept statelessness as a constraint on citizenship stripping. In the world as we know it, where all habitable space is already assigned to some state, the claim that a citizen, by virtue of his or her conduct, does not belong to this state must entail the claim that the person does belong to that state.[6] This exposes two related problems for conduct-based revocation. The first is that the people whom Joppke depicts as appropriate targets of denationalisation are not merely enemies of a particular state or government. On his view, they ‘explicitly posit themselves outside the political community of the nation-state’. In other words, they repudiate citizenship as such or, if one prefers, pose as ‘citizens’ of a non-state entity that every other state in the world rightly regards as deeply threatening and inimical to their security. One expects that they will be as ‘tactical and cynical’ in their connection to one citizenship as to another. The Canadian citizenship revocation law adopts this model of the global terrorist by making conviction for a terrorist-related offence in another country grounds for revoking Canadian citizenship. If another state regards a Canadian citizen as a terrorist, that is reason enough for Canada to conclude that his citizenship connection to Canada is inauthentic and warrants amputation.  

One problem with Joppke’s argument is that it is self-defeating. If terrorists disavow citizenship as such, and are indeed hostis humani generi (enemies of all humanity), the same facts that would allow Joppke to pronounce that the Kouachis (for example) did not really belong to France must also yield the conclusion that they did not belong to any other state either. As a practical matter, if one state declares that formal possession of legal status is normatively insufficient to attach the terrorist to that state, it can hardly press the claim that legal status is sufficient to attach him to another state.  

Joppke mocks Peter Spiro for making the sensible observation that neither al Qaeda nor Islamic State are states, which means that they are not deportation destinations. Hailbronner abets Joppke by musing about whether IS’ military control over patches of land in the midst of violent conflict could be ratcheted up into something approximating statehood. If this is meant to hint at a viable legal option for where to dispose of otherwise stateless citizens, one might as well explore the equally plausible (from a legal perspective) option of launching them into space to orbit the globe aboard some inter-galactic Flying Dutchmen.[7] Alternatively, perhaps we are meant to shrug off as a convenient fact that powerful states can denationalise their citizens while they are abroad in conflict zones and make them some other [failing] state’s problem.  

Bauböck’s contribution directs one to another dimension of belonging, which reveals the second problem with Joppke’s approach. Citizenship stripping’s revival traces back to the anxiety about so-called ‘home-grown’ terrorists who, unlike the iconic foreign menace, actually possess citizenship by birth. Revoking citizenship enables the state to recast them as the alien within, in order to then cast them out. Denationalisation serves the narrative of terrorism as always and essentially foreign to the body politic by literally transforming the citizen-terrorist into the foreigner. But the very term ‘home-grown’ refutes the premise. The Kaouchi brothers were French citizens. They were orphaned as children and raised as wards of the French state. It is difficult to see them other than as products of French society, even if the ideology that seized them originated elsewhere. Indeed, any viable anti-terrorism strategy must attend carefully and critically to the local conditions that produce a descent into disaffection, hatred and violence – whether of the Islamist, neo-Nazi or any other variety. The French assailants may have been alienated from France, but there is no state to which they belonged more.[8]  

Ultimately, arguments about citizenship revocation turn on underlying conceptions of what citizenship is for, and expectations about what citizenship as legal status can achieve. Citizenship signifies membership, but beyond that general descriptor, citizenship inhabits multiple registers across many disciplines which are not reducible to or fully commensurate with one another. Citizenship as legal status is powerful because it carries the force of law, but also limited in what it can achieve for precisely the same reason. It is enabled and constrained because it is citizenship law and because it is citizenship law.  

States can and do use law to promote and endorse commitment, patriotism and active citizenship. They do it through public education, programmes for social inclusion, support and assistance, support for the arts and recreation, and other policies that build solidarity and encourage ‘good citizenship’. These various spheres of public activity are enabled through legal frameworks, and so law plays an important role here. Citizenship law’s chief constructive contribution lies in imposing (reasonable) requirements for naturalisation (such as residence and language acquisition) that genuinely facilitate integration and commitment to the national community.  

The state must also be concerned about ‘bad citizenship’ and it falls to the criminal justice and national security regimes to address the most egregious conduct that endangers or harms the national community. To conclude that contemporary citizenship law is ill-suited to advancing punitive goals does not deny that some people are very bad citizens, or that law plays a crucial role in addressing that fact. It simply opposes the recruitment of citizenship law to punish bad citizens by demoting them to non-citizens[9]. A man who attacks his mother may be a terrible son who deserves to be prosecuted for his crime, but it is not the job of family law to disclaim him as the son of his mother. Citizenship law is not criminal law. Nor is it national security law. Nor should it be rigged to open up a trap door that shunts citizens to immigration law.  

Accounting for citizenship status’ specific legal character also guides us toward what law can (and cannot) achieve. A number of plausible accounts of citizenship’s normative foundation circulate in political theory. They typically involve some idea of commitment or allegiance, whether to the state, the constitution, or democratic self-government. I do not here express a preference among them, but rather observe that they tend to focus on the internal relationship between state and citizen, and the grounds upon which the relationship may be properly said to have ruptured. They do not attend to the external dimension of legal citizenship, namely the role of nationality in stabilising the international filing system for humanity, and they do not furnish a satisfactory normative explanation for why the ‘bad citizen’ should be assigned to another state.  

Citizenship law cannot subject to legal regulation the myriad values, practices and aspirations ascribed to citizenship-as-belonging. This is unsurprising: Citizenship status enfranchises citizens above the age of majority, but there is no legal compulsion to vote (except in Australia. Belgium, Brazil and a few other states) and citizenship law does not purport to penalise those who never exercise their right or duty of active citizenship. Citizenship law does not purport to regulate access to most types of civil and social citizenship (in Marshallian terms).  

Nevertheless, defenders of revocation insist that citizenship law can and should regulate ‘loyalty and allegiance’ of citizens. The criminal law can punish people for intentionally committing wrongful acts, including treason, murder, and all other forms of horrific violence that concern us here. Some assailants may openly express contempt for their country of citizenship, while others (like the Ottawa shooter Joppke cites) display a messy history of mental illness and petty criminality preceding recent conversion to Islam. The putative value added by citizenship revocation is precisely that it makes lack of allegiance and loyalty the central element in defining crimes against citizenship. But to paraphrase Aldous Huxley, loyalty and allegiance are like happiness. They are byproducts of other activities. Fostering love of country is a valid aspiration of states and worth cultivating. But it cannot be manufactured by the carrot of a citizenship oath (as Joppke has elsewhere acknowledged), nor will it be conjured by the stick of revocation. Law is not adept at producing sentiment on command.  

Space constraints have led me to focus on those submissions that directly challenge my own position, and I have not responded to the cogent, provocative and creative insights offered by so many contributors. My own thinking has been deepened and provoked by them, for which I express gratitude and appreciation. I admit that I took as my remit citizenship revocation only in the literal, legal sense. I also acknowledge the criticism that confining my focus to citizenship revocation does not pay due regard to the claim that deportation of non-citizens may also constitute banishment in some circumstances, with attendant human rights implications. I hope that nothing I have said here gives the appearance of foreclosing or prejudging broader or different conceptions of banishment. There is always more to be said, and much to be done.

[1] The law permits denaturalisation of dual nationals who commit terrorism offences within fifteen years of naturalisation. http://www.theguardian.com/world/2015/jan/23/moroccan-born-man-jailed-terror-lose-french-nationality-sahnouni

[2] Audrey Macklin, “Citizenship Revocation, the Privilege to Have Rights and the Production of the Alien”, (2014) 40 Queen’s Law Journal: 1-54

[3] Egypt does not extradite its nationals, and the Egyptian criminal justice system does not inspire confidence in its capacity to administer justice.

[4] This point draws on the insight of legal theorist Lon Fuller. He admitted that his principles of legality were formal in the sense that they did not stipulate any substantive moral content to law. But he also maintained that legal systems that were intent enacting morally repugnant laws would be hard pressed to reconcile achievement of those objectives with compliance with principles of legality. I extend Fuller’s intuition to suggest that a chronic pattern of non-compliance with principles of legality in relation to a particular law supports an intuition that the law is normatively defective in substance.

[5] States can and do use the criminal law to prosecute people for terrorist related offences committed at home and abroad. Expanded police powers of investigation and surveillance enable detection. Passport confiscation that prevents travel to conflict zones restrains a right of citizenship (exit), and some states prosecute citizens who participation in combat abroad when the return. Some states also restrict the right of citizens abroad to re-enter in the name of national security. I consider this less defensible as a matter of law, both in relation to the excluded citizen and other affected states but cannot develop that argument here.

[6] One could, I suppose, imagine a world where states re-appropriate statelessness in order to resurrect the figure of the global legal outcast (hostis humani, or perhaps homo sacer).  Stripped of law’s protection, this global outlaw could be killed or punished with impunity.  I will set this aside this possibility, and I am unsure if this is what Joppke has in mind.

[7] It seems more likely that the UK will simply continue the practice of depriving citizens of their UK citizenship while abroad, now accompanied with a statement that the Home Secretary believes that target can obtain citizenship elsewhere. Even if the person does not, in fact, have access to another citizenship, the individual’s physical location outside the UK and inside another state (to which they may have no legal relationship) will impose insuperable hurdles on challenging the decision or compelling the UK to repatriate him.  

[8] One might object that the sample set is too limited: After all, there are dual citizens (especially those who naturalised as adults) who might reasonably be understood as more connected to their country of origin. A short answer is that even if true, it would be a clear conflict of interest to let one state of citizenship make that determination. A fuller answer, which lies beyond the scope of this intervention, would explain why this type of calculus is inimical to the security that distinguishes citizenship from other statuses.

[9] The various legal strategies currently in use to detect, deter, prevent and respond to terrorism can and do fail, sometimes tragically and spectacularly. Is this because states have not arrogated to themselves sufficient coercive powers, or do inadequate human, technical and financial resources explain more about operational failure?


Read more on citizenship deprivation on the website of the ILEC project


European citizenship is the fundamental status of EU Member State nationals. What, then, are the implications when a state deprives a European citizen of his or her nationality? A new project funded by the European Commission’s DG Justice and coordinated by the Centre for European Policy Studies (CEPS) in collaboration with the University of Maastricht and EUDO CITIZENSHIP aims to investigate this under-researched field. Since 2012, ILEC (Involuntary Loss of European Citizenship: Exchanging Knowledge and Identifying Guidelines for Europe) has carried out a comparative study of the law and practices governing loss of nationality across the 28 Member States. A key question it seeks to answer is what impact the development of European citizenship is having on national competences covering the acquisition and loss of nationality in light of increasing jurisprudence from the Court of Justice and European Court of Human Rights. Consult the website of the ILEC project


Independence Referendums: Who should vote and who should be offered citizenship?


Kick-off contribution

by Ruvi Ziegler


Imagine that you are a Scottish-born recent graduate, fortunate enough (certainly in today’s economy) to be offered a job in Amsterdam, working for a multinational company. You relocated from Edinburgh to Amsterdam in 2012, and most of your family continues to reside in Scotland. You take a great interest in the Scottish independence referendum on 18 September 2014, when voters will be asked whether Scotland should ‘become an independent country’, and are concerned about its ramifications. If Scotland votes to stay part of the UK, you intend to vote from abroad in the UK general election in May 2015. The Scottish government pronounces that, if Scotland becomes independent, you will be automatically considered a Scottish citizen. Yet, in the referendum itself, you will not have your say: Scottish-born expatriates are excluded. In this kick-off contribution, I make the case for resemblance between the category of persons entitled to participate in an independence referendum and the initial citizen-body of a new state created by such a referendum.

The scope of my claim concerns only independence referendums which may result in ‘Succession of States’. This term is defined in Article 2 of the International Law Commission’s Draft Articles on Nationality of Natural Persons in Relation to the Succession of States as ‘the replacement of one state by another in the responsibility for the international relations of territory’. 

Independence referendums may result in the dissolution of an existing state, namely ‘[w]hen a State dissolves and ceases to exist and the various parts of the territory of the predecessor State form two or more successor States’ (Article 22); see, for instance, the breakup of Czechoslovakia (albeit without referendums). More commonly, perhaps, independence referendums may lead to the separation of part(s) of the territory of a predecessor state while the latter state continues to exist; recent examples include the two (unsuccessful) independence referendums in Quebec, as well as the cases of the Republic of South Sudan, and Timor-Leste. Scotland is the most pressing case thought to conform to the latter definition (see e.g. James Crawford and Alan Boyle, Referendum on the Independence of Scotland: International Law Aspects and House of Lords, Constitutional Committee, Scottish Independence: Constitutional Implications of the Referendum).

Article 1 of the ILC Draft Articles stipulates that ‘[e]very individual who, on the date of the succession of States, had the nationality of the predecessor State, irrespective of the mode of acquisition of that nationality, has the right to the nationality of at least one of the States concerned [predecessor and/or successor]’. In the event of dissolution, all citizens of a predecessor state are affected by a successful referendum, whereas in separation cases the legal status of many citizens of a predecessor state may not be affected.

A previous debate in this forum concerning electoral rights of ‘second country nationals’ in their EU state of residence provides a helpful context for normative questions posed in relation to the link between citizenship and the franchise. Notably, this debate concerns political membership and electoral participation in an existing political unit. Independence referendums are different: they may create new political entities, and require the attribution (as per the terminology employed by the ILC Draft Articles) or offer of citizenship to individuals.

Let’s leave aside for now the moral or political legitimacy of particular independence referendums and indeed the legality of referendums under particular national laws (though other contributors may wish to engage with these issues). Instead, following Article 3 of the ILC Draft Articles, my starting point is arguably less demanding, namely that the ‘succession of States [is] occurring in conformity with international law’ (note, in this regard, the International Court of Justice’s Advisory Opinion Regarding the Unilateral Declaration of Independence in Respect of Kosovo). As the 27 March 2014 UN General Assembly resolution regarding the ‘Territorial Integrity of Ukraine’ demonstrates, when referendums fail to meet the above criterion, international non-recognition may ensue.

In this introduction, my (main) point of reference is the Scottish Independence Referendum and the criteria for attributing citizenship on ‘day one’ of an independent Scotland. Other contributions will no doubt broaden the geographic scope of this debate, perhaps to Catalonia and elsewhere.  

I wish to put forward two propositions.

The first proposition is that putative ab initio citizens of a putative state (the initial citizen-body of a new state), whether its nationality is attributed to them or they are given the ‘right of option’, are clearly stakeholders (borrowing Rainer Bauböck’s seminal characterisation) in an independence referendum that may bring that putative state into being.

Moreover, as Bauböck argued (id) in support of external voting in national elections ‘[b]y virtue of their permanent membership, citizens have a life-long interest in the future of the polity, its survival and success’. The rationales for expatriate voting in national elections (in contradistinction from local elections) apply a fortiori to independence referendums in light of its fundamental nature and the long-term effects of its outcome.

Citizens enjoy internationally recognised rights, most prominently the right to return to and reside in their state of citizenship. This and other rights will be directly and meaningfully affected by the outcome of the referendum. The lives of putative citizens may be directly affected by subsequent electoral processes in the putative state (see e.g. the language employed by the EU Commission in its 29 January 2014 recommendation regarding EU citizens residing in another EU member state). In the context of the Scottish independence referendum, one only needs to point to uncertainties regarding EU membership, the UK/Irish common travel area, currency, and taxation to start appreciating the extent to which a ‘Yes’ vote may meaningfully affect the lives of putative Scottish citizens.

The establishment of a new state whose citizenship they may hold from day one, which triggers this life-long interest, follows the referendum as a constitutive act. Indeed, it could be argued that, even if there is an inclusive franchise in national elections of an existing state, ‘[s]elf-government, whether direct or through representatives, begins by defining the scope of the community of the governed, and thus the governors as well’. As Cormac Mac Amhlaigh recently noted, the decision to exist as an independent political entity is a political question with a capital ‘P’: it involves an existential choice in the life of the nation beyond small ‘p’ politics.

Turning to Scotland, the 2013 Scottish Independence Referendum (Franchise) Act determines eligibility for participation in the referendum. The Act lowers the voting age to 16, and disenfranchises all serving prisoners (discussed here). Crucially, the eligibility criteria do not mirror the criteria for participation in the UK general election, set in Section 1 of the Representation of the People Act 1985. According to the latter Act, UK citizens who have left the UK in the last fifteen years are eligible to vote in UK general election; their vote is cast in their last place of residence; for those formerly resident in Scotland, this means their Scottish constituency (the plausibility of the arrangements under this act are also questionable, not least regarding the 15 year rule challenged e.g. in the Shindler case, and the selective access to the national franchise that is given to qualifying commonwealth and Irish citizens).

By contrast, the franchise for the Scottish Independence referendum follows the criteria employed to determine eligibility for local government elections, set in Section 2 of the Representation of the People Act 1983. Hence, in addition to UK citizens habitually resident in Scotland, to Irish citizens, and to qualifying Commonwealth citizens (all of whom are also eligible to vote in general UK elections), EU nationals habitually resident in Scotland (see 2011 census data) are eligible to vote in the referendum. UK citizens formerly resident in Scotland are excluded wherever they currently reside (namely in Rump-UK or elsewhere) and regardless of the duration of their absence from Scotland.

I assert that the franchise in independence referendums ought to reflect the fact that the types of question addressed in such referendums are qualitatively different from the issues raised in elections for sub-units of a state, such as local government elections. Independence referendums share the fundamental and long-term characteristics of national elections, and their significance is enhanced by their capacity, from both a national and an international law perspective, to alter the legal landscape for individual citizens.

The second proposition is that congruence between eligibility for participation in independence referendums and eligibility for citizenship ab initio is highly desirable. Under-inclusiveness (exclusion of putative citizens) may undermine the legitimacy of the referendum, not least for disenfranchised persons affected by a new legal reality. Over-inclusiveness (inclusion of persons ineligible for citizenship ab initio) suggests that perhaps such persons ought to be offered citizenship of that putative state.

Achieving congruence is no mean feat: a putative state would have to determine ab initio citizenry at the time of the referendum. Indeed, beyond the category of citizens of the predecessor state habitually resident in the territory affected by the succession of states (who according to the ‘presumption of nationality’ in Article 5 of the ILC Draft Articles are presumed to acquire the nationality of the successor state on the date of such succession), the picture is rather complex.

As Jo Shaw helpfully noted, states have followed several models for determining their citizenship ab initio: the ‘zero option’ model, where citizenship was given to all permanent residents at the moment of independence; the ‘restored state’ model, recognising a historic statehood; the ‘mixed’ model, drawing on elements of each; and the ‘federal upgrading’ model, where a previous ‘republican’ or ‘provincial’ citizenship was upgraded to state citizenship at the moment of independence. Hence, it may be queried whether citizenship should be offered to all habitual residents of the putative state, and/or to expatriates of the predecessor state formerly residing in the successor states, regardless of the length of time they have been away and of their current place of residence.

As noted above, the ILC Draft Articles distinguish between citizens of the predecessor state, who must be offered citizenship of at least one of the successor state(s), and citizens of third states, towards whom such an obligation in international law does not arise, though the provisions stipulate that the status of such citizens as habitual residents should not be affected by the succession of states (the latter issue may be particularly pertinent regarding rights of residence of citizens of other EU member states, were an independent Scotland to remain temporarily or long-term outside the EU). Hence, from an international law perspective, while a putative state could extend an offer of citizenship to such persons, it would not be required to do so. 

In contradistinction, the ILC Draft Articles stipulate (in Articles 22 and 24 regarding dissolution and separation, respectively) that citizens of a predecessor state habitually resident elsewhere who have ‘an appropriate legal connection’ to a successor state should be offered citizenship of that successor state. The ILC opted for an arguably less demanding ‘test’ than the International Court of Justice’s stipulation in its 1955 Nottebohm (Lichtenstein v. Guatemala) case concerning the exercise of protection by Lichtenstein, Nottebohm’s state of nationality. In that case, the Court held that ‘nationality is a legal bond having as its basis a social fact of attachment, a genuine connection of existence, interests and sentiments, together with the existence of reciprocal rights and duties’. The choice to move away from Nottebohm can be explained by the reverence of a protection-enhancing rights-based approach to the link between the citizen and her state.

The ILC Draft Articles mandate that an offer of citizenship should be subject to a ‘right of choice’. Importantly, the choice is between two or more citizenships, most likely between retaining the citizenship of a predecessor state and obtaining the citizenship of a successor state. However, while international law has historically been hostile to multiple citizenships (see the Preamble to the 1930 Hague Convention on Certain Questions Related to the Conflict of Nationality Laws), it is presently considered to be neutral on this matter. An independent Scotland and Rump-UK may agree that their respective citizens may hold other nationalities, including that of the respective state north/south of the border. However, it takes two to tango: while the Scottish government has announced in its White Paper entitled ‘Scotland’s Future’ (published on 27 November 2013) its intention to follow ‘a[n] inclusive model of citizenship for people whether or not they define themselves as primarily or exclusively Scottish’, the current position of the UK government is less clear, and it is not implausible that putative citizens of Scotland may be forced to exercise a right of choice between obtaining Scottish citizenship and keeping their Rump-UK citizenship.  

Fraught with genuine difficulties as it may be, a determination of state citizenship ab initio is unavoidable: without entering the debate over the status in international law of the 1933 Montevideo Convention ‘criteria’ for statehood (a permanent population; a defined territory; government; the capacity to enter into relations with the other states), it may be legitimately expected that a putative state adopt (non-arbitrary) criteria for its ab initio citizenry. It stands to reason, then, that these criteria may be defined at the time of the referendum, and serve as the basis for its franchise.  

Indeed, the Scottish government agrees that determination of citizenship ab initio is required: Chapter Seven of the Scottish government’s White Paper offers a clear blueprint. In a helpful table, it stipulates under the heading ‘at the date of independence’ that ‘British citizens habitually resident in Scotland on day one of independence’ (projected for 24 March 2016) as well as ‘British citizens born in Scotland but living outside of Scotland on day one of independence’ will automatically obtain Scottish citizenship. In contradistinction, ‘after the date of independence’, migrants residing in Scotland legally and citizens of any country who have spent at least ten years living in Scotland at any time and have an ongoing connection with Scotland ‘may apply for naturalisation’. Intriguingly, the paper suggests that attribution of citizenship to children (and grandchildren) will follow a mixed model combining elements of ius sanguinis and ius soli.

Set against the proposed congruence model, the franchise for the Scottish Independence Referendum is both under and over-inclusive. It excludes ‘British citizens born in Scotland but living outside of Scotland on day one of independence’, whom the Scottish government clearly considers to be Scottish enough to be attributed citizenship on ‘day one of independence’, that is, to be part of the constituent body-polity that in due course will adopt a ‘modern [written] constitution’ for Scotland (following a constitutional convention). Concurrently, it enfranchises some residents in Scotland: those who happen to be citizens of other EU member states, Irish and qualifying commonwealth citizens. The latter persons are, apparently, well-placed to decide whether Scotland should be an independent country, but that does not make them part of ‘the people of Scotland’.

The Scottish government’s blueprint for the ab initio polity of an independent Scotland, while prima facie compliant with the ILC Draft Articles’ framework, is by no means immune from critique, which other contributors may wish to mount. Notwithstanding the question which criteria should have been proposed for citizenship ab initio, the claim that the determination of the franchise for the Scottish independence referendum was ill-conceived is underscored by the Scottish government’s vision of the people of an independent Scotland being so markedly different from the electorate that will determine in less than four months the coming into being of that polity. While in the Scottish case ‘the deed is done’, perhaps lessons can be learnt and applied to determine the franchise in future independence referendums.



The Scottish referendum franchise: Residence or citizenship?

by Bernard Ryan (University of Leicester)



Eligibility to vote in the 18 September 2014 referendum on Scottish independence will be based on the franchise currently used throughout the United Kingdom for devolved and local matters. Accordingly, the voters will be persons resident in Scotland who are British citizens, qualifying Commonwealth citizens, Irish citizens, or other EU citizens. With the Scottish case in mind, Ruvi Ziegler’s kick-off contribution to this forum develops two propositions: that the “ab initio citizens of a putative state” are “stakeholders” in an independence referendum; and, that there should be “congruence” between the franchise for an independence referendum and the initial citizenship of a potential new state.

As Ziegler’s contribution focuses on the Scottish case, my remarks will do so too.[1] I will firstly point to tensions between Ziegler’s two propositions in the Scottish case. I will then suggest that the residence-based franchise chosen for the Scottish referendum deserves respect because it was chosen by the main protagonists, and is calculated to ensure the legitimacy of the referendum. At the same time, the content of a future Scottish citizenship law is too indeterminate to provide the basis for the franchise in a referendum.

Internal tensions

Drawing upon his wider perspective, Ziegler offers two general comments on the Scottish referendum franchise. His first proposition leads to the claim that

“the franchise in independence referendums ought to reflect the fact that the types of question addressed in such referendums are qualitatively different from the issues raised in elections for sub-units of a state, such as local government elections. Independence referendums share the fundamental and long-term characteristics of national elections…”

His second proposition leads to criticism of a mismatch between those eligible to vote and those expected to become Scottish citizens at independence. From the perspective of “congruence”, the franchise is under-inclusive in not providing a vote for non-residents who are British citizens who can expect to acquire Scottish citizenship automatically, and over-inclusive in granting a vote to other resident EU and Commonwealth citizens.

It is notable that Ziegler’s two propositions with respect to the Scottish referendum franchise point in somewhat different directions. One tension arises from the fact that the United Kingdom does not confine the right to vote in parliamentary elections to British citizens, but instead extends it to resident Commonwealth and Irish citizens. It might be thought to follow from Ziegler’s first proposition that Commonwealth and Irish residents should have a vote in the referendum, as that is the position for United Kingdom parliamentary elections. Yet, his second proposition suggests otherwise, as these persons are not British citizens, and are unlikely to become Scottish citizens at independence.

A second issue concerns votes for British citizens who are resident outside Scotland and who would probably acquire Scottish citizenship at independence. Ziegler’s comment about national elections suggests one of two answers. If the parliamentary franchise in Scotland is the reference point, a vote should be available to expatriate British citizens outside the United Kingdom for up to fifteen years, provided they are registered in a Scottish constituency. Alternatively, if the reference point is the parliamentary franchise for the United Kingdom as a whole, one might conclude that British citizens resident in the rest of the United Kingdom should also have a vote in the first fifteen years. Neither approach however does enough to meet Ziegler’s test of “congruence”. That would suggest a vote for all British citizens who are on course for Scottish citizenship, wherever in the world they live, and without a requirement of residence in Scotland in the previous fifteen years.

The actual franchise

These tensions within Ziegler’s position show that it is one thing to set out a general position, and another to apply it coherently on the ground. In my view, a better approach would be to start, not from general principles, but from the United Kingdom’s constitutional order.

In the absence of a written constitution for the United Kingdom, the ground rules for the referendum were necessarily defined in the political sphere. Despite their different views on the substantive question, in the Edinburgh Agreement of October 2012, the United Kingdom and Scottish Governments reached a consensus concerning the terms on which the referendum would take place.[2] That paved the way for enabling legislation at Westminster, after the approval of both Houses of the United Kingdom Parliament, and of the Scottish Parliament.[3]

In relation to the franchise, the Edinburgh Agreement provided that all persons entitled to vote in devolved and local elections should also have a vote in the referendum. The Agreement permitted the extension of the franchise to others, if that was the decision of the Scottish institutions. That option was then used by the Scottish Parliament to grant a right to vote in the referendum to 16 and 17 year olds, something which was in line with the SNP’s policy more generally, and which has specifically been mentioned in the Edinburgh Agreement.[4]

A significant reason for basing eligibility to vote on the Scottish devolved franchise was that that franchise had been used in the 1997 devolution referendum. This argument from precedent appeared consistently in the consultation documents of both the Scottish and British Governments.[5]

A second reason, offered by the Scottish Government, was that the devolved franchise made residence in Scotland central to the right to vote.[6] Among its reasons for supporting a residence-based approach were that that principle was “internationally accepted…for constitutional referendums”, and that “sovereignty lies with that Scottish people”. It could equally have argued that residents, as a general rule, have the greatest stake in the outcome of the referendum decision.

A further argument, put forward in the United Kingdom Government’s consultation in 2012, was that it was essential to start from a pre-existing franchise in order to avoid the “the perception that changes were being made to favour one or other outcome.”[7] That argument might be thought especially relevant to the conferral of eligibility to vote upon some British citizens resident outside Scotland. Had that been done, it would have led to suggestions that voters likely to favour the Union were being added, and would have risked raising doubts about the legitimacy of the referendum, with problematic implications in the event of a close ‘no’ vote.

The Scottish referendum franchise was therefore arrived at through a process of constitutional decision-making led by the Scottish and United Kingdom Governments. That process, and the precedent- and residence-based outcome that it produced, are calculated to deliver a legitimate referendum. For these reasons, the franchise chosen must be considered robust to a critique which favours models based on the United Kingdom parliamentary franchise and/ or future citizenship.

Scottish citizenship

The discussion so far has assumed that there is an identifiable category of persons who would be what Ziegler terms “citizens ab initio”, from which a credible franchise could be constructed. The position is though more complex and indeterminate than that.[8] The Scottish Government’s plan is that a Scottish citizenship law would be elaborated in the eighteen-month period of transition to independence which would follow a positive vote in the referendum.[9] It is true that the Scottish Government outlined its own position as to the content of the citizenship law in Scotland’s Future, published in November 2013. But that position must be considered provisional, and anyway lacks detail in many respects.[10] A franchise based on future citizenship would therefore be difficult to operationalise in the Scottish case.

Consider first the two groups whom the Scottish Government’s intends would acquire citizenship automatically: those habitually resident in Scotland on the date of independence, and those who were born in Scotland, irrespective of their place of residence. Ziegler is clear that these groups should in principle have a right to vote in the referendum. Even in these cases, however, the post-referendum disposition of the British Government is a complicating factor. The current British Government has indicated that it might seek to withdraw British citizenship from those persons resident in Scotland at independence upon whom Scottish citizenship was conferred automatically.[11] As that prospect crystallised in a post-referendum context, the Scottish Government and Parliament might decide to narrow the category of those who acquired citizenship automatically, in order to permit others to retain British citizenship.

Beyond the ‘automatic’ cases, it is uncertain what Ziegler contemplates for those persons who would be eligible for Scottish citizenship around the time of independence, but upon whom it would not be conferred automatically. In his contribution, he suggests that persons given the “right of option” to become citizens are “stakeholders”, who ought to have a say over independence. He also hints that other EU and Commonwealth citizens should be normatively entitled to participate if they are “offered citizenship of [the] putative state.

Two questions of principle may be posed here. First, which kinds of “option” or “offer” are sufficient to bring someone within the normatively favoured group? Secondly – however the first question is answered – what is the justification for denying a vote to some persons who are residents but who are not eligible for citizenship (or not eligible in the right way)? Ziegler does not offer a clear position on either of these points.

In the Scottish case, there is in any event uncertainty as to who would have an option over Scottish citizenship around the time of independence. For example, Scotland’s Future stated that any person with “a parent or grandparent who qualifie[d] for Scottish citizenship” could register as a Scottish citizen – but it is unclear how many generations that would cover, or how it would apply to those born before independence. Scotland’s Future also recognised that persons of “good character” would be able to naturalise through residence in Scotland, but neither the qualifying period(s) nor any other requirements were specified. A further ill-defined possibility is that others could naturalise if they had “spent at least ten years living in Scotland at any time”, and had an “ongoing connection” to it. Finally, it is possible that some British citizens might be free to decline automatic Scottish citizenship: this was accepted in outline in Scotland’s Future, but the details would depend upon the interaction of British and Scottish law at the time of independence.

Final thoughts

In conclusion, I found Ziegler’s attempt to devise a general theory of entitlement to vote in independence referendums admirable, rather than compelling. In the Scottish case, Ziegler’s approach runs up against the enduringly pragmatic character of the United Kingdom’s constitutional arrangements. The right to vote is not generally limited to British citizens, even for parliamentary elections. The constitutional architecture gave the flexibility to the two main protagonists to agree upon a precedent and residence-based approach to the franchise. Equally, the content of a Scottish citizenship law remains uncertain, and would not be defined until after a referendum vote.

Viewed in the abstract, it may make sense to class an independence referendum as a ‘national’ matter and/or to aim for the equation of the franchise with a future citizenship. The lesson I would draw for other cases, however, is that it may be difficult to apply that approach in a real situation. This will be especially true if there are many non-citizen residents with a legitimate expectation – based on precedent - of a say over independence.



[1] For the avoidance of doubt, I do not take a position as to the merits of an independent Scotland. My view is rather that unnecessary obstacles - intellectual or otherwise - ought not to be put in its way, if that is the wish of the Scottish population, determined in accordance with a fair process.
[2] Agreement between the United Kingdom Government and the Scottish Government on a referendum on independence for Scotland, signed at Edinburgh on 15 October 2012.
[3] Scotland Act 1998 (Modification of Schedule 5) Order 2013, SI 2013 No. 242.
[5] Scottish Government, Scotland’s Future: Draft Referendum (Scotland) Bill Consultation Paper (February 2010), pp 24-25; Scottish Office, Scotland’s Constitutional Future (January 2012), Cm 8023, pp 16-17; Scottish Government, Your Scotland: Your Referendum (January 2012), para 2.10.
[6] Scottish Government, Your Scotland: Your Referendum (January 2012), para 2.10.
[7] Scottish Office, Scotland’s Constitutional Future (January 2012), Cm 8023, pp 17.
[8] The discussion here draws upon B. Ryan, ‘At the Borders of Sovereignty: Nationality and Immigration Policy in an Independent Scotland’ (2014) 28 Journal of Immigration Asylum and Nationality Law 146, 147-151.
[9] Scottish Government, Scotland’s Future: Your Guide to an Independent Scotland (November 2013), p 340.
[10] Ibid, pp 271-273 and 495-499.
[11] Home Secretary, Theresa May, declared in the House of Commons on 10 June 2013 that “Decisions on UK citizenship are for the UK Government. Any decisions on the retention of UK citizenship by Scottish citizens after independence would be affected by future Scottish Government policy decisions.” See too the discussion in HM Government, Scotland Analysis: Borders and Citizenship (Cm 8726, January 2014), p 9.


Regional citizenship and self-determination

by Rainer Bauböck (EUI, EUDO CITIZENSHIP co-director)


In his kickoff contribution, Ruvi Ziegler argues that those who would become citizens on ’day one’ of an independent Scotland should also be enfranchised in the referendum that will decide whether or not Scotland will become independent. 

Ziegler suggests that “putative ab initio citizens of a putative state (the initial citizen-body of a new state), whether its nationality is attributed to them or they are given the ‘right of option’, are clearly stakeholders … in an independence referendum that may bring that putative state into being.” Since I have defended a stakeholder principle for determining who should be offered citizenship and the franchise (Bauböck 2009), I need to explain why I disagree with Ziegler’s proposition.  I arrive at the same conclusion as Bernard Ryan, whose objections are based on legal norms and pragmatic reasons. My argument will instead be based on democratic principles that I consider as normatively coherent and attractive even if they are not fully recognised by current laws, academic scholars or political actors.

The short version of my objection is that independent states and autonomous regions within states are polities of different kinds. What it means to be a citizen in a polity depends not only on whether a person has a genuine link to that polity, but also on its nature. The Scottish referendum will decide whether the nature of the Scottish polity will be radically changed by transforming it from a self-governing region within the UK into an independent state. The outcome of this decision must not be preempted by enfranchising those who would become citizens of an independent Scotland. Instead, the only legitimate franchise is the existing one for Scotland as an autonomous territory of the UK.  

Consider first the puzzle why UK citizens who are born and reside in England, Wales and Northern Ireland do not have a vote in the referendum. They are certainly stakeholders in the UK-wide polity. If Scotland became independent, this polity would be quite radically changed. Isn’t an independence referendum exactly the kind of constitutional transformation in which all citizens have to be enfranchised, including those living permanently abroad, as David Owen has cogently argued (Owen 2010)? Ziegler points out how “a ‘Yes’ vote affect the lives of putative Scottish citizens" on Scottish independence. But doesn’t it at least equally affect the lives of citizens in what has somewhat prematurely been labelled rUK (the rest of the UK). Without Scotland on board, it is more likely that Conservative governments will stay in power in London and that a future referendum on UK membership in the EU will result in Brixit. Why is this not enough to give them a vote in the referendum? 

One answer is that if independence depends on a majority of all British citizens voting for it, then Scotland has of course no effective right to self-determination, since its residents will be outvoted by an overwhelming majority of R-UK citizens. This leads to the crucial follow-up question: why does Scotland have a right to self-determination that excludes votes from UK in the first place? 

Legally, the answer to this question is clear: because Westminster has accepted that Scotland can decide unilaterally on its territorial status. The Scottish right to self-determination has thus been agreed to by democratically elected representatives of both sides in the dispute: the British and the Scottish governments. This is a rare case. More frequent are conflicts in which the central government denies the right to self-determination of an autonomous territory, as the Spanish government currently does in response to Catalan demands for an independence referendum. The fact that the Spanish constitution rules out unilateral self-determination by any part of the “indivisible Spanish nation” does not settle the matter from the democratic perspective. If Scots can unilaterally determine the future status of their territory, why shouldn’t Catalans have the same right? In other words, in the absence of a consensus, what else could ground a right of self-determination through a democratic vote? 

The case of Kosovo provides an answer to this question. Unilateral secession from a state can be legitimate if the central government of a state violates fundamental rights of the inhabitants of a province, abolishes their previous rights of self-government or tries to change the demographic basis of their autonomy through ethnic cleansing and settlement policies. Self-determination becomes then a remedial right that can be exercised against the will of the central government.

I do not think that self-determination claims in Catalonia, Quebec, Flanders and other provinces that enjoy robust autonomy qualify as remedial. This does not settle the issue since it is possible to argue that the Madrid, Ottawa and Brussels governments ought to consent to self-determination referendums for the same pragmatic reasons as London. In Quebec two such referendums were held in 1980 and 1995 and in 1998 the Supreme Court of Canada stated that, although Quebec does not have a right to self-determination under international law or the Canadian Constitution, the federal government would have to enter negotiations if another referendum resulted in “a clear majority vote in Quebec on a clear question in favour of secession”[1].  I leave it to other contributors to further explore such non-consensual and non-remedial cases and will instead focus on how this distinction bears on the first question asked in this forum debate: Should the franchise in independence referendums depend on whether self-determination is remedial or consensual? 

Where an independence referendum is clearly remedial, as the one held in 1991 in Kosovo in response to Milosevic’s abolition of Kosovo autonomy was, the solution proposed by Ziegler seems the right one. When a constitutionally guaranteed right of a political community to self-government is abolished, those who would become citizens of an independent state should have the right to decide on whether they support this outcome. There are several qualifiers to this proposition: it may be practically difficult to create a voter registry that includes those driven into exile some time ago, especially in a situation where a central government tries to prevent the referendum from happening. And even a remedial right to self-determination does not imply a right to exclude residents from the franchise and future citizenship on ethnic or other discriminatory grounds. If secessionists claim jurisdiction over a certain territory they must grant the franchise and future citizenship to all legitimate residents in this territory – which still allows for the exclusion of recent settlers brought in to change the demographic composition of the population. But the normative reason remains clear: the decision about the future independence of a region need not be taken exclusively by those previously enfranchised in regional elections, if there is no more effective regional self-government and citizenship as a legitimate basis for this franchise.

Whereas remedial self-determination involves a decision whether to replace an abolished regional citizenship with citizenship in an independent state, consensual self-determination involves a decision whether to upgrade an existing regional citizenship into that of an independent state. And this decision should be taken by those who are currently voting citizens of the region. It seems to me illegitimate to preempt the outcome of such a decision by enfranchising putative citizens who do not have a right to vote in regional elections now and who will also not gain a future right to vote in case of a ‘No’ outcome. My concern is not how enfranchising persons born in Scotland who reside in London, Brussels or Boston would affect the referendum result. It may well be that a majority of these putative Scottish citizens would prefer Scotland to remain in the UK so that they do not risk losing their EU passports. I am raising a principled objection: The putative demos of an independent Scotland should not replace the existing demos of Scotland as part of the UK in a decision about independence because currently only the latter but not the former can be considered as democratically legitimate.

This brings me to the second question asked in this forum: Who should be the citizens of newly independent states? Our previous discussion would be pointless if the answer were: exactly the same as the current regional citizens. But regions and independent states are polities of different kinds and their membership rules differ accordingly. All citizenship laws of independent democratic states are based on birthright (ie some mixture of ius soli and ius sanguinis) and provisions for naturalisation and voluntary renunciation of citizenship. These rules create often large discrepancies between residents and citizens with significant numbers of non-resident citizens and non-citizen residents. By contrast, the citizens of autonomous regions in democratic states are in most cases the citizens of the wider state who take up residence in the region. There is no birthright, naturalisation, or renunciation, and thus also no external citizenship that would allow emigrants originating in the region who do not maintain a residence there to cast votes in regional elections. 

The Scottish case is slightly unusual because the franchise in Scottish Parliament elections includes not only British, Irish and Commonwealth citizens who can also vote in Westminster elections, but even citizens of other EU member states. As Ziegler points out, the latter would not automatically become Scottish citizens in case of independence but would have to opt in through naturalisation. Is this a problem for the legitimacy of the Scottish referendum? I do not think so. It is certainly not illegitimate for a region to extend its citizenship to residents who are not citizens of the larger state. We find this extension in local elections in fourteen European states where the franchise is granted to all residents independently of their nationality. If all of the currently enfranchised citizens of Scotland as a region of the UK have a legitimate vote in regional elections, it would be wrong to exclude them from the decision on a fundamental change of the territorial status of the region.

Who should then become citizens of an independent Scotland? On this point, I do not have any major disagreement with Ziegler’s helpful discussion of international legal norms for initial determination of citizenship in case of state succession or with the respective proposals in the Scottish government’s White Paper. What we need to understand is that these norms address the specific task of setting up a birthright citizenship regime for a newly independent state that necessarily differs from whatever rules had previously existed within an autonomous region. Just as we must distinguish the initial collective determination of automatically included citizens on day one of an independent state from the ongoing individual determination under a regular citizenship law, so we ought to distinguish existing regional citizenship from initial determination at independence. Independence referendums are procedures that provide democratic legitimacy to the transition from regional to independent-state citizenship. But they must not blur this distinction. If the moral basis for self-determination exercised in this way is consent rather than remedial justice, then the vote belongs to the regional citizens who should decide on whether they want to change their region’s status through secession. 



[1] Reference re Secession of Quebec, [1998] 2 S.C.R. 217. 


Bauböck, Rainer. 2009. "The Rights and Duties of External Citizenship." Citizenship Studies 13 (5):475-499.

Owen, David. 2010. "Resident Aliens, Non-resident Citizens and Voting Rights." In Citizenship Acquisition and National Belonging, edited by G. Calder, P. Cole and J. Seglow, 52-73. London: Palgrave.



A Matter of Legitimacy?

by Dimitrios Kyritsis (University of Reading)


Democratic rule aspires to an ideal of universal political participation. Within national political communities, this ideal has historically manifested itself in a demand for the extension of the franchise. But outside that framework the ideal is fraught with ambiguity. 

One of these ambiguities concerns procedures that aim to establish a national political community from scratch, as illustrated by debates about the franchise in the Scottish independence referendum. As things stand, if you were born and grew up in Scotland but reside elsewhere, say in England, you do not have a right to vote in the referendum. This is so despite the fact that, were Scotland to become independent, the Scottish government has committed to granting you citizenship ab initio. I do not doubt that the interests of political expediency and national myth-making might recommend that the franchise be extended to you. But it strikes me that no duty is breached if this does not happen. In his kick-off contribution, Ruvi Ziegler claims otherwise. He not only contends that giving the right to vote to potential ab initio citizens who are non-residents is ‘highly desirable’ but that not doing so ‘may undercut the legitimacy of the referendum’. The latter statement suggests that, for Ziegler, something in the vicinity of a mandatory norm will have been breached. Here, I shall offer some arguments against this thesis. 

Ziegler advances two reasons for his view. First, he says, we ought to extend the franchise to all potential ab initio citizens because independence referendums ‘meaningfully and directly’ affect their legal options. Granted, this feature of independence referendums may serve to distinguish them from political procedures that have no similarly momentous effect. But the principle that we ought to participate in all the political decisions that profoundly affect us is not a sound one. That principle would give Venezuelan citizens a say in US energy policy, for instance. It would require that (many) citizens of Central American states participate in the design of US immigration policy. Notice that in both examples US policy is foisted upon someone. The situation is markedly better for those potential ab initio citizens of an independent Scotland who are not eligible to vote in the referendum. According to the ILC Draft Articles, they must be given the right to accept or reject Scottish citizenship. Thus, if the principle is not valid in the case of the Venezuelans, surely it cannot be for the disenfranchised Scots-to-be. That is not to say that the latter have an easy choice to make. Some of them may face the dilemma of trading one nationality for another. Still, there is no general duty to give people easy choices, especially when those choices basically add new - morally permissible - options to the ones that existed before. 

Second, Ziegler maintains that the extension is required to remedy the mismatch between the group of people who will decide on independence and the group of ab initio citizens. Arguably, the mismatch smacks of arbitrariness, and arbitrariness is the kind of thing that ‘undercuts’ political legitimacy. But it is not clear why this particular mismatch is arbitrary in a sense that is crucial for legitimacy. Consider a restriction on the right to vote in the referendum along racial lines. What makes it illegitimate is its illicit ground. Ziegler owes us an analogous argument. Perhaps he thinks that the status of ab initio citizenship is a sufficient (though not necessary) condition for the right to vote on independence. I beg to differ. If anything, it is those who will become systematically subject to the coercive force of the new state from day one just by virtue of their residence that have the most pressing interest to decide on independence, whether they will be citizens or not. Conversely, for the non-residents who can become ab initio citizens, this status is - given their right to choose - little more than some sort of premium membership. You can throw in the right to vote if you want, but you don’t have to. 

Where do these arguments leave us? Even if I am right that Ziegler does not dent the legitimacy of the Scottish independence referendum, his proposal touches on an important issue. Ours is a world of states whose existence, boundaries and membership (not to mention relative wealth and power) are to a large extent the product of force and historical contingency. This raises a host of challenges, which we encounter time and again in debates about immigration, global justice, humanitarian intervention, the obligation to obey the law and so forth. Understandably, we want to ensure that at least new states will be built on more rational foundations. Ziegler’s proposal focuses our attention on this worthwhile project.



Who votes in a referendum? General comments and some facts concerning Québec

by Guy Laforest and Eric Montigny (McGill University and Université de Laval)


In critical essays on constitutional referendums, Patrick Taillon and Stephen Tierney have argued that the crucial difficulty is to identify the People, the Demos.[1] In other words, the first question is to determine who is substantively concerned by the ballot question (who are the legitimate stakeholders, as argued by Rainer Bauböck in his own commentary in this forum). In light of its own experience with referendums, Québec is a good case study. The aim of this contribution is to comment on some of the larger issues debated by Ruvi Ziegler and others in this forum, and further to explain the rules that are currently in place in Québec on voter eligibility.

General comments

On matters of referendums, consistent normative logic does not always work. To give but one example from the current debate, Rainer Bauböck in his own contribution recalls that, beyond the issue of the franchise, unilateral secession can be rendered legitimate if the central government of a state has violated the fundamental rights of the inhabitants of the seceding province or abolished their previous rights to self-government. Sometimes, reality can be more complex. It is possible to imagine a situation whereby the central government has substantially reduced, rather than abolished, the rights to self-government of the inhabitants of the seceding jurisdiction. Context, here, would require further normative reflections. On the issue of the franchise at the heart of the current debate, our core argument is as follows: gaining consensus between central and sub-state government is more important than maintaining consistency between pre- and post- independence enfranchisement.

Ruvi Ziegler, examining the Scottish case, argues that those who will be offered Scottish citizenship after independence should also be entitled to vote in the referendum in which the matter will be decided. He argues on the side of normative clarity and consistency. Bernard Ryan replies by siding with the pragmatic political and constitutional arrangements at work in the United Kingdom, which apply to the upcoming Scottish referendum the rules allowing European Union and Commonwealth citizens to vote at devolved and local elections. In the end, we believe that consistency between past and present democratic rules should prevail in this case over consistency about matters as they stand and democratic citizenship rules for the future. Rainer Bauböck makes a similar point in his comment, arguing that the rules guiding normal sub-state elections in Scotland should also apply to an independence referendum. However, the foundations of his reasoning are different from ours.  Bauböck starts from the idea that the sub-state demos is substantially different from the demos of an independent state, before moving to the issue of consistency. We would rather argue that the most important dimension, on the issue of the franchise and on many others, has to do with the existence of an overall agreement between the British and Scottish governments, arrived at on 15 October, 2012. Concerning independence referendums, nothing is more important than reciprocal consent and respect between the existing state and secessionist authorities. At least in part for reasons of consistency, we believe that the provision allowing young people, 16 and 17 years of age, to vote in the upcoming Scottish referendum, is wrong. This is, however, the kind of inconsistency we can live with because it was one element in the contours of the compromise between British and Scottish authorities. We will reformulate our core argument after having looked at the case of Québec.

Referendums in Québec and the electoral system

Over time, Quebeckers have experienced different types of referendums (always non-binding, following the logic of Westminster-based parliamentary democracy); three were initiated by the Canadian federal parliament and four by their own legislature.[3] The two most famous ones dealt with sovereignty-association in 1980 and with sovereignty-partnership in 1995.

As citizens of the Canadian federal state residing in the province of Québec, Quebeckers are entitled to vote at both regional and national levels.[4] Jurisdiction over the electoral process is shared by both levels of government, under the umbrella of the judiciary, including the Supreme Court of Canada which sits as Canada’s final court of appeal.

Two main laws stemming from the National Assembly of Québec govern the electoral system in the province. Adopted in 1978, the Québec Referendum Act is still in force. It includes provisions regarding the obligations to form a Yes and a No ‘camp’, to establish a Referendum Council, and to regulate financing. For other matters, such as the eligibility to vote, the Referendum Act refers to the Québec Election Act.

Qualified electors

According to the Québec Election Act, mentioned above, in order to be able to vote in a referendum in Québec, one has to be a qualified elector. The act applies age (18 years) and Canadian citizenship requirements.

To be a qualified elector, one must also have been domiciled in Québec for six months. For most electors, this is easy to demonstrate. This provision could face interpretive difficulties for newcomers or for students born elsewhere in Canada. The domicile of a person is the domicile established under the Québec Civil Code.[4] It means that it has to be one’s main address and that this person has expressed in practice her or his intention to consider it as such. A debate on the status of residency, initiated by McGill University students, occurred in Court before the 2014 election. Most of the cases submitted to the Court were rejected. Moreover, one year after the referendum of 1995, students of Bishop University were found guilty of voting without being Québec residents. Robert Ghiz, currently Premier of Prince Edward Island, was studying at Bishop University at the time and admitted to voting in the 1995 referendum.
In that perspective, a person who is deprived of voting rights pursuant to Québec laws (Election Act, the Referendum Act, the Act respecting elections and referendums in municipalities or the Act respecting school elections) is not allowed to vote.

A permanent list of electors

Since 1995, Québec has chosen to put in place a permanent list of electors. This list consists of the information contained in the register of electors and the register of territories. According to the Election Act, this information shall include the name, domicile-based address, gender and date of birth of each elector.

The information relating to electors is updated on the basis of the information transmitted to the Chief Electoral Officer. This information could come directly from electors, from the school boards, the Public Curator, the Chief Electoral Officer of Canada and the Department of Citizenship and Immigration of Canada. In practice, it comes essentially from the Régie de l'assurance-maladie du Québec. This agency has to notify the Chief Electoral Officer of any change in the name, address, date of birth or gender of a person whose name is entered on the permanent list of electors, as well as the date of the person's death. The same applies for a person who has informed the Régie that he or she has acquired Canadian citizenship or who is about to reach 18 years of age, at least six months before the person's eighteenth birthday.

Alternative voting procedures and voting outside Québec

The Québec Electoral Act provides many different ways for electors to express their vote beyond the regular practice of voting in one’s residential neighborhood. First, electors can vote at the returning officer's main office or branch offices. Second, the practice of early voting is gaining in popularity.[5] Since 2014, it is possible to vote on the campus of a vocational training centre or a post-secondary educational institution. Finally, postal voting is permitted. The latter method is available to electors who are incarcerated but also to electors residing outside Québec. The latter case is crucial to our current discussion.

Voting by electors residing outside Québec was in force for both the 1992 and 1995 referendums. Electors that register to vote outside Québec were deemed to be domiciled at their Québec address. They had to demonstrate that they had left Québec temporarily after being domiciled in Québec for 12 months. They were able to vote outside Québec for two years after their date of departure.

This two-year limit did not apply to an elector, and her or his spouse, posted outside Québec working for the governments of Québec or Canada, or to an employee of an international organisation of which Québec or Canada is a member and to which it pays a contribution. An elector who wished to vote outside Québec had to file and sign a request including the following elements: name, sex and date of birth;   domiciled address in Québec or last domiciliary address in Québec;  date of departure from Québec; projected date of return to Québec; and  postal address outside Québec. In addition, the person had to complete a declaration stating that he or she intended to return to Québec.

As shown on table 1, no more than 3,000 electors registered to vote outside Québec in the 1992 referendum. In 1995, 15,000 electors registered. Their participation rate was close to 80% in both referendums.

Referendum of 1992 1995
Number of constituencies 124 125
Number of electors 3 086 14 818
Yes option 1 343 2 533
No  option 1 089 9 016
Number of valid votes 2 432 11 549
Number of rejected votes 32 168
Total of votes 2 464 11 717
Participation rate 79,8% 79,1%











Since 1980, Québec’s democracy has experienced three referendums on its constitutional future. The Referendum Act and Election Act are permissive regarding eligibility. If one is a Canadian citizen, one can vote even if one has been a resident for only six months. Postal voting is also available for electors residing temporarily outside Québec. Nevertheless, to be a qualified voter, one must demonstrate that one has the will to contribute to the future of Québec and to be a member of the political community. In summary, one has to be a Quebecker, and this notwithstanding a temporary residence outside Québec. For normal elections and for referendums, those are the rules of the game. 

However, the legal and normative rules of the game are not everything in matters of referendums. Issues of political culture, dimensions related to the existence of a form of mutual respect and trust, however thin and limited, between key players, are of fundamental importance. At the time of the two secession referendums in Québec, there was no such trust and respect between the governing authorities of Québec and of Canada.[6] This is the greatest difference between the current Scotland-UK case on the one hand, the past Québec-Canada case and the current stalemate in Catalunya-Spain on the other hand. The politics of referendums requires consistency and as much consent as possible in the circumstances.


[2] It is likely that EUDO Citizenship will study the Canadian case at greater length in the future.
[3] Note that many electors in Québec consider their province to be the “national” level.
[5] 19% of electors used the early voting option in the Québec election that was held on 7 April, 2014. 
[6] See Tierney, p.144 and p.297, for a discussion of the background of the Québec referendums and of the 1998 Belfast (‘Good Friday’) Agreement.


A Catalan perspective: franchise in a forbidden referendum

by Jaume López (Pompeu Fabra University)


My contribution to this debate will offer what might be called "a Catalan perspective" in light of the impending 9 November 2014 vote on the political future of that region/nation.

Let me note at the outset that I agree with Rainer Bauböck’s position on the question who should vote in a referendum on the political future of a region or stateless nation (which includes the decision whether or not to become an independent state). Persons who participate in the regional elections of the territory that aims to change their political status should vote, not potential future citizens of a possible new state.

I agree with his reasoning, and I think it could be also applied to a referendum on the future of Catalonia, even if it is neither a classic case of remedial self-determination, nor a consensual one.  I do think it is a question about upgrading the status of a polity, for the question in Catalonia is a double one: 1) Do you want Catalonia to become a state (which can include a federal or confederal relation in/within Spain) 2) If the first question is answered affirmatively: Do you want Catalonia to become an independent state? The present autonomy can thus be upgraded in different ways. I also think that it has remedial features, for only the continuous negative response of the Spanish state to federal reform and its failure of recognition of the Catalan identity explain the massive increase in support for a referendum and also for independence (about the 50% of the Catalan populationsee here). In Albert Hirschman's terms[1],  the Catalonian voice has repeatedly been ineffective in bringing about reform and this has strengthened the exit option.

The inclusion of two questions locates the Catalan case in between an internal self-determination process that would create a federal state and external one that would create an independent state. The right to decide as a vague concept has played the main role in all the social and political demands. It has been at the centre of some of the most important demonstrations ever held in Catalonia. From my point of view we cannot simply equate an external right to self-determination with a right to decide. I think that the Advisory Opinion of the International Court of Justice on Kosovo's independence provides a better theoretical and legal basis for this new concept of the right to decide.  I disagree on this point with Bauböck. Although it is true that Kosovo was a case of remedial secession, the ICJ opinion offers three exhaustive bases of legitimation: non-productive negotiations, democracy, and peaceful ways, all of which are clearly present in the case of Catalonia. The Catalan sovereignty demands can be better understood as a right of the present demos to decide on its future rather than as a more traditional right to self-determination of peoples and nations. But this is certainly another debate.

Is the view defended by Bauböck the one that is going to be applied in Catalonia in order to define who is entitled to vote in the upcoming referendum? Possibly not. ‘Possibly’, because Catalonia has not yet adopted legislation defining who will be able to vote in the impending referendum. ‘Not’, because a draft bill was tabled on 16 July 2014 with a view to a parliamentary debate in the Catalan Parliament at September, and this draft bill does not draw a clear distinction between regional voters and future citizens.

The fact that, four months before such an important referendum, enabling legislation has not yet been passed by the Catalan Parliament reflects in part the particularity of the Catalan sovereignty process. At present, it is unclear whether a referendum will be held, in view of the Spanish government’s objection which it has repeatedly voiced in the Spanish parliament. The difference between the UK-Scotland and the Spain-Catalonia processes is remarkable and explains, in part, why the Catalan Parliament wants to pass an act authorising 'political consultations'. This strange name is due to the fact that, the Spanish constitution reserves the power to hold referendums to the Spanish state, and prohibits the holding of referendums by regional governments, including Catalonia’s. On 8 April 2014, the Spanish Parliament voted against authorising the Catalan Parliament to legislate for a referendum. 

Article 2 of the Spanish Constitution proclaims ‘the indissoluble unity of the Spanish nation, the common and indivisible country of all Spaniards’, on the one hand, and ‘recognises and guarantees the right to autonomy of the nationalities and regions of which it is composed’, on the other hand. The unresolved tension between these two principles has hampered full recognition of Spain as a pluri-national democracy. In his recent coronation speech, the new King Philip claimed that Spain is a 'united and diverse nation', emphasizing the common idea that Spain is a nation, and not a multinational state.

It is clear that the distinction between holding a consultation and holding a referendum is confusing and can only be explained by the current political context. On the one hand, according to the Spanish Catalan Statue, approved by Catalans in referendum in 2006 and amended by the Constitutional Court four years later, the Catalan government can develop ways to improve democracy and consultation with Catalans. On the other hand, according to the Spanish Constitution, only the Spanish Parliament can hold binding and non-binding referendums (as, for example, the 2005 referendum on the EU Constitutional Treaty). The question is: what does a ‘political consultation’ mean in the Spanish legal landscape? Is it in fact a non-binding referendum that cannot be called a referendum, but could have strong binding effect? This would be the case if a majority of Catalans clearly express their desire for Catalonia to become a new state. 

These elements affect the franchise. First, legislators must try to distinguish between a consultation and a referendum. Thus, the proposed eligibility for participation in the November consultation is not based on the general electoral register (which is ‘owned’ by the Spanish government), but on a new register that draws from all the municipal registers, basing the franchise on residence rather than citizenship. In addition, persons above age 16 will be able to vote, whereas in constitutionally regulated referendums and representative elections only those who are 18 years or older are entitled to vote.   

Furthermore, according to the draft bill, apart from regional electors (Spanish citizens ordinarily resident in Catalonia), citizens of other EU states resident in Catalonia for more than a year and registered therein would be eligible to vote, whereas citizens of non-EU states must satisfy a three years registration period from the day they have obtained a residence permit in Catalonia. Catalans residing abroad may vote in the consultation if they register in a voluntary registry, while Catalans who live and are registered in the rest of Spain will not be able to vote. Catalan MPs may have been inspired by the arrangements made for the Scottish referendum. Drawing a distinction between residents (the requirement of one or three years’ residence) based on their nationality (EU/non-EU) seems very questionable.

I would argue that the proposed franchise is fundamentally mistaken. On the one hand, it seems clear that the nature of a vote on the political future of Catalonia (which includes the possibility of independence) is very different from other types of referendums. On the other hand, it makes no sense that someone who cannot vote for the regional parliament can decide if Catalonia should change its present status.

Although there is no Catalan citizenship as such, the legal status of Catalans is clearly defined in Article 7 of the Spanish Catalan Statute: ‘1. Spanish citizens legally resident in Catalonia benefit from the political status of Catalans or citizens of Catalonia. Their political rights are exercised in accordance with this Statute and the law. 2. Spanish citizens resident abroad whose last legal place of residence was Catalonia also enjoy, as Catalans, the political rights defined by this Statute; their descendants, who maintain this citizenship, shall also enjoy these rights, if they so request, in the manner determined by law.’ Article 7 defines who is entitled to vote in the regional elections. In other words, the Catalan demos. Not the Catalan nation. Not putative citizens of a Catalan state. It is this demos which should have the possibility to vote on any upgrading of the current status as an autonomous community of the Kingdom of Spain, including the possibility to become a new independent state.


[1] Albert O. Hirschman (1970) Exit, Voice, and Loyalty. Cambridge, Mass., Harvard University Press


Polish migrants in Scotland

by Derek McGhee and Emilia Pietka-Nykaza (University of Southampton)


The Scottish Independence referendum is a historic event, as an independence referendum being held in an existing EU member state is unprecedented (Shaw 2013: 13). Independence referendums are unlike other sub-national elections, as they address questions that as Ziegler notes in this Forum ‘are qualitatively different from the issues raised in elections for sub-units of a state’. The fact that the outcome of the referendum could disrupt Scotland's and the UK's continuing EU membership is of particular concern for EU migrants living in Scotland. We will return to concerns over continuing EU membership and other related issues (including the link between sub-national election rights and naturalization) at the end of the contribution. 

This contribution primarily concerns  a particular aspect of what Bauböck (2005) might call EU migrants’ sense of having ‘a stake’, or being ‘stakeholders’. Thus, we examine how some of the Post-Accession[1]  Poles in Scotland (it should be noted that Post-accession Poles are Scotland's largest minority group[2] ) we interviewed[3]  perceive their eligibility to vote in this referendum. What we were particularly struck by in our interviews was the number of our participants who referred to their inclusion as eligible voters in this referendum either as a privilege or as a burden. Both of these perceptions offer opportunities for deepening our appreciation of the experience of being a ‘stakeholder alien’ in this historic referendum.

The participants who perceived their voting rights as a burden struggled with the decision whether to vote in the referendum. There were two main reasons that a number of our participants gave for this contemplation. The first reason was related to the question whether migrants in Scotland have a moral right to vote and decide about independence of a nation state they are not citizen of. The second reason was that migrants did not want to exercise a right that might contribute to an outcome that their hosts might not desire. Thus, their response was a matter of taking on the role of the considerate guest who does not want to be seen to be abusing their host's kindness and hospitality:

I know that I have a right take part in referendum, but do I have a moral right to do so? (…) If someone is asking about my personal opinion whether Scotland should be independent, I would say no, I think it should remain in the UK. But if someone is asking me do I feel that I should decide after 7 years of living in this country? I think I don’t. I think I won’t be voting because I can contribute to the decision that could make them [Scots] unhappy.

Jan, age 57, warehouse worker, Glasgow 

In a sense participants such as Jan are exhibiting a sophisticated understanding akin to Derrida's (2000a, 2000b, 2005) insistence on the impossibility of 'pure' or 'absolute' hospitality. For Derrida, hospitality is precarious and conditional. That is, conditional on the host’s continuing favourable attitude to their guest(s). As well as exposing the migrants' perception of the precariousness of hospitality, Jan is also articulating what Richard Sennett calls a code of honour. Following Bourdieu, Sennett considers honour to suppose that 'an individual who sees himself through the eyes of others, who has need of others for his existence, because the image he has of himself is indistinguishable from that presented to him by other people' (Bourdieu, in Sennett 2003: 55). In this context, being honourable and honouring the host is an expression of gratitude which acknowledges a conditional welcome and the risks of appearing ungrateful (to one’s host), as such gratitude has ‘survival value’ (Komter 2005: 57). From Jan's perspective, his decision not to vote is in a cycle of gift (the vote) and counter-gift (deciding not to vote) that from Jan’s perspective is essential in sustaining social ties and social cohesion (Komter 2005: 57) in his adopted country. 

In contrast, other participants viewed their inclusion in the referendum as a more straightforward and unconditional gift or privilege. That is, as something Scotland has given EU migrants voluntarily without them asking or demanding this right. For a number of participants this gift or privilege was seen as ‘form of gesture’ and recognition of migrants’ presence and contribution to hosts communities. 

I think it’s a form of gesture and a way of showing one’s trust, because, on one hand, I think if one has lived here for a number of years, one should be considered a citizen of this country (..). I think it was a very valid and positive gesture, because no matter how you look at it, the immigrants who come here not only join the army of labour but also settle down here and contribute to the economy, plan their lives here and shape the culture of the country, and so I think they should totally have the right to vote as well.

Marta, 28, Web developer, Glasgow

The gift or privilege of being eligible to take part in the referendum was also associated with the need to reciprocate, 'give back'. This sentiment was expressed by Anna:

I feel that taking part in referendum is my privilege because I am not a citizen of this country. In Poland this is my duty, but here this is my privilege (…) the implications of this referendum are huge, thus this is huge decision. Because I’m eligible to vote I want to learn and know more and be able to decide wisely and responsively. 

Anna, 42, Teacher of German language, Glasgow

Here, the perception of Marta's and Anna’s right to vote in the referendum is regarded 'as a sign of honour, respect, and appreciation' (Komter 2005: 45). This was articulated by another participant thus: “…it makes me feel appreciated that Scots decided that because I live in this country I am eligible to take part in the referendum.” Marek Psychotherapist, 44, Edinburgh. 

There are some similarities between the Polish Migrants who have decided to honour or show respect to their hosts through not voting, and others who feel 'honoured' by what they perceive as the gifted privilege (rather than the right) to vote in the referendum. Both responses feature the necessary ingredient of inequality in the gifting or exchange process which, from a Maussian perspective, leads to those who benefit from the gift wishing to give something back even if they cannot give back an equivalent (Wise 2009: 11). Thus, they reciprocate through voting or not voting, depending on what they perceive to be the proper way of honouring the gifted privilege they believe the Scots have bestowed on them. What is common to both responses is they want to 'do the right thing' with these gifts. According to Wise's reading of Mauss, these exchanges have the effect of turning people outward of producing a more general disposition of trust (Wise 2009: 17). Cheal takes this one step further when he says that the circulation of gifts underpins the moral economy, that is, a 'system of transactions which are defined as socially desirable (that is, moral) because through them social ties are recognized, and balanced social relationships are maintained' (Cheal 1988: 15 and 19). That being said, Jan’s response to the situation is more complex than Marta’s, Anna’s, and Marek’s. Jan did not want EU migrants' participation in the Scottish Independence Referendum to impact negatively on what he perceives to be the current pro-migration attitudes in Scotland. Jan’s response presents a degree of anxiety and powerlessness, which evokes the other side of the migrant experience where, there are concerns that conditional hospitality can turn to hostility, in the context of the unstable pairing of hospitality/hostility (Derrida 2000: 3). 

Many of our participants perceive what they consider to be the potential 'strings attached' to their inclusion in the referendum electorate. Just as Caplow observed, in terms of interpersonal gifts, the majority of gifts are given in order to ascertain and fortify relationships that are deemed important but have not yet been stabilized (Caplow, in Komter 2005: 47). Although we have found that, for a number of our participants who feel the warmth of recognition, honour and being part of this historic process, reciprocation of the perceived gift or privilege of referendum electorate inclusion could well have a stabilising effect, we note that this stabilising effect in terms of the obligations and the compulsion to 'give something back' to Scotland for those who intend to honour the perceived privilege of election right inclusion by voting in the referendum was not in turn articulated in longer-term naturalisation plans. What did emerge in our interviews was a yearning for clarification and certainty in the context of the uncertainty the referendum has generated for EU citizens as to what their status as EU citizens of an independent Scotland will be. Thus, clarification of their 'long-term alienage' (Shaw 2007: 70-71) was more salient than naturalisation for these particular stakeholders in this referendum. 

What we observed was that the Poles’ ‘stakeholdership’ as migrants did not seem to follow Bauböck's assumptions that limited (sub-national) voting rights should lead to naturalization by application (Bauböck 2005: 686)[4].  Shaw notes regarding Bauböck's definition of stakeholder citizens that 'long-term alienage' seems to be excluded as a possibility for migrants (Shaw 2007: 74). On the whole, our participants are intent on and content with remaining EU citizens living in Scotland. With regard to the referendum, they desire to have their legal status (in terms of rights and responsibilities) as EU citizens living in a potentially independent Scotland more clearly articulated in the future.




Bauböck, R. (2005) Expansive Citizenship - Voting beyond territory and membership; PS online, 683-687.

Cheal, D. (1988) The Gift Economy, Routledge: London.

Day, S. and Shaw, J. (2002) European Union electoral rights and the political participation of migrants in host polities, International Journal of Political Geography, vol. 8, 183-199 .

Derrida, J. (2000a)  Hospitality, Angelaki: Journal of the Theoretical Humanities, 5.3, 3-18.

Derrida, J. (2000b) Of Hosptality, Stanford University Press:Stanford.

Derrida, J. (2005) The Principles of. Hospitality, Parrallax, 11.1, 6-9.

Hammar, T. (1990) Democracy and the Nation State, Avebury: Aldershot.

Komter, A. (2005) Social Solidarity and the Gift, Cambridge University Press: Cambridge.

Scotland's Future (2013). Scotland's Future: Your Guide to an Independent Scotland, Scotreferendum.com.

Sennett, R. (2003) Respect: the formation of character in a world of inequality, Penguin: London.

Shaw, J. (2007) The Transformation of Citizenship in the. European Union, Cambridge University Press: Cambridge.

Shaw, J. (2013) Citizenship in an independent Scotland: Legal Status and Political Implications, CITEES Working Paper Series, 2013/34, University of Edinburgh.

Walker, N. (2008) Migrantship and the Deterritorialisation in the EU, European University Institute, Working Paper, http://cadmus.eui.eu/handle/1814/8082, 

Wise, A. (2009) Everyday multiculturalism: Transversal crossings and working class cosmopolitans, A. Wise and S. Velayutham (Eds) Everyday Multiculturalism, Palgrave: Basingstoke,  21-45.




[1] Post-accession migrants are the citizens of the eight countries that joined the EU in 2004 (Czech, Republic, Estonia, Hungary, Latvia, Lithuania, Poland, Slovakia and Slovenia). 

[2] According to the 2011 Census for Scotland, the number of Polish migrants living in Scotland was 55,231. See National Records for Scotland

[3] This was a small-scale mostly qualitative study. We do not intend to generalise findings, nor do we claim that outcomes are representative of all Polish migrants in Scotland. We conducted an online survey, 250 participants completed the survey between April-June 2014, and 24 individual in-depth interviews with Polish Adults, 12 in Glasgow and 12 in Edinburgh, between May and June 2014,. We anonymised the names of all participants.

[4] According to the White Paper, applications for naturalisation will be possible for migrants who can prove they have resided in Scotland for ten years at any time and have an on-going connection with Scotland (Scotland's Future 2013: 496). It may well be that for those who are resident the qualification period may be shorter, e.g. five years.



Scotland’s independence referendum, citizenship and residence rights:
Identifying ‘the people’ and some implications of Kuric v Slovenia

By Jure Vidmar (University of Oxford)


Writing in 1956, Ivor Jennings famously stated: “On the surface … [the right of self-determination] seemed reasonable: let the people decide. It was in fact ridiculous because the people cannot decide unless somebody decides who are the people.”(1) This statement can also be read more broadly and the question can be asked whenever we refer to the democratic ideal of decision-making in accordance with the will of the people – who is that? Voting results can be distorted if the concept of the people is either too exclusive or too inclusive. Ruvi Ziegler argues that in the case of Scotland it is both. In contrast to Ziegler, I will argue that there are in fact two concepts of the people of Scotland: one for the purposes of voting in the referendum, the other one for the purposes of obtaining Scottish citizenship. They regulate inclusion and exclusion differently. 


To illustrate Ziegler’s point, Hamish, who has just graduated in his native Scotland, packed up his bagpipes and moved to Amsterdam, cannot vote at the independence referendum. Nevertheless, should Scotland become independent, Hamish will become its citizen. On the other hand, Slawomir, a Polish plumber who has recently moved to Glasgow, can vote at the referendum. Yet, even if Slawomir wholeheartedly supports Scotland’s independence, he will not automatically become a citizen of the new state. Hamish and Slawomir belong to the people of Scotland for different purposes. I do not think that this differentiation is a bad thing or even uncommon in international practice. I will discuss the Scottish identification(s) of the concept of the people in light of this practice, and also pick up on Ziegler’s point on the implications of the vote for EU citizenship rights.


By reference to Jo Shaw, Ziegler mentions several models of awarding citizenship that have been followed in the wave of post-1990 new state creations. In the territory of Yugoslavia, for example, citizenship was generally extended to all permanent residents of a certain former federal republic. This conclusion needs to be qualified, though. Yugoslavia was a federation that also knew the concept of the so-called ‘internal citizenship’. The latter was not awarded territorially but by bloodline. Slovenia may be a particularly instructive example in light of the 2012 European Court of Human Rights (ECtHR) decision in Kuric v Slovenia. I will first present possible implications of this decision for Scotland and then develop an argument against an overlap between the two concepts of ‘the people’ of Scotland. 


In the 1990 independence referendum, Slovenia enfranchised both its ‘internal citizens’, wherever they lived, and all permanent residents of Slovenia. The outcome was that 88.5 per cent of all those eligible to vote favoured independence. When Slovenia became independent, it automatically extended citizenship to any person who had been its ‘internal citizen’. Permanent residents were given the opportunity and a window in which they could opt for citizenship, but they did not acquire it automatically. Those who did not apply or missed the deadline were simply erased from the registry of permanent residents. The group of the so-called erased residents was thus created, that is, people who lost their previously-acquired residency rights in Slovenia. 


It was Slovenia’s argument in Strasbourg that members of this group were offered full citizenship rights, and that failing to take up the offer was their fault. The Court did not accept this argument and held in para 357:


[A]n alien lawfully residing in a country may wish to continue living in that country without necessarily acquiring its citizenship. As shown by the difficulties faced by the applicants, for many years, in obtaining a valid residence permit, the Slovenian legislature failed to enact provisions aimed at permitting former SFRY citizens holding the citizenship of one of the other republics to regularise their residence status if they had chosen not to become Slovenian citizens or had failed to do so. Such provisions would not have undermined the legitimate aims of controlling the residence of aliens or creating a corpus of Slovenian citizens, or both.


Following this logic, once you have legally established permanent residency, you keep the right of residence, even if the legal status of either your home or your host state changes and, as a result of this change, your new citizenship status alone would no longer give you a right to residence. What matters is that you had the right at the moment of the change of the territorial status. It is notable that the Court established that non-citizen residents enjoy this guarantee under Article 8 of the European Convention on Human Rights (ECHR) (the right to private and family life) in their own right; it does not depend on, for example, a family relationship with a citizen of the host state. 


In the context of the Scottish referendum vote, this decision has implications for the following groups: (i) potential future Scottish citizens residing in rUK (the rest of the UK); (ii) potential future Scottish citizens residing in other EU member states; (iii) UK citizens residing in Scotland who will not opt for Scottish citizenship; (iv) non-UK EU citizens residing in Scotland who will not qualify or opt for Scottish citizenship. As I argue elsewhere, by declaring independence, Scotland would prima facie also exit the EU, unless negotiated otherwise.(2) It is thus not pre-guaranteed that the residency status of these groups would continue to be protected by EU citizenship. However, following the Kuric doctrine, it appears that these categories of people retain their present residence rights as an effect of the ECHR and regardless of what happens with Scotland’s EU membership. The Kuric doctrine thus answers Ziegler’s point that the referendum would also have meaningful effects on the citizenship rights due to uncertainty surrounding Scotland’s EU membership. Regardless of whether Scottish citizenship would carry EU citizenship and whether Scottish citizens would also keep UK citizenship, Hamish will be able to continue to reside in Amsterdam and Slawomir in Glasgow. It is true, however, that the ECHR-effect would only freeze the existing rights.  EU citizenship as such would not be retained via Kuric. If Scotland remains outside of the EU and Hamish does not retain his UK citizenship, he will only be able to continue to reside in the Netherlands, but will no longer be entitled to exercise EU free movement rights in e.g. Germany. 


What does this analysis mean for Ziegler’s point on enfranchisement being at the same time too inclusive and too exclusive? First of all, Scotland is not entirely comparable to dissolutions of socialist federations where the concept of internal citizenship existed. Where it exists, internal citizenship is a point of reference which identifies the core of the people who decide. As shown above, this core can be expanded by permanent residents. International practice is somewhat contradictory and there are no settled rules of (customary) international law that would govern the procedural standards of independence referendums. In the absence of internal Scottish citizenship in the UK, other means of enfranchisement had to be employed. Certainly, it would be possible to argue that everyone born in Scotland should be entitled to vote, especially if this person is ultimately entitled to become a citizen. However, even this solution could be challenged with ‘hard cases’. 


Imagine a person whose parents are both English, she has always lived in England, yet she was coincidentally born in Edinburgh. Why would this person be entitled to vote more than a Polish national who actually lives in Scotland, whose children go to school there and who intends to remain indefinitely? In essence, when you need to identify a group, you inevitably need to draw certain boundaries which can always be challenged by borderline examples. In some instances, the group will be too inclusive and in others it will be too exclusive. What is important, in my view, is that there is no deliberate manipulation of enfranchisement, that the rules of the game are not written in such a way that enables one football team to play with eleven players while the other team plays with only ten players. Scotland’s enfranchisement is defensible. In the absence of the concept of internal citizenship, a variant of a territorial approach appears to be a fair choice. Giving the right to vote to those born in Scotland and in residence elsewhere could also lead to problems. 


Ziegler specifically points to the discrepancy between ‘the people’ for the purpose of voting at the referendum and ‘the people’ for the purpose citizenship. However, this is not unusual in international practice. For example, many new states tend to extend eligibility for citizenship to the diaspora, immigrants and their descendants. As non-residents of the new state and often non-citizens of the predecessor state, these individuals are usually excluded from voting in the independence referendum. Yet, they qualify as citizens later. Once a state is created, it may wish to throw the ‘citizenship net’ broadly and catch non-residents with links to the new state, albeit these links can often be rather loose. Nascent states feel somewhat vulnerable when they first come out of their cocoons and try to keep ties with the diaspora for a number of reasons, political, economic and cultural. Awarding (dual) citizenship to such groups is a way of doing so. 


I would even go as far as to say that a complete overlap between future citizenry and the scope of the franchise at the independence referendum can be problematic. In principle, nothing is wrong if the citizenry ultimately includes people living abroad with loose ties to the (new) state. But including the diaspora too generously in the decision-making process on the future legal status of a territory can distort the results. At the end of the day, it should be, in principle, for the people who live in a certain territory to determine the destiny of that territory. Would it really be legitimate for the future of Scotland to be decided by a Scottish-born person, who feels very Scottish otherwise, but has lived in London or Sydney for 40 years? Should Scotland become independent, good reasons may exist to indeed give this person an option to claim Scottish citizenship. At the same time, good reasons exist why this person should not vote in the referendum. Certainly, Ziegler’s example of Hamish who has just moved to Amsterdam may tempt us to conclude otherwise and say this is different than being abroad for 40 years, but the line needs to be drawn somewhere. Arbitrariness can never be completely avoided, yet it seems reasonable to enfranchise on the basis of slightly modified voting eligibility rules in local elections. Enfranchisement in local elections is territorial, combined with a qualifying citizenship. An independence referendum is an eminently territorial question, so its rules of enfranchisement should also be, in principle, territorial.


In the end, I do not think that ‘the people’ for the purposes of the independence referendum should entirely overlap with ‘the people’ for the purposes of the citizenship of the future state. The decision who are the people is ultimately arbitrary and, on the first sight, illogical: Hamish can be Scottish for some purposes but not for others, and Slawomir the other way around. In fact, good reasons exist for such inconsistencies. In my view, two concepts of the people of Scotland exist: one concept for the purposes of the referendum, another concept for the purposes of future Scottish citizenship. They should be seen as two separate categories. 





(1) Ivor Jennings (1956) The Approach to Self-Government, Cambridge University Press, p. 56.

(2) Jure Vidmar (2014) The Scottish Independence Referendum in an International Context, forthcoming in 51 Canadian YBIL, SSRN Version, p. 25. 




Not All Who Are Enfranchised Need Participate

by Ben Saunders (University of Sterling)


Drawing democratic boundaries is always difficult, but this difficulty is particularly clear when the decision is one that potentially alters these boundaries, creating a new demos. Ruvi Ziegler’s proposal is that the franchise for the independence referendum ought to correspond to the proposed Scottish citizenship. If there are people who will be offered citizenship, then it seems that they have a stake in the issue and ought to have a say in the referendum. Conversely, if people are given a say in the referendum, then it seems that they have a good claim to citizenship as well.

Though this seems prima facie attractive, I share the worries voiced by others, notably Rainer Bauböck. Further, Ziegler’s proposal seems to assume that citizenship and the right to vote should go hand in hand, yet there is no logical nor obvious moral reason why this should be the case.[1]  The particular package of rights traditionally associated with citizenship is essentially a historical accident. We may wish to give the vote to some who will not be given citizenship, because we do not wish to accord them other rights that would be afforded by citizenship.

Some hold that everyone affected by a decision should be enfranchised in it, though this has radical consequences. A less radical proposal, which Dimitrios Kyritsis seems to accept, is that only those who will be subject to coercion as a result of a decision need be enfranchised. Both the ‘all affected’ and ‘all coerced’ principles suggest a franchise wider than current citizens or residents.[2]  However, neither principle requires that we grant these people other rights that are usually attached to citizenship, such as the right to enter and remain within our community.

I am inclined to think that residency is more important than citizenship in determining who should be enfranchised. Those who are long-term residents in Scotland should have a say over its future, since they will be part of that future, even if they are not (at least immediately) offered citizenship. If this is correct, then the Polish immigrants interviewed by Derek McGhee and Emilia Pietka-Nykaza need not feel that it is not their place to participate in the decision; provided that they are settled indefinitely in Scotland, and are not merely transient visitors, then they have a right to be included.

What McGhee and Pietka-Nykaza’s survey does highlight, however, is that we may distinguish between those who should be given the right to participate and those who should actually participate. Clearly, it makes no sense to say that someone ought to vote if she is not afforded the opportunity to do so, but it is possible that someone afforded the opportunity to vote ought not to exercise it.

Consider someone who moved to Scotland for a one-year work contract (or perhaps a Master’s degree) in October 2013. Suppose that she will definitely leave in October 2014 and has no intention of ever returning. This person, if an EU or Commonwealth citizen, will be entitled to vote in the September 2014 referendum, yet it seems plausible to say that she has no business doing so, given that she does not expect to have anything to do with an independent Scotland (which, by the Scottish Government’s own timetable, will not emerge until 2016 even if the vote is for independence).

I am not suggesting that she ought not to be given the vote. It would be difficult for government officials to determine who should and who should not be given the vote on grounds such as these. Rather, my suggestion is that, even if she is given the vote, perhaps she ought not to exercise it, on principled grounds. This is not an absolute or all things considered judgement; it might be that other reasons make it at least defensible for her to vote. But, absent other considerations, it may be that she ought not to participate in the vote, even though she has the right to do so (and perhaps even ought to have this right). Though low levels of democratic participation are frequently lamented, it has been argued that we ought to abstain, on principled grounds, when we are indifferent to the alternatives on offer.[3] 

It might be objected that this argument unreasonably assumes that voters are motivated solely by consideration of their own interests. If voting is better understood as an attempt to identify what justice demands, then there may be no objection to including more people in the vote if they help us to succeed. However, it is not clear that Scottish secession is a case where justice demands one particular answer. Further, even if justice does dictate what should be done, it need not license others to interfere in a community’s decision-making, any more than I can interfere with your conduct whenever it is immoral. Sometimes interference may be justified, but sometimes people have a right to act wrongly.

Robert Goodin has argued that we ought, ideally, to enfranchise every individual in every decision, in order to be sure of including those affected.[4]  Over-inclusiveness, he suggests, is less troublesome than under-inclusiveness, since if those who are unaffected vote randomly then they are likely to cancel out, leaving the matter to be decided by the votes of those who are affected. Unfortunately, unaffected voters cannot be trusted to vote randomly or to abstain; there is a danger that if we extend the franchise too widely then some will use the political power they are given to impose their preferences on others. Nonetheless, perhaps it is better to err on the side of generosity when allocating voting rights, since those with the right can always abstain, while those without it may be deprived of a legitimate voice.



[1] Heather Lardy (1997) ‘Citizenship and the Right to Vote’ Oxford Journal of Legal Studies 17(1): 75-100.

[2] Sarah Song (2009) ‘Democracy and Noncitizen Voting Rights’ Citizenship Studies 13(6): 607-20.

[3] Paul Sheehy (2002) ‘A Duty Not to Vote’ Ratio 15(1): 46-57.

[4] Robert E. Goodin (2007) ‘Enfranchising All Affected Interests, and its Alternatives’ Philosophy & Public Affairs 35(1): 40-68.




Different boundaries - different meanings

By Vesco Paskalev (University of Hull)


Ruvi Ziegler initiated this forum debate with two claims – that all prospective citizens of a putative state should be considered as stakeholders in its coming into being, and that there should be congruence between them and the persons enfranchised to vote in the independence referendum itself. While the stakeholding claim seems more or less acceptable to most subsequent contributors, the congruence claim was intensely contested by almost everyone. This is surprising, since the latter seems intuitively appealing, supported by international law Ziegler quoted and also fits well into the mainstream normative theory. On the other hand, the argument of Rainer Bauböck and others that the consensus behind the current franchise should be respected is very powerful too. From the beginning of the debate I felt that both sides are right, each with regard to a different referendum, and this was made obvious by the dilemma of the Poles resident in Scotland discussed by McGhee and Pietka-Nykaza.

Bauböck emphasises the normative significance of the difference between "the putative demos of an independent Scotland" and "the existing demos of Scotland as part of the UK". He argues that, currently, only the latter can hold a referendum which is democratically legitimate. There is a difference indeed, but this means that the referendum question: “Should Scotland become an independent country?” can be interpreted in two different ways. “Shall the Scottish nation become independent?” Or: “Shall Scotland decide upon its future status independently of the UK?” I suspect that the independence movement aims to ask the former question [1].  The vote for independence of Scotland is a matter of self-determination of certain people, and not merely as an upgrade of the status of certain territory, in the way that a referendum in say, Yorkshire, would be. Further, Scottish independence is qualitatively different from devolution, which is a good reason not to enfranchise the same people for both decisions. However we define 'people' in general and 'the Scots' in particular, if the referendum is to determine whether such a subject should become independent, it is precisely for the putative members of this subject to decide. This question is ontological and the putative demos comes into being by the act of the vote itself. If this is the question, the franchise should be narrowly tailored to minimise the discrepancy between the people for the purpose of self-determination and the people for the purpose of subsequent politics. On the other hand, the upgrade of the status of a territory is mostly utilitarian question – whether it would be better for an existing unit to separate or not. In such a referendum it is the people living in Scotland rather than the Scottish people that should be enfranchised. The questions may be similar, but the make-up of the enfranchised population would be identical only if 'Scottish people' are defined as the 'people living in Scotland'. If this were the present case, all current residents should become prospective Scottish citizens automatically and not merely given opportunity to naturalise. Thus, whichever of the two questions the SNP intended to ask at the referendum, it got it wrong.  

The odd position of the Poles resident in Scotland who have found themselves enfranchised reveals the flip side of the same coin: if a significant number of people who are not (and do not aspire to be) Scottish can vote, the answer which the referendum will yield on 18 September will be irrelevant for the ontological question. When too many putative Scots are excluded and too many putative non-Scots are included, the referendum can no longer be a legitimate act of self-determination of a people. However it will be still a legitimate answer to the territorial upgrade question. It turns out that the boundaries of the franchise determine the meaning of the result, and – given the necessarily cryptic wording on the question on the ballot – which of the questions is being asked. Who can vote effectively determines what exactly they are voting for. While presumably the Scots want to vote on the self-determination question, they are given the opportunity to answer on a territorial upgrade question. 

One may wonder if the distinction makes any practical difference at all. It certainly matters for the future determination of Scottishness, but as Bernard Ryan noted, for the most part this would be decided not by the referendum itself but by the prospective Scottish constitution and other foundational documents afterwards. Yet at the very least it may affect whether people would vote and how they would vote. For example some hard core nationalists may boycott and denounce a referendum if 'all those Poles' are included, while liberals may vote 'yes' if foreign residents are included and 'no' if they are not. The framing of the question is not only a theoretical concern, and as we saw, the Polish themselves are actually thinking hard over it [2].  With regard to this, movements for self-determination are well advised to make explicit which of the two questions they would like to put to test, and enfranchise the respective persons accordingly. This is a matter of political choice too.



[1] I would attribute the agreement of the SNP which initiated the referendum to this enfranchisement scheme to either failing to notice the difference, or – more likely - strategic calculation with regard to the outcome. Yet according to Jo Shaw many documents of the Scottish government support the interpretation that the question is one of territorial upgrade and its aim is not to break all of the 'unions' that hold the UK together, but only the political union and some aspects of the economic union. As the two questions were not explicitly distinguished, the different voters may be actually answering different questions.

[2] While I am sympathetic to the distinction Saunders makes between those who are entitled to vote, and those who ought to do so, I think that the very fact that certain people are enfranchised while others are not is already meaningful and affects the others, even if the former actually abstain. 



Types of Membership, Types of Processes and Types of Politics

By Dora Kostakopoulou (University of Southampton)


Ruvi Ziegler makes an important as well as interesting point in his kick-off contribution. By reflecting on the actual membership of the ‘single issue demos’ which will decide the future of Scotland on the 18th of September 2014, on the one hand, and the putative, and rather sketchy at this point, membership of an independent Scottish commonwealth, on the other, he notices the incongruence between the two. He has fully grasped the historical and political context, which Bernard Ryan also eloquently outlines, but he is convinced that the franchise in the Scottish independence referendum should not have reflected the criteria deployed for political participation in local government elections (Section 2 of the Representation on the People Act 1983). Instead, it should have followed other criteria since ‘independence referendums share the fundamental and long-term characteristics of national elections, and their significance is enhanced by their capacity, from both a national and international law perspective, to alter the legal landscape for individual citizens’ (kick-off contribution). 

Why Congruence?

Ruvi Ziegler furnishes two arguments in favour of congruence. The first is that the principle that ‘all affected interests should be considered’ is not complied with since individuals who will be deemed to be ‘ab initio’ citizens should Scotland become independent are excluded from casting their vote on the 18th of September. The second argument is that the under-inclusive character of the present arrangement undermines the legitimacy of the referendum – an argument that Dimitrios Kyritsis does not share. Ziegler notices the paradox that the ad-hoc referendum demos excludes persons who will be eligible for citizenship on day one of independence while the proposed citizenry of a future state will exclude those persons who are eligible to participate in the constitutive political act of establishing the new state. This leads him to argue that ‘the determination of the franchise for the Scottish independence referendum was ill-conceived’ (kick-off contribution). 

Ziegler’s reasoning is both plausible and insightful. It is plausible because we are confronted with a paradox. It is also insightful because, notwithstanding the special role of political contexts and historical processes, it is conceivable that political scientists and lawyers can develop clear and normatively justified criteria for franchise in independence referendums. Having said this, however, Ziegler’s argument also rests on two presumptions which may not be universally shared. I will call these i) the time-continuum and ii) the desirability of monism presumptions.

The time-continuum presumption projects a linear temporal line among the independence referendum, day one of independence and everyday political life post-independence. By so doing, it underscores the different political acts, processes, stages and politics that are involved in the transition of a political unit from regional self-determination to state formation to polity functioning and state maintenance. It thus subsumes multiplicity into an overarching monism; one type of political process, one type of politics, and one type of membership and electoral participation.      

Types of Process, Types of Politics and Types of Policy

Neither historical manifestations of a political unit’s secession nor imaginative constructions of state formations in contractarian political theory, such as, for example, the Hobbesian formation of a sovereign state, conflate the constitutive act of the formation of a state with law and policy making by, and within, the (new) state. The former is a state-generative act or an act of constitutive politics since it emplaces a political structure. Following the establishment of a state, a different political process, which could be either inclusive or elitist in character, normally commences with a view to designing a complex array of policies and their generative structures. Foreign policy and external action, distributive policies, regulative policies and constituent policies will be enacted thereby shaping the state’s functioning and maintenance. Defining who will be a citizen, or will be worthy to become citizen, is a constituent policy. And it is normally politics (and ideology) which will determine the scope of policies, including the scope of constituent policies, such as the citizenship policy of the new state. In other words, there is no continuum between ‘the constitutive’ and ‘the constituent’ and there is a lot of writing, a lot of re-writing and a lot of creativity in institutional design and policy formulation and implementation post-independence. 

All this is to say that ‘framers’, that is, those who will vote in the independence referendum, do not have to be identical with ‘the deciders’ in an independent Scotland. Similarly, having chosen the residence-based option for franchise on the 18th of September, there existed no obligation on the part of the Scottish Government to outline the scope and content of future policies in detail in a document.  The legitimacy of the referendum would not have been undermined if Scotland’s Future: Your Guide to an Independent Scotland, which was published by the Scottish Government in November 2013, contained no explicit, or very ambiguous, references to the content of the citizenship law of the newly independent state. Nor do any putative claims for inclusion into the body of citizens by Commonwealth citizens and EU citizens resident in Scotland derive their normative force and political weight from the fact that these persons will take part in the independence referendum. Such claims would have to be premised on the normative force of democratic considerations which make residence and participation in the socio-economic life generative of the entitlement to participate fully in the political sphere and to authorise the laws which govern one’s affairs. Their right to vote in the referendum stems from the above premise and could thus be a supplementary ground in favour for their inclusion into the permanent Scottish demos following independence.   

In the light of the foregoing discussion, the different political processes and politics involved in polity transitions must be put in proportion and perspective. True, a discussion of who should be part of the people of Scotland is rather premature at this point for the reasons that both Rainer Bauböck and Bernard Ryan outline in their contributions, but if predictions or a critical examination of what has been suggested thus far can be made at an institutional level, then one has to recognise the possibility of more than one pattern or policy option as well as the possibility that what might be chosen in the eighteen-month period that will follow a ‘yes’ outcome on the 18th of September might contain significant variations from what was proposed in November 2013. This is how politics works and almost any generalisation concerning policy design often proves to be inapplicable to most of the cases of concrete policy formulation. 

This is not to say that academics and policy practitioners should not engage with questions concerning the (rightful) membership of the Scottish demos and provide advice about the content of the future Scottish citizenship law and policy. Rather, one needs to recognise that any such intellectual endeavour will be unavoidably normative in the same way that any really good citizenship theory is unavoidably normative. And while scholars are mainly interested in neat designs and are attracted to settled patterns and the elimination of framework ambiguities, real politics is messy, complex and unpredictable. The making of a real law is almost never a linear path from point A to B, but an act of producing a mosaic where multiple models and different patterns are brought together in a single design. And the Scottish citizenship law mosaic remains yet to be configured.    

Different Membership Criteria for Different Demoi?

The different political moments involved in the Scottish independence story, the different political processes and types of politics unavoidably yield different types of ‘we, the people’ for the purposes of state formation and state functioning or maintenance following formation. While congruence between the membership criteria of the referendum demos, on the one hand, and the Scottish demos is not necessary since the connection between the types of demoi is a loose one, it is still desirable and important to reflect on the existing criteria of membership, characterise policy choices as good or bad and, generally speaking, to consider important questions about political membership, inclusion and democracy. This is precisely what Ziegler has invited us to do. If we value consensual and inclusive political processes and open and democratic politics, the choice of policies (of a citizenship policy in this case) becomes more limited. I fully agree with Ziegler that in formulating policies, governing elites need to: a) be inclusive; b) be as consistent as possible and c) give a political voice to all those who have made Scotland the hub of their lives and will be subject to Scotland’s jurisdiction on day one of independence – thereby according priority to democratic, as opposed to ethnocentric, considerations. Congruence between the different demoi would thus be normatively desirable; the Scottish citizenship law and policy should mirror the residence-based approach of the Scottish referendum franchise. 




Puerto Rico:  the Referendum Strategy and its Discontents

By Jaime Lluch (University of Pennsylvania)


Although I have some expertise on Canada (Quebec), the U.K. (Scotland), Spain (Catalonia), and Italy (South Tyrol and Valle d Aosta), I want to discuss in this forum the case of Puerto Rico.

On 6 November 2012, creative politicians in Puerto Rico arranged to hold a plebiscite on the territory’s future constitutional status. In the civil law tradition of Puerto Rico, the term used is “plebiscite,” not referendum. This was the fourth plebiscite of this sort following those held in 1967, 1993, and 1998. Politicians in Puerto Rico are now engineering a fifth plebiscite to be held sometime in the next two years. These have all been constitutive referendums in the sense that they sought to invoke the constituent power of the people to (potentially) adopt a very different constitutional status. Contrary to the upcoming referendum in Scotland, or the two ones that were held in Quebec, these plebiscites were not strictly speaking “independence referendums.” As defined by Ruvi Ziegler, the latter are referendums that may result in the succession of states, i.e., the “replacement of one state by another in the responsibility for the international relations of territory.” Instead, we can call them self-determination referendums, since the question posed to the voters has always presented all three major constitutional status options. Also, one of the constitutional options (and one which has been gathering electoral strength in the last few decades) involves asking the central state to accept Puerto Rico as one of the constituent units of the federation. From the point of view of the U.S. state, this involves state expansion, not state contraction, and thus the potential political effect is the opposite of a successful independence referendum. 

Finally, at no point has the choice for the voters been “independence yes or no” or “federation yes or no.” In that sense, the Puerto Rican plebiscites are more like the proposed Catalan referendum than the Scottish one. In Catalonia, the proposed text of the referendum is: “Do you want Catalonia to become a State?" and "In case of an affirmative response, do you want this State to be independent?" Although the first question is rather unclear and ambiguous, it would seem that the two questions taken together would offer voters the chance to vote for either the current Spanish State of Autonomies, federation, confederation, or independence. Jaume Lopez seems to recognize this in his contribution, although he casts a less critical eye on the ambiguity of the first question than I do.  By contrast, the referendums in Puerto Rico have been straightforward and clear in the choice presented to the voters. However, there has always been considerable political jostling regarding who defines the constitutional status options, and how the different formulas (federalism, autonomy, or independence) are defined. 

The most vital issues in Puerto Rico regarding these referendums have involved culture and national identity (both in the USA and Puerto Rico), the nature of the federal political system of the USA, the rigidity of U.S. constitutionalism, the nature of the post-1952 political status and its limitations, and the procedural aspects of these referendums, in particular whether the U.S. Congress is willing to provide for a federally-sponsored, binding vote, with the constitutional formulas pre-approved by Congress. 

After four plebiscites, nothing has fundamentally changed in Puerto Rico’s political status since 1952. For several years now, there has been an emerging consensus among political and social elites in Puerto Rico that the criollo plebiscite route is probably exhausted. Elites from all the political persuasions are now converging on the idea that instead of a plebiscite, a Constitutional Assembly on Status should be convened. Such an assembly would serve as a deliberative body, representing the people of Puerto Rico. It would seek to elaborate clear formulas for the three major constitutional options and would then demand from the U.S. Congress a clear commitment to the three formulas it is willing to support and eventually to follow through with a federally-sponsored solution. 

Surprisingly enough, the issue of the franchise for the vote in these referendums and the question of who would qualify for citizenship ab initio in the case of a victory of the independence option have not been prominent in the discussion either in the public sphere or in academic commentary on these issues. Nevertheless, the two issues that are discussed in this forum regarding who should have the right to vote (i.e., the legitimate franchise) in  such referendums, and who should be the citizens of a newly independent state, are intrinsically interesting and relevant to Puerto Rico. 

Regarding the first question, I agree with Rainer Bauböck that, where self-determination has been agreed between a territory and the central government, the only legitimate franchise in constitutive referendums is the currently existing one for the territory in question. Indeed, in the plebiscites held in Puerto Rico, the enfranchised persons have been those residing on the island and who are registered on the rolls of the Electoral Commission of Puerto Rico, usually because they participate in local  elections. 

In these plebiscites, there has been of course a rich discussion in the public sphere, and one of the questions posed has to do with who should have the right to participate in such momentous occasions. There are more than 4 million first and second generation Puerto Rican immigrants in the USA, and it has been suggested that perhaps they should be included in the franchise. However, as a matter of law, this has never been accepted and the Puerto Rican demos has been defined as consisting of those U.S. citizens who reside in Puerto Rico. Thus, Cuban immigrants, Dominican Republic immigrants, a sprinkling of European immigrants, and mainland U.S.-born immigrants, can vote and have voted in these plebiscites, as long as they are U.S. citizens registered in the electoral rolls of the Electoral Commission. The situation might be different if the demographics in Puerto Rico were similar to those in Hawaii: by the time Hawaii was accepted as a state in 1959, the native Hawaiians were a minority, and the majority were continental U.S.-born residents. 

No one has even tried to suggest that the other 317 million U.S. citizens living in the 50 states of the federation should be entitled to vote in these plebiscites. Yet in Spain this position has been seriously put forward by the Popular Party and its think tank FAES. They have argued that if they were to allow a referendum on the future of Catalonia, all 47 million Spanish citizens should be able to vote. In the case of Puerto Rico, at least three of the four political status options that Puerto Ricans have, require the consent of U.S. Congress: becoming a state of the USA federation, acquiring a free association status or some kind of enhanced autonomy would have to be approved by Congress. Even a negotiated form of independence would depend on the U.S. legislature’s good will and consent as a practical matter. Thus, no matter what Puerto Ricans decide in their criollo plebiscites, their collective will is subject to whatever substantive conditions and procedures the U.S. Congress decides to impose. 

This is not surprising, since Puerto Rico’s history is one of back-to-back colonialism. Spain created this state of affairs in 1493 and the U.S. took over at the close of the 19th century. Many would say that the status inaugurated in 1952 continues to be “colonial”. From a social science perspective, this is correct in my opinion. Thus, Puerto Rico has an inalienable right to self-determination. 

Moreover, as in Scotland, Catalonia, Euskadi, and Quebec, it would be a mistake to refer to these territories as mere regions, when in fact they have all the accoutrements of nationhood: a large proportion of the population in these nations exhibits national consciousness. As the president of the Puerto Rican Independence Party Rubén Berríos Martínez expressed it in a testimony before U.S. Congress: “The problem of Puerto Rico... is not a problem of disenfranchisement of a minority or an issue of civil rights, as some people believe. It is not a problem of individual rights. It is a problem of national rights, of the inalienable rights of a nation, of a people, to govern themselves.”

Regarding the second question, who should be the citizens of a newly independent state, this question has not been very relevant in recent years in Puerto Rico. Independentism was an important force during the first half of the 20th century but in the last few decades independentism has been at an all-time low point, barely reaching 5% of the electorate. Debate in the public sphere and academic commentary has tended to ignore this issue of who should be the citizens of a newly independent Puerto Rico.

A related question is more relevant for contemporary Puerto Rico. If Puerto Ricans were to choose a form of genuine free association as their preferred constitutional status, would the residents of Puerto Rico be able to maintain their U.S. citizenship and transmit it to the next generations? Puerto Ricans hold U.S. citizenship by virtue of a 1917 federal statute: theirs is not the usual Fourteenth Amendment citizenship. The Report by the President’s Task Force on Puerto Rico of 2005 cast doubts on whether they could do so [1],  while a leading constitutional law scholar in Puerto Rico [2] has argued that Puerto Ricans could retain their U.S. citizenship, could probably hold dual citizenship under a formula of free association and could continue transmitting their U.S. citizenship to their offspring.

Below, I give some further background on these plebiscites and the existing autonomy of Puerto Rico.

Background: The territorial autonomy of Puerto Rico

Since 1898, Puerto Rico has been an “unincorporated territory” of the United States, and the nature of its relationship with the U.S. has been set by federal statutes, especially the Foraker Act of 1900, the Jones Act of 1917, and the Federal Relations Act of 1950-1952. Although the U.S. Constitution provides for “states” and “territories,” the category of “unincorporated territory” was sculpted by the U.S. Supreme Court. Puerto Rico has a very peculiar form of territorial autonomy within the United Sates. It is part of the wider U.S. federal political system, but it is not one of the constituent units of the U.S. federation. Its autonomy is called “Estado Libre Asociado” (ELA), or free associated state, but it is neither free nor associated nor a state. There are now less than 4 million people on the Island and another 4 million plus Puerto Ricans on the U.S. mainland, many of whom circulate back and forth. Puerto Ricans are U.S. citizens by virtue of the 1917 federal statute, but they cannot vote in U.S. federal elections.

Autonomies such as Puerto Rico are non-federal arrangements because they are constitutionally subordinate to the center [3].  The “shared rule” component between the central state and the autonomous unit is weak or practically non-existent. The power to terminate or modify the Puerto Rico-USA relationship rests squarely within U.S. Congress. During 1952-53 the U.S. succeeded in getting Puerto Rico off the agenda of the UN Decolonization Committee in part by arguing that the Estado Libre Asociado was a compact of a bilateral nature whose terms may only be changed by common consent. However, soon thereafter Congress and the Executive branch started to behave “as if no compact of any kind existed and as if Puerto Rico continued to be a territory or possession of the United States, completely subject to its sovereign will. Puerto Rico leaders would spend the rest of the century unsuccessfully trying to convince the United States to allow full decolonization” (Trías Monge 1997) [4].  Supporters of the ELA have tried on several occasions to negotiate a “culminated ELA,” starting with the Fernós-Murray bill of 1959, but they have been unable to obtain the consent of U.S. Congress. However, Congress continues to assume that it can unilaterally exercise plenary powers over Puerto Rico under the territorial clause of the U.S. Constitution, and the U.S. government contends that sovereignty over Puerto Rico resides solely in the U.S. and not in the people of Puerto Rico.

For well over a century, politics in Puerto Rico has not been about political economy or the relation between the state and society. Instead, it has mainly been a debate about how the territory will relate to the U.S. federal political system and about how different conceptions of national identity shape this debate. These debates are dominated by three well-defined political orientations in the political party system: independentist, autonomist, and federalist (in the sense of becoming the 51st state of the USA).

Previous Plebiscites of 1967, 1993, and 1998

In all the previous plebiscites, the choice offered to the electorate has been to decide between independence, autonomy or federalism. In 1967, autonomism won with 60% of the vote, while the independentists boycotted the event. In 1993, 48.6% voted for autonomism, 46.3% for federalism, and 5.1 for independence. In 1998, 46.6% voted for federalism, 2.6% for independence, 0.3 for free association and 50.5% voted for “none of the above.” 

The plebiscite of 6 November 2012

Several years ago the independentist party offered a proposal for holding a plebiscite with two rounds. In the first round, the people would be asked whether they wanted to continue living under the present (colonial) “unincorporated territory.” If the answer chosen by a simple majority was “No”, then there would be a second round (months later) in which the people would be asked to choose among the three non-status-quo options, following an educational campaign. The idea was to narrow down the options to the ones that would decolonize the polity, given that the U.S. Congress had never shown an interest in doing so, or even in holding a federally-sponsored plebiscite in Puerto Rico. 

In 2012, the federalist party in power in Puerto Rico at that time modified the idea of a two-step referendum and proposed instead a two question referendum, to be held on the same day as the regular elections for selecting both the new Governor and the new legislature (with a view to boosting their support in the regular elections).

Thus, in the plebiscite of 6 November 2012, the people were asked two questions. The first was: “Do you agree that Puerto Rico should continue to have its present form of territorial status?” In the second question, they were asked to choose between federalism, independence, and a “sovereign ELA”, which is a light version of a genuine status of free association. The autonomist party actively and energetically campaigned for a Yes vote on the first question, and on the second question for a “blank vote.” The results showed that 54% of the people voted No in the first question – a No vote in this context suggesting a vote for change. For the first time in their history, Puerto Ricans voted to show their disapproval of their present political status. This was the most important result of this event.

Among the choices offered by the second question, federalism received 61%, sovereign ELA 33.3%, and independence 5.5%, but there were 480,918 blank votes, so if those votes were to be counted, federalism received only 46% of the vote.

U.S Congress, as the legislative branch of the central state government, needs to take up its responsibility to end the current territorial status of Puerto Rico. In light of these results, another vote should be taken among these three options, but this time in a Congressionally-sponsored plebiscite.

Every single one of the four plebiscites that have been held in Puerto Rico has been a locally-sponsored one. U.S. Congress has never agreed to provide for a binding referendum. This is not just a matter of respecting and implementing the result of the vote, but of exercising its responsibility as the dominant power. Moreover, only U.S. Congress can define what sort of conditions would be imposed if the people want to join the federation, what kind of transition period there would be were independence to be chosen, or what sort of expansive model of autonomy U.S. Congress would be willing to grant Puerto Ricans. 



[1] President’s Task Force Report on Puerto Rico (Washington DC, Government Printing Office: 2005).

[2] Álvarez González, José Julián, 'The Empire Strikes Out: Congressional Ruminations on the Citizenship Status of Puerto Ricans', 27 Harvard Journal on Legislation 309 (1990). 

[3] For further discussion of this point, please see Jaime Lluch, 'Autonomism and Federalism', in Publius: the Journal of Federalism, Vol. 42(1) Winter (2012).   

[4] Trías Monge, José, Puerto Rico: The Trials of the Oldest Colony in the World (New Haven: Yale University Press, 1997).


Who can vote on a referendum and who can be granted nationality of new states? Theory, practice and interests

By Vincent Laborderie (Université Catholique de Louvain)

In this contribution, I will move away from the Scottish case and generalise about who is entitled to vote in the context of an independence referendum. I shall therefore discuss different theoretical possibilities and see how they were applied in previous referendums in Quebec and Montenegro). Unlike most contributors on this subject, I will not focus on what attitude is normatively preferable but rather on the feasibility of different options and their impact on the referendum outcome. These two questions are, as we will see, closely linked.

The last section of this contribution will deal with the issue of citizenship eligibility. In disagreement with Ruvi Ziegler’s kick-off contribution, I will argue that this question should not be linked with the referendum franchise.

1.Theory about franchise in independence referendums

In this contribution, I will use the phrase “host state” to identify the state in which a regional independence referendum took place or will be organised. This region will be referred to as a “secessionist region”. 

Concerning the franchise issue, the question at stake could be the following: 

Should the most pertinent criteria be residence in the secessionist region (including foreigners) or citizenship (including “nationals” living abroad)?

Basically we can distinguish three groups of people concerned by the question at stake. The first one relates to citizens of the host state living in the secessionist region. This group must, without any question, be able to vote in an independence referendum. They represent the core of the electoral body. Participation of the two other groups is more questionable. 

The second group consists of people living in the secessionist region but who are not citizens of the host state. The third group is formed by citizens of the host state who are linked to the secessionist region but who live abroad or in another part of the host state. Including the second or third groups in the electoral body of an independence referendum depends on which criterion is taken into account: nationality or residence. 

Further criteria could specify which members of these two groups are allowed to participate in the referendum. For foreigners living in the secessionist region, a certain period of residence in the region could be required. The nationality of these foreigners could also be a criterion – as it is, for example, in the Scottish referendum where only EU and Commonwealth citizens could vote. 

Concerning nationals residing outside the secessionist region, time and space criteria could be applied. For having the right to participate, one must have left the secessionist region for less than a certain time. Current residence can also be a criterion. The most relevant distinction here is certainly between people who live in the host state but not in the secessionist region and people who live abroad. 

2.Franchise practice in Quebec, Montenegro and Scotland

Far from being only theoretical, the scope of the franchise can determine the outcome of the referendum and, therefore, the future of host states, secessionist regions and of their inhabitants. I will refer to the franchise criteria used in referendums that occur in a context comparable to Scotland and Catalonia (or any other part of the European Union), i.e. in economically developed and democratic countries. Historically speaking there are very few instances. The old cases of Norway (1905) and Iceland (1944) will not be of any help. The questions being debated in this forum were irrelevant at the time of these votes. But taking these cases into account helps us to realize that questions about the franchise are closely linked with the process of globalisation and of massive migration that can be observed all over the world and especially in Europe. 

Recent cases are limited to the two Quebec referendums (1980 and 1995) and the Montenegro referendum (2006) [1]. We can add the two referendums scheduled for 2014 in Scotland and Catalonia. But only the former will take place for sure and franchise questions have already been settled there. 

Concerning citizens of the secessionist region residing elsewhere, we observe that they get the right to participate in an independence referendum only in Quebec. But criteria are restrictive. As pointed out by Guy Laforest and Eric Montigny in their contribution, “they have to demonstrate that they left Québec temporarily” and for less than two years. The main reason for this situation is probably the difficulty to distinguish between “citizens” of the secessionist region and those of the rest of the host state. As Jure Vidmar points out in his contribution, internal citizenship was specific to socialist federations, such as Yugoslavia and the USSR. For nationals living abroad, the criterion would be their last residence in the host state. For nationals living elsewhere in the host state it is less obvious how to take into account this criterion. Should persons who were born in the secessionist region be considered? How about those who left their region for less than a certain period or those who have relatives there?

Before the Montenegro independence referendum (2006), Serbia presented to the Venice Commission a list of 260.000 “Montenegrin citizens” living in Serbia. The majority of them would have probably voted against independence. Since only 460.000 voters were recorded in Montenegro, participation of these “Serbian-Montenegrins” could have had a decisive effect.[2] 

We touch here upon an important issue: advantages for “yes” and “no” camps to include different groups of voters. Coupled with the difficulty to determine who among citizens of the host state should count as a citizen of the secessionist region, this is probably the reason why people outside secessionist regions are so rarely authorised to participate in the referendum. 

Whereas in most cases the majority of people living elsewhere in the host state would vote against independence of the secessionist region, the preferences of people living abroad are more uncertain. But here one faces the problem of material and practical organisation. Most secessionist regions do not have facilities abroad to organise the vote of expats and would be able to build such capacities only in a few countries. So they would be obliged to use facilities of the host state. Even if, as in the Scottish case, an agreement with the host state exists, there might not be enough confidence to let central state authorities organise the vote. 

Concerning non-citizen residents in a secessionist region, other contributions offer various arguments for and against offering the franchise to them. I will not enter this discussion but rather underline that, also in this case, independentist forces would be better served by a restrictive approach. This is illustrated by the second Quebec referendum in 1995. The franchise in the independence referendum was restricted to Canadian citizens. Indeed, the Quebec government, organising the referendum without any participation of federal authorities, assumed that foreigners would vote against independence. It seems that the Canadian government shared the same belief as it offered the Canadian nationality to an unusually large number of foreigners residing in Quebec just before the referendum [3]. This is partly reflected in the outcome. Only 33% of voters in Montréal, favoured independence in the 1995 referendum whereas the ‘yes’ option got a total of 49.5% for the entire Quebec province. Even if the Anglophone community represents a substantial proportion of the population of Montréal, it is obvious that the large number of immigrants living in the city played a decisive role in this outcome. In his famous and shocking speech after the referendum, Jacques Parizeau, leader of the “yes” camp, proclaimed that they had been defeated “essentially by money and the ethnic vote”.

This very pragmatic empirical observation can be supported by research which points out that support for independence is closely linked to level of identification with the secessionist region. [4] We can assume that newcomers and non-citizens identify less with a secessionist region than natives of that region. 

Decisions about the franchise are strongly influenced by the advantage each camp could gain. If there is no agreement with the host state, the government of the secessionist region will organise the referendum itself and will obviously choose a franchise that favours the vote for independence. If there is an agreement between authorities of the host state and of the secessionist region – as in the Scotland and Montenegro cases – the franchise criteria result from a negotiation in which both involved parties would logically try to favour their preferred outcome.

As a conclusion concerning the franchise, we can stress two elements. The first is that there are many options how to determine the scope of the franchise. The second is that, in most cases, independentists have better chances if they narrow the right to participate in the referendum to official citizens of the host state living in the secessionist region. The less restrictive the criteria are, the more difficult to obtain a majority of votes in favour of independence. Far from being only a theoretical or normative question, the scope of the electoral body is of paramount importance in determining the outcome of a referendum. As the Quebec and Montenegro cases show, both parties are generally aware of it. In this respect, it will be very interesting to study the vote of non-UK citizens in the Scottish referendum on 18 September 2014.

3.The citizenship issue

I disagree with Ruvi Ziegler’s argument about the desirability of congruence between the criteria for the franchise in independence referendums and those determining who will obtain the nationality of the new state in case of independence.

In fact, it seems natural that the group of people who would automatically obtain citizenship on day one of independence remains smaller than the group allowed to vote in the referendum. The difference between these two groups – those who could vote but not automatically obtain the new citizenship – would consist of foreigners living in the secessionist region. As many other contributors have already argued, it is appropriate that they can participate in the referendum if they have lived in the region for a certain period of time and/or want to remain here. But there is no reason that this participation is sufficient for their claim to citizenship status. 

I would insist on the fact that a government of a secessionist region must be cautious about the citizenship issue. If a secessionist government promises to offer citizenship to everybody living in the region’s territory – or adopts very inclusive rules for acquiring it – and also gives the right to foreigners to vote for independence, there is a risk that this could be seen as buying votes. In other words, non-citizens would be encouraged to vote for independence, not because they want it but because of the promise of citizenship. 

In an EU context, giving nationality to a large number of non-EU citizens could also create a problem with other member states. Indeed, every citizen of a member state possesses rights in all other member states, such as the right to settle and to work there. As a consequence, if a new country is too permissive in granting citizenship, it could complicate relationship with other EU members. Ultimately, its recognition or new accession as EU member state could be threatened. 

For all these reasons, I suggest that the citizenship issue and the right to participate in a referendum on independence should be disconnected. Concerning the former issue, it seems essential to distinguish citizens of the host state who live in the secessionist region from other groups. This first group of people should be offered the new nationality on day one of independence. 

For non-nationals, a new naturalisation policy could be implemented later, after the new country becomes independent and the new national government holds the possibility to define new criteria for acquiring citizenship. These criteria can differ from those of the host state. They can be more or less restrictive. But it is important that they are legally disconnected from the independence referendum. Indeed, it is generally agreed that a decision taken by referendum must be as clearly stated as possible. An essential aspect of this clarity is that there is only one question at stake. Adding immigrant integration and citizenship issues to the question of independence in the same referendum certainly won’t favour clarity.



[1] Other recent referendums took place in countries that we can not qualify as “democratic” (South Sudan, East Timor and Eritrea). Taking the democratic criterion into account also excludes referendums organised in former Soviet or Yugoslav republics in the early 1990’s. Even if these referendums could have met democratic standards, they did not take place in well-established democracies. We can recall that there was no referendum in Kosovo before declaration of independence in 2008 nor before the Czechoslovak “velvet divorce” in 1993.

[2] Amaël, Cattaruzza, Territoire et nationalisme au Montenegro, Paris: L’ Harmattan, 2010, p. 132.

[3] Federal government always denied there was a planified policy to accelerate naturalization. But it was noticed that the number of new canadian citizens in Quebec was 87% higher in 1995 than in 1993 to reach 43.850. This figure drop by 40% in 1996. During the month before referendum, 11.500 people were naturalized. Federal civil servant were requested to work on weekends to obtain these results.

[4] Van der Zwet, Arno and Craig McAngus, ‘How Different are Assessments of Independence and Devolution Max? An Analysis of the Role of National Identity and Party Affiliation’, Scottish Affairs 23, no 1, 2014, pp. 1 26 ; Jérôme Couture and Vincent Laborderie, ‘Identité régionale et soutien à l’indépendance dans quatre États subnationaux (Québec, Écosse, Catalogne, Flandre)’, paper presented at 23rd World Congress of Political Science (IPSA, 2014).



In a secession referendum the franchise should depend on what you do, not what you are

By Daniel Weinstock (McGill University)


Imagine that the government of a federated entity with recognized borders launches a referendum to secede from the federation to which it belongs. The argument I want to make in this very short intervention is that citizenship in the federation, and residency in the potentially seceding entity, are individually necessary, and jointly sufficient conditions for participation in the referendum vote. This conclusion places me at odds with some of the contributors to this forum, including Ruvi Ziegler, whose excellent contribution kicked it off.

Many proposals have been made to specify who the demos should be in order to determine the democratic franchise. At one extreme lies what some have called the all affected principle. That principle, as its name indicates, suggests that all those who are affected by a democratic decision should have some say in the decision-making process. 

This principle is clearly over-inclusive in the case of a secession referendum. After all, all citizens of the federation in question will be affected in substantial ways by the decision of a federated entity to secede. But it seems inappropriate to give them all a right to vote. After all, the desire to secede is most often born of the sense on the part of a substantial number of those living in the federated entity that all is not as it should be in their relations with their federal partners. To give those federal partners an effective veto would be simply to import the logic of the problems that have triggered the desire for secession into the decision-making process itself. (1)

At the other end of the spectrum lies the nationalist principle, according to which all those people who trace their origins back to the “founding people” of which the territory of the federated entity is seen as the national homeland should be allowed to make such an existential decision as whether to secede or not.

This principle would on broadly liberal-democratic grounds be unacceptable, by ruling in people who oughtn’t to have a say, and by ruling out people who ought to have one. Let me explain.

The defender of a nationalist principle would consider that someone who traces his origins back to the founding national group should have a say in whether secession should occur or not, even if he has not resided on the territory for years, indeed even if his parents or grandparents had not done so. It would however rule out people who reside on the territory, even though they have only arrived recently, and/or are not members of the founding national group.

It seems to me that excluding “blood nationals”, even blood nationals who are still citizens of the larger federation, is a requirement of liberal democratic ethics. The basic idea is that the right criterion to use in order to determine who gets to vote in secession referendums is a commitment to contribute to the (economic, political, cultural) life of the putative new country, and that simply being related by blood to the founding national group constitutes no evidence of that.

The case of blood nationals who are still citizens of the federation is a difficult one, and my exclusion of them might seem to put me in contradiction with an argument that I have put forward elsewhere, according to which dual nationals should be allowed to vote in elections of the country or countries of residence of which they are not residents at time of election. 

There is however a principled reason to include such people in national and sub-state election, and to exclude them from secession referendums. My argument for the former claim is, in a nutshell, that diversifying epistemic perspectives can help a polity get it right in choosing between political parties’ policy proposals. Citizens who live outside the polity making a decision may be able to “see” aspects of policy questions that are overlooked or under-emphasized by people inside.

But there is no “getting it right” in the case of a secession referendum. Assuming that the group that is considering secession is not doing so frivolously (and what that standard involves is a vexed question indeed!), it is an existential one, about the kind of polity that they want to be. That the group may or may not come to regret a decision does not mean that secession should be treated as a question to which there is a right answer.

Citizens of the federation who are residents, where that term denotes a legal status, rather than a merely factual one, are in virtue of their satisfaction of the criteria for the granting of the status of resident those who best satisfy the criterion that I am putting forward as most appropriate for the granting of the right to vote in secession referendums. The granting of that status is the best index we possess to track the willingness and commitment to contribute to the society in the requisite ways. It is, like any institutional criterion, an imperfect one. But it is, I would argue, as close as we can get.

Two categories of persons constitute interesting limit cases. They are, first, those who have been accepted as immigrants, and who have thus been received as candidates for citizenship , but who have not yet acquired that citizenship, either because they have not satisfied a temporal criterion, or because they have not yet undertaken steps to acquire it in cases in which they have satisfied that criterion. And second, there are those who possess international mobility rights within the potentially seceding territory, and who reside there in virtue of those rights, rather than in virtue of having been accepted as immigrants. The clearest example of this latter category would be citizens of the European Union when they find themselves in one of the EU states of which they are not citizens. I think that there is no “right” answer as to whether these two groups should be enfranchised in secession referendums or not. For what it’s worth, I would opt for enfranchising the first group before I would the second, because the process of applying for immigration is a more demanding sign of one’s commitment to one’s new home than is simply availing oneself of one’s mobility rights. But I don’t think that a grave injustice is committed if the reverse prioritization were to be made (as is the case in the Scottish referendum).

At basis, my view is that for existential questions such as whether to form a new country or remain a part of an already existing one, participation in secession referendums should be determined by what you do, rather than by what you are. Legal residency tracks that moral idea tolerably well.



 (1) Gustaf Arrhenius, "The Boundary Problem in Democratic Theory”.




Catalonia: Will Catalans be permitted to hold a legally binding referendum on independence?

By Montserrat Guibernau, Queen Mary University of London


It feels a bit odd to write on who should be entitled to vote in a referendum on Catalan independence at a time when the Spanish State, invoking the Spanish Constitution, strictly forbids it. The key arguments invoked by the State are: Article 2 of the Constitution that reads: ‘the Constitution is based upon the indissoluble unity of the Spanish nation, common and indivisible patria of all Spaniards’, and Article 8 which states that ‘the Army’s mission is to guarantee the sovereignty and independence of Spain, to defend its territorial integrity and the constitutional set up’.

Currently, the Spanish Constitution strongly emphasizes the ‘unity’ of Spain, this is a point that unavoidably brings about references to the historical background of contemporary Spain including the heritage and memories of the civil war, the dictatorship, the transition to democracy and the coups d’état against the new democratic Spain, taking place as late as the 1980s. It also brings to the fore images of the continuous repression of Basques, Catalans and Galicians as national minorities that managed to obtain some political and cultural recognition in the new democratic Spain; that was a condition for the country to be regarded as a Western liberal democracy by the EU and other international organizations.

Spain has traditionally displayed a ‘centralist view’ of the State in direct confrontation with the aspirations of its national minorities, in particular Catalonia, and attempts to foster a plural image of Spain have not been successful. For instance, former prime minister J.L. Rodriguez Zapatero sought to defend the idea of a ‘plural Spain’ thus emphasizing diversity within a progressive Spain. But probably Spain was not as progressive as he had envisaged, since he obtained limited backing for his views within his own party (the Spanish Socialist Workers Party or PSOE) and little support outside Catalonia – the community that overwhelmingly had supported his view.

Will Catalans be permitted to hold a legally binding referendum on independence?

In Spain, national minorities have a voice – access to Congress and the Senate – however, they have no veto power, and only acquire distinctive relevance whenever one of the main political parties is short of a majority and needs their votes to form a government.  

The 1979 Statute of Autonomy of Catalonia was approved in a legally binding referendum on 25th October 1979. The Preliminary Section of the Statute defines Catalonia as a nationality which ‘in order to accede to self-government, constitutes itself as a Self-Governing Community in accordance with the Constitution and with this Statute’ (Article 1.1) (it should be underlined that the sovereignty of Catalonia is implied in this declaration) and the Generalitat as ‘the institution around which the self-government of Catalonia is politically organized’ (Article 1.2).¹

The powers of the Generalitat ‘emanate from the Constitution, this Statute and the people’ (Article 1.3). These provisions make clear that the Constitution defines the scope and number of devolved powers and confirms the existence of a single sovereign demos in Spanish democracy, constituted by all Spanish citizens, including the Catalans. This interpretation considers the Catalan people to be a ‘sub-group’ of the demos formed by all the citizens of Spain. For instance, it is precisely this interpretation that is invoked by those arguing that all Spanish citizens should be able to vote in an eventual referendum on Catalan self-determination; a view that, if effected, would reinforce further the status of Catalans as a ‘constant minority’ within Spain.² Although this may work from a normative point of view, it will not from a democratic perspective. 

The most important issue is how to unlock the current situation and introduce dialogue between the Spanish and the Catalan governments at a time when Catalan society continues to display a strong bottom up social movement in favour of the so called ‘right to decide’, this is, Catalans demand the right to hold a legally binding referendum on whether Catalonia should become independent or not. The date for this referendum, supported by a range of Catalan civil society associations, is 9th November 2014. At present, it looks unlikely that the Catalans would be permitted to hold a legally binding referendum on that date, however the strength and number of people participating in the forthcoming 11th September 2014 (Catalan Day) demonstration could play a key part in contributing to the unlocking of the current situation. Democracy is by nature a dynamic process, it is not fixed, and is the outcome of a constant dialogue: it is not possible to justify a continuous lack of engagement in dialogue while upholding democratic credentials.

Who should vote and who should be offered citizenship?

Rainer Bauböck highlights a fundamental difference between ‘independent states’ and ‘autonomous regions within states’ as polities of different kinds. In my view, this is very important because ‘independent states’ are able to decide on who should and who should not be allowed/entitled to vote in an eventual binding referendum on self-determination affecting a part of that state. In contrast, ‘autonomous regions within states’ usually lack the power to call for a referendum on self-determination unless this is endorsed by the state. This tension is illustrated by the current imbalance of power between Catalonia and Spain and reflects the continuous relevance of the nation-state as key political actor.

According to Article 6 of the Catalan Statute of Autonomy, all Catalans as well as all Spanish citizens who are legally resident in any of the municipalities of Catalonia are Catalans and in my view, they should be offered ‘Catalan citizenship’ on day one of an independent Catalonia. Participation in a binding referendum on Catalan independence should be limited to Catalan citizens. A new naturalization policy for long-term residents should make citizenship available after 5 years of  residence. Further details should be the outcome of dialogue and debate among political forces in the Catalan Parliament, after the referendum takes place.  If possible, steps should be taken towards a common EU policy on these matters so that similar policies could be implemented at the EU level. 

Who is a Catalan? A former president of Catalonia, Jordi Pujol, defined as ‘Catalan’ a person ‘who lives and works in Catalonia and wishes to be a Catalan’³, a definition that encompasses lieu of residence, workplace, and the ‘will’ to become a member of the Catalan nation as a modern, strongly pro-European prosperous polity. This definition points at some expectations regarding Catalonia as a particular type of nation –diverse, open and inclusive, where civil society has traditionally played a key role. It also highlights the ‘will’ of the individual to belong to that nation and often engage in the construction of a shared collective identity.4



(1) The Catalan Statute of Autonomy (1979) 3rd edition, Generalitat de Catalunya, Barcelona, 1986. Preliminary section, Article 1.

(2) I have first developed the concept of ‘constant minority’ in ‘Catalan Secessionism: Young People’s Expectations and Political Change’, The International Spectator, DOI 10.1080/03932729.2014.952955.

(3) Pujol, J. Tot compromis comporta un risc, Edicions 62, Barcelona 1977 p. 25.

(4) See Pujol, J. La Força Serena i Constructiva de Catalunya, Generalitat de Catalunya, 1990. 



Why Flanders is unlikely to have a referendum on independence anytime soon

By Dirk Jacobs, Université Libre de Bruxelles


Belgium does not have particularly fond memories of organising a referendum on its political future. In March 1950, a referendum was held about the potential return of exiled king Leopold III, who had surrendered to German forces during World War II against the will of the government. In this referendum on the so-called ‘King’s question’, Flanders and rural parts of Wallonia in majority accepted the return, while a majority of the population in the industrial areas of Wallonia rejected it. When the King returned to the country in July 1950, a general strike broke out (mainly in Wallonia), bombs exploded and protestors were shot by the police, bringing the country to the brink of civil war. The government forced the King to abdicate in favour of his son, in order to avoid a potentially violent march on Brussels and to counter a serious attempt to form a separatist Walloon government. These events are part and parcel of the collective memory of the Belgian political elites and have for decades made the idea of organising a national referendum a no-go zone. 

Sixty-five years later, the memories of the ‘King’s question’ have faded, the Flemish nationalist movement is thriving, and Walloon separatism has almost disappeared. The Flemish nationalists explicitly support the Scottish and Catalan causes for independence and applaud the referendums , (1) but do not call for a referendum on independence of Flanders. For outsiders, this might seems strange, as politically Flemish nationalists have never been in a stronger position. In June 2014, the N-VA, the “new Flemish alliance”, a right-wing nationalist party striving for Flemish independence, had a landslide victory in the national elections. In the elections for the Flemish parliament, the nationalists obtained 32% of the vote, and the N-VA is now leading the Flemish government, in coalition with Christian-democrats and right-liberals. In the federal parliament, the N-VA obtained 33 of the 150 seats, making it the largest party in the hemicycle. N-VA is currently negotiating the formation of a federal government with the Flemish Christian-democrats, the Flemish right-liberals, and the Francophone right-liberals. Even if the N-VA has at several occasions expressed their support to the Scottish and Catalan initiatives for a referendum on independence, they are not employing a similar strategy in Belgium. Organising a referendum on the future of Belgium and the independence of Flanders is not on the negotiating table. 

There are several plausible reasons for this reluctant attitude of the Flemish nationalists towards the idea of a referendum. The most important reason is perhaps that all opinion polls consistently show that a majority of Flemish voters are not in favour of independence.(2)  Even among the electorate of the N-VA, there is insufficient support for independence. Indeed, even though the first article of the party statutes of the N-VA clearly states that the goal of the party is Flemish independence,(3)  the party leadership has downplayed the urgency of independence, partly in an attempt to attract right-wing voters not holding nationalist views. In the last election campaign and during current government formation, the nationalists have made it crystal clear their priority is establishing a fiscally conservative government and keeping left-wing parties out of government. This does not mean they have given up on their nationalist agenda, but it is no longer centre-stage. N-VA keeps repeating the mantra that the Belgian federal state is an artificial construction bound to fail, but at the same time are stating that they prefer a gradual ‘evolution’ rather than a ‘revolution’. They believe that Belgium is to gradually evaporate in a process of devolution of powers to sub-federal entities and further transfer of competencies to the European level. They wish to replace the current complex system of Regions and Communities by so-called ‘confederalism’ instead of immediately pushing for independence.(4)  Independence will come, but slowly. 

A major obstacle for any quick separation is that there is no clear-cut scenario for the federal capital of Brussels. The Brussels Capital Region is an officially bilingual territory, where both the Flemish and the Francophone community governments have prerogatives. It is geographically an enclave surrounded by municipalities that are part of the territory of the Flemish Region and where Dutch is the official language (even if the majority of inhabitants of the Flemish municipalities surrounding Brussels are French speakers). Flemish independence would in theory remain relatively easy if it were to be limited to the territory of the Flemish region but then the Flemish would have to give up their shared capital. Given that Brussels is historically a Flemish city, and that the Flemish minority of Brussels is heavily protected as a result of decades of political struggle along linguistic lines, for parts of the Flemish movement it would be considered treason to give up the capital city. Now that the soft power of the Flemish and the status of the Dutch language has considerably increased in the capital compared to some decades ago, entirely giving up on Brussels would be a severe blow to the ‘Flemish cause’. Integrating Brussels in an independent Flanders is, however, unimaginable for Francophone political elites and for large parts of the Brussels population. Brussels, in other words, is the glue that keeps Belgium together and a clear ‘solution’ for Brussels, which would be acceptable for all, is not within reach in case Belgium would disappear as a country.(5)  

Organising a referendum and launching debates on the phrasing of questions would be opening up a Pandora’s box. None of the major linguistic groups can afford to give up on Brussels, but Flemish independence is probably only realistic if they would give up the capital. This ‘sacrifice’ would, however, not remove all obstacles. Even if the referendum were limited to independence of the current Flemish regional territory, large regional discrepancies in voting patterns might arise, in particular in municipalities in the Brussels periphery where sizeable groups of francophones live in Flemish territory. The fact that the linguistic border fixed in 1963 no longer corresponds to sociological reality has been an ongoing source of conflict. Organising a referendum on Flemish independence would fuel demands in the Brussels periphery to be disconnected from Flanders and linked to Brussels. It would put at center stage again the demands of Francophones living in Flanders to be recognised as a ‘national minority’ and be granted all rights foreseen in the Framework Convention for the Protection of National Minorities which Belgium has signed but never ratified. Until now the official discourse is that Belgium only has ‘national majorities’ which all have their own “turf” and that the current language legislation hence makes it obsolete to think in terms of national minorities.(6)  The Flemish can defend this reasoning in the Belgian framework, but will no longer be able to do so in a credible way in an independent Flemish state. 

In sum, a referendum on Flemish independence would just make things more complicated for Flemish nationalists. Tough choices would have to be made about secession scenarios and deals would have to be struck with Francophone compatriots, leading to the necessity to give up on Brussels and taking risks of losing parts of current Flemish territory in the bid for independence. Most importantly, as polling shows, finding a majority in Flanders for independence seems to be close to impossible. That is why federal Prime Minister Elio Di Rupo, a francophone socialist, in December 2012 challenged the Flemish nationalists to call for a referendum on Flemish independence. That is why Flemish nationalists stress they do not want a revolution and see no need for a referendum: it is a battle they cannot win.



(1) See for instance the text « No Rules, Great Scotch! » on the N-VA official homepage.

(2) Abts, K., Swyngedouw, M. Billiet, J. & Meuleman, B. (2014) Vlaanderen kiest. Trends in stemgedrag en opvattingen over politiek, staatshervorming en kerk. Leuven : LannooCampus. 

(3) In the first article of the party manifesto it is literally stated that N-VA opts for an independent Flemish Republic. 

(4) This discursive strategic framing is well illustrated in the English FAQ page of the N-VA website.

(5) Jacobs, D. (2007) "Brussels, do you speak-a my language?", pp.223-238 in Buelens, G., Goossens, J. & Van Reybrouck, D. (eds.) Waar België voor staat. Een toekomstvisie. Antwerpen: Meulenhoff/Manteau.

(6) A justification in English on the non-ratification of the treaty by Belgium, under pressure by the Flemish, is provided on the website of 'de rand'. This is a quasi-non governmental organisation sponsored by the Flemish government which has as a mission to defend the Flemish character of the Brussels periphery. 


Voting in the referendum on Scottish independence: some observations from the front line

By Jo Shaw (University of Edinburgh, EUDO-CITIENSHIP co-director)


Ruvi Ziegler's elegantly worded argument for the congruence of the referendum franchise and of the initial determination of the citizenry in a new independent Scotland, aka iScotland, along with his thorough rejoinder to the responses of a variety of commentators coming from different territorial and disciplinary backgrounds, may not offer the final word on this important question, but they certainly go a long way towards airing most of the key points that need to be made. Moreover, I have had the privilege of reading Ziegler’s rejoinder before writing this short text, and this has confirmed my view that my own contribution to the debate should take another angle. For I should confess that I have thought a lot more about this issue since Ziegler initiated this debate, having originally broadly subscribed to the 'hey ho, it's just a messy model like the rest of citizenship issues in the UK' school of thought, when it came to the question of how the Scottish referendum franchise was conceived in the Edinburgh Agreement and instantiated in legislation adopted by the Holyrood Parliament in the summer of 2013. But Ziegler has not only provoked some further thinking on the franchise, but also broader reflections on the matter of the Scottish referendum vote (and the campaign leading to the vote).

Of course, the conception of the franchise adopted for the referendum on Scottish independence, largely based on the regional franchise and including, as a peculiar emanation of UK constitutional law not only qualifying Commonwealth and Irish citizens but also EU citizens, but excluding non-residents whatever their citizenship, has been contested. The strongest contestation has come from persons born in Scotland but now living elsewhere in the UK, rather than those - such as Ziegler's example of the Scottish born graduate working in Amsterdam as an EU citizen - who live outside the UK altogether. This is hardly surprising, since they represent numerically by far the largest group who could lay claim to be disenfranchised. Many of them will have noticed that at the same time they have been 'offered' future citizenship in the various Scottish government documents that have come out which sketch the outlines of the initial determination of the citizenry. Indeed, it would seem that they become automatically Scottish citizens, although one would assume that those who are resident in rUK will retain UK citizenship, even if rUK were to decide to redefine its citizenry to exclude some groups of new Scottish citizens resident in Scotland (e.g. those born in Scotland who have never resided elsewhere in the UK).

It is worth noting that the settlement of the franchise was - as Lord Kerr of Kinlochard (a distinguished Scottish former diplomat, occasional commentator on EU affairs, and member of the disenfranchised UK-based Scottish-born (and educated) diaspora) put it - a 'casual concession' by David Cameron to Alex Salmond, and it was a concession with considerable repercussions. The casualness of that concession, which received no public debate, is regarded by many resident in Scotland as typical of the casualness with which 'Westminster' deals with most Scottish matters, and thus has provided further fuel for the independence movement.

The disenfranchisement of the group of 'non-resident Scots' - against the promise that they would be ab initio citizens of iScotland - was, however, probably the only workable outcome, since drawing up a register of electors based on either birth in Scotland or previous residence there (for how long?) would have been an extraordinarily expensive and possibly rather inaccurate exercise. If you accept the power of its premise of congruence, what Ziegler's argument does achieve, however, is that it shines a light on the perhaps over-inclusive ab initio citizenship condition of birth in the territory as a UK citizen (with no additional connections to the territory being necessary). But as Dora Kostakopoulou has pointed out, the contestation of that is to come later, and will not occur unless there is a Yes vote on 18 September 2014. My experience with trying to generate interest in my earlier research has shown that, while people are a bit puzzled about who would be Scottish citizens in the event of independence, it has definitely not been an animating factor in the debate. It seems likely that the relatively over-inclusive scope of the ab initio definition, not to mention the proposal for a vaguely specified group of 'citizens by connection', who can apply to be naturalised even if they are not resident, tells us something about the main audience for documents such as Scottish government consultation papers on these matters, and that there would be a quietly supportive external diaspora who may prove to be rather influential in an iScotland.

This leads me to the main comment I would like to make in this short contribution, and that concerns my observations from the front line. I wrote this note on the last day (2 September 2014) when qualifying referendum electors could apply to join the electoral register. We do not yet know the final tallies, but anecdotal evidence suggests that a lot more people will be registered to vote in Scotland for the referendum than were for the latest elections using approximately the same franchise, namely the Scottish Parliament elections of 2011. Of course, part of that surplus will stem from the additional 16 and 17 year olds who have been enfranchised in the referendum, but it is also clear that a large number of people who have allowed their voter registration to lapse (or who have never been registered) have now sought to make sure they will be able to vote. Some of the new registrations will also include EU citizens who were the subject of the research reported by Derek McGhee and Emilia Pietka-Nykaza, some of whom undoubtedly feel burdened by this 'privilege' of participating in the referendum, although others have been animated by this opportunity to contribute to a potentially life changing decision. One must presume that this surge of late registrations is also likely to signal a very high turnout, perhaps even between 80-85%. Speaking personally, I should state that I have never in my lifetime experienced anything akin to the political engagement which has been engendered by the opportunity that the independence referendum has given people in Scotland to discuss their future. Confirmed postal voters (an increasingly large percentage of middle class middle aged people who are likely to vote but are often not available on polling day) report on social media having rescinded their postal votes for this one time, just so they can experience the thrill of going to the voting booth. 

Much of the public politics of the referendum debate has been rather uninspiring, especially the two set piece televised debates between male, macho leaders (First Minister Alex Salmond and Alastair Darling, representing the 'Better Together' campaign). But I concur with others who say that this hardly represents the true tenor of the debate, which has been largely carried on outside the normal realms of political debate in the modern world, in families, workplaces, streets, town meetings, and festivals. And of course on social media where debate has been, in turns, both ‘shouty’ and quite inspiring.

The engagement with the Yes campaign of a large number of independent grassroots groups often aligned with the political left has been responsible for much of this change from the normal fodder of political party dominated local, regional, national, and European parliamentary elections. Much of the 'new politics' has been animated by people involved in the cultural sectors, and has been tinged with a high degree of whimsy and humour. Perhaps the best example is offered by the series of conversations animated by the Scottish playwright David Greig, all through the Edinburgh Festival Fringe in August 2014, being an ironic tribute to David Bowie's intervention in the debate to suggest that Scots should vote no in the referendum: All Back to Bowie's. As the website states:

In response to David Bowie's famous declaration at the Brit Awards, a group of Scottish artists are setting up camp in Bowie's (metaphorical) living room for an irreverent lunchtime show exploring the 2014 Scottish independence referendum, and what it might mean for the country to stay with - or leave - the UK.

If this is the 'regional electorate' which Rainer Bauböck in his contribution argues is best placed to consider whether or not to upgrade a regional citizenship into the citizenship of an independent state, then anecdotally I would concur that confining the electorate by reference to residence has been successful. It has given a powerful sense of common destiny to Scottish residents, even if, of course, we are making a decision which affects many more people than that group alone for all the reasons that the contributors to this debate have made clear. To what extent that sense of destiny would carry over into iScotland, or back into a continuation of the three hundred and seven year Union of the Parliaments which must, surely, even in the event of a no vote, be ripe for radical constitutional reform, remains to be seen.



Independence Referendums and Citizenship ab initio – A Rejoinder

By Ruvi Ziegler (University of Reading)


I am most grateful to all the contributors for taking the time to engage with the topic and offer eye-opening and thought-provoking perspectives on contexts ranging from Catalonia/Spain, Flanders/Belgium, and Scotland/UK to Quebec/Canada and Puerto-Rico/USA. The nature of this rejoinder makes it challenging to address the many incisive points raised in the contributions. My aim is to address principal points of contention – and agreement – that emerged from the debate, and to clarify several issues related to putative citizenship claims from an international law perspective.

In my kick-off contribution, I argued that it is normatively desirable for the enfranchised population in independence referendums to resemble the citizenry on ‘day one’ of a successor state that may come into being following an affirmative vote. I also submitted that, citizens ab initio of a successor state are significant stakeholders in an independence referendum that may bring that state into being. The distinction between the initial determination of the citizenry and the rules of acquisition (naturalisation) and loss of citizenship after independence should be emphasised: my claims concerned only the former. Notably, the above claims are not derived from an expressed preference for one of the models for attribution of successor state citizenship ab initio (see e.g. Jo Shaw’s research), as long as the selected model is compliant with international law standards as per the ILC Draft Articles.  

Independence referendums: background conditions 

My contribution steered clear of determining the legitimacy of particular external self-determination claims (see this encyclopaedic entry for select sources on self-determination in international law). However, as became evident during the debate, three background conditions need to be satisfied before such claims are brought forward by way of an independence referendum, and an additional condition may be critical for its implementation.

First, a sufficiently determined political movement possessing the will to pose the independence question. Dirk Jacobs’ contribution demonstrated the implications of the absence of such a political will in Flanders. Jaime Lluch explained that previous ‘plebiscites’ in Puerto Rico were not strictly speaking ‘independence referendums’ as three of the four political status options involved remaining within the U.S. constitutional framework. Second, a defined territory. The considered case-studies (Catalonia, Quebec, Scotland, Puerto Rico and Flanders) indicate that an interim stage of self-governance (be it devolution, federalism, and/or regional autonomy) is likely to precede an independence referendum. This legal and political reality may affect the question of eligibility, both for participation in the referendum and for citizenship ab initio. Third, and perhaps most critically for this debate, an identified ‘people’ on behalf of whom the claim for external self-determination is made. Finally, even though the acquiescence of the state from which secession is sought is not necessarily normatively required (Rainer Bauböck’s observation), its absence is a recipe for uncertainty. Note Jaume Lopez’s and Montserrat Guibernau’s portrayals of current tensions between the Spanish State and the Catalan Government, leading the Catalan Parliament to refer to the 9 November 2014 vote as a 'political consultation' and asking a modular two-stage question rather than a straightforward independence question.    

I readily concede that my claims are normative; even if, as Guy Laforest and Eric Montigny observed, ‘on matters of referendums, consistent normative logic does not always work’, I believe that we should not give up on trying. I also acknowledge Dora Kostakopoulou’s insightful remarks regarding the dynamism of political processes: it may very well be that the White Paper’s framework for citizenship ab initio will be altered following an affirmative vote and that, politically, the referendum would have probably gone ahead even if the SNP had refrained from setting out a citizenship agenda. However, despite my critique of the incongruence between the blueprint for citizenship ab initio and the referendum franchise, I think we ought to appreciate the fact that a plan has been presented; at least in this aspect, a standard has been set for future independence referendums.

The franchise in independence referendums and putative citizenship: four categories

In the main, the contributions addressed questions arising in the context of potential separation of a territory from an existing state rather than dissolution thereof. Hence, from an international law perspective, citizenship of the predecessor state remains unaffected. In contrast, dissolution of a predecessor state affects all its citizens. Against this background, the eligibility of persons belonging to the following categories of persons to vote in an independence referendum was considered: First, citizens of the predecessor state habitually resident in the putative successor state (including persons temporarily absent). Second, non-citizens habitually resident in the putative successor state. Third, citizens of the predecessor state formerly resident in the putative successor state or having other ‘appropriate legal connection’ with the successor state. Fourth, other citizens of the predecessor state. 

A consensus has emerged that persons in the first category form the ‘core’ electorate in an independence referendum. Indeed, such persons will be considered citizens ab initio of a successor state (unless they choose to decline the offer): according to Article 24(a) of the ILC Draft Articles, concerning separation of a territory, a successor state shall attribute its nationality to persons concerned [defined as citizens of the predecessor state] habitually resident in its territory, subject to granting them the right of option (Article 26 thereof). As Jacobs noted, the reluctance of Flemish leaders to pose the independence question stems in part from the (electoral) implications of enfranchising Francophones currently residing in Brussels.

A similar consensus has emerged that, although persons in the fourth category will be affected in substantial ways by secession, it would be inappropriate to enfranchise them in an independence referendum, as this would effectively grant the majority of the predecessor state veto power over secession (a point which Guibernau highlighted in relation to Catalonia). I contend that it would arguably be (at least equally) implausible to consider such persons as citizens ab initio of a successor state.

Contributors diverged as to whether persons belonging to the second and third categories should be enfranchised and/or be considered/offered citizenship ab initio, offering normative, pragmatic, strategic, and (national) law arguments to support their position.  

Debate themes

The following (non-exhaustive) themes have emerged from the debate: (1) external self-determination; (2) the ‘core electorate’ option; (3) the challenge of the regional franchise (4); Scotland: a tale of two unfitting franchises; (5) over-inclusiveness and abstention; (6) stake-holding and putative citizenship; (7) strategic ex/inclusion; (8) the significance of an agreement with the central government. 

Whose external self-determination is it, anyway?

International documents refer to the (external) self-determination of peoples. Article 1(2) of the UN Charter considers the development of ‘friendly relations among nations based on respect for the principle of equal rights and self-determination of peoples’, and the identically worded Article 1 of the ICCPR and ICESCR refers to peoples’ right to self-determination, by virtue of which ‘they freely determine their political status…’ In line with Vesco Pakalev’s contribution, I contend that a ‘people’ (or a ‘definable group’, to borrow the term used by the Canadian Supreme Court judgment, in Re Secession of Quebec, [138]) making a self-determination claim needs to be self-defined, so that it can plausibly answer this question: in the name of whom is external self-determination sought? The answer to this question should determine the category of persons entitled to participate in the referendum, since ‘the application of the right to self-determination requires a free and genuine expression of the will of the people concerned’ (International Court of Justice, Advisory Opinion concerning Western Sahara, 16 October 1975, [55]). 

Hence, when external self-determination is sought via an independence referendum, inclusion and exclusion therein are not just symbolically significant: a flawed designation of the electorate may cast a normative shadow on the legitimacy of the referendum which, in turn, may affect the extent to which the outcome is recognised. Notwithstanding Jennings’ paradox of self-determination with which Jure Vidmar opened his contribution, self-definition is an inevitable part of an external self-determination process.  

I further assert that, consistently with international law, a putative state should have an initial body-polity at the moment of coming into being. By setting a blueprint for citizenship ab initio, the Scottish government attempted to define the people of Scotland on behalf of whom external self-determination is sought. It could have made other choices reflecting different encapsulations of ‘the people’; however, unlike Kostakopoulou, I believe that, acting in good faith, the Scottish government had to make a choice and duly present it for public scrutiny. Lopez’s contribution noted that the Catalan government has not yet presented a similar blueprint; nevertheless, absent contrary indications, it may perhaps be presumed that the definition of a ‘Catalan’ in Article 7 of the 2006 Statute of Autonomy of Catalonia (which, in its preamble, refers to self-government of ‘the Catalan people’) reflects the category of persons whose external self-determination the independence movement seeks: Spanish citizens legally resident in Catalonia (subsection 1) and Spanish citizens resident abroad whose last legal place of residence was Catalonia (subsection 2).

Herein lies the rub of my congruence claim: that, normatively, the designation of a ‘people’ whose external self-determination is sought should be reflected both in the franchise of a referendum on independence and in the initial citizen-body of a putative state.

The ‘core electorate’ option

Daniel Weinstock posited that (in a federation) citizenship of the predecessor state and habitual residence in the putative successor are individually necessary, and jointly sufficient for enfranchisement in an independence referendum; namely, that the franchise should not extend to persons in the second or third categories outlined above. Following my earlier discussion, Weinstock’s argument suggests that ‘the people of X’ for the purpose of external self-determination is coterminous with its citizen-residents, and that the franchise should be based on both criteria. I submit that the same logic, if adopted, suggests that this should be the initial body-polity on day one of a successor state.

Weinstock observed that this proposition puts him at odds with mine. Now, this may be the case insofar as he views the exclusion of ‘“blood nationals’, even blood nationals who are still citizens of the larger federation’ to be ‘a requirement of liberal democratic ethics”, whereas my position is contingent on whether non-residents will be considered citizens ab initio. Hence, Weinstock opposes enfranchisement of non-residents in an independence referendum; in contrast, I believe that it depends on whether they are considered part of ‘the people of X’ for the purposes of external self-determination, and that there may be sound reasons to include non-residents who have ‘an appropriate legal connection’ with the seceding territory. Indeed this is why the ILC Draft Articles encourage successor states to give such persons the option to opt for their citizenship ab initio. 

Nevertheless, it is noteworthy that, at least on my reading, Weinstock’s proposition is compatible with my congruence claim. Moreover, notwithstanding the above, his proposition is compatible with the minimum requirements of the ILC Draft Articles. Article 8(1) states that ‘a successor state does not have the obligation to attribute its nationality to persons concerned [read: citizens of the predecessor state] who have their habitual residence in another state and also have the nationality of that or any other state’.   

The challenge of the regional franchise    

Bauböck (with whom several other contributors concurred) argued that, when a region which enjoys a degree of self-governance is considering external self-determination, the independence referendum franchise should be based on the regional franchise, which is (according to Lluch) the only legitimate franchise in a constitutive referendum. Notably, while the scope of Bauböck’s contention is limited to non-remedial secession, classification of some cases may be challenging (note Lopez’s different characterisation of Catalonia). 

Bauböck’s claim is two-fold: first, the existing regional franchise enjoys constitutional legitimacy. Second, ‘the [independence] vote belongs to the regional citizens who should decide on whether they want to change their region’s status through secession’. The first argument is important for obtaining the acquiescence of the central government; however, as independence operates outside the state’s existing constitutional order, it carries limited weight. 

The second leg of the argument is compelling, yet it too has limited force: the regional franchise ought to be the prima facie basis for determining the referendum franchise insofar as it defines ‘the people’ in the name of whom external self-determination is sought. This will often be the case when ‘people’ of that region are defined as a subset of the body-polity of the state, as seems to be the case in Catalonia, where Article 7 of the Catalan Statute defines and enfranchises all ‘Catalans’. Here, it seems plausible to base both the independence referendum franchise (rather than the over-inclusive franchise that is currently proposed) as well as citizenship ab initio on the regional franchise, a position which Lopez and Guibernau endorsed. Bauböck, however, argued that the regional franchise should not determine citizenship ab initio of the successor state, since ‘regions and independent states are polities of different kinds and their membership rules differ accordingly’. But why should citizenship on ‘day one’ (as opposed to naturalisation and voluntary renunciation) reflect a different polity than that which exercised external self-determination? I remain unconvinced.

The more challenging scenario is when the regional franchise reflects considerations that suit local (rather than national) elections. This is, arguably, the case of Scotland, as the franchise for UK regional assemblies, including the Scottish Parliament, is identical to the franchise for local elections. Being both over and under-inclusive, the Scottish regional franchise is a misfit, bearing insufficient correlation to ‘the people of Scotland’. In such circumstances, both options are unfavourable: one can either adopt the regional franchise, thereby weakening the external self-determination claim (see Vesco Pakalev’s cogent analysis); or, alternatively, adopt a different franchise which reflects ‘the people of X’ but which could suffer from an internal legitimacy deficit. In the UK, a layer of complication is added by the fact that the general election franchise would not be a suitable option either, for reasons which I discuss below.    

Scotland: a tale of two unfitting franchises 

Although the claims in the kick-off were intended to be of general applicability, the emphasis I placed on the Scottish franchise and ab initio citizenship prompted defences thereof by Vidmar and Bernard Ryan. Vidmar argued in reference to Scotland that ‘a variant of the territorial approach appears to be a fair choice’; it baffles me how a variant that is based on select list of nationalities among Scotland’s residents is considered a fair choice (see also Lopez’s critique of the proposed franchise for the Catalan vote on 9 November 2014 which favours EU residents over other migrants in terms of their residence eligibility requirement).

Ryan pointed to a supposed tension between my two propositions with respect to the Scottish referendum franchise; this may be accurate had I proposed the UK general election franchise as the basis for the independence referendum franchise. However, the kick-off does no such thing. In fact, it points out two follies of the latter franchise: the (arbitrary) 15 year out-of-country residence rule that disenfranchised many UK external citizens, and the enfranchisement of qualifying Commonwealth and Irish citizens. Rather, my proposition is that because certain non-resident UK citizens will be considered citizens ab initio of an independent Scotland, their exclusion from participation in the referendum is incongruent.  

Ryan argued that my general approach “runs up against the enduringly pragmatic character of the UK’s constitutional arrangement”. Yet, enfranchisement of qualifying Commonwealth and Irish citizens is a peculiar feature of the UK and its colonies’ shared imperial past. Even if the UK chooses to retain this historical anomaly in future elections (notably, the 2008 Goldsmith report entitled ‘Citizenship: Our Common Bond’ proposed “limiting in principle the right to vote in Westminster elections to UK citizens”; page 75, [17]), I struggle to see the logic and utility behind ‘exporting’ this arrangement to the franchise and citizenship arrangements of a newly established state.    

Over-inclusiveness and abstention 

Ben Saunders argued that “we may wish to give the vote to some who will not be given citizenship, because we do not wish to accord them other rights that would be afforded by citizenship.” I submit that, an election cycle and an external self-determination referendum determining the international legal status of a territory are qualitatively different electoral processes; while Saunders’ rationale may be applicable in the former, small ‘p’ politics processes, it is not appropriate in the latter, because it understates the legal significance of the creation of a new independent state, including with regard to the right to enter and remain.  

Saunders suggested that “[t]hose who are long-term residents in Scotland should have a say over its future.” Now, perhaps a claim for self-determination of X could theoretically be made by ‘the people living and working in X’ (as long as no arbitrary distinctions are made between residents). Why, then, should only some of these people be considered the initial members of the independent political community which their collective act of participation in an independence referendum has brought into being?  According to Saunders, non-citizens ‘will be part of that future [of the successor state] even if they are not (at least immediately) offered citizenship’. From an international law perspective, this proposition is ill-advised: if the territory remains part of the larger, predecessor state, then their future in that territory (and in the country as a whole) depends on their immigration status in the country. In the event of an affirmative vote, the question whether they will be part of the future of that successor state will be determined by their immigration status in that state which, in turn, will be determined by the body-polity of that state. 

Vidmar cited the ECtHR Kurić judgment which should presumably calm the anxieties of Polish migrants in Scotland. However, this case involved the residence rights of citizens of the predecessor state (Socialist Federal Republic of Yugoslavia); indeed, Article 14 of the ILC Draft Articles enunciates that “the status of persons concerned as habitual residents shall not be affected by the succession of states.” From current international law perspective, the security of residence of non-citizens of the predecessor state, even long-term residents, may be a different matter altogether (although, like Vidmar, I would like to see the logic of the judgment extended in these situations in future).

It is quite tempting to share Saunders’ intuition that ‘it is better to err on the side of generosity when allocating voting rights’, since over-inclusiveness guarantees that all putative citizens ab initio are enfranchised, and eligible voters may abstain. However, independence referendums may have a pass threshold, in which case failing to vote (or abstaining, where voting in a referendum, like in elections, is considered mandatory) is effectively a vote against independence; in such referendums, there is no neutral position. More fundamentally, the decision to enfranchise (certain) non-citizens or (certain) non-residents is a symbolic public statement that they are considered part of the people seeking self-determination. It carries normative weight and affects behaviour. It is hardly surprising that the Polish migrants in Scotland (surveyed by Derek McGhee and Peitka-Nykaza) are unsure whether to participate, as they have been sent conflicting messages: on the one hand, they are invited to take part in a historic referendum potentially terminating a 307-year-old union. On the other hand, they are not considered part of the initial citizen-body of a putative Scotland, and (as noted above) their security of residence is indeterminate. No wonder many of them refer to their inclusion as a ‘privilege’ or a ‘burden’ (rather than as a right or an entitlement) and some of them see themselves as guests who ought to vote in a manner that would satisfy the wishes of their kind hosts lest hospitality turns to hostility; the problem being, of course, that these wishes are hardly uniform. One could anticipate similar dilemmas arising for EU and other migrants in Catalonia, should the franchise follow the proposed model.

Stake-holding and putative citizenship

In his ‘mid-debate’ summary, Paskalev observed that most contributors have not contested the claim that putative citizens of a successor state are significant stake-holders. Kyritsis offered a limited normative challenge, arguing that even if it were unproblematic (perhaps even desirable) to enfranchise citizens ab initio, their exclusion does not dent the legitimacy of an independence referendum. He argued that, the stake-holding of ab initio citizens is not qualitatively different from that of citizens of neighbouring states who may be vicariously affected by electoral results (e.g. citizens in Latin America affected by U.S. elections). Bauböck made a similar claim with respect to citizens of rUK (persons in the fourth category outlined above). Now, it is possible that the discernible, daily effect of a change in U.S. administration on a citizen of Venezuela may be greater than the effect of Scottish independence on our (now famous) Amsterdam-based former Scottish graduate. However, from an international law perspective, the individual status and citizenship-contingent rights of the former Scottish graduate are meaningfully affected. Kyritsis was right to note that, according to Article 8(2) of the ILC Draft Articles, a successor state shall not confer its citizenship on our graduate against her will, seeing that she is a UK citizen habitually resident elsewhere. Yet the independence of Scotland may force her to choose between UK and Scottish citizenships, and that choice may in turn affect her life and work possibilities in the territory to which she has an ‘appropriate legal connection’ (elsewhere, Nick Barber discussed some of the challenges that non-residents may face in the event of independence). 

Kyritsis further argued that ‘there is no general duty to give people easy choices’; this is undisputed but beside the point: it is not the fact that our graduate will be faced with the choice that is at stake here but the fact that she was excluded from the process that has brought about this scenario. Kyritsis’ analogy to premium membership fails to encapsulate the significance of exclusion. A more suitable analogy would be that of children being faced with (hard and/or exciting) choices that were pre-determined by their parents. The ability to choose ex-post does not compensate for the symbolic harm caused by exclusion ex-ante. Do those seeking external self-determination consider the Amsterdam-based Scottish graduate merely a potential beneficiary of a decision made in her (electoral) absence, or part of the ‘people’ on behalf of whom self-determination is sought? I submit that, like the Polish migrant, she has been receiving a mixed message from the independence movement.        

Strategic ex/inclusion

Vincent Laborderie’s contribution endorsed adopting an effectively over-inclusive franchise in order to discourage political calculations: he proposed that ‘the group of people who would automatically obtain citizenship on day one of independence remains [sic] smaller than the group allowed to vote in the referendum’. He asserted that the prospect of citizenship ab initio may entice a non-citizen resident (e.g. the Polish migrant in the Scottish independence referendum) to favour independence, and so the independence movement may be seen to be buying votes; hence, the migrant should be enfranchised, but should not constitute part of the initial citizen-body of a successor state. I contend that, the normative dilemma whether non-citizen residents should participate in a referendum should not be resolved by predictions regarding their voting intentions; surely, no-one would argue that the citizenship ab initio of resident citizens should be dependent on their voting intentions, and it is not a principled basis for denying membership to others. 

Similarly, Vidmar’s caution against “including the Diaspora too generously” lest it “distort the result” is unconvincing: if, as he posited “an independence referendum is an eminently territorial question”, then this should be the reason to exclude non-residents (and enfranchise residents), rather than the way they are predicted to vote. Interestingly, it has been suggested that the extension of the franchise in the Scottish referendum to 16 and 17 year olds may had been driven by a nationalist preference (see this response); however, not only can predictions backfire but, critically, the important question is whether it is right to extend the franchise.

The significance of an agreement with the central government 

The question whether the referendum - and the referendum franchise - is agreeable to the self-determination movement and to the central government (as in the case of Scotland via the Edinburgh agreement) is no doubt politically significant, and may determine the likelihood that an affirmative vote in an independence referendum will result in independence, de facto and de jure. Laborderie observed that, absent an agreement, the independence movement may have an incentive to try to tweak the franchise so that it suits its purposes, implying that the need to reach an agreement with the central government increases the chance of a genuinely representative referendum. This seems intuitively plausible. Indeed, all else equal, an independence referendum that takes place pursuant to an agreement is preferable. Laforest and Montigny argued that “gaining consensus between central and sub-state government is more important than maintaining consistency between pre- and post- independence.” From a realpolitik perspective, this is very sensible. However, while the Edinburgh agreement facilitated the smooth running of the referendum, a different franchise could have been pursued and agreed between the parties. Saying that consensus trumps consistency is question-begging. More fundamentally, I maintain that a normatively flawed franchise is not saved by the fact that it is endorsed by the central government. Ideally, this examination ought to be conducted before an agreement is concluded, should be transparent, and open to public scrutiny.  

In closing

A historical glimpse may be quite insightful on this occasion. The UN ‘Future Government of Palestine’ resolution (General Assembly resolution 181/II of 29 November 1947), commonly known as ‘the Partition Plan’, set out a clear blueprint for participation in the elections of the constitutive assemblies of the prospective Jewish and Arab states, and for citizenship thereof (section B, entitled ‘steps preparatory to independence’). Like the ILC Draft Articles, the base-layer for citizenship was (Mandate) Palestine citizens resident in the respective states. In turn, the resolution enunciated [10] universal suffrage for adult citizens of both genders. The franchise was also intended to include non-citizen residents, provided that they have “signed a notice of intention to become citizens of such State.” As is all too well-known, the arrangements in the partition plan (including partition itself) have not materialised, but the blueprint for participation and membership is noteworthy.

I am much obliged to the EUDO Citizenship forum for hosting this online discussion, to Rainer Bauböck and Jo Shaw for scrupulous editing, and to Jean-Thomas Arrighi and Vesco Paskalev for providing administrative support. The debate will certainly continue, not least in view of the forthcoming vote in Catalonia. Wherever it takes place, let it be informed, respectful, and good-mannered as this one has been. 




Further reflections on the scope of the franchise and citizenship in Scotland and Catalonia can be found at the following links:


Citizenship in an independent Scotland: Legal Status and Political implications, CITSEE Working Papers 2013/34, by Jo Shaw 

After the vote, the citizenship question, by Nick Barber (blog entry)

Should Scotland vote for what is best for Scotland, by Kieran Oberman (blog entry)

Catalan citizenship: the pending debate, by Ricard Zapata-Barrero (blog entry) 




Should citizenship be for sale?

On 12 November 2013 the Maltese Parliament decided to offer Maltese and European citizenship at the price of € 650,000, but implementation of the law has been postponed due to strong domestic and international critiques. On 23 December, the Maltese government announced significant amendments, including a higher total amount of € 1,150,000, part of which has to be invested in real estate and government bonds. Several other European states have adopted ‘golden passport’ programmes. Should citizenship be for sale? In November 2013 EUDO CITIZENSHIP invited Ayelet Shachar of the University of Toronto Law School to open a debate on these controversial policies. Twelve authors have contributed short commentaries, most of which refer to the initial law adopted by the Maltese Parliament. An executive summary by Rainer Bauböck provides an overview over the main questions raised in our forum. For further information on investor citizenship programmes see Jelena Dzankic’s EUDO CITIZENSHIP working paper on the topic and consult the news section of our observatory.


Dangerous Liaisons: Money and Citizenship

Kick-off contribution by Ayelet Shachar

Vogue predictions that citizenship is diminishing in relevance or perhaps even vanishing outright, popular among jetsetters who already possess full membership status in affluent democracies, have failed to reach many applicants still knocking on the doors of well-off polities. One can excuse the world’s destitute, those who are willing to risk their lives in search of the promised lands of migration in Europe or America, for not yet having heard the prophecies about citizenship’s decline. But the same is not true for the well-heeled who are increasingly active in the market for citizenship: the ultra-rich from the rest of the world. They are willing to dish out hundreds of thousands of dollars to gain a freshly-minted passport in their new “home country.” That this demand exists is not fully surprising given that this is a world of regulated mobility and unequal opportunity, and a world where not all passports are treated equally at border crossings. Rapid processes of market expansionism have now reached what for many is the most sacrosanct non-market good: membership in a political community. More puzzling is the willingness of governments—our public trustees and legal guardians of citizenship—to engage in processes that come very close to, and in some cases cannot be described as anything but, the sale and barter of membership goods in exchange for a hefty bank wire transfer or large stack of cash.

Everybody knows that immigration is among the most contentious policy issues of our times, and recent years have witnessed a “restrictive turn” (Joppke 2007; Orgad 2010) with respect to ordinary immigration and naturalisation applicants, such as those who enter on the basis of a family reunification claim or for humanitarian reasons. The situation is different, however, for the world’s moneyed elite, who can sidestep many of the standard requirements for settlement by “buying” their way into the political community. The public act of naturalisation—of turning a non-member into a citizen—has always borne an air of legal magic, with the result that it is the “most densely regulated and most politicized aspect of citizenship laws” (Bauböck and Goodman, 1). At stake is the regulation of the most important and sensitive decision that any political community faces: how to define who belongs, or ought to belong, within its circle of members. Not everyone knows, however, that governments are now proactively facilitating faster and smoother access to citizenship for those who can pay. Revealing insights about the current state of citizenship can be gained, I will argue in this short essay, by examining who is given this red-carpet treatment, and on what basis.

Consider the following examples. Affluent foreign investors were offered citizenship in Cyprus as “compensation” for their Cypriot bank account deposit losses (the value of which was set at €3M in the aftermath of the EU bailout). In 2012, Portugal introduced a “golden residence permit” to attract real estate and other investments by well-to-do individuals seeking a foothold in the EU. Spain recently adopted a similar plan. On 12 November 2013, Malta approved amendments to its Citizenship Act that put in place a new individual investor legal category that will allow high-net-worth applicants to gain a “golden passport” in return for €650,000. Under these cash-for-passport programmes, many of the requirements that ordinarily apply to those seeking naturalisation, such as language competency, extended residency periods or renunciation of another citizenship, are waived as part of an active competition, if not an outright bidding war, to attract the ultra-rich. Portugal, for example, offers a fast track for qualified applicants that entitles them to a 5-year permanent residence permit, visa-free travel in Schengen countries, the right to bring in their immediate family members, and ultimately the right to acquire Portuguese citizenship and with it the benefits of EU citizenship. This package comes with a hefty price tag: a capital transfer investment of €1M, a real estate property purchase at a value of €500,000, or the creation of local jobs. The investment needs to remain active in Portugal for the programme’s duration. Alas, the individual who gains the golden permit bears no similar obligation. Simply spending 7 days in Portugal during the first year and fourteen days in the subsequent years is enough to fulfil the programme’s requirements. So much for the conclusion of the International Court of Justice, in the 1955 Nottebohm decision, that “real and effective ties” between the individual and the state are expected to undergird the grant of citizenship.

In Malta, recipients of the golden passport will be vetted in accordance with a discretionary ministerial act that puts in place little transparency and accountability. Government officials have made clear that applicants can expect an expedited treatment, meaning that they will not have to “stand in the queue” like everyone else. In addition, the names of golden passport recipients will remain confidential, making it close to impossible ever to know to whom the polity has sold a precious part of its soul. This last provision has raised the ire of the opposition. Their concern is that concealing the identity of those who gain membership by literally purchasing citizenship makes it so that “Maltese [a]re now being denied the right to know who is Maltese” (Malta Today 2013). The secrecy provision was eventually withdrawn in the eleventh hour, but the basic structure of the programme remains intact: privileged and fast-track naturalization, allowing any Tom, Dick and Harry … [to] buy a Maltese passport without ever setting foot on Maltese soil.” A recent survey shows that the vast majority of the population opposes the sale of citizenship in principle, and rejects this scheme in particular, detached as it is from any residence or other requirements that would establish ties with the passport-granting country and society.

Beyond Europe, those seeking a new passport can look to St. Kitts and Nevis, where economic citizenship can be purchased for as low as $250,000 (for a lump sum) or $400,000 (if monies are directed to a real estate project), and issued within months. They might also consider Antigua and Barbuda, which is the latest in a growing list of countries to roll out a citizenship-by-investment programme, or the Commonwealth of Dominica. Whereas ordinarily the law requires significant residence periods for those seeking naturalisation in these island nations (fourteen years in St Kitts and Nevis, seven years in the Commonwealth of Dominica and in Antigua and Barbuda, respectively), the residency requirement is reduced to merely seven days—a short vacation under the tropical sun—or even waived altogether for those who purchase their fast-tracked passport.

The citizenship-by-investment programmes that I have just described fall into the category of what we might call unfettered cash-for-passport exchanges. No “nexi”between the country and the passport recipient are required; only the investment monies must “reside” in the country for a fixed term. This is to be distinguished from more traditional programmes, themselves the subject of perennial critique, under which migrant millionaires (to borrow David Ley’s apt term) can receive an admission visa through a designated business-investment stream, but would then have to more or less comply with standard residency and naturalisation requirements (Dzankic 2012). Such programmes are found in, among other places, Australia, New Zealand, Hong Kong, the United Kingdom and the United States. Both kinds of programme raise serious ethical quandaries, but the unfettered cash-for-passport programmes are far more extreme and blatant than the traditional investment programmes. They contribute to some of the most disturbing developments in 21st-century citizenship, including the emergence of new forms of inequality and stratification. Instead of retreating to the background as some theorists had forecasted, states are proactively creating and exacerbating inequalities through their selective and managed migration policies, setting up easy-pass citizenship for some while making membership more restrictive and difficult to achieve for others. This new world order reveals tectonic pressures and introduces urgent dilemmas about the proper scale, scope and relations of justice and mobility, citizenship and (selective) openness. These developments also bear a profound impact on immigration law and policy on the ground, since they entail processes through which the boundary between state and market is constantly being tested, eroded, and blurred.

It is these intricate and underexplored interactions between state and market that are at the heart of my inquiry into emerging selective migration regimes and transactional visions of citizenship (Shachar 2006; Shachar forthcoming). Legally, the sovereign prerogative to issue a valid and internationally recognized passport is reserved in our international system to states alone. Governments and only governments—not markets—can secure and allocate the precious legal good of membership in the political community. But what happens when the logic of capital and markets infiltrates this classic statist expression of sovereignty? The proliferation of what I have called unfettered cash-for-passport programmes is a dramatic example of this pattern at work and it invites our critical scrutiny, especially since governments that use these programmes often do so in the name of advancing their country’s national interest while paradoxically setting up dangerous connections between money and access to citizenship, possibly to the detriment of the basic egalitarian and participatory thrust of political membership as we currently know it. These developments raise core ethical and legal questions. Why are states putting citizenship up for sale? And what precisely is wrong with easy-pass naturalisation along the lines of the cash-for-passport programmes? Is it the queue jumping? The attaching of a price tag to citizenship? The erosion of something foundational about political membership itself? Or, perhaps, all of the above?

Surely, zealous free-marketeers will enthusiastically defend such programmes as freeing us from the shackles of culture, nation and tradition and moving citizenship forward to a new and more competitive global age of transactional contracting in which, as Nobel Prize laureate Gary Becker once put it, a price mechanism substitutes for the complicated criteria that now determine legal entry (Becker 1992). As much as Becker would like to deny it, though, these programmes have something of a “whiff of scandal” not only due to frequent accusations of money laundering and fraud (Economist 2013), but also because of something deeper and more profound. Citizenship as we know it (at least since Aristotle) is comprised of political relations; as such, it is expected to both reflect and generate a notion of participation, co-governance, and a degree of solidarity among those included within the body politic. It is difficult to imagine how these values could be preserved under circumstances in which insiders and outsiders are distinguished merely by the ability to pay a certain price. The objection here is to the notion that everything, including political membership, is “commensurable” and reducible to a dollar value. This is what makes cash-for-passport exchanges, even if they account for only a limited stream or quota of entrants per year, deeply problematic and objectionable. The sale and barter of citizenship, even if initially reserved only for a small stream of recipients, nevertheless sends a loud and unmistakable message in both law and social ethics about whom the contemporary market-friendly state gives priority to in the immigration and naturalisation line and whom it covets most as a future citizen. This expressive conduct and the new grammar of market-infused valuation it entails tell us something important about the volatile state of citizenship today and the direction in which we may be heading.

Although economists will be quick to note that cash-for-passport programmes can create a hefty stream of revenue for governments, this is a hardly a strong enough justification to endorse them. The desire to enlarge their coffers may, as a matter of real-life experience, explain why some countries offer these programmes. As a normative matter, however, such an exchange threatens to corrupt the good that is put on sale: what changes when we “sell” citizenship is not just the price tag of membership, but its substantive content as well. As it plays a more and more important role in countries’ immigration and naturalisation policies and priorities, citizenship-for-sale may also gradually reshape the greater class of those who are likely to enjoy political membership. Reliance on a price mechanism alone, to the exclusion of other important considerations, would not only prevent the vast majority of the world’s population from ever gaining a chance to access citizenship in well-off polities. Taken to its logical conclusion (as reductio) it might also lead, corrosively and over time, to a world where anyone included in the pool of members must pay up, or risk “falling helplessly to the wayside” (Spiro 2008, 134).

Several scholars have taken up the task of imagining how our world might look were the market—rather than the state—to govern access to, and the acquisition of, political membership. As one study explains, “[i]f we take the basic incidents of citizenship to be protection of members and participation in modes of governance, the market for citizenship could form around offer of and demand for these services. Indeed, the offer of broader packages of citizenship services would be the basis for product differentiation” (Downes and Janda 1998, 55). “Product differentiation,” it should be noted, is a euphemism for providing lesser rights and services in exchange for lower fees (Jordan and Düvell 2003). Farewell, then, to the hard-earned ideal of inclusive citizenship as equal membership. In its absence, auction mechanisms and supply-and-demand rules may well replace our (however imperfect) procedures of exerting some degree of democratic governance and collective decision-making on what it means to belong to a political community, how to obtain a secure legal status of citizenship, and on what conditions.

Even staunch defenders of the market approach to citizenship understand that they are facing a hard sell. Becker, for one, admits that “people object to the sale of permits because, as they say, ‘citizenship is not to be for sale’” (Becker 1992), and this is a moral intuition that runs deep. As evidenced by recent debates over the instalment of cash-for-passport programmes, most people have strong reservationsagainst attaching a price tag to citizenship (Borna and Stearns 2002, 197). The reasons are many. As already mentioned, such a move may cause irreparable harm to the vision of citizenship as grounded in long-term relations of trust and shared responsibility, and may prefigure the conflation of the political and ethical with the economic and calculative. It may also undermine membership bonds grounded in co-authorship, cross-subsidisation of risk, and even sacrifice that might be expected in times of need. What is more, citizenship currently involves making collective decisions, and translating those decisions into binding commitments, in the context of a political project that is far larger than oneself, and that extends well beyond the lifespan of each generation of members—a time horizon that will be extremely hard to sustain under a regime of strategic transactions, according to which “wealth buys membership.” Turning citizenship into a money-based prize also contradicts any notion of complex equality through blocked exchange according to which advantage in one sphere (here, wealth) cannot be legitimately transferred to another (in this case, membership) (Walzer 1983). This makes the idea of selling membership unnerving for anyone who objects to the ultimate triumph of economics over politics, the reduction of our public life and ethics into mere pecuniary transactions, or the imperialistic idea that “trades” occupy the full terrain of human value and meaning (Radin 1987; Sunstein 1997; Sandel 2013).

Another set of concerns arises in the context of supranational citizenship, as in the derivative structure of European citizenship. The actions of those member states that take the liberty to put their national citizenship “on sale” indirectly affects the supranational political membership good that is shared by other countries, which may resist such commodification. There are also complex questions about to whom (beyond its own citizenry) the transacting government is obliged to provide justificatory reasons concerning its selective admission and naturalisation policies. Need it justify itself to other member states? To the Commission of the European Union? To would-be entrants who might have had a shot at admission through standard migration streams (family, employment, and humanitarian) but who are priced out of the advantage given to those who can afford a “golden passport”? From a global perspective, cash-for-passport programmes clearly exacerbate pre-existing inequalities rather than alleviate them. Should the sedentary populations of the migrant millionaires’ countries of origin, which are typically less stable or poorer than the destination countries, get to weigh in as well? Or, if an expansive all-affected-interests principle is applied, perhaps anyone at all who may be unfairly and arbitrarily affected should have a voice in these decisions (Goodin 2007). And what about migrants who are already settled in the country but ineligible to benefit from naturalisation schemes that require no knowledge or familiarity with the political structures, main civic institutions, history or language of the country, and who are subject instead to ever more demanding civic integration requirements? If civic integration is a required precondition to the bestowment of full membership by the state (as restrictive citizenship tests increasingly indicate), how can this demand only apply to some and not to others?

After all, there is no rational connection between delivering a stack of cash or sending in a bank wire transfer and establishing the kind of participation and equal standing among fellow citizens that the political bonds of membership are meant to represent and foster. From this vantage point, the very transaction in citizenship, even if carefully regulated and implemented by monopolistic governments or their authorised delegates, should be prohibited. Taken to its dystopian extreme, this approach may lead to a situation whereby the size of their wallets, and nothing else, distinguishes suitable from unsuitable candidates for initial entry and eventual citizenship. This kind of transaction, as lawyers and philosophers like to put it, is value-degrading: the trading in citizenship “taints,” “degrades” or outrightly “corrupts” (in the moral sense) its value as a good. We might in the same vein say that these cash-for-citizenship programmes detrimentally affect the “character of the goods themselves and the norms that should govern them”(Sandel 2012, 113). As critics of commodification have been at pains to clarify in other contexts (Cohen 2003), it is not that €1M is too high or too low a price, but that placing a “for sale” tag on citizenship, no matter what amount is written on it, has a corrosive effect on non-market relations, eroding the ties that bind and altering our view of what it means to belong to a political community. Just as I think we should be critical of granting citizenship according to nothing but the fortuitous and arbitrary circumstances of station of birth (Shachar 2009; Shachar 2011), I believe we must resist, with even greater force, the notion that money can buy “love of country”—or secure membership in it. 

If governments and activists are listening, they should heed the warning signs. The ideal of equal citizenship has been inflicted with many wounds over the past decades, and has always been more of an aspiration than a reality. However, the dangerous and increasingly frequent links between money and access to political membership reflected in the more calculated, mercantilist-like perceptions of citizenship that have given rise to unfettered cash-for-passport programmes threaten not only the implementation of the ideal, but the ideal itself. Courting the world’s moneyed elite by relaxing standard admission and naturalisation requirements may enrich the coffers of a country in the short run, but in the long haul it risks cheapening something far more important: citizenship itself.



Bauböck, Rainer and Sara Wallace Goodman, Naturalisation, EUDO Citizenship Policy Brief No. 2. 

Becker, Gary (1992), ‘An Open Door for Immigrants – the Auction,’ Wall Street Journal, Oct. 14, 1992, A1.

Borna, Shaheen and James M. Stearns (2002), ‘The Ethics and Efficacy of Selling National Citizenship’, Journal of Business Ethics, 37: 193-207.

Cohen, I. Glenn (2003), ‘The Price of Everything, the Value of Nothing: Reframing the Commodification Debate’, Harvard Law Review, 117: 689-710.

Downes, Daniel M. and Richard Janda (1998), ‘Virtual Citizenship’, Canadian Journal of Law and Society, 13: 27-61.

Dzankic, Jelena (2012), ‘The Pros and Cons of Ius Pecuniae: Investor Citizenship in Comparative Perspective’, EUI Working Papers, RSCAS 2012/14.

The Economist (2013), ‘Selling Citizenship: Papers Please’, Sept. 28, 2013.

Goodin, Robert (2007), ‘Enfranchising All Affected Interests, and Its Alternatives.” Philosophy and Public Affairs, 35: 40-68.

Joppke, Christian (2007), ‘Beyond National Models: Civic Integration Policies for Immigrants in Western Europe,’ West European Politics, 30: 1-22.

Jordan, Bill & Frank Düvell (2003), Migration: The Boundaries of Equality and Justice. Cambridge: Polity Press.

Ley, David (2010), Millionaire Migrants: Trans-Pacific Life Lines. Wiley-Blackwell.

Malta Today (2013), ‘Opposition Proposes Change of Name to Individual Donor Programme’ http://m.maltatoday.com.mt/newsdetails/news/national/Opposition-proposes-change-of-name-to-individual-donor-programme-20131109.

Orgad, Liav (2010), ‘Illiberal Liberalism: Cultural Reflections on Migration and Access to Citizenship in Europe’, American Journal of Comparative Law, 58: 53-106.

Radin, Margaret J. (1987), ‘Market-Inalienability’, Harvard Law Review, 100: 1849-1937.

Sandel, Michael J. (2013), What Money Can’t Buy: The Moral Limits of Markets. New York: Farrar, Straus and Giroux.

Shachar, Ayelet (2006), ‘The Global Race for Talent: Highly Skilled Migrants and Competitive Immigration Regimes’, NYU Law Review, 81 (2006): 148-206.

Shachar, Ayelet (2009), The Birthright Lottery: Citizenship and Global Inequality. Cambridge, MA: Harvard University Press.

Shachar, Ayelet (2011), ‘Earned Citizenship: Property Lessons for Immigration Reform’, Yale Journal of Law & the Humanities, 23: 110-158.

Shachar, Ayelet (forthcoming), Olympic Citizenship: Migration and the Global Race for Talent. Oxford: Oxford University Press.

Spiro, Peter J. (2008), Beyond Citizenship: American Identity After Globalization. Oxford: Oxford University Press.

Sunstein, Cass R. (1997), ‘Incommensurability and Kinds of Valuation: Some Applications in Law’, in Incommensurability, Incomparability, and Practical Reason, Ruth Chang ed. Cambridge, MA: Harvard University Press.

Walzer, Michael (1983), Spheres of Justice: A Defense of Pluralism and Equality. New York: Basic Books.




Cash-for-passports and the end of citizenship

By Peter Spiro

Investor citizenship programs are becoming increasingly commonplace in state practice. What was once the province of outlier Caribbean microstates is gaining traction among more substantial states. As an instrumental tool, states see citizenship-for-sale as a way to help get out of an economic hole on the cheap. There is no marginal material cost to minting new citizens, especially those with deep enough pockets to afford the price of admission. Hence the adoption of investor citizenship programs by such countries as Cyprus, Malta, and Portugal.

          I sympathize with Ayelet Shachar’s powerful framing of these programs. There is something unseemly, at least, about putting membership in the polity up for sale. Cash-for-passports, as Shachar derisively labels the phenomenon, clashes with our received understandings of citizenship as a marker of social solidarity in a Walzerian sense. The emerging market for citizenship literally commodities the status, the tip of an iceberg that Shachar is describing in other work as states come to see immigration as a talent-pool competition.

          But where Shachar sees investor citizenship programs as a threat to robust citizenship ties, I see them more as a manifestation of citizenship that is already being hollowed out. If citizenship still meant what it used to mean, if it still represented special ties as a sociological matter, then investor citizenship schemes would not exist. In that context, citizenship-for-sale would have implicated serious symbolic societal costs by breaking the social contract, understood not as an arm’s-length market transaction but rather as the locus of morally-inflected rights and responsibilities. In the old world, such programs would have been inconceivable.

          Today, far from inconceivable, they are becoming an accepted element of strategic immigration policy. Investor citizenship programs remain controversial (perhaps especially in a small, distinctive society such as Malta, which may more represent the old norm rather the new). But they are obviously gaining traction. States have something to sell. There must be some sentiment in adopting states that the revenues will exceed costs, social or otherwise.

          Investor programs give the lie to notion that citizenship is sacred, in a civic sense. The programs evidence the descent of citizenship from its former pedestal. Shachar extols a “vision of citizenship as grounded in long-term relations of trust and shared responsibility, . . . membership bonds grounded in co-authorship, cross-subsidisation of risk, and even sacrifice that might be expected in times of need.” That’s the citizenship of the past, and passport-for-sale schemes supply another data point to prove it.

          This notwithstanding externalities imposed on other states. In some contexts, these externalities will be miniscule (a citizen of Malta can travel to the United States visa free where the citizen of Russia cannot, but the numbers will be low, and the number who abuse visa-free entry will be even lower). In the European context they are potentially greater, as the EU member states become subject to lowest-common-denominator citizenship policies. Those who buy Maltese citizenship are less likely to settle in Valletta (one wonders how many could even name the capital city before – or perhaps even after – they have made the purchase) than in Berlin or Paris or London. When one buys Maltese citizenship one gets EU citizenship included in the price; it opens a backdoor to the rest of Europe. But the EU seems unlikely to complain. There is no legal basis for opposition, citizenship policy remaining exclusively within Member State discretion. Nor is there likely to be much pushback as a policy matter, so long as the price is high enough to depress numbers and maintain economic quality (as it were).

          In material terms, the programs are not much of a threat to provider states, either. The numbers will be low. (Portugal had only 330 takers in the first year of its program.) Because many buyers will remain non-resident, they will be invisible to the existing citizenry. They will not be politically engaged, to the extent they will feel no interest beyond protection of their bought-and-paid status. One possible cost would be with respect to diplomatic protection. It will be interesting to see whether that is a part of a bargain – whether in fact states will intercede with other states on behalf of their paying members (and whether international tribunals would recognize protection of cash-only nationals).

          Shachar is correct that the investor programs show that citizenship is still worth something. As the market thickens, we will see how much. With the reference point of states that sell permanent residency, we will be able approximately to isolate the value of citizenship itself – the premium states will be able to extract with the passport. Will investor programs like Malta’s, which offer citizenship, be priced much higher than  Hungary’s, which extends residency status only? (I will leave to the economists to deal with asymmetries among the various packages.) I suspect that premium will not be great, especially insofar as permanent residency includes the possibility of future eligibility for naturalization. Finally, there is the possibility of price competition as more states enter the market and some seek to maximize revenues by attracting more buyers at a lower price point.

          Investor citizenship programs are a symptom, not a cause. Shachar sees citizenship as something that can be rescued through citizenship policy. As material forces of globalization fragment citizen solidarities, citizenship law cannot revive them.


Citizenship for those who invest into the future of the state is not wrong, the price is the problem

By Raul Magni-Berton, University of Grenoble

Roughly two thousand years ago, Roman citizenship began to be sold to rich foreigners. As a consequence, rather than a way to share equal duties and rights, citizenship by the third century C.E. had become an aristocratic title. It divided people instead of rallying them. It increased inequalities instead of reducing them. 

          The current situation is somewhat similar. Rich people have access to rich countries’ membership, and poor people remain on the wrong side. Thus, I sympathise with Shachar’s concerns and I think we should avoid to reproduce what we have already experimented in our ancient history. 

          However, I do not agree with the way in which either Shachar or Spiro has identified the problem. Consider, for example, a situation in which a foreigner asks for access to citizenship in those terms: “I want to share the responsibility of my failures and achievements with you, and I’d like to invest in you and to be partly responsible of your achievements and your failures.” This is a touching statement of solidarity and identification with a group. I have called it the stockholder principle: individual citizens are like a joint-stock company in which fellow-citizens invest. The consequence of these collective investments is a shared responsibility for individuals’ achievements. Moreover, the right to benefit from public support is associated with the duty to invest in other fellow-citizens’ life projects. These duties are embodied in specific taxes for public investment. Thus, each citizen is also a stockholder with respect to other citizens. 

          Thus I would not say that the Maltese Parliament voted to “sell” the Maltese passport when it granted citizenship for € 650,000. From a foreign investor’s point of view, given that she makes the above statement and is ready to invest in the future of Maltese citizens, she acquires a moral claim to become citizen. She does not only give a sum of money in exchange for rights; she also becomes more largely committed to the duties of a Maltese citizen. In other words, she gains access to the Maltese nationality with an investment, which is a way to link her destiny to that of other Maltese.

          So what is wrong with this beautiful story? Why are the Maltese people sceptical and international opinion critical? Of course, we could agree that the argument of externalities, mentioned by Shachar, is relevant: European citizens should also benefit from those new investments. Thus, the problem is identifying who decides the allocation of those investments: the government of Malta or the EU. Although these externalities are expected to be low, as Spiro points out, it can be argued that Maltese citizens free-ride because they alone benefit from the foreigners’ will to become European, and this could be morally disputable. 

          Beyond that, the main argument I would like to develop here is that € 650,000 seems, at first sight, a lot. Not in absolute terms, of course. Suppose, for example, a society in which people spend € 200 on a movie. Several others things are likely to be true in such a society. Firstly, there are some people that can afford to pay this price. Secondly, there are no other less expensive goods which are substitutable; such as for example theatre, sport or other entertainment. Perhaps this is because technological progress has improved cinema so that it delivers a specific pleasure one cannot find elsewhere; or, alternatively, because theatres or circuses have simply gone bankrupt.

          Analogically, in the case of naturalisation, several other things are likely to be true in virtue of the fact that people prefer to pay a considerable amount of money, rather than to proceed with alternatives. For example, in a society where people are ready to pay € 650,000 for a passport, many of these alternatives are likely to be extremely burdensome, impractical, or unfair. 

          Let me assume that, until now, the Maltese way to naturalise foreigners has been fair according to the stockholder principle. In other words, a “poor foreigner” can be naturalised, if she is ready to share the responsibility for her failures and achievements with Maltese citizens, as well as invest in them and become partly responsible for their achievements and failures. Under this assumption, investing money in Malta, whatever the amount, is one fair way, among others, to gain access to citizenship. There is no reason, after all, to distinguish between financial and human investments. 

          But, if the Maltese law was fair, people would not be likely to invest € 650,000 to be naturalised. Of course, they could love Malta. They also could be so wealthy that they prefer to pay this amount rather than spend time in human investments. More probably, however, the fact that people are ready to pay this amount reveals that the law is in fact too restrictive and does not provide other reasonable ways to become citizen. 

          Naturalisation in Malta is possible after five years of residence, but it includes discretionary conditions, the severity of which can vary across time. In other European countries specific conditions and varying periods of residence are required. The greater the severity, the greater the price for passports. Investor citizenship programmes should be used to create a fruitful community, not to maximise price. 

          To conclude, I do not believe that investor citizenship programmes in themselves are unfair. On the contrary, they can reveal, via a financial argument, how hard the naturalisation process is. All European countries are concerned with this issue: too restrictive laws prevent motivated people to give their contribution to the host country and they divide humanity into rich and poor, rather than into different united groups. Exactly as the Roman Empire did.