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

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.