On 18 September 2014, Scottish citizens voted 55%-45% to stay in the United Kingdom. The turnout of 84.65% was the highest for any UK electoral event since the introduction of universal suffrage.¹xThis number significantly trumped the 65.1% who voted in the 2010 UK general election and the 50.6% who bothered to turn out for the 2011 Scottish parliamentary elections. This process has brought into vogue the referendum as a mode of settling issues of nationalism. It has been watched keenly in other countries such as Canada, Belgium and Spain. In this paper, I address the connection between the referendum as a decision-making device and sub-state nationalism. It is certainly the case that when sub-state nationalists wish to assert constitutional claims, in particular the most fundamental claim - to statehood itself - they invariably turn to the referendum, and indeed the referendum is becoming an ever more prominent aspect of sub-state nationalist politics.

It is important to set sub-state nationalism in context and so I begin by discussing the tenacity of modern nationalism in Western Europe since the constitutional demands of sub-state territories such as Scotland and Catalonia show no signs of abating. Secondly, I discuss the wider proliferation of the referendum around the world, of which the context we are looking at today is only part. This itself raises very interesting questions about the health of representative democracy. The referendum is becoming a fixed feature in many states for many purposes and is arguably part of a broader demand by citizens for a greater role in constitutional decision-making. This consideration has led me to conduct a theoretical re-evaluation of the role of the referendum in constitutional politics, which is a somewhat neglected topic. In this paper I review the classical critique of referendums and I make the case that the use of referendums as a device in constitution-framing and changing can be defended within democratic theory. I then turn to the Scottish process in 2014, asking if it is indeed a case study in good democratic practice which others might use as a future benchmark. In the final part of the text, I return to the specific application of referendums in the context of sub-state nationalism, addressing what might be called ‘the demos question’. This question has now been addressed by the Supreme Court in Canada and it is implicit in the preparedness of the UK Government to allow a referendum in Scotland; it does however remain unresolved in Catalonia and I end with some remarks about this broader context.

• 1 Sub-state nationalism in a globalising age: a paradox?

  Sub-state nationalism in Europe remains strong. We see this for example in the Scottish referendum, in on-going demands in Wales for more powers, and in the current volatile situation in Catalonia. One question which is often asked about modern sub-state nationalism, particularly in Europe, is why this is the case? Why has nationalism emerged as such a strong force since the late 1960s, when nationalism had acquired such a bad name in the first half of the twentieth century and when the trajectory of Europe appears to be one of inter-state or even supra-state federalisation?

  In general, the nationalist movements at sub-state level in Europe are not associated in any way with the kinds of ideologies that shaped the dark legacy of the twentieth century; this is certainly the case in Spain and the UK. The political programmes of these movements have helped make this clear, but there is of course also a burgeoning scholarship across a range of disciplines that helps to explain the nature of contemporary nationalism, and in doing so helps also to explain why the relationship between sub-state nationalism and contemporary globalisation is perhaps not as paradoxical as it might seem. It is all too easy to understand the term nationalism as having only one meaning, and then trying to lump in the Scottish or Catalan cases with some of the abhorrent movements we saw in the 1930s; some people continue to make this mistake. But various strands of new work have helped explain pretty fully that we can’t proceed on this basis and that we need to rethink the late twentieth century stereotype of the nation as a backward and out-dated vehicle for democratic politics and instead focus upon the democratic opportunities which come with more localised vernacular politics.

  We see the first challenge to the negative stereotyping of sub-state nationalism in the work of sociologists who have, since the 1960s, demonstrated the resilience of national identity and surprisingly the strengthening of it particularly at the sub-state level within democratic, liberal, tolerant states. Sociologists have found national identities to be resilient, but they have also found them not to be particularly thick, with markers of membership based decreasingly on ethnic markers and more on civic models of belonging.²xDavid McCrone, Understanding Scotland: The sociology of a nation, 2nd ed. (London: Routledge, 2001). They have also found that national identity remains strong even as cultural distinctions both within multinational states and around the world seem to diminish in an era of cosmopolitanism. The ever closer alignment of values among nations within states and the growing strength of nationalist sentiment within these nations, has been called ‘Tocqueville’s paradox’ by Canadian politician Stéphane Dion.³xStéphane Dion, ‘Le nationalisme dans la convergence culturelle: le Québec contemporain et le paradoxe de Tocqueville,’ in L’engagement intellectuel: Mélange en l’honneur de Léon Dion, ed. Raymond Hudon and Réjean Pelletier (Québec: Presses de l’Université Laval, 1991).

  Secondly, political scientists have addressed the constitutional aspirations of this new form of sub-state nationalism and found, contrary to many expectations, that political actors adopting the nationalist mantle are for the most part not backward-looking or reactionary, but espouse values wholly consistent with the plurality of opinion in modern, Western societies, for example on issues such as social welfare, citizenship and human rights.⁴xMichael Keating, Nations Against the State - The New Politics of Nationalism in Quebec, Catalonia and Scotland, 2nd ed. (UK, Palgrave, 2001); Alain-G. Gagnon and James Tully, ed., Multinational Democracies (Cambridge: Cambridge University Press, 2001). Furthermore, they have advanced political programmes that run largely with the grain of changing state power, supra-state integration, and internationalisation of previously monopolistic state functions. Nationalists in Scotland, Catalonia and Quebec situate themselves within the context of their respective integrating continents in ways similar to state nationalists, and in some ways are in fact more pro-integrationist.⁵xMichael Keating, Plurinational Democracy: stateless nations in a post-sovereignty era (Oxford: Oxford University Press, 2001). We certainly saw this in the Scottish referendum where the nationalist platform was built upon an ‘independence in Europe’ project and which attempted at every turn to contrast its position with the Euro-scepticism of the UK Conservative Party.

  And thirdly we find in political theory, including perhaps surprisingly liberal political theory, a subtle turn that has served to question the notion that nationalism and liberalism are inherently incompatible. Political philosophers, most comprehensively Will Kymlicka, have found the aspirations of these national groups to be wholly consistent with liberalism and have in fact argued that liberalism has a duty to accommodate these political and constitutional aspirations, if it is to be true to its own values of liberty and equality.⁶xWill Kymlicka, Multicultural Citizenship: A Liberal Theory of Minority Rights (Oxford: Oxford University Press, 1995); Margaret Moore, ‘Normative Justifications for Liberal Nationalism: justice, democracy and national identity,’ Nations and Nationalism 7 (2001): 1-20; Ferran Requejo, ed., Democracy and National Pluralism (London: Routledge, 2001). A main strand of this approach is the contention that it is only in the context of individual people’s societal culture that they can advance their own life goals in a fulfilled way, achieving the liberal ideal of free and equal citizens.⁷xWill Kymlicka, Politics in the Vernacular (Oxford: Oxford University Press, 2001); Stephen Tierney, ‘The Search for a New Normativity: Thomas Franck, Post-Modern Neo-Tribalism and the Law of Self-Determination,’ European Journal of International Law (2002): 941-60. By this argument, the nation becomes the essential vehicle for the fulfilment of individual self-determination through the collective ties each individual shares with his or her primary political community.

  Therefore, just as in empirical terms national identity has shown itself to remain strong, within the social sciences there is considerable evidence that it is also adaptable, and can remain fit for purpose in an internationalising world. It is a story in which national identities remain resilient, with people finding primary identity in their societal cultures, while at the same time becoming open to the opportunities that an increasingly cosmopolitan world has to offer. This should perhaps not be too surprising; it is also the story of on-going state nationalism in Europe. It seems that the error is in trying to understand state and sub-state nationalism as categorically different phenomena; they are not. Instead what we have seen is rival nation-building projects ongoing within the same state, with each approaching modernity and indeed post-modernity in similar ways.

• 2 The age of referendums

  I will now turn to the second element in the story of sub-state nationalism: the referendum. Notably the increased use of referendums or at least the attempted use of referendums is part of a wider phenomenon. Over the past few decades the referendum has become a fixed feature of state and constitution-building across the globe. In Table 1 above, I offer a breakdown of how referendum use has grown in four main areas of constitutional practice.

  Table 1

  4 types of referendum in the ‘new wave’	Examples
  Founding of new states	New states emerging from the former USSR and SFRY, Eritrea (1993), East Timor (1999), Montenegro (2006), South Sudan (2011), Scotland (2014).
  Creation and amendment of new constitutions	Former republics of the USSR and SFRY, Iraq: ratification of the Constitution (2005), Egypt: constitutional reforms (2011).
  Sub-state autonomy	Spain referendums: e.g. Basque Country (1979), Catalonia (1979), Galicia (1980). United Kingdom referendums: Scotland (1997), Wales (1997, 2011), Northern Ireland (1998).
  European Union: treaty-making processes in respect of both integration and accession	Malta, Slovenia, Hungary, Lithuania, Slovakia, Poland, Czech Republic, Estonia, Latvia (all in 2004), Croatia (2013).

  The aspirations of sub-state peoples for constitutional change have been key to at least two of these processes. First, in the founding of new states, the referendum was widely used in the early nineties in the break-up of the USSR and SFRY and its use is now the default mechanism for the emergence of most new states as exemplified by Eritrea (1993), East Timor (1999), Montenegro (2006) and South Sudan (2011). And of course a referendum was held in Scotland in September 2014 with the question: ‘Should Scotland be an independent country?’

  Secondly, referendums were once very rarely used in the creation or amendment of constitutions, but again throughout Central and Eastern Europe and more recently in Iraq and Egypt, we see the referendum emerge both in the founding of new constitutions and within the text of these constitutions as part of future amendment procedures. This can also be traced to sub-state processes since many of those sub-state peoples in Central and Eastern Europe who achieved statehood by way of a referendum, either had a subsequent referendum to ratify the new constitution and/or included the referendum in the new constitution as central to future processes of constitutional amendment.

  In some sense of course these two examples show how the referendum is used in processes that lead to the break-up of plurinational states. But referendums are also central, as a sub-set of the second group, in establishing complex new models of sub-state autonomy as we have seen in Spain and the UK in the late 1970s and 1990s respectively, and in ongoing processes of constitutional change as we saw in a referendum on further devolution for Wales in 2011. A related example is the referendum on the draft Charlottetown Accord in 1992 where distinct referendum processes were held respectively in Quebec and the rest of Canada.

  Fourthly, it is also notable that we have seen a major proliferation in referendum use in the accession to and the transfer of sovereign powers from European states to the European Union. To take an example, of the first 15 countries to join the EU only Ireland and Denmark held referendums to ratify the decision. Of the 10 accession countries in 2004 only Cyprus did not; and we see the trend continue in January 2012 with Croatians voting in a referendum for accession in 2013.

• 3 Democratic issues

  The rejuvenation of referendums in so many different directions led me to reassess the theoretical argument for and against referendums in my book Constitutional Referendums: The Theory and Practice of Republican Deliberation.⁸xStephen Tierney, Constitutional Referendums: The Theory and Practice of Republican Deliberation (Oxford: Oxford University Press, 2012). There are other legitimacy questions at stake, to which I will return in the final question, but a crucial issue when we think about the referendum in the context of sub-state nationalism is: is the referendum at all an adequate device for the making of significant constitutional decisions? I take it that there are within political theory three main objections, on democratic grounds, to referendums.

  The first objection is what can be called the elite control syndrome. It is argued that referendums lend themselves by definition to elite control and hence to manipulation by the organisers of the referendum. This suggests that the executive staging of a referendum has wide discretion to set the rules for the referendum, without much in the way of oversight or control by the legislature. In particular, the initiation power lies in the hands of the executive which can decide when and on what issue to hold a referendum.

  The second main criticism of referendums is that there is an in-built tendency of the referendum process merely to aggregate pre-formed opinions rather than to foster meaningful deliberation. This is the so-called deliberation deficit. In other words, referendums engage the voters simply at the time of voting which they do unreflectively without real reflection or collective discussion of the issues.

  A third criticism of referendums, which I call the majoritarian danger, is that they represent a model of majoritarian decision-making that imperils the interests of dissenting individuals and minorities. For many, this is the main objection to the referendum: not only is it a poor way of making decisions, it can be deeply dangerous. Referendums usually involve a simple 50% plus 1 majoritarian model leading to a winner takes all outcome; and in the end, a majority may simply vote to harm a minority. Of course, it is widely held that the point of democracy is not simply to give expression to the will of the majority but also to protect individual rights, and where possible to arrive at decisions that enjoy the broadest levels of support (or at least acquiescence) possible.

  My research suggests that these criticisms are more issues of practice than of principle and that by applying the recent ‘deliberative turn’ in political theory and practice, it may be possible to overcome these concerns adequately. In other words, by trying to build more scope for citizen deliberation into the referendum process it is possible to enhance its democratic legitimacy. To achieve this end, legal regulation is crucial: a detailed legal framework for referendums, far from being a restraint on the popular will, can arguably help to facilitate a more profound engagement of citizens by offering a clear and properly regulated process for meaningful democratic deliberation that goes well beyond the mere casting of ballots. In this sense, law and constitutionalism can facilitate for citizens a meaningful expression of their voice, freeing them from easy manipulation.

  The turn to deliberative democracy is to some extent traceable back to the work of John Rawls in the early 1970s,⁹xJohn Rawls, A Theory of Justice (Cambridge, Mass.: Harvard University Press 1971). but it was given a sustained push in the past 10 to 15 years in a number of directions, particularly among those who want to see a greater role for the citizen in democratic decision making. The work on deliberative democracy is by now a broad church and there are many areas of disagreement among theorists as to the key values of deliberative democracy. But a common commitment is that political decisions should be preceded by ‘authentic deliberation’, or what John Rawls calls ‘public reason’.¹⁰xJohn Rawls, ‘The Idea of Public Reason Revisited,’ University of Chicago Law Review 64 (1997): 765. According to this principle, people engaged in making decisions should reflect authentically and honestly before they do so, engage publicly with others, be prepared to defend their views and be open to persuasion by the arguments of others.

  The key principles I have distilled from this work on deliberative democracy and which can be translated to referendum design, are the following ones:
  

  • Popular participation: a referendum reminds us more than any other type of electoral event that the fundamental point of democracy is a determining role for the people. It seems that a key aim in any attempt to regulate a referendum adequately should be to maximise popular participation not only in terms of voter turnout but also in terms of awareness-raising and engagement with the issues.

  • Public reasoning: a democratic referendum should be conducted in such a way that it provides a meaningful environment for reflection and discussion. This is important first at the elite level: do we see referendum rules being framed in a deliberative way, facilitating a fair debate between elites over the substantive issues? And secondly, do we see it fostered at the level of the citizen in a meaningful attempt to engage individuals and civil society throughout the campaign? There is also a role for civil society here, and in particular for the media. To what extent do private actors seek to inform and enlighten public debate, rather than to take an overtly partisan line?

  • Inclusion and parity of esteem: is the referendum process inclusive? Are the franchise rules fair? Are sufficient efforts made to maximise both voter registration and turn-out? Do ordinary citizens, including marginalised groups and minorities, feel they have a voice and are being listened to?

  • Consent in collective decision-making. Does the referendum lead to a result that reflects public opinion? Can it be said to be, in the end, a satisfactory exercise in collective decision-making? Is the process fair and democratic, so that the result is one which even the losing side can agree to, if not with?

  In Table 2 below, I set out how the key principles mentioned above, drawn from the theory on deliberative democracy, can be applied to referendum design in order to remedy the democratic objections against referenda.

  Table 2

  The contribution of deliberation to the decision-making process	Principles of civic republican deliberation	Addressing the following criticisms of referendums
  Who deliberates?	Popular participation within a broader system of representative government	Elite manipulation: ‘the elite control syndrome’
  How should they deliberate?	Public reasoning: reflection and discussion	Aggregation: ‘the deliberation deficit’
  Under what conditions?	Equality and parity of esteem	Majoritarianism: ‘the majoritarian danger’
  To what end?	Consent and collective decision-making	All of the above

• 4 The Scottish referendum

  The referendum in Scotland offers an ideal case study to test the hypothesis of deliberative democracy as a remedy for criticism of referendums, as formulated in the previous section. Firstly, because it was organised within a healthy and fully-functioning democracy. Secondly, it was long in the planning: the Scottish Government announced its intention to hold a referendum in January 2012,¹¹x‘Your Scotland - Your Referendum - A Consultation Document’, Scottish Government white paper, January 25, 2012, http://www.scotland.gov.uk/Publications/2012/01/1006. some two and a half years before the vote itself, thus offering a lengthy span of time within which channels of deliberative participation might be fostered. Thirdly, the UK already had a model of detailed regulation of referendums in place (Political Parties, Elections and Referendums Act 2000) which, inter alia, created an independent Electoral Commission and invested it with a detailed oversight role in UK referendums. The existing UK legal regime was very influential in the framing of the Scottish referendum process. Fourthly, the referendum process was framed against, and given additional legal authority and political credibility by the ‘Edinburgh Agreement’ between the UK and Scottish governments, the aim of which was to ensure the referendum delivered ‘a fair test and a decisive expression of the views of people in Scotland and a result that everyone will respect’.¹²x Agreement between the United Kingdom Government and the Scottish Government on a referendum on independence for Scotland, October 15, 2012, http://www.scotland.gov.uk/About/Government/concordats/Referendum-on-independence. And finally, the referendum itself was regulated by two statutes passed by the Scottish Parliament - the Scottish Independence Referendum (Franchise) Act 2013 (hereafter ‘Scottish Franchise Act’), and the Scottish Independence Referendum Act 2013 (hereafter ‘Scottish Referendum Act’) - which together offered a comprehensive framework of rules and constraints.

  I will, therefore, assess how well this legislation has served, first to constrain elite control, and second to help foster meaningful deliberation, particularly among citizens. Scotland is not a divided society with readily identifiable minorities whose interests are clearly imperilled by an exercise in majoritarian decision-making. Therefore, I will not discuss the third criticism of referendums (the majoritarian danger) further here. It is notable however that the referendum decision was reached upon the basis of a simple 50% plus 1 majority rule. This has been a controversial issue elsewhere, particularly in Canada,¹³xStephen Tierney, Constitutional Law and National Pluralism (Oxford: Oxford University Press, 2004), 320-2. but it was never a focus of the debate in the Scottish context, and again for this reason I will not address it further here.

  Constraining elite control?

  The ‘elite control syndrome’ is generally the main objection to referendums. The charge is that referendums are organised by governments to effect political goals and are therefore only staged when the prospects of a successful outcome (from the government’s viewpoint) are favourable. To this end an executive is able to shape – indeed manipulate – the various elements of process design to achieve the desired result. Besides the fact that they have the power of initiation, they can also set the question, choose the date, fix the franchise, decide whether and if so how the referendum will be regulated, and determine the funding and spending rules for the campaign.

  How did the Scottish process measure up in response to this problem? It would seem that in fact elite power was well dispersed between the UK and Scottish Governments, with the two Parliaments also having significant roles to play. On 15 October 2012, the two governments signed ‘the Edinburgh Agreement’.¹⁴xKeating, Nations Against the State; Gagnon and Tully, Multinational Democracies. In this agreement, the UK Government accepted the principle that the Scottish Government could hold a referendum and handed over most of the process issues to the Scottish Parliament.¹⁵xThis provided that the referendum should have a clear legal base, to be legislated for by the Scottish Parliament and to be conducted so as to command the confidence of parliaments, governments and people. This was formalised by an Order in Council (per Scotland Act 1998 s30) which devolved to the Scottish Parliament the competence to legislate for a referendum on independence which had to be held before the end of 2014. Scotland Act 1998 (Modification of Schedule 5) Order 2013, para 3, http://www.legislation.gov.uk/uksi/2013/242/made. In the course of 2013, the referendum was then regulated by two statutes passed by the Scottish Parliament that built upon existing UK law. For over a decade UK referendums have operated on the basis of a dedicated referendum law (PPERA 2000). Notably the PPERA 2000 only applies to referendums organised by the Westminster Parliament and so it did not regulate the proposed referendum in Scotland. Its terms however acted as an important benchmark for the Scottish Government in drafting the Scottish Franchise Bill and the Scottish Referendum Bill, and for the Scottish Parliament deliberating upon these drafts.¹⁶xThe Edinburgh Agreement (para 2) provided: ‘Both governments agree that the principles underpinning the existing framework for referendums held under Acts of the UK Parliament – which aim to guarantee fairness – should apply to the Scottish independence referendum. Part 7 of the Political Parties, Elections and Referendums Act 2000 (PPERA) provides a framework for referendums delivered through Acts of Parliament, including rules about campaign finance, referendum regulation, oversight and conduct.’

  PPERA also created the independent Electoral Commission which has various responsibilities, most of them related to funding and spending rules, but also in relation to the intelligibility of the referendum question. This body had no automatic role in relation to the Scottish referendum but such a role was in time guaranteed by the Scottish Franchise Act and Scottish Referendum Act, and by the actions of the Scottish Government. Eventually, the Commission played a major role in a number of areas, including the setting of the question, the provision of information to citizens, the regulation of funding and spending, and the management of the mechanics of registration and voting.

  So in the end, by dint of the limited competence of the Scottish Parliament and the subsequent uncertainty concerning its power to hold a referendum, the initiation power for the Scottish referendum was in fact regulated more fulsomely and by a more plural array of actors than would an equivalent process organised by the UK Government. In short we can say that elite control was pluralised, that it was regulated by different pieces of legislation and that it was overseen by an independent commission.

  Facilitating deliberation?

  The second main criticism of referendums, by which they are often held in contrast to the purported merits of representative democracy, is that public reasoning – which allows for the informed reflection and discussion of ideas before decisions are reached – is absent from referendum processes. Various assumptions underpin this idea: referendums tend to be held quickly by way of a snap poll organised at the behest of the government; voters are presented with an issue which is itself confusing and can be made worse by an unintelligible question; voters themselves lack the time, sufficient interest in the matter at stake or the competence to understand or engage properly with the issue, and in effect turn up at the polling station, if they bother to do so at all, in an unreflective manner often following party cues in determining how to vote.

  The regulatory design of the Scottish referendum answers a number of these criticisms. The franchise is the first of these mechanisms. In a mass popular engagement with democracy, both participation and deliberation are vital. It is not enough that those who make the decision do so in a reflective and discursive way, it is also essential that the process should generate the widespread engagement of citizens across the polity if the exercise is to be truly legitimate. That the franchise is defined in a properly inclusive way, is therefore the first step in achieving this goal. The body of voters in the Scottish referendum was largely uncontroversial: the franchise for the referendum was the same as for Scottish Parliament elections and local government elections,¹⁷xScottish Independence Referendum (Franchise) Act 2013, s. 2. mirroring the franchise used in the Scottish devolution referendum in 1997. One consequence of this franchise was that EU citizens who are resident in Scotland were able to vote in the independence referendum.

  One major difference from the 1997 franchise, however, was the provision in the Scottish Franchise Act extending the vote to those aged 16 and 17.¹⁸x Ibid., s. 2(1)(a). Another notable provision of the Scottish Franchise Act excluded convicted persons from voting in the referendum if they were detained in a penal institution.¹⁹x Ibid., s. 3. This has been a controversial topic in the UK ever since the European Court of Human Rights ruled that the blanket ban on prisoner voting in UK elections violated Article 3 of Protocol 1 of the European Convention on Human Rights.²⁰x Hirst v. the United Kingdom (No 2) [2005] ECHR 681. It seemed clear, however, that section 3 of the Franchise Act did not violate the Convention, since A3P1 guarantees ‘the free expression of the opinion of the people in the choice of the legislature’ [emphasis added], which is generally taken to refer exclusively to parliamentary elections and to exclude referendums.²¹xStephen Tierney, ‘Possible vires issue in relation to section 3 of the Scottish Independence Referendum (Franchise) Bill, Advice to the Scottish Parliament, Scottish Referendum Bill Committee paper, March 21, 2013, http://www.scottish.parliament.uk/S4_ReferendumScotlandBillCommittee/20130321_Letter_to_DFM_on_prisoners_voting_rights.pdf. This view has been endorsed by Lord Glennie in the Outer House of the Court of Session,²²x Moohan and others, Petitioners [2013] CSOH 199. and subsequently by the Inner House²³x Reclaiming Motion: Moohan and others v. Lord Advocate [2014] CSIH 56. and the UK Supreme Court.²⁴x Moohan and Another v. Lord Advocate UKSC 2014/0183 (24 July 2014).

  Franchise is one thing, citizen engagement is quite another thing. Astonishingly 4,285,323 people (97% of the electorate) registered to vote and in the end 84.7% turned out, the highest figure for any UK electoral event since the introduction of universal suffrage, significantly trumping the 65.1% who voted in the 2010 UK general election and the 50.6% who bothered to turn out for the 2011 Scottish parliamentary elections. Turnout is of course only one marker of participation. Yet the story that could be heard time and time again from voters and campaigners alike is that citizens felt greatly empowered by the referendum and the role they had in making such a huge decision. There is evidence emerging of the extent to which people sought out information about the issue at stake and engaged vociferously with one another at home, in the workplace, in pubs and public meetings, and, to an unprecedented degree in British politics, on social media through online newspaper comment sections, Twitter, Facebook, blogs etc.²⁵xSee Ailsa Henderson et al. ‘Risk and Attitudes to Constitutional Change,’ ESRC Scottish Centre on Constitutional Change Risk and Constitutional Attitudes Survey, August 16, 2014, http://www.futureukandscotland.ac.uk/sites/default/files/news/Risk%20and%20Constitutional%20Attitudes%20Full%20Survey%2014%20Aug.pdf and AQMen project https://www.aqmen.ac.uk/project/socialmedia. My own evidence is merely anecdotal, but as someone who lived through the referendum campaign, I can say that especially in the month before the vote, I experienced a level of public engagement with a major political issue the like of which I had never known before.

  An intelligible question?

  Do citizens understand the question posed in the referendum? If not, then meaningful deliberation in referendum is not possible. Here again the Scottish process worked well. The Scottish Government originally proposed ‘Do you agree that Scotland should be an independent country? Yes/No’. The Scottish Government decided to send its proposed question for review to the Electoral Commission. This process was concluded quickly and the Commission reported back suggesting a change to the question.²⁶xThe Electoral Commission took the view that ‘based on our research and taking into account what we heard from people and organisations who submitted their views on the question, we consider that the proposed question is not neutral because the phrase “Do you agree …?” could lead people towards voting “yes”.’ They therefore recommended the following question: ‘Should Scotland be an independent country? Yes/No.’ ‘Referendum on independence for Scotland: Advice of the Electoral Commission on the proposed referendum question,’ The Electoral Commission, 2013, http://www.electoralcommission.org.uk/__data/assets/pdf_file/0007/153691/Referendum-on-independence-for-Scotland-our-advice-on-referendum-question.pdf. The suggestion was accepted by the Scottish Government and the new question was included in the Scottish Referendum Act.²⁷xScottish Independence Referendum Act 2013, s. 1(2). See also ‘Scottish independence: SNP accepts call to change referendum question,’ BBC News, January 30, 2013, http://www.bbc.co.uk/news/uk-scotland-scotland-politics-21245701.

  The Electoral Commission also offered the view that the clarity of the question hinged not only on its syntax, but also upon the content of the independence proposal: ‘Clarity about how the terms of independence will be decided, would help voters understand how the competing claims made by referendum campaigners before the referendum will be resolved.’²⁸x‘Referendum on independence for Scotland: Advice of the Electoral Commission on the proposed referendum question,’ The Electoral Commission, January 30, 2013, http://www.electoralcommission.org.uk/__data/assets/pdf_file/0007/153691/Referendum-on-independence-for-Scotland-our-advice-on-referendum-question.pdf, para 5.41. See also generally paras 5.41-5.44, e.g., ‘We recommend that both Governments should agree a joint position, if possible, so that voters have access to agreed information about what would follow the referendum. The alternative - two different explanations - could cause confusion for voters rather than make things clearer.’ (para 5.43). This is an interesting comment, reflecting as it does the requirement that a fully deliberative process is only possible if citizens know what they are voting for.²⁹xOne of the main criticisms of the Quebec referendum in 1995 was that the proposal of sovereignty and partnership was not well understood by citizens. Tierney, Constitutional Law and National Pluralism, 293-9. The Scottish Government published a White Paper in November 2013 which sets out its vision of independence,³⁰x‘Scotland’s Future,’ Scottish Government white paper, November 26, 2013, http://www.scotland.gov.uk/Publications/2013/11/9348. followed by a White Paper and a draft Scottish Independence Bill, which sought to lay out proposals for both an interim and a permanent constitution for an independent Scotland.³¹x‘Scottish Independence Bill: A consultation on an interim constitution for Scotland,’ Scottish Government white paper, June 2013, http://www.scotland.gov.uk/Publications/2014/06/8135/downloads. Each of these were of course heavily criticised by the UK government and by the Better Together campaign. The November White Paper in particular led to a series of papers by the UK Government contesting many of the claims made in the White Paper. In the end, and surely inevitably, citizens were left with a debate in the context of the referendum campaign, rather than any agreed set of ‘facts’ about what independence would look like. These events highlight that public deliberation in a real democracy can never take place in a controlled vacuum. Citizens have to try to distil facts from the political debates, and even in some cases from the political deceit. But this obviously this goes for ordinary elections as well.

  Sufficient time to understand the issue?

  Another prerequisite of serious public deliberation is that people have time to consider the issue upon which they are being asked to vote. An element of control which was left to the Scottish Government by the Edinburgh Agreement was the timing of the referendum. In January 2012, it set its course for a referendum some two and a half years hence. This timing was of course strategic. The Scottish Government saw the autumn of 2014 as a propitious time. It allowed the Scottish National Party sufficient time to make the case for independence and it would coincide with a number of significant events. It would also come shortly before a UK General Election which would distract the UK parties and perhaps make them less inclined to work together. But, from the perspective of deliberative participation, a beneficial side effect of this procrastination was that the election debate was conducted over a very long period of time, permitting each side to make its case in full and allowing citizens the time and space necessary to consider the issues. Other important timing issues included in the legislation are the 16 week ‘referendum period’³²xScottish Independence Referendum Act 2013, sched. 4, part 3. and the four week ‘purdah period’³³x Ibid., sched. 4, para 25. to which I will return.

  Sufficient information available?

  Citizens cannot deliberate properly without access to sufficient, reliable information. In many ways the existence of such a favourable environment depends more upon the health of a particular democracy and of its civil society than it does upon legal regulation. Indeed, excessive regulation can in fact serve to inhibit the free flow of ideas. In the Scottish referendum there were many sources of information: both governments,³⁴xWe have noted ‘Scotland’s Future’, Scottish Government white paper November 26, 2013 op. cit. In June 2014 the UK government sent a 16-page booklet called ‘What Staying in the UK Means for Scotland’ to every household in Scotland, setting out the case for the No side. See ‘UK “fact booklet” to be sent to Scots households’, The Scotsman, June 12, 2014. each campaign group, other registered participants in the referendum campaign, and various other sources in civil society³⁵xSee for example on the franchise and mechanics of voting issued by the Citizens Advice Bureau, http://www.adviceguide.org.uk/scotland/law_s/law_civil_rights_s/law_government_and_voting_s/the_scottish_independence_referendum_s.htm. including the media and academia.³⁶xSee for example the major ESRC investment: ‘Future of the UK and Scotland,’ http://www.esrc.ac.uk/research/major-investments/future-of-uk-and-scotland/index.aspx. As noted above, a particularly notable source of information was also social media, with Twitter, Facebook and blogs playing a major role in the dissemination of ideas and also in their discussion and debate.

  The provision of information by the two campaigns has been touched upon in the context of the November White Paper, but another related issue is the provision, and indeed the very possibility of providing independent or neutral information to voters through the Electoral Commission. Among a number of statutory duties, the Commission was given by the Scottish Referendum Act the task of promoting public awareness and understanding in Scotland about the referendum, the referendum question, and voting in the referendum.³⁷xScottish Independence Referendum Act, s. 21. From the outset, this was going to be a very challenging duty given the deep disagreement between the two campaigns about what exactly was meant by ‘independence’.³⁸xStephen Tierney, ‘Why is Scottish Independence Unclear?,’ U.K. Const. L. Blog, February 25, 2014, http://ukconstitutionallaw.org/. It was hard to see how the Electoral Commission could attempt to produce an objective account of a number of highly technical and fiercely contested issues, concerning not only international relations but also defence, economic relations, the question of a currency union, the disentanglement of the welfare state, national debt etc., particularly when so many features of the post-referendum landscape would be contingent upon negotiations between the two governments in the event of a majority ‘Yes’ vote. It was argued with evidence before the Scottish Parliament Referendum Bill Committee that it was simply not possible to perform such a role in a neutral way.³⁹xWilliam Norton evidence before the Referendum (Scotland) Bill Committee, May 9, 2013, Official Report, http://www.scottish.parliament.uk/parliamentarybusiness/28862.aspx?r=8140&i=73882&c=1482666.

  In the end, the Electoral Commission intimated that it would ‘not seek to explain to voters what independence means’ but would offer information ‘aimed at ensuring that all eligible electors are registered and know how to cast their vote.’⁴⁰xJohn McCormick, the Electoral Commissioner for Scotland, evidence to: Scottish Parliament Referendum (Scotland) Bill Committee, May 23, 2013, Official Report, col 431, http://www.scottish.parliament.uk/parliamentarybusiness/28862.aspx?r=8326&mode=pdf. The Electoral Commission also concluded that both governments had a duty here to help explain the implications of either a ‘Yes’ or a ‘No’ vote and that a joint position on this would benefit voters. This was a brave attempt to improve the deliberative environment for citizens, even if in political terms such a joint position was never likely to be achievable. In the end though a document was published which contained a two page joint statement that sets out a useful summary of what would happen following a ‘Yes’ or a ‘No’ vote.⁴¹x https://www.electoralcommission.org.uk/__data/assets/pdf_file/0012/170400/The-2014-Scottish-Independence-Referendum-Voting-Guide.pdf. It does not try to explain what independence means, but it does discuss the need for negotiations in the event of a Yes vote, and the balance of devolved and reserved competences in the interim period.

  Promotion, funding and spending

  Other areas of regulation were a 16 week period in which the funding and spending of both campaigns was tightly regulated, and a four week purdah where the Scottish Government and a wide range of other public bodies could not engage in promotional activity. The UK Government also committed to be bound by equivalent restrictions in the Edinburgh Agreement.⁴²xFor comment on this by Deputy First Minister Nicola Sturgeon, see Scottish Parliament Referendum (Scotland) Bill Committee, June 13, 2013, Official Report, cols 554 and 560. In terms of funding and spending rules, the Scottish Referendum Act sought to ensure equality of arms between the two campaign groups. Final audited accounts are not yet available, but it seems to have been largely successful in ensuring a fairly level playing field.

  As a conclusion, one could say that the Scottish referendum seems to have been quite successful in helping to provide an environment for democratic engagement and this precedent is likely to be of interest to many countries in which the referendum is a growing feature of constitutional politics, and not only in the sub-state context.

• 5 The demos question: the ultimate challenge of sub-state nationalism

  In the final part of this paper, I return to the specific issue of sub-state nationalism and the fact that the referendum is a device that is becoming more and more popular for sub-state nationalists. One obvious reason for this popularity is that the referendum is a means to advance political claims. In the cases discussed in this text, the sub-state polity already has a level of autonomy, and with autonomy come institutions such as a sub-state government and legislature that are able to organise referendums, as we have seen in Quebec and Scotland, although not yet in Catalonia. But why has the referendum been chosen specifically? There are other ways to make political claims – for example, the victory of a nationalist party in elections. The answer seems to be that the referendum is seen to provide a particular form of validation. The people are expressing a specific view on a specific issue, which helps to endorse the particular political claim (i.e. of independence). People voting for a nationalist political party may do so for many reasons and may not want independence.

  But is not just that the referendum brings to the table one single issue; crucially it also involves the people speaking directly and in doing so they can be seen to be engaging in a collective constituent act. This has tremendous symbolic and rhetorical power. Yet it also leaves us with one major question: it is all very well to claim that a referendum lets the people decide, but who are the people? This was the question Ivor Jennings asked in the decolonisation context of the 1950s. The very existence of any democratic polity and any political act made in its name implies a manifestation of ‘the people’. But how do we legitimately demarcate the political space within which a people can be said to exist and act politically? Political theorists call this the ‘democratic boundary problem’. It is perhaps not surprising that this preliminary question of democratic thought – who are the people? – invariably goes unarticulated in the day to day life of representative democracy. The most obvious and indeed practical reason is that there is no need to dwell on this: it is simply assumed that ‘the people’ is a synonym for the citizenry of the state.⁴³xFor Dahl, the great philosophers have ignored this question because ‘they take for granted that a people has already constituted itself’. See Robert A. Dahl, After the Revolution? Authority in a Good Society (New Haven: Yale University Press, 1970), 60-1. Others consider its neglect to be more surprising. Canovan has observed: ‘Unlike “freedom”, “justice”, or even “nation”, “people” has attracted hardly any analysis, even by theorists of democracy’, a fact she deems ‘astonishing’. See Margaret Canovan, ‘Populism for Political Theorists?’ Journal of Political Ideologies (2004): 241, 247. And Goodin comments: ‘the first question for any democratic theory is who is included’, but notes, ‘[t]his is a strangely under-explored issue within democratic theory’. See Robert Goodin, Innovating Democracy. Democratic Theory and Practice after the Deliberative Turn (Oxford: Oxford University Press, 2008), 3. The latter can be taken to encompass the former in most cases quite happily; and if the state works, leave it be. Another reason is that to explore this question too closely can in some cases be dangerous. In a number of polities the nature of the demos and even the territorial boundaries of the state are contested affairs. In such a situation, why not let sleeping dogs lie rather than invite a confrontation over inclusion and exclusion?

  In referendums, however, and in constitutional referendums particularly, a more explicit identification of the people cannot be avoided even, or indeed especially, in hard cases. The constitutional referendum by definition implicates an anterior act of democratic border drawing – the framing of the self who will perform an act of constitutional self-determination and who will in doing so expressly articulate itself as a constitutional people. In one sense, any constitutional referendum as an act of popular self-determination requires a direct attribution of constitutional authorship. In this respect, all referendums are potentially constitutive, since they bring back the people – the ultimate source of democratic legitimacy for the polity.

  There are two dimensions to this boundary-drawing: the first conceptualises the polity as a physical space; the second imagines it as a group of people. In this light, the constitutional referendum sets the boundary of the people by way of both territorial demarcation and franchise rules. The franchise issue can raise complex problems even when the territorial question is settled. There may be broad consensus that a referendum, if it is to be held, should take place within a particular territorial space, but the issue of who is entitled to vote can raise its own problems, as I have already briefly touched upon in relation to the Scottish referendum. Yet it is the territorial issue that is more relevant to the very fact of constitutive referendums in Quebec, Scotland, and possibly Catalonia. Of course the referendum has been important in other periods of territorial change as well – the post-WWI plebiscites; post WWII decolonisation; the collapse of the USSR and SFRY. It is about the assertion of a separate demos within the state and the legitimacy of that demos to determine its own future. It has a symbolic role in bringing together and hence declaring the voters to be – directly, equally and communally – a determining ‘self’. What is now at the heart of the debates in the UK and Spain is the moral legitimacy of a sub-state people presenting itself as a self-determining entity. Why was the referendum only held in Scotland and not throughout the UK? Why is the Scottish people treated as a people separate for the purposes of constituent power from the people of the whole UK?

  The territorial conceptualisation sees the people in terms of space, the participants are those who can vote in a particular geographical area. The territorial boundaries of the demos are often taken for granted, as they are of course in a representative democracy, particularly when the referendum is held throughout the state. But this need not be the case and when there is disagreement, the referendum has the potential to escalate already existing territorial disagreements. My focal point here is indeed sub-state nationalists who argue that the territory of the sub-state nation contains within it a self-determining people. We see in this challenge how the demos question lies unconfronted in a world where we take for granted that the principles of democracy and formal equality of citizens apply automatically in each polity in the same way, and that each state contains one demos. The referendum is increasingly used by modern nationalism to challenge this orthodoxy, asserting the existence, or at least the possibility of multiple demoi within the state.⁴⁴xNormative challenges to the injustice of particular stately instantiations of the people have in recent times been offered by political theorists. See Will Kymlicka, Multicultural Citizenship (Oxford: Oxford University Press, 1995); Margaret Moore, Ethics of Nationalism (Oxford: Oxford University Press 2001); Ferran Requejo, ‘Liberal Democracy’s Timber is Still Too Straight: The Case of Political Models for Coexistence in Composite States,’ in Europe’s Constitutional Mosaic, ed. Neil Walker, Jo Shaw and Stephen Tierney (Oxford: Hart Publishing, 2011).

  Interestingly, international law does not come to the aid of Scottish or Catalan nationalists here. The right to secession is very tightly circumscribed and almost certainly does not include these territories. As a result, the issue is situated within constitutional law. The right of a province to negotiate secession from Canada was read into the Canadian constitution by the Supreme Court in the Secession Reference:
  

  [W]e are… unable to accept… that a clear expression of self-determination by the people of Quebec would impose no obligations upon the other provinces or the federal government. The continued existence and operation of the Canadian constitutional order cannot remain indifferent to the clear expression of a clear majority of Quebecers that they no longer wish to remain in Canada. This would amount to the assertion that other constitutionally recognized principles necessarily trump the clearly expressed democratic will of the people of Quebec. Such a proposition fails to give sufficient weight to the underlying constitutional principles that must inform the amendment process, including the principles of democracy and federalism. The rights of other provinces and the federal government cannot deny the right of the government of Quebec to pursue secession, should a clear majority of the people of Quebec choose that goal, so long as in doing so, Quebec respects the rights of others. Negotiations would be necessary to address the interests of the federal government, of Quebec and the other provinces, and other participants, as well as the rights of all Canadians both within and outside Quebec. (para 92)

  
  The UK has arguably gone further in that the central government itself has accepted Scotland’s right to leave in the Edinburgh agreement without the need for any judicial intervention; and indeed facilitated the process. A similar right exists for Northern Ireland in relation to reunification with the Republic, something which is often overlooked. Within this final agreement lies, in effect, a recognition that the UK is a plurinational union state. The idea of a union state is one which accepts that the state was forged from a union of nations, each of which continued to consolidate and build its national identity within the state. Processes of homogenisation did not occur. In other words its national pluralism is part of the very identity of the state and of its on-going lived experience. The principles of democracy and rights demand that this association continue to be voluntary.

  This may be a difficult concept for some wedded to Westphalian views of the state as receptacles of one unitary people and one unitary source of sovereignty in the tradition of Bodin and Hobbes. But it should be less hard to appreciate in light of nearly 60 years of European Union within which national pluralism is lived reality, and in which the treaties now expressly enshrine a right of exit; a path down which the people of the United Kingdom have chosen to go after another referendum on 23 June 2016.

  We see that national pluralism was an implicit understanding in the minds of the Scottish citizens as well. People in Scotland were mobilised in huge numbers to debate the issue and to vote. There was no boycott on the basis of illegitimacy or even any serious question about the legitimacy of the referendum; it was accepted that a nationalist party had been elected on a manifesto promise to hold a referendum. People may have disagreed about independence and questioned the prudence of a referendum on the issue, but there was and remains wide consensus that Scots have the right to decide.

  Notably this is not a concession which Spain is prepared to make. Here the constitution itself or at least how it is widely interpreted, does not recognise this logic of national pluralism. The constitution does make a category distinction between its national societies and other regions. Spain is thus a mixture of a plurinational and a regional state. Catalonia is recognized as a nationality (‘nacionalidad’) and it is constituted as a self-governing unit of Spain (‘comunidad autónoma’). But Article 1.2 of the Spanish constitution provides: ‘National sovereignty belongs to the Spanish people, from whom all State powers emanate.’ And Article 2 establishes: ‘The Constitution is based on the indissoluble unity of the Spanish Nation, the common and indivisible homeland of all Spaniards; it recognises and guarantees the right to self-government of the nationalities and regions of which it is composed and the solidarity among them all.’ Secession is not ruled out, but the central state asserts that this would require a constitutional amendment, which is not out of line with the Canadian or even British approaches to negotiated secession. Crucially however it is maintained by the central state that a referendum of secession would also need an amendment. The Spanish Constitutional Court, in its Judgment 103/2008 of 11 September 2008, the leading case on referendums and secession, decided indeed that this would require an amendment to the Constitution at the beginning of the process. Also this would have to be done through ‘constitutional amendment provided in Article 168, the hardest amending process of the Spanish Constitution.’⁴⁵xThe amending procedure of Article 168 of the Spanish Constitution requires to follow these successive steps: (1) a two-thirds majority of the members of the Spanish Congress and the Spanish Senate would be required, (2) both of these houses would have to be dissolved, (3) elections would need calling to constitute a new Congress and a new Senate, (4) the decision passed by the previous houses would require ratification by a two-thirds majority of the members of each house and, finally, (5) the amendment would need to be ratified by a referendum submitted to the Spanish people. Both Canada and the UK accept that their sub-state territories are potentially self-determining and that they cannot be kept in the state against their will, which has not been conceded in Spain.

• 6 Conclusion

  National pluralism is a reality in Europe today. In acknowledging it, we must adjust our constitutional understandings to accept that sovereignty can not only be pooled but that it can also be divided. A sharing of power among equal sovereign entities is the reality of the European Union, but it is also the reality of states within it. This reality brings with it a constitutional right of self-determination. To this end, the referendum has intervened in a crucial way and the Scottish process is likely only to enhance its legitimacy as a feature of constitutional debate in plurinational states. The Scottish Referendum Act, building upon the Edinburgh Agreement principles, helped set the conditions for a fair, lawful and democratic referendum.⁴⁶xA significant task for the Electoral Commission once the regulated period began, was to monitor how well the legislation in the Scottish Referendum Act and Scottish Franchise Act was implemented and how responsibly all of those engaged in referendum campaigning behave. So far the evidence is that it has approached its regulatory role in a vigorous way. As we cast the Scottish process into wider perspective, it is indeed notable that the leading strategist from the Yes campaign in the Quebec referendum of 1995, a referendum which suffered from a rancorous relationship between the two campaigns and the absence of agreed and independently overseen process rules, has written recently commending the UK for the way in which the Edinburgh Agreement fostered a mutually acceptable referendum process. He sees this breakthrough to be of great significance to other countries facing similar referendum processes: ‘Nations that have been through this wrenching debate recently or who, especially in Catalonia, will navigate these waters soon, need the British government to keep offering a template of fair play and respect for democracy.’⁴⁷xJean-François Lisée, ‘Well done, Britain, for a fair referendum – it’s a shame Canada didn’t manage it,’ The Guardian, September 9, 2014, http://www.theguardian.com/commentisfree/2014/sep/09/britain-referendum-canada-scottish-independence-vote. The Scottish referendum is indeed an opportunity to provide a model of citizen engagement at a time when the referendum is proliferating around the world like never before.⁴⁸xLawrence LeDuc, The Politics of Direct Democracy: Referendums in Global Perspective (Peterborough, Ontario: Broadview Press, 2003), 29. In the end the quality of Scottish, and indeed British, democracy was boosted by the way the decision was reached.

  But the British process was important for a second reason. It dealt with the demos question by accepting that the Scottish citizens had the right to decide. Something which, ironically, prominent actors in the European Union were less inclined to concede. Admittedly, the Catalan experience is very different because there is a real impasse in the prior question of the constitutional right to decide.

  Finally, the Scottish experience may be illuminating as well, because in the end the people voted to stay. Part of me wonders how many people, even subconsciously, made the decision that the kind of state which allows you the right to express sub-state constituent power, which allows sub-state institutions to organise the referendum, set the timing, franchise and question, is not the kind of state you want to leave. This may in fact be the ultimate lesson for a democratic Europe. Peoples can be governed only with their consent. Otherwise no polity, old or new, can claim for itself the ultimate source of legitimacy upon which the democratic state depends.

Noten

• * This text was presented in the spring of 2015 as a part of the lecture series Sovereignty Yesterday, Today, and Tomorrow? organised by the Law Faculty and the Institute of Philosophy at the University of Leuven. It draws upon my published work.

• 1 This number significantly trumped the 65.1% who voted in the 2010 UK general election and the 50.6% who bothered to turn out for the 2011 Scottish parliamentary elections.

• 2 David McCrone, Understanding Scotland: The sociology of a nation, 2nd ed. (London: Routledge, 2001).

• 3 Stéphane Dion, ‘Le nationalisme dans la convergence culturelle: le Québec contemporain et le paradoxe de Tocqueville,’ in L’engagement intellectuel: Mélange en l’honneur de Léon Dion, ed. Raymond Hudon and Réjean Pelletier (Québec: Presses de l’Université Laval, 1991).

• 4 Michael Keating, Nations Against the State - The New Politics of Nationalism in Quebec, Catalonia and Scotland, 2nd ed. (UK, Palgrave, 2001); Alain-G. Gagnon and James Tully, ed., Multinational Democracies (Cambridge: Cambridge University Press, 2001).

• 5 Michael Keating, Plurinational Democracy: stateless nations in a post-sovereignty era (Oxford: Oxford University Press, 2001).

• 6 Will Kymlicka, Multicultural Citizenship: A Liberal Theory of Minority Rights (Oxford: Oxford University Press, 1995); Margaret Moore, ‘Normative Justifications for Liberal Nationalism: justice, democracy and national identity,’ Nations and Nationalism 7 (2001): 1-20; Ferran Requejo, ed., Democracy and National Pluralism (London: Routledge, 2001).

• 7 Will Kymlicka, Politics in the Vernacular (Oxford: Oxford University Press, 2001); Stephen Tierney, ‘The Search for a New Normativity: Thomas Franck, Post-Modern Neo-Tribalism and the Law of Self-Determination,’ European Journal of International Law (2002): 941-60.

• 8 Stephen Tierney, Constitutional Referendums: The Theory and Practice of Republican Deliberation (Oxford: Oxford University Press, 2012).

• 9 John Rawls, A Theory of Justice (Cambridge, Mass.: Harvard University Press 1971).

• 10 John Rawls, ‘The Idea of Public Reason Revisited,’ University of Chicago Law Review 64 (1997): 765.

• 11 ‘Your Scotland - Your Referendum - A Consultation Document’, Scottish Government white paper, January 25, 2012, http://www.scotland.gov.uk/Publications/2012/01/1006.

• 12 Agreement between the United Kingdom Government and the Scottish Government on a referendum on independence for Scotland, October 15, 2012, http://www.scotland.gov.uk/About/Government/concordats/Referendum-on-independence.

• 13 Stephen Tierney, Constitutional Law and National Pluralism (Oxford: Oxford University Press, 2004), 320-2.

• 14 Keating, Nations Against the State; Gagnon and Tully, Multinational Democracies.

• 15 This provided that the referendum should have a clear legal base, to be legislated for by the Scottish Parliament and to be conducted so as to command the confidence of parliaments, governments and people. This was formalised by an Order in Council (per Scotland Act 1998 s30) which devolved to the Scottish Parliament the competence to legislate for a referendum on independence which had to be held before the end of 2014. Scotland Act 1998 (Modification of Schedule 5) Order 2013, para 3, http://www.legislation.gov.uk/uksi/2013/242/made.

• 16 The Edinburgh Agreement (para 2) provided: ‘Both governments agree that the principles underpinning the existing framework for referendums held under Acts of the UK Parliament – which aim to guarantee fairness – should apply to the Scottish independence referendum. Part 7 of the Political Parties, Elections and Referendums Act 2000 (PPERA) provides a framework for referendums delivered through Acts of Parliament, including rules about campaign finance, referendum regulation, oversight and conduct.’

• 17 Scottish Independence Referendum (Franchise) Act 2013, s. 2.

• 18 Ibid., s. 2(1)(a).

• 19 Ibid., s. 3.

• 20 Hirst v. the United Kingdom (No 2) [2005] ECHR 681.

• 21 Stephen Tierney, ‘Possible vires issue in relation to section 3 of the Scottish Independence Referendum (Franchise) Bill, Advice to the Scottish Parliament, Scottish Referendum Bill Committee paper, March 21, 2013, http://www.scottish.parliament.uk/S4_ReferendumScotlandBillCommittee/20130321_Letter_to_DFM_on_prisoners_voting_rights.pdf.

• 22 Moohan and others, Petitioners [2013] CSOH 199.

• 23 Reclaiming Motion: Moohan and others v. Lord Advocate [2014] CSIH 56.

• 24 Moohan and Another v. Lord Advocate UKSC 2014/0183 (24 July 2014).

• 25 See Ailsa Henderson et al. ‘Risk and Attitudes to Constitutional Change,’ ESRC Scottish Centre on Constitutional Change Risk and Constitutional Attitudes Survey, August 16, 2014, http://www.futureukandscotland.ac.uk/sites/default/files/news/Risk%20and%20Constitutional%20Attitudes%20Full%20Survey%2014%20Aug.pdf and AQMen project https://www.aqmen.ac.uk/project/socialmedia.

• 26 The Electoral Commission took the view that ‘based on our research and taking into account what we heard from people and organisations who submitted their views on the question, we consider that the proposed question is not neutral because the phrase “Do you agree …?” could lead people towards voting “yes”.’ They therefore recommended the following question: ‘Should Scotland be an independent country? Yes/No.’ ‘Referendum on independence for Scotland: Advice of the Electoral Commission on the proposed referendum question,’ The Electoral Commission, 2013, http://www.electoralcommission.org.uk/__data/assets/pdf_file/0007/153691/Referendum-on-independence-for-Scotland-our-advice-on-referendum-question.pdf.

• 27 Scottish Independence Referendum Act 2013, s. 1(2). See also ‘Scottish independence: SNP accepts call to change referendum question,’ BBC News, January 30, 2013, http://www.bbc.co.uk/news/uk-scotland-scotland-politics-21245701.

• 28 ‘Referendum on independence for Scotland: Advice of the Electoral Commission on the proposed referendum question,’ The Electoral Commission, January 30, 2013, http://www.electoralcommission.org.uk/__data/assets/pdf_file/0007/153691/Referendum-on-independence-for-Scotland-our-advice-on-referendum-question.pdf, para 5.41. See also generally paras 5.41-5.44, e.g., ‘We recommend that both Governments should agree a joint position, if possible, so that voters have access to agreed information about what would follow the referendum. The alternative - two different explanations - could cause confusion for voters rather than make things clearer.’ (para 5.43).

• 29 One of the main criticisms of the Quebec referendum in 1995 was that the proposal of sovereignty and partnership was not well understood by citizens. Tierney, Constitutional Law and National Pluralism, 293-9.

• 30 ‘Scotland’s Future,’ Scottish Government white paper, November 26, 2013, http://www.scotland.gov.uk/Publications/2013/11/9348.

• 31 ‘Scottish Independence Bill: A consultation on an interim constitution for Scotland,’ Scottish Government white paper, June 2013, http://www.scotland.gov.uk/Publications/2014/06/8135/downloads.

• 32 Scottish Independence Referendum Act 2013, sched. 4, part 3.

• 33 Ibid., sched. 4, para 25.

• 34 We have noted ‘Scotland’s Future’, Scottish Government white paper November 26, 2013 op. cit. In June 2014 the UK government sent a 16-page booklet called ‘What Staying in the UK Means for Scotland’ to every household in Scotland, setting out the case for the No side. See ‘UK “fact booklet” to be sent to Scots households’, The Scotsman, June 12, 2014.

• 35 See for example on the franchise and mechanics of voting issued by the Citizens Advice Bureau, http://www.adviceguide.org.uk/scotland/law_s/law_civil_rights_s/law_government_and_voting_s/the_scottish_independence_referendum_s.htm.

• 36 See for example the major ESRC investment: ‘Future of the UK and Scotland,’ http://www.esrc.ac.uk/research/major-investments/future-of-uk-and-scotland/index.aspx.

• 37 Scottish Independence Referendum Act, s. 21.

• 38 Stephen Tierney, ‘Why is Scottish Independence Unclear?,’ U.K. Const. L. Blog, February 25, 2014, http://ukconstitutionallaw.org/.

• 39 William Norton evidence before the Referendum (Scotland) Bill Committee, May 9, 2013, Official Report, http://www.scottish.parliament.uk/parliamentarybusiness/28862.aspx?r=8140&i=73882&c=1482666.

• 40 John McCormick, the Electoral Commissioner for Scotland, evidence to: Scottish Parliament Referendum (Scotland) Bill Committee, May 23, 2013, Official Report, col 431, http://www.scottish.parliament.uk/parliamentarybusiness/28862.aspx?r=8326&mode=pdf.

• 41 https://www.electoralcommission.org.uk/__data/assets/pdf_file/0012/170400/The-2014-Scottish-Independence-Referendum-Voting-Guide.pdf.

• 42 For comment on this by Deputy First Minister Nicola Sturgeon, see Scottish Parliament Referendum (Scotland) Bill Committee, June 13, 2013, Official Report, cols 554 and 560.

• 43 For Dahl, the great philosophers have ignored this question because ‘they take for granted that a people has already constituted itself’. See Robert A. Dahl, After the Revolution? Authority in a Good Society (New Haven: Yale University Press, 1970), 60-1. Others consider its neglect to be more surprising. Canovan has observed: ‘Unlike “freedom”, “justice”, or even “nation”, “people” has attracted hardly any analysis, even by theorists of democracy’, a fact she deems ‘astonishing’. See Margaret Canovan, ‘Populism for Political Theorists?’ Journal of Political Ideologies (2004): 241, 247. And Goodin comments: ‘the first question for any democratic theory is who is included’, but notes, ‘[t]his is a strangely under-explored issue within democratic theory’. See Robert Goodin, Innovating Democracy. Democratic Theory and Practice after the Deliberative Turn (Oxford: Oxford University Press, 2008), 3.

• 44 Normative challenges to the injustice of particular stately instantiations of the people have in recent times been offered by political theorists. See Will Kymlicka, Multicultural Citizenship (Oxford: Oxford University Press, 1995); Margaret Moore, Ethics of Nationalism (Oxford: Oxford University Press 2001); Ferran Requejo, ‘Liberal Democracy’s Timber is Still Too Straight: The Case of Political Models for Coexistence in Composite States,’ in Europe’s Constitutional Mosaic, ed. Neil Walker, Jo Shaw and Stephen Tierney (Oxford: Hart Publishing, 2011).

• 45 The amending procedure of Article 168 of the Spanish Constitution requires to follow these successive steps: (1) a two-thirds majority of the members of the Spanish Congress and the Spanish Senate would be required, (2) both of these houses would have to be dissolved, (3) elections would need calling to constitute a new Congress and a new Senate, (4) the decision passed by the previous houses would require ratification by a two-thirds majority of the members of each house and, finally, (5) the amendment would need to be ratified by a referendum submitted to the Spanish people.

• 46 A significant task for the Electoral Commission once the regulated period began, was to monitor how well the legislation in the Scottish Referendum Act and Scottish Franchise Act was implemented and how responsibly all of those engaged in referendum campaigning behave. So far the evidence is that it has approached its regulatory role in a vigorous way.

• 47 Jean-François Lisée, ‘Well done, Britain, for a fair referendum – it’s a shame Canada didn’t manage it,’ The Guardian, September 9, 2014, http://www.theguardian.com/commentisfree/2014/sep/09/britain-referendum-canada-scottish-independence-vote.

• 48 Lawrence LeDuc, The Politics of Direct Democracy: Referendums in Global Perspective (Peterborough, Ontario: Broadview Press, 2003), 29.