OUTER HOUSE, COURT OF SESSION
	[2013] CSOH 199
P643/13 P901/13 P903/13	OPINION OF LORD GLENNIE in Petition of LESLIE MOOHAN, GARY GIBSON and ANDREW URQUHART BLACK GILLON Petitioners; for judicial review of sections 2(1)(a)(i), 2(2)(a) and 3 of the Scottish Independence Referendum (Franchise) Act 2013 ________________

Petitioners: O'Neill QC, Pirie; Balfour + Manson LLP

Respondent: Moynihan QC, Ross; Scottish Government

19 December 2013

The independence referendum

[1] On 18 September 2014 there will be a referendum in Scotland ("the independence referendum") to answer the question: "Should Scotland be an independent country?" This is laid down in section 1 of the Scottish Independence Referendum Bill which passed the Scottish Parliament in November 2013. It has not yet received the royal assent, but there is no doubt that it will in due course become law. It has the support of the United Kingdom government.

[2] This opinion, covering the arguments raised in three petitions, is concerned with the right to vote in that independence referendum. In particular, it concerns the right of convicted prisoners to vote. This is presently excluded by a blanket ban on prisoners voting. This was enacted by the Scottish Parliament. The question before the court is as to the legality of that blanket ban.

The franchise for the independence referendum

[3] Even before the passing of the Referendum Bill, the franchise for the independence referendum was the subject of legislation in the Scottish Parliament. The Scottish Independence Referendum (Franchise) Act 2013 ("the Franchise Act") makes provision for those who are and are not entitled to vote. The petitions in this case concern the "blanket ban" on convicted prisoners voting in the referendum if on that date they are detained in a penal institution in pursuance of the sentence imposed on them. The ban does not apply to prisoners on remand awaiting trial, nor to persons imprisoned for contempt of court; nor, obviously, to persons who have completed their sentence of imprisonment. Accordingly, when I refer in this opinion to "prisoners", I mean convicted prisoners currently serving sentences of imprisonment; and when I refer to the blanket ban on prisoners voting, I use that as shorthand for the ban which excludes from voting any convicted prisoner who, at the time of the vote, is serving his or her sentence of imprisonment.

[4] The relevant sections of the Franchise Act are sections 2 and 3, which provide so far as material as follows:

"2. Those entitled to vote in an independence referendum

(1) A person is entitled to vote in an independence referendum if, on the date on which the poll that the referendum is held, the person is -

(a) aged 16 or over,

(b) registered in either -

(i) the register of local government electors maintained under section 9(1)(b) of the [Representation of the People Act 1983] for any area in Scotland, or

(ii) the register of young voters maintained under section 4 of this Act for any such area,

(c) not subject to any legal incapacity to vote (age apart), and

(d) a Commonwealth citizen, a citizen of the Republic of Ireland or a relevant citizen of the European Union.

(2) For the purposes of this Act, a person is, on any date, subject to a legal incapacity to vote if the person -

(a) would be legally incapable (whether by virtue of any enactment or any rule of law) of voting at a local government election in Scotland held on that date, or

(b) is legally incapable, by virtue of section 3, of voting in an independence referendum held on that date.

3. Offenders in prison etc. not to be entitled to vote

(1) A convicted person is legally incapable of voting in an independence referendum for the period during which the person is detained in a penal institution in pursuance of the sentence imposed on the person."

The blanket ban on prisoners voting is achieved by the Franchise Act in three separate ways. First, since the Representation of the People Act 1983 (RPA 1983) prohibits serving prisoners from being registered on the register of local government electors, section 2(1)(b)(i) is not satisfied. Second, in terms of the RPA 1983, a serving prisoner is legally incapable of voting at a local government election in Scotland and is therefore subject to a legal incapacity to vote in the referendum by virtue of section 2(2)(a). Third, a serving prisoner is expressly stated by section 3(1) to be legally incapable of voting in an independence referendum and is therefore subject to a legal incapacity to vote in the referendum in terms of section 2(2)(b). It will be apparent that in terms of those sections no convicted prisoner may vote while serving a sentence of imprisonment, no matter how long or short is his or her sentence and no matter how serious is the offence for which he or she was sentenced. It is for this reason that it is colloquially but accurately referred to as a "blanket ban".

The petitioners

[5] I heard three petitions for judicial review at the instance of three individual petitioners. Each of the petitioners is a prisoner in HM Prison Addiewell in West Lothian. The petitioner in petition P643/13 is Leslie Moohan, currently serving a sentence of life imprisonment with a punishment part in terms of section 2 of the Prisoners and Criminal Proceedings Act 1993 of 15 years, backdated to February 2008. He is not eligible for release until, at the earliest, February 2023. Petition P901/13 concerns Gary Gibson, currently serving a sentence of imprisonment of seven years and four months with effect from July 2012. The third petition P903/13 concerns Andrew Gillon, who was sentenced to life imprisonment in May 1998. In terms of Schedule 1 to the Convention Rights (Compliance) (Scotland) Act 2001, the punishment part of his sentence has been fixed at 12 years, backdated to 13 January 1998. That punishment part expired in January 2010. Since that time he has been detained on the ground of the risk that he poses, as assessed by the Parole Board for Scotland, rather than punishment for the offence. The Parole Board for Scotland will not consider his case again until 22 September 2014, after the date of the referendum. He will not be released until then at the earliest.

[6] Accordingly, none of the petitioners is due to be released until after the date fixed for the referendum. They will all still be serving prisoners at the time of the referendum, and in terms of the sections 2 and 3 of the Franchise Act will be ineligible to vote. They each wish to vote in the referendum. By these petitions they each contend that the blanket ban on prisoners voting in the independence referendum is unlawful on ECHR and other grounds. Nothing turns for present purposes on the different offences of which they have been found guilty or the different periods of imprisonment to which they have been sentenced. As is made clear in the decided cases (see Hirst v United Kingdom (No.2) (2006) 42 EHRR 41 at para 72), and as was accepted in argument before me, a prisoner is entitled to challenge the statutory removal of or restriction on his right to vote, on ECHR grounds at least, notwithstanding that if the legislation were redrafted in an ECHR compliant manner he would not necessarily be entitled to vote. So this case is not about whether prisoners serving lengthy periods of imprisonment for particularly serious crimes should be allowed to vote. It is about whether a blanket ban on all prisoners voting is lawful.

Grounds of challenge

[7] The challenge to the blanket ban on voting is made on a number of different grounds. First, there are grounds based upon article 1 of the First Protocol ("A3P1") and article 10 of the European Convention on Human Rights and Fundamental Freedoms ("ECHR"). A challenge was originally advanced also on the basis of article 14, but in light of the decision of the Supreme Court in McGeogh, to which I refer below, this was not insisted upon. Second, there are grounds based upon what are said to be "fundamental" or "constitutional" rights enshrined in the common law, namely: the rule of law; the right to vote; and respect for international obligations. Third, it is said that, in circumstances where the result of the referendum would or might lead to the loss of European Union rights, the blanket ban is contrary to EU law.

[8] I propose to consider these various grounds of challenge in that order. In doing so, I have not overlooked the point made by Lord Reed in Osborn v The Parole Board [2013] UKSC 61 at paragraphs 54- 63 that it is an error to maintain a clear separation between common law and Convention rights and to suppose that because an issue falls within the ambit of a Convention guarantee, the legal analysis should begin and end with the Convention and the Strasbourg case law on it. Nonetheless, it is convenient to discuss the different issues raised by the various challenges separately.

ECHR challenges

The Human Rights Act 1998

[9] The Human Rights Act 1998 was passed by the United Kingdom Parliament to give effect in the United Kingdom to the "Convention rights" set out in Articles 2 to 12 and 14 of the European Convention on Human Rights and Fundamental Freedoms ("ECHR"), Articles 1 to 3 of the First Protocol thereto and also, by amendment, article 1 of the Thirteenth Protocol: see section 1. Section 6 of the Act makes it "unlawful for a public authority to act in a way which is incompatible with a Convention right", unless it is required so to act by "primary legislation" (mainly but not exclusively Acts of the UK Parliament). In relation to any act of a public authority which the court finds is unlawful, the court is given the power to grant such relief or remedy or make such order within its powers as it considers just and appropriate: see section 8. In addition, the court has been given power to determine whether a provision of primary legislation is compatible with a Convention right: see section 4. Accordingly, when judges give effect to Convention rights, they do so in accordance with the wishes and at the direction of the United Kingdom Parliament. But they cannot strike down primary legislation of that Parliament. Their powers are limited in that respect. If satisfied that a provision of primary legislation is incompatible with a Convention right, the most that the court can do is that it "may make a declaration of that incompatibility": see section 4(2). The word "may" makes it clear that that is remedy is discretionary.

The Scotland Act 1998

[10] The Scotland Act 1998 provides that it is beyond the legislative competence of the Scottish Parliament to legislate in a manner which is incompatible with Convention rights. Convention rights are as defined under the Human Rights Act: section 29(2)(d). An Act of the Scottish Parliament "is not law" in so far as any provision thereof is outside the legislative competence of the Scottish Parliament: section 29(1). The power of the court under the Scotland Act goes beyond making a declarator of incompatibility - it may, indeed must, strike down a provision of an Act passed by the Scottish Parliament if it is outside its legislative competence and therefore "not law".

The ECHR challenge - outline

[11] The ECHR challenge invokes A3P1 and article 10. So far as voting is concerned, only A3P1 has so far been considered in detail in the domestic courts. I shall therefore deal with that first, before moving on to deal with article 10. First, however, I should consider the approach which a domestic UK court is required to adopt in dealing with a challenge on ECHR grounds.

The correct approach to an ECHR challenge

[12] It has been established in a long and consistent line of cases decided by the Supreme Court that in construing the scope of the Convention rights brought into effect as part of domestic law by the Human Rights Act and the Scotland Act the court is required to follow the approach recently articulated by Lord Neuberger giving the judgment of the court in Manchester City Council v Pinnock [2011] 2 AC 104 at para 48:

"Where, however, there is a clear and consistent line of decisions whose effect is not inconsistent with some fundamental substantive or procedural aspect of our law, and whose reasoning does not appear to overlook or misunderstand some argument or point of principle, we consider that it would be wrong for this court not to follow that line."

The courts are required, in the words of Lord Bingham in R (Ullah) v Special Adjudicator [2004] 2 AC 323 at para [20], "to keep pace with the Strasbourg jurisprudence as it evolves over time: no more, but certainly no less" or, as Lord Brown put it in R (Al-Skeini) v Secretary of State for Defence [2008] 1 AC 153 at para [106], "no less, but certainly no more". Lord Brown's inversion of the aphorism is important. As he explains in that paragraph, if the domestic court misinterprets the Convention and interprets it more generously than the Strasbourg court has done or would have done, the mistake will stand in that member state (though not in others), since the member state cannot itself go to Strasbourg to have it corrected. By contrast, however, if the domestic courts have interpreted the Convention too narrowly, the aggrieved individual can seek to have that decision corrected in Strasbourg. Parliament did not intend by enacting the Human Rights Act and the Scotland Act to give the courts of the United Kingdom the power to give a more generous scope to the Convention rights than that which was to be found in the jurisprudence of the Strasbourg court, since that would have the effect of changing those rights from Convention rights into freestanding rights of the court's own creation: per Lord Hope in Smith v Ministry of Defence [2013] 3 WLR 69 at para [43]. Hence the need for caution.

[13] The need for caution has been emphasised in other authorities, including Ambrose v Harris 2012 UKSC 53 per Lord Hope at paragraphs [18]-[20] and per Lord Dyson at para [105]. However, two points should be noted. First, the Supreme Court has made it clear that it is not bound to follow every decision of the Strasbourg court. Section 2 of the Human Rights Act requires the domestic courts to "take into account" decisions of the Strasbourg court, not necessary to follow them. But it will only be in the rare case that the domestic courts will feel free to depart from the Strasbourg jurisprudence. Nor, secondly, does the need for caution mean that the court is necessarily tied to an "Ullah-type reticence" (per Lord Kerr in Ambrose at paragraphs [126] and [130]) whenever there is an absence of a clear and constant line of jurisprudence from Strasbourg. There are cases, of which Rabone v Pennine Care NHS Trust [2012] 2 AC 72 is an example, where the domestic court has arguably construed Convention rights more generously than the existing Strasbourg jurisprudence, "going rather further than the evolving jurisprudence of the European Court of Human Rights ("Strasbourg") has yet clearly established to be required": per Lord Brown at para [111]. In that case the court believed its decision "to flow naturally from existing Strasbourg case law (albeit that it could be regarded as carrying the case law a step further)": per Lord Brown at para [112]. Another such case is In re G (Adoption: Unmarried Couple) [2009] 1 AC 173 where the House of Lords reached its decision on the basis that "it seems now to be not at all unlikely that if the issue in this case were to go to Strasbourg the court would hold that the discrimination of which the applicants complain violates article 14": see per Lord Hope at para 53, echoing Lord Hoffman at para 27. However, such cases are likely to be rare. For the present, the approach in Ullah and Al Skeini holds good. In British Broadcasting Corporation v Sugar (No 2) [2012] 1 WLR 439, Lord Wilson, dissenting on the main issue, said that he would welcome an appeal in which it was appropriate for the Supreme Court "to consider whether, of course without acting extravagantly, it might now usefully do more than to shadow the European Court of Human Rights in the manner hitherto suggested ...": see para 59. Lord Mance agreed with this: see para 113. It may be, therefore, that Ullah and Al Skeini may one day not represent the last word on the subject. But they do now.

[14] It was submitted on behalf of the petitioners that because Convention rights have become domestic statutory rights by reason of the Human Rights Act and the Scotland Act, their meaning was a matter for the domestic courts rather than the Strasbourg court. That submission was based primarily on the speech of Lord Hoffmann in In re McKerr [2004] 1 WLR 807 at paras 63-64. Lord Hoffman was addressing an argument which appeared to confuse the United Kingdom's adherence to the Convention with its incorporation of certain Convention rights into domestic law. He pointed out that, once incorporated into domestic law, the source of the rights is the statute, not the Convention. He went on to say (in para 63):

"They are available against specific public authorities, not the United Kingdom as a state. And their meaning and application is a matter for domestic courts, not for the court in Strasbourg".

In para 64, however, he expressly recognised that the obligation on the domestic court was to "take into account" any judgment of the Strasbourg court, though he emphasised, as have many before and since, that in adjudicating in litigation in the United Kingdom about a domestic "Convention right", the domestic court is not bound by a decision of the Strasbourg court; it only has to take it into account. That is quite consistent with the evolving case law both then and now. But the courts have gone on to emphasise that in taking into account the Strasbourg case law, it will normally follow clear and constant jurisprudence emerging from the Strasbourg court and will not innovate by a generous interpretation of the Convention in the absence of some clear understanding of that being the direction in which the Strasbourg jurisprudence will inevitably follow. Accordingly, although it is right to say that the meaning of the Convention rights in the Human Rights Act and in the Scotland Act is a matter for the domestic courts rather than the Strasbourg court, that is subject to the qualification to which I have already referred.

A3P1

[15] A3P1 is entitled "Right to free elections". It provides as follows:

"The High Contracting Parties undertake to hold free elections at reasonable intervals by secret ballot, under conditions which will ensure the free expression of the opinion of the people in the choice of the legislature."

The petitioners argue that A3P1 guarantees an individual his right to vote not only in Parliamentary elections but also in a referendum, such as that which is due to take place in September 2014, which has a fundamental effect on the future governance of Scotland.

[16] In making this submission the petitioners emphasise the importance of the referendum for the future of Scotland; the constitutional imperative for there to be a referendum on a subject as important as this; and the fact that agreement has been reached between the Scottish and United Kingdom Governments guaranteeing, in effect, that the result of the referendum will be binding, in practice if not in law. None of that was disputed. Accordingly it is not necessary to go through and set out the materials which were put before the court on this matter.

Prisoners' voting rights - the story so far

[17] This is not the first occasion on which the issue of prisoners' voting rights has come before the courts. The Grand Chamber of the European Court of Human Rights has held on two occasions that a blanket ban on prisoners voting in parliamentary elections is contrary to A3P1 and therefore unlawful. In Hirst v United Kingdom (No.2), a case concerning the right of a prisoner to vote in UK parliamentary and local elections, the Grand Chamber explained that A3P1 guaranteed individual rights, including the right to vote and to stand for election: see para 57. It underlined the importance attached to democratic principles which underlay the interpretation and application of the Convention; and it emphasised that the rights guaranteed under A3P1 were "crucial to establishing and maintaining the foundations of an effective and meaningful democracy governed by the rule of law": see para 58. The right to vote is not a privilege. Universal suffrage has become the basic principle, but there is room for implied limitations on it. Contracting States had to be given a wide margin of appreciation in this sphere, there being "numerous ways of organising and running electoral systems and a wealth of differences, inter alia, of historical development, cultural diversity and political thought within Europe which it is for each Contracting State to mould into its own democratic vision". However, exclusion of any group or category of persons from the right to vote had to be imposed proportionately and in pursuit of a legitimate aim: see paragraphs 59-62. While the United Kingdom's stated aims in enacting the blanket ban in section 3 of the RPA 1983, namely the aim of preventing crime, conferring additional punishment, and enhancing civic responsibility and respect for the rule of law were all potentially legitimate aims to be taken into account, those aims must be pursued by proportionate means and must not be such as to impair the essence of the right. A blanket ban on convicted prisoners voting in an election was a disproportionate and arbitrary method of pursuing those aims: see paragraphs 74-82.

[18] After noting that the RPA 1983 had been amended in 2000 to allow remand prisoners to vote while continuing the blanket ban on convicted prisoners, the court in Hirst said this (at para 82):

"[Section 3 of the 1983 Act] remains a blunt instrument. It strips of their Convention right to vote a significant category of persons and it does so in a way which is indiscriminate. The provision imposes a blanket restriction on all convicted prisoners in prison. It applies automatically to such prisoners, irrespective of the length of their sentence and irrespective of the nature or gravity of their offence and their individual circumstances. Such a general, automatic and indiscriminate restriction on a vitally important Convention right must be seen as falling outside any acceptable margin of appreciation, however wide that margin might be, and as being incompatible with Art.3 of Protocol No.1."

Those principles were re-affirmed by the Grand Chamber in Scoppola v Italy (No.3) (2013) 56 EHRR 19, rejecting submissions to the contrary by the United Kingdom as third-party intervener: see paragraphs 75-80 and 93-96. The court drew a contrast with systems in which loss of the right to vote depended on the period of imprisonment to which the convicted person was sentenced.

[19] In the United Kingdom, the courts have followed those decisions. In Smith v Scott 2007 SC 345, the Inner House of the Court of Session, sitting as a Registration Appeal Court hearing a complaint by a prisoner that he was prevented by a blanket ban from voting in elections to the Scottish Parliament, ruled that a blanket ban on prisoners voting in parliamentary elections was incompatible with A3P1 and pronounced a declaration of incompatibility. The relevant legislation in that case was the RPA 1983. The issue came again before the court in the conjoined appeals of McGeoch v Lord President of the Council and Chester v Secretary of State for Justice ("McGeoch") [2013] 3 WLR 1076. Those two cases concerned voting in elections to the European Parliament. The Supreme Court followed the Grand Chamber decisions in Hirst and Scoppola and confirmed that a blanket ban on prisoners voting in parliamentary elections was incompatible with A3P1. It did not pronounce a declaration of incompatibility, but only because the court in Smith v Scott had already pronounced such a declaration and nothing would be gained by repeating it.

Application of A3P1 to Acts of the Scottish Parliament

[20] It is clear, therefore, that a blanket ban on prisoners voting in parliamentary elections is contrary to A3P1 and unlawful. In both Scott and McGeoch the legislation imposing a blanket ban (the RPA 1983) was an Act of the UK Parliament. Under the Human Rights Act the courts had no power to strike it down. Their powers were limited to making a declaration of incompatibility. Under the Scotland Act 1998, however, the position is different, for the reasons explained above. It being beyond the legislative competence of the Scottish Parliament to legislate in a manner which is incompatible with Convention rights, any provision of an Act passed by the Scottish Parliament which is incompatible with Convention rights is necessarily ultra vires and will be struck down by the courts. Accordingly, if the legislation imposing the blanket ban considered in Scott and McGeoch had been passed by the Scottish Parliament, rather than by the United Kingdom Parliament, the finding that the blanket ban infringed A3P1 would have meant that the relevant provisions of the Act of the Scottish Parliament which purported to impose the blanket ban were not law, were ultra vires and would have been declared to be unlawful and struck down by the courts. This is not in doubt. The argument that the domestic legislature should be afforded a "margin of appreciation" and allowed, in its discretion, to impose a blanket ban was considered in McGeoch and rejected in light of the clear reasoning of the Grand Chamber in Hirst and Scoppola.

The application of A3P1 to the independence referendum franchise

[21] The petitioners argue that this reasoning applies also to the independence referendum franchise. The Scottish Parliament has sought to impose a blanket ban on prisoners voting in that referendum. Whatever the arguments in favour of some restriction on prisoners voting, a blanket ban is disproportionate for the reasons given in Hirst and discussed in McGeoch. It infringes A3P1. It should be struck down. There was no reason in principle why a blanket ban on prisoners voting, which is regarded as disproportionate in the context of voting in elections for a legislature, should not similarly be regarded as disproportionate in the context of voting in a referendum. The respondents do not suggest otherwise. It is no doubt because they could anticipate that the domestic court would be bound to regard the blanket ban as disproportionate that the respondents have not attempted to justify the ban in these proceedings; but it would be unfair to them to suggest that they do not continue to assert that it is justified.

Strasbourg authority

[22] The difficulty for the petitioner, however, is that it has been held consistently both by the Commission and by the European Court of Human Rights that A3P1 applies to voting in elections for the legislature and has no application to voting in a referendum or, for that matter, in an election to elect a president or head of state. There is a long line of cases all to this effect. The first is X v United Kingdom (Application 7096/75, decided on 3 October 1975) which was concerned with the referendum on EEC membership held in the United Kingdom in 1975. A prisoner who was excluded from voting in that referendum sought to challenge that exclusion in the European Court of Human Rights. That challenge was held to be inadmissible. A3P1 applied only to elections for the choice of a legislature. It did not apply to a referendum. The reasoning contains a reference to the fact that the referendum was only consultative, and that it was not required by law. That latter point is probably best understood as a reference to it not being required by the terms of A3P1. It may be that whether a referendum is properly regarded as consultative or binding will vary from case to case and will depend upon the legislation and political consensus (if any) pursuant to which it is held. The petitioners say that as a matter of fact in the present case it has been agreed between the Scottish Government and the United Kingdom Government that the referendum will be binding. That may be a proper ground of distinction. However the decision in X v United Kingdom has been followed by the Commission and by the Strasbourg Court, both directly and indirectly, in at least 10 subsequent decisions on admissibility: Bader v Austria (Application 26633/95, 15 May 1996), Nurminen v Finland (Application 27881/95, 26 February 1997), Castelli v Italy (Applications 35790/97 and 38438/97, 14 September 1998), Hilbe v Lichtenstein (Application 31981/96, 7 September 1999), Borghi v Italy (Application 54767/00, 20 June 2002), Santoro v Italy (Application 36681/97, 16 January 2003), Comitato Promotore Referendum Maggioritario v Italy (Application 56507/00, 8 July 2003), Bocellari v Italy (Application 399/02, 28 October 2004), Z v Latvia (Application 14755/03, 26 January 2006), Niedzwiedz v Poland (Application 1345/06, 11 March 2008) and McLean & Cole v UK [2013] 57 EHRR SE8 (11 June 2013). In all those cases, stretching in an unbroken line from 1975 to 2013, the principle has been summarised briefly: A3P1 applies to elections to the legislature; it has no application to voting in other elections nor to voting in a referendum.

[23] The petitioner in this case seeks to gain some assistance from a sentence found in para 33 of the judgment of the court in McLean: "there is nothing in the nature of the referendum at issue in the present case which would lead the Court to reach a different conclusion here". It is suggested that this shows that the court is open to persuasion that A3P1 might be applicable to a particular referendum in a particular case. A referendum such as that with which the court is presently concerned, which is not merely consultative but is agreed to be binding and decisive on the issue, might now be regarded by the Strasbourg court as being subject to A3P1. I do not consider that that sentence in McLean is a proper basis for effectively ignoring the consistent body of decisions to date to the effect that A3P1 does not apply to protect the right to vote in a referendum. That sentence is doing no more than recognising that, as with every general rule, there is always the possibility of an exception to it. One can possibly envisage a situation where a country held a referendum on a regular basis to determine important issues, relegating the role of the legislature to relatively minor issues. In such circumstances there might be scope for an argument that the referendum should be treated as the equivalent of an election to which A3P1 applied. But it does not go further than that if, indeed, it goes that far.

[24] The decision of the Commission and the Strasbourg court in these cases to the effect that A3P1 only applies to elections to a legislature and does not apply to voting in a referendum, is consistent with the wording of the article itself. A3P1 contains an undertaking by High Contracting Parties to hold elections at reasonable intervals to ensure the free expression of the opinion of the people in the choice of the legislature. It enshrines the obligation to hold regular elections as a means of choosing members of the legislature. By contrast, a referendum is typically a one-off event dealing with a single issue. While it might possibly be called an election, it is not held at regular intervals. Nor has it anything to do with the choice of the legislature. That is so even where the referendum has been about whether the country should accede to the EU, notwithstanding that it might be argued that that is a vote about the choice of legislature to which the people of that country will, in the event of a "Yes" vote, be subject. Z v Latvia and Niedzwiedz v Poland were both cases of this sort, and both were decided after the decision in Matthews v United Kingdom (1999) 28 EHRR 361 in which it was held that the European Parliament was a "legislature" for this purpose. That argument was therefore available in those cases, though the short case reports do not indicate whether such an argument was advanced. In short, therefore, neither the language of A3P1 nor the Strasbourg jurisprudence supports the petitioner's argument on this point.

[25] For these reasons I am unable to accept the petitioners' argument based on A3P1. I could only uphold their contention that A3P1 applied to voting in a referendum (as well as to voting in elections for a legislature) if there was a clear and constant line of Strasbourg jurisprudence to that effect; or if, in the absence of a clear and constant line, there were clear pointers in that direction which would enable me to conclude that when the case next came before the court in Strasbourg it would undoubtedly, or at least very probably, come to that conclusion. The domestic court should keep pace but not get ahead. Not only is there no such clear and constant line of Strasbourg jurisprudence, but such indications as there are - albeit only on the question of admissibility - almost all point towards the opposite view, namely that A3P1 does not protect the individual's right to vote in a referendum.

[26] Drawing on what was said in Manchester City Council (supra), it was submitted that the court had no duty to follow a line of Strasbourg decisions that disclosed no reasoning. Nor was the court required to follow a line of Strasbourg jurisprudence, even if reasoned, which was not decided by the Ground Chamber. Accordingly, the decisions to which I have referred, holding that A3P1 did not protect the right to vote in a referendum, did not have to be followed. Since there is no absolute obligation to follow decisions of the Strasbourg court, I can accept the first part of this submission to this extent, that if the unreasoned decisions did not appear to accord with any reasoned analysis, they might carry little weight. As to the second submission, since there is no obligation to follow any Strasbourg line of authority, it must follow that there is no obligation to follow decisions which are not Grand Chamber decisions. However, the basic proposition that A3P1 does not protect the right to vote in a referendum seems to me to be entirely consistent with the wording of the Article. That point is made in X v United Kingdom. It is a short point, and any reasoning in support of the decision is not required to be lengthy. Further, it is re-iterated in decision after decision. Although those decisions are not decisions of the Grand Chamber, I can see no reason to doubt their correctness or to think that, were it to come before the Grand Chamber at some time in the future, the decisions would be reversed.

[27] It may be that at some date in the future the Strasbourg court will wish to reconsider its interpretation of A3P1 and extend it to protect the right to vote in a referendum. Three arguments were advanced as to why it might do so.

[28] First, it was suggested that the holding of a referendum on a specific issue of importance, usually but not always on some constitutional issue, was becoming more frequent. The result of such a referendum was now often regarded as decisive, de facto if not de jure. If it became the practice by such a means to obtain democratic legitimacy for the making of law, the legislature might to that extent be bypassed, and the guarantee enshrined in A3P1 as currently interpreted might not achieve the aim of ensuring full participation in the democratic process.

[29] Secondly, it was observed that certain international conventions and codes of practice pointed against the appropriateness or proportionality of a blanket ban on prisoners voting not only in regular elections for the legislature but also in a referendum. The International Covenant on Civil and Political Rights 1966 ("ICCPR"), which was ratified by the United Kingdom in 1976, provides by article 25 that every citizen shall have the right and the opportunity and without discrimination or unreasonable restrictions "(a) to take part in the conduct of public affairs, directly or through freely chosen representatives, [and] (b) to vote and to be elected at genuine periodic elections which shall be by universal and equal suffrage and shall be held by secret ballot, guaranteeing the free expression of the will of the electors...". This has been interpreted by the Human Rights Committee, the body established to consider complaints concerning violation of the rights set out in the Covenant, as conferring on the citizens of signatory states a right to vote in a referendum as well as in elections to legislative bodies: see General Comment No 25 (57) set out at Annex V to the General Comments under article 40, para 4 of the International Covenant on Civil and Political Rights, adopted by the Committee at its 1510^th meeting (fifty-seventh session) on 12 July 1996; but and the subsequent decisions of the Committee in Gillot v France (Communication No.932/2000) and Yevdokimov v Russia (Communication No.1410/2005).

[30] Similarly, the Venice Commission's Code of Good Practice in Referendums (Resolution 235 (2007)) declares at para 1.1 that any deprivation of the right to vote in a referendum based on a criminal conviction must be limited to the case where the criminal conviction is for "a serious offence" and the withdrawal of the right to vote may only be imposed by "express decision of a court of law". It also emphasises that the proportionality principle must be observed. The Venice Commission is merely an advisory body of the Council of Europe, of which the United Kingdom is a member, and its Code of Good Practice in Referendums is not, of course, binding on the United Kingdom. It is certainly not incorporated into English or Scots law. Nonetheless, it was submitted that it represents an example of good practice.

[31] It is clear that the Strasbourg court has on occasion been influenced in its interpretation of the Convention by other conventions or covenants to which some or all of the High Contracting Parties are signatories; and also by any emerging consensus amongst High Contracting Parties. For example, in Bayatayan v Armenia (Application No. 23459/03) the applicant was a Jehovah's Witness who was convicted and sentenced to imprisonment for failing to report for military service. He successfully applied to the European Court of Human Rights on the basis of a violation of his article 9 rights (freedom of thought, conscience and religion). In the course of its decision, the Grand Chamber considered whether there was a need for a change in its case law. It recognised (at para 98) that it was in the interests of legal certainty, foreseeability and equality before the law that it should not depart without good reason from precedents laid down in previous cases. However, it expressed a concern that a failure by the court to maintain "a dynamic and evolutive approach" would risk rendering it a bar to reform or improvement. The convention had to be interpreted and applied in a manner which rendered its rights practical and effective, not theoretical and illusionary. It noted an earlier interpretation of article 4 ECHR but was not convinced that it reflected its true purpose and meaning. It was also conscious (at para 1) that the restrictive interpretation of article 9 previously applied "was a reflection of the ideas prevailing at the material time". It then said this (at para 102):

"102. The Court reiterates in this connection that the Convention is a living instrument which must be interpreted in the light of present-day conditions and of the ideas prevailing in Democratic States today .... Furthermore, in defining the meaning of terms and notions in the text of the Convention, the Court can and must take into account elements of international law other than the Convention and interpretation of such elements by competent organs. The consensus emerging from specialised international instruments may constitute a relevant consideration for the Court when it interprets the provisions of the Convention in specific cases ...."

After going on to observe a trend amongst European countries to recognise the right to conscientious objection, the Court went on in para 105 to say this:

"105. The Court would further point out the equally important developments concerning recognition of the right to conscientious objection in various international fora. The most notable is the interpretation by the UNHRC of the provisions of the ICCPR ... which are similar to those of the Convention. Initially the UNHRC adopted the same approach as the European Commission .... However, in 1993, in its General Comment No.22, it modified its initial approach and considered that a right to conscientious objection could be derived from Article 18 of the ICCPR ..."

I need not cite that passage in greater detail. It is sufficient to note for present purposes that the Grand Chamber clearly considered that in interpreting the relevant Articles of the Convention it was of assistance to note developments taking place outside the narrow confines of its own case law. In that case it made specific mention of a change of approach by UNHRC in its interpretation of certain Articles of the ICCPR. It went on in the following para to mention the coming into force in 2009 of the Charter of Fundamental Rights of the European Union which, on the particular point at issue, it considered "reflects the unanimous recognition of the right to conscientious objection by the member States of the European Union, as well as the weight attached to that right in modern European society." It concluded, in para 108, that since its earlier decision the domestic law of the overwhelming majority of Council of Europe member states, along with the relevant international instruments, had evolved to the extent that there was already virtually general consensus on the question in Europe and beyond, in the light of which it could not be said that a shift in its interpretation of article 9 was not foreseeable. In the light of that, "and in line with the 'living instrument' approach", it took the view that it was not possible to confirm the earlier case law.

[32] I cite this case simply to illustrate the fact that the Strasbourg court will, in an appropriate case, look outside the literal construction of the Articles of the Convention and beyond its own case law to see what developments there have been internationally about the relevant issues; and what consensus, if any, has grown up so as to find expression in other outlets. Many other cases in Strasbourg have reflected the same approach in interpreting specific Articles of the Convention: see, for example, Bankovic v Belgium (2007) 44 ECHRR SE5 at para 57; Demir and Baykara v Turkey (2009) 48 EHRR 54 at paras 76 and 78; Sitaropoulos v Greece (2013) 56 EHRR 9 at paras 67-71; and James v United Kingdom (2013) 56 EHRR 12. The case of In re McCaughey [2012] 1 AC 725 illustrates how the Strasbourg court may alter its interpretation of rights guaranteed by the Convention in light of such materials; in the first six paragraphs of his judgment Lord Phillips traces the changes made by the Grand Chamber as to the nature of the obligations imposed by article 2 of the Convention (right to life).

[33] It is therefore quite possible - I put it no higher or lower than that - that at some time in the future the Strasbourg court may revisit the ambit of A3P1 and hold that it protects the right to vote not only in elections to the legislature but also in referendums, or at least those which meet certain defined criteria. But it has not done so yet, despite the fact that it is some years now since the United Nations Human Rights Committee in its General Comment and in its decisions in Gillot and Yevdokimov gave that interpretation to article 25 of the ICCPR. It is, in my view, clear that the point has not yet been reached where a domestic court can say with any degree of confidence that the Strasbourg court will take that step.

Article 10

[34] Article 10 of the Convention is concerned with freedom of expression. It provides as follows:

"1. Everyone has the right to freedom of expression. This right shall include freedom to hold opinions and to receive and impart information and ideas without interference by public authority and regardless of frontiers. This Article shall not prevent States from requiring the licensing of broadcasting, television or cinema enterprises.

2. The exercise of these freedoms, since it carries with it duties and responsibilities, may be subject to such formalities, conditions, restrictions or penalties as are prescribed by law and are necessary in a democratic society, in the interests of national security, territorial integrity or public safety, for the prevention of disorder or crime, for the protection of health or morals, for the protection of the reputational rights of others, for preventing the disclosure of information received in confidence, or for maintaining the authority and impartiality of the judiciary."

I have set out in full the terms of para 2 of article 10 simply to make it clear that the article is not purporting to entrench an unfettered right to freedom of expression. In the present case no reliance is placed on those qualifications set out in para 2. That is not surprising, since the objectives sought to be pursued by the Scottish Parliament in enacting the blanket ban in the Franchise Act are no doubt similar to those sought be pursued by the United Kingdom Parliament in enacting the blanket ban in the RPA 1983. It having been held both in the Grand Chamber of the European Court of Human Rights and in the Supreme Court, in relation to United Kingdom legislation, that a blanket ban is not a proportionate way of pursuing those objectives, however legitimate those objectives might be, it is difficult to conceive that any other answer would be given to a similar question, should it arise, in respect of Scottish legislation such as the Franchise Act.

[35] Accordingly, the question for decision in the present case is simply this: does the right to freedom of expression as confirmed by para 1 of article 10 extend to protect the right to vote in a referendum? The petitioners say that it does. The respondents say that it does not and, what is more, that it does not extend even to protect the right to vote in an election to the legislature. That, they say, is the exclusive province of A3P1.

[36] The petitioners submitted that article 10 should be construed as protecting the right to vote and, by necessary inference, the right to vote in a referendum called to determine the manner in which the country was to be governed for the foreseeable future. They submitted that it had been recognised in Castelli v Italy (Applications 35790/97 and 38438/97, 14 September 1998) that casting a vote was an act of expression. It followed that any restriction on the right to cast a vote was an interference with freedom of expression on grounds which were proportional and in pursuit of a legitimate aim or objective. Article 10 required to be interpreted so that the rights conferred by it were practical and effective. The referendum to be heard in September 2014 would be binding in practice, even if not in law. Accordingly, to make the freedom of expression guaranteed by article 10 effective, everyone should be allowed to vote unless there were legitimate objections. It was submitted that the Grand Chamber in Hirst and the court in Anchugov and Gladkov v Russia (Applications 1157/04 and 15162/05, decided on 4 July 2013) had recognised that a complaint under article 10 against a prohibition on prisoners voting in an election to the legislature was admissible. There was no reason why article 10 should not also apply to voting in a referendum. The reasons for holding that A3P1 did not apply to a referendum had no application to article 10.

[37] It was recognised on behalf of the petitioners that there were a number of past decisions on admissibility of both the European Commission of Human Rights and the Strasbourg court in which it had been said that the right to freedom of expression protected under article 10 did not include the right to vote. It was submitted, however, that those cases did not constitute the sort of clear and constant jurisprudence which the domestic courts were bound to follow. None of them was a decision of the Grand Chamber. None of them disclosed any real reasoning. Thus, in X v Netherlands (Application 6573/74) the commission simply asserted that in its opinion "this provision does not guarantee the right to vote as such". That was followed, either directly or indirectly, without any additional reasoning, in X v United Kingdom (Application 7096/75), X v Germany (Application 6850/74), Luksch v Italy (Application 27614/95), Baskauskaite v Lithuania (Application 41090/98) and Borghi v Italy (Application 54767/00). In Liberal Party v United Kingdom (1980) 4 ECHRR 185, however, the Commission held the article 10 claim to be inadmissible because article 10 did not guarantee the right to vote nor the right to stand for election "or any other right already secured by [A3P1]". That was consistent with Holland v Ireland (Application 24827/94), Zdanoka v Latvia (Application 58278/00) and Hirst, where the court did not consider whether there was a breach of article 10 because it had applied A3P1 as the "lex specialis".

[38] It was submitted that these cases, showed that the Strasbourg jurisprudence had moved on. In Hirst (Application 74025/01) the decision on admissibility on 8 July 2003 was that the complaints under both A3P1 and article 10 raised "serious issues of fact and law... which should be examined on their merits". The decisions on the merits of the application, by the Chamber (on 30 March 2006) at para 56 and the Grand Chamber (at paragraphs 65, 71 and 89), was that, having found there to be a violation of A3P1, no separate issue arose for decision under article 10. In the combined decision in Anchukov and Gladkov v Russia, after making enquiries of the Russian Government in that and other applications - viz. Gladkov (App 15162/05), Bolsunoviskiy (Application 16824/10) and Kuchmara (Application 4664/10 - as to the applicability of article 10, the court ruled on admissibility and on the merits at the same time and, rejecting the arguments of the Russian government, held that the article 10 and A3P1 complaints were both admissible. On the merits the court found it unnecessary to deal with article 10 because it decided in favour of the applicant on A3P1. This was authority, it was argued, for the proposition that Article 10 (the "lex generalis") could apply notwithstanding the possible application of A3P1, the "lex specialis". If A3P1 did not apply, the position was a fortiori; there was no reason to hold that article 10 was excluded.

[39] For the respondents it was submitted that article 10 did not mention the right to vote and did not appear to protect it. That was the view taken in Clayton & Tomlinson, The Law of Human Rights, at para 15.241. The Convention had to be read as a whole. The general provisions should not be construed as intruding into areas covered by specific provisions. This was the approach adopted by the court in two recent decisions. In Austria v United Kingdom (2012) 55 EHRR 14 the applicant complained that he had been deprived of his liberty by the police using "kettling" or containment tactics to control a demonstration. Article 2 of the Fourth Protocol guaranteed the right to liberty of movement, but the United Kingdom had not ratified the Fourth Protocol. Accordingly, the applicant brought his complaint under article 5(1) (right to liberty). The Court held that the complaint was admissible but that it failed on the merits. It considered that article 5(1) should not be interpreted as applying to matters covered by article 2 of the Fourth Protocol to which the United Kingdom had not acceded. It therefore did not cover mere restrictions on liberty as opposed to total deprivation of liberty. On the facts there was no such total deprivation. Misick v United Kingdom (2013) 56 ECHRR SE13 was an admissibility decision by the Fourth Section, where for various reasons the United Kingdom Government temporarily suspended representative government in the Turks & Caicos Islands. The applicant brought a complaint by way of judicial review in the domestic courts, founding upon A3P1. This was answered by the United Kingdom Government withdrawing A3P1 from the Islands (as it was entitled to do). The application to the Strasbourg court was based upon article 8 (respect for private and family life). The court ruled that the application was inadmissible. It held that article 8 should not be used to fill the gap left by the withdrawal of A3P1. In both cases the unwillingness of the court to use the general provision (the lex generalis) to cover much the same ground as the specific provision (the lex specialis) and to extend its range was based upon an insistence that "the Convention must be read as a whole, and interpreted in such a way as to promote internal consistency and harmony between its various provisions". In Gladkov, on which the petitioners relied, the applicant had been excluded from voting in any elections. In the combined decision in Anchukov and Gladkov the court ruled that that part of the complaint which related to the barring of the applicant voting in the election for president of the Russian Federation was not within A3P1 (see para 54). Nevertheless it saw no separate issue arising under article 10. In the absence of a claim within A3P1, article 10 would have been directly relevant had it protected the right to vote. Accordingly, that case was a recent decision to the effect that article 10 did not guarantee the right to vote, whether in elections to the legislature or in any other form of election. Such a decision was consistent with the long line of decisions on admissibility referred to above.

[40] On the material before me, I have come to the view that there is no clear and constant jurisprudence in Strasbourg to the effect that article 10 protects the right to vote, whether in legislative elections or in any other forum. On the contrary, in so far as there is any consistency in the jurisprudence created by the series of decisions on admissibility, they support the view that article 10 does not have any application in this sphere. Complaints about restrictions on voting based upon article 10 have frequently been held to be inadmissible. In those cases, therefore, the proposition that article 10 applies to protect the right to vote has been considered to be, in effect, unarguable. While it is true that the court in Hirst appears to have accepted that the complaint under article 10 was admissible, it did not need to decide that it was anything more than that for the purpose of its decision in the case. At best, therefore, Hirst provides some support for the proposition that a complaint based upon article 10 is not unarguable. I cannot, however, regarded it as providing support for the proposition that the Strasbourg jurisprudence has shifted significantly on this point. As I say, it was not necessary for the court in Hirst to decide the case on the basis of article 10. The subsequent decision in Anchukov and Gladkov suggests that had it had to decide the article 10 complaint it would have rejected. In any case, Hirst was concerned with the right to vote in elections to a parliament. It was not concerned with voting in a referendum. Given the reiteration by the court of the need for the Convention to be read as a whole, and interpreted in such a way as to promote internal consistency and harmony between its various provisions, it would be odd if article 10, the general provision, could be interpreted as giving a right to vote in a referendum where A3P1, the specific provision, did not. That would fail to give any coherence and consistency to the Convention read as a whole and would, in effect, render A3P1 virtually irrelevant.

[41] Far from there being a clear and constant line of Strasbourg jurisprudence to the effect that article 10 can be relied upon in a case such as this, it seems to me that the decided cases point the other way. In those circumstances, following the approach that this court should keep pace with Strasbourg jurisprudence but not go further, I reject this part of the petitioners' case.

[42] I was referred in argument on this point to the decision of the Supreme Court of Canada in Haig v Canada [1993] 2 RCS 995 in which it was held that the right to vote was not protected by section 3 of the Charter which guaranteed freedom of expression. There is, to my mind - and I understood both parties to be of the same mind when it suited their arguments - a danger in a domestic court relying on Commonwealth or other non-European authority to interpret the ECHR. It may well be that the Strasbourg court itself might regard a reference to such non- Strasbourg and non-European case law as being of some assistance, but that is a matter for it and not for this court.

[43] I should add that I have considered in the context of the article 10 argument the same materials as were relied upon in connection with A3P1 to suggest that the Strasbourg court might be influenced to alter its approach to the protection of the right to vote in a referendum (see para [] above). I do not need to repeat the arguments here. Suffice it to say that it is for the Strasbourg court, if so minded, and not for this court to take such materials into account in going that one step further than it has done at present in interpreting the Convention.

Conclusion on ECHR challenge

[44] For the reasons set out above, I reject the petitioners' challenge to the blanket ban based upon A3P1 and article 10.

Common law challenges

[45] The petitioners' common law challenges were presented under three separate heads. They submitted first that the blanket ban on prisoners voting in the referendum was contrary to the rule of law; second that it was contrary to the fundamental or constitutional right to vote; and third that it was contrary to the United Kingdom's international obligations. The first and second submissions are essentially different ways of making the same point. I propose to deal with them together. First, however, it is necessary to identify the grounds upon which an Act of the Scottish Parliament may be reviewed at common law.

Grounds of review at common law

[46] The susceptibility of an Act of the Scottish Parliament to review at common law was the subject of detailed consideration by the Supreme Court in AXA General Insurance Company Ltd v Lord Advocate 2012 SC (UKSC) 122. For present purposes it is not necessary to look beyond that decision. In AXA it was held that an Act of the Scottish Parliament was not reviewable on grounds of irrationality, unreasonableness or arbitrariness: see per Lord Hope at para [52] and Lord Reed at paragraphs [142]-[148]. However, the judgments of Lord Hope (at para 51), Lord Mance (at para 97) and Lord Reed (at paragraphs 149-153) provide some support for the proposition that the Scottish Parliament has no power to legislate contrary to the rule of law or "fundamental rights". Were it to do so, the courts could intervene.

[47] I put to one side for present purposes the question of what is meant by the rule of law and what is embraced within the expression fundamental rights. I come back to that later. Nor is it necessary for present purposes to seek to resolve the question, "still under discussion" (per Lord Hope at para [50]), of whether there are any common law constitutional limits on the legislative powers of the United Kingdom Parliament. This case is not concerned with the legislative powers of the United Kingdom Parliament. The Scottish Parliament, though plenary, is a devolved legislature. There is nothing heretical about the notion that its powers are limited. They are expressly limited by reference to the ECHR and European Union Law. There is no reason in principle why they should not also be limited by reference to the rule of law and certain fundamental rights. For present purposes I am prepared to proceed on the basis that they are so limited.

[48] In order to understand the argument advanced by the petitioners, it is important to note two things.

[49] First, and unsurprisingly, the petitioners did not contend that the legislative powers of the United Kingdom Parliament were in any way limited. They proceeded on the basis that the United Kingdom Parliament was sovereign, and that it was within the legislative powers of the United Kingdom Parliament both to restrict the right to vote, whether in an election to the legislature or in a referendum, by imposing a blanket ban on prisoners, and also to give the Scottish Parliament a similar power to restrict the right to vote by imposing a blanket ban. Such a concession was, perhaps, inevitable, at least in respect of the United Kingdom Parliament's power to impose a blanket ban, given that the Supreme Court in McGeoch had not been invited to venture into that territory; and the concession that the United Kingdom Parliament had the power to give the Scottish Parliament a similar power in setting the franchise for the referendum perhaps flowed inevitably from that.

[50] Secondly, however, and this follows logically from the first point, they did not contend that the restriction on the power of the Scottish Parliament to legislate in breach of the rule of law or of some fundamental right was unqualified. The submission was a more limited one, namely that the Scottish Parliament could not legislate in breach of the rule of law or of some other fundamental right unless it was expressly given the right to do so by the United Kingdom Parliament in the Scotland Act.

[51] Formulated in this way, it seems to me that what was really being contended for was what is sometimes known as the "principle of legality", explained by Lord Reed in AXA at para [152] as meaning "not only that Parliament cannot itself override fundamental rights or the rule of law by general or ambiguous words, but also that it cannot confer on another body, by general or ambiguous words, the power to do so." Thus, if the court accepts the petitioners' arguments that there is a constitutional or fundamental right to vote, interference with which requires to be proportional and in pursuit of a legitimate aim or objective, the argument for the petitioners on these aspects of the common law challenge comes down to a question of construction of the Scotland Act, namely: does the Scotland Act expressly or by clear implication give the Scottish Parliament the power to restrict the franchise for the independence referendum in breach of that fundamental right?

[52] Put in this way, there is, to my mind, a certain artificiality about the argument. Statutory interpretation involves an attempt to ascertain the intention of Parliament. In the case of the Scotland Act, that means the intention of the United Kingdom Parliament. All statutes require to be construed in their context. The petitioners' argument involves the proposition that in conferring powers on the Scottish Parliament, the United Kingdom Parliament cannot have intended to allow the Scottish Parliament to restrict the franchise in an election or referendum by imposing a blanket ban on prisoners voting. That, it is said, would be contrary to the rule of law and contrary to a fundamental right; and, therefore, since Parliament cannot be taken casually to have legislated to override fundamental rights or the rule of law, if it intends to do so, it must do so in the clearest words. Accordingly, unless the Scotland Act contains such clear words, it must follow that Parliament did not intend to allow the Scottish Parliament to legislate contrary to fundamental rights and the rule of law. However, if one thing is clear in this field it is that the United Kingdom Parliament is of the clear view that no convicted prisoner serving his sentence of imprisonment should be allowed to vote in elections or in a referendum. That is its position in the RPA 1983, and it was its position in respect of the franchise for the EEC referendum in 1975 and the alternative vote referendum in 2012. The notion that the United Kingdom Parliament, by not including a specific provision in the Scotland Act empowering the Scottish Parliament to enact a blanket ban on prisoners voting, manifested an intention not to give the Scottish Parliament that power is, in those circumstances, somewhat far-fetched.

[53] Nonetheless, I deal with this submission on its merits in the following paragraphs. This makes it necessary to identify in principle what is embraced within the rule of law for this purpose and what rights are for this purpose to be regarded as fundamental; and then to focus more specifically on the question whether the blanket ban on prisoners voting contained in the Franchise Act calls into question those limits.

Rule of law/ fundamental rights

[54] In the course of argument I asked junior counsel for the petitioners what he meant by the "rule of law" in this context. Counsel was initially unable, or perhaps reluctant (I do not mean this critically), to offer any definition. But it is an important starting point for any argument that the Scottish Parliament cannot pass particular legislation because it breaches the rule of law. The phrase is often used to mean different things. For example, in R (Cart) v Upper Tribunal [2011] QB 120, for the purpose of considering the whether the Upper Tribunal was amenable to judicial review, Laws J felt it necessary to discuss the concept of the rule of law, which he described as protean. He identified the rule of law, so far as relevant to that issue, as being the principle that "statute law has to be mediated by an authoritative judicial source, independent both of the legislature which made the statute, the executive government which (in the usual case) procured its making, and the public body by which the statute is administered". That is clearly not what is meant in this case. I was referred to Lord Bingham's lecture on The Rule of Law (The Sixth Sir David Williams Lecture). Lord Bingham offered some thoughts on the definition of the rule of law. He thought that the core of the existing principle was "that all persons and authorities within the state, whether public or private, should be bound by and entitled to the benefit of laws publicly and prospectively promulgated and publicly administered in the courts". To put more flesh on that definition, he identified eight sub-rules. Some of them have no application to the present case. Perhaps the most relevant is in his fourth sub-rule, which was that "the law must afford adequate protection of fundamental human rights." I was also referred to his eighth sub-rule which required "compliance by the state with its obligations in international law, the law which, whether deriving from treaty or international custom and practice, governs the conduct of nations". This latter sub-rule perhaps is more relevant to a different part of the common law challenge in this case. For present purposes, the affording of adequate protection of fundamental human rights appears to me to lie at the heart of the petitioners' contentions. An essential feature of that was the guarantee of the right to vote.

[55] In this connection, counsel referred me to the following statement in the decision of the Grand Chamber in Hirst at para 58:

"58. The Court has had frequent occasion to underline the importance of democratic principles underlying the interpretation and application of the Convention and it would use this occasion to emphasise that the rights guaranteed under Art.3 of Protocol No.1 are crucial to establishing and maintaining the foundations of an effective and meaningful democracy governed by the rule of law."

That is a clear and unequivocal declaration that the right to vote guaranteed under A3P1 is an essential part of an effective democracy governed by the rule of law. A similar statement was made by the majority of the Supreme Court of Canada in Sauvé v Canada [2002] 3 RCS 519, a case to which I will return, at para 9:

"The right to vote is fundamental to our democracy and the rule of law and cannot be lightly set aside".

Other relevant passages are at paragraphs 27 (referring to the "centrality of the right to vote to Canadian democracy [and] the rule of law") and 58-9 ("Denial of the right to vote to penitentiary inmates undermines the legitimacy of government, the effectiveness of government, and the rule of law. ... what is at stake is the denial of the fundamental rights of every citizen to vote"). This emphasis on the rule of law guaranteeing the right to vote in a democracy suggests that there is a complete overlap between the principle of the rule of law as applied in this case and the principle of respect for fundamental rights which is the other limb of the petitioners' case.

[56] In support of the proposition that the right to vote is a fundamental right in the sense which I have described, I was taken to certain statements by Lord Bingham and Lord Rodger in Watkins v Secretary of State for the Home Department [2006] 2 AC 395 at paragraphs 24 - 25 and 58 - 61. That case concerned the conduct of prison officers in opening prisoners' mail and involved a consideration of what was described as the "constitutional right" to protection of confidentiality of legal correspondence. To that extent, any statements about other fundamental or constitutional rights were obiter. It is nonetheless of importance and deserving of respect that Lord Bingham expressed the view that we would "now, of course, regard the right to vote as basic, fundamental or constitutional" (see para 25); and that Lord Rodger (at para 61) talked of the right to vote as falling "within everyone's notion of a 'constitutional right'".

[57] Accepting, for present purposes, that the right to vote may be described as "fundamental" or "constitutional", it is still necessary to consider what that entails. This is far from easy. Watkins concerned the question of whether proof of damage was a necessary ingredient of the tort of misfeasance in public office, and the question whether the right to confidentiality of legal correspondence was to be described as "constitutional" arose in this context. The Court of Appeal, founding on Ashby v White and other election cases from the 19^th century, took the view that certain "constitutional" rights might be vindicated in an action in tort without proof of special damage. Despite that different context, however, some of the observations in the House of Lords are instructive. I would highlight the following matters as being of some relevance to the present case. They overlap to some extent. First, it is necessary to know for what purpose the right is or is not to be categorised as "constitutional" or "fundamental" (see para 58); in other words, what impact such characterisation is intended to have. Second, "there ... is no magic in the term 'constitutional right'": see para 58. "Fluctuations in terminology are only to be expected": see para 61. The terms "constitutional" or "fundamental" cover a number of different concepts. They tend to be used, in particular, in the sphere of interpretation of statutes (the principle of legality referred to above): see para 61. It is, therefore, necessary to know in what sense the expressions are intended to be used. Third, although the terms "work well enough, alongside equivalent terms, in the field of statutory interpretation", if it is necessary to define them for other specific purposes, the substantive rights to which the label attaches require to be identified with rather more precision: see para 62. Fourth, where the rights which are said to be constitutional or fundamental are not already established at common law but fall within the scope of the ECHR, "a claimant can be expected to invoke his remedy under the Human Rights Act rather than to seek to fashion a new common law right": see para 64. Subject to the point made by Lord Reed in Osborn v The Parole Board at paragraphs 54- 63 about the gradual development of the common law in line with statute and Strasbourg jurisprudence, caution should be exercised before seeking, by reference to ECHR principles, to develop freestanding common law constitutional or fundamental rights which have not existed before.

[58] The existence of a constitutional or fundamental right to vote can now be asserted at a level of generality which would, I suspect, brook no contradiction in a modern western democracy. But it was not always so in any meaningful sense. In centuries past, the right to vote depended upon property rights, gender and, occasionally, educational qualifications. As late as the early twentieth century the universality of the right to vote amongst adults regardless of gender was not recognised in the United Kingdom. In terms of our unwritten constitution, even now the right to vote in elections is limited to one of the two Houses of Parliament, the other being made up of hereditary and appointed peers. There is no general right to vote on issues in a referendum, the system throughout the United Kingdom being one of representative democracy where, exceptionally, Parliament may legislate to allow specific questions to be put to a popular vote. If one were to talk of a constitutional right to vote in other than general terms, it would have to be circumscribed by reference to these (and, no doubt, other) matters. If the right to vote is to be held up as a constitutional principle entitling the court to question the validity of legislation of the Scottish Parliament, and this is the petitioner's case on this limb of the argument, it needs to be defined rather more carefully.

[59] In this connection I was taken to a number of Commonwealth common law judgments, viz three decisions of the Supreme Court of Canada, namely Haig v Canada (Chief Electoral Officer) [1993] 2 RCS 995, Libman v Quebec (Attorney General) [1997] 3 RCS 569 and Sauvé v Canada (Chief Electoral Officer) [2002] 3 RCS 519; a decision of the South African Constitutional Court, namely August v Electoral Commission 1999 (3) SA 1 (CC); a decision of the High Court of Kenya, namely Kanyua v Attorney General and Interim Independent Electoral Commission [2010] eKLR; and a decision of the High Court of Australia, namely Roach v Electoral Commissioner [2007] HCA 43. It was submitted on behalf of the petitioners that these cases confirmed that the right to vote was regarded as a constitutional or fundamental right at common law, and therefore supported the proposition advanced by them in this case. It is necessary to look briefly at these cases and, in doing so, to quote more extensively from the judgments.

[60] I consider first the Canadian cases. They involved questions relating inter alia to section 3 of the Canadian Charter of Rights and Freedoms which provided that

"Every citizen of Canada has the right to vote in an election of members of the House of Commons or of a legislative assembly and to be qualified for membership therein."

It goes without saying, but it is nonetheless important to recognise it, that any statements in the cases about the importance of the right to vote were made against the background of this written constitutional guarantee. The case of Haig concerned a referendum which the federal government directed should be held on a question relating to the Constitution of Canada in all provinces and territories except Québec. Québec was to hold a separate referendum on the same date and on the same question, but in accordance with the provincial legislation. As a result of different requirements as to residency in the federal and the provincial legislation, the appellant, who had moved from Ontario to Québec in mid-1992, was not qualified to vote in the Québec referendum because he had not resided there for six months prior thereto or in the federal referendum because on the enumeration date he was not ordinarily resident within one of the polling divisions established for the purpose of the referendum. He brought an application seeking a declaration that, on its proper interpretation, section 3 of the federal Referendum Act included a resident who was ordinarily resident in a province at any time in the six month period prior to that referendum; and in the alternative, and this is more relevant to the present case, a declaration that denying him a vote in the federal referendum violated his rights under inter alia section 3 of the Charter. His application failed on all counts.

[61] In giving the judgment on behalf of the majority of the court, L'Heureux-Dubé J drew the following distinction between the right to vote in regular elections to the legislature and the right to vote in a referendum (at page 1031-2):

"The purpose of s.3 of the Charter is, then, to grant every citizen of this country the right to play a meaningful role in the selection of elected representatives who, in turn, will be responsible for making decisions embodied in legislation for which they will be accountable to their electorate. ...

The democratic rights guaranteed in the Charter are also positive ones. Federal and provincial governments have a mandate to hold regular elections to allow citizens to select their representatives. The failure to hold such regular elections would violate the Charter, would open the government to account for such constitutional infringements, and would undoubtedly provoke a constitutional crisis. Since the results of an election are clearly binding upon citizens in a Democratic society, failure to act upon such results would entail a serious constitutional breach.

A referendum, on the other hand, is basically a consultative process, a device for the gathering opinions. Voting in a referendum differs significantly from voting in an election. First, unless it legislatively binds itself to do so, a government is under no obligation to consult its citizens through the mechanism of a referendum. It may ... bind itself to conduct a specific referendum but, in the absence of such legislation, there is no obligation to hold this type of consultation. Second, though a referendum may carry great political weight and a government may choose to act on the basis of the results obtained, such results are non-binding in the absence of legislation requiring a government to act on the basis of the results obtained. In the absence of binding legislation, the citizens of this country would not be entitled to a legal remedy in the event of non-compliance with the results. Were a government to hold a referendum and then ignore the results, the remedy would be in the political and not the legal arena. These differences provide further evidence that the constitutionally guaranteed right to vote does not contemplate the right to vote in a referendum.

Section 3 of the Charter is clear and unambiguous as is its purpose: it is limited to the elections of provincial and federal representatives. Consequently, since a referendum is in no way such a selection, the citizens of this country cannot claim a constitutional right to vote in a referendum under s.3 of the Charter. ..."

L'Heureux-Dubé J returned to the topic at page 1040-1 in the context of an argument based on section 2(b) of the Charter, which protected the right to freedom of expression. This passage is of some interest, because the right guaranteed by that section was not, in terms of interference with the right to vote, restricted by the narrow parameters of section 3. He said this:

"... As I stated at the outset, there is no dispute concerning the importance of freedom of expression. Nor is it disputed that voting is a form of expression. Further, in the context of religious objections, it is clear that voting as a means of expression is constitutionally entrenched in s.3 of the Charter. However, there is just as clearly no constitutionally entrenched right to vote in a referendum.

A referendum is a creation of legislation. Independent of the legislation giving genesis to a referendum, there is no right of participation. The right to vote in a referendum is a right accorded by statute, and the statute governs the terms and conditions of participation. The Court is being asked to find that the statutorily created platform for expression has taken on constitutional status. In my view, though a referendum is undoubtedly a platform for expression, s.2(b) of the Charter does not impose upon a government, whether provincial or federal, any positive obligation to consult its citizens through the particular mechanism of a referendum. Nor does it confer upon all citizens the right to express their opinions in a referendum. A government is under no constitutional obligation to extend this platform of expression to anyone, let alone to everyone. A referendum as a platform of expression is, in my view, a matter of legislative policy and not of constitutional law."

[62] The petitioners here sought to draw a distinction between the referendum in issue in that case, which was described as purely consultative, and that which is to take place on the question of Scottish independence, where it is said, and I have no reason to doubt that this is true, that the outcome of the referendum will in practice and as a matter of agreement be binding. There may be some force in that distinction, although it is to be noted that the comments in Haig about the referendum being consultative only were made in the context of the discussion of section 3 (right to vote) rather than in the context of the discussion of section 2(b) (right to freedom of expression). Perhaps of more relevance, however, is the point made in the discussion of section 2(b), to the effect that the constitutional right to vote is satisfied by the exercise of the right to vote in elections to the legislature. There is no separate entitlement, constitutional or otherwise, to be consulted in a referendum on any particular aspect of the Constitution, or more generally, on other aspects of government or administration. I come back to this point later.

[63] I should also mention the judgment of Cory J who agreed in the result though not with all the reasoning of the majority. He emphasised that the right to vote was of "fundamental importance to Canadians and to our Canadian democracy" and, because of this, considered that the enfranchising statutes should be given the broadest and most liberal interpretation: see page 1058. He would have been prepared to interpret the residency requirements in the federal statute as covering the case of someone in the position of Mr Haig. However, he considered that Mr Haig's application must fail, since he could have taken a different route and applied to be added to the list of voters in his former residence: see page 1059. For present purposes, it is only necessary to quote from two passages in his judgment. The first is at page 1048-1049, where he deals with the right to vote as affecting the approach to interpretation of the statutory franchise provisions:

"All forms of democratic government are founded upon the right to vote. Without that right, democracy cannot exist. The marking of a ballot is the mark of distinction of citizens of a democracy. It is a proud badge of freedom. While the [Charter] guarantees certain electoral rights, the right to vote is generally granted and defined by statute. That statutory right is so fundamental that a broad and interpretation must be given to it. Every reasonable effort should be made to enfranchise citizens. Conversely, every care should be taken to guard against disenfranchisement. ...

The courts have always recognised the fundamental importance of the vote and the necessity to give a broad interpretation to the statutes which provide for it. This traditional approach is not only sound it is essential for the preservation of democratic rights. ..."

The second is at page 1050 in which he stresses the importance of the right to vote on the referendum issue:

"During the course of the hearing an argument was advanced that a referendum was distinct from and less important than an election. It was argued that, as a result, the generous principles applicable to the right to vote in elections should not apply with the same force to a referendum. I cannot accept that contention. A vast amount of public study, effort and time was expended in drafting the terms of the Charlottetown Accord. Every effort was made to advise Canadians of the importance of the referendum pertaining to it and the significance of the vote of every citizen. The number of voters exercising their franchise in the referendum was comparable to the turnout in federal elections. In the minds of most Canadians, the referendum was every bit as important as an election. If it was not, then Canadians would be clearly justified in wondering what all the fuss was about. The same principles applicable to the right to vote in elections should be applied in the same manner to the right to vote in a referendum."

The passage at page 1048 was cited with approval by Sachs J in August. The passage at page 1050 was cited with approval in Libman.

[64] In Libman the issue was not voting but concerned the constitutional validity of provisions in the Referendum Act channelling funding and campaign expenditure through national committees. The appellant challenged the legislation on the basis that if he wished to conduct a referendum campaign independently of the national committees, his freedom of political expression would be limited to unregulated expenses; whereas if he wished to incur regulated expenses he would have to join or affiliate himself with one of the national committees. He brought his challenge on the basis of section 2(b) of the Charter (freedom of expression) and also under section 2(d) (freedom of association). His challenge to the legislation was successful under both heads. So far as concerned freedom of expression, the appellant wished to express his opinions on the referendum question and convey meaning independently of the national committees by means of regulated expenses. That was a form of political expression which was clearly protected by section 2(b). Similar reasoning applied to the argument under section 2(d). The case was cited to me for the following passage in para 46 of the judgment of the court:

"46. Although the referendum system is different from the electoral system, in that the popular vote concerns a specific question and is not necessarily binding on the government whereas in an election the people vote to elect their political representatives for a specific mandate, the same principles underlying election legislation should in general be applicable to referendum legislation (see Cory J's comments on the subject in Haig v Canada [1993] 2 SCR 995 at p.1050). There are enough points of similarity between the two systems to draw such a parallel. Both involve setting up a procedural structure allowing for public discussion of political issues essential to governing the country or province and designed to ensure that the majority principle is adopted. In both elections and referendums, voters can freely express their choice after being informed of the issues during the election or referendum campaign, as the case may be."

The reference to Cory J's comments is a reference to the second quotation from his judgment in para [63] above. It is unnecessary to look at this case in more detail.

[65] The third of the Canadian cases is Sauvé. This concerned the right of a prisoner to vote. Section 51(e) of the Canada Elections Act denied the right to vote in an election to every person "who is imprisoned in a correctional institution serving a sentence of two years or more". It was argued successfully that that provision was unconstitutional as contravening section 3 of the Charter. Section 1 of the Charter made it clear that the rights and freedoms set out in the Charter were not unfettered but were guaranteed "only to such reasonable limits prescribed by law as can be demonstrated be justified in a free and democratic society". The majority of the court held that, to justify the infringement of section 3 of the Charter, it had to be shown that the infringement achieved a constitutionally valid purpose or objective, and that the chosen means were reasonable and demonstrably justified, i.e. invoking what in ECHR terms would be called "proportionality". Deference to the social and political aims sought to be achieved by the government was not appropriate where the decision was concerned with limiting fundamental rights. I was referred to a number of passages in the judgment of McLachlin CJ giving the judgment of the majority. At para 9, having considered the argument in favour of deference being owed to Parliament's judgement of what infringements were appropriate, she said this:

"9. I must, with respect, demur. The right to vote is fundamental to our democracy and the rule of law and cannot be lightly set aside. Limits on it require not deference, but careful examination. This is not a matter of substituting the Court's philosophical preference for that of the legislature, but of ensuring that the legislature's proffered justification is supported by logic and common sense."

I was taken to a number of other passages which I need not quote in full. For example, in para 13, she said that while deference may be appropriate on a decision involving competing social and political policies, "[it] is not appropriate ... on a decision to limit fundamental rights." She went on to describe the right of citizens to vote as "one of the most fundamental rights guaranteed by the Charter". In para 14 she referred to Charter rights as being not a matter of privilege or merit but a function of membership in the Canadian polity "that cannot lightly be cast aside." That, she said, was "manifestly true of the right to vote, the cornerstone of democracy ...". Later, in para 27, she contrasted the "vagueness of the government's justificatory goals" with the "centrality of the right to vote to Canadian democracy [and] the rule of law ...." In para 58 she went on to say, in the context of addressing proportionality, that denial of the right to vote to serving prisoners:

"undermines the legitimacy of government, the effectiveness of government, and the rule of law. It curtails the personal rights of the citizen to political expression and participation in the political life of his or her country. It countermands the message that everyone is equally worthy and entitled to respect under the law - that everybody counts [...]. It is more likely to erode respect for the rule of law than to enhance it, and more likely to undermine centre singles of deterrence and rehabilitation than to further them."

At para 59 she added that the government's plea of no demonstrated harm to prisoners rang hollow "when what is at stake is the denial of the fundamental right of every citizen to vote." I need not refer to any other passages in the judgement.

[65] August, a decision of Sachs J sitting as a judge of the Constitutional Court of South Africa, also concerned the voting rights of prisoners in a general election. The legislation declared that no person was entitled to vote if detained in a prison after being convicted and sentenced of various serious offences. The Constitution provided for universal adult suffrage and guaranteed that every adult citizen had the right to vote in elections for any legislative body established in terms of the Constitution; but it did not provide for any power to exclude prisoners from voting. It was held, in effect, that prisoners should be entitled to register as voters on the voters' roles and vote in the election, provided they were otherwise eligible in terms of citizenship, mental capacity, etc.. At para [17], Sachs J said this:

"[17] Universal adult suffrage on a common voters roll is one of the foundational values of our entire constitutional order. The achievement of the franchise has historically been important both for the acquisition of the rights of full and ineffective citizenship by all South Africans regardless of race, and for the accomplishment of an all-embracing nationhood. The universality of the franchise is important not only for nationhood and democracy. The vote of each and every citizen is a badge of dignity and of personhood. Quite literally, it says that everybody counts. In a country of great disparities of wealth and power it declares that whoever we are, whether rich or poor, exalted or disgraced, we all belong to the same democratic South African nation; that our destinies are intertwined in a single interactive polity. Rights may not be limited without justification and legislation dealing with the franchise must be interpreted in favour of enfranchisement rather than disenfranchisement."

In a footnote, the judge noted that the point had been strongly emphasised in Canada in the judgment of Cory J in Haig and by the court in Sauvé.

[66] A similar approach was taken by the High Court of Kenya, consisting of five judges, in Kanyua, which was concerned with the right of prisoners to vote in Presidential and National Assembly elections. The court held that prisoners who were of sound mind and above the voting age and had not committed electoral offences were entitled to vote. The following passage appears in the judgment of the court, expressing the same sentiments as in the Canadian and South African cases:

"On the balance of proportionality we hold that there is no legitimate governmental objective or purpose that would be served by denying the inmates the right to vote in a referendum. The Njoya Case has demonstrated that the people's constituent rights to vote in a referendum is a basic human right. A right that ushers in or refuses to usher in a new Constitution. A constituent power higher than the Constitution and the National Assembly and Presidential Election Act. Can the Constitution and the National Assembly and Presidential Act Cap 7 prevent inmates from taking part in a referendum if the prisoners are deemed to be part of the people? In our view they cannot."

[67] Finally I should mention the decision of the High Court of Australia in Roach. It concerned an amendment to the Commonwealth Electoral Act 1918. Before then, there had been a number of changes to the eligibility of prisoners to vote, but as at the date of the amendment the position was that prisoners serving a sentence of three years or longer were not entitled to be on the electoral register or to vote, while those serving a sentence of less than three years were entitled to remain enrolled (or apply for enrolment) and vote. The amendment was to the effect that all prisoners serving a sentence of full-time detention were disqualified from voting, though they might remain on the electoral roll. In effect, the amendment introduced a blanket ban on convicted prisoners while serving a sentence of imprisonment. The High Court held that this blanket ban was unconstitutional. Gleeson CJ, in the course of his judgment, referred to the decision of the Canadian Supreme Court in Sauvé as well as to that of the Strasbourg court in Hirst, but made the point (in para 17) that there was a danger of "uncritical translation" of the concept of proportionality from one legal system to another. Nonetheless, he found aspects of the reasoning in the Canadian and Strasbourg cases "instructive". Gummow, Kirby and Crennan JJ, who with Gleeson CJ formed the majority in favour of declaring the amendment to be invalid, gave a single judgment, deciding the case, in effect, on grounds of proportionality. Their approach can be found in the following passages from paragraphs 81 to 95:

"81. Voting in elections for the Parliament lies at the very heart of the system of government for which the Constitution provides. This central concept is reflected in the detailed provisions for the election of the Parliament of the Commonwealth in what is otherwise a comparatively brief constitutional text.

...

85. The question with respect to legislative disqualification from what otherwise is adult suffrage (where 18 is now the age of legal majority throughout Australia) thus becomes a not unfamiliar one. Is the disqualification for a 'substantial' reason? A reason will answer that description if it be reasonably appropriate and adapted to serve an end which is consistent or compatible with the maintenance of the constitutionally prescribed system of representative government. When used here the phrase 'reasonably appropriate and adapted' does not mean 'essential' or 'unavoidable'. Rather, ... in this context there is little difference between what is conveyed by that phrase and the notion of 'proportionality'. What upon closer scrutiny is disproportionate or arbitrary may not answer to the description reasonably appropriate and adapted for an end consistent or compatible with observance of the relevant constitutional restraint upon legislative power.

...

93. The [amendment] treats indifferently imprisonment for a token period of days, mandatory sentences, and sentences for offences of strict liability. It does not reflect any assessment of any degree of culpability other than that which can be attributed to prisoners in general as a section of society. ...

94. The Solicitor-General of the Commonwealth accepted that, for example, manslaughter is a striking example of an offence which involves an extensive range of moral culpability down to little more than negligence; this may be reflected in the term of the sentence imposed. He responded that the [amendment] operated with a valid degree of precision by limiting the period of disqualification to that for which the law provided incarceration. The difficulty with that proposition is the scope thereby provided for the particularly capricious denial of the exercise of the franchise.

95. The legislative pursuit of an end which stigmatises offenders by imposing a civil disability during any term of imprisonment takes [the amendment] beyond what is reasonably appropriate and adapted (or 'proportionate') to the maintenance of representative government. ..."

[68] There is no doubt that these cases assert in no uncertain terms the fundamental or constitutional importance in a democracy of the right to vote. There can be little quarrel with this. However, they are all set in the context of the specific constitutional arrangements of the country in question. In those constitutional arrangements, the right to vote is enshrined, and the important issue for the court in each case, as I understand it, was to see to what extent a restriction on such a right could be justified. Each case therefore involved a consideration of the doctrine of proportionality. But that is not the issue here, since if it comes to the point that the blanket ban has to be justified as a proportionate means of achieving legitimate social and political objectives, the respondents concede that they cannot succeed, standing the decision of the Grand Chamber in Hirst and the decision of the Supreme Court in McGeoch. The issue here is a different one, namely whether the right to vote can be said to be a common law constitutional principle in the United Kingdom, so that without clear authority from the United Kingdom Parliament the Scottish Parliament cannot restrict the franchise in the Independence Referendum save to the extent that such restrictions are justified and proportionate. Therefore the focus of the enquiry is not whether the blanket ban is proportionate, but whether there is at common law a constitutional right to vote with which the blanket ban interferes. In that context, though the Commonwealth cases are a necessary and forthright reminder of the importance attached to the right to vote in a democratic society, they do not help fix it as a common law constitutional principle in the United Kingdom so as to have this effect on the power of the devolved legislature in Scotland to impose a blanket ban on prisoners voting in the Referendum.

[69] If there is such a constitutional right in some specific way so as to have this effect, it must be clearly founded in the common law. Having considered the arguments advanced by the petitioners, though I accept the existence of a fundamental or constitutional right to vote in general terms, I have come to the conclusion that that right does not extend to voting in a referendum. I have come to this conclusion for three main reasons.

[70] The first reason is that I cannot find any basis for saying that a common law right to vote has developed separately from the legislation which has, over the years, enshrined within the unwritten United Kingdom constitution the modern participatory representative democracy. Consideration of the history of the way in which the franchise has been extended over the centuries does not suggest that the legislative process has been underpinned by a notion, so embedded as to amount to a constitutional principle, that everyone has a right to vote. In so far as the 17^th century case of Ashby is now held up as supporting the idea of a constitutional right to vote - and Lord Rodger in Watkins at paras 49-57 explains convincingly why it should not be so understood - it does not support the theory that that constitutional right is a right belonging to everyone. That is obvious from its time and place, and specifically from the limited franchise at the time it was decided. Nor can it apply to a vote in a referendum, a thing unheard of at the time. Nor, so far as I am aware, was the right of everyone to vote a theory which drove the extension of the franchise in legislation in the 19^th and 20^th centuries. The various Acts of Parliament under which the franchise was extended were enacted not against a background of an accepted constitutional right to universal suffrage but rather as a result of political calculation, sometimes principled, sometimes not, sometimes simply a pragmatic response to extra-Parliamentary pressure. It seems to me that the correct analysis is that, on this question, the common law has developed in step with domestic legislation. I do not accept the respondent's argument, under reference to the remarks of Sir John Donaldson MR in Hipperson v Newbury Electoral Officer [1985] 1 QB 1060 at 1067, that the right to vote is a creature of statute, as readily taken away as conferred. It seems to me that the common law has developed to recognise the right to vote as a fundamental or constitutional common law right. But it has done so alongside and in step with the statutory extension of the franchise and with the incorporation into domestic law of the ECHR, particularly A3P1. It has now reached the point where it can be described as constitutional, so much so that in the unlikely event of parliament now seeking to legislate to restrict it severely or to abolish it entirely, the question would again be raised as to whether there were any limits to the doctrine of parliamentary sovereignty. But that is quite different from saying that there is (and presumably has been for some time) a constitutional right of universal suffrage for all purposes, in referendums as well as elections to the legislature, and in terms of which the franchise may only be restricted so far as is necessary and proportionate in furtherance of legitimate concerns and aims.

[71] The second reason is grounded in principle. If there is a constitutional right to vote in the United Kingdom to the effect that everyone is entitled to vote subject to restrictions which are both proportionate and in pursuit of a legitimate objective, what is the extent or content of that right? As is clear from the Commonwealth authorities to which I have referred, in many democracies there is a constitutional right to vote for the legislature as a whole. But, as I have already pointed out, in the United Kingdom the right to vote is more limited. The legislature consists of two Houses of Parliament, only one of which is elected. If the right to vote is a constitutional principle of the common law, existing without reference to the parliamentary legislation giving effect to it, how can this be justified? By the same token, the right to vote in a representative democracy finds expression in the right to vote for the elected representatives to be sent to the legislature. The principle of representative democracy, with the necessary qualification in the United Kingdom by reference to the position of the House of Lords, is that the elected legislature decides on the issues of the day and passes such legislation as it considers appropriate. Members of the public have no separate right, constitutional or otherwise, to vote directly on the issues of the day, however important they may be, save to the extent that that right is given to them by their elected representatives in parliament. In my opinion, any constitutional right to vote is, in principle, limited to the right to vote for the members of parliament. In Scotland it extends, by virtue of the Scotland Act, to voting for members of the Scottish Parliament. The franchise for electing members to the Scottish and United Kingdom Parliaments is fixed by legislation passed by the United Kingdom Parliament (section 3 of the RPA 1983). The constitutionality of the United Kingdom Parliament's blanket ban on prisoners voting in parliamentary elections (as opposed to its compatibility with A3P1) has not been challenged. Nor has the legitimacy of the Scottish and United Kingdom Parliaments elected on such a franchise, or the legislation passed by them. It seems to me that the constitutional right to vote, being a right to vote for elected members of the legislature, has been satisfied by the legislation affording the right to vote in those parliamentary elections, at least until the next such elections. Once elected, it is up to the two parliaments to determine what legislation to pass. If, on a major issue, one or both parliaments decides to hold a referendum, that decision is a decision taken by the parliament or parliaments as the representative legislatures chosen by the people. It can do so on whatever basis and on whatever terms it chooses. On what basis can it then be said that there is a further constitutional right for everyone to vote in the referendum? If there was no constitutional right to insist on a referendum, why should there be a constitutional right to vote in a referendum if the elected parliament decides to hold one?

[72] The third reason is this. The constitutional right to vote which is argued for by the petitioners is, as I have said, a right for everyone to vote unless excluded by restrictions on the franchise which are both proportionate and in pursuit of legitimate aims and objectives. It is said, on the basis both of principle and of the Strasbourg jurisprudence, that a blanket ban on prisoners voting is not and would not be a lawful restriction on the basic right of everyone to vote. But that is the position which the United Kingdom Parliament has adopted both for elections to that parliament and for elections to the Scottish Parliament. In those circumstances I am being asked to hold, in effect, that the United Kingdom Parliament in enacting section 3 of the RPA 1983 and, no doubt, in enacting previous legislation to a similar effect, has been acting unconstitutionally. And I am being asked to decide this not on the basis of A3P1 or any other part of the ECHR, but as a matter of common law independent of that. Standing my views as to the development of common law, which I have expressed earlier, this seems to me to be an untenable position.

[73] For all these reasons I reject the petitioners' first and second common law challenges.

Contrary to UK international obligations

[74] The third ground of challenge at common law is that the imposition of the blanket ban by the Franchise Act is contrary to certain international obligations of the United Kingdom. The argument is that the Scotland Act, which set up the Scottish Parliament and gave it its powers, should be construed consistently with the United Kingdom's international obligations. It created a legitimate expectation that the Scottish Parliament would not act contrary to international obligations undertaken by the United Kingdom. On that basis, the Scottish Parliament has no power to act contrary to such international obligations; alternatively any legislation which it passes which is inconsistent with such international obligations is contrary to the legitimate expectations created by the Scotland Act and is capable of being judicially reviewed. Alternatively it was submitted under reference to Padfield v Minister of Agriculture, Fisheries and Food [1968] AC 997, that acts of public bodies such as the Scottish Parliament were subject to review on the ground that they run counter to the policy and objectives of the enabling legislation, which policy and objectives are identified by construing the legislation as a whole.

[75] The international obligations in question here are the obligations set out in the International Covenant on Civil and Political Rights 1966 ("ICCPR"), to which I have referred earlier, which was ratified by the United Kingdom in 1976. As I have explained, article 25 has been interpreted as conferring on the citizens of signatory states a right to vote in a referendum as well as in elections to legislative bodies. That being the case, the petitioners argue that any restriction on the right to vote must be proportional and in pursuit of some legitimate aim or objective, a test which, if applicable, the respondents accept that they must fail.

[76] It is, of course, well established that in construing domestic legislation, and more generally in considering the extent of domestic legal rights, the court will have regard to international treaties or other international obligations to which the United Kingdom has subscribed. This point arose frequently before the enactment of the Human Rights Act 1998, where the court would be influenced by the ECHR in interpreting ambiguous statutory provisions, in exercising its discretion and sometimes developing the common law: see, for example, R v Lyons [2003] 1 AC 976, per Lord Bingham at para 13 and per Lord Hoffman at para 27. The same approach has been applied in respect of the Aarhus Convention: see Forbes v Aberdeenshire Council [2010] CSOH 1 at para 11. Indeed there are dicta to the effect that rules of international law are rules of Scots law: see, for example, Lord Advocate's Reference No 1 of 2000 2001 JC 143 at paragraphs 23-24. But this line of authority only goes so far. It does not allow the courts to ignore the will of Parliament and amend, repeal or strike down unambiguous legislation.

[77] There is, however, no ambiguity in the provisions of the Franchise Act of which the petitioners complain. That is absolutely clear: there is a blanket ban on prisoners voting. What is relied on by the petitioners to counter this is the enabling legislation, in other words those provisions of the Scotland Act which provide for the legislative competence of the Scottish Parliament. The argument is that where that enabling legislation is silent, the court may fill the gap and hold that the United Kingdom Parliament cannot or should not act in a manner contrary to the relevant international obligation, here the obligations contained in the ICCPR, or hold that it would be contrary to the legitimate expectation created by the Scotland Act that it should not do so.

[78] Section 29 of the Scotland Act provides for the "legislative competence" of the Scottish Parliament. I have already discussed this in connection with incompatibility with Convention rights. Apart from Convention rights and EU law, no other international obligations are referred to in that section. But the Scotland Act does contain other provisions which deal with international obligations undertaken by the United Kingdom and their effect on the ability of the Scottish Parliament to legislate. Section 35 of the Act is headed "power to intervene in certain cases". It provides in subsection (1)(a) that if a Bill contains provisions "which the Secretary of State has reasonable grounds to believe would be incompatible with any international obligations ..." he may make an order prohibiting the Presiding Officer of the Scottish Parliament from submitting the Bill for Royal Assent. Section 58(1) provides that if the Secretary of State has reasonable grounds to believe that any action proposed to be taken by a member of the Scottish Government would be "incompatible with any international obligations" he may by order direct that the proposed action shall not be taken. The same applies mutatis mutandis to subordinate legislation of the Scottish Parliament: see section 58(4). International obligations are defined in section 126(10) of the Act, and clearly would include the ICCPR.

[79] It was argued for the petitioners that these references to international obligations strengthened their case that the Scottish Parliament was not entitled to legislate in a manner which was contrary to international obligations undertaken by the United Kingdom or, at least, that such legislation would be contrary to the legitimate expectation created by the Scotland Act and/or contrary to the purpose for which the Scottish Parliament was created.

[80] I cannot accept this argument. Indeed, I take the opposite view. It seems to me that whatever might have been the position in the absence of references in section 35 and 58, the argument is unstateable given the terms of those sections. Those sections do make it clear that the United Kingdom has an interest in whether Acts of the Scottish Parliament, or acts of the Scottish Government, conform to the United Kingdom's international obligations. But they do not make incompatibility with such international obligations a ground for challenge other than by the Secretary of State. In terms of those sections, read with section 29, it is not a bar to the legislative competence of the Scottish Parliament that the Act which it proposes to pass would be incompatible with international obligations undertaken by the United Kingdom. On the contrary, it is within the legislative competence of the Scottish Parliament to pass such Acts. If it does so, or proposes to do so, the Secretary of State can decide whether or not to make an order prohibiting the Presiding Officer from submitting the Bill for Royal Assent. But the Secretary of State need not take any action at all and may decide not to. It would be odd if the court had the power to intervene in those circumstances by judicial review when the Secretary of State, representing the United Kingdom government, had taken the view that it did not wish to stop the Scottish Parliament passing such legislation.

[81] A similar argument was advanced and rejected in Whaley v Lord Advocate 2008 SC HL 107 at paras [8]-[9]. Lord Hope expressed his conclusion in this way (at para [9]):

"What these provisions do is enable the Secretary of State, who is a minister of the UK government, to intervene if he thinks it appropriate to do so in the interests, for example, of international comity. They do not limit the legislative competence of the Scottish Parliament in a way that can be decided upon by a court."

It was suggested that the argument dealt with in Whaley was a different argument to that advanced by the petitioners in this case, in that it focused on the legislative competence of the Scottish Parliament in terms of the Scotland Act rather than the implied restriction on the right to act "unconstitutionally" without express power to do so being given by the United Kingdom Parliament. I accept that to some extent, but it seems to me that the same answer can be given to both submissions. The terms of sections 35 and 58 make it clear that the intervention by the Secretary of State in a case where he or she thinks it appropriate to intervene is the only restriction on the Scottish Parliament in this respect.

[82] Accordingly, I reject this third, common law ground of challenge.

Incompatibility with EU law

[83] As has already been noted, section 29(1) of the Scotland Act provides that an Act of the Scottish Parliament is not law so far as any provision of the Act is outside the legislative competence of the Parliament. Subsection (2)(d) provides that a provision is outside that competence if it is incompatible with EU law.

[84] The petitioners here contend that EU law is engaged because the outcome of the Independence Referendum will affect Scotland's membership of the EU and the status of Scottish nationals as citizens of the EU. The Referendum is therefore not simply a domestic matter. It affects the EU rights of Scottish nationals entitled to vote in the Referendum in two ways: first, because an independent Scotland would not, for a while at least, be a member of the EU, and therefore individuals resident in Scotland would not have access to EU rights for the period when Scotland was not a member state; and secondly, because Scottish citizens, not being citizens of a member state, for a while at least, would for that period cease to be EU citizens. The points are separate but obviously closely related. The first of those points, namely that Scotland would cease for a while at least to be a member state of the European Union is predicated on the proposition that Scotland as an independent state would have to apply for membership from outside the EU, and unless and until she became a member she would remain outside the EU. The second point is based on an interpretation of article 20 of the Treaty on the Functioning of the European Union ("TFEU"). That article establishes citizenship of the Union and provides that "every person holding the nationality of a Member State shall be a citizen of the Union". It goes on to say that citizenship of the Union is additional to and does not replace national citizenship. The argument is that upon the Scotland attaining independence and becoming an independent state outside the EU and applying to get in, Scots who had previously been nationals of the United Kingdom but had become Scottish nationals would no longer be nationals of a Member State and, as a result, would lose their EU citizenship. The argument assumes that those becoming Scottish nationals would be required to give up their United Kingdom nationality.

[85] I asked counsel for the petitioners whether their argument depended upon it being established that, in the event of a "Yes" vote leading to independence, Scotland would no longer be a member state of the EU, and the petitioners, as the Scottish nationals, would lose their EU citizenship and rights, or whether it was sufficient for his argument that there was a possibility of those things happening. Neither the three petitions nor the Common Note of Argument for the petitioners made it entirely clear whether the petitioners were saying that this would happen or simply that it might happen. Counsel for the respondents intervened, in my view quite understandably, to say that they had interpreted the petition as raising the same point as had been rejected by the Supreme Court in McGeoch; but, perhaps more importantly, had not understood the petition to be seeking a judicial determination of the question of Scotland's continued membership of the EU on attaining independence. The only materials before the court on this question were the Opinion by Professors Crawford and Boyle entitled "Opinion: Referendum on the Independence of Scotland - International Law Aspects", which was commissioned by the United Kingdom Government; and a letter dated 10 December 2012 from the President of the European Commission, a copy of which was lodged in process. One would expect a full argument about such important matters of international law to be conducted on the basis of rather fuller materials. Counsel for the respondents made it clear that if the court was being asked to determine that question, he would require much longer to place the appropriate materials before the court.

[86] In the circumstances, I decided to continue hearing the argument on the basis that if I considered I needed to reach a decision on the question of whether Scotland's independence would mean that it ceased for some time to be a member of the EU, and that in consequence Scottish nationals ceased for some time to be citizens of the EU, I would put the case out By Order for further argument on this question. It seemed to me more likely that the petition could be answered by assuming, without deciding it and without requiring the respondents to deal with the point in detail, that the petitioners had shown that there was at least a doubt as to Scotland's continued membership and the consequences in terms of the loss of EU citizenship for Scottish nationals. I have come to the conclusion that that is a course that I can safely adopt without any unfairness to either party and I have decided that there is no need, therefore, to seek further assistance on the legal point involved. In effect, I have decided to determine the matter as though it were a debate on the relevancy of the petitioners' averments on this point, making the assumption in their favour for the purpose of these proceedings only - and I emphasise without deciding the issue - that independence would lead to a cessation of EU membership and citizenship, albeit possibly only temporarily. This course can cause no prejudice to the petitioners' argument and avoids the need for lengthy further argument on matters of international law. It is apparent from the Crawford and Boyle Opinion that in legal terms there is no simple answer, either in law or in fact. They correctly describe the legal situation as sui generis, pointing out that the various EU Treaties do not specifically cover the point. But the difficulty goes further than that. As is apparent from the terms of their Opinion, and as is perhaps obvious to anyone with even a passing acquaintance with the arguments for and against, the decision on continued membership will not ultimately be decided solely as a legal question but will, to a greater or lesser extent, involve questions of hard politics. The court is not in a position to know fully what political considerations will be brought to bear on the issue, and with what leverage. Accordingly, to ask the court to decide the question whether upon achieving independence Scotland would cease for a time to be a member of the EU, with consequences for its nationals in terms of their citizenship of the EU, would be to ask the court to predict the outcome of robust and complex negotiations. That is a question largely of fact. It is not a task which the court is equipped.

[87] I should simply add that during the course of the hearing before me the Scottish Government produced its White Paper entitled "Scotland's Future: your guide to an independent Scotland" (November 2013 ISBN: 978-1-78412- 068 -9). In his Speaking Note provided to assist his submissions, Mr O'Neill QC, for the petitioners, quoted extensively from that document and made a number of representations about the position of the Scottish Government as set out in the White Paper. For the reasons given above, I do not propose to go into any of those points. However, in fairness to the respondents I should say that in certain material respects Mr O'Neill appeared to have misunderstood and as a result misstated the position of the Scottish Government as set out in that document.

[88] I therefore approach the matter on the basis that if there were to be a "Yes" vote on the question posed in the independence referendum, that would lead to a process of negotiation both with the United Kingdom Government and with the EU and, no doubt, various individual member states of the EU. Assuming the negotiations with the United Kingdom Government resulted in there being agreement on all the prerequisites of independence, and agreement on a timetable for progress towards independence, I have to assume that there might come a time when Scotland as an independent state was outside the EU and applying to join. In those circumstances, the possibilities are practically endless. For example, there might well, in the unique circumstances envisaged, be some temporary arrangement under which Scotland was afforded some quasi-EU rights and status pending the conclusion of negotiations; and Scottish nationals might either retain their UK citizenship (acquiring dual citizenship) or, as former EU citizens, be granted some continuation of that status until the negotiations were concluded one way or the other. I do not know. That is pure speculation. But, for the purpose of dealing with this issue, I have to assume, as the petitioners contend, that the eventual outcome might be that Scotland was outside the EU without any temporary "protection" and its citizens were no longer EU citizens. All that will depend upon negotiations, but that is what I must assume for the purpose of addressing the issue raised by the respondents.

[89] In those circumstances the petitioners rely upon the decision of the European Court of Justice in Rottman v Freistaat Bayern [2010 [QB 761. In that case the applicant, an Austrian national, applied for and was granted naturalisation as a German national and, in consequence, lost his Austrian nationality. Subsequently it transpired that he had obtained his German nationality by deception, and the German authorities decided to withdraw his naturalisation with retroactive effect. Under Austrian law he could not revert immediately to his Austrian nationality. He applied to have the decision by the German authorities annulled, on the grounds that loss of his German nationality would render him stateless. A reference was made to the European Court of Justice on the question whether it was contrary to EU law for a member state to withdraw from a citizen of the Union the nationality of state acquired by naturalisation and obtained by deception inasmuch is that withdrawal deprived the person concerned of the status of citizen of the Union by rendering him stateless. The European Court of Justice held that a member state's power to lay down the conditions for acquisition and loss of nationality had to be exercised with due regard to European Union law; and therefore the exercise of that power, in so far as it affected the rights conferred protected by the legal order of the European Union, was amenable to judicial review carried out in the light of EU law. It went on to decide that withdrawal of German nationality was not necessarily contrary to EU law provided that the decision observed the principle of proportionality. On this basis the matter was remitted to the German courts. In the course of its judgment, the court made the following observations:

"41 ... the fact that a matter falls within the competence of the member states does not alter the fact that, in situations covered by European Union law, the national rules concerned must have due regard to the latter ...

42 It is clear that the situation of a citizen of the Union who ... is faced with a decision withdrawing his naturalisation ... and placing him ... in a position capable of causing him to lose the status conferred by article 17EC and the rights attaching thereto falls, by reason of its nature and its consequences, within the ambit of European Union law.

43 As the court has several times stated, citizenship of the Union is intended to be the fundamental status of nationals of the member states ...

44 Article 17(2)EC attaches to that status the rights and duties laid down by the Treaty, including the right to rely on article 12EC in all situations falling within the scope rationae materiae of Union law ...

45 Thus, the member states must, when exercising their powers in the sphere of nationality, have due regard to European Union law ...

46 In those circumstances, it is for the court to rule on the questions referred by the National court which concern the conditions in which a citizen of the Union may, because he loses his nationality, lose his status of citizen of the Union and thereby be deprived of the rights attaching to that status.

47 In this regard, the National court essentially raises the question of the proviso formulated in the court's case law cited in para 45 above, to the effect that the member states must, when exercising their powers in the sphere of nationality, have due regard to European Union law, and also a question of the consequences of that proviso in a situation such as that in the case in the main proceedings.

48 The proviso that due regard must be had to European Union law does not compromise the principle ... that the member states have the power to lay down the conditions for the acquisition and loss of nationality, but rather enshrines the principle that, in respect of citizens of the Union, the exercise of that power, in so far as it affects the rights conferred or protected by the legal order of the Union, as is in particular the case of a decision withdrawing naturalisation such as that at issue in the main proceedings, is amenable to judicial review carried out in the light of European Union law. ...

55 In such a case, it is, however, for the national court to ascertain whether the withdrawal decision at issue in the main proceedings observes the principle of proportionality so far as concerns the consequences it entails for the situation of the person concerned in the light of European Union law..."

[90] As is apparent from the above passages, the decision under review in Rottman was a decision which directly affected the retention by the applicant of his EU citizenship and all that that entailed. As the respondents pointed out in argument, that is far removed from the circumstances of the present case where all that can be said is that a "Yes" vote in the independence referendum will set in train a process of argument and negotiation at the end of which a decision may be taken which affects Scotland's membership of the EU and the status of the petitioners and others as EU citizens. In other words, whereas the decision impugned in Rottman was a decision directly affecting the applicant' s EU citizenship and therefore his EU rights, the decision impugned here, namely the blanket exclusion of prisoners from the franchise in the independence referendum, has no such effect. Nor does the referendum itself pose the question whether Scotland should leave the EU or its nationals lose their EU citizenship. The independence referendum relates to the internal affairs of the United Kingdom. What may follow upon the result of the referendum is a matter one or more steps removed from that.

[91] In the course of argument I was referred to a number of cases more often cited in an immigration and asylum context, namely Zambrano v Office National de l'Emploi [2012] QB 265, Dereci v Bundesministerium Fur Inneres [2012] 1 CMLR 45 and Harrison v Secretary of State for the Home Department [2013] 2 CMLR 23. I referred to these cases in my Opinion in MIK Ptr [2013] CSOH 176. Those cases suggest, as I put it at paragraph 32 of my Opinion in that case, that "unless an EU national is forced to leave the EU, then EU law is not engaged". Applied to the present case, it might be said that that this lends force to the argument that for EU law to apply there has to be an established factual link with a matter covered by EU law, here the alleged potential loss of EU membership and/or EU citizenship. Even putting the petitioners' case at its highest, I do not accept that that factual link has been established.

[92] In short, therefore, I accept the respondent's argument that, by enacting the Franchise Act, the Scottish Parliament is not exercising competence in the sphere of nationality. It is not purporting to make a decision about EU membership or EU citizenship. The process which it is putting in place by the independence referendum is not a process which will have any direct impact on the question of EU membership or EU citizenship. The point may arise in the future where decisions are taken which might affect those questions. But that time has not yet come.

[93] I should revert briefly to the decision of the Supreme Court in McGeoch. A number of EU arguments were run in that case. All were rejected. Mr O'Neill QC also appeared as counsel for Mr McGeoch. He emphasised to me, and I accept that this is correct, that the EU arguments rejected in McGeoch were different from those advanced here. Nonetheless, Lord Mance's general discussion of the European law issues at paras 43-58 is of interest. In rejecting the "wider submission" advanced by counsel for Mr Chester, the other applicant in the case, he considered aspects of the interrelationship between the scope and conditions of the right to vote under the European Treaties and the principles established in the Strasbourg jurisprudence in cases such as Hirst and Scoppola. He concluded at para 58 that the decisions of the European Court of Justice did not import the Strasbourg jurisprudence into the general provisions of Community and Union law about voting in European Parliamentary elections. He pointed out that eligibility to vote under the Treaties is a matter for national parliaments. He added this:

"There is no sign that the European Commission has ever sought to involve itself in or take issue with voting eligibility in member states or specifically with the restrictions on prisoner voting which applied in a number of such states. The Strasbourg jurisprudence operates as the relevant control, albeit one that has itself proved in some respects controversial. It would not only unnecessarily duplicate that control at the European Community or Union level, it could also lead to further conflict and uncertainty. ... Further, even in the form into which they have been shaped by the Treaty of Lisbon, it is notable that such provisions as the European Treaties contains concerning individual voting rights are notably limited in scope. They relate to the core Treaty concerns of equality between nationals or Union citizens and freedom of movement within the European Union ...."

[94] That passage is instructive. Even if I had found that EU law was engaged, on the basis that the independence referendum was a referendum directly affecting the question of EU membership and EU citizenship, I would not as a result necessarily have concluded that the franchise arrangements put in place by the Scottish Parliament in the Franchise Act contravened EU law. As was pointed out in argument, the issue of proportionality mentioned in Rottman was proportionality in the result rather than proportionality in the process. The issue before me is as to process; and I am not persuaded that EU law would necessarily replicate the Strasbourg jurisprudence on the question of a blanket ban on prisoners, let alone go further than the Strasbourg courts have gone to date and apply the principles underlying that jurisprudence to the question of voting in a referendum rather than simply voting in elections for a legislature.

[95] I therefore reject the petitioners' argument based upon incompatibility with EU law. I have considered whether I should make a reference in this case for a preliminary ruling by the European Court of Justice. Had the matter turned on the point discussed in paragraph [94] above, I might have done so. But it does not. I have rejected the petitioners' EU law argument on the basis that even taking the most benevolent view of the material put before the court there is no direct link between the independence referendum and any decision as to future membership or citizenship of the EU. That seems to me to be clear. I do not consider that there is any doubt about the matter. In those circumstances, where the matter is acte clair, not only does the court not have to refer it but it should not refer it. I have therefore decided not to make a reference.

Disposal

[96] For the reasons set out above, the petitions must be refused. At the end of the hearing it was agreed that if I was in favour of the petitioners' argument in any respect I should put the case out By Order so that the details of the interlocutor could be considered. Since I have rejected their arguments there is no need for me to take this course. I shall therefore simply refuse the petitions in each case. Meanwhile I shall reserve all questions of expenses.

[97] Finally, I would like to thank counsel for their very full and helpful submissions. In particular, I am grateful for the detailed notes of argument submitted by both parties and for the careful and methodical approach adopted by junior counsel on both sides in laying out the materials and arguments. This enabled the submissions by senior counsel to be relatively concise. I have not referred to all the authorities cited to me. Should the matter go further, most (though not all) of those authorities can be found referred to in the written notes of argument and lists of authorities.