IN November 2022, the UK Supreme Court concluded that the Scottish Parliament did not have the competence to enact legislation to hold a second referendum on Scottish independence (Reference by the Lord Advocate of devolution issues under paragraph 34 of Schedule 6 to the Scotland Act 1998 [2022] UKSC 31, [2022] All E.R. (D) 64 (Nov)). That an independence referendum would relate to matters reserved to the UK Parliament, specifically the Union and the Westminster Parliament, is in many senses the least remarkable aspect of the judgment. It does, of course, have important political ramifications for Scotland and the independence movement. From a legal perspective, the case has consequences for UK constitutional law and the relationship between domestic and international law.
The reference came to the Supreme Court in a novel manner. The Scotland Act 1998 provides a specific mechanism through which the government Law Officers, both those advising Scotland and the UK, may refer a question to the Supreme Court as to whether Acts of the Scottish Parliament are within competence. According to section 33, such references are made within four weeks of the passing of the Bill through the Scottish Parliament, prior to the Bill receiving royal assent.
However, this option was not available as the draft Bill had not yet been laid before the Scottish Parliament. Section 31 of the Scotland Act requires that the person in charge of a Bill states that, in their view, the Bill is within the competence of the Scottish Parliament. The Scottish Ministerial Code requires that any statement must first be cleared by the Scottish Law Officers. However, the Lord Advocate had concluded that, in her opinion, the Bill was not within the competence of the Scottish Parliament. This placed the Scottish First Minister in a quandary. She could either drop the Bill or change the Ministerial Code enabling her to make a statement that the Bill was within the competence of the Scottish Parliament without this needing first to be cleared by the Law Officers. The First Minister, however, found an alternative solution, with the cooperation of the Lord Advocate, who agreed to refer the matter to the Supreme Court under paragraph 34 of Schedule 6 to the Scotland Act. This empowers the Lord Advocate to refer a “devolution issue” to the Supreme Court.
The reference raised three questions. First, could paragraph 34 of Schedule 6 be used to refer an issue as to whether a draft Bill was within the competence of the Scottish Parliament? Second, even if this reference were possible, should the Supreme Court nevertheless exercise its discretion to decline to hear it? Third, did the Scottish Parliament have the power to legislate to hold a referendum on Scottish independence? In addition, the Scottish Nationalist Party (SNP) intervened in the case, arguing that the provisions of the Scotland Act should be interpreted in line with international law, specifically the right to self-determination. Read in this manner, they argued, legislation to hold an independence referendum would be within the competence of the Scottish Parliament.
The Supreme Court concluded that it could hear the reference and that there were no reasons to exercise its discretion not to do so. The Scottish Parliament did not have the competence to hold an independence referendum and this lack of competence did not undermine the right to self-determination as understood in international law. In reaching these conclusions, the Supreme Court reaffirmed current case law relating to statutory interpretation generally and the interpretation of the Scotland Act 1998 specifically. Its judgment also has consequences for devolution and the role of the Law Officers, as well as when courts interpret legislation in a manner that reinforces the rule of law.
Recent decisions of the Supreme Court have referred to what Lord Burrows has christened the “correct modern approach to statutory interpretation” (R. v Luckhurst [2022] UKSC 23, at [23], noted Feldman [2022] C.L.J. 460). This requires courts to interpret legislation according to the natural or ordinary meaning of the words. In reaching this determination, the context and purpose of the legislation is also important. This applies to the Scotland Act 1998 in the same way as any other legislation, confirming yet again – if such confirmation were needed – that “constitutional” statutes (or at least the Scotland Act 1998) are not interpreted differently from “ordinary” statutes.
This approach determined the interpretation of whether the Lord Advocate was raising a “devolution issue”. Paragraph 34 of Schedule 6 of the Scotland Act sets out five specific examples of a devolution issue, before adding a further category of “any other question about whether a function is exercisable within devolved competence or in or as regards Scotland and any other question arising by virtue of this Act about reserved matters”. This section of the paragraph is broadly worded, particularly through the use of the words “any other question”. Read against the context of the other provisions of the paragraph, this section “has the appearance of a sweeping-up provision, designed to supplement the more precise provisions which precede it, so as to ensure that no gap is left” (at [37]). It was not designed, as argued by the Advocate General on behalf of the UK Government, to refer only to narrow additional matters, its scope being limited by earlier sections of the paragraph.
Nor was the Supreme Court persuaded that the broader context of this provision, read against the scheme of legislative scrutiny for Acts of the Scottish Parliament according to sections 31 and 33 of the Scotland Act 1998, would dictate against reading the Scotland Act to permit these references. The context of this scheme of legislative scrutiny would prevent the Lord Advocate from referring a Bill that had been introduced in the Scottish Parliament. It would not, however, prevent the reference of a proposal for a Bill, or a draft Bill, that had not yet been introduced. Whilst recognising that this may give rise to the possibility of two references being brought on the same Bill, this was not a sufficiently strong reason to narrow down the reading of paragraph 34. The wording of the legislation may change between its proposal and its final form. Moreover, the Lord Advocate could be expected to exercise her powers appropriately, avoiding a frivolous second reference of a Bill whose proposal had already been referred to the Supreme Court.
In reaching this conclusion, the Supreme Court was not persuaded by arguments made by the Advocate General, in reliance on Notes on Clauses – now Explanatory Notes – that paragraph 34 was not meant to include a reference as to whether Bills were within the competence of the Scottish Parliament. The Court was not persuaded that this reading of the statute was supported by the Notes on Clauses. Nor were they of a mind to give too much weight to such Notes which are prepared by government officials and have no parliamentary endorsement. This approach to explanatory notes is more cautious than that of the Supreme Court in R (O) v Secretary of State for the Home Department [2022] UKSC 3, [2022] 2 W.L.R. 343, which indicated that there may be some situations in which explanatory notes may cast light on the meaning of statutory provisions. This case also suggested that explanatory notes may reveal ambiguity or uncertainty. However, even in R (O), it was clear that any reference to external sources was secondary to the text and could not displace the wording of a clear statutory provision.
The slightly more cautious approach taken in the Scottish reference case is probably more of emphasis than of degree. It also pays attention to constitutional reality – explanatory notes are not expressions of the legislature and to prioritise this wording over that of the statute could favour the intention of the Government over that of Parliament, as recognised with regard to the use of Hansard post Pepper v Hart [1993] A.C. 593. However, it is to be hoped that this caution does not prevent the use of explanatory notes in other cases where they do illustrate ambiguity in otherwise clear words, particularly when a statutory provision granting power to the executive goes beyond the justification of that power provided in the explanatory notes. To use explanatory notes in this way would uphold, rather than potentially undermine, the proper relationship between the legislature and the executive.
The Advocate General also argued that there was no justification for providing the Lord Advocate and other Law Officers with a power to refer proposed legislation to the Supreme Court in order to determine whether this was within the competence of the Scottish Parliament. Here, the Supreme Court recognised that even Lord Advocates were not infallible. If they were mistakenly to conclude that a Bill was within legislative competence when it was not, this error could be corrected through a reference under section 33 of the Scotland Act. If, however, the Lord Advocate were to mistakenly conclude that a Bill was not within competence, thereby preventing the Bill from being introduced, there would be no means through which this mistake could be corrected. This may inadvertently limit the powers of the Scottish Parliament. Enabling the Lord Advocate to refer a proposed Bill to the Supreme Court, therefore, would have the purpose of upholding the rule of law, ensuring the Scottish Parliament acted within competence and preventing the possibility of an overly cautious Lord Advocate limiting the Scottish Parliament's legislative competence. This approach is to be welcomed and shows a clear understanding of the importance of the role of the Law Officers. It helps to ensure legal opinion, sound though it may be, and indeed was in this case, does not inadvertently become legal fact.
However, despite concluding both that this procedure can be used to refer a question as to the competence of the Scottish Parliament to enact a proposed Bill, and that the court should hear this particular reference, there is no suggestion that the Supreme Court will automatically accede to all references made under this procedure. The Supreme Court was willing to hear this reference given its exceptional nature. The particular issue was one of huge practical importance. It was plain that this was not a moot issue as the Scottish Government would introduce the proposed Bill were it to be found to be within competence. A draft Bill had been produced and it was evident that there was only one clear issue as to whether the proposed Bill would be within legislative competence. Consequently, this did not raise hypothetical or moot issues and could be resolved by the court.
The conclusion that a referendum on independence was beyond the scope of the Scottish Parliament is unsurprising in the context of previous decisions determining when the purpose or effect of legislation “relates to” a reserved matter. The purpose of the Bill was to hold a referendum. The effect of the Bill was not confined to the holding of a referendum. Even a referendum that is not self-executing can have a huge impact on political decision-making – as the Brexit referendum clearly illustrated. Lawfully held referendums are clear expressions of democratic will which carry greater import than opinion polls. The legislation required to hold a national referendum provides democratic oversight of the question asked, the franchise, campaign length and expenditure, and voting rules. It also authorises expenditure and establishes official oversight of voting and counting of votes. These legitimating elements not only distinguish referendums from opinion polls, but also explain why politically advisory referendums have enormous constitutional importance. This meant, in turn, that a Bill for holding a referendum would have more than a loose or consequential connection to the Union and to the powers of the Westminster Parliament. In reaching this conclusion, the Supreme Court relied on impeccable legal reasoning which was nevertheless sensitive to and understanding of political realities.
One of the notable aspects of the decision is the extent to which the international right of self-determination limits the UK Government's ability to resist the Scottish request for a second referendum. As an intervenor, the SNP argued that there is such a right in international law and a presumption that UK law should be interpreted in a way that does not place the UK in breach of international obligations, which together meant that “relates to” in section 29(2)(b) should be given a meaning that would bring a non-self-executing referendum on independence within Scottish competence. The Supreme Court accepted both propositions in general terms but rejected their application in the present case and to section 29(2)(b) in particular.
First, the Court held that the right of self-determination is “not in play here”, avoiding the question whether the Scottish were a “people”. It placed significant reliance on the Supreme Court of Canada's opinion in Reference re Secession of Quebec [1998] 2 S.C.R. 217 that the right of self-determination generates, at best, a right to external self-determination where a people is governed as part of a colonial empire; subject to alien subjugation, domination or exploitation; and potentially where “a people” is denied any meaningful exercise of its right to self-determination within the state of which it forms a part. As with the Quebecois, none of these situations reflect that of the Scottish people. In any case, the UK Supreme Court reasoned that no interpretation of section 29(2)(b) could place the UK in breach of the right of self-determination in international law. Nothing in the allocation of powers in the Scotland Act as between Westminster and Scotland infringes that right: “[o]n the contrary, the legislation establishes and promotes a system of devolution founded on principles of subsidiarity” and it would be “inappropriate to apply any interpretive presumption with the purpose of achieving a greater or lesser devolution of powers” (at [90]).
The decision demonstrates the complex interaction between international law and constitutional law on issues of self-determination. Its relatively light treatment of the right of self-determination is hardly surprising, but not unimportant. Affirmation of the Supreme Court of Canada's opinion provides further support for the view that the right of self-determination may entitle a people to external self-determination outside the colonial context in very limited circumstances. Beyond the above-mentioned situations, the possibility of a right to unilateral secession or independence is (even more) deeply controversial, not least because it threatens the status quo for states and because it goes to the heart of what it means to be a member of the international community. The Kosovo Advisory Opinion (I.C.J. Rep. 2010, p. 403) famously avoided discussing self-determination, concluding instead that Kosovo's unilateral declaration of independence was “in accordance with international law” (this perhaps explains why the UK Supreme Court relies on – and thus ends up deferring to – the UK's submissions to the ICJ during the Kosovo proceedings; exceptionally, the UK supported Kosovo's right to self-determination). In other words, the authors of that declaration – considered to be non-state actors outside the apparatus of the state – were not said to have violated any positive rule of international law (i.e. territorial integrity, right of non-interference or Security Council resolution calling for non-recognition of a territorial unit as a state). As Crawford, acting then as counsel for the UK, famously said in the oral hearings: “I am a devoted but disgruntled South Australian. ‘I hereby declare the independence of South Australia.’ What has happened? Precisely nothing. Have I committed an internationally wrongful act in your presence? Of course not” (CR 2009/32, p. 47). Crucially, the ICJ did not recognise a right to unilateral secession in international law in explicit terms or set out whether/when violation of the right to self-determination might generate a right of external self-determination: that is, the right to choose one's own political form, including independent statehood.
This uncertainty and paucity of authority leaves unresolved the deeply difficult democratic elephant in the room: if the Scottish pro-independence vote represents a robust and stable majority view, then we are left with a constitutional system in which there is devolution on most matters within a structure that allows the central Government to resist the Scottish wish to have a say on a fundamental question of their own future. The limited international jurisprudence that does exist suggests that the right of self-determination in international law extends to a right of external self-determination in very narrow circumstances. For a pro-independence majority in Scotland, the burning question, then, is whether ordinary political representation, a degree of devolution, and respect for rights and so on, are sufficient to overcome the inability of a posited “people” to advance negotiations within the state – within a “voluntary” union – on their own future.
If there is a robust majority view in favour of independence that has not been given effect within constitutional arrangements, should this give rise to a right of external self-determination on the basis that it denies a say in a fundamental question of how a people are governed? Some find this to be an attractive argument, but the difficulty is that it has a somewhat circular logic: any group in any state may assert a right of self-determination and point to the central Government's refusal to entertain that claim as grounds for invoking a right of external self-determination. Moreover, such an argument would likely have controversial political implications for similar ongoing situations (e.g. Catalonia, Ireland) and might be abused (as Russia's reliance on self-determination to justify illegal uses of force demonstrates: Sanger [2022] C.L.J. 217). On the other hand, the Scottish claim is marked out by the fact that it draws strength from two constitutional and historical particularities – namely that, (1) at least in theory, Scotland forms part of a voluntary union; and (2) the EU – a strong, regional, supranational entity/structure – is deeply implicated in the claim for independence, helping both to justify a referendum now that the UK is no longer part of the EU, and giving political stability and plausibility to Scotland operating as an independent state.
How then to proceed? In the Quebec Reference Opinion, the court recognised that “the continued existence and operation of the Canadian constitutional order cannot remain unaffected by the unambiguous expression of a clear majority of Quebecois that they no longer wish to remain in Canada”, and “[t]he primary means by which that expression is given effect is the constitutional duty to negotiate in accordance with the constitutional principles [of federalism, democracy, constitutionalism, rule of law and respect for minorities]” (p. 273, emphasis added). Moreover, if the federal Government failed in its duty to negotiate, this could undermine its legitimacy, while failure on the part of Quebec might undermine any attempt at recognition as a state at the international level. In this respect, “adherence of the parties to the obligation to negotiate [in the Canadian Constitution] would be evaluated in an indirect manner on the international plane” (p. 273).
The attractiveness of the Canadian court's position is that it seeks to accommodate a difficult political reality within the existing constitutional framework, while also recognising the relevance of the international right of self-determination. It would be extremely difficult to articulate hard rules on when a political entity has a right to break away from a state but focusing on process and on a duty to negotiate allows for fact-specific dimensions to be considered. In the Scottish situation, this includes the 2014 “once in a lifetime” referendum on Scotland's status, which in turn must be set again the fact that it took place before Brexit, which Scotland did not vote for, together with Westminster's attempts to disempower Scotland, and the other devolved nations, during the post-Brexit transition. We might then conclude that international law operates directly, in very limited circumstances and in exceptional and clear-cut cases, to confer a right on a people to choose their own political form, but otherwise leaves political and democratic questions about the relationship of “a people” within a unified constitutional entity to be worked through at the national level, with the international right helping to strengthen the need for negotiation on both sides.
For Scotland, this will have to be worked out in the political and not the legal arena. The SNP are proposing to campaign for the next general election to the Westminster Parliament solely on Scottish independence. If, following this election, the UK Government of whatever political persuasion continues to refuse to grant the Scottish Parliament the power to legislate for a second independence referendum, we may see further novel means of bringing the issue to the courts. It is hard to predict whether any such challenge would be brought and, if so, what the court would say and whether that would have a larger constitutional impact than this decision, particularly to resolve the democratic elephant in the room.
