Two weeks ago, the UK Government published its consultation document on Scotland’s Constitutional Future, in which it stated its view that the Scottish Parliament has no power to enact legislation authorising a referendum on the question whether Scotland should become independent from the United Kingdom. Last week, the Scottish Government published its own consultation paper, Your Scotland, Your Referendum, claiming that the Scottish Parliament can validly authorise the asking of at least some questions about independence, although the document is ambiguous as to whether the Scottish Government believes that it has power to ask its preferred referendum question, namely ‘Do you agree that Scotland should become an independent country?’.
The legality or otherwise of an independence referendum is, from one perspective, a narrow point: a matter merely of process, which could be authoritatively resolved by an express grant of power from either the UK Government (under s.30 of the Scotland Act 1998) or the UK Parliament (under primary legislation), and a distraction from the more important issues of where a referendum might be taking both Scotland and the rest of the UK.
From another perspective, though, the legality issue is crucial because the answer to the question tells us a great deal about where we are currently in constitutional terms, and hence about how great or small a constitutional leap we might be about to take. Furthermore, since the recognition claims and aspirations that are so central to constitutional politics are often as much concerned with who is entitled to set or influence the agenda of debate as they are about the outcome of debate, we cannot dismiss the question of background legal authority as of merely academic interest. Accordingly, the legality issue remains important even if it becomes practically irrelevant, and it has significance not only for Scotland, but for our understanding of the nature of the UK constitution as a whole.
Contrary to the views of the UK Government and a number of influential commentators, on this blog and elsewhere, we believe that the legality of a referendum Bill passed under the Scotland Act as it currently stands is a more open question than has been generally acknowledged. In other words, we believe that a plausible case can be made that such a Bill would be lawful, and believe it is important that these arguments are clearly set out. The case for legality rests on a particular reading both of the purposes of a referendum Bill, and of the purposes of the Scotland Act.
The UK Government’s argument that a referendum Bill would be unlawful rests on two premises: first, that section 29(2)(b) of the Scotland Act, which provides that an Act of the Scottish Parliament is outwith competence if it “relates to” a reserved matter, is to be interpreted literally; and, second, that the purpose of a referendum Bill, having regard to its effect in all the circumstances (s.29(3)), would be to dissolve the Union. Since the Union is a matter reserved to the UK Parliament, the Bill would, they argue, therefore relate to a reserved matter and would be unlawful.
Both premises of this argument are contestable. Taking the second premise first, there are a number of difficulties with the UK Government’s interpretation of the purpose of a referendum Bill. In the first place, they rely on a broad interpretation of its purpose, effectively treating the undoubted political aspiration of the SNP government, as promoters of the Bill, to bring about an end to the Union, as the purpose of the legislation. By contrast, the purpose as stated in the long title of the draft Bill published by the Scottish Government is simply to seek “the views of people in Scotland on a proposal about the way Scotland is governed.” In taking a broader rather than a narrower view of the legislative purpose, the UK Government seems to be relying on the approach adopted by the Supreme Court in Martin and Miller v HM Advocate  UKSC 10, in which Lord Hope said:
“One of the circumstances to which it is proper to have regard is the situation before the provision was enacted, which it was designed to address. Reports to and papers issued by the Scottish Ministers prior to the introduction of the Bill, explanatory notes to the Bill, the policy memorandum that accompanied it and statements by Ministers during the proceedings in the Scottish Parliament may all be taken into account in this assessment.”
It is, however, important to note that the reason for adopting a broad approach to purpose in Martin and Miller was not to hold that the legislation in question was ultra vires, but in fact to hold that it was intra vires. Arguably, this approach would not be appropriate where its effect would be to render legislation beyond competence. This is supported by s.101(2) of the Scotland Act, which instructs the courts that, where possible, they should interpret Bills as narrowly as is required to allow them to be upheld.
A second objection to the UK Government’s broad reading of the purpose of a referendum Bill is that it conflates the intention of the Scottish Government with the intention of the Scottish Parliament. It is perfectly conceivable that some MSPs may vote for a referendum, not because they support independence, but rather because they expect that the vote will be lost and that the issue of independence will thereby be removed from the political agenda, at least for the foreseeable future. Indeed, this seems to be the UK government’s own reason for wishing a referendum to take place.
Finally, it is a curiosity of the UK Government’s position that, in assessing the effect of a referendum, they seem to be assuming that the yes campaign will win. Hence it will confer a political mandate to negotiate terms for independence and will lead inexorably to the dissolution of the Union. Should the outcome be different, though, the effect, it would seem, would be to reinforce the Union. If the key test of legality is practical effect, would the referendum legislation in that case be lawful? Of course, it would be absurd if the legality of a referendum Bill were to depend on outcome of the referendum, not least because the outcome is unknowable at the point at which any challenge is likely to be heard. This then points to the true meaning of the ‘effect’ of a Bill as being its legal effect, rather than its practical effect. Since the legal effect of a referendum Bill is indisputably simply to seek the views of people in Scotland – and any further effect is both non-binding and speculative – this again points to the narrower, consultative, interpretation of the legislative purpose as being the correct one. It should be added that, if this is the correct approach to the identification of the Bill’s purpose, then the precise wording of the referendum question would appear to be a red herring; the legal effect of the referendum is not altered by asking an indirect rather than a direct question about whether Scotland should become independent.
However, even if the purpose of the Bill is correctly characterised as consultative, the second limb of the UK Government’s argument remains to be addressed. Would a consultative Bill nevertheless ‘relate to’ the Union? Literally, of course, it would. However, the weight of authority (see Tarun Khaitan’s helpful post on this issue) suggests that devolution statutes, as constitutional measures, are not to be interpreted literally, but rather “generously and purposively” (Robinson v Secretary of State for Northern Ireland  UKHL 32). In other words, the statutory phrase “relates to a reserved matter” is itself capable of expansive or restrictive interpretation. Arguably, in order to give effect to the purpose of the Scotland Act, it should therefore be read as requiring those challenging the competence of a Bill to show more than it merely “has something to do with” a reserved matter, in this case the Union.
What, then, is the purpose of the Scotland Act? Here we come to the nub of the difficulty of predicting how the courts would respond to a challenge to the legality of a referendum Bill: the contested, and potentially evolving, nature of (a) the purpose of devolution; (b) the nature of the UK constitution; and (c) the impact of devolution upon the UK constitution. There are at least three possibilities. One is to regard devolution as a mere delegation of authority from the UK Parliament. On this view, the Scottish Parliament is politically and legally subordinate to Westminster, and the latter remains the sole font of sovereign authority within the state (the unitary state narrative). The second possibility is that devolution represents a move towards a quasi-federal constitution. On this view, the Scottish Parliament is the political equal of Westminster, within its sphere of competence, but it is bound by the norms of the UK constitution as a whole (the federalist narrative). The third view sees devolution as a renegotiation of the terms of Union on the part of the sovereign Scottish people, and hence sees the Scottish Parliament as a legitimate representative of the Scottish people in the course of any further renegotiation in which the interests of the Union as a whole and of its various parts are put at issue (the union state narrative).
Both the unitary state and federalist understandings of the purposes of devolution might suggest that questions about the future constitutional position of Scotland should indeed be reserved to the UK Parliament. However, the union state narrative – strongly endorsed by the Scottish Constitutional Convention prior to devolution – suggests a much broader role for the Scottish Parliament in debating, and facilitating public deliberation on, questions about the future governance of Scotland, albeit the Scottish Parliament does not have power to unilaterally secede from the Union or to enlarge its own powers. On this view of the purposes of the Scotland Act, then, it may be argued that a consultative referendum would be a lawful exercise of the Scottish Parliament’s legislative powers.
Which of these constitutional narratives would the courts – and particularly the Supreme Court – endorse? The unitary state narrative would appear to be ruled out by the recent decision in Axa General Insurance Ltd v Lord Advocate  UKSC 46, which rejected the argument that, at least for the purposes of judicial review at common law, the Scottish Parliament is to be understood as a subordinate legislature. More generally, judicial dicta and extra-curial writings questioning the continued accuracy of the traditional understanding of Westminster’s sovereignty also suggest that a crude, unitary account of the UK constitution would not be accepted.
The thickening sense of UK constitutional law that this brings in its wake might point to a quasi-federal account of the UK constitution as being most likely to be endorsed. However, the union-state narrative also gains support, not only from the political background to devolution, but also from the decision in MacCormick v Lord Advocate 1953 SC 396 and subsequent cases, to the effect that the fundamental principles of Scots constitutional law are not necessarily the same as those of English constitutional law, a view apparently endorsed by Lord Hope in R (Jackson) v Attorney-General  1 AC 262. The choice between these two options might turn in the end on whether the Supreme Court sees itself as a court interpreting and upholding a single and unified corpus of constitutional law, or as one recognising and seeking to reconcile the somewhat different constitutional traditions of the various jurisdictions of the UK.
The questions ultimately at stake in any challenge to the legality of a referendum Bill are, then, large ones indeed, and such a case would have the potential to be a defining moment in Scots – and UK – constitutional law. Should we, as constitutional lawyers, therefore welcome the prospect of a legal challenge? Does this have the potential to be our Reference re the Secession of Quebec, laying down authoritative principles as to how questions about Scotland’s – and the UK’s – constitutional future is to be determined? Or is the better analogy Bush v Gore? In other words, would any decision about what the law is on this point inevitably be seen as being tainted by the judges’ views as to what the law ought to be? The fraught political context in which such a challenge would be heard, as well as the widely publicised antipathy to the Supreme Court within the Scottish government, suggests that the latter analogy is more apt.
The risk to the authority of the courts – a risk which, it should be noted, would not be avoided by taking refuge in a literal interpretation of the Scotland Act – therefore suggests that the UK and Scottish Governments would indeed be wiser to agree on an express transfer of powers. Nevertheless, because of the fundamental nature of the issues at stake, and the inherent contestability of constitutional law questions of this kind, it is important that any such agreement should be not taken as an unequivocal endorsement of the view that Westminster alone is entitled to authorise a referendum on the constitutional future of any part of the UK.
Gavin Anderson, Senior Lecturer, University of Glasgow
Christine Bell, Professor of Constitutional Law, University of Edinburgh
Sarah Craig, Lecturer, University of Glasgow
Aileen McHarg, Senior Lecturer in Public Law, University of Glasgow
Tom Mullen, Professor of Law, University of Glasgow
Stephen Tierney, Professor of Constitutional Theory, University of Edinburgh
Neil Walker, Regius Professor of Public Law and the Law of Nature and Nations, University of Edinburgh
Editor’s update: This post was the subject of a report in The Herald. The report can be read here: