Published on 6 October 2022
Share Tweet  Share

The Scottish referendum question: what might the Supreme Court decide?

In advance of the Supreme Court hearing on the query made by the Scottish Government about whether it can hold a second referendum on the question of Scottish independence from the United Kingdom, Professor of Constitutional Theory Stephen Tierney looks at the merits of the case and considers how the court is likely to respond.

In June of this year, the Scottish Government put forward a proposal to hold a second referendum on the question of Scottish independence from the United Kingdom. It published a draft bill and has sought to refer this to the Supreme Court of the United Kingdom for a ruling on whether the Scottish Parliament has the authority under the Scotland Act 1998 to pass it. A date for the hearing has been provisionally set by the Supreme Court for 11-12 October 2022.

The Supreme Court is confronted with two questions: one concerns whether the time is right for the Court to answer the question put to it by the Scottish Government; the other is the substance of the question itself. This post seeks to understand how the Court is likely to answer these questions.

Will the Court decide the case is too early?

Both the Lord Advocate, the Scottish Government’s chief law officer, and the Advocate General, who advises the UK Government on Scottish legal matters, have put forward their written arguments. The Advocate General argues that the Lord Advocate’s reference procedure (under Schedule 6, para 34) cannot be used to test the legality of a proposed bill. The Scotland Act 1998 contains a set of procedures through which the lawfulness of a bill can be both assessed upon its introduction to the Scottish Parliament (s.31) and tested by the courts upon its passage through that chamber (s.33). This process should be followed before any case is brought to court.

In her submission, the Lord Advocate surprisingly expresses a lack of confidence in the legality of the proposed draft bill and uses this lack of confidence to justify the use of the reference procedure, arguing that without a reference at this stage the draft bill would be unlikely to reach the courts, presumably because she would not be able to ‘sign off’ on the legality of the Bill before it was introduced into the Scottish Parliament.

It is unsurprising therefore that, on behalf of the UK Government, the Advocate General is robust in refuting the appropriateness of the Reference:

“It is hard to see why this should be a matter of legal concern. Parliament is unlikely to have intended the time of the Scottish Parliament, or the resources of the Supreme Court on a Schedule 6 reference, to taken up with matters which the Scottish Government’s own Law Officer considers to be outside competence.”

It would appear that the Supreme Court could well find that the use of the reference procedure is inappropriate to determine the potential legality of a draft proposal when a bill could be brought before the Scottish Parliament and subjected to parliamentary pre-legislative review. It would set a possibly very expansive and problematic precedent, allowing devolved governments to use the Supreme Court as a pre-legislative reviewing chamber or indeed allowing the UK government to hold up policy initiatives of the devolved governments in references of its own. If the Court does refuse the Reference, it may also decide not to comment on the merits of the case.

The merits of the case

The Lord Advocate asks this question of the Court:

“Does the provision of the proposed Scottish Independence Referendum Bill that provides that the question to be asked in a referendum would be “Should Scotland be an independent country?” relate to reserved matters?

At the heart of the Lord Advocate’s argument is the contention that the proposed referendum would be ‘advisory’, ‘consultative’ or ‘non-binding’. To understand why this is so central to the case, it is necessary to look at s.29 of the Scotland Act which defines what the Scottish Parliament can and cannot do.

Section 29 of the Scotland Act 1998 provides that an Act of the Scottish Parliament is not law if it is outside the legislative competence of the Parliament. A provision is outside that competence so far as it “relates to reserved matters” (s.29(2)(b)). And whether or not it relates to a reserved matter is to be determined by “reference to the purpose of the provision, having regard (among other things) to its effect in all the circumstances” (s.29(3)). Reserved matters are set out in Schedule 5 to the Act and include “the Union of the Kingdoms of Scotland and England” and “the Parliament of the United Kingdom”.

It is clear that any decisive act which would end the Union is a reserved matter, something the Lord Advocate readily concedes (Lord Advocate’s Written Submission, para 102). The question then is whether it is possible to construe s.29 in such a way that a referendum on the subject of independent statehood does not “relate to” the Union due to its purportedly advisory or non-binding nature.

To seek to understand how the Court might approach this issue, it is necessary first to address the general approach taken by the courts to interpret the Scotland Act, before turning to their specific treatment of s.29. The courts have emphasised repeatedly that the Scotland Act must be interpreted by way of “the ordinary meaning of the words used” and that “the Scotland Act provides its own dictionary”, meaning the British courts should not take lessons Commonwealth federal courts’ division of powers decisions. The Supreme Court has also taken the view that the Scotland Act is to be interpreted like any other statute (see Imperial Tobacco, EU Law Continuity Bill and Rights of the Child). This approach, which construes the Scotland Act narrowly, emanates from the fact that the Scotland Act does not create a ‘division’ of powers in a federal sense: Parliament’s own authority to legislate for Scotland on any matter remains unlimited (SA s28(7)).

On its face, the term “relating to” a reserved matter would appear, on the ordinary meaning of the words, to be a potentially very broad test (“to relate: to find or show the connection between two or more things” – [Office for Economic Development]). To take a referendum on the question of independent statehood as an example, relying upon the ordinary meaning of “to relate”, it would seem that any bill proposing a referendum on this matter would “relate” to the Union, in the sense of having a connection with it, whether the proposed referendum purported to be either decision-making or advisory.

The Lord Advocate however suggests that an ‘advisory’ or ‘non-binding’ referendum would not relate to the Union or to Parliament because its effect would be, in legal terms, nil. It would not end the Union. The Supreme Court has however confirmed that relevant “effects” in s.29(3) include both “legal and practical effects”. It would appear, given the significance of referendums within the British constitution, that any referendum would be considered to have important practical effects in terms of creating a political obligation to implement the result, a conclusion reached in Miller I concerning the Brexit referendum.

In its general approach to interpretation of s.29, the Supreme Court has also made the following point: “It is … important to bear in mind the central aim of the provisions in the Scotland Act concerning reserved matters… that matters in which the UK as a whole has an interest should continue to be the responsibility of the UK Parliament.” The Union of the Kingdoms of Scotland and England seems clearly to be one of these matters, and arguably the reservation upon which all other matters in which “the UK as a whole has an interest” depend.

This approach helps surmount the potential ambiguity introduced into the section by the terms “purpose” and “effect”. In the first place, the Court seems to elevate the notion of “relates to” as the over-arching test, with “purpose” and “effect” performing an ancillary function. If a bill clearly relates to a reserved matter, it is irrelevant that its purpose (e.g. presenting a referendum as a consultation exercise) would appear to be within the competence of the Scottish Parliament. As Lord Hope has ruled: “[A]n Act may have a perfectly lawful object but may seek to achieve that object by invalid methods.”

Of course, a further argument could be made that a referendum bill has two or more purposes, at least one of which falls within competence: secession on the one hand and mere consultation on the other. Lord Hope has also addressed this matter and made clear, albeit obiter dicta, that where an unlawful purpose exists, even if it sits alongside a potentially lawful one, the provision would still be incompetent.

Underpinning much of this case law is the clear distinction, drawn by the courts, between the status of the United Kingdom Parliament on the one hand, and that of the devolved legislatures including the Scottish Parliament, on the other. The limited powers of the Scottish Parliament, in contrast to the UK Parliament’s unlimited authority, has been frequently restated by the courts in judgments concerning devolution matters (see AXA General Insurance, Martin v Most, Imperial Tobacco, and EU Law Continuity Bill).  This distinction was made very forcefully in a recent reference case. Again, this point argues against the kind of expansive reading of the Scotland Act that would be necessary to find that the proposed draft Bill does not relate to the Union.

Section 101(2) of the 1998 Act provides that a Bill of the Scottish Parliament “is to be read as narrowly as is required for it to be within competence, if such a reading is possible…” Although this may seem to invite a generous interpretation of the Scotland Act, the Supreme Court has pointed out that this provision cannot be used to circumvent the limits of the Scottish Parliament’s powers, a point supported by the Explanatory Notes which accompanied the Scotland Bill 1998: “if a provision clearly cannot be read to be within competence, for example an Act of the Scottish Parliament providing only for a referendum on independence, then the section will not allow it to be read as being within competence.”

Finally, the Court of Session has already been asked to assess the question whether the Scottish Parliament has the authority to hold a referendum on Scottish independence. The Lord President, Lord Carloway, declined to consider the matter as hypothetical and premature but he did give a strong hint that in his view such a bill would be unlawful:

it may not be too difficult to arrive at a conclusion, but that is a matter, perhaps, for another day.” (emphasis added)

Lord Carloway’s conclusion is of course the latest chapter of a complex legal story, but his comment reflects a strong line of cases in which the courts have carefully and strictly policed the limits of devolved powers.

To conclude, the Supreme Court may find that the Reference has been submitted at too early a stage because it was open to the Scottish Government to bring a bill to the Scottish Parliament and have its legality tested at the beginning and end of the legislative process. If it does reach this conclusion it may refuse to address the substance of the Lord Advocate’s question. If, however, the Supreme Court does decide to assess the substantive question, previous cases suggest that the draft Bill will most likely be declared to fall beyond the powers of the Scottish Parliament.

This blog post is written in a personal capacity. It draws upon a paper to be published by the Review of the UK Constitution Project exploring the legal and constitutional issues surrounding the independence referendum proposal.

The views and opinions expressed in this post are those of the author(s) and not necessarily those of the Bennett Institute for Public Policy.


Stephen Tierney

Stephen Tierney is Professor of Constitutional Theory, Edinburgh Law School and Legal Adviser to the House of Lords Constitution Committee. He is the author of Constitutional Referendums: The Theory and Practice...

Related Publications

Back to Top