Cormac Mac Amhlaigh: … yes, but is it legal? The Scottish Independence Referendum and the Scotland Act 1998

The question of the legality of a referendum on Scottish independence pursuant to an Act of the Scottish Parliament has been a hobbyhorse of Scottish constitution- watchers since before the Scotland Act 1998 (SA) came into force.  It has now entered the national (British) political fray front and centre, in the form of the spat between Alex Salmond and David Cameron over ownership and control of an independence referendum process and the timing and wording of any such referendum.

Both parties have (unsurprisingly) taken opposite views on the legality of an Act of the Scottish Parliament (ASP) facilitating a referendum on independence (or devo max; devolution of powers just short of full independence) in the absence of a specific grant of the power to hold the referendum by Westminster.  Thus, arguably one of the most significant constitutional issues since the foundation of the United Kingdom over 300 years ago could founder on legal technicality:  would an ASP authorising a referendum, touch upon prohibited grounds under s. 29 of the SA.

As is well known, the model of devolution of powers to the Scottish Parliament under the Scotland Act took place according to an exclusionary as opposed to an inclusionary model.  That is, unlike the devolution proposals in 1978, the Scotland Act devolves plenary powers to Holyrood except for the competence to define the limits of its own competence, as well as specific prohibited grounds reserved to the Westminster parliament.  Most of the litigation on the powers of the vires of measures taken under the Scotland Act (most notably executive acts) in the first decade and a half of devolution have related to (some high profile) human rights cases, most notably the recent Cadder v. HM Advocate [2010] UKSC 43 Frazer v. HM Advocate [2011] UKSC 24 decisions.

The relevant wording of s. 29 for the purposes of an ASP facilitating a referendum, is that it must not ‘relate to’ reserved matters (s. 29(2)(b)).  S. 29(3) provides guidance on the meaning of ‘relates to’, which will 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) SA).  Reserved powers, contained in Sch. 5 include the Constitution of which ‘the Union of the Kingdoms of Scotland and England’ is part.  So, it all turns on whether an Act which facilitates a referendum on Scottish independence or devo max ‘relates to’ the Union of Scotland and England within the meaning of s. 29 SA.

The Coalition government, in its recent report on the a referendum which includes its interpretation of the legal position, claims that such a referendum clearly relates to the constitution, and particularly the Union between Scotland and England, in that an independence referendum could lead to the severance of the Union.  The Scottish government, for their part, argue that any such referendum would be merely advisory or consultative, have no legal effect within the terms of the Scotland Act and therefore does not violate the terms of devolution.  The implication being that part of the reserved powers model of devolution entails the devolution of the power to hold referendums given that it is not specifically reserved or limited by the Scotland Act, regardless of the subject matter of the plebiscite.

The question of the validity of an ASP facilitating such a referendum is an archetypal ‘hard case’, due to the fact that it involves a dispute about the level of analysis the Supreme Court should adopt in assessing the vires of ASPs under s. 29 SA.  Should the scrutiny of the purpose of a referendum ASP be narrow and limited, not moving beyond the immediate purpose of the Act itself; i.e. to hold a referendum?  Or should the scrutiny be more contextual, one could say ‘anxious’, taking into account, not only what the Act purports to do in the immediate term, but also the social and, perhaps more importantly political, context within which the Act would operate and its longer term consequences?

On a narrow reading of s. 29, which coincides with the Scottish Government’s reading, the purpose of such a provision would be to hold a referendum and there the analysis by the court would stop.  The holding of referendums is not explicitly prohibited in the SA and therefore the facilitating Act would clearly be valid.  If pushed, it could be argued on this view that its ‘effect in all the circumstances’ would be null in that it would not have any legal effect; its simply taking the temperature of Scottish public opinion.

A broader level analysis, one which seems to be reflected in the coalition government’s reading of the SA, is that the analysis must not stop at the immediate purpose of the ASP to hold a referendum, but that the purpose of the referendum itself must also be considered.  On this reading, the purpose of the ASP is thus to hold a referendum in order to subsequently secede from the Union which therefore ‘relates to’ a reserved matter.  The referendum, on this view, should not be considered an end in itself, but rather an instrument to achieve a further goal, that of secession or the devolution of further powers to the Scottish Parliament.

Whereas there has been much discussion (and disagreement) about the legality of an independence ASP, ultimately  the legality of the bill will be determined by the Supreme Court’s interpretation of s. 29 and so the best predictor of the legality of a proposed bill is an attempt to predict how the Court would approach s. 29.  To put it bluntly in the context of the current debate, a broad contextual approach to scrutiny of the purpose a purported Bill by the Supreme Court would favour the coalition government’s position whereas a narrow, limited scrutiny of the purpose of the purported bill favours the Scottish government’s interpretation of the Parliament’s powers.  The question is then, what indicators are there of the Supreme Court’s view?

There has been a surprising dearth of litigation reaching the Supreme Court based on the competence of ASPs with regard to reserved powers as opposed to violations of EU or Human Rights law under s. 29(2)(d). In fact, to date, only one case has been decided by the Supreme Court where it was forced to interpret the relevant provisions of s. 29 SA; Martin and Miller v. HM Advocate [2010] UKSC 10.  Although the Court was unanimous in finding that the Act of the Scottish Parliament (ASP) in question was not in violation of s. 29(2) read in conjunction with s. 29(3), that is that it did not ‘relate to’ reserved powers, the case does reveal some clues as to how the Supreme Court would interpret s. 29 SA for the purposes of a referendum bill, should it ever have the occasion to.

The case involved an ASP, the Criminal Proceedings (Reform) Scotland Act 2007 , and in particular s. 45 which purported to raise the maximum penalty for offences tried summarily to 12 months imprisonment.  The joint appellants were convicted of road traffic offences for which the maximum penalty had, before the 2007 Act, been 6 months imprisonment.  The Court sentenced them to   12 and 7 months imprisonment respectively (i.e. more than the pre-2007 reform limits).  The Road Traffic Acts, pursuant to which the offences were prosecuted, is a reserved matter under Schedule 5 of the Scotland Act and so they claimed, among other things, that the ASP which augmented the sentences was beyond the powers of the Scottish Parliament due to the fact that it ‘related to’ a reserved matter, namely offences under the Road Traffic Acts.

The case was eventually decided according to the more arcane provisions of s. 29(2)(c) and s. 29(4) SA where the court divided 3 to 2 on the outcome, details which need not detain us here.  Of more interest was the unanimous approach of the Court as to whether the amendments of the 2007 Act ‘related to’ a reserved matter under s. 29(2) SA read in conjunction with s. 29(3) SA.  The court found that the ‘purpose test’ contained in s. 29(3) SA, necessitated a broad contextual approach to the provision of the 2007 and not simply a narrow, facial reading of the provision which augmented the sentences for offences tried summarily.  Thus, Lord Hope found that in applying the purpose test, the Court should have regard to the ‘situation before the provision was enacted, which it was designed to address’.  The Court liberally referenced the expert committee reports which recommended the change in the maximum penalties for offences tried summarily, reports and papers issued by the Scottish Ministers prior to the introduction of the Bill, explanatory notes to the Bill, policy memoranda, statements made by ministers and the Parliamentary committees during the passage of the Bill through the Scottish Parliament to conclude that, in the light of the broader context of the provision, its purpose was to reform the summary justice system by reducing pressure on higher courts (which was unambiguously a devolved matter), and not to raise sentencing limits for road traffic offences generally (which would be a reserved matter).  The increase in the sentencing powers of offences tried summarily in the lower courts was deemed to be a necessary part of this reform.  As Lord Kerr remarked, the greater maximum penalty introduced by the relevant provisions was not ‘an end in itself’ but a mechanism of a broader purpose of the reform of the jurisdiction of the Scottish courts (a devolved matter) in order to enhance efficiency.

Analogizing then from this case to the legality of an ASP facilitating a referendum, it would appear that were the Court to follow the approach it took in Miller and Martin, this would militate in favour of the coalition government’s interpretation of s. 29.  As such, the referendum introduced pursuant to the bill would not be read as an ‘end in itself’ but rather would have to be seen in its broader political context of enactment, including consultations, debates and previous ministerial statements.  This could include the purpose of the bill constituting a means to a broader end of Scottish independence which would clearly fall foul of s. 29(2) SA.  As much seems to have been anticipated by the coalition government’s legal analysis of the situation in its recent Consultation Document, where it refers to the Scottish Government’s 2010 Draft Referendum Bill Consultation which stated that ‘The [Scottish] Parliament’s powers should be extended to enable independence to be achieved’.  The Consultation document therefore concluded that ‘both purpose and effect [of a referendum bill] therefore related directly to the reserved matter of the Union of the Kingdoms of Scotland and England’ and that the ‘underlying purpose’ of any advisory or consultative referendum would be Scottish independence.

However, if the reports, statements and arguments advanced by the Scottish Government as well as the explanatory notes to any forthcoming referendum bill contained a ‘saving clause’, explicitly stating that the purpose of the Bill was not to give effect to secession nor directly affect the Union and therefore encroach upon a reserved power, but rather was the simply exercise of a devolved power (the power to hold referendums), then this would muddy the waters considerably.  It would remain to be seen what the Supreme Court would make of such a savings clause, were it to be introduced.

Whichever way the Court swings on this question should it ever actually reach its docket, what is clear is that the law can play only but a minor role in the resolution of this potential constitutional crisis.  Analogies to the Canadian situation with Quebec should be made with caution (after all, the Scotland Act is not the Canadian constitution), but if we are permitted to draw anything from the Quebec Secession Reference decision of the Canadian Supreme Court in 1998, it is surely that these questions are, at root, political questions which must, in the final analysis, be resolved politically.

Dr. Cormac Mac Amhlaigh is Lecturer in Public Law, University of Edinburgh.