*Editors’ Note: This post is part of the ‘Unwritten Constitutional Norms and Principles Blog Series’*
Throughout the life of devolution, the courts appear to have diverged on how to properly determine the scope and limits of devolved lawmaking power, and the extent to which norms and principles not expressly contained in the text of the devolution statutes can usefully sculpt or inform those statutes’ interpretation. The current dominant approach is one in which principles and considerations not expressly reflected in those statutes play a far less significant role in that analysis. Although that approach has certain intellectual strengths, it also has some significant weaknesses, and this post argues in favour of a moderate approach which can help ameliorate some of these weaknesses. This approach requires that (i) particular provisions are interpreted in light of the statutory framework of which they form part, and (ii) the implicit role played by wider principles within that framework is acknowledged. Such an approach is not necessarily novel, nor is it a significant departure from the courts’ current position, but it is worth explicating because it may have significant (corrective) implications for the treatment of the devolution settlements.
Interpretive Approaches
In some cases, the courts’ interpretation of the devolution statutes has been informed partly by reference to principles and considerations which, though often arguably functions of those statutes, are not expressly reflected within their texts. For example the courts have drawn upon the constitutional significance of the devolution framework itself, or the particular qualities of its relevant institutions, as providing interpretive guardrails for the devolution statutes, and the powers and limitations these provide. Observations of these kinds have sometimes been connected with deeper constitutional principles, some of which have served to determine both the appropriate scope of judicial review, and have provided hard limits on devolved lawmaking power. In AXA, for example, Lord Reed said that, in its enactment of devolution, Westminster ‘cannot be taken to have intended to establish a body which was free to abrogate fundamental rights or to violate the rule of law.’ Lord Hope similarly explained in that case that the Scottish Parliament’s ‘firm rooting in the traditions of a universal democracy’ ‘suggests that the judges should intervene, if at all, only in the most exceptional circumstances’ (‘if legislation enacted by the Scottish Parliament was incompatible with the rule of law’). However, it might be argued that reference to the qualities and status of devolution, as well as the meaning and purchase of background constitutional principles themselves, are contestable, nebulous, or subject to – rather than determinative of – a court’s interpretation of the relevant statutory scheme. Such considerations might, therefore, be objected to on the basis that they can provide a vehicle through which ‘a judge’s view of the aim and purposes of devolution can inform the interpretation of specific provisions’.
In some more recent judgments the courts have, by contrast, noted the limited utility of principles and considerations of these kinds, or of ‘any broader understanding of the constitutional context in which devolution is situated’, which may not always (and depending on the route through which certain questions reach the courts) be instructive or determinative. For example, in Continuity Bill the Court said that a s 33 reference ‘is concerned only with the extent of the Scottish Parliament’s “legislative competence”’, ‘a term of art … [which] refers back to section 29’. That provision was, in the Court’s view, ‘exhaustive’ and, as a consequence, ‘[t]he UK Law Officers’ case on these points is not assisted by reference to the constitutional framework underlying the devolution settlement or the principles of legal certainty and legality. The constitutional framework underlying the devolution settlement’ the Court said, ‘is neither more nor less than what is contained in the Scotland Act construed on principles which are now well settled’ (emphasis added).
Adopting an analogous, if starker, approach in the s 35 Reference Lady Haldane held that recourse to principles of the separation of powers or the rule of law would be of little assistance: ‘Section 35 does not, in and of itself, impact on the separation of powers or other fundamental constitutional principle. Rather it is itself part of the constitutional framework.’ This appears to present a dichotomy between principles that are ‘part of the constitutional framework’, and those which do have such wider impacts or effects. Foran argues that, on Lady Haldane’s approach, ‘background principles are themselves concretised and afforded democratic legitimacy in the process of legislating’ in the relevant devolution statute. Importantly, on this account the substantive effect given to, and balance between, any relevant constitutional principles ‘is worked out within the legislative framework designed by Parliament.’
Weaknesses of the Current Approach
In recent practice this latter approach has, however, arguably exhibited two important weaknesses. First, in some cases the courts have focused on particular provisions in relative isolation, without considering how those provisions might have their meaning coloured or conditioned by other parts of the statute in which they are found. Second, this focus has arguably neglected to account for the implicit role inevitably played by wider constitutional principles.
As to the first, I have argued elsewhere that the devolution settlements contain within them some important tensions that the courts, in interpreting the scope of their various provisions, are tasked with balancing. The courts themselves have recognised this, with Lord Hope explaining in Imperial Tobacco that, when interpreting the Scotland Act, the Court should be mindful that that Act’s ‘concern must be taken to have been that the Scottish Parliament should be able to legislate effectively about matters that were intended to be devolved to it, while ensuring that there were adequate safeguards for those matters that were intended to be reserved.’ This dictum recognises that the courts need to give effect to the limitations within the devolution statutes in a way that is not blind to, and does not undermine the full scope of, the empowerment intended for the devolved institutions (and vice versa). This task is especially important because, as Lord Rodger noted in Martin v Most, each relevant devolution statute ‘concentrates on identifying the matters lying outside the competence of the [relevant] Parliament and… Executive’ rather than those lying within. Giving effect to a provision in a way that does not take the other parts of the relevant settlement into account would therefore be problematic. This concern is partly reflected in the courts’ (generally accepted) recognition that the devolution scheme is intended to create ‘a coherent, stable and workable system within which to exercise … legislative power’.
In both Continuity Bill and Treaty Incorporation, however, the Supreme Court gave an effect to a particular provision – s 28(7) of the Scotland Act – which is difficult to reconcile with the settlement within which that provision sits, especially the power of the Scottish Parliament to amend or repeal Acts of the Westminster Parliament within its competences. The interpretation of that provision in those cases – creating a notion of ‘unqualified legislative power’ – was all the more remarkable given that this provision has, in orthodox terms, been understood as a reflection of (an aspect of) parliamentary sovereignty. The Court interpreted that principle in flexible terms in both of these judgments, but did not extend that treatment to their interpretation of s 28(7) itself.
Indeed, the second weakness of this approach is its propensity to consider the interpretation of certain provisions as divorceable from principles which inform, at least impliedly, the text of the settlement itself. In Allister, for example, Lord Stephens said simply that ‘the answer to the question as to how the conflict or inconsistency [between article VI of the Act of Union with Ireland 1800 and the Northern Ireland Protocol] is to be resolved has been answered by Parliament in section 7A of the 2018 Act’ (emphasis added), purportedly rendering any reference to the constitutional statutes doctrine and, commensurately, the constitutional weight or value of the relevant norms at play, ‘academic’. As Mark Elliott and I have argued, it is, by contrast, the constitutional ‘weight’ of principles reflected in certain provisions that often determines how statutory language,and particularly the interaction between different pieces of legislation, is to be understood.
A Moderate Approach
There are, however, two ways that these weaknesses might be ameliorated. First, it should be accepted that the meaning and scope of each individual provision is illuminated by the other parts of the statute of which it forms part. This reasoning was evident in the Court’s treatment, in Treaty Incorporation, of a different set of provisions to those considered above. There, the Court was faced with provisions of the Scotland Act 1998 which were in tension with one-another: the ‘pre-enactment safeguards’ (ss 31(1), 31(2), 32, and arguably 33, the procedure by which the reference was made) appeared to require that the relationship between the Scottish Parliament’s legislation and its competences was static and clear at the point of enactment. Section 101(2), by contrast, appeared to allow the Court to read down legislation so that it was within competence (so far as it was possible to do so) dynamically. One set of provisions appeared to require the Bill be on its face within competence; another appeared to anticipate, or at least be capable of tolerating, a more dynamic, ‘cookie-cutter’ or ‘circumscribed’ approach.
Importantly, rather than interpreting these provisions in isolation, the Court construed s 101(2) narrowly so as to avoid ‘rendering nugatory the pre-enactment safeguards provided by the Scotland Act’. In this way, the Court acknowledged the tension within the settlement that needed interpretive resolution and narrowly interpreted one set of provisions to reconcile them with the other components of the statute. Interpreting the devolution framework so as to be internally coherent has also been advocated for by McHarg and McCorkindale, who have argued that a ‘[j]ustification for extending [a particular approach to repeal] to devolved legislation could have been found in the other constitutional provisions contained in the Scotland Act’ (emphasis added) (ss 63A and 28(8)).
Although the Court’s approach to s 101(2) in Treaty Incorporation can be criticised at least because the ‘pre-enactment safeguards’ were, if anything, better suited to a narrower interpretation than s 101, recognising, as the Court did, that this provision ‘has to be construed in the context of the Scotland Act as a whole’ nonetheless represents an intellectually coherent and defensible approach.
Sandro argues that ‘any legislative provision might have to be read in light of its co-text, which ranges from other provisions in the same statute, to relevant provisions in other statutes, all the way to constitutional principles which are engaged by the application of the provision in question.’ Although perhaps narrower than Sandro’s, the approach advanced here similarly does not exclude relevant norms and principles which are not expressly reflected in the devolution statutes from playing an important role. Indeed, the Court in Treaty Incorporation appeared to note that its interpretation of these different provisions should also recognise the wider constitutional environment in which they operate, and the settlement to which they give effect. The Court said that a further ‘reason for giving section 101(2) a less expansive construction’ reflected ‘the constitutional settlement effected by the Scotland Act’, emphasising that s 101(2) of the Scotland Act should be narrowly construed in order to avoid a degree of legal uncertainty intolerable to, or incompatible with, the rule of law (as reflected inter alia in the ECHR).
Thus, provisions which are part of the legislative framework can nonetheless have their meaning to some extent coloured by implicit principles, especially as far as these can assist the courts in determining how to reconcile the internal tension within the statutes. This kind of approach is also apparent in Continuity Bill. Although the Court said there that ‘[c]onsistency with the rule of law or the constitutional framework underpinning the devolution settlement… are not independent grounds of challenge available in these proceedings’, the Court did still recognise that consistency with those matters remained ‘relevant’, if ‘only so far as it assists in resolving [the competence] issue [raised in a s 33 reference].’
Conclusion
The moderate approach advocated for here is not necessarily novel, nor far removed from the current position which is dominant in the courts. Rather, it is explicated as a way to ameliorate some of the weaknesses in that approach, offering a path between an overly narrow approach, and one which relies too heavily on concepts that might be deeply contested. As Lord Rodger put it in Martin v Most, ‘[m]atters become clearer, however, when the provisions are seen in their setting in life.’
Thanks are owed to Se-shauna Wheatle and Aileen McHarg for their helpful comments on an earlier draft. Any errors or omissions remain my own.
Nicholas Kilford, Postdoctoral Research Associate at Durham Law School.
(Suggested citation: N. Kilford, ‘Interpreting The Devolution Statutes’, U.K. Const. L. Blog (28th May 2024) (available at https://ukconstitutionallaw.org/)
