For the Balkinization symposium on Aileen Kavanagh, The Collaborative Constitution (Cambridge University Press, 2023).
Erin F. Delaney
Aileen Kavanagh has written a book that is beautiful in both words and in sentiment. She deftly paints a vision of the separation of powers that goes “beyond the forms to norms” (p. 261) and privileges “inter-institution comity [or] ‘that respect which one great organ of the State owes to another’” (p. 98). She has a normative and a positive claim: In her view, the “long-term working relationship” (p. 102) among the three branches of government ought to be based on mutual self-restraint and mutual support, focused on reciprocity and reputation. And Kavanagh argues, in the United Kingdom, that is the nature of the constitution.
Drawing on the United Kingdom as “an illustrative case-study,” Kavanagh develops her account by diving into the nuances of the Human Rights Act (1998), including executive agenda-setting, parliamentary process, the role of the civil service, and the response of the courts. Her description of the interaction of the branches in this context is persuasive, and the “presumption of partnership” is a normatively attractive result. Her definition of collaboration is expansive—including “when one branch checks, corrects, or even vetoes the decision of another branch” (p. 107)—but she may be right that it is possible to distinguish between constructive (or instructive) comment and destructive critique.
In Kavanagh’s words, the United Kingdom is an “instantiation of the collaborative ideal” (p. 23), but the norms of the “Collaborative Constitution” would be unrecognizable to those who understand the UK constitution to go beyond her Diceyan description. Kavanagh’s book does not encompass (or acknowledge at any point) the territorial constitution, as is unfortunately true in many segments of British public law. Expanding the constitutional frame to include devolution raises questions about the nature of the collaborative constitution and—under Kavanagh’s terms—how to understand devolution itself.
The collaborative constitution devolves in the context of devolution. Along with the HRA, another plank of the New Labour reforms of the late 1990s was devolution, instantiated in the Scotland Act (1998), the Government of Wales Act (1998), and the Northern Ireland Act (1998). These Acts—like the HRA—also incorporated “collaborative” elements. For example, in the course of negotiations over the Scotland Act, Lord Sewel stated in the House of Lords that the Westminster Parliament “would not normally legislate with regard to devolved matters in Scotland without the consent of the Scottish Parliament.” This principle became known as the Sewel Convention.
Over the ensuing decade, a scholar could tell a wonderful collaborative story about the interactions between Westminster and Holyrood (Scottish parliament) under the Sewel Convention. During that time, Sewel was “engaged more than 140 times in Scotland, but consent had been withheld only once and was followed by a compromise.” Others could show how the UK Supreme Court (after the Constitutional Reform Act (2005)) worked to reinforce and support devolution. Indeed, a Memorandum of Understanding was agreed in 2013, along with Supplementary Agreements and more Devolution Guidance Notes, among the devolved governments and the UK. In 2016, an updated Scotland Act put the Sewel Convention in statute. Much like Kavanagh’s description of the process of protecting rights under the HRA, the devolution context is similarly latticed with frameworks, committees, and admonitions to act responsibly and with respect. Trust is necessarily the watchword.
Given the similarities in the processes and protections of the territorial constitution to those in the rights-protecting Diceyan constitution described by Kavanagh, one would expect a similarly rosy picture of continued collaboration. Needless to say, those expectations would be misplaced.
The process of exiting the European Union has undermined the Sewel Convention dramatically. Although Leave won in England and Wales, Remain carried the day in Scotland and Northern Ireland. And repatriating powers from the EU has meant creating a new approach to the internal market within the UK, in order to address regulatory divergence. It has not gone well. Over the twenty-five years of devolution, “consent has only been denied [by a devolved government] from 28 acts,” but “19 of these cases occurred within the 2019-24 parliament.” Of course, this assumes consent has even been sought, and in the words of Dr. Chris McCorkindale, there “seems to be no guiding constitutional principle as to when it is appropriate for UK Ministers to [take powers to act in devolved areas] and as to the consent mechanisms (if any) that should attach to the exercise of those powers.”
Indeed, to the extent that there has been partnership, one might suggest it has been among the central branches against the devolved governments: more collborateur than collaborative. The UK Supreme Court decided, in the 2018 Miller v. Secretary of State for Exiting the European Union, that (notwithstanding its statutory footing) Sewel was a nonjusticiable constitutional convention, and thus the Court could not opine on whether the May Government was obligated to seek agreement from Holyrood on triggering Article 50 to withdraw from the European Union.
What to make of this devolutionary fly in the collaborative ointment? One possible answer is that devolution might present a useful test case for Kavanagh: in this context, the framework for collaboration buckled under Brexit. How much of the pre-Brexit collaborative (territorial) constitution was due to an alignment in political viewpoints? Was the Brexit/post-Brexit misalignment merely a lapse, or evidence of a cycle? If party politics matters for collaborative devolution—if Keir Starmer’s Labour government ushers in a new dawn for partnership with Scotland—how can we be sure that the structures, processes, and nudges of the HRA (as opposed to broadly shared views among elites as to rights protections) are what led to successful collaboration in the Diceyan constitution?
Another response, of course, is that the territorial constitution is not really part of the UK constitution at all. As Kavanagh points out, “[b]eneath the constitutional architecture of legal rules lies constitutional attitudes as political norms” (p. 10). If “constitutionalism [is] mindset, grounded in norms and beliefs,” perhaps devolution is not sufficiently rooted in the minds of the Westminster parliamentarians, the Prime Minister, and the judges of the UK Supreme Court. If it is not collaborative, can it be constitutional?
Erin F. Delaney is the Leverhulme
Professor of Comparative Constitutional Law at UCL. You can reach her at e.delaney@ucl.ac.uk.