Brexit has posed a profound challenge for Northern Ireland’s fragile constitutional arrangements. Those very arrangements provided for a delicate balance of Northern Ireland’s constitutional position. Northern Ireland is part of the UK only with the continuing consent of the people of Northern Ireland, and the 1998 Agreement provided for a range of measures which connected Ireland and Northern Ireland together. This smoothing of some of the harder edges of statehood was long opposed by sections of Unionism within Northern Ireland. 

Indeed, the playing of the “Orange Card”, or threatening to make the constitutional space of Northern Ireland ungovernable by widespread demonstrations, was as prominent in Northern Ireland’s creation as it was in the aftermath of the Sunningdale Agreement, the Anglo-Irish Agreement and now in the response to the Protocol on Ireland and Northern Ireland. A legal action in which the lead protagonists storm against the UK-EU Withdrawal Agreement reducing Northern Ireland to the status of “Vichy France” or to “Vassalage” can be seen to be using the judicial review process to stoke wider resentments. That is the backdrop to today’s important decision in Re Jim Allister’s application for Judicial Review in the Northern Ireland High Court.

The entire challenge is based upon a paradox. The applicants insisted that it is unconstitutional to treat Northern Ireland differently with regard to trade in goods from other parts of the United Kingdom. The argument they present to the public is one of integrationist Unionism; to paraphrase Margaret Thatcher, that Northern Ireland is as British as Finchley. The argument presented to the High Court, however, was that Northern Ireland is a unique constitutional space in which ordinary rules have been modified. For if Northern Ireland is just like Finchley, then Parliament can pass legislation to rearrange trading rules in different parts of the UK (China, for example, goes further by holding multiple WTO memberships). 

The challenge is also laced with irony, particularly the sight of Jim Allister QC, one of the 1998 Agreement’s staunchest critics, browbeating other Unionist politicians into a legal action which in large part turns on the Agreement’s terms. In irony heaped upon irony, the person making the arguments that Northern Ireland is governed under special constitutional rules which restrict parliamentary sovereignty is Northern Ireland’s former Attorney General, John Larkin QC, one of the most strident critics of what he regards as the legalisation of the political sphere. Perhaps it is only lawfare when someone else is doing it. 

In response to these claims, Colton J offers a much more detailed reflection on what is distinct about Northern Ireland’s constitutional order than the UK Supreme Court managed in its afterthought on Agnew/McCord as part of the 2017 Miller decision. Those scant few paragraphs on Agnew/McCord, in saying little more than that the 1998 Agreement’s principle of consent did not block Brexit, amounted to an acknowledgement that some form of Brexit could be compatible with the 1998 Agreement. The challenge to the Protocol instead deals with whether the concluded Withdrawal Agreement actually landed in a space which is acceptable under Northern Ireland’s constitutional arrangements, and the Court’s emphatic conclusion is that it did.

The first ground of challenge related to the Act of Union and specifically Article VI on people living in Ireland and Great Britain being ‘on the same footing’, at least with regard to trade. As Colton J declared, tongue firmly in cheek, ‘much constitutional water has passed under the bridge’ since the Act of Union [96]. Indeed, many stage 1 law students will be able to dive back to 1872 and Ex parte Canon Selwyn, on the disestablishment of the Church of Ireland to show that Parliament can modify the Act of Union. Colton J did not dispute that the terms of the Protocol cannot be squared with Article VI [62]:

Although the final outworkings of the Protocol in relation to trade between GB and NI are unclear and the subject matter of ongoing discussions it cannot be said that the two jurisdictions are on “equal footing” in relation to trade. Compliance with certain EU standards; the bureaucracy and associated costs of complying with customs documentation and checks; the payment of tariffs for goods “at risk” and the unfettered access enjoyed by NI business to the EU internal market conflict with the “equal footing” described in Article VI.

This might sound like a significant admission regarding Northern Ireland’s post-Brexit constitutional status, but the rights and privileges of the people of Northern Ireland have repeatedly been changed (relative to those applicable in Great Britain) in the centuries of Union, not least by the Government of Ireland Act 1920 and Northern Ireland Act 1998. In short, devolution would be impossible if the Act of Union could not be varied. 

The challenge, however, proceeded on the basis that any modification to such a constitutionally significant statute could only have been made in express terms [87-88]. Colton J took the starting point that the later statutes (the European Union (Withdrawal) Act passed in 2018 and modified by the European Union (Withdrawal Agreement) Act in 2020) supersede the Act of Union. That is a classical assertion of implied repeal, and that classical account has mattered in the past; the Act of Union has previously been impliedly altered in its operation [102].

It is, however, more difficult to square with some of the discussion in the Supreme Court’s HS2 decision, which maintained that where two “constitutionally significant” statutes conflict, the court should undertake a substantive evaluation of the relationship between the statutes. Although that decision is not explicitly referenced, Colton J proceeds to explore the context of the two enactments, contrasting the generality of the Act of Union with the specific requirements of the Withdrawal legislation (comparable to Beatson J’s weighing up of statutes in Brynmawr). Colton J thus recognised that the Act of Union was impliedly repealed insofar as the Withdrawal legislation conflicted with it.

The issue of clarity of law is important – the terms of the Withdrawal legislation might be specific in legal terms, but they are opaque. The Protocol too often seems to say one thing in terms of trade, but operate in a different way in practice. In not fully explaining these terms in the hurried passage of the Withdrawal Agreement Act 2020, the UK Government fell into the pattern of historic involvement in the European project set by the European Communities Act 1972 and its concealment of the impact of EEA and then EU membership on parliamentary sovereignty. Boris Johnson himself, the Court notes, added to this confusion in ministerial statements [117]. This obfuscation is not, for Colton J, unlawful, but it is hardly a shining example of good governance.

The second ground was that the Protocol conflicted with the Northern Ireland Act 1998, in transferring powers over certain areas of law making for Northern Ireland to the EU without the consent of the people of Northern Ireland. This issue, however, was unanimously settled by the UK Supreme Court in Miller; the reach of the 1998 Agreement’s principle of consent does not extend further than Northern Ireland’s place in the UK. There really was no scope for Colton J to strike out on his own against that position [137].

The third ground related to the “Stormont lock”, under Article 18 of the Protocol, which allows the Northern Ireland Assembly to vote on the continued application of the Protocol’s trade provisions in 2024. The challenge was on the basis of its operation as a majority vote in the Northern Ireland Assembly, rather than a cross-community vote (as would be normal on devolved issues, but would have allowed Unionists to veto the continued application of these Protocol provisions). For the UK Government, as summarised by Colton J, the overarching trade rules applicable to Northern Ireland were not ordinarily a devolved issue, and this therefore allowed Westminster to put in place a specific rule that departed from cross-community voting in this instance [181]:

Because it was not a devolved matter or a matter within the legislative competence of the Assembly as a matter of principle it did not require cross-community support. Consistent with the approach to the referendum concerning exit from the EU itself it was felt that a simple majority would be sufficient, albeit that cross-community support would be encouraged and achieved if possible.

The Court broadly agreed that this was a matter of international relations, and thus not a transferred or devolved matter, and that a special provision could thus be put in place (even by statutory instrument). The terms of the Protocol on Ireland/Northern Ireland (Democratic Consent Process) (EU) Regulations 2020 can hardly be said to be outwith the parent Withdrawal legislation [206].

The fourth ground, on the application of Article 3 of Protocol 1 ECHR and the democratic deficit around the application of EU law rules, was more speculative. It is impossible to conclude that Northern Ireland has as much say in the creation and operation of these rules as when it was part of a Member State; but the Court concluded that the Withdrawal Agreement was concluded as a democratic exercise by the UK Parliament [263]. It is, moreover, worth noting that mechanisms like Article 15’s Joint Consultative Working Group continue to provide mechanisms by which Northern Ireland’s representatives can be involved and feed into, EU decision making over the Single Market in goods. If the rules applicable to Northern Ireland were not within the margin of appreciation available under the ECHR, then this would have dramatic consequences for countries like Norway and Switzerland which have long been “rule takers” in their relationship with the EU. 

The final ground for challenge was that the Protocol breached Article 50 of the Treaty on European Union, in providing for the continued application of EU law outside the EU. In short, Article 50 did not provide a legal basis for concluding an ongoing legal relationship. But under the Protocol Northern Ireland is not part of the Single Market for goods, the UK as a non-Member State is choosing to apply the rules of the Single Market to this territory as if it were part of it. The Court found nothing within Article 50 to prevent the conclusion of such terms [291].

The argument in the round was thus not over whether Northern Ireland’s constitutional order is special; if it was not special then much of the last five years of debate over Brexit would not have happened and there would have been no more impetus to reach a special Brexit accommodation for Northern Ireland than there was for Scotland. The votes in favour of remain in these two constituent parts of the UK, in short, mattered less than the distinct legal basis of their current constitutional arrangements. Alongside exploring aspect of that order, Colton J placed considerable weight on parliamentary sovereignty and the UK Parliament having endorsed these Brexit outcomes for Northern Ireland [297]:

The court should not interfere with or ignore the clearly expressed will of Parliament in passing primary legislation to implement a valid agreement between contracting parties, both of which endorsed that Agreement through their respective constitutional orders.

The High Court’s decision therefore comes as little surprise. The Protocol was designed to accommodate the terms of the 1998 Agreement and to operate without prejudice towards it (as is made explicit in Article 1). It would have been a colossal failure of years of negotiations for all parties to have missed a hard and fast rule of the 1998 Agreement that conflicted with the Protocol’s terms. Some of these arguments, however, speak to the ambiguous nature of centuries old constitutional documents and also to myth making and even misinformation surrounding aspects of the 1998 Agreement. 

The Agreement was concluded against a backdrop of EU membership and a UK constitutional order which was only introducing devolution and had largely outsourced the management of rules that impinged on trade between its constituent parts to the rules of the EU Single Market. If this ecosystem had not existed, the 1998 Agreement would have had to be much more extensive to provide for it. But just as the Agreement did not preclude the UK’s Withdrawal from the EU, neither do its explicit terms preclude the UK from exercising its statehood over Northern Ireland by agreeing that it be subject to distinct goods regulations and trading rules to other parts of the UK. For Northern Ireland, it must be remembered, is not in the EU Single Market, much less the EU. The rules for the Single Market in goods continue to apply only because this was accepted by the UK and EU as supporting the functioning of multiple elements of North/South co-operation which had developed under Strand 2 of the 1998 Agreement. This deal has undoubted become one of the ‘essential elements of the architecture of the modern United Kingdom’ (Somerville v Scottish Ministers, [169]).

The outcome of this case provides an opportunity for the Unionist parties in Northern Ireland to re-evaluate their approach to the Protocol. As Colton J noted in his decision, talk of Northern Ireland as Vichy France is ‘wide of the mark’ [218]. This rhetoric helped neither the argument in this case, nor public discourse over Northern Ireland’s post-Brexit governance. Talk of an appeal has already begun, but in the face of this decision, continuing to insist on the unlawfulness of the Protocol will do little more than ratchet up tensions in Northern Ireland. This is as good an opportunity as there is for some Unionist politicians to say that now that the High Court has clarified the position in law it is time to move on with the important business of engaging in discussions over mitigating some of the problems which have emerged with the Protocol’s workings. 

Colin Murray, Reader in Public Law, Newcastle Law School, Newcastle University.

(Suggested citation: C. Murray, ‘Vichy France and Vassalage: Hyperbole versus the Northern Ireland Protocol’, U.K. Const. L. Blog (1 July 2021) (available at https://ukconstitutionallaw.org/))