On 13 May 2024, Humphreys J in the Northern Ireland High Court delivered his judgment in The matter of Northern Ireland Human Rights Commission, JR295 and the Illegal Migration Act 2023 [2024] NIKB 35. The case dealt with the impact of the Withdrawal Agreement and the Windsor Framework. Re NIHRC had the effect of disapplying most of the Illegal Migration Act 2023 in respect of Northern Ireland. It has had political implications as it seriously undermines the Government’s Rwanda policy. But it also has profound constitutional implications. Firstly, the doctrine of the supremacy of EU law continues to apply in Northern Ireland. Supremacy doesn’t however apply to EU laws generally, but only to the Withdrawal Agreement and to EU laws that are to be implemented under the Windsor Framework. Secondly, the Charter of Fundamental Rights of the European Union is entrenched, to a limited extent, in Northern Ireland. By implication, this also means that those parts of the European Convention on Human Rights which overlap with the Charter are also entrenched. This is not a full entrenchment, only entrenchment where Northern Ireland is acting “within the material scope of EU law” ([63] of Re NIHRC) under the Windsor Framework. This means that any primary legislation that breaches these human rights will be disapplied in Northern Ireland. This paper analyses these two constitutional implications and argues that the judgment is simultaneously orthodox and radical.

Brexit texts affecting Northern Ireland

The Withdrawal Agreement (WA) sets out the arrangements for the UK leaving the EU. That part of the WA which deals with the particular arrangements for Northern Ireland was originally called the Protocol on Ireland / Northern Ireland. Following political developments, the Protocol was renamed the Windsor Framework (WF).

Article 4 of the WA states:

The provisions of this Agreement and the provisions of Union law made applicable by this Agreement shall produce in respect of and in the United Kingdom the same legal effects as those which they produce within the Union and its Member States.

Accordingly, legal or natural persons shall in particular be able to rely directly on the provisions contained or referred to in this Agreement which meet the conditions for direct effect under Union law.

Article 2 of the WF states that as a result of Brexit, there should be no diminution of the rights, safeguards and equality of opportunity set out in the 1998 Good Friday Agreement (expressly referring to protection from discrimination). It then lists six EU anti-discrimination laws, as set out in Annex 1 to the WF, of which there shall be no diminution: “the anti-discrimination laws”. Article 5 states that EU laws set out in Annex 2 to the WF also apply in Northern Ireland. Annex 2 contains a long list of these laws: “the trade laws”.

As Colin Murray points out, “the non-diminution commitment is not simply an international law obligation upon the UK.” The WA and WF were given effect in the UK via two key domestic legislative texts. Firstly, the European Union (Withdrawal) Act 2018 which very broadly governs withdrawal from the EU. Secondly the European Union (Withdrawal Agreement) Act 2020 which very broadly implements the Withdrawal Agreement.

Supremacy of EU law under the terms of the WA

The domestic mechanism for giving effect to the WA and WF is set out in section 7A of the 2018 Act:

What does this mean? On the face of it, it means that EU laws set out in the WF must be given full effect in Northern Ireland, and furthermore, all other laws must be read in such a way as to give them full effect.

S. 7A was not a startingly legislative innovation, but rather an old legislative friend. Any EU or UK constitutional lawyer could not fail to recognise that it is rewrite of section 2 of the European Communities Act 1972. Section 2 was the original “conduit pipe” (see [80] of Miller) that allowed EU law to flow into the UK. The explanatory notes to the 2020 Act make it clear ([109] to [118]) that s. 7A is also intended to act as a conduit pipe. The significance of s. 7A was noted by the Supreme Court in [66] and [108] of Allister [2023] UKSC 5. In Fitzsimmons [2024] NIKB 11, the Northern Ireland High Court clearly stated that by virtue of s. 7A, “any provisions of the [Northern Ireland Troubles (Legacy and Reconciliation) Act 2023] which are in breach of the WF should be disapplied” (at [527]).

With this backdrop, Humphreys J was on firm ground when he stated that “Section 7A mirrors the language of section 2 of the 1972 Act” (at [53]). The 2018 Act used some modern formatting techniques to make the provision admirably clearer, but it is in substance the same provision as the original. Under the terms of the 1972 Act, most memorably set out in Factortame [1990] UKHL 7, the EU law doctrine of supremacy was recognised as part of UK domestic law. That is to say, that an Act of Parliament could be disapplied by the UK courts if it was inconsistent with an EU law provision. One major question for the court in Re NIHRC was therefore – did using the same words in section 7A give the same legal effect as the 1972 Act?  “Yes” was the clear answer, as required by the recent case law set out in the above. “Factortame is still in place since the rights and obligations under the WA must prevail over any inconsistent domestic law” (at [57]). Colin Murray thinks the outcome so predictable that he questions “why this decision came as such a shock?”

Re NIHRC was an orthodox decision. If Parliament uses the same words, it means the same thing.  Parliament chose in the 1972 Act to limit its sovereignty, and it chose to do so again in the 2018 Act. The original logic was that if you want to be in the EU club, you follow club rules. The same logic continues. If you want Northern Ireland to remain (to a limited extent) in the EU club, you still have to follow those limited club rules. Oliver Garner classifies this as “conditional supremacy” of EU law, which is one way of saying that EU law is not supreme in its own right, but is only supreme because the UK Parliament has agreed to recognise it as supreme. Because Parliament has enacted s. 7A, the courts are not placed in the awkward position envisaged by Conor Crummey where they have to judicially disapply statutes on some fundamental common law grounds. Instead, they are acting on clear legislative instruction.

What EU law is supreme?

Under the WF, only some EU law is supreme, and it is only supreme as respects Northern Ireland. My own taxonomy of Humphrey J’s decision is that there are four areas of supremacy:

1. The WA and the WF itself;

2. The trade laws set out in Annex 2 to the WF;

3. The anti-discrimination laws set out in Annex 1 to the WF; and

4. The Charter on Fundamental Rights, as it applies in the context of these three areas.

Art 5(2) of the WF clearly makes the trade laws apply – “the provisions of Union law listed in Annex 2 to this Protocol shall also apply …. to and in … Northern Ireland”. The Government argued that there was merely an obligation of result to ensure that the anti-discrimination laws applied. But Humphreys J ruled against this as the WA, and in particular Article 4, was designed to ensure that people could rely directly upon WF rights.

But arguably the area of EU law whose supremacy is most controversial is the Charter on Fundamental Rights. The ruling was that the CFR was not supreme in all cases, instead “the CFR continues to have effect in UK law in circumstances where ‘Union law’ continues to be implemented” at [65]. In reaching this decision, Humphreys followed the decision in Rugby Football Union v Consolidated Information Services [2012] UKSC 55, in particular at [28] that the CFR has direct effect only when it was “implementing EU law” or “whenever a member state is acting within the material scope of EU law.” Therefore, where: the WA / WF is being implemented, the anti-discrimination laws are being implemented, or the trade laws are being implemented (all in Northern Ireland), then the CFR also applies, and these laws must be implemented in a way that gives full effect to the CFR. Section 5(4) of the 2018 Act states very clearly that the CFR is not part of UK domestic law as a general rule. But under s. 7A, this must give way to the particular provisions of the WF as it applies in Northern Ireland. (And see also Article 6 of the Treaty on European Union, which accords the CFR the same status as the EU treaties.)

In my view, this is the most radical effect of the judgement. The Human Rights Act 1998 requires all legislation to be interpreted, so far as possible, to accord with the European Convention on Human Rights. The most that can be done if an Act of Parliament breaches these rights is that a court can make a declaration of incompatibility under section 4. Under section 6 of the Northern Ireland Act 1998, a provision which breaches these rights is beyond the legislative competence of the Northern Ireland Assembly. But neither of these laws prevented an Act of Parliament having effect. However, what the 2018 and 2020 Act have managed to do is establish a hierarchy of law in Northern Ireland to such an extent that if an Act of Parliament breaches certain human rights (within the context set out above) that Act of Parliament must to that extent be disapplied. Brexit has done more than anything else in entrenching human rights in Northern Ireland. This is not judicial law-making, according to Humphreys J: “this outcome does not occur at the whim of the courts but represents the will of Parliament as articulated in the Withdrawal Act” (at [175]).

The bulk of Re NIHRC then assessed the Illegal Migration Act 2023 against these standards. It concluded that applying the 2023 Act would create a diminution of rights, that this would be contrary to Article 2 of the WF, and that in accordance with s. 7A of the 2018 Act, the court must then disapply those provisions of the 2023 Act. On the application of the NI Human Rights Commission, the court also ruled under its powers in the Human Rights Act 1998, that much of the 2023 Act was in breach of the ECHR, and that it would also be appropriate to make a declaration of incompatibility.

The immediate consequence is that the second applicant in the case must not be removed from Northern Ireland under the terms of the Illegal Migration Act 2023. He is a 16-year-old asylum seeker from Iran who arrived in the UK as an unaccompanied child who claims that he will be killed or sent to prison if returned to Iran. In an abstract discussion on the doctrine of supremacy, it is worthwhile pausing to note that the judgment has effects on real people.

Conclusion

Assuming the case is not overturned on appeal, there are three longer term consequences. Firstly, many sections of the Illegal Migration Act 2023 are disapplied in Northern Ireland – they remain on the statute book but are not properly law. Possibly by extension of this logic, the same may apply to the Safety of Rwanda (Asylum and Immigration) Act 2024 if that Act also is in breach of the CFR. Secondly, and more profoundly, within the material scope of EU law that applies in Northern Ireland as agreed to in the WF, including matters within the CFR, EU law remains supreme and any conflicting Act of Parliament must be disapplied. Although arguably a radical outcome, this has been done in a very orthodox way. Adam Tucker has previously argued, in discussing the Rwanda Act “this may even be a situation where the courts go further: either by straightforwardly refusing to honour the legislation, or by adopting an interpretation so hostile that it is effectively disingenuous and best understood as a disapplication”. But Re NIHRC has disapplied an Act of Parliament by the very conventional route of another Act of Parliament, with judicial authority stretching back to Factortame. Thirdly and more prosaically, the Northern Ireland statute book is now even more complicated. As well as NI Assembly Acts (can be struck down if beyond legislative competence), Westminster Acts (can’t be struck down under sovereignty of Parliament), EU law (was previously inviolable under doctrine of supremacy, but post-Brexit capable of being repealed or reinterpreted), we have the new category of EU law covered by the Windsor Framework (the new top of the hierarchy, which cannot be struck down).  Even if Parliament were to enact a new law which expressly stated that it applied “notwithstanding” s. 7A, it is difficult to see this having effect without reneging on the entirety of the WA and revoking s. 7A.

Dawn Oliver famously characterised Factortame as fishing on the incoming tide.  Post-Brexit, that tide is now going out, but it still applies to asylum seekers in Northern Ireland.

Dr Ronan Cormacain, Consultant Legislative Counsel

Acknowledgements: I am grateful to Adam Tucker and the editors for comments on a draft of this paper, all views and errors contained in the paper are my own.