The myriad problems with the Safety of Rwanda (Asylum and Immigration) Bill, as well as the policy that the Bill is supposed to facilitate, have been clearly documented. One common criticism is that the Bill would precipitate a ‘constitutional crisis’ by provoking the courts into refusing to recognise its legal effect. Adam Tucker argues that the Bill’s most problematic sections could very well ‘count as a novel entry in our canon of possible limits of parliamentary sovereignty’. Jeff King argues that the House of Lords would be justified in radically amending or voting the Bill down, precisely because of the danger that it might incite a constitutional crisis. The Lords suggested sweeping amendments to the Bill, all of which were rejected by the Commons. The Bill will continue to go through a ping-pong process between each House before a final wording is settled.
I agree both with Tucker’s view that the disapplication of the Bill by the courts might be justified and King’s view that the Lords would be justified in amending or voting down the Bill in advance. In this post, however, I wish to take the argument a step further, and challenge the assumption that judicial disapplication of the Rwanda Bill would represent a fundamental shift in the constitutional order. This assumption, which I will call the Orthodox View, seems to underpin most of the commentary on ‘disapplication’. Whether or not we think the courts would be justified in denying legal effect to the Rwanda Bill, it is assumed that doing so would represent the crossing of a constitutional Rubicon. Implicitly, this seems to count as a reason against denying legal effect to the Bill. Perhaps disapplication would be justified, but the courts should think long and hard before taking this step.
There is, however, an alternative view available, according to which the disapplication of a statute would simply represent the application of familiar principles of statutory interpretation in a novel context. On this view, judges faced with a statute purporting to empower the government to deport asylum seekers to a country with which they have no prior connection, without any effort to guarantee their life and safety, have always been not just empowered but required to hold that such a statute has no legal effect. It’s just that up until now, they have never been faced with such a statute. The constitutional principles are not new; the factual context is.
‘Disapplication’
In several obiter dicta, judges of the UK Supreme Court indicated that there might be legislation purporting to license a violation of common law principles so egregious that the courts would hold that such legislation has no legal effect. In R (Jackson) v Attorney General, Lord Steyn gave the example of legislation that sought to abolish judicial review, noting that in such circumstances, the Court ‘may have to consider whether this is a constitutional fundamental that even a sovereign Parliament acting at the behest of a complaisant House of Commons cannot abolish’ [102]. Lady Hale and Lord Hope made similar statements in the case. In AXA v Lord Advocate, Lord Hope said, ‘The rule of law requires that the judges must retain the power to insist that legislation of that extreme kind is not law which the courts will recognise’ [51]. More recently, in R (Privacy International) v Investigatory Powers Tribunal (PI), Lord Carnwath noted that ‘binding effect cannot be given to a clause which purports wholly to exclude the supervisory jurisdiction of the High Court’ [144].
It is difficult to describe the eventuality foreseen by the courts in language familiar to UK public law. From one point of view, they could be seen as envisaging something like strong-form constitutional ‘strike down’. But this comparison is imperfect. In the US, the effect of a successful facial challenge to the constitutionality of legislation is that future courts are directed to treat the offending legislation as though it never existed. The political act of legislative enactment is in effect erased, and the government is pre-empted from being able to use the same legislation again in a different context. There is nothing for future courts to interpret. In the UK context, these institutional consequences would not be the same. As a technical legal matter, since the legislation would remain on the statute books, future courts could still conceivably interpret it as having some impact on our legal rights and obligations if litigated in some new context. This seems more akin to an ‘as applied’ constitutional challenge than a facial one.
Tucker and King, in their recent posts, both speak of the possible ‘disapplication’ of the Rwanda Bill, invoking the kind of language used in cases where domestic legislation violated EU law. There is an important difference, though, between the kind of judicial action contemplated in Jackson, AXA and PI and the kind of ‘disapplication’ established in R v Secretary of State, ex p Factortame (No 2). The latter kind of ‘disapplication’ was contingent. Primary legislation that was ‘disapplied’ because of a conflict with EU law could remain on the statute book, and could in principle regain its legal effect were the relevant EU laws to change. The Rwanda Bill is different. While the words could remain on the statute book, it is difficult to see how it could ever have legal effect, given that its conflict is not with another body of law, but with deeper constitutional principles.
The judicial action contemplated in Jackson, AXA and PI, then, seems to occupy a theoretical space somewhere between strong-form ‘strike down’ and a Factortame-style ‘disapplication’. The statute would remain on the books, but it is difficult to see how it would ever in practice have legal effect. This does not clarify matters entirely, but it is enough to go on for now. We can say that the Rwanda Bill, if ‘disapplied’, would remain on the books, but that it would have no legal effect in the range of cases to which it was tailored to apply. I will continue to use ‘disapplication’ as shorthand.
The Provisions
The most objectionable provisions of the legislation that was brought through the Commons are well documented elsewhere. I will touch on them only briefly here.
Clause 2(1) of the Bill requires that decision-makers ‘must conclusively treat Rwanda as a safe country’ for the purposes of immigration decisions. Tucker describes this clause as an attempt to ‘legislate Rwanda to safety’, an effort which he correctly notes is impossible. But in fact the tenor of this clause is even more troubling. It does not say that Rwanda, as a matter of UK law, is a safe country. It merely instructs both decision-makers and courts to ‘conclusively treat’ Rwanda as safe for those being deported to it, whether it is or not. The actual safety of Rwanda, as far as the Bill is concerned, is neither here nor there.
The Bill thus purports to remove one of the most basic procedural rights in any legal system: the right to have the merits of one’s case considered by an administrative decision-maker. The merit of one’s case, in this particular context, turns on precisely the question that decision-makers and courts are instructed to disregard: the safety of Rwanda. The importance of this procedural right is significantly heightened in cases where the applicant’s life and safety are at risk. The importance of the right to petition a court to review such a decision is in turn significantly heightened.
The House of Lords recommended amending this clause to add the proviso: ‘unless presented with credible evidence to the contrary’. The Lords also removed important sub-clauses that sought to oust judicial review of immigration decisions brought on the basis that Rwanda is not a safe country. These amendments were rejected by the Commons.
Clauses 4(1) and 4(2) of the Bill state that decision makers and courts are not precluded from deciding whether Rwanda is a safe country ‘based on compelling evidence relating specifically to the person’s particular individual circumstances (rather than on the grounds that the Republic of Rwanda is not a safe country in general)’. It might be argued that this guards against the kind of rule of law violation identified above. As the Court in R (AAA) v Home Secretary found, however, Rwanda’s asylum system is systematically inadequate. (See in particular [77]-[94]. Tucker also summarises the key findings here.) The changes necessary to remedy Rwanda’s asylum system, the Court noted, ‘may not be straightforward, as they require an appreciation that the current approach is inadequate, a change of attitudes, and effective training and monitoring’ [104]. The dangers posed to individuals by Rwanda’s asylum, in other words, will generally arise because of the inadequacy of Rwanda’s asylum system ‘in general’, not because of that individual’s personal circumstances.
Moreover, one of the most serious dangers identified is the possibility that those sent to Rwanda will face clandestine refoulement to a third country. After Rwanda entered into a similar scheme with Israel, more than 100 individuals were transferred to Uganda in violation of the agreement (R (AAA) at [96]). Clause 4(2) of the Bill as originally brought through the Commons provided that decision-makers and courts, when considering under clause 4(1) whether a person’s own circumstances meant that Rwanda would be unsafe for them, could not take into account the likelihood that Rwanda would transfer them to a third country.
Clause 4 was removed in its entirety by the House of Lords. A new clause was inserted explicitly stating that decision-makers may take into account whether Rwanda is a safe country ‘for the person in question or for a group of persons to which that person belongs’. A subsection of that clause also explicitly provides that decision-makers and courts may take into account the risk of Rwanda removing the person to another state. Another new clause provides, ‘The court or tribunal may grant an interim remedy that prevents or delays, or that has the effect of preventing or delaying, the removal of the person to the Republic of Rwanda’. Again these changes were rejected by the Commons.
These are only what might be viewed as the domestic constitutional concerns. Daniella Lock, in an analysis now vindicated by the Supreme Court decision in R(AAA), convincingly demonstrates the Rwanda Bill’s incompatibility with the ECHR and the UN Refugee Convention. As King points out, the Rwanda Bill, in facilitating the policy, would entail ‘a systematic breach of the UK’s international law obligations’. The respect for such obligations, it has been accepted by various constitutional actors, is fundamental to the rule of law.
It is not difficult to see why talk of judicial disapplication of the statute has arisen. Fundamental human rights, administrative justice as a subset of the rule of law, profound separation of powers concerns and international legal obligations are nested together in this context. The Rwanda Bill runs roughshod over each of them.
Below, I sketch two theoretical models of judicial disapplication that can be used when thinking about the Bill. It should though be noted that the Bill will now go through a ‘ping-pong’ process between the Commons and the Lords. It is not yet clear what any final legislation will look like, and so we cannot yet say whether its disapplication by the courts would be justified. My aim here, therefore, is not to show conclusively that the Rwanda Bill’s disapplication would be the correct decision. Rather, it is to show that the courts holding that a statute has no legal effect need not be viewed as a constitutional shift or revolution. This defuses one important argument against disapplication.
The Orthodox View – A New Constitutional Order
Suppose the Supreme Court was to hold that the Safety of Rwanda Act had no legal effect. On the Orthodox View (OV), this would represent a fundamental shift in the constitutional order.
Underpinning the OV is the view that Parliament is still, as a legal matter, sovereign; it can, through authoritative acts of will, create any legal obligation it sees fit. Cases in which statutory interpretation seems to depart significantly from parliamentary intentions pose a challenge to this picture. In these cases, courts seem to hold that a statute’s legal impact is not solely determined by the intentions of the legislature (assuming for the sake of argument that a coherent conception of legislative intention exists). These cases can be explained away in various ways. We might say, as long as the legislative wording is sufficiently imprecise, that such interpretations are consistent with a more abstract parliamentary intention to legislate consistently with the rule of law or common law principles. Alternatively, we might draw a distinction between ‘legal’ and ‘constitutional’ constraints on Parliament’s authority (see for example Mark Elliott, ‘Parliamentary Sovereignty’ in J. Jowell and C. O’Cinneide (eds) The Changing Constitution (9th edn, OUP 2019), 54).
These sorts of explanations, however, would not be available were the courts to hold that a precisely-worded statute had no legal effect. According to the OV, this would see the court crossing a constitutional Rubicon. It would bring about a shift in the constitutional order itself; where before it was accepted that Parliament could still, as a purely legal matter, ‘make or unmake’ any law that it wished, in the new constitutional world this would no longer be the case.
This shift would, on this view, prefigure a ‘constitutional crisis’, as different political institutions argued over which vision of the constitution was the correct one. This is not to say that scholars who implicitly adopt the OV cannot argue in favour of a statute’s disapplication. The precipitation of a constitutional crisis seems to count as one reason against disapplication, but this reason could be outweighed by other, weightier reasons. The Rwanda Bill might be so egregious in its disregard for the rule of law that the courts would be justified disapplying it notwithstanding the constitutional crisis that would be triggered.
The Alternative View – A New Context
An Alternative View (AV) of this picture is available. The starting point for the AV is that Parliament has neverbeen absolutely sovereign in the way that public law theorists generally mean. Law-making authority, according to the AV, has always been constrained by other constitutional principles.
To understand this, it helps to think of the principle of parliamentary sovereignty not as a ‘political fact’, but rather as a principle of political morality; a democratic principle. This is not to say that parliamentary sovereignty was, as a historical matter, understood in these terms; only that the doctrine can in the present day be understood as underpinned by such a value. We see this kind of view of parliamentary sovereignty at play in both Miller decisions, for example, in which Parliament’s democratic credentials militate against attempts to undermine its sovereignty with the royal prerogative. According to the AV, Parliament has law-making authority because of its democratic mandate. For this reason, legislative intentions matter a great deal and must be taken seriously. When interpreting a statutory provision, courts ask what legal effect Parliament intended the legislation to have, because democracy makes that intention relevant.
This, however, does not make that law-making authority absolute. If legislative authority is underpinned by a democratic principle, then it is natural to think that that authority will be constrained by other principles. It would be a very thin, majoritarian conception of democracy that justified unconstrained legislative authority. When a judge tries to figure out the legal impact of a statute, according to the AV, they appeal both to this democratic principle and to other relevant principles. Typically, these other principles are expressed in terms of the ‘rule of law’.
To illustrate this further, it may be helpful to consider the following, well-known extract from Lord Steyn’s judgment in ex p Pierson:
There is no ambiguity in the statutory language. The presumption that in the event of ambiguity legislation is presumed not to invade common law rights is inapplicable. A broader principle applies. Parliament does not legislate in a vacuum. Parliament legislates for a European liberal democracy founded on the principles and traditions of the common law. And the courts may approach legislation on this initial assumption. But this assumption only has prima facie force. It can be displaced by a clear and specific provision to the contrary. ([1998] AC 539, 587-588.)
Lord Steyn is here setting out an account of the principle of legality. His statement reflects the idea that living in a liberal democracy requires commitment to a particular set of principles and values. It reflects the idea, further, that the UK has committed to those values in its common law tradition. This includes a strong commitment to the democratic principle that gives Parliament law-making authority, but it includes other principles as well. This is why the courts will, where possible, interpret legislation consistently with these other principles. This is typically framed in terms of a ‘presumption’ about Parliament’s intentions, but this might be viewed as shorthand for a more complex process of normative reasoning, one that appeals to the kind of principles referred to by Lord Steyn.
Lord Steyn is of course explicit that this presumption can be displaced by specific wording. One might view this as support for the OV. Lord Steyn’s subsequent judgment in Jackson, however, takes his earlier reasoning a step further:
[The UK constitution] is a legal order in which the United Kingdom assumes obligations to protect fundamental rights, not in relation to other states, but towards all individuals within its jurisdiction. The classic account given by Dicey of the doctrine of the supremacy of Parliament, pure and absolute as it was, can now be seen to be out of place in the modern United Kingdom. Nevertheless, the supremacy of Parliament is still the general principle of our constitution. It is a construct of the common law. The judges created this principle. If that is so, it is not unthinkable that circumstances could arise where the courts may have to qualify a principle established on a different hypothesis of constitutionalism. ([2005] UKHL 56 [102].)
Here we see the same sorts of concerns that animated his earlier decision in Pierson, namely the interplay between various constitutional principles, and the ways in which those principles might constrain one another. What is new is the recognition of a new set of (at the time still hypothetical) circumstances, one that would justify the conclusion that a statute had no legal effect.
According to the AV, we can view the judgment in this way. Parliamentary sovereignty is the ‘general principle’ of the Constitution in the sense that I described above, i.e. a democratic principle that holds great weight in the UK constitutional order and will be determinative in most cases. It is, however, not the only law-determining principle. Its role in particular cases is determined by the context in which it operates, and by the weight of other relevant principles. This is why the courts ‘may have to qualify’ it where it impacts other fundamental principles.
To repeat, it might be that the democratic principle underpinning legislative supremacy is so important that the correct interpretation of a statute almost always corresponds with the intentions of the legislature that enacted it. But even if that is the answer in almost every case, the Court in every case silently asks the question of whether that intention is constrained by other constitutional principles. This is why Trevor Allan, for example, argues that viewing Parliamentary Sovereignty as ‘absolute’ can only ever mean a ‘very rough approximation of the requirements of legislative supremacy, apt to describe the likely outcome of certain kinds of political action’ (The Sovereignty of Law (OUP 2013), 152).
According to the AV, then, the reason that the courts have never held that a statute has no legal effect is not that they are constitutionally disempowered from doing so. Rather, the reason is that such a judgment has never yet been justified. The democratic principle that gives Parliament its strong law-determining role must be taken seriously. When interpreting a statute’s legal impact, the courts must first look to what the legislature intended. But the effect of this intention has always been silently constrained by the rule of law (or not so silently, as in Pierson and other similar cases). The courts have never been faced with a statute so egregious in its violation of the rule of law that the weight of legislative intention is entirely nullified. If the courts were faced with such a piece of legislation, however, their holding that it had no legal effect would not represent a radical constitutional shift. Rather, it would represent the application of ordinary principles of statutory interpretation of the kind we see when judges appeal to the principle of legality, applied in a novel context. It is the factual context that would be different; not the constitutional principle.
A judge that held that the Rwanda Bill had no legal effect, on this view, would be doing the same thing that judges always do when they appeal to the principle of legality: trying to work out the legal impact of a statute by appealing to the relevant principles.
Conclusion
I have not sought to make a full argument that disapplication of the Rwanda Bill would be the correct legal outcome. It is my view that this would be the case if the courts were presented with the Bill as it was brought through the Commons. My aim with this post, however, has been a narrower one: to draw attention to a view according to which the disapplication of any statute is in principle justifiable according to the principles of the Constitution as it currently is. Such a move, on this view, would not represent a radical constitutional shift. This removes one important argument against disapplication of the Rwanda Bill.
An important consequence of this view is that disapplication would not justify any constitutional retaliation from other institutions. The OV, recall, views the situation in essentially gladiatorial terms: the courts strike the first blow by refusing to acknowledge a statute’s legal effect, the government responds either by ignoring this judgment or by enacting legislation to curb the courts’ powers; the court has to decide how to respond in turn, etc. My concern is that casting the debate in these terms effectively legitimates the second move – the restriction of judicial review – as an inevitable response to the judicial ‘provocation’ of holding that a statute had no legal effect. But determining precisely what effect a statute has is what judges are always required to do. According to the AV, that the courts have not yet held that a statute had no legal effect is merely a consequence of the fact that they had until now never been presented with a statute that would have justified that conclusion. The courts making such a finding would in no way justify constitutional retaliation from other institutions of government. The constitutionally objectionable move in all of this is not denying the legal effect of a statute directing decision makers to ignore the question of a country’s safety when deciding whether to deport someone there. The constitutionally objectionable move is bringing such legislation forward to begin with.
The author would like to extend his warm thanks to Donal Coffey, Dimitrios Kyritsis, Daniella Lock, Paul Scott and Se-shauna Wheatle for generous comments and feedback.
Conor Crummey is an Assistant Professor at Maynooth University School of Law and Criminology. Some of the theoretical issues explored in this post are considered in more detail in a forthcoming monograph: The Principle of Legality: A Moral Theory (under contract with OUP).
(Suggested citation: C. Crummey, ‘The Safety of Rwanda (Asylum and Immigration) Bill and the Judicial ‘Disapplication’ of Statutes’, U.K. Const. L. Blog (26th March 2024) (available at https://ukconstitutionallaw.org/))
