Irena Sabic KC and Alex Grigg of the Garden Court Chambers Public Law and Immigration Teams provide reflections on the Rwanda Bill (personal views only), with thanks to Liam Mckenna for his contribution.
1. On 14 April 2022, the then Prime Minister Boris Johnson MP announced the ‘UK and Rwanda Migration and Economic Development Partnership’ . Under that agreement, signed that same day by the then Home Secretary Priti Patel MP and Rwandan Minister of Foreign Affairs Vincent Biruta, migrants who entered the UK without permission and via an ‘illegal and dangerous’ route could be removed to Rwanda, rather than being given any right to seek asylum in the UK.
2. The viability of that arrangement depended upon Rwanda being a safe country to receive the migrants sent there for the determination of their refugee claims. The UK government did not attempt – and still has not attempted – to suggest that it should be permitted to go ahead with removals regardless of whether or not Rwanda is in fact a safe place for asylum seekers.
3. By late May 2022, a little over a month after the agreement was announced, the Home Office made the first decisions selecting certain asylum seekers to be removed to Rwanda. It certified their asylum claims as ‘inadmissible’ into the UK procedure because of their method of arrival via a safe third country and rejected claims by those asylum seekers that Rwanda was not a safe place, either because of a risk of harm to them within Rwanda itself  or because Rwanda could not be relied upon to comply with the Refugee Convention and other international conventions to which it was a signatory, leading to a risk of refoulement.
4. Those claims were supported by perhaps the highest authority on issues of refugee status determination, the UNHCR and the international body charged with supervisory status of the 1951 Refugee Convention. However, the Home Office not only rejected any representations resisting removal to Rwanda, but routinely certified those human rights claims as being ‘clearly unfounded’.  Such certification is appropriate only where the claim “cannot on any legitimate view succeed.” 
6. Far from being ‘clearly unfounded’, claims that Rwanda is not a safe country for the removal of refugees have now been vindicated by a unanimous Supreme Court.  There was substantial evidence that refugees sent to Rwanda would face a real risk of refoulement, i.e. of removal to states where they would suffer persecution. Claims about risk of serious harm within Rwanda itself, dismissed in the High Court, were left open by both the Court of Appeal and the Supreme Court.
6. Undeterred by the Supreme Court conclusion, within days, the government has chosen to meet that court’s judgment by going yet another step further. Whether or not Rwanda was safe before, we are told, it certainly is now, the new ‘Rule of Law Partnership’ with Rwanda achieving what the MEDP did not.
7. So certain is the government’s view as to Rwanda’s safety in reliance on that treaty that the courts are prohibited from entertaining any suggestion that it might not be. That is the effect of the Safety of Rwanda (Asylum and Immigration) Bill. Where certification of a human rights claim as ‘clearly unfounded’ precluded a right of appeal to the Immigration Tribunal but left open the possibility of a claim for judicial review in the High Court, now – if the Bill is passed into law – there is to be no judicial scrutiny whatsoever of the government’s position that Rwanda is a safe country.
8. Given how badly wrong the government got its previous assessment of Rwanda’s safety, there seems little reason to be confident that this time its views can be safely relied upon. The Supreme Court not only disagreed with the government’s position, but made a point of noting that pressures placed upon officials led to them reaching conclusions “without close examination of supporting evidence” and that, however well-qualified and informed officials might be, “[m]inisters do not, however, necessarily act on the advice of their officials.” If matters of life and death at stake did not forcefully mark the need for judicial scrutiny, this government’s record of disregarding evidence it does not like in order to justify the conclusions it wishes to reach  would do so just as well. The government in the Supreme Court sought to undermine the UNHCR’s position by suggesting that the UNHCR had an “institutional interest” in the Rwanda scheme failing ; the government’s own interest in the scheme succeeding, and so in glossing over the evidence that Rwanda has been and remains unsafe, seems not to trouble it in the same way.
9. It is therefore to be hoped that the new Bill is rejected by Parliament, so that the issues to which it gives rise remain hypothetical only.
10. For the moment, however, the Bill has passed two votes in the House of Commons. It is impossible to say whether it will falter in the House of Lords, or whether as the Bill progresses the Commons may subject it to a more critical eye.
11. There being a real possibility that the Bill will pass into law, we therefore offer these thoughts on its constitutional and legal ramifications.
Does the Bill break international law?
12. The UK has committed itself by way of numerous instruments, most notably the Refugee Convention, the Convention against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment (1984), the International Covenant on Civil and Political Rights (1966) and the European Convention on Human Rights not to send any person – including any migrant, whether or not they otherwise have a right to remain in the UK – to a state where (i) they will suffer persecution or other serious harm such as torture, or (ii) be sent on to another state where they will suffer that harm.
13. The effect of the Supreme Court’s finding in the Rwanda judgment just last month was that removing asylum seekers to Rwanda would be a breach of all of those Conventions. While a range of issues were raised, the court flagged as particularly relevant to its conclusion the following:
a. Rwanda had a history of disregard for human rights obligations, having been described as recently as 2017, by the High Court in England and Wales, as “a state which, in very recent times, has instigated political killings, and has led British police to warn Rwandan nationals living in Britain of credible plans to kill them on the part of that state”,  and as recently as 2021, criticised by the UK government, for “extrajudicial killings, deaths in custody, enforced disappearances and torture”.
b. Rwanda’s asylum system was not adequate. Claims might be summarily rejected before reaching the main decision-making body; decision-makers were non-specialist and gave perfunctory, if any, reasons for their decisions; there was no example of an appeal ever being brought to the High Court, and there was evidence that the Rwandan judiciary would be reluctant to make findings contrary to the Rwandan government; there were 100%, or very close to that, rejection rates for certain nationalities; and there was evidence of forcible expulsion of asylum seekers without consideration of their claims. While the Rwandan government professed its compliance with the Refugee Convention, it maintained a number of positions clearly contrary to the Convention.
c. There had previously been a removals agreement between Israel and Rwanda which had contained similar assurances of human rights compliance as those now relied upon by the UK government as establishing Rwanda’s safety. That agreement had been found unconstitutional by the Israeli Supreme Court, and it was undisputed that “persons who were relocated under the agreement suffered serious breaches of their rights under the Refugee Convention”.
14. Whatever Rwanda’s good faith in assuring the UK that it would comply with all relevant human rights and other international law, the evidence was thus that Rwanda could not be trusted to make good on those assurances. Whether by practical deficiencies in training, persistent misunderstanding of what international law in fact required, or active disregard of international obligations by individual officials, there was a real risk of Rwanda refouling refugees to countries either directly or indirectly where they would face persecution.
15. The UK government does not suggest that Rwanda has of its own motion fundamentally reformed its asylum system in the past year, so that the Supreme Court’s findings were already out of date when they were made. Rather, its position is that the issues raised by the Supreme Court are completely and unanswerably resolved by its new UK-Rwanda treaty. In particular, the government relies on the following:
a. By contrast to the MEDP, the treaty is binding in international law.
b. There is now more detailed provision in relation to the reception of individuals removed to Rwanda, dealing in granular detail with the quality of accommodation, food, healthcare etc. to be provided.
c. There is now detailed provision for a bespoke ‘Claims process’, in Annex B of the treaty. This involves the establishment of a new ‘Appeal Body’, to be composed of judges from a mix of nationalities, and which will hear claims de novo, i.e. will in effect decide any asylum claim for itself.
d. Regardless of the quality of decision-making in Rwanda, and of the new Appeal Body, in any event no relocated individual is to be removed from Rwanda other than to the UK, where that is requested by the UK: Article 10(3).
16. If the Supreme Court’s finding had been simply that Rwanda’s decision-makers are not good enough to meet the state’s obligations, then this might have been sufficient to answer that concern. Supposing that Article 10(3) of the agreement is complied with, there would not then be a risk of refoulement, as removals would only ever be back to the UK.
17. However, the Supreme Court’s findings were not so limited. As summarised above, there was evidence of regular breach by Rwanda of its international obligations, including treaty obligations, and including human rights commitments. Those breaches occurred even in the context of the similar removals agreement between Rwanda and Israel. Whether or not ‘Rwanda’ at an abstract level is taken to be seeking to comply with its international obligations in good faith, the evidence is that at the level of the administrative decision-making on which compliance depends there is widespread ignorance or disregard of Rwanda’s obligations.
18. Put shortly, the government’s position depends on the treaty to sufficiently conclude there is no risk of Rwanda deviating from its terms. But all of the obligations which Rwanda has previously breached were likewise contained in binding international law. There is no obvious reason to suppose that the obligations in this treaty will fare any better than the others which have in the past been breached by Rwanda. It is not enough, as the government did during the Rwanda litigation, to refer to reputational incentives which will encourage Rwanda to comply. Much the same incentives apply to compliance with human rights law at large, and have not in the past been sufficient to prevent Rwanda breaching its international obligations.
There seems therefore, putting it at its lowest, at the very least an arguable case that the new treaty is not a sufficient basis on which to conclude that Rwanda is now a safe country to which to remove migrants.
19. If it is right that the treaty does not resolve the safety issues identified by the Supreme Court, then removals to Rwanda will remain incompatible with international law. The new Safety of Rwanda Bill, by preventing any legal challenge to those refusals, will be integral in its operation in placing the UK in breach of its obligations.
20. Further, whatever is ultimately made of Rwanda’s safety in light of the new treaty, it seems plainly contrary to the ECHR to exclude any right to challenge removal on the basis of generic safety concerns as the bill expressly seeks to do. The ECHR supplements its substantive rights with a right to an effective remedy at the domestic level. In the past, efforts to designate particular countries as safe have been couched as rebuttable presumptions, to ensure that an effective remedy remained available in the event that the presumption of safety proved false. By excluding all challenge on the basis of generic safety concerns, and by disapplying the Human Rights Act 1998, the Bill ensures that no domestic remedy at all is available, let alone any ‘effective’ remedy.
21. The Joint Committee on Human Rights makes the point (at para 21 of their preliminary briefing dated 11 December 2023) that the Bill cannot act as a practical and effective means of ensuring that the risks identified by the Supreme Court are addressed. This is essentially an evidential question which concerns, amongst other things, the integrity of the Rwandan asylum system. It is not an issue that the UK Parliament should legislate about and certainly not indefinitely into the future as the Bill provides. What if there was compelling evidence (such as that which led to the SC decision) that Rwanda was not safe? Under the Bill, the Courts would still be prohibited from hearing any arguments about such evidence and the implications for asylum seekers because the Bill proposes to legislate for a state of affairs which is conclusive and unquestionable, even if it is, in fact, a fiction. Further and critically, if the UK Gov was able to show that the UK-Rwanda Treaty adequately addresses the SC concerns, the Bill would be entirely unnecessary. It therefore seems that the aim of this Bill is to speed up the operational policy and to do so by prohibiting any judicial scrutiny of the asylum system in Rwanda. This is entirely unprecedented under the common law and in human rights terms and a deeply concerning feature of the Bill.
Is the Bill unconstitutional?
23. The UK legal system does not, as some other countries’ systems do,  automatically import international legal obligations into domestic law. While we might expect the government to honour the its international commitments, a failure to do so is not per se domestically unconstitutional.
24. Similarly, unlike in many other countries, in the UK ‘unconstitutional’ does not necessarily entail ‘unlawful’ (and by the same token, ‘lawful’ does not necessarily entail ‘constitutional’).
25. Whatever the international lawfulness of the Bill and prospective future removals to Rwanda, it is therefore important to ask both:
a. Is the Bill constitutional, i.e. compatible with the norms governing the UK constitution?
b. If unconstitutional, is the Bill also unlawful, i.e. do the norms it contravenes include legal ones?
26. There are plainly significant objections to the Bill. In the face of the Supreme Court’s finding that Rwanda is not a safe country for removals, the Bill seeks to declare that exactly the contrary is the case. While the government presents at least a fig leaf of argument to justify that position, in the form of the new treaty with Rwanda, there is no obvious, or indeed non-obvious, reason why it should not be willing to have that argument tested by the courts. In particular, in circumstances where the government has previously insisted that its contentions about Rwanda’s safety are unanswerable, only to have them comprehensively answered, the need for that scrutiny seems particularly important.
27. In constitutional terms, we see two major, and related, objections to the Bill.
28. First, it is a means for the government to say that it is doing one thing whilst really doing another. The government says that that Rwanda is a safe country; that the government remains committed to its international obligations under the Refugee Convention and the other treaties discussed above; and that as Rwanda is a safe country so that removals can proceed compatibly with those international obligations. But the Bill does not ensure Rwanda’s safety – whatever Parliament’s powers in law, it cannot change reality. Rather, it simply declares that Rwanda’s safety is not to be called into question. What the Bill amounts to in truth is a statement that it is lawful to proceed with removals to Rwanda regardless of whether those removals are contrary to international law.
29. The Bill thus presents itself as deciding a factual issue – the safety of Rwanda – when its effect is to change the applicable principles – the permissibility of removing a person to a country where they will face serious harm or onwards refoulement.
30. It is a basic principle of the constitution that a Parliament which wishes to interfere with fundamental rights must “squarely confront what it is doing and accept the political cost”. This is commonly called the ‘principle of legality’. If Parliament wishes to change the law so as to abrogate the prohibition on refoulement, and the constitutional right of access to justice, it should say so clearly, not seek to achieve that effect by the sleight of hand of a presumption of safety and a confining of a remedy only to ‘individual circumstances’. The Bill in its present form thus flies against the principle of legality.
31. Second, the Bill is contrary to the rule of law. In its most basic, and uncontroversial, formulation, the rule of law requires that we be governed by the application of established principles. It is for Parliament to establish those principles, but for the courts to decide on their application in particular cases.
32. The Rwanda Bill, as it presents itself, is a subversion of that principle. It is purported by the government not to disturb the prohibition on refoulement, but to declare that removals to Rwanda do not breach that prohibition, by means of an irrebuttable presumption of fact combined with an ouster of legal challenge. Moreover, it does so in the face of a recent Supreme Court judgment to the contrary, and in reliance on the views of a government which has shown itself an unreliable judge of exactly this issue.
33. The Bill is thus contrary to the rule of law. It enshrines as law a factual judgment made by the government, in an attempt to immunise that judgment from any legal oversight. In any area this would be objectionable. In such a politically-charged area, where the government has obvious incentives to favour the view that Rwanda is safe, the importance that such questions be decided by the independent judiciary is fundamental, but overridden by the Bill.
34. We do not consider that any of the defences of the Bill on constitutional terms are persuasive. It is not enough to defend the Bill to refer to Parliamentary sovereignty. Whether or not that is enough to make the Bill legally effective, Parliament’s sovereignty is not the only principle on which our constitution rests. While our constitution depends upon political rather than legal enforcement of certain of its norms, that does not make those norms any less important.
35. Nor is it enough to observe that legal fictions, such as conclusive presumptions, are commonplace in other areas of law, such as taxation. There is a world of difference between a presumption intended to anticipate and avoid evidential difficulties, where what is at stake is an amount of tax payable, and a presumption intended to enshrine and insulate from scrutiny the government’s assessment of its own policy.
36. The UK’s political constitution is commonly justified on the basis that political and constitutional controls, and the self-restraint of the various actors involved, avoid any need for legal enforcement of constitutional norms. Passage of the Rwanda Bill into law would mark a substantial departure from our basic constitutional principles.
Lawfulness / legal effectiveness
37. Is the above sufficient to make the provisions of the Bill, if passed into law, somehow unenforceable? There has been substantial discussion of that possibility since the Bill’s publication, drawing upon obiter in Jackson  UKHL 56,  1 AC 262 and Privacy International  UKSC 22,  AC 491.
38. While plainly contrary to the conventional understanding of Parliamentary sovereignty, that Parliament may “make or unmake any law whatever”, the logic of these contentions is not faulty. Whether expressed as a principle of the common law or a more fundamental ‘rule of recognition’ or Grundnorm, the scope of Parliamentary sovereignty depends upon the extent to which it is accepted by all the ‘players of the game’. The courts have never in modern times been asked to hold that Parliament may lawfully send people, potentially, to their deaths in the interest of upholding immigration controls. No purely legal norm dictates the answer that the courts would give if now presented with that stark question.
39. That said, there are strong reasons to consider that the courts would not accept any invitation to depart from the conventional understanding of Parliamentary sovereignty. First, there is the weight of authority and tradition. While never asked to pronounce on this situation, the obiter in Jackson and Privacy International are matched by as many or more assertions that what Parliament dictates, however contrary to basic principle, must be faithfully observed by the courts. The most notable suggestion that there may be limits on Parliament’s sovereignty prior to Jackson was by Sir Edward Coke in Dr Bonham’s Case, some 400 years ago, forcefully refuted at the time and probably recanted by Coke himself in subsequent years.
40. Second, for the courts to decline to enforce Parliament’s legislation would itself raise rule of law issues. In the absence of a written constitution, how are the courts to judge by what norms Parliament is limited, and how can those norms be characterised as ‘legal’ rather than political?
41. Third, for the courts to set themselves against Parliament would be to set up a contest that there is a substantial chance that the courts would lose. Politicians might plausibly claim that any decision declining to enforce the Rwanda Bill (or Act, as it would then be) itself lacks legitimacy so may lawfully be disregarded. In seeking to expand their constitutional role, the courts might end up only undermining themselves.
42. But an express invitation to disapply the Bill is not the only, or even most likely, way in which the courts might be asked to enforce the constitutional norms breached by the Bill. Traditionally, the courts have found ways to police these norms by way of ‘interpretation’ of difficult provisions, as happened most notably in Privacy International.
43. It is difficult to see how the Bill on its current language could be interpreted away as was the ouster clause at issue in Privacy International. Indeed, the Bill expressly anticipates such a challenge, containing provision clearly intended to respond to the reasoning in Privacy International.
44. If no convincing interpretative argument can be found, plausibly the courts might instead be invited on ‘principle of legality’ grounds to find not that Parliament can never disapply the principle of refoulement, or the constitutional right of access to justice, but that if it wishes to do so it must do so clearly and expressly, and that the Rwanda Bill as currently drafted does not achieve this (see our position on the principle of legality set out above). That would still be a substantial extension of existing principle – it would require finding that the apparently conclusive presumption of safety and the ouster of review in the Act are ineffective – but it would have the benefits of (i) preserving Parliament’s ultimate sovereignty and (ii) being firmly grounded in the existing principle of legality.
Alex Grigg was instructed by Duncan Lewis on behalf of a number of clients the Home Office had selected for the Rwanda scheme. All but one of Alex’s individual clients were removed from the scheme following the lodging of grounds. Alex was then instructed for his one remaining client HTN as sole counsel in the ultimately successful High Court, Strasbourg and Court of Appeal interim relief proceedings and was instructed in the generic AAA claim (up to drafting for the Court of Appeal). Toufique Hossain of Duncan Lewis described Alex’s role in the 2022-23 Rwanda litigation as ‘invaluable’.
 Rwanda has a record of responding to protests by refugees and others with lethal violence.
 Section 94(3) of the Nationality, Immigration and Asylum Act 2002.
 ZL and VL  EWCA Civ 25,  1 WLR 1230 at §57, Lord Phillips MR.
 R (AAA & Ors)  UKSC 42,  1 WLR 4433.
 The treaty was described as the ‘Rule of Law Partnership’ by the Home Secretary in an oral statement to Parliament on 6 December 2023.
 Clause 2 of the Safety of Rwanda (Asylum and Immigration) Bill.
 AAA (SC) §§52-53.
 See in particular the government’s dismissal of the past agreement between Israel and Rwanda as irrelevant, AAA (SC) §§95-100.
 AAA (SC) §69.
 If so, at present the most likely source of rebellion appears not to be the more liberal elements in the Conservative party but its right wing, a number of whose members chose to abstain on the Bill’s second reading on the basis that the Bill does not go far enough in excluding legal challenges. Whilst threatening to go further and vote against the Bill at its third reading, it is difficult for the moment to see that as more than a negotiating tactic.
 It is plausible that these obligations are also to be found in customary international law, as the UK has itself previously contended, see §25 of AAA (SC).
 AAA (SC) §76, quoting Government of Rwanda v Nteziryayo  EWHC 1912 (Admin) at §370.
 AAA (SC) §76.
 AAA (SC) §§77-94.
 AAA (SC) §§95-100.
 Underhill LJ in the Court of Appeal at §156 described “a culture of, at best, insufficient appreciation by DGIE officials of Rwanda's obligations under the Refugee Convention, and at worst a deliberate disregard for those obligations”, a remark approved by the Supreme Court at §88.
 ECHR Article 13.
 Following the decision in R (Nasseri)  UKHL 23,  1 AC 1, where a declaration of ECHR incompatibility was issued in relation to a statutory irrebuttable presumption of safety.
 The Bill makes limited provisions for challenges brought on the basis of ‘individual circumstances’, clause 4, but not insofar as those ‘individual circumstances’ are said to create a risk of refoulement, clause 4(2).
 Clause 3.
 Including Rwanda, where the new Treaty is automatically incorporated into domestic law (see Article 3(6)).
 R v SSHD, ex parte Simms  2 AC 115 (HL), Lord Hoffmann at 131E-G.
 Clause 4(1) of the Bill.
 A. V. Dicey, Introduction to the Study of the Law of the Constitution (1885).
 See e.g. AF  UKHL 28,  2 AC 269 at §93, Lord Scott of Foscote.
 See the quotations from Coke in relation to Parliamentary sovereignty given by William Blackstone in his Commentaries, in the passage available online. It is possible to read the remarks in Dr Bonham’s case as in any event limited to suggesting control of Parliament by way of muscular interpretation, rather than the endorsement of express judicial limits on Parliament’s ability to legislate.
 See this view expressed by Dawn Oliver in ‘Parliamentary sovereignty: a pragmatic or principled doctrine?’ (3 May 2012) on the UK Constitutional Law Association blog.
 Clause 7(1) defines ‘decision’ as including a ‘purported decision’, evidently intended to disapply the reasoning in Privacy International. Similar language in the Judicial Review and Courts Act 2022 has been found effective in relation to ‘Cart’ judicial reviews, see R (Oceana)  EWHC 791 (Admin); R (LA (Albania))  EWCA Civ 1337; and in Scotland Sooy  CSOH 93. Notably, however, that provision applies in a context where significant judicial scrutiny will already have been available, and the review sought is of a judicial decision; that is very far from the total exclusion of oversight proposed in the Rwanda Bill.