Whither the Refugee Convention? R (AAA) v SSHD and a vision of refugee law that applies only to the Global South

Blog post by Nicola Palmer, Reader in Criminal Law, Kings College London, Dickson Poon School of Law and author of ‘Courts in Conflict: Interpreting the Layers of Justice in Post-Genocide Rwanda’.

On 29 June 2023, the Court of Appeal found, in a majority decision, the UK’s Asylum Partnership Arrangement (APA) with Rwanda to be unlawful. The court concluded that:

there were substantial grounds for thinking that there were real risks that the asylum seekers…would be refouled or subject to breaches of article 3, or that their asylum claims would not be properly or fairly determined in Rwanda (para 13: iv).

The focus of the judgment is on the safety, not of Rwanda generally, but of its individual “Refugee Status Determination” (RDS) process in particular.

The decision has been greeted with loud applause and a sense of relief by those fighting against the current UK’s government systematic political and legislative campaigns targeting vulnerable people migrating to the UK. The fight for a humane migration system is a crucial one, but the Court of Appeal decision is a deeply constrained victory in this regard. The focus on Rwanda, with a scathing critique of its systems and processes, stands in some contrast to the very narrow reading of the UK’s own international legal obligations to offer refuge to people arriving on its shores.

If we focus on the UK instead of on Rwanda, the Court of Appeal has unanimously upheld the UK’s externalisation of its legal obligations under the 1951 Refugee Convention. Throughout my observation of the oral arguments in R (AAA) v SSHD, Lord Pannick KC acting on behalf of the Secretary of State foregrounded the arguments in favour of a very narrow reading of the UK’s obligations under this Convention. They have won on this ground. The Vice-President of the Court of Appeal Civil Division, Lord Justice Underhill at para 316 states, “the Refugee Convention does not prohibit a receiving state from declining to entertain an asylum claim where it can and will remove the claimant to another non-persecutory state”.

In the wake of the Court of Appeal’s decision, this displacement or externalisation of the UK’s international legal obligations to assess an asylum claim has received much less attention than the decision on the safety of Rwanda. However, it has very serious implications for the racialised global inequality of mobility that increasingly dictates where the world’s most vulnerable people can move to. The Court of Appeal’s interpretation of the Refugee Convention displaces the UK’s obligations to offer refuge onto other countries if they are deemed safe. Such a reading coheres with, and indeed reinforces, the current UK Illegal Immigration Bill and renders less significant the House of Lords’ most recent amendment stating that this Bill cannot conflict with the UK’s international legal obligations.

The Court of Appeal upheld the Divisional Court’s reasoning that the Refugee Convention only establishes a narrow obligation on the UK to avoid refoulement:

Mr Drabble KC submitted that the Refugee Convention imposes an obligation on contracting states to determine all asylum claims made, on their merits. We disagree. There is no such obligation on the face of the Convention. The obligation that is imposed is the one at article 33, not to expel or return a refugee to a place where his life or freedom would be threatened by reason of any of the characteristics that the convention protects. (para 321 referring to para 121 of the earlier judgment)

It is worth stating in full the purposive reasoning of the Refugee Convention that was put before the Court of Appeal by Mr. Richard Drabble KC, acting on behalf of two of the claimants. In the Court’s own account, he argued that there was an implied obligation on a state to assess the refugee status of those people who arrive on its territory. This arises

“as a combination of the declaratory nature of refugee status (which requires investigation of the individual’s circumstances and the nature of his claim) and Convention provisions, including the prohibition on refoulement (Article 33), the prohibition on penalties for illegal entry or presence (Article 31), the duty to afford refugees the same treatment as ‘aliens’ (Article 7), the prohibition on discrimination on grounds of country of origin and the right of access to courts (Article 16). Even Article 9, which permits provisional measures against an individual where ‘essential to national security’, does so only ‘pending a determination by the Contracting State that that person is in fact a refugee’.” (para 310).

The Court’s basis for dismissing this submission and supporting an interpretation that the Refugee Convention only establishes an obligation to ensure that a person is not sent to a third county in which they would face persecution and other forms of serious harm turns largely on the emphasis placed on articles 25-27 of the EU Council Directive 2005/85/EC.

In the Court’s account, Directive 2005/85/EC recognises the legitimacy of treating an asylum application as inadmissible, and refusing to consider it, if a country other than a member state “is considered as a safe third country for the applicant” (article 25.2 (c)).  This goes beyond the “Dublin system”, which since 1997, has permitted member states to decline to entertain asylum applications from claimants who had previously applied to another member state.

The Directive extends the externalisation of the obligation to assess an asylum claim to any other ‘safe third country’, however, it recognises that this practice is constrained by criteria of what constitutes a safe third country and would be subject to rules laid down in national legislation, including:

rules requiring a connection between the person seeking asylum and the third country concerned on the basis of which it would be reasonable for that person to go to that country.

The Court of Appeal does not discuss this rather minimal but potentially important constraint in which people seeking asylum can only be sent to a ‘safe third country’ with which they have a connection. Instead, it focuses entirely on how this Directive supports an exclusive focus on non-refoulement as the only obligation that the UK owes to people who arrive on its territory.

The Court’s focus on refoulement then determines its decision on whether or not sending someone to Rwanda constitutes a penalty under Article 31 of the Convention. In a chilling paragraph, Lord Justice Underhill states:

In short, it is in my view inconsistent with the well-recognised scheme of the Convention that the expulsion of a migrant to a safe third country should be treated as a penalty within the meaning of article 31 (1), whatever the reasons for taking that course may be and however unwelcome it may be to the migrant in question. (para 329)

The implications of this reading are far-reaching. Indeed, it seems to confirm David Cantor’s fears which he wrote in the wake of the first draft of the Illegal Immigration Bill, 

if the treaty only prevented the removal of refugees to a country where they feared persecution (as this government claims), every country in the world could spend all its time trying to ship its refugees to other countries. This would make the Convention meaningless and leave refugees in indefinite stasis.

In finding the UK’s displacement of its obligations under the Refugee Convention lawful, the Court of Appeal has simultaneously found that Rwanda is, in its view, not capable of meeting its obligations under this same international Convention.  It is hard not to read the decision as a statement of “We don't think the obligations to offer refuge apply to us but they do apply to you and you aren’t quite meeting them, but of course, improvement is possible.”

To reach its overall decision, the judgment offers a detailed breakdown of the Court’s assessment of the credibility of Rwanda’s current ‘Refugee Status Determination’ (RSD) process. There are an estimated 130,000 refugees currently in Rwanda, a figure in line with the global picture in which 76% of the world’s refugees are hosted by lower- and middle-income countries. In Rwanda, refugee status has generally been granted on a prima facie basis determined by the place people have come from, most commonly neighbouring Burundi and the Democratic Republic of Congo. An individualised RSD process has only been used in Rwanda to determine claims made by people coming from outside the region. This is a small number of cases – by the UNHCR’s estimate only 300 in the last three years (para 40). Based on the UNHCR’s assessment of this RSD practice, the Court of Appeal has found substantial grounds for thinking that there were, at the time of the APA, real risks that the asylum seekers would be refouled or subject to breaches of article 3 through refoulement.

In reaching this conclusion, one criticism raised of the Rwandan RSD system concerned whether Rwandan officials have sufficient knowledge of the Refugee Convention. This criticism turns on a claim by the UNHCR that in the recent past, there have been several instances where individuals seeking asylum in Rwanda have been denied the opportunity to make asylum claims at all (para 145). The Rwandan Government’s response to this claim was, on the first occasion, that these asylum seekers’ applications were not assessed because the Government requires an asylum seeker to present his/her need for protection immediately upon arrival. On the second occasion, the Rwandan Government countered that this failure to assess an asylum application arose because the applicant was using forged documents and thus did not meet immigration entry requirements.  In discussing this, the Court correctly and favourably states the Claimants’ position that there is no obligation for immediate disclosure of an asylum application and the use of forged documents is not a reason for refusing to consider an asylum claim (para 147).

My concern here is that the UK itself, in the same judgment, refuses to consider any asylum claims from people who come to the UK by a dangerous journey. Lord Justice Underhill goes on to say:

In my view the [Government of Rwanda’s] responses do not satisfactorily answer the allegation that it acted in breach of the Refugee Convention by declining to consider the asylum claims made and that its expulsion of the claimants led to their (indirect) refoulement in some of the cases and risked it in the others; and they do indeed show an imperfect understanding of the requirements of the Convention. (para 149)

Yet on the Court of Appeal’s own reading of the Refugee Convention the flaw here is only with the risk of refoulement and that alone. On its own reading of the Refugee Convention, all obligations to assess asylum applications fall away entirely if the individual concerned is sent to ‘another safe third country’. If deemed safe, at some other point in time, Rwanda could, technically, process the claims of people sent from the UK by sending these same people to another country that it determines to be ‘safe’. In doing this, it would be acting lawfully within this judgment’s interpretation of the Refugee Convention.

The Court of Appeal’s interpretation of the Refugee Convention uses international law at its civilising best. As Antony Anghie’s work so usefully illuminates, in this decision, Rwanda is set up as the aberrant society. The binary logic of this racialised ‘othering’ creates a bright legal line between the ‘civilised’ and the ‘uncivilised’. International law then provides the civilising bridge, offering the legal ‘techniques to normalize the aberrant society’. The UK courts have unanimously stated that the UK government has no obligation to assess an asylum claim. Instead, they place that legal obligation onto Rwanda. The Court then determines that Rwanda has failed to meet that standard and that its government officials require more legal training and capacity building.

With the Court’s attention only on the safety of Rwanda’s asylum procedures, it has lost a crucial opportunity to offer a purposive reading of the Refugee Convention that upholds the UK’s own obligations to offer refuge to people fleeing war and persecution. The hope would be to see the claimants’ proposed appeal to the Supreme Court directly address this issue.

I would like to thank Prof Mary Bosworth, Prof Phil Clark and Prof Miles Jackson for their comments on this piece and the Leverhulme Trust’s grant RPG-2022-265 - ‘Renewing Positive Complementarity: Reframing Justice Relations after Atrocity’ which is supporting my time on this research.

Any comments about this post? Get in touch with us! Send us an email, or post a comment here or on Facebook. You can also tweet us.