Beyond Containment – Confronting Racial Hierarchies in International Law: The Past and Present of the Refugee Regime – Portending a Problematic Future?

[Jinan Bastaki is an Associate Professor of Legal Studies at New York University, Abu Dhabi]

During the UN General Assembly high-level week (September 2025), the US Deputy Secretary of State, Christopher Landau, delivered a presentation entitled “The Global Refugee and Asylum System: What Went Wrong and How to Fix It.” He made five points: a) every nation has the right to control its own borders; b) refugees are not entitled to choose which country they receive asylum in; c) refugee status is temporary; d) sovereign states and not transnational bodies determine whether the conditions in a country of origin permit return; and e) that every country must accept the return of its nationals. This is with a context of severe budget cuts by the US to refugee supporting agencies, and a slash in refugee quotas.

In fact, the international refugee regime already enshrines many (if not all) these principles. It is predicated upon the delicate balance between state sovereignty and humanitarian concerns, with the former usually dominating. While the refugee regime has worked well for some categories of refugees, and refugee law has evolved in ways to include persecuted categories not originally envisioned by the drafters of the Convention (women for example, a category that still comes with its exclusions), my chapter in Emancipating International Law: Confronting the Violence of Racialised Boundaries builds upon works by Tendayi Achiume, James Hathaway, Antony Anghie, Lucy Mayblin and others, to analyse today’s refugee regime through the lens of its origins. 

The chapter demonstrates that the 1951 Convention was adopted in a colonial context and informed by questionable conceptions of race. Legal doctrines that we may consider neutral, such as state sovereignty, were imbricated with these prejudices, as only certain states were actually sovereign and independent. Only one type of refugee was envisaged and, while refugees make up a category of non-nationals that states in principle have agreed to protect when they reach their borders, it is still states who decide who falls under the definition. Indeed, only about 1% of refugees worldwide are resettled to wealthy third-states, while low- and middle-income countries host around 70% of refugees, and 66% of those in need of international protection reside in countries neighbouring their home states.

Sovereignty and Borders

Any discussion on refugees begins with the operation of borders and sovereignty in international law. Since the 1648 Treaty of Westphalia, state sovereignty developed into a fundamental right of states that centres freedom from interference. Yet, imperial states justified their colonisation of other countries on civilisational grounds, which were informed by socially constructed notions of race. Within this world, colonisers enjoyed freedom of movement, while the colonised were restricted (unless they were enslaved persons being forcibly transferred). Borders functioned as gateways of inclusion and exclusion, and their entrenchment policed the movement of people through a racialised spectrum.

To illustrate, the International Refugee Organization (IRO) was created in 1946, but it only tended to refugees created by World War II. This was not limited to refugees in Europe, but was intended for European refugees wherever they were. For example, the IRO operated in India to care for 4000 WWII Polish refugees, as well as Polish refugees who were being held in British administered camps in Southern Rhodesia. Within emergent international refugee law, the international was confined to Europeans. It was against this backdrop the 1951 Convention on the Status of Refugees was formulated.

States convened under the auspices of the UN to discuss the creation of an international treaty for the protection of refugees. In the context of the end of WWII, many regions had to contend with hundreds of thousands to tens of millions of refugees, including India-Pakistan (more than 14 million displaced persons due to Partition), the Middle (Near) East (almost one million displaced by the creation of the state of Israel), and Europe (around 65 million, a mix of refugees and internally displaced persons). With so many refugees spread throughout the world, it was an opportunity to create a truly global and universal instrument. India and Pakistan were part of the discussions, as well as the Arab states that were independent by then. Yet, it quickly became evident that the more dominant voices wanted to maintain focus on European refugees. India and Pakistan eventually withdrew from the discussions when they saw that their refugees would be excluded from the ambit of the Convention (neither state is party to the Convention today).

The 1951 Convention, with more than 140 signatory states, provided the first individualised definition of a refugee, distinct from the group approach that previously dominated. Consistent with the IRO, the Convention imagined one type of refugee: the individual fleeing persecution in Europe. Article 1A(2) stated that a person would be recognised as refugee if:

As a result of events occurring before 1 January 1951 and owing to well-founded fear of being persecuted for reasons of race, religion, nationality, membership of a particular social group or political opinion, is outside the country of his nationality and is unable or, owing to such fear, is unwilling to avail himself of the protection of that country.

Universality appeared to be on the table. Article 1B clarified that states could choose whether “events occurring before 1 January 1951” would mean in a) Europe or b) Europe and elsewhere (Europe was still at the centre). However, even those states favouring the more expansive definition did so because they felt that non-European refugees would not easily reach them and their colonies were explicitly excluded. The UK, for example, extended the Convention only to the Channel Islands and the Isle of Man, while the Netherlands maintained that the Convention did not apply to its Indonesian colonies. The international remained European.

The Excluded

While India and Pakistan withdrew from the discussions, the Arab states persisted to advocate on behalf of Palestinian refugees, who at that time numbered 750,000 outside of what became the expanded borders of Israel (there were many internally displaced as well). Due to the Partition Resolution — UNGA resolution 181 —  Arab states insisted that the UN retained responsibility for the plight it orchestrated. Two agencies were thus created to assist and protect Palestinian refugees: UNRWA, tasked with carrying out direct relief and works programmes, and the UN Conciliation Commission for Palestine, tasked with finding a durable solution conforming to UNGA Resolution 194. 

Here the objectives of the Arab and European states converged: both wanted Palestinians to be excluded from the 1951 Convention, but for different reasons. Since the Palestinians’ main aim was repatriation, the Arab states did not want them to be subsumed within a general category of refugees. European states, for their part, regarded Palestinian refugees as “different”. What was different about them was not made explicit; they were, after all, persons who were displaced from their homes due to war, and were denied return due to their identity, similar to many refugees in Europe. What was different about them was implied: they were not European. 

A compromise was proposed by the Egyptian representative to square the circle: according to Article 1D, if the specialised UN agencies “ceased for any reason,” those refugees would “ipso facto be entitled to the benefits of this Convention.”

Why did states agree to this clause, which seemed to offer extra protection to Palestinians only? The Iraqi representative at the Conference of Plenipotentiaries reassured European states that, regardless, it would be too costly for Palestinians to seek refuge in Western states. The UK delegate agreed and emphasised that even if they did, each state would be able to deal with it individually – a nod to their sovereign power to exclude. Instead of heightened protection, the specialised arrangement resulted in Palestinian confinement in camps in the countries where they first fled. Courts initially insisted that Article 1D was solely an exclusion clause unless UNRWA was fully terminated (the UNCCP had been defunct since the 1950s). Denmark was the only state to depart from this conventional interpretation, asserting in the mid-1980s that Palestinians automatically came within the protections of the Convention, including those who voluntarily left UNRWA’s areas of operation, though this was short-lived.

While the Convention enables sovereign states to decide how to interpret its provisions, due to the tireless work of advocates and lawyers, more recent jurisprudence in the EU treats the cessation of protection and assistance as an inability of UNRWA to provide protection or assistance, allowing Palestinians to apply for refugee status. They do not have to demonstrate persecution, but must show that UNRWA’s assistance has ceased for them. This has enabled some Palestinians to obtain refugee status, but it also remains that during the 2023-2025 genocide, some states closed their doors to Palestinains fleeing Gaza.

The 1967 Protocol and the rise of non-entrée regimes

While the 1951 Convention was initially meant for Europeans, the 1967 Protocol removed the temporal (pre-1951) and geographical limitations. This once restrictive definition would now be open to anyone, regardless of national origins. Or would it?

On the one hand, the universalisation of the definition made anyone eligible for refugee status in principle, as long as they fulfilled the conditions. On the other hand, since it would be more difficult for states to be selective, the Protocol also precipitated a marked increase in ‘remote controls’. While these remote controls are diverse, my chapter focuses on deportation of asylum seekers to so-called safe third-countries or for offshore processing (e.g. the UK’s attempts at spiriting asylum seekers to Rwanda). The use of these remote controls intensified after the Cold War and the rise in Third World refugees. Since the 1951 Convention does not require States to process asylum cases on their territory and does not comment on the legality of relying on ‘protection’ elsewhere, states have exploited these silences. 

One such country is Australia, although it is by no means the only one. For example, while Australia’s “external territories” are considered Australian territory in the same way as the Australian mainland, Christmas Island — which is closer to Indonesia than Australia — is classified as ‘excised territory’. Australian law prohibits asylum seekers from applying for protection on the island. For those who do reach the mainland, they were at one point confronted by Australia’s ‘Malaysia solution’: the transfer of asylum seekers to Malaysia to have their applications assessed by the UNHCR. A legal challenge led an Australian court to rule that even though Malaysia did not satisfy the requirements of an offshore processing country, Article 33.1 of the Refugee Convention “nevertheless permits removal of a refugee to a ‘safe’ third country.” Thus, while challenges to certain policies persist, with some notable victories, the legality of offshore processing remains. This enables the state to be selective, warding off most refugees while welcoming those it deems desirable or deserving.

Conclusion

Despite its origins, the 1951 Convention has evolved into the main instrument for protecting refugees. Landau’s comments regarding how to fix the refugee regime only identify features that already exist which enable the exclusion of most applicants. Non-refoulement, the jus cogens norm that prevents states from returning a refugee to where they may be persecuted, does not seem to prevent states from deporting them to third states. As Hathaway states:

While laudable efforts are made by a variety of agencies to attend to the immediate needs of refugees often unmet by the narrow scope of legal protection, both emergency assistance and durable solutions are beyond their exclusive control and can be frustrated by the exercise of negative state discretion.

Ultimately, the Refugee Convention reproduces racially inequitable outcomes not as a by-product, but as a feature of a system committed to state sovereignty and a global racial hierarchy. The result is that many refugees continue to receive subpar protection, while states exploit the silences of the Convention to deport undesirable refugees to ‘safe’ third countries and intensify their use of non-entrée regimes. This is most evident when it comes to Palestinians, that despite the dire situation in Gaza, few states created special arrangements in the same way that they did for Ukrainian refugees