From the 1951 Convention to the 1967 Protocol: how the refugee regime was globalized


 

How should we understand the globalization of the international refugee regime? A conventional understanding is that the 1951 Refugee Convention, although it put in place a universal definition of ‘refugee’ for the first time, remained limited to European refugees. The text refers to refugees displaced as a result of ‘events occurring in Europe before 1 January 1951’, because it was intended to deal with Europeans still displaced after the second world war. This was understood to be a finite task, and the mandate of UNHCR, the agency set up to assist states in implementing the convention, was due to last only eight years. Two new refugee crises in the 1950s changed this. In Europe, Austria called on UNHCR’s assistance in managing tens of thousands of Hungarian refugees fleeing Soviet repression after the failed 1956 revolution. And in north Africa, Tunisia and Morocco called on its assistance in managing tens of thousands of Algerian Muslim refugees fleeing French repression of the independence movement. These events placed more strain on the convention’s limited definition of ‘refugee’ than it could bear, and set in train the removal of both the geographic and temporal limitations through the 1967 Protocol. The refugee regime was then able to respond to the ‘new refugees’ of the 1960s during decolonization in the Global South.

This account is misleading, and overdue for revision. The 1951 text was not as limited as people think. Most of the initial signatories intended the convention to be applied to people displaced from anywhere, not just from within Europe. And the decisive momentum for globalization was created by African states newly independent from France.

The 1951 text, in article 1.B(1), offered contracting states the choice between two definitions of ‘refugee’. Alternative ‘a’ was limited to people displaced by ‘events occurring in Europe’ (already a formulation designed to include wiggle room); alternative ‘b’ was not. Most scholars, if they register this fact, seem to assume that the predominantly European states that negotiated and then signed the convention text must have plumped for the limited definition. But they didn’t. Of the twenty initial signatories—that is, the states whose signatures made the convention an active piece of international law that other states could sign up to—no fewer than twelve adopted the unlimited definition from the start. The Conference of Plenipotentiaries that agreed the text had included representatives of states from the Americas, the Middle East, and Australasia as well as Europe. The actual signatories were more clustered in Europe, but most of these, from Austria and Belgium to the UK and Yugoslavia, adopted the unlimited definition. The smaller number of signatories that adopted the more restrictive definition included non-European states—Brazil, Colombia, and Turkey—as well as European ones. So as well as the text including the possibility of a global definition of ‘refugee’, most of the initial signatories announced that this was how they would apply it. Two more of them, Colombia and the Holy See, shifted from ‘a’ to ‘b’ in 1961.

But it was the arrival of new contracting states after the convention entered into force in 1954 that made the momentum for globalization unstoppable. The decisive push came from a group of states that didn’t yet exist when the convention was drafted in 1948–51: former colonies of France that became independent in the 1950s and 60s. Several of the European states that signed the convention in 1951 or acceded to it later in the 1950s and 60s had colonial empires when they did so: Belgium, the Netherlands, the UK (all initial signatories on 28 July 1951); France, which signed a little later; Portugal, which acceded in 1960. Most colonial powers either didn’t apply the convention to their colonies at all (Belgium, Portugal) or did so selectively at a later date (the UK, the Netherlands). France was unusual because it informed the High Commissioner that it would apply the convention to ‘all territories for the foreign relations of which France is responsible’—that is, to all of its colonial possessions.

Why was this important? Because as France’s colonies became independent later in the 1950s and 60s, they did not need to undergo any lengthy accession process to join the international refugee regime: as successor states to the French empire, they could simply succeed to a convention that already applied on their territories. France signed the convention in September 1952 and ratified it in June 1954. For France and its colonies, it therefore came into force in September 1954. As colonies became independent states in north and then west Africa, they succeeded to the convention, automatically joining the refugee regime—and then they reshaped it, in ways that neither France nor the UN institutions could have predicted, let alone controlled. When France signed the convention in 1952 it did not expect to leave Indochina within two years, withdraw from Tunisia and Morocco in under five, cede a large part of its legally incorporated national territory (Algeria) within ten, or see west Africa reach independence in fifteen.

Former colonies of France that succeeded to the convention remade the refugee regime first by extending UNHCR’s activities ‘beyond Europe’, then by creating momentum for the globalization of the text of the convention. The agency’s first major venture beyond Europe was triggered by a request for assistance from  Tunisian prime minister Habib Bourguiba: the vast ‘joint relief operation’ that it coordinated in 1959–62 with the League of Red Cross Societies to assist as many as 300,000 refugees who had fled to Tunisia and Morocco during the Algerian war of independence. The Algerian National Liberation Front cooperated with this operation, and after Algeria achieved independence in 1962 it succeeded to the convention within a year. France had adopted the limited definition of refugee in the convention text, and would stick to it until 1971. But Morocco and Tunisia had adopted the unlimited definition, so that they (and UNHCR) could treat Algerian Muslims as refugees, and Algeria did too. By the early 1960s, ‘decolonization’ had also come to French west Africa. From 1960 to 1965, a group of newly independent west African states succeeded to the convention as former colonies of France—and they too immediately informed the High Commissioner that unlike the former colonial power they would adopt the expansive definition (‘b’). This meant that several years before the 1967 Protocol, a large majority of contracting states in both the former colonial north and the decolonizing Global South were already working with a global, not European, definition of ‘refugee’. 

The way to the 1967 Protocol was already open in the text of the 1951 Convention, but the decisive shift towards the globalization of the refugee regime came from the Global South. Just as the arrival of India and Pakistan as member states of the United Nations in 1948 decentred the new UN institutions and ensured that they would not act as the club of imperial powers that they had been intended to be, so too did the arrival into the refugee regime of a group of north and west African contracting states change it in ways that the initial signatories had neither expected nor intended.

This blog post was first published on refugeehistory.org.


Previous
Previous

Refugees and racial hierarchies in Lebanon: How refugee protection policies in the country ignore the prevalent problem of racism

Next
Next

State succession to the 1951 Refugee Convention: the curious case of Mauritius