Notice & Comment

The 1967 Refugee Protocol and the Challenges of a Regional Agreement in Latin America, by Robert F. Barsky

In a series of blog posts, I have suggested that envisioning the 1967 Refugee Protocol as a simple amendment to the 1951 Convention, and fixating solely upon it eliminating the words ‘as a result of events occurring before 1 January 1951’, doesn’t capture the complexity of the Protocol’s creation, the context within which it was negotiated, its importance for current debates, or the challenge of negotiating an international treaty when many state officials were leaning towards regional agreements as a means of addressing contemporary refugee issues. In this blog, I will set forth the arguments employed by Latin American officials who advocated on the one hand for a universalization of the 1951 refugee Convention, and on the other for a regional agreement that would address the issues they considered most urgent for their region.

The UNHCR’s core concern by the mid-1960s was to eliminate the dateline (pre-1951) to ensure the continued relevance of that agency, and, moreover, to build an international refugee law regime that could address future refugee crises. At the same time, there was also widespread dissatisfaction amongst many UN member states that the Convention’s refugee definition wasn’t well-suited to the specific circumstances of specific states or regions, especially Africa, Asia, and Latin America. UNHCR officials had worked to adapt the Convention to new refugee flows, notably the 200,000 Hungarians who fled the Soviet invasion in 1956. This required an approach that backdated the causes of this crisis so that they would conform to the requirement that they be related to events preceding 1951. This rather creative approach wasn’t going to keep working, particularly as the UNHCR faced new refugee flows from Africa, in particular, which didn’t fit at all within the prescribed limits of the Convention.

Furthermore, and consistent with their mandate, UNHCR officials were anxious to negotiate a new, more liberal international treaty that would allow them to fulfil their obligations to refugees under the UNHCR Statute without having to continue to defer to the ‘good offices’ mechanism[1]. One of the crucial challenges that negotiators faced was to find a way to negotiate an international treaty a time when entire regions of the world were skeptical of what they considered to be the eurocentrism of international refugee law.[2] There was also a Cold War dimension to these negotiations, as the Soviet bloc officials and those in the West fought a war of words, with each side claiming the other as a source of refugee flows, and as guilty of disregarding human rights law. For these and other reasons, it is my contention that without looking back at the history of the Protocol’s drafting, we cannot adequately account for the historical backdrop against which refugee protection was being discussed.

In this blog, I will focus upon reservations that were being expressed by reluctant state officials from Latin America in regards to the new Protocol. Most state representatives from that region considered that the Convention was a European treaty that could (or should) be superseded by regional agreements better suited to specific refugee circumstances. This view fed the fears, expressed by many UNHCR officials at the time, that negotiations towards the new treaty needed to move forward through low-key direct negotiations with interested parties, so as to generate a minimum of fanfare and dissension. There was a widespread fear amongst negotiators that opening-up a new refugee treaty for debate amongst states in the General Assembly, or through the kind of Conference of Plenipotentiaries that had produced the 1951 Convention, could be disastrous. It was obvious to them, and to most state officials, that there didn’t exist the kind of universal accord that had driven the formation of the United Nations, and the passage of the 1948 Universal Declaration of Rights, in the immediate postwar period. On the other hand, there were some encouraging signs at the time, most notably that the United States, which had been unsuccessful in gaining passage of the 1951 Convention through the Senate, was anxious to sign onto a new international refugee treaty in 1968, the year that the General Assembly had named the International Year for Human Rights.

Despite the many advances made in the realm of civil rights legislation and human rights law in the 1960s, particularly in the United States, Cantor and Barichello (2013)[3] have noted a general skepticism about refugee law amongst Latin American state representatives. This tendency is confirmed in the memos, letters, minutes and communiqués related to the Protocol’s negotiation[4]. The Draft Protocol was completed by a group of 19 refugee law experts during their sojourn at the Villa Serbelloni in April of 1965, and it was submitted for consideration to the High Commissioner, with recommendations on how to move forward. The High Commissioner spent considerable time, beginning in 1966, soliciting input from various governments, to determine the pathway towards effecting the recommendations of the Colloquium. Paul Weis noted in a memo to Francisco Urrutia, who had become the UNHCR Representative for Latin America on February 17th, 1966, that the High Commissioner was awaiting replies from Argentina, Brazil, Colombia, Ecuador, Peru and Venezuela. Weis herein asked Urrutia if he has ‘heard of any reactions by these countries to the High Commissioner’s letter’, and requested that Urrutia ‘take such action as [he] may consider appropriate vis-à-vis the authorities of these countries to secure positive replies’. Weis also warned that ‘a discussion at this stage might also involve the danger of the Secretariat reaching a too hardened standpoint in favor of a Conference before the views of Governments have been ascertained’. At that point, only 13 of the 58 State representatives had replied.

A few months later, a memo was produced documenting the ‘Visit of Dr. Urrutia to Legal Division on 13 April 1966’. Therein, we find a note that  ‘Dr. Urrutia thought it was unlikely there would be many new accessions by Latin American countries, most notably of Venezuela which had never signed the 1951 Convention (and still hasn’t). Chile, he thought, ‘would probably do so in due course’ and so might Uruguay, ‘although parliamentary procedure in the latter country was excessively slow’. Most importantly, officials from Latin American states felt that ‘the question of asylum is regulated by their inter-state agreements. They are interested in travel documents to enable refugees to travel outside Latin America’ because, as he noted, within Latin America, travel was possible for both refugees and aliens on the basis of residence permits. Urrutia also thought that Latin American countries were generally ‘favorable to an extension of the dateline’, but them, the only relevant provision in the Convention related to travel documents. From his reading, ‘the Bellagio Colloquium recommendations were… well-received’, but ‘the Latin American States feel that the question of asylum is regulated by their inter-state agreements’.

In the end, every Latin American country except Cuba and Guyana came to accede to the 1951 Refugee Convention and/or the 1967 Protocol. The refugee definition was expanded by the non-binding 1984 Cartagena Declaration, which builds upon the Organization of African Unity (OAU) agreement,  adding to it the threat of generalized violence, internal aggression, and massive violation of human rights. Unlike the case of the OAU agreement, a refugee according to the Cartagena Declaration must show a link between herself or himself and the real risk of harm and demonstrate that ‘their lives, safety or freedom have been threatened’. From the perspective of the documents relating to the negotiation of the Protocol, this progression from the Convention to the Protocol to the regional agreement in Africa and the declaration in Latin America follow from the anticipated movement towards an increasingly liberal approach to refugee protection, although how they are being operationalized raise growing challenges in the current context.


Robert F. Barsky is a Professor of Law, French, English, and Jewish Studies at Vanderbilt University.

[1] James C. Hathaway, A Reconsideration of the Underlying Premise of Refugee Law, 31 Harv. Int’l. L. J. 129, 184 (1990), p. 158.

[2] The Handbook of Procedures and Criteria for Determining Refugee Status notes that: ‘In addition to the 1951 Convention and the 1967 Protocol, and the Statute of the Office of the United Nations High Commissioner for Refugees, there are a number of regional agreements, conventions and other instruments relating to refugees, particularly in Africa, the Americas and Europe. These regional instruments deal with such matters as the granting of asylum, travel documents and travel facilities, etc. Some also contain a definition of the term ‘refugee’, or of persons entitled to asylum.’ UNHCR, Handbook and Guidelines on Procedures and Criteria for Determining Refugee Status under the 1951 Convention and the 1967 Protocol Relating to the Status of Refugees, UN doc HCR/1P/4/ENG/REV.3 (1979, reissued 2011).

[3] David James Cantor & Stefania Eugenia Barichello (2013) ‘The Inter-American human rights system: a new model for integrating refugee and complementary protection?’ The International Journal of Human Rights, 17:5-6, 689-706,

[4] Unless otherwise noted, all of the original documents to which I refer can be found in the Paul Weis repository at Oxford, assembled in a large file labeled PW/PR/HCN/KSN/1-58 (133).