Article body

Introduction

As a refugee protection crisis engulfs Africa,[1] the Organization of African Unity’s 1969 OAU Convention Governing the Specific Aspects of Refugee Problems in Africa (1969 Convention)[2] has remained largely beyond serious scrutiny. Understandably, in the wake of the convention’s adoption, attention focused on its remarkable legal innovations.[3] However, almost half a century later and amid declining standards of refugee protection in Africa, the discussion has scarcely moved on. When it receives any attention at all—which usually occurs around significant anniversaries[4]—the 1969 Convention is either uncritically praised[5] or else analysis remains focused on its novelties,[6] in particular the unique refugee definition, at the expense of scrutiny of that definition or of the broader protection framework that the convention establishes.[7] Africa is the only region of the developing world to have adopted a binding, regional refugee legal instrument, and it hosts almost a quarter of the world’s refugees,[8] yet the 1969 Convention remains poorly understood or misunderstood.[9]

This article on the 1969 Convention seeks to contribute to remedying this problem. It is characterized by its alternative approach. While recognizing the 1969 Convention’s significant contributions to refugee protection in Africa, this paper focuses equally on what the 1969 Convention is not—in terms of both widely held misconceptions about it and omissions from it—as opposed to the usual approach that focuses almost exclusively on what the 1969 Convention is. The paper begins by surveying the elements of the 1969 Convention that are commonly hailed as its major legal innovations: its unique refugee definition, its progressive development of an individual right to asylum, the broadened nature of non-refoulement under the convention, and its formalization of responsibility sharing, a type of temporary protection, and voluntary repatriation. This part provides an overview of the convention while at the same time surveying much of the literature on the 1969 Convention. The paper then goes on to address the most misunderstood of these innovations—the unique refugee definition—in greater depth, with an emphasis on dispelling the most common misconception surrounding the definition: that it is much broader than the refugee definition found in the 1951 Convention Relating to the Status of Refugees[10] (1951 Convention). Finally, the paper investigates a glaring yet often overlooked omission: the 1969 Convention’s silence regarding refugees’ civil and political, as well as socio-economic rights, and how the African convention works as the “regional complement”[11] to the universal refugee instrument in that regard. In so doing, this paper argues that refugees recognized only under the 1969 Convention (whether for practical reasons or because they do not meet the 1951 Convention refugee definition) are nevertheless entitled to the same standards of treatment as refugees recognized under the 1951 Convention. Taken together, this paper’s survey of the innovations in, misconceptions about, and omissions from the 1969 Convention provides a unique critical overview of the African refugee protection regime.

The critical approach adopted here should not be taken to suggest that the 1969 Convention should be interpreted in any way other than in good faith and in line with its object and purpose,[12] the humanitarian nature of which is made explicit in the convention’s preamble.[13] Rather, the overarching purpose of this paper is to form part of a movement toward more serious, critical legal engagement with the 1969 Convention and with refugee protection in Africa more generally. There is a remarkable dearth of critical legal analysis of the 1969 Convention,[14] which is all the more stark in relation to the sheer volume of analysis to which the 1951 Convention has been subject.[15] Serious academic analysis of the 1969 Convention is a critical component of full engagement with it as a tool of refugee protection. Indeed, Rankin maintains that the failure to provide an interpretive framework for the 1969 Convention “may ultimately undermine the flexibility of the [refugee] definition by limiting the situations in which it can be applied.”[16] If the 1969 Convention begins to receive even a fraction of the critical attention that has been devoted to its universal counterpart, it will represent an important contribution to the legal protection of refugees in Africa at a time when such a contribution is sorely needed.

I. Innovations

The 1969 Convention is the regional legal instrument governing refugee protection in Africa. It was adopted on 10 September 1969 at the sixth ordinary session of the OAU’s Assembly of Heads of State and Government, when it was signed by forty-one heads of state or government. It entered into force on 20 June 1974[17] after ratification by one-third of OAU member states.[18] It has since been ratified by forty-five of the fifty-four member states of the African Union (AU),[19] the successor organization to the OAU.

A. Overview of the 1969 Convention

The 1969 Convention is a relatively short instrument, containing a preamble and fifteen articles. The first article provides two refugee definitions, which are discussed in detail below,[20] and includes paragraphs on cessation[21] and exclusion.[22] These two paragraphs closely follow the 1951 Convention provisions, with three additions. Two additional cessation clauses provide that the 1969 Convention shall cease to apply to any refugee who has “committed a serious non-political crime outside his country of refuge after his admission to that country as a refugee”[23] or has “seriously infringed” the 1969 Convention’s purposes and objectives.[24] A further point of distinction is that the 1969 Convention does not include the clause present in the 1951 Convention preventing cessation in respect of a refugee who can “invoke compelling reasons arising out of previous persecution for refusing to avail himself of the protection of the country of nationality.”[25] Finally, an additional exclusion clause adds “acts contrary to the purposes and principles of the OAU as a further ground for exclusion.[26]

Article II of the 1969 Convention relates to asylum; each of its paragraphs is addressed in detail below.[27] The third article articulates refugees’ duty to respect the laws and regulations of the host state, echoing article 2 of the 1951 Convention, and prohibits them from engaging in subversive activities against any OAU member state. States party to the convention undertake to support this duty by prohibiting refugees “residing in their respective territories from attacking any State Member of the OAU, by any activity likely to cause tension between Member States.”[28] The prohibition on subversive activities is operationalized by the cessation clauses described above, which terminate the refugee status of an individual who commits a serious non-political crime after the acquisition of such status or who has seriously infringed the 1969 Convention’s purposes and objectives. Article IV, on non-discrimination in the application of the convention, follows article 3 of the 1951 Convention, however discrimination is prohibited on the additional grounds of nationality, membership of a particular social group, or political opinion.[29] The fifth article relates to voluntary repatriation, which is addressed in detail below.[30] Article VI, like article 28 of the 1951 Convention, mandates that contracting states provide refugees with travel documents. In view of article II(5), on temporary protection, which is discussed in detail below, article VI(2) provides, “Where an African country of second asylum accepts a refugee from a country of first asylum, the country of first asylum may be dispensed from issuing a document with a return clause.” Articles VII and VIII relate to state co-operation with the OAU and the office of the United Nations High Commissioner for Refugees (UNHCR), respectively. Article VIII(2) provides that the 1969 Convention “shall be the effective regional complement in Africa” of the 1951 Convention. This clause is analyzed in detail below.[31] The final seven articles are technical provisions.

It is apparent that many of the 1969 Convention’s eight substantive provisions represent significant departures from the 1951 Convention. This reflects the 1969 Convention’s objective, as announced by its title: to address aspects of the refugee problem singular to Africa. Indeed, Rwelamira explains that the final text of the 1969 Convention “settled for only the specific aspects of the African [refugee problem] which were not adequately catered for under the 1951 Convention.”[32] Accordingly, many of the 1969 Convention’s provisions are considered major innovations in the field of refugee law. Each innovation is surveyed below.

B. A Unique Refugee Definition

The 1951 Convention defines a refugee as someone with a well-founded fear of persecution on the basis of his or her race, religion, nationality, membership of a particular social group, or political opinion.[33] The 1969 Convention includes that same definition[34]—minus the 1 January 1951 date limit in the 1951 Convention that most states later agreed, by way of the 1967 Protocol,[35] not to apply—and provides at article I(2):

The term “Refugee” shall also apply to every person who, owing to external aggression, occupation, foreign domination or events seriously disturbing public order in either part [or] the whole of his country of origin or nationality, is compelled to leave his place of habitual residence in order to seek refuge in another place outside his country of origin or nationality.

Both definitions are employed by UNHCR in its operations in Africa,[36] though given the relative ease of applying the 1969 Convention’s article I(2) in the situations of mass influx that so often characterize refugee movements in Africa,[37] in practice, UNHCR and states often recognize refugees in Africa only under article I(2), even though the 1951 Convention’s refugee definition may equally apply.

The 1969 Convention’s additional refugee definition will be addressed at length below, as part of the analysis of common misconceptions surrounding it.[38] For the moment, it suffices to note that article I(2) of the 1969 Convention explicitly introduces objective criteria, based on the conditions prevailing in the country of origin, for determining refugee status,[39] and “requires neither the elements of deliberateness nor discrimination inherent in the 1951 Convention definition.”[40] Additionally, article I(2) was globally influential in that it contributed to the 1984 adoption of the Cartagena Declaration, which recommended that the traditional refugee definition be expanded in Latin America to include

persons who have fled their country because their lives, safety or freedom have been threatened by generalized violence, foreign aggression, internal conflicts, massive violation of human rights or other circumstances which have seriously disturbed public order.[41]

Furthermore, UNHCR’s Executive Committee of the High Commissioner’s Programme (ExCom) concluded that the 1951 Convention refugee definition should be broadened to take account of mass displacement, and in so doing used—without attribution—the exact wording of the 1969 Convention’s article I(2).[42]

C. Advancing an Individual Right to Asylum?

Asylum is “the first and most fundamental of the refugee’s needs and to grant him this constitutes the preliminary condition for him to have all the other rights.”[43] The 1951 Convention does not, however, establish any individual right to asylum. The Universal Declaration of Human Rights (UDHR), by contrast, enshrines the right of individuals to “seek and to enjoy” asylum,[44] but stops short of recognizing any individual right to asylum at international law. The UN Declaration on Territorial Asylum[45]—which is recalled at paragraph 7 of the preamble to the 1969 Convention—is similarly circumscribed. This resolution was followed, ten years after its adoption, by the UN Conference on Territorial Asylum, which failed to recognize or codify any individual right to asylum.[46] In 1981, however, the African Charter on Human and Peoples’ Rights (African Charter) recognized for the first time the right of persecuted individuals to “seek and obtain asylum.”[47] The ramifications of this provision have yet to be analyzed in depth, and it is beyond the scope of this paper to do so here. Suffice it to note that, despite the African Charter’s advance, the general consensus remains that the grant of asylum is within the exclusive discretion of states; as they have no obligation to grant it, individuals have no right to asylum corresponding to their UDHR right to “seek and to enjoy” it.[48]

While the 1969 Convention reflects this general consensus, it nevertheless significantly “strengthens the institution of asylum”[49] by providing:

Member States of the OAU shall use their best endeavours consistent with their respective legislations to receive refugees and to secure the settlement of those refugees who, for well-founded reasons, are unable or unwilling to return to their country of origin or nationality.[50]

This urging of states to grant asylum is “a further inroad into the traditional international law perspective which has tended to regard asylum as an exclusive right of the sovereign state,” but is “certainly not a right to be enforced by an individual against a state.”[51] The convention does not stop there; mirroring part of the preamble to the UN Declaration on Territorial Asylum, it characterizes the grant of asylum as a “peaceful and humanitarian act” that “shall not be regarded as an unfriendly act by any Member State.”[52] The language encouraging states to grant asylum is, however, only recommendatory. Thus the 1969 Convention incrementally advances, but does not enshrine, an individual right to asylum.[53]

D. Broadened Non-refoulement?

While the 1969 Convention’s contribution to the advancement of an individual right to asylum may be characterized as modest, its role regarding non-refoulement—a major aspect of the concept of asylum—is somewhat more significant. The general rule of non-refoulment provides that an individual should not be returned to a state where there is a real chance that he or she will face persecution, other ill-treatment, or torture. This principle is codified in, or has been judicially read into, a number of international refugee[54] and human rights instruments.[55] Most commentators even view the norm as having the status of customary international law.[56] Human rights-based non-refoulement is discussed below;[57] the focus here is on non-refoulement under refugee law. In that context, the norm as articulated at article 33(1) of the 1951 Convention, prohibits states from returning a refugee to territory where there is a risk that his or her life or freedom would be threatened on account of his or her race, religion, nationality, membership of a particular social group, or political opinion. The second paragraph of article 33, however, permits a national security exception.

The 1969 Convention’s non-refoulement provision closely follows article 3(1) of the UN Declaration on Territorial Asylum.[58] The 1969 Convention provides:

No person shall be subjected by a Member State to measures such as rejection at the frontier, return or expulsion, which would compel him to return to or remain in a territory where his life, physical integrity or liberty would be threatened for the reasons set out in Article I, paragraphs 1 and 2.[59]

This is broader than the 1951 Convention’s non-refoulement provision in two important respects; however, the 1969 Convention does not expand non-refoulement as greatly as is often suggested.

First, the 1969 Convention expands non-refoulement because it does not include a national security exception like the one found in its universal counterpart. However it does not render non-refoulement absolute, as many scholars have suggested.[60] Pursuant to articles I(4)(f) and (g), the application of the 1969 Convention, and hence protection from refoulement, ceases if the individual concerned commits a serious non-political crime outside the country of refuge after admission as a refugee or seriously infringes the convention’s purposes and objectives. This, according to D’Sa, implies that the 1969 Convention, like the 1951 Convention, allows expulsion in limited circumstances, “although the OAU appears to deal with the latter somewhat indirectly.”[61]

Second, the 1969 Convention’s non-refoulement provision applies at frontiers, while the 1951 Convention makes no such explicit provision. As a result, many commentators view non-refoulement under the 1969 Convention as broader than under the 1951 Convention.[62] State practice, however, has aligned the universal refugee regime with the standard of the 1969 Convention. According to Goodwin-Gill and McAdam, “By and large, States in their practice and in their recorded views, have recognized that non-refoulement applies to the moment at which asylum seekers present themselves for entry, either within a State or at its border.”[63] At present, therefore, the 1969 Convention’s conception of non-refoulement is no broader than that of the 1951 Convention as far as applicability at frontiers is concerned.

E. Formalization of Responsibility Sharing, Temporary Protection, and Voluntary Repatriation

The 1969 Convention formalized for the first time versions of three important refugee law concepts: responsibility sharing, temporary protection, and voluntary repatriation. Article II(4) articulates a very early notion of responsibility sharing, providing:

Where a Member State finds difficulty in continuing to grant asylum to refugees, such Member State may appeal directly to other Member States and through the OAU, and such other Member States shall in the spirit of African solidarity and international co-operation take appropriate measures to lighten the burden of the Member State granting asylum.

Such “appropriate measures” include regional resettlement, financial support, and political responsibility sharing. Each possible method of responsibility sharing has, however, been constrained in practice by the limited resources of African states.[64]

Temporary protection describes a variety of practices. Fitzpatrick describes it as “a magic gift, assuming the desired form of its enthusiasts’ policy objectives.”[65] The dual meaning attributed to the notion of temporary protection articulated in the 1969 Convention reflects Fitzpatrick’s description. The concept finds expression at article II(5), which provides, “Where a refugee has not received the right to reside in any country of asylum, he may be granted temporary residence in any country of asylum in which he first presented himself as a refugee pending arrangement for his re-settlement.” York University’s Centre for Refugee Studies interpreted article II(5) as implying that the nature of the protection granted under the 1969 Convention is of limited duration:

The debate about temporary versus permanent refugee protection has no real currency in the South, where protection has almost always been assumed to be temporary, even if it lasted for a long time. Protection has usually been provided by neighbouring countries with the clear understanding that the refugees would eventually return home. In fact, in Africa, temporary protection is not only common practice, it is given prominence in the Organi[z]ation of African Unity’s 1969 Convention Governing the Specific Aspects of Refugee Problems in Africa, Article 11(5) [sic].[66]

The version of temporary protection actually posited by the 1969 Convention does not, however, imply limited protection. Rutinwa explains that article II(5)

applies to persons who have been recognised as refugees but for one reason or another have not been granted the right of residence for any duration at all. It is not intended to determine the duration of residence for all refugees who have been recognised and granted asylum. ... [Furthermore,] where a person is resettled from one African country to another on account of the first country not being able to continue to provide him or her asylum, the function of resettlement in this case is not to terminate but to continue the refugee status of that person but in a different country.[67]

Put this way, it becomes clear that the 1969 Convention’s notion of temporary protection is more akin to responsibility sharing than it is to later versions of temporary protection designed to limit states’ obligations toward refugees. Under the 1969 Convention, it is the sojourn in the first country of asylum, not the protection, that is temporary.[68]

While the notion of temporary protection articulated by the 1969 Convention is a humanitarian one,[69] it seems premised on an idea that is fundamentally less so. Article II(5) exists to remedy a situation where a refugee has received asylum but no corresponding right of residence. That a refugee could be recognized as such but could also be lawfully deprived of a right of residence must be queried. A state’s realization of its obligations under the 1951 Convention—which applies co-extensively with the 1969 Convention[70]—to ensure refugees’ rights clearly depends on the refugees’ presence in the territory of the asylum state.[71] Indeed, article II(1), in urging states to grant asylum, conceives of such asylum in terms of reception and securing the “settlement” of refugees.

Article V of the 1969 Convention addresses voluntary repatriation. Its first paragraph articulates the core principle: “The essentially voluntary character of repatriation shall be respected in all cases and no refugee shall be repatriated against his will.” This principle is an important corollary of article II’s provisions on asylum, particularly article II(3) on non-refoulement. The clauses that follow the core principle are premised on the assumption that the conditions for safe return have been met[72] and detail the duties of countries of asylum and origin and refugee-assisting agencies. The sending state, in collaboration with the receiving state, must “make adequate arrangements for the safe return of refugees who request repatriation,”[73] while the country of origin must “facilitate their resettlement and grant them the full rights and privileges of nationals of the country, and subject them to the same obligations.”[74] The convention mandates countries of asylum, countries of origin, voluntary agencies, and international and intergovernmental organizations to assist refugees with the process of return.[75] It provides in particular that states of origin should use the news media and the OAU to invite refugees home, and should provide assurances regarding the circumstances prevailing there, and that host countries should ensure that such information is received.[76] Article V also provides that, upon return, refugees must not be penalized for having fled.[77]

The 1969 Convention is the first, and remains the only, international legal instrument to formally insist on the voluntariness of refugee repatriation;[78] however, previous articulations of the concept appear in the UNHCR statute,[79] an early UN General Assembly resolution,[80] and the constitution of the International Refugee Organization,[81] the predecessor institution to UNHCR. Furthermore, that repatriation should be voluntary is evidenced by state practice.[82] Its originality aside, article V(1) is a “powerful statement of principle,”[83] which despite isolated critiques,[84] is hailed as representing an early articulation of a principle that became a cornerstone of the international regime for refugee protection.[85] Unfortunately, it has been misinterpreted to suggest that repatriation is the primary solution for refugees on the continent.[86] Rutinwa explains that, in fact, article V is “much more about elaborating the principles and the modalities of effecting voluntary repatriation than a prescription of it as the only solution.”[87]

The 1969 Convention introduced several important legal innovations. It included a new refugee definition, advanced an individual right to asylum, broadened the norm of non-refoulement somewhat, and formalized the concepts of responsibility sharing, temporary protection, and voluntary repatriation. The preceding part detailed the latter three innovations and highlighted the ways in which some of them have been misunderstood. By far the most misunderstood legal innovation of the 1969 Convention, however, is its unique refugee definition, to which the following part is devoted.

II. Misconceptions

The article I(2) refugee definition is without a doubt the most celebrated feature of the 1969 Convention. Okoth-Obbo notes that the provision has “generated a reputation which boarders [sic] on the mythical.”[88] However, the provision’s mythical status is arguably the result of several misconceptions about it, which taken together, have led to a somewhat erroneous “interpretive consensus”[89] characterized by the almost universal propensity to view the article I(2) refugee definition as remarkably “expansive”,[90] “extensive”,[91] “wide”,[92] or “broad”,[93] especially in relation to the 1951 Convention refugee definition.[94] This interpretive consensus has precluded critical analysis, thereby perpetuating the misunderstanding. Indeed, according to Rankin, the focus in the literature on the definition’s broadness “tends to gloss over ... [its] vagueness and ambiguity.”[95]

While the article I(2) refugee definition has certainly extended international protection to individuals who would not otherwise qualify for refugee status,[96] it is not necessarily quite as inclusive or broad as most commentators suggest;[97] indeed, the analysis below suggests that the 1969 Convention’s unique refugee definition likely only extends refugee protection incrementally. The handful of scholars who have engaged in serious critical analysis of the 1969 Convention have identified three particularly common misconceptions about the article I(2) refugee definition, which contribute to the flawed interpretive consensus about the definition’s breadth: first, that all four events justifying flight under the 1969 Convention (1969 Events) remain equally relevant today;[98] second, that the article I(2) refugee definition is entirely objective;[99] and third, that it applies only to groups[100] or was drafted with a view to promoting the group determination of refugee status.[101] Each of these misconceptions is addressed in turn below.

A. Continued Relevance of All 1969 Events?

There is little consensus regarding the meaning of each of the 1969 Events—external aggression, occupation, foreign domination, and events seriously disturbing public order—which “lacked a firm definition under international law”[102] when the 1969 Convention was drafted and adopted. Since then, scholars have made excellent efforts at elaborating the terms.[103] Ultimately, however, any authoritative consensus around their meaning will depend on the weight of reported judicial interpretation, of which there is a paucity in Africa, and on an exhaustive survey of state practice that is beyond the scope of this paper. Accordingly, no attempt is made here to further articulate the terms’ significance. Suffice it to highlight the lack of an interpretive consensus and the fact that three of the four 1969 Events—external aggression, occupation, and foreign domination—largely ceased to be relevant with the end of colonialism and apartheid,[104] narrowing the scope of the article I(2) definition.

While Okoth-Obbo argues that external aggression, occupation, and foreign domination could be viewed as “vessels still possessed of the capacity for the legal transcription of Africa’s refugee realities of today”[105]—the war in the Democratic Republic of the Congo, in which Uganda was held to be occupying power,[106] comes immediately to mind—on the whole, the terms no longer carry the import that they once did. The article I(2) definition “was very appropriate ... [in the 1960s] in that it addressed the immediate concerns of people fleeing from the colonial territories ... and from the racist regimes in Southern Africa,”[107] but it is less relevant in the contemporary context. Indeed, Okoth-Obbo ultimately concludes that the definition “should be upgraded to more properly reflect the actual situations which today cause people to flee as refugees in Africa.”[108] With external aggression, occupation, and foreign domination being less relevant as causes of refugee flight today, “events seriously disturbing public order” assumes increased significance, and reaching an interpretive consensus about the term’s precise meaning becomes of even greater importance.[109]

B. Extent of the Article I(2) Refugee Definition’s Objectivity

In moving away from the 1951 Convention’s well-founded-fear standard in favour of a focus on the disruptive conditions in the country of origin or nationality, the article I(2) refugee definition certainly introduces an objective element. According to Hathaway, it “acknowledges the reality that fundamental forms of abuse may occur not only as a result of the calculated acts of the government ... but also as a result of that government’s loss of authority.”[110] This outward orientation has led to a consensus among most scholars of the 1969 Convention[111] that the article I(2) refugee definition is “based solely on objective criteria”[112] and therefore mandates a completely objective test of refugee status. However, this consensus is overstated for two reasons. First, the focus on the objectivity of the article I(2) refugee definition overestimates the subjectivity of the 1951 Convention definition[113] and underestimates the extent to which this universal definition can apply to victims of war and civil strife.[114] Second, views of the article I(2) refugee definition as entirely objective overlook elements of the definition that may mandate a subjective test of refugee status.[115] Each of these reasons is addressed in turn below.

1. Subjectivity and the 1951 Convention

The view that the article I(2) refugee definition is objective is largely a relative one, as the regional definition is almost always assessed in relation to its universal counterpart. The latter is usually viewed as including both objective (“well-founded”) and subjective (“fear”) elements.[116] Oloka-Onyango, for example, notes that

by moving away from the Geneva Convention’s ‘. . .well-founded fear of persecution . . .’ standard, the OAU Convention explicitly gave credence to the fact that a refugee exodus could be the result of factors of a more general nature, intrinsic to the particular country in question, rather than to the individual subjective status or fears of the refugee.[117]

Such comparisons do the 1951 Convention refugee definition a disservice because they overemphasize its subjectivity, which some maintain was never intended. Hathaway, for example, argues that “[t]he concept of well-founded fear is ... inherently objective.”[118] He maintains that the fear element was only intended to introduce a prospective risk assessment into the refugee definition.[119] Indeed, the view that the 1951 Convention includes a subjective element merely because it mandates an inquiry into an individual’s (objective) circumstances seems an exaggeration of the concept of subjectivity.[120] Nevertheless, leading jurisprudence[121] and UNHCR’s interpretive guidance[122] have affirmed the importance of the subjective aspect of the 1951 Convention refugee definition. A more salient point, therefore, is the extent to which focus on the objectivity of the article I(2) refugee definition in applying it to victims of war and civil strife—who often arrive in groups—underemphasizes the extent to which the 1951 Convention refugee definition is also applicable to such individuals. Indeed, “[t]he very existence of the OAU Convention has been used by some to justify a conservative reading of the 1951 Convention.”[123]

The misconception that the 1951 Convention does not apply in situations of mass influx has arisen from problematic interpretations of the 1951 Convention both in relation to the 1969 Convention and on its own. In the latter case, Durieux and McAdam have argued that “[t]o assert that the [1951] Convention does not apply in cases of mass influx is tantamount to saying that the individual does not exist in a group.”[124] Indeed, Kälin has argued convincingly that the 1951 Convention can provide refugee status to individuals fleeing civil war.[125] In relation to the 1969 Convention, Jackson maintains that while the wording of the article I(2) refugee definition is certainly broader than that of the 1951 Convention, “[t]here must ... necessarily be a considerable amount of overlapping and, as regards their practical application the difference between the two definitions is probably not as great as might at first sight appear.”[126] Okoth-Obbo endorses this view when he notes that as the article I(2) refugee definition was increasingly “pointed to as the unique example of positive law enabling the consideration of victims of war and civil strife as refugees ... the more it became possible to validate and reinforce the argument that the 1951 Convention did not apply to those categories.”[127] If the article I(2) refugee definition is continually assessed in relation to the 1951 Convention refugee definition, and if such comparisons misunderstand the subjectivity and applicability of the 1951 Convention refugee definition, the result is an exaggerated view of the novelty of the article I(2) refugee definition’s objectivity.

2. Subjectivity in the Test for Refugee Status Under the 1969 Convention

It is far from clear that refugee status under the 1969 Convention should be assessed on an entirely objective basis for two primary reasons. First, the article I(2) refugee definition is framed in terms of individual status. Edwards argues that “this necessitates inquiring into the individual or subjective reasons for flight of each applicant.”[128] The second, more persuasive, reason is textual, beginning with the use of the word “compelled.” The “compelled” aspect of the article I(2) refugee definition seems to have much in common with the subjective “fear” aspect of the 1951 Convention definition. Particular events that may compel one person to flee his or her place of habitual residence may not result in such compulsion in another individual whose appreciation of the risk of the events differs. This is the view of Edwards[129] and Okoth-Obbo. The latter notes:

[T]he OAU expanded definition is predicated mainly on the compulsion to leave the place of habitual residence in order to seek refuge. Ironically, in so doing, it reintroduces the problematic question of motive for flight which it is otherwise credited with having disabused from the refugee definition.[130]

Rankin approaches the nature of the term “compelled” with some ambivalence but concedes that it may indeed import subjectivity into the analysis:

Although compelled may be objective, no one has offered a satisfactory explanation for why it does not contain a subjective element. The word could relate to a subjective feeling or preference. ... Is the mere existence of an OAU event enough to demonstrate that someone has been compelled? Or is it necessary to show linkages between an asylum seeker and a particular event? Until these questions are answered compelled remains ambiguous.[131]

Edwards, Okoth-Obbo, and Rankin thus agree that analysis of the 1969 Convention, to date, has overlooked the importance of the element of compulsion in the article I(2) refugee definition, thereby underestimating the role of subjectivity in determining refugee status. But what would a subjective test for refugee status under the article I(2) refugee definition actually look like? The Rankin extract above highlights the connection between, on the one hand, an objective test and the mere existence of a 1969 Event, and on the other hand, a subjective test and linkages between the refugee and such event. Edwards is more explicit about what a subjective test would entail, providing an example:

[I]t would not be unreasonable in status procedures to require a claimant from Kinshasa, DRC, to justify why he or she departed the city on the basis of generalised violence occurring in North Kivu, an area thousands of miles from the capital. In all likelihood, it would be quite difficult for him or her to prove that he or she was “compelled” to leave as a result of those events if one only considered the objective facts. Importing a subjective analysis does not mean that an individual needs to prove that flight was the only alternative available, but it does introduce a causal connection or nexus between the flight and the event in question.[132]

In other words, the subjective test of refugee status under article I(2) of the 1969 Convention is an assessment of whether the 1969 Event caused the individual refugee’s flight—as opposed to an objective test that looks for the mere existence of a 1969 Event in the refugee’s country of origin or nationality. Subjectivity under the 1969 Convention is thus primarily about the nexus required between the 1969 Event and flight.

Commentary on the 1969 Convention has rarely addressed explicitly whether a nexus between the disruptive events and flight is required. Rather, there seems to be an implicit interpretive consensus presuming that an individual would not flee a 1969 Event without a nexus between the event and a risk of harm. According to Hathaway, “Because the African standard emphasizes assessment of the gravity of the disruption of public order rather than motives for flight, individuals are largely able to decide for themselves when harm is sufficiently proximate to warrant flight.”[133] Implicit in this view is the assumption that, in the determination of refugee status, flight itself is sufficient evidence of the proximity of harm. While in most cases this assumption will be borne out, it obfuscates the importance of, first, the words “owing to” and, second, the fact that the individual must be compelled to leave his or her “place of habitual residence”. These terms are additional textual elements of the article I(2) refugee definition that seem to mandate a subjective test of refugee status, because they suggest a requirement of linkages between the 1969 Event and the refugee’s flight. Indeed, the deliberate inclusion of “owing to” and “place of habitual residence” suggest that the nexus between the disruptive events and flight ought to be more than merely presumptive.

The ordinary meaning of “owing to” is analogous to “as a result of” or “due to”. Accordingly, under the article I(2) definition, a refugee is someone who, as a result of, or due to, a 1969 Event, is left with no choice but to flee his or her place of habitual residence. Put this way, it becomes clear that flight must be the direct consequence of a risk of harm to the individual stemming from the 1969 Event. Furthermore, the article I(2) refugee definition specifically provides that a refugee must have fled his or her place of habitual residence, as opposed to his or her country of origin or nationality. According to Rankin, this clause is used to focus “attention on those who face danger because of the state of their communities,” resulting in “an implied relationship or geographic nexus between an OAU event and a person’s place of habitual residence.”[134] That the article I(2) definition requires physical proximity between the putative refugee and the 1969 Event certainly suggests that the unique refugee definition demands an explicit nexus between the risk of harm and the refugee’s flight.

The requirement that flight be from the place of habitual residence also explains why the fact that the harm may be in “either part [or] the whole of”[135] the country of origin or nationality does not expand the refugee definition as much as might initially be expected. The specific mention that the harm may be in part or the whole of the country of origin or nationality makes it at least initially plausible that an individual may be recognized as a refugee if his or her flight is prompted by an event taking place anywhere in his or her country of origin. In context, however, it becomes clear that there is “a necessary link between the asylum seeker and the OAU event. ... [T]he nexus is created by the fact that an asylum seeker is compelled to leave his or her place of habitual residence.”[136] Moreover in article I(2), “either part [or] the whole of his country of origin or nationality” likely applies only to the final 1969 Event, “events seriously disturbing public order.” This is so for interpretive reasons—the lack of a comma in article I(2) between “events seriously disturbing public order” and “in either part [or] the whole of,” and the result of applying the doctrine of ejusdem generis[137]—and because the necessity of specifying “either part [or] the whole of” attaches only to “events seriously disturbing public order.” External aggression, occupation, and foreign domination, even if only prevailing in one part of a state (most likely the capital city), will almost by definition affect the country as a whole. Edwards puts it as follows: “[T]he international requirement associated with the ... three terms suggests that they are experienced throughout the whole of the territory de jure, even if the actions are limited to specific parts of the territory de facto.”[138] Thus there can be no suggestion that the inclusion of “either part [or] the whole of his country of origin or nationality” implies that a refugee may justifiably flee events that are not directly connected to him or her.

Additionally, a mere assumption that flight is necessarily the result of threatened harm does not reflect the targeted nature of the 1969 Convention. It is axiomatic that the 1969 Convention only extends refugee protection where such protection is necessary to safeguard a particular individual from a 1969 Event. That the convention does not extend international protection indiscriminately suggests that its refugee definition should not be applied in an indiscriminate manner. Relying solely on objective criteria to determine refugee status “could (and does) give rise to situations where asylum is sought by persons who flee for reasons unconnected to the event in question, but who can use that event to claim asylum.”[139]

Thus the terms “compelled” and “owing to”, and the precise requirement that an individual have fled his or her place of habitual residence, coupled with the limited applicability of “either part [or] the whole of his country of origin or nationality” and the 1969 Convention’s obvious purpose of offering protection only to those at risk of harm, strongly suggest that, in principle, refugee status should depend on an assessed, as opposed to presumed, nexus between the 1969 Event and the refugee’s flight. This calls into question the sufficiency of determining refugee status on a purely objective basis, which in any case would often be little more than an exercise of determining nationality.[140] It is more likely that refugee status under article I(2) of the 1969 Convention was intended to be determined both objectively—involving an assessment of whether a 1969 Event is prevailing in the country of origin or nationality—and subjectively—involving an assessment of the risk of harm that the 1969 Event actually posed to the individual concerned. Yet, it should be noted that, in specific situations where the 1969 Event is widespread and affects the whole of the country or territory from which the individual has fled, the existence of such a nexus may, purely as a matter of procedure, be presumed. Any other approach would belabour the obvious.

The view of the test for refugee status under the 1969 Convention’s unique refugee definition as both objective and subjective is bolstered by a close examination of the third common misconception about article I(2): that it applies only to groups or was drafted with a view to mandating the “group” determination of refugee status. The view that refugee status under article I(2) of the 1969 Convention was meant to be determined on a group basis is very compatible with a completely objective assessment of the compulsion to flee, based entirely on the conditions prevailing in the country of origin or nationality. However, upon close examination, it becomes clear that the group determination of status emerged solely as a result of practical considerations and was never specifically intended. This issue and others are explored below.

C. Group Determination of Refugee Status

There are two major misconceptions about the 1969 Convention and the group determination of refugee status. The first is highlighted by Durieux and McAdam, who assert that there is a popular perception that “the OAU Convention applies only to groups, not to individuals.”[141] This perception is clearly erroneous for two reasons. First, the 1969 Convention reproduces, at article I(1), the 1951 Convention refugee definition. It is beyond dispute that this definition applies to individuals.[142] Second, both the article I(1) and article I(2) refugee definitions contained in the 1969 Convention refer to “every person”, thereby mandating the individual determination of refugee status.[143]

The second, more common, misconception is that the 1969 Convention was purposely designed to address situations of mass influx[144] and therefore provides the legal foundation for the group determination of refugee status.[145] Nobel, for example, notes that the convention’s article I(2) refugee definition is “the legal basis for admitting refugee masses upon a group determination of their status.”[146] Similarly, the Lawyers Committee for Human Rights maintains that the 1969 Convention “introduced the notion of group determination of refugee status”[147] and suggests that “the preferable practice of group eligibility ... is provided for under the OAU Convention.”[148] Additionally, Milner explains that

by making refugee status contingent on generalized situations in the refugee’s country of origin, the 1969 OAU Convention allows states to recognize entire groups of individuals as refugees on the basis of shared characteristics and common cause of flight.[149]

Finally, d’Orsi notes that the 1969 Convention “is meant to promote the prima facie recognition of groups of refugees.”[150]

Such views are incorrect.[151] Rather than being based on an analysis of the 1969 Convention, they arise out of the common practice in Africa of determining refugee status on a prima facie basis in situations of mass influx.[152] While the article I(2) refugee definition does not preclude such a practice—especially because, as compared to the 1951 Convention definition, it includes elements that are arguably easier to apply in situations of large-scale influx[153]—a close reading the 1969 Convention reveals no specific intent to introduce or promote prima facie determination.[154] To the contrary, as noted above, the language of the convention is in the singular, providing that a refugee is “every person” who has a well-founded fear of persecution on account of a particular characteristic[155] or who was compelled to leave his or her country of origin or nationality as a result of a 1969 Event.[156] Furthermore, the 1969 Convention is not unique in its applicability in situations of mass influx. Durieux and McAdam maintain that the 1951 Convention “contains nothing to suggest its inapplicability in cases of mass influx.”[157] Nor is article I(2) of the 1969 Convention the only refugee definition that permits prima facie refugee status determination (RSD): if the objective circumstances that triggered the application of the presumption of eligibility for refugee status “are those under Article IA(2) of the 1951 Refugee Convention ... then the persons recognised to be prima facie refugees are refugees within the meaning of”[158] that article. Indeed, refugees fleeing Hungary for Austria and Yugoslavia following the 1956 Soviet invasion and occupation were recognized on a prima facie basis under the 1951 Convention.[159] The determination of refugee status on a prima facie basis in situations of mass influx is not a product of the 1969 Convention, nor is it inherently or exclusively linked to that instrument. Rather, it arose as a matter of practical necessity in situations of mass influx,[160] in which “the numbers of the asylum seekers involved and the urgency to provide assistance ... make it impracticable and forbiddingly costly to administer individual status determination.”[161]

The false attribution of the group determination of refugee status to the 1969 Convention has caused confusion between the latter concept and the prima facie recognition of refugee status in situations of mass influx.[162] Prima facie RSD is a process whereby individual refugee status is recognized on the basis of a presumption.[163] That such a presumption is often applied to groups in situations of mass influx is the source of the erroneous conflation of prima facie RSD with the group determination of refugee status. In fact, there is no such thing as group RSD; it is just an imprecise way of expressing the need to resort to prima facie status determination in situations of large-scale influx. Under a prima facie process of status determination, “it is not the refugee quality ... of the entire group that is determined, but that of each individual in the group. Groups do not accrue refugee status, be it prima facie or by other means. Only individuals do.”[164] Albert puts it similarly when he maintains that prima facie RSD is “better described as being an expedited form of individual RSD, not ‘group’ RSD.”[165] This view is confirmed by UNHCR’s handbook, which explains:

[S]ituations have ... arisen in which entire groups have been displaced under circumstances indicating that members of the group could be considered individually as refugees. ... Recourse has therefore been had to so-called “group determination” of refugee status, whereby each member of the group is regarded prima facie (i.e. in the absence of evidence to the contrary) as a refugee.[166]

Thus the ascription of group RSD to the 1969 Convention has not only perpetuated a misconception about the convention, it has also contributed to the propagation of a legal concept with no actual basis in law, or at the very least, to the proliferation of inaccurate legal terminology.

A related misconception has to do with the nature of the refugee status that recognition on a prima facie basis confers. There is a view that such recognition creates only a “presumption of refugeehood and therefore entails an incomplete (or secondary) refugee status,”[167] in terms of both its durability and the post-recognition rights that attach. Okoth-Obbo evidences this view when he notes, “The prima facie concept refers to the provisional consideration of a person or persons as refugees without the requirement to complete refugee status determination formalities to establish definitively the qualification or not of each individual.”[168] This implies that refugees recognized on a prima facie basis receive only temporary protection[169] and do not qualify for the full range of rights conferred by individual recognition. Indeed, according to the Lawyers Committee for Human Rights, “the notion of group eligibility under the OAU Convention was based on the premise of legally institutionalized temporary protection.”[170]

Durieux and Hurwitz,[171] Durieux and McAdam,[172] Jackson,[173] and Rutinwa[174] have each, however, rejected this view of prima facie refugee status as temporary and incomplete, on the basis of the conclusiveness of the prima facie presumption of refugeehood. While each author accepts that prima facie refugee status is indeed presumptive, the “operation of the presumption provides full and sufficient evidence”[175] of refugee status, unless “the State decides to subject it to scrutiny on an individual basis, and finds against the individual asylum seeker.”[176] That the presumption of refugee status within a process of prima facie RSD is conclusive suggests that refugees recognized pursuant to such a process are refugees like any other. Indeed, Rutinwa argues:

If persons recognised as refugees on a prima facie basis are presumed to be refugees within the definitions found under the relevant instruments, it logically follows that their treatment should be in accordance with the standards stipulated under those instruments.[177]

Once one understands what prima facie RSD is and how it works, it becomes clear that it is merely a procedural tool that can have no effect on the substantive rights conferred. Indeed, Durieux and McAdam maintain, “One must conclude that prima facie recognition entails full refugee status, and beneficiaries of it are entitled, in Contracting States, to the standards of treatment stipulated by the 1951 Convention.”[178] While this resolves the question of the nature of the status that attaches to refugees recognized on a prima facie basis, it raises the related, wider issue of the nature of refugee status that results from recognition—whether on a prima facie basis or individually—under article I(2) of the 1969 Convention. This critical issue is addressed in the part that follows.

III. Omissions

Refugees within the meaning of article I(1) of the 1969 Convention are clearly also refugees under article 1A(2) of the 1951 Convention—the two definitions are almost identical—and therefore benefit from the standards of treatment guaranteed among articles 3-34 of the latter instrument in states that have ratified it. The 1969 Convention, however, enumerates no such standards of treatment, nor does it explicitly incorporate the 1951 Convention’s rights framework. This fact is surprisingly overlooked by almost all of the limited number of critical analyses of the 1969 Convention.[179] The lack of comprehensive standards of treatment in the 1969 Convention raises the question of the specific refugee rights framework applicable—in addition to the one provided by human rights law more generally—to refugees who are recognized under article I(2) of the 1969 Convention, but who do not meet the 1969 Convention article I(1) or 1951 Convention article 1A(2) criteria or who, for practical reasons, are not recognized under them.

The analysis above, demonstrating that the 1969 Convention’s unique refugee definition is not as broad as is often posited,[180] suggests that, in most cases, a refugee recognized under article I(2) will also meet the 1969 Convention article I(1) or 1951 Convention article 1A(2) definition. However, there will remain a very limited number of individuals who meet the 1969 Convention’s unique definition but not the more traditional 1951 Convention definition. Such individuals would include, for example, those fleeing the limited neo-colonial situations that Okoth-Obbo points out continue to occur in Africa,[181] or those who, for evidentiary reasons, do not qualify under the 1951 Convention but whose country of origin is clearly in the grip of events seriously disturbing public order. Moreover, as explained above, in the situations of mass influx that so often characterize refugee movements in Africa, for reasons of expediency, UNHCR and states will often conduct RSD on a prima facie basis under article I(2) of the 1969 Convention.[182] That is not to say that individuals recognized on a prima facie basis would not also qualify for refugee status under the 1951 Convention, but it does imply that the resources required to conduct such an assessment were not, in the circumstances, available. The practical effect of the absence of status determination under the 1951 Convention is that the rights and benefits that flow from that instrument will not automatically inhere. This is why resettlement countries usually reassess the refugee status of article I(2) refugees before accepting them into their resettlement programmes, which under domestic law, usually depend on the resettlement candidate having met the international refugee definition.[183] Host states will clearly not re-evaluate refugee status for the purposes of determining the post-recognition rights to which a refugee is entitled. For this reason, it is critical to determine the rights applicable to refugees recognized only under article I(2).

Remarkably little analysis has been devoted to this issue, perhaps because it seems obvious to many that, in states that are party to both the 1951 and 1969 Conventions,[184] all refugees—regardless of the definition applicable—benefit from the 1951 Convention’s rights framework. Indeed, the scholars who have expressed this view have done so rather matter-of-factly or with very little legal analysis.[185] Among them, Fitzpatrick notes only that the extension of the refugee definition under the 1969 Convention was done “without any suggestion that the quality or durability of ... protection should be diminished as compared to that enjoyed by persons meeting the definition in the 1951 Convention.”[186] In discussing the domestic refugee laws of South Africa and Tanzania, Mandal mentions that the states’ obligations under the 1969 Convention required them to guarantee the same rights to article I(2) refugees as to 1951 Convention refugees.[187] McAdam maintains simply that the 1969 Convention, “as a regional complement to the [1951] Convention, applies [1951] Convention rights to persons fleeing external aggression, occupation, foreign domination or events seriously disturbing public order.”[188] Van Hövell tot Westerflier argues, “[T]he majority of Africa’s refugees, though not falling within the stricter scope of the 1951 Convention, is, at least in theory, entitled to claim the rights set forth in this Convention.”[189] Weis maintains, “The 1951 Convention, the 1967 Protocol and the African Convention, together, constitute a codification of the rights—and duties—of refugees in Africa.”[190] Rutinwa is of the same opinion as Fitzpatrick, Mandal, McAdam, van Hövell tot Westerflier, and Weis, and provides some historical support for his perspective:

That refugees recognised under section I(2) of the OAU Convention are entitled to the same standards of treatment as those recognised under the 1951 Refugee Convention was confirmed by the Arusha Conference which recognised the definitions of the term ‘refugee’ contained in Article I, paragraphs 1 and 2 of the 1969 OAU Refugee Convention as the basis for determining refugee status in Africa and stressed “the essential need for ensuring that African refugees are identified as such, so as to enable them to invoke the rights established for their benefit in the 1951 Refugee Convention and the 1967 Refugee Protocol and the 1969 OAU Refugee Convention”. This was irrespective of the procedure by which they were recognised.[191]

While Rutinwa marshals conference proceedings in support of his view, Durieux and Hurwitz rely on article VIII(2) of the 1969 Convention, which describes the 1969 Convention as the “regional complement” to the 1951 Convention. They argue that

[o]n its face, the only possible interpretation of this provision is that a person recognized as a refugee under either branch of the definition in the complementary OAU Convention is entitled to the rights contained in the primary 1951 Convention.[192]

This view is reiterated in Durieux’s later article with McAdam,[193] and by McAdam herself.[194] Finally, Rwelamira explains that in states that have ratified both the universal and African refugee conventions:

The African refugee would then be able to enjoy the specific and well-defined rights relating to gainful employment, freedom of movement, welfare as well as rights relating to economic pursuit such as, labour legislation, acquisition of property, and other benefits related to employment.[195]

Rwelamira concludes that “[i]n essence ... one should regard the two Conventions as cumulative.”[196]

However, just as the scholars mentioned above have taken the view that article I(2) refugees benefit from 1951 Convention rights, others have adopted the opposite perspective, with a similar near absence of legal argument. Barutciski, describing the 1969 Convention as a protection system “meant to address mass flows”[197]—an erroneous attribution that was discussed above—goes on to maintain that such systems “do not give refugees significant rights beyond non-refoulement guarantees.”[198] He further explains that, while the 1969 Convention’s unique refugee definition

means that it is the sole international refugee treaty which applies to most African refugees, it does not include the elaborate socio-economic rights found in the 1951 Convention. ... Limited rights apparently encourage a more liberal admission policy in situations of mass inflow, while elaborate rights that may lead to integration tend to discourage Governments from allowing refugees to access their territories.[199]

Barutciski is not alone in his perspective. Mendel maintains that

the 1951 Convention may be described as guaranteeing extensive benefits, tending towards residence rights, to a narrowly defined class of individuals. The OAU Convention, on the other hand, provides relatively limited benefits, broadly consistent with maintenance in camps, to those fleeing a wide range of situations.[200]

De la Hunt surely had views such as these in mind when she noted the emergence of an erroneous trend toward viewing the 1969 Convention as proposing “an entirely different kind of refugee regime that [gives] fewer rights to more refugees.”[201] Chartrand had earlier anticipated the emergence of such a trend:

The broadening of the definition of the term refugee could ... raise problems if it leads to the emergence of different classes of refugees—those who qualify for refugee status under all the relevant international instruments and those who qualify under only one—with confusion and disagreement among states and international agencies as to whom to accord which standard of treatment.[202]

While the majority of scholars who have addressed this issue, however briefly, agree that article I(2) refugees benefit from 1951 Convention rights in states party to that instrument, the existence of an opposite minority point of view and the sparse legal reasoning on both sides of the divide suggest that the issue deserves sustained attention. Three principal arguments are advanced in support of the view that article I(2) refugees in host states party to both the 1951 and 1969 Conventions are entitled to the full range of rights guaranteed by the former instrument: a lex specialis argument, an equality argument, and finally, a treaty interpretation argument that is supported by certain general principles of international law. Each of these arguments is addressed in turn below; they are in addition to the obvious point that viewing article I(2) refugees as not entitled to 1951 Convention rights empties article II of the 1969 Convention of all but the most base content.

Before proceeding, however, it is necessary to examine Conclusion No. 22 of the Executive Committee of the High Commissioner’s Programme (ExCom) on the “Protection of Asylum-Seekers in Situations of Large-Scale Influx,”[203] which may, at first blush, appear to support the perspective of Barutciski and Mendel. It could equally appear to support the false proposition discussed above, namely that refugees recognized on a prima facie basis (in situations of mass influx) enjoy fewer rights than their counterparts recognized via individual status determination. The ExCom conclusion begins by noting that large-scale influxes may include refugees within the meaning of the 1951 Convention—persons “who, owing to external aggression, occupation, foreign domination or events seriously disturbing public order in either part or the whole of their country of origin or nationality are compelled to seek refuge outside that country.”[204] It then goes on to enumerate sixteen standards of treatment for “asylum-seekers who have been temporarily admitted to a country pending arrangements for a durable solution.”[205] The list does not include certain rights guaranteed by the 1951 Convention, such as rights to education,[206] to gainful employment,[207] to social security,[208] and to identity papers[209] or a travel document.[210]

The protection regime established by ExCom Conclusion No. 22 applies only, however, to large-scale influxes in which “[s]tates, although committed to obtaining durable solutions, have only found it possible to admit asylum-seekers without undertaking at the time of admission to provide permanent settlement.”[211] In other words, the conclusion applies only to situations of temporary protection—not the circumstance at issue here—and even in such situations, according to Goodwin-Gill, the conclusion provides “a point of departure only.” In certain circumstances, “even a ‘temporary’ solution may require more substantial provision for refugees, including the opportunity to earn a living and to have access to education, housing, and social assistance.”[212] This view was affirmed during UNHCR’s Global Consultations on International Protection, during which it was noted that ExCom Conclusion No. 22 “was never intended as a substitute for standards of protection under the 1951 Convention.”[213] Moreover, McAdam explains that the protection regime established by the ExCom conclusion must be understood within the particular context in which it was adopted, namely the mass exodus from Indochina beginning in the mid-1970s, when numbers overwhelmed individual processing and front-line states were not parties to the 1951 Convention.[214] The conclusion, therefore, “filled a gap by identifying existing normative standards for States not bound by the [1951] Convention or [1967] Protocol.”[215] In considering the very particular circumstances to which it applies, it becomes clear that ExCom Conclusion No. 22 provides no support for the proposition that article I(2) refugees enjoy fewer rights than those recognized pursuant to article I(1) of the 1969 Convention or under the 1951 Convention. Nor does it support the idea that refugees recognized on a prima facie basis enjoy more circumscribed rights than those recognized individually.

A. The 1951 Convention as Lex Specialis

“Complementary protection” describes the protection from refoulement “granted by States on the basis of an international protection need outside the 1951 Convention framework.”[216] The source of the prohibitions against refoulement subsumed within the concept of complementary protection generally arise directly from, or have been read into, international and regional human rights instruments, such as the Convention Against Torture and Other Cruel, Inhuman or Degrading Treatment or Punishment (CAT) and the European Convention on Human Rights.[217] In her groundbreaking study of complementary protection, McAdam argues that the 1951 Convention “functions as a form of lex specialis (specialist law) for all those in need of international protection, and provides an appropriate legal status irrespective of the source of the State’s protection obligation.”[218] In other words, McAdam’s view is that the content of complementarily protected status derives from the rights framework contained in the 1951 Convention, regardless of the fact that the beneficiaries of such status are not refugees within the meaning of its article 1A(2). McAdam advances several propositions in support of this thesis, but the core of her argument is that the 1951 Convention’s

application has been extended through the expansion of non-refoulement under human rights law ... rather than by the conventional means of a Protocol. ... Since the scope of non-refoulement has been broadened by subsequent human rights instruments, this necessarily widens the Convention’s application.[219]

If the 1951 Convention provides the rights blueprint for all beneficiaries of complementary protection, then such rights must apply equally to refugees within the meaning of article I(2) of the 1969 Convention. McAdam argues that the progressive development of human rights-based non-refoulement, in extending the range of individuals entitled to international protection, also extended the applicability of the 1951 Convention. Surely the 1969 Convention’s extension of protection from refoulement to individuals fleeing a 1969 Event extends the 1951 Convention’s applicability in a similar fashion, especially because, as refugees, individuals recognized under article I(2) of the 1969 Convention are in a situation even more conceptually similar to 1951 Convention refugees than beneficiaries of complementary protection. This argument is of even more force when one considers McAdam’s reasoning regarding why human rights instruments do not themselves provide status in addition to protection from refoulement. She argues that it would be “futile for instruments like the CAT to enumerate the legal status arising from the application of non-refoulement, since the Refugee Convention (as the lex specialis) already provides an appropriate status for any person protected by that principle.”[220] By analogy, it would be similarly futile for the 1969 Convention to enumerate an exhaustive list of refugee rights, since the 1951 Convention, as the lex specialis on the content of refugee status, already does so.

The argument that beneficiaries of complementary protection, and hence article I(2) refugees, enjoy 1951 Convention rights by virtue of that convention’s function as the lex specialis for individuals in need of international protection must, however, be approached with some caution in light of Hathaway’s critique of McAdam’s work.[221] Hathaway rejects McAdam’s thesis on two main grounds. First, he argues that “there is no basis to suggest that the Refugee Convention exists to delineate the entitlements of persons granted protection against refoulement.”[222] Second, he maintains that McAdam’s contention is premised on an incorrect application of the concept of lex specialis. The notion of lex specialis is used primarily to resolve conflicts between competing norms of international law but can also be employed “to assist in the construction of a general provision in relation to a matter also governed by a more specific norm.”[223] McAdam employs lex specialis in quite a different sense: to extend the 1951 Convention’s “beneficiary class to embrace persons outside its textual ambit.”[224] According to Hathaway, “Because there is simply a legal void to be filled in relation to non-refugees, there is no conflict of rules that lex specialis can assist to resolve.”[225] Hathaway further argues that the secondary usage of lex specialis similarly provides no support for McAdam’s position:

[T]he importance of interpreting general rules in harmony with more specific rules does not advance McAdam’s thesis that the absence of rules defining the status of the broader class of non-returnable persons must be filled by effectively recasting the Refugee Convention’s beneficiary class.[226]

If, as Hathaway contends, the 1951 Convention is not the lex specialis for all individuals in need of international protection, then it clearly becomes impossible to argue that 1951 Convention rights apply to article I(2) refugees on that basis. In light of Hathaway’s critique, it seems that McAdam’s position is merely lex ferenda. Resolving the issue of whether article I(2) refugees can enjoy 1951 Convention rights does not, however, depend on McAdam’s view being lex lata. A range of other bases exists, detailed below, leading to the conclusion that article I(2) refugees do enjoy 1951 Convention rights, one of which stems from a point on which Hathaway and McAdam agree. Both authors, and others,[227] have found that the law of non-discrimination provides a basis for guaranteeing 1951 Convention rights to beneficiaries of complementary protection.[228] The thesis that the legal duty of non-discrimination mandates the equal treatment of refugees and beneficiaries of complementary protection was developed by Pobjoy,[229] while Hathaway has explored the role of non-discrimination regarding equal treatment between citizens and non-citizens,[230] and between and among 1951 Convention refugees.[231] These approaches can equally be employed for the benefit of article I(2) refugees, as is demonstrated below.

B. Equality

The legal duty of non-discrimination requires that “irrelevant criteria not be taken into account in making allocations.”[232] Article 26 of the International Covenant on Civil and Political Rights (ICCPR) articulates this duty with particular force because the ambit of its guarantee is not limited to the ICCPR alone;[233] rather, it applies to the “allocation of all public goods, including rights not stipulated by the Covenant itself.”[234] Article 26 provides:

All persons are equal before the law and are entitled without any discrimination to the equal protection of the law. In this respect, the law shall prohibit any discrimination and guarantee to all persons equal and effective protection against discrimination on any ground such as race, colour, sex, language, religion, political or other opinion, national or social origin, property, birth or other status.

This guarantee ensures both formal equality (“equality before the law”), as well as substantive equality (“equal protection of the law”). It does not, however, “establish an unconditional guarantee of equality.”[235] Indeed, not every instance of differential treatment amounts to discrimination; in many circumstances, it is perfectly reasonable for a state to differentiate between groups. Rather, equality requires that any unequal treatment be “properly justified, according to consistently applied, persuasive, and acceptable criteria.”[236] Accordingly, the Human Rights Committee (HRC), which is the ICCPR’s treaty-monitoring body, has read the following proviso into article 26:

[T]he Committee observes that not every differentiation of treatment will constitute discrimination, if the criteria for such differentiation are reasonable and objective and if the aim is to achieve a purpose which is legitimate under the Covenant.[237]

Thus, to amount to discrimination under article 26, unequal treatment must be based on criteria that are neither reasonable nor objective, nor in pursuit of a legitimate aim.

Pobjoy has distilled article 26 and its proviso into a convenient three-pronged test for establishing unlawful discrimination:

  1. Has there been differential treatment between individuals in similar circumstances? In other words, is there an inequality basis for a discrimination claim?

  2. Is the unequal treatment based on a ground captured by art 26?

  3. Is the unequal treatment based on “reasonable and objective” criteria?[238]

This test, with the “pursuit of a legitimate aim” appended to its third prong,[239] will be applied to determine whether article 26 prohibits the differential allocation of rights as between refugees recognized pursuant to article 1A(2) of the 1951 Convention or article I(1) of the 1969 Convention, and refugees recognized under article I(2) of the latter instrument.[240] As a preliminary matter, however, it must be established that an article I(2) refugee can invoke the ICCPR vis-à-vis his or her host state, assuming that state has ratified the ICCPR and its optional protocol permitting individual communications.[241] The HRC has confirmed that as a general rule, the rights enshrined in the ICCPR “must be guaranteed without discrimination between citizens and aliens.”[242] All refugees, therefore, benefit from the protections afforded by the ICCPR.

As a starting point, assume that state X—which is party to the 1951 Convention, the 1969 Convention, and the ICCPR and its optional protocol—guarantees the full range of 1951 Convention rights to refugees under that instrument and under article I(1) of the 1969 Convention, but guarantees refugees recognized under article I(2) of the 1969 Convention only the few rights contained therein (non-refoulement[243] and the right to a travel document[244]). The first prong of Pobjoy’s test, which must establish that similarly situated individuals have been treated differently, is clearly answered in the affirmative. Once arrived in state X, a refugee who fled individualized persecution is no different from one who fled a 1969 Event. The two, as refugees, are in similar circumstances, yet they do not enjoy the same allocation of rights.

To establish discrimination, the test’s second prong must demonstrate that the unequal treatment is based on a ground captured by article 26. The unequal treatment by state X is based on the source of the individual’s refugee status. This is not one of the grounds enumerated in article 26; however, the listing there is not exhaustive. The source of the unequal treatment may fall within “any other status”. The HRC has approached the meaning of “any other status” on a case-by-case basis. The only general requirement is that the status in question must capture a distinct group, as opposed to an individual.[245] Refugees recognized under article I(2) of the 1969 Convention are clearly a distinct group. Moreover, according to Pobjoy, because the HRC has repeatedly affirmed that distinctions based upon nationality or citizenship fall within the notion of “other status” in article 26, it “seems reasonable to assume that this principle would apply between different categories of non-citizens.”[246] The second prong of the test is therefore also answered affirmatively.

Under the test’s third prong, to avoid running afoul of the ICCPR, the unequal allocation of rights must be based on reasonable and objective criteria and must be in pursuit of a legitimate aim. Owing to the paucity of jurisprudence on the issue, “[t]he extent to which the Committee is likely to consider differential treatment between categories of non-citizens to be reasonable and objective is unclear.”[247] In the absence of case law, Pobjoy finds that the most sensible approach is to

identify the potential bases which a state may invoke to justify the differential allocation of rights ... and ... critically examine these bases to pre-empt the likelihood that they will be considered reasonable and objective criteria justifying the differential allocation of rights.[248]

The only conceivable basis on which state X might decide to circumscribe the rights of article I(2) refugees would be the belief that their plight is inherently more temporary than refugees fleeing individualized persecution. Such logic has been employed to justify the lower standards of treatment accorded to beneficiaries of complementary (subsidiary) protection under the European Union’s Qualification Directive.[249] This belief is not, however, borne out in practice. An ample number of studies have documented the increasingly protracted nature of refugee situations in Africa[250] and the persistence of the conflicts giving rise to them.[251] Moreover, to accord fewer rights to refugees fleeing a 1969 Event based on the theory that their plight is inherently temporary misconceives the nature of the universal regime of refugee protection, which itself is premised on temporariness. Because the denial of 1951 Convention rights to article I(2) refugees cannot be based on criteria that are objective and reasonable, there is no need to proceed to the test’s second sub-prong regarding legitimate aims. Thus the test’s third prong is answered in the negative.

The application of Pobjoy’s test suggests that distinguishing between the allocation of rights to article I(2) refugees and refugees recognized under article 1A(2) of the 1951 Convention or article I(1) of the 1969 Convention constitutes unlawful discrimination under the ICCPR. UNHCR takes a similar position in more general terms, finding it “doubtful that international law would permit selective provision of international protection according to category.”[252] Indeed, even the 1969 Convention itself recalls “that the Charter of the United Nations and the Universal Declaration of Human Rights have affirmed the principle that human beings shall enjoy fundamental rights and freedoms without discrimination.”[253] There is arguably, therefore, a duty to guarantee 1951 Convention rights to all article I(2) refugees in African states party to the ICCPR and both refugee conventions. This conclusion is, however, tentative, as the HRC has never ruled on the issue and “there is very limited jurisprudence concerning the differential allocation of rights to categories of non-citizens generally.”[254] Indeed, it is unlikely that the issue of discrimination against article I(2) refugees will ever come before the HRC, since states in Africa do not, in practice, distinguish between refugees recognized either under article 1A(2) of the 1951 Convention or under article I(1) of the 1969 Convention, and those recognized only under article I(2) of the 1969 Convention. Nor does domestic refugee legislation in Africa generally make such a formal distinction. These facts are considered in the discussion of state practice in the part that follows.

C. The Law of Treaties and General International Law

That the 1969 Convention is a regional instrument complementing the 1951 Convention is “a novelty in itself.”[255] Notwithstanding the sui generis nature of the relationship between the 1951 and 1969 Conventions, several principles emanating from the law of treaties, which were codified with the 1969 adoption of the Vienna Convention on the Law of Treaties (Vienna Convention), are instructive in determining the rights to which refugees recognized only under article I(2) of the 1969 Convention are entitled. These are the rule on the application of successive treaties relating to the same subject matter,[256] the general rule of interpretation,[257] and the rule on supplementary means of interpretation.[258] Certain general principles of international law are also relevant. Each is addressed in turn below.

1. Successive Treaties Relating to the Same Subject Matter

Article 30 of the Vienna Convention addresses the question of the relationship between successive treaties relating to the same subject matter. This provision, while primarily invoked to resolve explicit conflicts, extends its scope by addressing “more generally the rights and obligations of States part[y] to successive treaties relating to the same subject-matter.”[259] However, doctrinal agreement on the strict construction[260] to be applied to “relating to the same subject-matter”[261]—language that appears in the subparagraph addressing the general applicability of article 30—and to the residuary character of article 30 in relation to conflict clauses[262] raises the threshold question of whether article 30 even applies to the issue at hand.

The 1969 Convention seeks only to regulate matters not already covered by the 1951 Convention,[263] calling into question whether it and the 1951 Convention can be construed as relating to the same subject matter.[264] In this regard, the relationship between the 1951 and 1969 Conventions is an example of the main basis on which Sinclair critiques article 30. He notes:

Article 30 of the Vienna Convention is in many respects not entirely satisfactory. The rules laid down fail to take account of the many complications which arise when there coexist two treaties relating to the same subject-matter, one negotiated at the regional level ... and another negotiated within the framework of a universal organisation.[265]

Furthermore, article 30 is residual in the sense that it only applies “in the absence of express treaty provisions regulating priority.”[266] That is to say that, where a treaty expressly provides how it should relate to another instrument, article 30 is not invoked. The 1969 Convention indeed contains such conflict clauses.[267]

Yet article 30 may nevertheless be instructive in the instant case, because it confirms the primacy of the 1969 Convention’s conflict clauses. Article 30(2) of the VCLT concerns conflict clauses aimed at giving priority to another treaty. It provides, “When a treaty specifies that it is subject to, or that it is not to be considered as incompatible with, an earlier or later treaty, the provisions of that other treaty prevail.” According to Villiger, “[O]n the contractual level, the residual character of Article 30 vis-à-vis the conflict clause of another treaty may be based on its own para. 2—but only when that conflict clause grants priority to other treaties.”[268] The 1969 Convention’s conflict clauses indeed acknowledge the primacy of the 1951 Convention: the ninth preambular paragraph recognizes that the 1951 Convention “constitutes the basic and universal instrument relating to the status of refugees,” and article VIII(2) states that the 1969 Convention “shall be the effective regional complement in Africa” of the 1951 Convention. Article 30(2) therefore mandates that these conflict provisions govern the resolution of the issue at hand.[269] Their proper interpretation necessitates recourse to the general rule of interpretation.

2. The General Rule of Interpretation

The general rule of interpretation provides, “A treaty shall be interpreted in good faith in accordance with the ordinary meaning to be given to the terms of the treaty in their context and in the light of its object and purpose.”[270] According to Sinclair, the ordering of elements within this provision is important: “The initial search is for the ‘ordinary meaning’ to be given to the terms of the treaty in their ‘context’; it is in light of the object and purpose of the treaty that the initial and preliminary conclusion must be tested and either confirmed or modified.”[271] The clauses relevant to resolving the issue at hand are, as discussed above, the 1969 Convention’s conflict clauses.[272] Also relevant is the 1969 Convention’s tenth preambular paragraph, which calls

upon Member States of the Organization [of African Unity] who had not already done so to accede to the United Nations Convention of 1951 and to the Protocol of 1967 relating to the Status of Refugees, and meanwhile to apply their provisions to refugees in Africa.

In interpreting the relevant provisions in line with their ordinary meaning and in context, it becomes clear that article I(2) refugees are entitled to the rights contained in the 1951 Convention. In recognizing the 1951 Convention as “the basic and universal instrument relating to the status of refugees,”[273] the 1969 Convention establishes its predecessor instrument as the global reference point for refugee status in a general sense. In going on to recognize that the 1951 Convention “reflects the deep concern of States for refugees and their desire to establish common standards for their treatment,”[274] the 1969 Convention specifically establishes the 1951 Convention as the universal source of refugee rights. This is why, in the preambular paragraph that follows, the 1969 Convention calls on states that have “not already done so to accede to the United Nations Convention of 1951 and to the Protocol of 1967,” and to meanwhile “apply their provisions to refugees in Africa.”[275] Indeed, according to D’Sa:

[T]he fact that the OAU Convention expressly recognizes the legal status and validity of the UN Convention and encourages the OAU Member States to accede to the latter is ... evidence that the OAU Convention does not preclude the application by its Member States of the additional provisions of the UN Convention relating to such matters as gainful employment, welfare, housing, public education, administrative assistance and so on.[276]

The co-application of the 1951 and 1969 Conventions that emerges from the latter instrument’s preamble is the meaning that must be similarly attributed to the final relevant provision: the conflict clause at article VIII(2). It, again, provides that the 1969 Convention is the “effective regional complement in Africa” of the 1951 Convention. As already mentioned, Durieux and Hurwitz find that “[o]n its face, the only possible interpretation of this provision is that a person recognized as a refugee under either branch of the definition in the complementary OAU Convention is entitled to the rights contained in the primary 1951 Convention.”[277] Durieux and McAdam,[278] and McAdam[279] reach the same conclusion, while Okoth-Obbo explains that, by describing itself as the regional complement to the 1951 Convention, the full scope of application of the 1969 Convention “must be considered as also including the 1951 Convention.”[280] Similarly, Awuku views the language of article VIII(2) as implying that the 1969 Convention does not supersede, but rather that it supplements, the 1951 Convention.[281] Holborn considers article VIII(2), as well as preambular paragraphs 9 and 10, and finds that

[t]he final text of the 1969 OAU Refugee Convention makes clear that it was drawn up to supplement and not to supersede or conflict with the 1951 Convention and the 1967 Protocol. ... The substantive articles of the OAU Convention create obligations to be assumed by contracting states in addition to those they have accepted by becoming parties to the 1951 Convention and 1967 Protocol.[282]

These views suggest that the 1951 and 1969 Conventions are co-extensive in Africa: an individual can be recognized as a refugee under article I(2) of the 1969 Convention, and the rights associated with such status can then be derived from the 1951 Convention.

This initial interpretation is not vitiated when considered in light of the 1969 Convention’s object and purpose, which can be gleaned from the convention’s preamble. The preamble begins by “[n]oting with concern the constantly increasing numbers of refugees in Africa and [by stating that signatories are] desirous of finding ways and means of alleviating [refugees’] misery and suffering as well as providing them with a better life and future,”[283] and goes on to articulate the “need for an essentially humanitarian approach towards solving the problems of refugees.”[284] These preambular paragraphs suggest that a form of asylum featuring only the few rights explicitly recognized by the 1969 Conventionnon-refoulement and the right to a travel document—would be manifestly inconsistent with its object and purpose of alleviating the suffering of refugees in a humanitarian manner. The preamble confirms the preliminary interpretation reached above: refugees recognized only under article I(2) of the 1969 Convention benefit from the rights enumerated in the 1951 Convention.

The general rule of interpretation includes an enumeration of the other elements that must be taken into account, together with the context. Among them are “[a]ny relevant rules of international law applicable in the relations between the parties”[285] and “[a]ny subsequent practice in the application of the treaty which establishes the agreement of the parties regarding its interpretation.”[286] The consideration of each of these elements affirms the conclusion reached above.

That the Vienna Convention mandates the consideration of relevant rules of international law reflects the principle of systemic integration, pursuant to which treaties, as incarnations of international law, are “limited in scope and ... predicated for their existence and operation on being part of the international law system. As such, they must be ‘applied and interpreted against the background of ... general principles of international law.’”[287] Thus “[e]very treaty provision must be read not only in its own context, but in the wider context of general international law.”[288] The international law that is relevant consists of those rules that touch “on the same subject matter as the treaty provision or provisions being interpreted or which in any way affect that interpretation”[289] and that are in force at the time of interpretation.[290] The 1951 Convention and the 1967 Protocol, the Universal Declaration of Human Rights, the ICCPR, and the International Covenant on Economic, Social and Cultural Rights[291] may thus be relevant to the interpretation of the 1969 Convention.[292] Their applicability is by no means, however, a foregone conclusion.

International refugee and human rights law cannot necessarily be applied to the interpretation of the 1969 Convention, because the relevant rules of international law must be applicable in the relations between the parties. It is not clear whether this reference is to the parties to a dispute over the meaning of a particular treaty or to all the parties to the treaty being interpreted. Gardiner’s view is that article 31(3)(c) likely refers to the latter.[293] This raises the question of

whether the focus is on examination of relations between all the parties to the treaty, whether the situation is similar to that of subsequent practice (where the practice must be the concordant practice of a sufficient number of parties coupled with the acquiescence and imputed concurrence of the rest), or whether there is some other interpretation to be given.[294]

Ultimately, the jurisprudence and doctrine on this issue have not produced a clear answer.[295] At the very least, it is arguable that the international human rights law prevailing at the time of interpretation of the 1969 Convention is generally relevant to such interpretation, in particular because “[t]reaties that affect human rights cannot be applied in such a manner as to constitute a denial of human rights as understood at the time of their application.”[296] Invoking international refugee and human rights law to determine whether article I(2) refugees benefit from 1951 Convention rights reaffirms the conclusion reached above; finding otherwise would be nonsensical, as it would effectively deny article I(2) refugees the human rights from which they would otherwise benefit. Of particular note in applying human rights law to the interpretation of the 1969 Convention is the preamble to the 1967 Protocol, which provides that “equal status should be enjoyed by all refugees,” including those who were recognized as a result of “new refugee situations [that] have arisen since the [1951] Convention was adopted.”[297] Applying this provision to an interpretation of the 1969 Convention clearly suggests that the rights framework under the 1969 Convention should be derived from the 1951 Convention.

The subsequent state practice referred to by article 31(3)(b) includes domestic legislation.[298] Among African states that have domesticated the 1951 and 1969 Conventions, many have done so in such a way that affirms that 1951 Convention rights apply to article I(2) refugees.[299] Uganda’s Refugees Act, 2006,[300] for example, includes both the 1969 Convention article I(1) and the 1951 Convention refugee definition, as well as the one articulated at article I(2) of the 1969 Convention, and guarantees both types of refugee a range of rights derived from the 1951 Convention.[301] The same is true in South Africa[302] and Tanzania,[303] among other states. It must be noted, however, that commentary on the Vienna Convention has clarified that the subsequent practice referred to by article 31(3)(b) must be common to all the parties to the treaty sought to be interpreted.[304] Subsequent practice that is not common to all parties may nevertheless “constitute a supplementary means of interpretation within the meaning of Article 32 of the [Vienna] Convention.”[305] This provision is addressed below.

3. Supplementary Means of Interpretation

Recourse may be had to supplementary means of interpretation in order to confirm the meaning flowing from the application of the general rule of interpretation.[306] Supplementary means include, but are not limited to, the preparatory work of the treaty and the circumstances of its conclusion.[307] Several sources are relevant in this regard. First, in 1980, an OAU and UNHCR working group promulgated guidelines to assist OAU member states in domesticating their obligations under the 1969 Convention.[308] According to section 11 of those guidelines, “persons considered as refugees according to the ‘extended’ OAU refugee definition are entitled to the rights and are subject to the duties defined [in] the 1951 United Nations Refugee Convention.”[309] A second supplementary means of interpretation, already discussed above, is the domestic refugee legislation of certain states that are party to the 1969 Convention. Third, and most important, is the history behind the 1969 Convention. While no official set of travaux préparatoires is available for the 1969 Convention, its drafting history supports the interpretation reached here:[310] the 1969 Convention does not create second-class refugees excluded from the 1951 Convention’s rights framework.

4. General Principles of International Law

This conclusion, reached by interpreting relevant clauses of the 1969 Convention in line with the Vienna Convention, is confirmed by relevant general principles of international law. The 1969 Convention’s legal regime may be characterized as a special regime in relation to the 1951 Convention’s more general regime. According to the International Law Commission:

The scope of special laws is by definition narrower than that of general laws. It will thus frequently be the case that a matter not regulated by special law will arise in the institutions charged to administer it. In such cases, the relevant general law will apply.[311]

In this case, the special 1969 Convention does not detail the rights to which refugees are entitled once they are recognized as such; the 1951 Convention’s general regime thus serves to fill the gap.

The principle of lex posterior derogat legi priori, though not strictly applicable because the parties to the 1969 Convention are not identical to those to of the 1951 Convention, is nevertheless also instructive. The principle provides that an earlier treaty applies only to the extent that it is not inconsistent with a later treaty concluded among the same parties.[312] The application of 1951 Convention rights to article I(2) refugees is not inconsistent with the 1969 Convention; indeed application of the 1951 Convention’s rights framework is in keeping with the objects and purposes of the 1969 Convention.

Conclusion

This paper began by surveying the 1969 Convention’s legal innovations. It went on to analyze one such innovation—the unique refugee definition—in order to question the common view that it is far broader than the international refugee definition. Finally, this paper addressed the omission of a rights framework from the 1969 Convention. The discussion of the article I(2) refugee definition revealed that three of the four 1969 Events are now largely irrelevant, that the definition is not entirely objective, and that the 1969 Convention does not apply only to groups, nor was it drafted with a view to allowing the group determination of refugee status. These insights about the article I(2) refugee definition will, I hope, go some way toward correcting common misconceptions about it and, taken together, indicate that the definition is not quite as expansive as is often suggested.

The analysis of the standards of treatment to which article I(2) refugees are entitled suggests that, in states party to both the 1951 and 1969 Conventions, there can be no suggestion that article I(2) refugees enjoy fewer rights than their article I(1) or 1951 Convention counterparts. Ultimately, it seems that the omission of a rights framework from the 1969 Convention was premised upon the specific objectives of the treaty and the futility, to use McAdam’s language,[313] of including a cumbersome rights framework when a simpler legal option existed. Thus the 1969 Convention was not only innovative in, among other things, incrementally expanding the range of individuals who could qualify for refugee status. In applying the content of the existing universal instrument’s status to a broadened class of individuals, the drafters found a pioneering yet pragmatic way of ensuring that refugees in Africa are guaranteed the same rights as refugees elsewhere, but that issues particular to African refugees are not neglected. Indeed, “[t]he substantive articles of the ... regional agreement are intended to create obligations to be assumed by states which are parties to it in addition to those undertaken by becoming parties to the [1951] Convention and the [1967] Protocol.”[314]

However, this approach leaves certain refugees—those in states party to the 1969 Convention but not its universal counterpart, or in states that are party to the 1951 and 1969 Conventions but not the 1967 Protocol, or in states party to the 1951 and 1969 Conventions that have retained the former instrument’s geographical limit[315]—in a precarious situation. Rwelamira explains:

[B]ecause the OAU Convention is regarded as complementary to the UN Convention it has no provisions dealing with substantive or minimum rights, a problem which becomes real when a state member of the OAU Convention is not a party to the UN Convention.[316]

This is why Goundiam argues that to be effective, the 1969 Convention must be “completed by states ratifying ... other more exhaustive instruments.”[317]

It is, however, largely a theoretical problem, which, in practice, extends well beyond the borders of states such as Libya (which is party to the 1969, but not the 1951, Convention). Despite 1951 Convention guarantees, refugees all over Africa find their rights systemically violated; moreover, they are usually without recourse for such violations. In addition to widespread ratification of the 1951 Convention, this convention and its regional counterpart must be implemented and enforced.