Negotiating rights: the politics of local integration.
Subject: Peacekeeping forces (Case studies)
Peacekeeping forces (Political aspects)
Refugees, African (Surveys)
Refugees, African (Case studies)
Refugees, African (Political aspects)
Refugees, African (Political activity)
Actors (Surveys)
Actors (Case studies)
Actors (Political aspects)
Actors (Political activity)
Actresses (Surveys)
Actresses (Case studies)
Actresses (Political aspects)
Actresses (Political activity)
Bribery (Case studies)
Bribery (Political aspects)
International cooperation (Case studies)
International cooperation (Political aspects)
Forced migration (Case studies)
Forced migration (Political aspects)
Migrant labor (Surveys)
Migrant labor (Case studies)
Migrant labor (Political aspects)
Migrant labor (Political activity)
Economic conditions (Case studies)
Economic conditions (Political aspects)
Charities (Case studies)
Charities (Political aspects)
Emigration and immigration law (Case studies)
Emigration and immigration law (Political aspects)
Human rights (Case studies)
Human rights (Political aspects)
Warehousing (Case studies)
Warehousing (Political aspects)
Author: Polzer, Tara
Pub Date: 09/22/2009
Publication: Name: Refuge Publisher: Centre for Refugee Studies Audience: General Format: Magazine/Journal Subject: Sociology and social work Copyright: COPYRIGHT 2009 Centre for Refugee Studies ISSN: 0229-5113
Issue: Date: Fall, 2009 Source Volume: 26 Source Issue: 2
Topic: Event Code: 290 Public affairs Canadian Subject Form: Migrant labour; Migrant labour; Migrant labour; Migrant labour
Product: Product Code: 9100034 Bribery; 9103700 Immigration & Emigration; 9103550 Human Rights; 4220000 Warehousing NAICS Code: 92812 International Affairs; 493 Warehousing and Storage
Organization: Organization: United Nations. Research Institute for Social Development; United Nations. Research Institute for Social Development
Geographic: Geographic Scope: South Africa; Guinea Geographic Code: 6SOUT South Africa; 6GUIN Guinea
Accession Number: 258439534
Full Text: Abstract

Advocacy campaigns against the "warehousing" of refugees in camps suggest the facilitation of local integration as a preferred policy option for states, international organizations, and non-governmental organizations. This paper argues that the institutions, assumptions, and habits that the international refugee protection system has developed over the past forty years hinder our understanding of local integration as a fundamentally political and refugee- and host-driven process. The paper uses African case studies to show how local integration is part of broader processes of local politics. It proposes alternatives to three key assumptions of conventional policy-oriented approaches to local integration: (1) that local integration is a form of local politics rather than institutionalizable process for an exceptional category of people (e.g., refugees); (2) that local integration is negotiated by refugees based on a range of legitimacy claims and forms of exchange rather than primarily based on "refugee rights"-related claims; and (3) that local integration is enabled by hosts for a variety of reasons rather than mainly for reasons related to the idea of "refugee protection."

Resume

Les campagnes contre le <> des refugies dans des camps suggerent la facilitation de l'integration locale comme option a privilegier par les Etats, organisations internationales et organisations non gouvernementales. Le present article soutient que les institutions, hypotheses et habitudes developpees par le systeme de protection internationale des refugies au cours des quarante dernieres annees entravent notre comprehension de l'integration locale en tant que processus fondamentalement politique mis en oeuvre par les refugies et les hotes. La recherche s'appuie sur des etudes de cas africains pour montrer comment l'integration locale fait partie du processus plus large de politique locale et propose des alternatives a trois hypotheses cles des approches classiques a l'integration locale axees sur les politiques : 1) que l'integration locale est une forme de politique locale plutot qu'un processus institutionnalisable pour une categorie exceptionnelle d'individus (par ex., les refugies); 2) que l'integration locale est negocie par les refugies a partir d'une serie de revendications de legitimite et deformes d'echange et non plus principalement a partir de revendications liees aux << droits des refugies >>; 3) que l'integration locale est permise par les hotes pour une variete de raisons, plutot que pour des raisons essentiellement liees a l'idee de la <>

Introduction

The "local integration" of refugees is usually conceived of as a policy option: something which professional institutions could and should plan and implement as a response to displacement. From the perspective of these institutions, such as United Nations agencies, governments, and nongovernmental organizations, this conception is understandable, given that they are debating their own programming options and impacts. However, this paper argues that a policy-oriented approach significantly limits the possibility of understanding the process of local integration and the contexts which facilitate or hinder it. I follow Oliver Bakewell in arguing that academics must move beyond the assumptions and categorizations of policy-oriented thinking and bring the interaction between refugees and others "back into history" (1) by applying "broader social scientific theories of social [and political] transformation and human mobility" rather than "privileging their position as forced migrants as the primary explanatory factor" for protection outcomes. (2)

This paper starts by outlining several key assumptions which the "refugee protection industry" has adopted about itself and about refugees in the past forty years which predispose institutions against local integration. It then discusses a prominent advocacy campaign which critiques some of these assumptions: namely the "anti-warehousing" campaign which promotes local integration as an alternative policy solution to the encampment of refugees. While this campaign against camps is valuable within the policy field. I argue that in critiquing camps and advocating for integration its policy focus nonetheless maintains three related conceptual blinkers. Firstly, it implies that local integration is an institutionalizable process for an exceptional category of people (e.g., refugees); secondly, that refugees integrate through claiming "refugee rights"; and thirdly, that hosts enable integration with the aim of providing "refugee protection."

In contrast, this paper uses predominantly rural African case studies to show how local integration is in practice part of broader processes of local politics. By local politics I mean a process through which individuals and groups negotiate with local power holders for access to needed resources. This approach places refugees and hosts at the centre of the process, rather than professional refugee protection institutions, and assumes that refugees are political actors, using political strategies and tactics just like other individuals and groups. This argument is based on an analysis which includes recognizing a range of legitimacy claims used locally by refugees, apart from "refugee rights," and a variety of reasons why hosts allow and enable integration, apart from reasons relating to "refugee protection."

The paper ends by returning to the policy field and assessing potential policy-based critiques of such an empirical and political understanding of local integration. It concludes that as local integration is largely a process which happens without or in spite of currently dominant institutional interventions, future interventions which wish to support rather than undermine local integration must first have the conceptual tools for understanding its locally specific logics.

Framing Local Integration

Before outlining the policy debate about local integration as an alternative to refugee encampment, let me clarify my understanding of local integration. The term has been used in many, often conflicting, ways. (3) I am not referring to the various types of purportedly self-sufficient refugee settlements, where refugees are largely isolated from local populations by host governments and international actors. (4) Even though these settlements may reduce some of the worst economic dependency problems of fully-catered camps, as it were, they do not change the essential separateness of refugees and therefore their removal from local political life.

Some authors define "local integration" in terms of a final state of similarity to (although not necessarily of assimilation with) local populations. Jacobsen, for example, describes what she calls de facto integration as "where the lived, everyday experience of refugees is that of being part of the local community." This includes lack of physical danger; freedom of movement in the host country and freedom to return to the home country; access to sustainable livelihoods; access to government services like education, health, and housing; social inclusion through intermarriage and social interactions with the host community; and comparable standards of living in comparison with the host community. (5) Jacobsen also emphasizes the importance of formal legal status, ideally permanent residence or citizenship in the host country, as the final step to full integration, since without it de facto integrated refugees remain vulnerable. (6) Crisp defines local integration primarily as reflecting the "assumption that refugees will remain indefinitely in their country of asylum and find a solution to their plight in that state," (7) in contrast to the assumption of temporariness inherent in camps and repatriation programs.

Jacobsen's description of "being integrated" is valuable, as is Crisp's focus on "indefiniteness," especially since both include the understanding that integration need not preclude eventual repatriation or cross-border livelihoods and identities. However, these authors retain many of the assumptions about the refugee protection system which I analyze below--such as the primacy of international and national law, and the "refugee" label--by underemphasizing the social and political process of integration and the local actors involved (not only "refugee protection" professionals). I believe that a very broad, process-focused definition of integration is necessary in order to overcome many of the thought-blinkers "refugee-studies" academics have developed. I define local integration as a process of negotiating access to local legitimacy and entitlement on the basis of a variety of value systems determined by local power holders in dialogue with refugees. Such a broad analysis of negotiation strategies, local value systems and local actors is necessary to develop a subtle, rather than a blueprint, debate on integration. I will return to this below.

There is an extensive literature on the integration of refugees in northern countries, focusing on the interaction between the individual/small group and the host state/society around issues of cultural assimilation, economic access, etc. (8) There are fewer studies on refugee integration in the context of "mass" movements in the "South," but it is definitely a phenomenon that occurs more often than is academically observed. (9) The examples of local integration I will discuss are all African and mainly rural or small-town based, (10) including on my own research among Mozambican refugees in South Africa (2002-2006). (11) The analysis applies equally, however, to locally integrated urban-based refugees, who now make up over half of recorded refugees worldwide. (12)

Finally, my understanding of local integration applies at all stages of the displacement process. (13) Refugee protection institutions commonly distinguish between interim or "temporary" responses to displacement, where the main policy options are seen to be encampment or local integration, and "durable solutions" to displacement, which are voluntary repatriation, local integration or third country resettlement. The a priori distinction between "temporary" and "durable" solutions is largely an institutional and policy construction, since the displaced themselves rarely have a clearly phased or linear interpretation of the temporariness or durableness of their current life situation.

There are many refugees, especially in African host countries, who self-settle and start integrating as soon as they cross the border, usually without any organized intervention on their behalf. Some authors with an intervention-focus have also suggested that local integration should be planned for by refugee protection agencies even in the immediate emergency phase of refugee influx, and not only as a second-phase, medium-term option. (14) Jacobsen explicitly focuses on protracted conflicts during which refugees integrate in the host country while waiting for many decades to "return home." (15) Finally, integration is a concern for those refugees whose countries of origin have achieved some peace and stability but where the country of asylum offers different opportunities, has become "home" in various ways, (16) or has become enmeshed with the country of origin as a locality for cross-border livelihoods. (17)

While local integration is potentially an option at all stages of the refugee experience and for individual as well as large-scale refugee movements, this is not to suggest that it is always an option, or always the best option, or that all refugees wish to integrate in the short or long term. However, it is an option which is often overlooked or actively prevented by intervening institutions because of the assumptions about refugees and refugee protection outlined below. In order to understand in what situations refugees and hosts are likely to succeed in integrating on their own, when targeted outsider interventions are likely to facilitate or hinder the process, or when other forms of intervention are needed, we need a clearer understanding of the integration process itself.

Before developing my approach to local integration as political negotiation, I briefly outline the assumptions about refugee protection interventions that dominate mainstream discussions, including the anti-warehousing debate.

Encampment and Local Integration as Policy Options

Since the 1970s, the major international actors in refugee protection and the majority of academic commentators have seen local integration of refugees in poor host countries as problematic. (18) In terms of international expenditure, (19) academic attention, (20) and media and popular perception, refugee camps have become the dominant response to most mass displacement situations by both host states and the international community. This is especially the case in Africa and increasingly in Asia. (21) The Office of the United Nations High Commissioner for Refugees (UNHCR) and many states see camps as the best interim or preparation stage where displaced people are held and helped until the conditions are ripe for the preferred "durable solution" of repatriation to their country of origin.

In reality, this has not always been the case. For thousands of years, it has been the norm for people who have moved away from conflict (just like people who moved for other reasons) to integrate locally, and local integration was espoused as the natural and ideal solution at the inception of the international refugee regime in the 1950s. (22) Furthermore, a large proportion of displaced people today, whether refugees or internally displaced, still self-settle and integrate, often without contact with official state or international assistance programs. Estimates of independently self-settled refugees are notoriously unreliable, (23) but they have ranged from 30 per cent to 75 per cent of the total number of refugees in Africa. (24)

There are currently approximately 2.5 million refugees living in camps around the world, making up one-third of the refugees counted in the UNHCR's global statistics. In contrast to global percentages, 60 per cent of UNHCR-counted refugees in Africa are in camps. (25) This does not include the much larger numbers of internally displaced persons also living in camps or camp-like settlements. The dominant characteristics of encampment are that camp residents are segregated from the local population and that they have limited freedom of movement or economic independence. Camps are often financed and managed by international organizations, while some are managed by host states.

The policy and practice of refugee encampment has developed historically in explicit opposition to the settlement of refugees among the local population of the host country. Crisp describes the historical and political reasons why host states in Africa increasingly rejected local integration of refugees from the 1970s onward in favour of encampment. These included concern about economic and environmental burdens in poor countries (and in richer countries), security concerns, anger at being "abandoned" by richer nations, fear of the domestic ramifications of popular xenophobia, and the perceived need to reassert sovereignty over porous borders. (26) Other writers note that host governments benefit from the international aid associated with encampment, which would not be forthcoming for self-sufficient, integrated refugees. (27) These arguments recognize a realist state-centric perspective and are based on the view that state rights (sovereignty) trump individual rights, and that citizen rights trump human rights. In this perspective, encampment is a legitimate means to prevent perceived threats and gain desired benefits for the state and for citizens without injuring any significant interests (since refugee interests are not considered a priori significant).

The trend toward camps and away from local integration comes not only from states but also reflects an institutional logic within the international "refugee protection industry" today. (28) This logic has seven elements which make it difficult for professional refugee protection institutions to recognize the localized and political nature of local integration.

First, refugee protection has become a specialized and bureaucratized industry with international, national, and non-governmental organizations dependent on recreating interventions for themselves. (29) Social and political processes which cannot be clearly traced to institutional interventions cannot be used to raise funds or claim positive impact. Second, beyond the immediate material incomes of specific organizations and specialists, there is a deeply held assumption that the responsibility for, as well as the cost of, refugee protection should lie with a centralized institution (the state or an international organization) rather than being diffused among the communities in which refugees might settle. Third, the idea that refugees can bring benefits to host communities, rather than only costs, is a common refrain in the mainstream refugee literature and in the publicity material of refugee assistance institutions, but interventions are rarely conceptualized around the resources and capacities which refugees have independently. Especially encampment is based on the assumption that all refugees are a problem and have problems; not that only those refugees who have problems and are problematic require assistance or intervention.

A fourth consequence of the bureaucratic focus is a tautological definition of who is a refugee: only those people who fit into an intervention-driven definition of refugee are counted and assisted; therefore a perception arises that all refugees are counted and assisted. Those refugees who self-settle and integrate are often not counted at all in the official statistics. (30) Even in situations like Guinea, where only 20 per cent of Sierra Leonean and Liberian refugees are in camps, the greater visibility of camps "strongly [shapes] the image outsiders have of all the refugees." (31)

Fifth, bureaucratization has led to a predilection for blueprint "solutions" and standardized procedures, of which refugee camps are particularly replicable examples. The problems associated with importing camp models without consideration for local specificities have been discussed elsewhere. (32) Sixth, in parallel to the programmatic specialization and blueprints, there have developed deep discursive specializations and blueprints. This includes the assumption that a discourse of refugee identification and refugee rights, as defined by international conventions, will in all cases be beneficial to the refugees concerned and therefore is desired by them. As Andrews points out,

Finally, the international agencies mandated with refugee protection, as well as many academics, have accepted some of the host states' arguments against local integration. (34) Some authors argue convincingly that refugee protection is only possible with the co-operation of host states, and that their priorities and concerns must be taken into account. (35) This is clearly correct in terms of immediate advocacy goals. However, by accepting states' fears of permanence as associated with local integration, and therefore supporting "temporary" encampment interventions, international (and national) non-state actors have contributed to the construction of the refugee as a liminal and apolitical category, a temporary aberration in the "national order of things." (36) In addition to having wide-ranging practical implications for the lives and livelihoods of refugees, as well documented by authors like Smith and Harrell-Bond (37) among others, this depoliticization and dehistoricization of refugees has deep disciplinary and theoretical effects on our ability to understand how refugees actively respond on arrival in a new place. (38)

These assumptions about refugees made by host states and international refugee protection actors must be abandoned in order to understand local integration as an empirical, contextual, historical, and political process. Before describing this integration process in the next section, I briefly discuss the contribution of the "anti-warehousing" campaign, which for several years has been advocating against encampment and for increased use of local integration as a policy option in response to displacement. (39) I argue that while this campaign has merits, it is misleading to present local integration as a different but functionally equivalent policy to encampment, in that it is something which professional refugee protection institutions do in response to large-scale displacement of people. The "anti-warehousing" campaign is focused on identifying the problems with encampment, rather than understanding the process of integration per se. The campaign therefore challenges some of the assumptions underlying the logic of encampment and professionalized refugee protection, but not others.

In contrast to a state-centric logic, the dominant proponents of the anti-warehousing campaign explicitly advance the primacy of international law (such as the 1951 UN Convention relating to the Status of Refugees and the 1948 Declaration of Human Rights) and human and individual rights in informing their judgments. (40) The refugee rights codified in the Refugee Convention are presented as a minimum standard to which host states and especially international institutions such as the UNHCR have subscribed. Measured against this standard, refugee encampment is presented as not only illegitimate, but also "unnecessary, wasteful, hypocritical, counterproductive, unlawful, and morally unacceptable." (41)

As part of this argument, sovereignty and citizenship rights, as emphasized by host states, are acknowledged, bur evidence is presented to debunk many of the security and resource competition concerns routinely expressed by host governments. (42) Smith, who identifies long-term camps not only as dehumanizing in general but also as racist, spells out the moral element very clearly. He points out that when Europeans were designing a system of refugee protection for other Europeans in the 1950s, camps were considered an unconscionable throwback to dictatorial practice and local integration was the assumed and natural route for refugees. If the architects of today's camps, including international actors and host governments, considered refugees as their own equals, they would not consign them to "warehouses." (43)

Regarding the assumption that professional institutions are the best placed to protect refugees, the anti-warehousing campaigners document rights abuses by the UNHCR and international NGOs in detail. They criticize the controlling approach to refugees in camps which often stems from narrowly interpreted organizational imperatives and the pressure on institutions to continue interventions for their own benefit rather than a consideration of refugees as human beings with rights and desires. (44) The conviction that refugees are a potential asset and not a burden is central to the anti-camps argument, which they document with extensive examples. Smith and Harrell-Bond and Verdirame also explicitly refer to the greater psychological well-being and health of self-settled and integrating refugees in comparison with camp-based refugees. (45)

I agree with the anti-warehousing campaigners and most commentators on refugee issues that human rights and the rights enshrined in the Refugee Convention are expressions of what every human being (which is what refugees are in the first instance) should have access to in their lives. What I am arguing is that conventional assumptions about refugee protection limit our ability to understand how refugees actually access and enjoy these rights and resources. There are three key assumptions, all related, which the anti-warehousing debate does not move away from sufficiently, and which are critical to answering the empirical question of how refugees access rights and resources and therefore to understanding the process of local integration.

The first assumption is that refugees are a particular group of people, categorically different from "locals" or other kinds of migrants, with different means of accessing rights and resources, whose actions can be understood through a specialized knowledge. I suggest that refugees negotiate access to rights in relation to local power holders like any other political actor. A second, related, assumption is that a discourse of refugee rights is always the strategically best way of assisting refugees. I submit that often refugees use other identities to claim rights to resources and power in a local context, such as ethnic identification, kinship networks, political clients or allies, etc. In some contexts, the refugee identification does indeed facilitate access to resources, but in others it can be counterproductive and alienating. The third assumption is that "durable solutions" and "refugee protection" are the result of specific interventions by specialized refugee protection organizations based on internationally defined refugee rights. I suggest that local integration often happens as a function of local relationships, in the absence of specific refugee protection interventions, and that it can be actively hindered by them. In fact, refugee protection can be seen as a by-product of successful integration processes, rather than integration being a result of refugee protection. These three points will be discussed in the following sections on the basis of African case studies.

Local Integration as Political Negotiation

My argument is that empirically, local integration is in the first instance a social and political process of incorporating newcomers into an existing political community. This process may or may not be connected to a discursive and institutional framework of "temporary" or "durable" solutions to displacement that assumes certain things about refugees, refugee rights, the role of the state and the role of international organizations.

The underlying questions for understanding integration as a political process is: how do refugees (actively) get what they need to live? This question is empirical and situational, not normative and universal: what does enable refugees to access the things they need, rather than who should protect them. Academics should consistently be looking at how refugees gain access to actual power and resources locally, as achieved through negotiation with a variety of actors, some of whom bur certainly not all may be agents of the state, international organizations, or NGOs. Within this framework, I agree fully with Verdirame and Harrell-Bond's assertion that "socio-economic integration is the best solution from the point of view of the refugees' enjoyment of their fundamental rights" (46) because it gives them more space to actively negotiate access to rights and resources with whatever local power holder can actually enable those rights at any particular time.

There are three guiding questions for a political analysis of refugee integration:

1. Who are the local actors who control access to power and resources and what are their interests?

2. What is the relationship of (different categories of) refugees to those actors--what resources do refugees have which these actors can use, or which threats do they (are they perceived to) present?

3. What tactics and strategies do refugees use within the framework of these relationships to negotiate access to power and resources? (47)

Local Power Holders and Interests

Local actors are all actors who impact directly on the local conditions in which refugees live. In the rural context of my own research in South Africa, powerful actors have been village headmen and religious leaders but also locally based officials of national state organs, such as the Department of Home Affairs, which is responsible for identity documentation. In this conception, local actors may also include staff members of international organizations, if they are present in that locality. I am counting state and international actors as "local" in this context, since the relevant points of interaction for refugees are the locally-based individual representatives of such larger institutions, and the ways in which those individuals act empirically, which may be quite different from the legal mandates or institutional interests their structures theoretically represent. (48) A focus on the specifics of the local is important, since, as numerous case studies point out, the conditions for integration are different from one district to another and from a rural area to the adjoining town, not least because different local actors hold power. (49)

In spite of much debate on "weak" and "failed" states in Africa and elsewhere, the state always remains a central power holder in refugee affairs. Particularly the question of legal documentation, as controlled by the state, is often a crucial constraining factor for refugee options. (50) The inclusion of a variety of actors in the analysis is not to negate the importance of state power to constrain or enable access to rights in many situations. However, the particular character of state power in a particular locality, including the specific meaning and use of documentation, cannot be taken for granted.

Who can access documentation or who needs it is not always a question of the law and formal state policy. As Hovil points out about self-settled Sudanese refugees in Moyo district of Uganda, many have avoided the entire concept of refugeedom by paying tax and carrying "graduated poll tax tickets that act as a means of official identification" virtually equivalent to citizenship. (51) This is effective even though the government requires refugees to be registered and live in camps. Bakewell writes that legal documentation was only relevant for Angolan refugees self-settled in remote and rural Zambian villages if they wanted to travel to towns to work, and that many "locals" had a range of different identity documents, including refugee cards, for different purposes. (52) In South Africa, many Mozambicans with the requisite social or financial capital were "adopted" by South African families of the same ethnic group and acquired citizenship documentation through them. (53)

There are also cases where the state has little influence on border areas where refugees self-settle, such as in Zambia in the early 1990s. (54) In the absence of the state, other significant actors may be traditional leaders who control access to land and have an interest in increasing their political following; international organizations and NGOs who offer resources and have an interest in docile clients; and armed groups who offer or withdraw security and have an interest in recruits.

Matching actors and their interests also illustrates that different local actors will be most useful (or dangerous) to different groups of refugees depending on what resources the refugees bring with them and what strategies they are pursuing. In the Guinean case, for example, rural Mano refugees easily integrated in rural areas, where traditional authorities and the general population were the relevant local actors. Urban refugees from different ethnic groups did not have the resources, such as political clientage traditions, agricultural knowledge, and labour power, to negotiate with these actors and did not want to, and therefore addressed themselves to international actors and the state for aid by establishing a "spontaneous camp" in Thuo town. (55)

Verdirame and Harrell-Bond provide a sophisticated analysis of the roles and power relations between a plethora of actors involved in refugee protection in Kenya and Uganda. (56) They include not only the mandates of the various institutions, but their on-the-ground practices and individual belief systems of staff members. What is needed in studying other refugee contexts, including local integration, is a similarly complex understanding of the powerful actors and their interests and assumptions, with the understanding that the roles of government and organized humanitarian actors may be quite marginal in certain contexts.

Refugee Resources

In the ideal international system of refugee protection, the main resource of refugees in relation to host states and refugee rights organizations is their ability to claim legal and moral capital as refugees, which is seen as sufficient to access resources and protection. In reality, however, refugees can rarely access rights purely on this basis, and many have to (or choose to) make use of other resources they bring with them. These may be material resources of exchange (financial resources, labour power and skills, trading relationships) or various forms of legitimacy claims (kinship, client-patron relationships, ideological affiliations). Hansen's early work on self-settled refugees in Zambia especially shows the relevance of cultural and social resources. (57) An analysis of resources enables us, for example, to see cases of co-ethnic integration not as "natural" occurrences, but rather as examples where negotiation for rights and resources is enabled through various existing legitimacy resources including language, cultural norms such as reciprocity, kinship ties, and common spiritual-religious origins (ancestors) which act along with other, more material considerations.

These resources, however, only facilitate integration if they are deemed valuable by local counterparts or used by refugees for the purpose of integrating. In Malkki's well-known work on Burundian refugees in Tanzania, for example, both camp-based and "town" refugees were of the same ethnic group, which was related to some of the local ethnic groups. For the "town" refugees this affinity assisted with an integration process based on remaining invisible to authorities and locals, while for the camp refugees it was irrelevant in their quest for a separate "mythico-historical" identification as exiles. (58)

Other useful resources which refugees can use to facilitate integration are labour power or their contribution to shifting demographic patterns. An increase in population density in a previously sparsely populated area can help locals attract state services. Bakewell's comment on Zambia applies just as well to Mozambican refugees in South Africa:

In camp situations, the refugees bring the same potential economic, social, and political resources with them, but there is usually no demand from the side of those in power, leaving refugees with fewer bargaining tools. Kinship, language, and symbolic ties are usually non-existent with international actors, whose own prestige is not dependent on attracting political clients except to the extent that there are a sufficient number of "beneficiaries" who remain passive, needy, and controlled. Financial exchanges or individual patron-client relationships that develop between humanitarian staff and refugees in camps, which are so often decried as corrupt and deviant, should in fact be understood as part of the same process by which refugees, just like any other social group, use the resources they have and which are in demand by those in power to negotiate desired outcomes.

Arguing that refugees claim rights as a negotiation for mutual gain with powerful actors, rather than based on abstract principles such as "refugee rights," is not necessarily a purely realist argument. It is in fact a crucial, but often unstated, element of idealist debates on accountability. Rights are only truly protected if those claiming them have a means of sanctioning those with the power to grant or withdraw positive freedoms and rights and the power to infringe or protect from infringement of negative freedoms. Such sanction presupposes a mutual, not entirely one-sided, relationship. By being based on local political negotiation, which requires a measure of mutuality if not necessarily equality, local integration by definition includes more accountability in the relationship between refugees and power-holders than camp-based situations, even as refugee rights advocates and academics are arguing for greater accountability in humanitarian practice. (60)

Refugee Strategies and Tactics

Of course, there is no simple or automatic supply-demand mechanism of resources for rights; resources are deployed by refugees according to particular strategies, depending on the desired outcomes and depending on often significant constraints. Malkki's example is useful once again, as the "camp refugees," in contrast to the "town refugees," did not want to integrate locally but rather to develop and maintain a group identification centring on return to Burundi. They therefore targeted their claims to resources and rights at international actors on the basis of their "refugeehood." We do not know what would have happened if those same refugees had not had powerful actors like the UNHCR who were responsive to this strategy. What other actors would they have turned to for food and shelter? Would they have had to adopt a different strategy?

Refugees also employ more subtle tactics, such as the day-to-day construction of moral legitimacy through the invocation of shared history by Shangaan-speaking Mozambican refugees in relation to their Shangaan-speaking hosts in South Africa. (61) While many, even most, Mozambican refugees remained extremely poor for twenty years after their arrival in South Africa, the mutually reinforced identity link with their hosts has meant that they have not experienced the often violent xenophobia which plagues refugees and migrants in other parts of post-1994 South Africa. (62)

Although I have been emphasizing negotiation for resources and power as an active process, there are clearly different levels of individual or collective agency in the choice of strategies and tactics. Some interactions (such as bribing a UNHCR status determination official to arrange resettlement or offering group allegiance to a chief) are more purposeful and active than others (such as increasing the population density of an area).

This broad model of political negotiation which looks at actors, resources, and strategies has several benefits. It allows us to describe and compare refugee responses across a variety of contexts and times and with a variety of other groups of actors, such as vulnerable citizens, internally displaced persons, and economic migrants. It is only seemingly paradoxical that a situationally defined approach would facilitate comparison; refugee studies has been obsessed with classification and labelling so that camp-based refugees are studied differently from returnees, who are again imbued with different characteristics to internally displaced. This kind of classification without a unifying theoretical model does not help to assess the relative achievement of rights. I am not proposing negotiated rights as a means of understanding local integration only; it is a means of seeing local integration as one scenario on a continuum of relative freedom to negotiate rights which stretches from warehoused refugees, slaves, trafficked children, etc. on the one end, to fully enfranchised citizens on the other. Along this continuum there may be more powerful refugees and vulnerable ones, as well as more or less marginalized citizens.

Sometimes, marginal citizens may have fewer rights than some refugees living among them, especially if we consider prominent political exiles or members of rebel groups befriended with the host state's government. Sudanese refugees in northern Uganda and Kenya enjoy certain rights not by virtue of being refugees, but through their association with the Sudan People's Liberation Army (SPLA) rebel group, which is in political allegiance with the respective host governments. SPLA membership cards allow freedom of movement in much of Uganda, (63) while, in contrast, Acholi citizens of Uganda have been resettled to government "villages" through a violent constraint of movement in the past twenty years. The warehousing critique correctly challenges the treatment of all refugees as a homogeneous, vulnerable, and dependent group, and the approach of analyzing the negotiation of rights gives us a means of acknowledging the more powerful as well as the vulnerable.

This negotiating-rights perspective also allows us to compare current negotiations for rights with historical periods when international refugee law and human rights concepts did not exist as benchmarks. This gives the study of refugee flows and responses a much longer historical context and allows for more cross-fertilization with historical studies in other disciplines. An active, political perspective therefore brings refugees "back into history" in more than one sense, comparatively over time and by not taking them out of a normal process of political change and development through refugee labelling and sequestration in camps. (64)

Local Integration through Refugee Rights?

Using the model of politically negotiated rights, "refugee rights" become one of many possible strategies that refugees can use to access rights within a set of structural constraints. However, claiming "refugee rights" will only be effective as a strategy if there are powerful actors present who are likely to react positively to this strategy. Some host governments, international organizations and NGOs are clearly responsive to this strategic claim. (65) But even in cases where the organization most mandated to respond to the claim of refugee rights--the UNHCR--is powerful, the legal rights of refugees as set out in the Refugee Convention have often been ignored in favour of UN-supported "warehousing." So the invocation of the refugee identification towards institutions mandated to respond to that label has just as often led to a direct constraint of rights as to an expansion of rights.

As noted above, other common strategies which refugees use for claiming access to rights include historical association, political allegiance, and ethnic identification. In the case of Angolan refugees in the Zambian border regions, Bakewell points out that there was an existing historical community spanning the formal border. In the absence of a strong central state presence in the border communities, this historical and ethnic connection was more important than differences of formal citizenship in accessing relevant resources such as land. (66) In terms of political allegiance, the case of SPLA members in Uganda is mentioned above.

The case of Mozambican refugees in South Africa serves as an example of ethnic identification, claimed in parallel, at different points in time, with other forms of legitimacy. Shangaan-speaking Mozambicans escaping the civil war in the mid-1980s were welcomed and integrated in South Africa by the Shangaan-speaking "homeland" government (located immediately on the border with Mozambique) as ethnic and black brothers in explicit opposition to a racist central state. (67) Even though it did not obviate all discrimination or conflict with locals, ethnic identification was therefore an important strategy through which Mozambican refugees could access basic local rights (such as access to land, low-level employment by local residents, etc.) in the first ten years of their stay in South Africa.

After the 1994 transition to democracy in South Africa, the ethnic "homelands" were dissolved. Faced with a central state that still only had a fledgling refugee rights legislation or culture, Mozambicans' strategy shifted to claiming rights from the state (rather than only from local authorities), but by identifying themselves as long-term residents from a neighbouring country rather than as refugees per se. In 1995 and 1996, over 90,000 Mozambican refugees received permanent resident status by virtue of being counted as miners or migrants from within the Southern African Development Community (SADC). Only in 1999, fifteen years after their arrival, did the strategy of claiming "refugee rights" bear fruit for Mozambicans in South Africa when a government amnesty was specifically oriented around the right to legal regularization for Mozambican refugees. (68)

The negotiation for documentation and rights from the state as "migrants" and "refugees" continued to be paralleled by a local negotiation for rights from the host community on the basis of common ancestry and tradition, as mentioned above. (69) In fact, trying to claim local rights on the basis of being a refugee was seen by many Mozambicans as being counterproductive. While South African hospitality initially included empathy with the horrors the Mozambicans had experienced during the civil war and their extreme poverty on arrival in South Africa, this charity soon underpinned an unequal hierarchical relationship. To gain expanded access to resources and rights, many Mozambicans have worked hard to renounce the need for charity, pity, or special treatment on the basis of their refugeehood and have made claims for local equality on the basis of long-term residence, education level, or South African citizenship documents.

Local Integration as Refugee Protection Intervention?

People have moved to escape persecution and wars for thousands of years and have always found ways, sometimes with ease and more often with much difficulty, of establishing new homes among new people, or else moving on until they did. Only rarely were those fleeing war welcomed and integrated merely because they were displaced by war. Even since the invention of the modern refugee regime in the 1950s, most local integration has not been the result of a host-country or host-community strategy based predominantly on the aim of providing refugee protection.

The European Cold War-era refugee regime was centrally about geopolitical and ideological rivalry, rather than about humanitarianism. (70) In Africa, some refugees were welcomed by host states and societies as African brothers and neighbours (not as refugees qua refugees) during the wars of independence. (71) Some, such as the Mozambicans in South Africa (72) or the Liberians in Cote d'Ivoire, (73) have more recently been welcomed as co-ethnic brothers (again not as refugees per se). Other refugees have chosen to evade the host state's official refugee protection system and try their luck as "undocumented migrants" or pose as locals, calculating that there was some greater benefit in this arrangement. (74) In sum, those who were integrated officially were not integrated on the basis of being refugees per se, and those who are integrating unofficially feel that their lives and chances are better by not being recognized or identified as refugees. Neither is therefore really integration on the basis of intentional refugee protection.

There are many case studies of local integration where the absence of state and international interventions has been instrumental in facilitating integration. In Sierra Leone of the early 1990s, for example, Leach found that, "in local terms, 'self-settlement' and 'integration' were not special, but an inevitable and well-precedented way of dealing with [the arrival of Liberian refugees]." (75) Even in the European context, Zetter et al. note that "in Italy, the lack of a nationally coordinated framework of policies for settlement and control of geographical mobility appears to have the effect of facilitating integration and enabling local networks to consolidate and support labor market access and social mobility." (76) When refugees decide not to live in camps but rather to self-settle and integrate, they are often avoiding not only the specific strictures of the camp setting, but just as much the refugee label and concept itself and the involvement of external institutional actors in their lives.

There are three questions implicit in the relationship between local integration and intentional refugee protection interventions. One is whether local integration interventions explicitly motivated by refugee protection aims have had a good record in facilitating local integration. We have very few empirical examples of how well locally integrated refugees would do if they were supported by both governments and the UNHCR without the presence of additional political or institutional interests apart from refugee protection. As mentioned above, in those cases where the government explicitly welcomed refugees (such as Cote d'Ivoire in the early 1990s, (77) Tanzania of the 1960s, (78) or Malawi (79)) there was little UNHCR or NGO assistance and governments had other strategic imperatives. We do have evidence of cases where interventions intended for refugee protection have constrained refugee rights and imperiled existing local integration, including various examples of forced encampment of refugees living independently in villages or towns. (80)

The second question is whether local integration has had a good record in protecting refugees--i.e., whether integrated refugees enjoy more rights and freedoms than refugees in other situations. Compared with long-term camp sequestration, the anti-warehousing advocates have convincingly argued that local integration is generally preferable in terms of basic rights such as the right to free movement, the right to work and self-sufficiency, and the right to association. (81) Others have argued that integration is also preferable to nominally self-sufficient settlements which nonetheless constrain free movement. (82) Still others note that health and socio-economic indicators for self-settled refugees are generally better than for those in camps. (83) Even where the economic conditions are more difficult for refugees outside camps, many choose to stay outside the camps because the freedom to determine their own lives is deemed more important than easy access to some services. (84)

We are left with the third, essentially normative, question concerning the relationship between local integration and refugee protection: what about those refugees who do not have the resources to negotiate for successful integration and what about situations where the local actors offer no space for refugees to negotiate or indeed attack them? What happens/ would happen to refugees in such situations in the absence of an international or national refugee protection intervention such as a camp? What about those subgroups considered the most vulnerable, and with the least independent resources for negotiation, such as unaccompanied children, women with small children, or refugees who are ill or disabled? What about Goma or Kosovo or Darfur, or any other large-scale (and high-profile) mass refugee exodus, especially where there is ongoing and widespread violence in the areas to which people are displaced?

There are undoubtedly contexts in which particular refugees do not and cannot survive without external assistance and without the kind of protection provided by a spatially separate and securitized camp. However, the existence of such particular needs and contexts does not invalidate the point of needing to empirically study and understand local integration where and when it happens. Furthermore, on the normative question of which type of protection (local integration or camp) is on average or aggregate better for particularly vulnerable refugees in contexts of ongoing violence or very large-scale movement, this may be impossible and indeed dangerous to answer, given the context-specific nature of refugee needs, desires, and options.

If we do attempt a review of the available evidence comparing the record of camps versus local integration in these extreme contexts, we have little dear evidence, as noted above, whether encampment interventions have a positive or negative effect in balance on refugee welfare, including for particular subgroups of refugees. It is even more difficult to make counterfactual suppositions; i.e., what would it have been like if there were no camps. In order to make a counterfactual argument, we would have to take away not only the economy of the camps (for the international organizations and host states involved), but also the political option of camps which host states, host communities, and some refugee groups can now use to fall back on international actors. We would also have to imagine away the insidious discursive construction of helpless, dependent and burdensome refugees which comes with camps and which affects even those refugees who are currently not in camps.

We can look at cases where large-scale refugee movements have been absorbed by hosts without the need for camps. Guinea is, to a large extent, such a host country, as were Malawi, Tanzania, and also many European countries after the Second World War. We can also look at cases where refugees, including supposedly vulnerable subgroups, have chosen to remain outside available camps and brave the fight for survival on their own, or at least without handouts based on "refugee rights." Here we do have evidence that seems to show that self-settled refugees suffer less hardship than camp-based ones, or are willing to suffer greater economic hardship for the added freedoms of self-settlement. We should also be doing more comparative work with historical migrations, to see how refugees survived, or not, before the invention of an international system of refugee rights.

Conclusion

To pre-empt a range of criticisms which a focus on refugee agency and the social and political "normality" of the integration process may potentially and to some extent legitimately provoke, let me outline four dangers of this approach. First, it may seem to underplay the structural constraints which refugees face, most significantly the role of the state in deciding on who has rights within its borders. Second, it may seem to gloss over the hardship that many refugees certainly experience. Third, it could be misunderstood as a fatalistic perspective, which holds that if integration happens "naturally" it is inevitable, and if it does not happen "naturally" there is nothing which concerned institutions can do about it, even if vulnerable people are suffering. Finally, it may seem to underestimate the power and importance of the legal framework as a potentially progressive and protective or constraining and discriminatory force. It is certainly not my intention to suggest these things.

In relation to the first two points, I am simply arguing that we need to look at both structural constraints and hardship situationally in each case of local integration (or where local integration is not happening), and how refugees empirically react to and manage them. In looking at how refugees negotiate protection, rights, and access to resources, we should not overlook informal non-state actors (such as community leaders, individual local residents, etc.) and their roles in protecting and enabling access to rights and the management of hardship in the local context. There is no suggestion that constraints and hardships do not exist where local integration takes place, or that they are a priori any greater or lesser than in encampment contexts.

In relation to the following two points--fatalism and law--we should be aware that both the dominant power structures and the laws which form the basis of the ideal refugee protection framework which is often held up as a model are, in fact, quite recent inventions, that they are not universally experienced, and that they have changed significantly in the past fifty, twenty and even five years. They will change again. Therefore, in understanding what actually happens on the ground, we should not start from a prescriptive idea of what should be, how states should act, and how the law should function, but rather start from a descriptive analysis of how states and communities act and how the law functions.

Finally, any comparison of local integration and encampment as ways of living needs to be clear on what benchmark is being applied. The "enjoyment of rights" by refugees is often touted as a measure, assessed through welfare indicators such as access to food, health, education, employment, etc. But this may be missing much of the point of how refugees experience local integration. The freedom to act politically and strategically, i.e., the "freedom to pursue normal lives," (85) may be the more important criterion from their perspective. The defining characteristic of encampment is the constraint of this freedom.

Hovil expresses the significance of freedom well in her study of Sudanese refugees in Moyo District, Uganda:

The actual physical constraint of the camp is matched by a much more subtle constraint, which I am arguing against. This is the assumption that it is up to "us," as professionals and institutions mandated with refugee protection, to "give" refugees freedom to negotiate access to their own rights. Justice Albie Sachs, in the introduction to Verdirame and Harrell-Bond's new book, shows how deeply this perception sits when he says: "there is a need for giving refugees a far more active role in deciding on their future." (87)

I have made an argument primarily about how we think about refugee responses and responses to refugees, not about what specific policy should be followed in response to displacement or to facilitate integration. Anti-warehousing advocacy campaigners are doing valuable work presenting arguments for a change in policy around encampment, within the context of the institutional arrangements (and the concomitant interests) that we have today. This debate against camps is appropriately based on a combination of arguments about human rights (contributing to showing up the hypocrisy and double standards of an aid system supposedly based on this concept), basic welfare, and cost-benefit analysis, thereby showing that camps do not provide a good service to refugees and only an illusory and morally indefensible service to states and international agencies. This campaign is a step in the right direction.

When the follow-on policy question arises, "Well, if not camps, what else shall we do with them then?" one can make many arguments about human rights, basic welfare, and cost-benefit analysis in favour of local integration, and can define government and international aid interventions which are likely to facilitate or constrain it. These are mainly to do with documentation, access to labour markets, and access to/ investment in basic services for all residents of a particular area, as has already been well-argued in documents such as the UNHCR Standing Committee's "Framework for Durable Solutions for Refugees and Persons of Concern," (88) and various academic papers. (89) It is definitely desirable to raise the profile of local integration with refugee rights and advocacy organizations who are arguing within the current, dominant "durable solutions" paradigm.

But I am also suggesting that academics must take the additional step of looking beyond the current institutional arrangements and their demands for particular policy justifications. (90) Empirical research which does not start from an explicit policy perspective is not merely a luxury. Bakewell has argued that "research which is designed without regard to policy relevance may offer a more powerful critique and ironically help to bring about more profound changes than many studies that focus on policy issues from the outset." (91) This is partly because institutions, especially ones specialized in crisis management, tend to have short historical memories and therefore tend to assume that things are as they are because they cannot be otherwise (or at least that only minor policy tweaks here or there are possible). (92) More importantly, by circling around a policy objective, the expert refugee industry is largely talking to itself and telling itself what it should do rather than considering all the actors and options available in real life. This is not in itself wrong, as institutions should be taking responsibility for monitoring their own interventions and holding themselves accountable as much as possible.

However, the problem remains that policy talk is inevitably and by definition skewed toward a consideration of existing institutional responses and logics. It is a simple fact that the majority of people we may call refugees does not actually experience the institutional interventions and logics as institutions assume they do. A significant proportion of refugees is not in camps and has never been "captured" in statistics or target-group specifications. For them, creatively negotiating power relations with local power holders is not a policy option; it is simply what they do. Even of those who are "captured" in the system, only a minority internalize it to the extent that they conform to the models institutions provide--either in camps or during repatriation drives or in local communities. By taking these models as the exclusive (explicit or implicit) benchmarks for refugee actions, we, as academics, run the risk of "warehousing" refugees not only with barbed wire and tents but also with words.

NOTES

(1.) Lisa Malkki, "Speechless Emissaries: Refugees, Humanitarianism and Dehistoricization" in Siting Culture: The Shifting Anthropological Subject, ed. K. F. Olwig and K. Hastrup (London: Routledge, 1997.)

(2.) Oliver Bakewell, "Research Beyond the Categories: The Importance of Policy Irrelevant Research into Forced Migration" Journal of Refugee Studies 21, no. 4 (2008): 432.

(3.) Jeff Crisp, "The Local Integration and Local Settlement of Refugees: A Conceptual and Historical Analysis" Working Paper No. 102, New Issues in Refugee Research (Geneva: UNHCR, 2004).

(4.) See Art Hansen, Refugee Self-Settlement Versus Settlement on Government Schemes: The Long-Term Consequences for Security, Integration and Economic Development of Angolan Refugees (I966-1989) in Zambia (Geneva: United Nations Research Institute for Social Development, 1990); Gaim Kibreab, "Local Settlements in Africa: A Misconceived Option?" Journal of Refugee Studies 2, no. 4 (1989); Karen Jacobsen, "The Forgotten Solution: Local Integration for Refugees in Developing Countries,' Working Paper No. 45, New Issues in Refugee Research (Geneva: UNHCR, 2001), 7.

(5.) Jacobsen, 9.

(6.) See also Kibreab.

(7.) Crisp, "Local Integration" 3.

(8.) Roger Zetter et al., "Summary Report: Survey on Policy and Practice Relative to Refugee Integration" (commissioned by European Refugee Fund Community Actions and conducted by Department of Planning, Oxford Brooks University, Oxford, 2002); Maya Korac, "Integration and How We Facilitate It: A Comparative Study of the Settlement Experiences of Refugees in Italy and the Netherlands" Sociology 37, no. 1 (2003); Michael Fixet al., "Los Angeles on the Leading Edge: Immigrant Integration Indicators and Their Policy Implications" (Washington, DC: National Center on Immigrant Integration Policy, Migration Policy Institute, 2008); Jill Rutter et al., "From Refugee to Citizen: 'Standing on My Own Two Feet': A Research Report on Integration, 'Britishness' and Citizenship" (London" Refugee Support, 2007).

(9.) Jacobsen.

(10.) Wim Van Damme, "How Liberian and Sierra Leonean Refugees Settled in the Forest Region of Guinea 1990-96," Journal of Refugee Studies 12, no. 1 (1999); Oliver Bakewell, "Repatriation and Self-Settled Refugees in Zambia: Bringing Solutions to the Wrong Problems" Journal of Refugee Studies 13, no. 4 (2000); Lisa Malkki, Purity and Exile: Violence, Memory and National Cosmology among Hutu Refugees in Tanzania (Chicago: University of Chicago Press, 1995); Lacey Andrews, "When Is a Refugee Nota Refugee? Flexible Social Categories and Host/Refugee Relations in Guinea" Working Paper No. 88, New Issues in Refugee Research (Geneva: UNHCR, 2003); Lucy Hovil, "Free to Stay, Free to Go?: Movement, Seclusion and Integration of Refugees in Moyo District," Working Paper No. 4 (Kampala: Refugee Law Project, 2002); D. Stone and M. De Vriese, "Refugee Livelihoods: Livelihood Strategies and Options for Congolese Refugees in Gabon. A Case Study for Possible Local Integration" EPAU Report (Geneva: UNHCR, 2004); Tania Kaiser, Lucy Hovil, and Zachary Lomo, "'We Are All Stranded Here Together': The Local Settlement System, Freedom of Movement and Livelihood Opportunities for Refugees in Arua and Moyo Districts,' Working Paper No. 14 (Kampala: Refugee Law Project, 2005); Ana Low, "Local Integration: A Durable Solution for Refugees?" Forced Migration Review 25 (2006): 64-5; Lacey Andrews Gale, "The Invisible Refugee Camp: Durable Solutions for Boreah 'Residuals' in Guinea" Journal of Refugee Studies 21 (2008): 537-52.

(11.) My research was conducted as part of the Refugee Research Project, now part of the Forced Migration Studies Programme (FMSP) of the University of the Witwatersrand. From 2002-2006 I was resident in Bushbuckridge District, South Africa, bordering Mozambique, where approximately 30 per cent of the population in the east of the district are refugees from the Mozambican civil war. My methodologies included repeated structured and unstructured interviews with members of fifty-seven households in three villages, half of South African and half of Mozambican origin, as well as key informant interviews and participation in public village events.

(12.) UNHCR, "2009 Global Trends; Refugees, Asylum-Seekers, Returnees, Internally Displaced and Stateless Persons" (Geneva: UNHCR, 2009).

(13.) I do not use the term "refugee cycle" here, since it has been so strongly associated with repatriation as the "natural" end of the "cycle."

(14.) Tom Kuhlman, "Responding to Protracted Refugee Situations: A Case Study of Liberian Refugees in Cote d'Ivoire" ed. UNHCR EPAU (Geneva: UNHCR EPAU, 2002), 40.

(15.) Jacobsen, 1.

(16.) Tara Polzer, "'We Are All South Africans Now': The." Integration of Mozambican Refugees in Rural South Africa," Working Paper No. 8 (Johannesburg: FMSP, 2004).

(17.) Crisp, "Local Integration."

(18.) Crisp, "Local Integration"; NGO Statement on Local Integration Global Consultations on International Protection 22-24 May 2002, http://www.icva.ch/ doc00000865.html. The UNHCR 1995 report The State of the World's Refugees: In Search of Solutions ignored local integration almost entirely.

(19.) Merrill Smith, "Warehousing Refugees: A Denial of Rights, a Waste of Humanity" in World Refugee Survey, ed. Merrill Smith (2004); Alexandra Fielden, "Local Integration: An Under-Reported Solution to Protracted Refugee Situations," Working Paper No. 158, New Issues in Refugee Research (2008), 3.

(20.) Jacobsen; Fielden, 1.

(21.) UNHCR, "2009 Global Trends"

(22.) See 1951 UN Convention relating to the Status of Refugees, Article 34.

(23.) Tara Polzer and Laura Hammond. "Editorial Introduction: Invisible Displacement," Journal of Refugee Studies 21, no. 4 (2008): 417-31.

(24.) Crisp, "Local Integration"

(25.) UNHCR, "2009 Global Trends" 1, 15. The estimates of up to 75 per cent of African refugees being self-settled assume that most of these refugees are not included in UNHCR statistics.

(26.) Crisp, "Local Integration," 5.

(27.) Jacobsen; Smith, "Warehousing Refugees."

(28.) Clearly, there are many different actors concerned with refugee protection and refugee rights, and there are significant variations in approach, influence, and impact. I do not wish to suggest that all intervening organizations or all host states share all of the assumptions listed in these two paragraphs. Nonetheless, I believe there is a core logic that informs most of the large-scale, high-visibility refugee protection institutions and interventions, such as the UNHCR and large NGOs involved in refugee camps, and that this core also structures what other, smaller actors do and think.

(29.) B.E. Harrell-Bond, Imposing Aid: Emergency Assistance to Refugees (Oxford: Oxford University Press, 1986).

(30.) Bakewell. "Research Beyond the Categories"

(31.) Van Damme, "How Liberian and Sierra Leonean Refugees Settled in the Forest Region of Guinea 1990-96," 36.

(32.) See for example Wim Van Damme, "Do Refugees Belong in Camps? Experiences from Goma and Guinea" The Lancet 346 (1995).

(33.) Andrews, "When Is a Refugee Not a Refugee?" 1; also Gale, "The Invisible Refugee Camp"

(34.) Guglielmo Verdirame and B. E. Harrell-Bond, "Rights in Exile: Violations of Refugee Rights in Camps" (paper, 9th IASFM, Sao Paolo, Brazil, 9-12 January 2005). See also Bakewell, "Research Beyond the Categories."

(35.) Jacobsen, 3.

(36.) Lusa Malkki, "Refugees and Exile: From 'Refugee Studies' To the Natural Order of Things," Annual Review of Anthropology 24 (1995).

(37.) Smith, "Warehousing Refugees"; Harrell-Bond, Imposing Aid.

(38.) Malkki, "Refugees and Exile"; Malkki, "Speechless Emissaries"

(39.) See the anti-warehousing campaign website, http://www .refugees.org/warehousing. See, most recently (late 2008), the initiative to entrench self-reliance as a standard of refugee care: http://www.refugees.org/conclusion. The extensive campaign to date has the support of over one hundred NGOs from around the world, predominantly from the South, as well as significant support from within the UN system. See Merrill Smith, ".Warehousing Refugees"; Jacobsen, "The Forgotten Solution"; Jeff Crisp, "No Solutions in Sight: The Problem of Protracted Refugee Situations in Africa" Working Paper No. 75, New Issues in Refugee Research (2003).

(40.) Smith, "Warehousing Refugees"; B. E. Harrell-Bond and Guglielmo Verdirame, Rights in Exile: Janus-Faced Humanitarianism (New York and Oxford: Berghahn Books, 2004); Jacobsen, "The Forgotten Solution"

(41.) Smith, "Warehousing Refugees":38.

(42.) Ibid.

(43.) Ibid., 39.

(44.) Harrell-Bond and Verdirame, Rights in Exile: Janus-Faced Humanitarianism.

(45.) Smith, "Warehousing Refugees"; Verdirame and Harrell-Bond, "Rights in Exile: Violations of Refugee Rights in Camps."

(46.) Verdirame and Harrell-Bond, "Rights in Exile: Violations of Refugee Rights in Camps"12.

(47.) Certeau defines "tactics" as "small manoeuvres" that subvert and play with dominant meanings. "Strategies" refer to more calculated everyday efforts to advance one's position from within a given set of structural constraints; Certeau 1984: 37, cited in Graeme Rodgers, "When Refugees Don't Go Home: Post-War Mozambican Settlement across the Border with South Africa" (PhD dissertation, University of the Witwatersrand, 2002), 4.

(48.) Michael Lipsky, Street-Level Bureaucracy: Dilemmas of the Individual in Public Services (New York: Russell Sage Foundation, 1980).

(49.) Van Damme, "How Liberian and Sierra Leonean Refugees Settled in the Forest Region of Guinea 1990-96"; Oliver Bakewell, "Returning Refugees or Migrating Villagers? Voluntary Repatriation Programmes in Africa Reconsidered" EPAU Working Paper (Geneva: UNHCR, 1999), Malkki, "Purity and Exile"; Andrews, "When Is a Refugee Not a Refugee?"; Kaiser, Hovil, and Lomo.

(50.) Verdirame and Harrell-Bond, "Rights in Exile: Violations of Refugee Rights in Camps," 12; Crisp, "The Local Integration and Local Settlement of Refugees: A Conceptual and Historical Analysis"; Jacobsen, "The Forgotten Solution: Local Integration for Refugees in Developing Countries."

(51.) Hovil, "Free to Stay, Free to Go?," 6

(52.) Bakewell, "Returning Refugees or Migrating Villagers?" 12.

(53.) Chris Dolan, "The Changing Status of Mozambicans in South Africa and the Impact of This on Repatriation to and Re-Integration in Mozambique" (Maputo: Norwegian Refugee Council, 1997); Tara Polzer, "Akdapting to Changing Legal Frameworks: Mozambican Refugees in South Africa--an Historical Overview," International Journal of Refugee Law 19, no. 1 (2007).

(54.) Bakewell, "Repatriation and Self-Settled Refugees in Zambia"

(55.) Van Damme, "How Liberian and Sierra Leonean Refugees Settled in the Forest Region of Guinea 1990-96," 39.

(56.) Harrell-Bond and Verdirame, Rights in Exile: Janus-Faced Humanitarianism.

(57.) Art Hansen, "Once the Running Stops: Assimilation of Angolan Refugees into Zambian Border Villages," Disasters 3, no. 4 (1979):370-371.

(58.) Malkki, Purity and Exile.

(59.) Bakewell, "Returning Refugees or Migrating Villagers?" 12; see also Polzer, "Adapting to Changing Legal Frameworks."

(60.) See the Humanitarian Accountability Partnership International, http://www.hapinternational.org/.

(61.) Rodgers; Polzer, "Adapting to Changing Legal Framework"; Tara Polzer, "Invisible Integration: How Bureaucratic, Academic and Social Categories Obscure Integrated Refugees," Journal of Refugee Studies 21, no. 4 (2008): 476-97.

(62.) Tara Polzer, "Lessons from a Border Area: Why There Is Relatively Little Xenophobia against Mozambicans in Bushbuckridge, Mpumalanga," In Submission to Open Hearing on Xenophobia and Problems Relating to It, Hosted by the South African Human Rights Commission with the Portfolio Committees of the Departments of Foreign Affairs and Home Affairs, Johannesburg, 1-3 November 2004 (Johannesburg: Acornhoek Advice Centre, Wits University, 2004).

(63.) Hovil, "Free to Stay, Free to Go?," footnote 69.

(64.) Malkki, "Refugees and Exile."

(65.) By using the term "strategic" I do not intend to question that many refugees who use this strategy are genuine refugees tender the Convention definitions. I am merely pointing out that genuine refugees can also choose to use other bases for claiming rights.

(66.) Bakewell, "Repatriation and Self-Settled Refugees in Zambia."

(67.) E. Ritchken, "Leadership and Conflict in Bushbuckridge: Struggles to Define Moral Economies within the Context of Rapidly Transforming Political Economies" (PhD dissertation, University of the Witwatersrand, 1995).

(68.) Polzer, "Adapting to Changing Legal Frameworks."

(69.) Rodgers, "When Refugees Dofft Go Home."

(70.) Laura Barnett, "Global Governance and the Evolution of the International Refugee Regime" International Journal of Refugee Law 14, no. 2/3 (2002): 238-62.

(71.) J. R. Rogge, "Repatriation of Refugees: A Not Simple 'Optimum' Solution," in When Refugees Go Home, ed. Tim Allen and Hubert Morsink (London: James Currey, 1994): 18.

(72.) Polzer, "Adapting to Changing Legal Frameworks."

(73.) Kuhlman, "Responding to Protracted Refugee Situations."

(74.) See Malkki, Purity and Exile; Hovil, "Free to Stay, Free to Go?"; Harrell-Bond and Verdirame, Rights in Exile: Janus-Faced Humanitarianism for some studies of refugees who have consciously avoided camp settlement.

(75.) M. Leach, "Dealing with Displacement: Refugee Host Relations, Food and Forest Resources in Sierra Leonean Mende Communities During the Liberian Influx 1990-91," in Research Report No. 22 (Brighton: Institute of Development Studies, 1922), 2, as quoted in Rodgers.

(76.) Zetter et al., "Summary Report: Survey on Policy and Practice Relative to Refugee Integration" (emphasis added).

(77.) Kuhlman, "Responding to Protracted Refugee Situations"

(78.) Loren B. Landau, "Beyond the Losers: Transforming Governmental Practice in Refugee-Affected Tanzania" Journal of Refugee Studies 16, no. 1 (2003).

(79.) Harri Englund, From War to Peace on the Mozambique-Malawi Borderland (Edinburgh: Edinburgh University Press for the International African Institute, 2002).

(80.) Guglielmo Verdirame, "Human Rights and Refugees: The Case of Kenya" Journal of Refugee Studies 12, no. 1 (1999); Simon Turner, "Suspended Spaces--Contested Sovereignties in a Refugee Camp" in Sovereign Bodies: Citizens, Migrants, and States in the Postcolonial World, ed. T. B. Hansen and Finn Stepputat (Princeton: Princeton University Press, 2005); Harrell-Bond and Verdirame, Rights in Exile: Janus-Faced Humanitarianism.

(81.) Harrell-Bond and Verdirame, Rights in Exile: Janus-Faced Humanitarianism; Smith, "Warehousing Refugees."

(82.) Kaiser, Hovil, and Lomo.

(83.) Van Damme, "Do Refugees Belong in Camps?"

(84.) Smith, "Warehousing Refugees."

(85.) Ibid., 38.

(86.) Hovil, "Free to Stay, Free to Go?" 22.

(87.) Harrell-Bond and Verdirame, Rights in Exile: Janus-Faced Humanitarianism (emphasis added).

(88.) UNHCR Standing Committee, "Framework for Durable Solutions for Refugees and Persons of Concern," ed. EC/53/ SC/INE3 (September 16, 2002).

(89.) Jacobsen; Crisp, "Local Integration" 6; C. Dolan, "Aliens Aboard: Mozambicans in the New South Africa" Indicator SA 12, no. 3 (1995).

(90.) Bakewell. "Research Beyond the Categories"

(91.) Bakewell. "Research Beyond the Categories" 433.

(92.) Larry Minear, "Humanitarian Action and Peacekeeping Operations" (background paper for UNITAR/IPS/NIRA Conference, Singapore, 24-26 February, 1997).

Tara Polzer is a senior researcher with the Forced Migration Studies Programme, University of the Witwatersrand (tara.polzer@wits.ac.za). She is also affiliated with the Development Studies Institute (DESTIN) at the London School of Economics and Political Science.

An early version of this paper was presented at the 9th Conference of the International Association of the Study of Forced Migration, Sao Paulo, Brazil, 9-12 January 2005. The author would like to thank the Wits Humanities Graduate Centre Publications Project for its support and Susan van Zyl as well as two anonymous reviewers for their comments.
UNHCR and humanitarian agencies commonly use the category of
   "refugee" in order to determine the population eligible for aid or
   resettlement. However, for understanding ... how the displaced
   themselves negotiate their survival with their hosts, this
   demographic category obscures more than it reveals. (33)


The arrival of the refugees was regarded positively by Zambian
   villagers as not only did they cultivate the bush, but they also
   boosted the population to levels better able to draw in services
   such as schools and clinics. For the chiefs and headmen, the
   increased population also increased their prestige. (59)


The most striking contrast between those refugees living in
   settlements and those who are self-settled is not the difference in
   relative standards of living, but the response they have to their
   predicament. The feeling of powerlessness pervading the interviews
   with settlement refugees stands in direct contrast to those who had
   opted out of the refugee assistance structures and were taking
   responsibility for their lives. There is obviously a danger of
   over-romanticising the lives of self-settled refugees. By no means
   are their circumstances easy--not least of all because they are not
   recognized by the refugee assistance structures in operation.
   However, the fact remains that their ability to move freely has a
   positive impact on their lives, allowing them to utilize fully the
   resources around them and make choices based on where they exist.
   (86)
Gale Copyright: Copyright 2009 Gale, Cengage Learning. All rights reserved.