Camp and settlement issues

Historical overview

Human settlements appear as a relatively natural form of human life, both during peacetime and war. The origins of (refugee) camps are more difficult to trace. Malkki (1995) , for instance, has traced their lineage within the international refugee regime to the very origins of the latter, that is the camps for the displaced in post-war Europe.

In Africa, where the debate between proponents of both self-settlement and planned settlements as well as relief-type camps has been most vocal in the past, historical debates about the mechanisms and methods of refugee assistance can be traced through a number of landmark conferences and events.

Many observers credit the 1967 conference on the Legal, Economic, and Social Aspects of the African Refugee Problems, which was convened in Addis Ababa under the auspices of the UN Economic Commission for Africa, the Organization for African Unity ( OAU, now the African Union or AU), the UNHCR, and the Dag Hammarskjöld Foundation, with providing the first big step towards an integrated approach to refugee assistance. The desire to link refugee assistance to the development needs of the host country was implicit in the final recommendation, which called for a zonal development approach based on the sharing of responsibility by host governments, UNHCR, UNPD, and non-governmental organizations ( NGOs).

However, Integrated Rural Development ( IRD) as a model for refugee assistance preceded Addis Ababa: Betts (1981) describes similar projects set up in Kivu, Zaire, and Burundi, which were based on close cooperation between the UNHCR and the International Labour Office ( ILO) and, in Burundi, the League of the Red Cross and the Food and Agriculture Organization of the United Nations (FAO). The Zairan project was administered as a joint initiative of UNHCR and ILO, as the main agency showed some signs of economic success, but fell prey to political disturbances that caused the death of the two main administrators. In Burundi the lack of expertise that was required for long-term planning as opposed to emergency relief posed problems. Overall, Betts concludes that these early attempts at IRD failed because of poor definition of ultimate objectives, general project mismanagement, discontinuity created by rotation of personnel, and the deteriorating political situation in 1972. Whatever the reasoning, it is at this stage that Pitterman observes a more fundamental move in UNHCR's budget, from an emphasis on rural settlements to emergency relief (Pitterman 1984) .

Despite setbacks, the idea of linking refugee relief explicitly with the overall social and economic dynamics of the host countries survived in small circles and was to become an issue again. In line with the recommendation of an internal UNHCR Seminar held in 1976, the Pan-African Conference on the Situation of Refugees in Africa, held in Arusha, Tanzania, in 1979, reiterated the themes evoked in Addis Ababa and came out in favour of spontaneous rather than formal settlement.

The first International Conference on Refugees in Africa ( ICARA I) was eventually convened in 1981 by UNHCR, the UN Economic Commission for Africa, and the OAU Committee of Fifteen on Refugees. But only with ICARA II in 1984 were integrationist projects given something of a new boost. ICARA II was called for partly because ICARA I had not raised enough funds for infrastructure projects (Kibreab 1991) . Its purposes were defined as threefold: (1) to thoroughly review the results of ICARA I; (2) to consider providing additional international assistance to refugees and returnees in Africa for relief, rehabilitation, and resettlement; and (3) to consider the impact imposed upon the national economies of the concerned countries and to provide assistance to strengthen their social and economic infrastructure to cope with the burden of refugees and returnees.

All such attempts were based on the belief that the provision of relief based on large-scale administration to refugees in camps or settlements isolated from the host societies was an inappropriate form of assistance, and that refugees could serve as resources of development. At ICARA II, 128 different RAD project proposals were presented, requesting a total amount of US$362 million. Most project proposals focused on large infrastructure projects (Clark and Stein 1985) . However, issues that loom high in the camp–settlement debate today, such as the rights to employment, security of status, and other socio-economic and political rights, were not discussed.

ICARA II stands as the last large and visible attempt to organize concerted action for RAA. Among the reasons for its failure, Kibreab notes the actors' divergent interpretations of the ultimate aim of developmental refugee assistance and a failure to guarantee the principle of additionality ("additionality" refers to the idea that any investment in RAD should be supplementary instead of substituting for development aid) as guidelines for pledges made for ICARA II projects. Furthermore, divisions and rivalries among the assistance agencies, NGOs, and host-government departments, as well as a failure to set out a framework for their co-ordination, played a role (Kibreab 1991) . Gorman also points out that great famine in sub-Saharan Africa converged to focus donor and media attention on emergency relief (Gorman 1987) .

[TOP]

Rights and legal standards

As far as legal aspects are concerned, scholars have focused on the way in which camp settings themselves are conducive, or not, to the maintenance of refugee rights. Some observers maintain that camps can provide both security and effective material assistance to refugees, thereby not only assuring the most basic of rights, the right to life, but also facilitating the monitoring of protection issues (Jacobsen 2001). Jamal in particular has made a strong argument that "camps strengthen asylum by encouraging hosts to accept the presence of refugees". (Jamal 2003:4) This argument is based on the belief that "host fatigue" in many refugee-hosting countries is only held in check through the material presence of refugee camps. Camps are thus part of international "burden sharing".

Critics argue that the maintenance of camps does not only involve direct breaches of basic human and refugee rights, but also creates situations in which other rights are more likely to be endangered. For instance, in its campaign on refugees launched in 1997, Amnesty International ( AI) attacks primarily the restrictions on freedom of movement that some camps represent (Amnesty International 1997) . Human Rights Watch ( HRW), on the other hand, has written on the problems emerging especially for women in refugee camps. A more recent topic concerns the ways in which a variety of different and often parallel legal systems inter-relate in camp settings. These include so-called traditional courts and conflict-resolution mechanisms inside camps, the legal system of the host country, and lastly the international legal framework of the refugee, which is the frame of reference for UNHCR protection officers. Such debates are of course closely linked to debates about the protection mandate of the UNHCR and its relationship to the provision of material assistance.

[TOP]

Security

A common argument in favour of camp-based assistance is that it serves to contain the security problems introduced by refugees, to reduce conflict between host and refugees, and/or to control the potential of refugees from civil war to use their host country as a sanctuary from attack. Other security issues also include raids by rebel groups, pursuit of refugees by military forces of the country of origin, the importation of small arms, and generally increasing levels of "banditry" and crime that are related to the current condition of refugee populations.

In Africa, many host states therefore justify control on the movement of refugees by citing Article 2(6) of the OAU Convention, which is interpreted as giving states full rights to decide on refugee settlement and the settlement patterns of the refugees. The article actually states that "for reasons of security countries of asylum shall, as far as possible, settle refugees at a reasonable distance from the frontier of their country of origin". This contrasts with Article 26 of the Convention: "each state shall accord to refugees lawfully in its territory the right to choose their place of residence and to move freely within its territory, subject to any regulations applicable to aliens in the same circumstances".

Especially since the 1990s, security-based arguments for encampment have been viewed with more scepticism. As Jacobsen puts it: "Camps do not solve security problems. They are in fact added sources of instability and insecurity … because they aggravate existing security problems and create new ones" (Jacobsen 2001). These arguments hold that camps may create conflict between refugees and their hosts where refugees are perceived as privileged by the members of the host population, which is sometimes as poor as or poorer than the refugees. They also, as Durieux (2000) among others has pointed out, provide fertile ground for recruitment of young men and woman for military activities by rebel groups. Bulcha (1988) shows moreover that more often the conflict within the refugee populations exceeds the potential conflict between them and their hosts. He specifies that, whilst differences of religion, ethnicity, and politics partially account for the latter conflicts, the most frequent causes are "relief-induced", arising from frustration and idleness.

[TOP]

Health

For many, the professionalization of refugee assistance and the parallel development of today"s refugee camps is at its most basic level an answer to high mortality rates in Africa's refugee crises. Refugee health is one of the most studied aspects of refugee assistance and encapsulates issues ranging from nutrition to reproductive health, mental health, and trauma treatment.

Immediately following a refugee influx, an initial emergency phase is identified by a crude mortality rate ( CMR) of one or more deaths daily per 10,000 people. In general such mortality rates are at least double pre-displacement levels. "Most deaths result from preventable and treatable infections, often exacerbated by malnutrition, caused mainly by diarrhoeal disease, respiratory tract infections, measles, and malaria" (Spiegel 2002:1,927) . During this period the focus in on immediate life-saving interventions: "Acute refugee crises such as those that have occurred recently in Goma, Bosnia, Somalia, Kosovo, East Timor, Angola, Sierra Leone, the Democratic Republic of Congo, to name but a few, are the emergency rooms of international public health" (Waldman 2002) .

A set of different questions emerge from an increasing awareness that many populations affected by complex humanitarian emergencies have been displaced for long periods, living relatively settled lives (Spiegel et al. 2000). Here especially, the range of issues involved in the health sector is ever-expanding, especially as public health concerns – as opposed to curative care – become important. Waldman (2001) has recently observed that more and more emergency health care is affected by "confusion on the issue of priorities". In regards to the debate about forms of settlement the main health questions relate to:

1. The effectiveness of emergency health care and how it is affected by different spatial settings.
2. The changes necessary for post-emergency settings: "Humanitarian organisations often provide similar services in the emergency and post-emergency phases of complex humanitarian emergencies, despite increasing evidence and consensus that needs differ between phases" (Spiegel et al. 2002:1,932) .
3. The proper way to manage health services for refugees in a variety of settings, namely via the establishment of parallel centres or attempts to work through local health systems.

In answering these questions, research is frequently hampered by real difficulties of measuring performance. This is so even where mortality rates are accepted as the prime dependent variable. A number of reasons for this exist, among them unavailability of accurate data due to poor record-keeping or underreporting of deaths in the camps. In addition, "data have mainly been obtained from the acute phase of complex emergencies, in which excess mortality as well as political interest, media attention, and funding is greatest. The post-emergency phase has been little studied, and no comprehensive programme guidelines exist for this phase" (Spiegel et al. 2002:1,927) . While there is much work done on health services in refugee camps, there are only a few treatments of refugee health care in more dispersed settlement situations (see notably Van Damme 1998) .

With the above caveats in mind, there is wide-spread evidence that camps do indeed allow quick detection and treatment of health problems for refugees and that refugee camp health services are usually better supplied and organized than pre-existing services for the host population. Yet, inversely, adverse health effects of the camp environment can be numerous (see for instance Spiegel and Quassim 2003) . The main factors causing epidemics and high mortality rates among displaced people are frequently: overcrowding in large settlements, poor access to water, and inadequate shelter. The effect of these conditions varies according to the condition in which refugees arrive, and the hardships they faced in their home countries and in flight.

At the extreme, based on research conducted in the Wad Sherifei refugee camp, Kassala, Sudan, during 1989, De Waal ( 1990, 1989) comes to conclude that refugee camps in countries such as Sudan provide one of the most pathogenic environments imaginable. Common deficiencies in refugee camps are protein and vitamins A and C. Camps which restrict movement and economic and personal freedoms of the inhabitants in some cases encourage malnutrition arising from, as Wilson (1992) asserts, the principal shortcomings of the predominant current food distribution system. This includes insufficient quantity, erratic supply, insufficient micronutrient composition, and lack of variety and palatability ( WHO 1988) . Wilson concludes that much improvement in refugee nutritional needs could be obtained by removing constraints on the survival strategies often employed. These include the ability to trade food; make earnings through access to labour markets and/or the production of crafts and establishment of petty commerce; the gathering of wild foods; gardening; farming; or livestock-raising.

Overcrowding that occurs on camps as the result of massive refugee influxes or repeated regrouping is at the origin of many epidemic diseases (e.g. measles, cholera, dysentery, and meningitis) and avitaminoses occur mainly in camps where diseases such as beriberi, pellagra, and scurvy are still widespread (Van Damme 1995) . Data on refugee health has shown a clear correlation between camps of increasing size and elevated mortality (NOHA 1994, but see Spiegel et al. 2002) . Data collected by MSF on the 1991 pellagra epidemic in Malawi indicate in addition that camp populations, once infected, were more severely affected than self-settled refugees (Malfait 1991) , despite the fact that health services established for refugees often far exceed the quality of health services normally available for local populations.

More recently a study of mortality data in fifty-one "post-emergency phase camps" which focused on the associations between mortality and health indicators found mortality rates in more recently established camps to be higher than in longer-established ones. In addition, local health workers numbered fewer than in longer-standing camps. Water provision and rates of diarrhoea seemed to increase under-five mortality rates. Importantly, the study also shows an association between increased trauma morbidity in camps situated closer to the border or area of conflict than in those situated further away (Spiegel et al. 2002) .The study concentrated on post-emergency phase camps in which refugees received outside food aid and health care. The authors note, however, other, very similar mortality data from rather different situations, namely where refugees no longer receive such services, either because they have integrated locally or for other reasons. At the very least this seems to indicate that simple mortality rates may not be the most informative measure of refugee health care in a post-emergency setting.

Before refugee camps became the dominant mode of refugee assistance, an integrative approach to refugee health care seems to have dominated. Yet little is known about the comparative effectiveness of such programmes. Van Damme et al.'s study deals with a rare contemporary case in which refugee health care was integrated into the local health care system. In Guinea, the resources of the refugee-assistance programme not only served the refugees but also significantly improved the local health system and transport infrastructure. The authors conclude that "the non-directive refugee policy in Guinea ... may be a cost-effective alternative to camps" (Van Damme et al. 1998:1,609) . Moreover, as Toole and Waldman (1993) have argued, mortality might have been lower because many refugees were self-settled, "avoiding the problems associated with crowded and unsanitary camps".

[TOP]

Social aspects

Immediately following a large-scale refugee influx, camps provide life-saving services, most clearly in terms of health care and food but also by focusing attention on a crisis situation. Yet where the goals of refugee assistance in camps are defined by "minimum standards", "larger questions of needs and freedoms" (Jamal 2003:5) may be ignored. The wider social and socio-economic consequences of different types of settlement have increasingly been the focus of concerns by academics and practitioners alike. In operational terms they have tended to be put under the somewhat uneasy label (and frequently vaguely defined sector) "community services" (Bakewell 2003) .

Dependency and coping mechanisms

In Somalia, Waldron observed that "almost every functional prerequisite of society is defined radically differently in the refugee camp as compared with the self-sustaining, kinship-based rural communities of the Somali and Oromo refugees" (Waldron 1992) . Pushing this argument further, Ryle (1992) , in his observation of Somali Refugees in Ethiopian Camps, observes how "in compensation for the loss of skills as farmers and stockmen they have become skilled manipulators of the international welfare system".

Success of refugee assistance and protection, especially in protracted refugee situations, encompasses at least the facilitation of "functioning communities" and livelihoods. In this respect, two problems are often discussed in the debate about settlement patterns, that of dependency and the issue of "negative" and "positive" coping mechanisms.

The creation of passive dependency among refugees is often perceived as the real spectre of camps. In his well-documented State of the Art Review of Refugee Studies in Africa (1991) , Kibreab notes the "general consensus in the literature that prolonged residence in camps fosters “dependency syndrome” among refugees". From another angle, this has been echoed in arguments to move away from a provision of "minimum [emergency] standards" towards the broader notion of "basic needs" in protracted refugee situations. Both emphasize the need to expand the social and economic capacities of refugees in an assistance setting after the immediate emergency phase ( UNHCR 2000) .

This latter point is frequently taken up in the debate about "coping mechanisms", a term that seems to be used to refer to all and any ways in which refugees organize themselves to sustain their livelihoods. As noted above, restriction associated with camp settings may foreclose economic opportunities for refugees. They may also lead to so-called "negative coping mechanisms" such as prostitution or theft. One of the most obvious cases between "coping mechanisms" and the logic of emergency assistance is that of food aid. The (mainly illicit) attempts by refugees to acquire second or increased rations is a frequent problem for the equitable distribution of resources, not to speak of accounting issues. Similarly, agencies often see the sale and export of food aid as sign of excess when further study has frequently shown it to be a coping strategy to accommodate other material, cultural, or micronutrient needs that may come at a high cost to the energy content of their diet (Reed and Habicht 1998).

In a pointed reminder of the wider political problems that are part and parcel of the settlement debate, Malkki (1995) has studied the differences between Rwandan Hutu refugees that remained in refugee camps in Tanzania since 1972 and those who settled in an urban environment. She argues that there were substantial differences in a variety of social dynamics and wider quality-of-life indicators as a result of different types of settlement options. In addition, the first group of refugees, which remained isolated from their host community, created an imaginary and idealized image of the "Hutu Nation" which they both incarnated and would eventually return to, preserving, justifying, and reproducing the antagonism between Hutu and Tutsi in Rwanda. Even though Malkki's work has since been criticized, among other things, for a lack of focus on the political dynamics inside the camps and an underestimation of the linkages between both groups of refugees, it remains highly influential and one of the only studies of its kind.

Refugee women

When evaluating the broader social and economic effects of assistance patterns on refugees, long-standing concerns with "dependency" have since the 1990s been joined by the realization of the considerable implications that modes of assistance have for refugee women and children. These issues have moved to the forefront at least partially because, as noted above, a larger proportion of assisted refugees tend to be women and children (Crisp 2002) and because increasing attention to their conditions has revealed a number of glaring cases of abuse and exploitation in camp settings. The public discovery of widespread sexual exploitation in the refugee assistance programme in Guinea had the dubious honor of pushing these issues to the fore, and similar cases have been documented in Tanzania and elsewhere. While there is ample evidence that the incidence of sexual and gender-based violence is not confined to camps – a study on the situation of refugees in Congo Brazaville in 1999 noted 1,600 cases of rape reported between May and December 1999 from the hospitals of Brazzaville, highlighting "the high prevalence of sexual violence directed against women and girls during migration" (Legros and Brown 2001) – the perversity of a system of protection that undermines its very ambition has caused much debate within policy circles.

Moreover, while there is little documentation of the extent to which previously encountered gender conditions affect women's post-flight circumstances, it is broadly accepted that refugee women are highly vulnerable in camps, especially in regards to sexual exploitation. This is partly because family protection and traditional authority structures are less reliable, and new power-relations are created and sustained by the introduction of new rules and material relationships brought about by international relief. Even in camp situations where more participatory approaches have been tried, women tend to stay largely excluded from these supposedly democratic structures set up in ignorance of pre-existing social patterns. Hyndman's (1997) interviews in the Dadaab camps at the Kenyan border point to the correlated fact that, as a consequence, women often create their own community-based arrangements outside official circuits of (socio-economic) refugee participation and power-arrangements, which portray their own set of hierarchies and conflictive relations.

Again, it should be noted that the debate here is concerned not so much about the desirability of these circumstances, but about the appropriate means to change them. Maximalists will hold that a camp setting with its regulation and rules almost unavoidably creates the space for such "perverse effects" of assistance programmes. Others believe that camp-based assistance can be improved up to a satisfactory or tolerable level and its advantages be made to outweigh their negative sides.

Economic impact and development

The question of the economic impact of refugee populations on their hosts is deserving of a separate guide on its own, and it is very difficult to parse out the independent effect of settlement patterns in this respect. There is evidence that both camps and settlements have provided benefits as well as costs to their host countries. However, as Landau puts it: "… whether the aggregate effects on host populations and land are positive and negative … is next to impossible and would require an elaborate indices of gains and losses and considerable more longitudinal data than are typically available for the areas involved" (Landau 2003:3) . It is useful moreover to distinguish between short-term economic impact and long-term transformatory effect of the presence of both refugees and relief (Landau 2003) .

Camps, which generally restrict the exercise of economic activities much more than self- or planned settlement options, tend to benefit host countries primarily through the temporary capital influx that comes from relief agencies running the camps. Phillips (2003) argues that the direct and indirect impact of this financial impact has remained largely unexplored. Her study shows that one of the reasons for this is doubtlessly the difficulties in tracing both material input and impact. Where refugee assistance is camp-based, a smaller economic impact is also felt through those refugees who manage to circumvent the restrictions placed upon them and engage in trading or work in the surrounding communities.

As far as the overall costs of refugee programmes are concerned (which are, at least in cash terms, mainly carried by the "international community"), the biggest costs of camps probably lie in the large funds that are required for food aid. Proponents of self-settlement schemes hold that these costs far exceed the funds needed for a regional economic stimulus package in refugee-affected areas that would increase local absorption capacity as well as benefit the hosts. Self-settlement or more open planned settlement, the argument goes, allow for a more long-term developmental, multiplier effect on the local economy (Zetter 1995) . Some planned settlements have in the past been significant economic centres for surrounding villages, when they were integrated into a larger economic development strategy of the host country and when the economic potential of refugees was tapped into. Studies that show how self-settled refugees have positively impacted on sectors of the local economy range from the Afghan case to Zambia and Honduras, and are indicated in the bibliography below. Often a positive economic impact is only acknowledged after refugees leave an area. As Phillips (2003:16) notes: "While Afghan refugees were seen by many as a burden on the economy, their rapid repatriation from Pakistan, particularly from NWFP has caused a sharp downturn in the local economy, with many businesses recording severe losses and facing possible closure after the massive exodus." This is echoed in parts of Tanzania as well as in other refugee hosting regions. It indicates the way in which an accurate assessment of the refugee impact is frequently complicated by the political and economic stakes of the actors involved.

[TOP]

Environment

The impact of refugees on the environment and, inversely, that of different environmental conditions on refugees is part and parcel of any refugee situation. Refugees may exert additional pressures on environmental resources in a hosting area in a variety of ways, for example through poaching, deforestation (for fuel wood or purposes of farming), water use, and, when refugees own livestock, additional pollution and overuse of rangeland.

The specific effects of a refugee presence have also been seen as a function of the different types of settlement policies adopted. The environmental impact of camps is arguably more concentrated and therefore more easily amenable to policy intervention. Moreover, specialized agencies are frequently employed to limit environmental damage and set in place remedial measures. Yet it has been debated whether the impact of camps is therefore necessarily smaller or even more easily reversible (Jacobsen 1997) .

[TOP]