Social and Ritual Activity In and Out of Place: the ‘Negotiation of Locality’ in a Sudanese Refugee Settlement1

1. Appadurai in Turton, 2004.

Abstract

This article argues that peoples' affective relationships with the specific physical territories that they inhabit are informed by and constructive of the social relations and practices which are enacted in them. When people are forced to leave their homes, the ways in which they engage with their physical, socio-cultural, political and spiritual landscapes are necessarily transformed. Based on ethnographic research with a group of long term Sudanese refugees in Uganda, the article shows how challenges to socio-cultural, ritual and political identities and activities are just as great as the more tangible challenges to protection and subsistence for refugees.

The article examines a number of key socio-cultural activities including funeral rituals and agricultural practices, exploring the extent and ways in which ’place making‚ in exile involves the active mediation of external factors at a several levels as well as processes of compromise and substitution with respect both to material culture unavailable in the settlement, and also with in relation to social relations and practice.

Keywords:

• refugees
• place
• socio‐ritual
• agency
• material culture

Introduction

In all, 2.7 million of the world's 9.5 million‐strong refugee population are to be found in sub‐Saharan Africa (Crisp, 2006). Of these, increasing numbers are trapped in protracted exile in countries next to or close to their countries of origin, unable either to return home or to find any alternative durable solution elsewhere.

The reluctance of EU states to host refugees from sub‐Saharan Africa has not gone unnoticed by African host states, many of whom have learned restrictive asylum practices from northern countries and are applying these with various negative consequences for refugee protection. The absence of meaningful burden‐sharing in the south certainly remains one of the main drivers of this trend. The most obvious consequences of the pressures on the asylum system in Africa are an increasing reliance on the encampment of refugees, leading to populations experiencing protracted exile contained in frequently remote, inhospitable and insecure locations. The absence of an adequate supportive response from states of the international donor community means that basic needs can remain unmet while the rhetoric of developmental approaches remains just that. Within this context, in the 1990s there was increasing pressure on the principle of voluntary repatriation which saw numerous refugee populations obliged to return home under far from ideal conditions (Chimni, 1999). These material, legal and political deprivations, however, do not constitute all of the challenges with which refugees are faced while in exile.

Forced Migration, Landscape and Culture

When people become refugees, they suffer the loss not only of the land, property, assets and material possessions that they have been unable to carry with them when they flee, but also of the less tangible networks, relationships, socio‐cultural practices and identities associated with the social and physical landscapes they are obliged to abandon (Davis, 1992; Parkin, 1999). They thus also lose the benefits of a broad familiarity and intimacy with places previously occupied and known as ‘home’, and the ways in which material objects and other cultural forms were formerly used, managed, negotiated, transacted and understood in home locations. These intangible losses speak to the specific social networks, relationships, authority structures and institutions through which people previously engaged with each other and with their environment on a day‐to‐day basis (Kaiser, 2006b). As other writers have shown us, material and non‐material losses of different kinds are inseparable from each other in the sense that they are experienced relationally (Davis, 1992; Loizos, 1981).

Despite these losses and the structural limitations of existing in what is often a hostile asylum setting, refugee groups have no choice but to assert themselves and respond to their new environments. Such a response is made not only in terms of their legal, material and subsistence statuses, but also in relation to their individual and collective subjectivities, identities and all aspects of their existential experience. In ‘refugee studies’, as well as in broader interdisciplinary studies of refugees, the analysis of the former has been largely privileged in comparison to the latter. A conviction that both elements are crucial and should be understood in relation to each other, has led to the research approach adopted here, in which both have been carefully addressed (see also Turton, 2005, p. 276).2

Forced migration has sometimes been characterized as an opportunity for expedited social change in the sense that for those who survive flight it provides a fresh backdrop against which individuals and sub‐groups within a population can seek to assert themselves, effecting or exploiting social and political change where this is a possibility. But what, precisely, has changed for refugees who have crossed an international border? At the most obvious and literal level, refugees' physical location is different in the pre‐ and post‐flight period and the shift from a familiar, ‘home’ territory to a ‘foreign’ place of exile has been much debated as a source of challenges to identity, livelihood, the enjoyment of human rights and other important aspects of life (Malkki, 1995; Kibreab, 1999).

‘Place’ has thus been rightly recognized as a crucial analytical and explanatory factor in relation to experiences of forced migration. This paper sets out to show that place should be understood as a multi‐dimensional category with meaning and significance located in its non‐material as well as material aspects. Peoples' affective relationships with the specific physical territories that they inhabit are informed by and constructive of the social relations and practices which are enacted in them. Also important is that we consider the wider political economies in which refugee lives and regimes are embedded, recognizing that refugee environments, far from being discrete and closed, have exceedingly blurred boundaries and are immediately connected to the wider world (Appadurai,1996; Feld & Basso, 1996; Novak, 2007; Turton, 2005). This dimension, while crucial, remains to be explored in further work in this project and is not covered in detail in this paper.

One of the main interests of this paper in empirical terms is to address the varied and complex links between interlocking refugee locations and the diverse actions and socio‐cultural activities they host, facilitate, enable and generate. It is evident from research findings that with respect to some socio‐cultural activities, transformed landscapes and their associated socio‐political configurations of exile have dramatically affected what refugees consider it is possible or desirable to do in them. In relation to other activities or practices, however, it has been relatively simple to find an accommodation or compromise version which it has been possible to perform in exile. This paper examines a number of key socio‐cultural or ritual activities performed and neglected in the Kiryandongo Refugee Settlement in Uganda, seeking to understand what variables are motivating diverse responses and strategies on the part of the refugee population and the extent to which changing ‘place’ and the transformed ‘landscapes’ which they inhabit has been influential to this process.

Whether or not important social and cultural activities are enacted in the refugee setting must evidently be determined by a range of factors. Noteworthy and complicating variables are questions relating to the temporariness of exile, the liminal nature of their refugee existence and the uncertainty they face, manage and contribute to by their responses to exile. Refugees variously mitigate, manage and exploit such temporariness and uncertainty in ways that will be discussed in relation to the ethnographic case study discussed below.

We begin with a vision of cultural activity – material and non‐material – that emphasises its embeddedness in and constructiveness of the everyday experience and action of all dimensions of life: social, economic, political and spiritual. By definition, the reproduction of culture is an active, dynamic process in which culture is continually re‐invented and is amenable to transformation. Change is consequently part of the system. This paper raises questions about the ways in which people displaced by conflict draw on a range of resources, including internal cultural resources, to help them to comprehend and manage difficult circumstances and to protect and transform those same important social relations and practices.

Following Richards (2004), Davis (1992) and Benoist & Voutira (1994), this paper understands conflict and forced migration as products of and continuous with ‘normal’ social relations in the sense that they are generated by social and political relations that are amenable to understanding and analysis. Peoples' responses to these challenging phenomena are no less so. Conflict and exile are also ‘cultured’ processes – the shape they take, their meaning and the impact they have need to be considered in relation to socio‐cultural specificity.

This paper continues with a brief introduction to the Ugandan refugee situation before identifying some of the main themes in the theoretical literature which have informed thinking on this topic. The majority of the remaining space is devoted to a discussion of the ethnographic material and focuses largely on non‐material socio‐cultural forms while making the case that these are practically and analytically amenable to analysis in similar ways as material culture.

Uganda's Refugees and the Kiryandongo Refugee Settlement

Ugandans have a long history of both hosting Sudanese refugees and of seeking refuge in Sudan themselves. Large numbers of Sudanese spent much of the 1960s in exile in Uganda while many northern Ugandans, in turn, fled their country after the fall of President Idi Amin in 1979. The second Sudanese civil war which began in 1983 led to a further flight to Uganda both by Ugandans and Sudanese in the 1980s. In 2006, many of those Sudanese refugees still lived in Uganda – approximately 200,000 were formally registered by the government and lived in UNHCR‐run settlements while an unknown number had settled independently outside them. The Ugandan security context has been hostile for refugees during the greater part of their exile as insurgencies led by the West Nile Bank Front (WNBF), Lord's Resistance Army (LRA) and other groups have endangered them as well as local populations in refugee hosting areas of northern Uganda.

In January 2005, the Government of Sudan and the main Sudanese rebel movement, the Sudanese Peoples' Liberation Army (SPLA), signed a ‘Comprehensive Peace Agreement’, bringing to an end over two decades of conflict. After an early reluctance to risk further challenges, many Sudanese have now repatriated on a voluntary basis. Approximately 100,000 refugees remain in Uganda. Renewed activity in this direction was partly made possible by some improvement in security in South Sudan occasioned by progress in peace talks between the Government of Uganda and rebels of the LRA who had previously held bases in South Sudan (ICG, 2007).

Since their arrival in Uganda from the late 1980s onwards, refugees have been registered by the state and required to live in government‐defined refugee camps and settlements. These have in most cases been located in the underdeveloped northern part of Uganda, in many cases within 50 kilometres of the international border with Sudan. The Ugandan state and its organs are very present in these refugee locations.

This paper relies most heavily on ethnographic research carried out with a predominantly Acholi population of Sudanese refugees in the Kiryandongo Refugee Settlement in Uganda's Masindi District. Initial year‐long doctoral research in the mid‐1990s was followed by several shorter research trips between 2002 and 2008. The material on which this particular paper is based was gathered both in the course of the 1996–97 research and in the context of a more focused research visit during January–February 2006.

Situated south of the Victoria Nile in a maize‐growing area, Kiryandongo is relatively privileged in that it has never been directly affected by rebel attacks and its inhabitants have benefited from reasonably good‐quality land. These refugees largely fled from the border areas of Sudan to Uganda in 1989 and were first accommodated in a transit camp in Kitgum district, before being relocated in 1990, after being attacked by the LRA. At Kiryandongo, each refugee family was allocated a plot of land and the majority now live on the crops which they are able to grow for sales and consumption, food rations having been withdrawn for the majority some years ago. Although the Acholi group dominate numerically, several other ethnic and linguistic groupings – including Madi, Latuko, Bari speakers, Zande – are to be found in the settlement which is home to people who have arrived from different places at different times. Nevertheless, after such a protracted exile, a significant community feeling has been generated despite some significant episodes of conflict within the population. The settlement is fairly well established with links into the local host population, with people who remained in Sudan, and with those members who have been resettled to countries as far afield as the USA, Australia, the UK and Scandinavia. It should be noted that moving to Uganda has for most of the refugees represented a move from a much less developed environment to a more developed one and involvement with the international humanitarian aid regime and the organs of the Ugandan state have also enlarged the experience of many of the refugees who previously lived fairly quiet, rural lives. Some of the leading Christian churches with whom refugees were already associated in Sudan have either moved with them to Uganda or are already present there and their influence on the refugees' lives continues to be great in many cases.

Place and Landscape – Situating the Research

‘The supposed isomorphism of space, place and culture’ (Gupta & Ferguson, 1992, p. 7) has been well and truly shattered. As Gupta and Ferguson themselves note, it fails to deal with the challenges posed by people living in borderlands, with cultural differences within a given locality, with post‐colonial cultures and with the power relations inherent in processes of social change (ibid., pp. 7–8).

Nevertheless, ‘place‐making’ as a social and political project remains a meaningful and significant activity (Appadurai, 1996; Turton, 2005; Hammond, 2004). Turton cautions that ‘we must treat place, not as a stage for social activity, but as a “product” of it’ (2005, p. 258). There is frequent emphasis in the anthropological literature on the fact that place‐making activities are explicitly processes of production; something new is generated in terms of relationships, including power relations, and these may be influenced by the physical context or environment itself but also by a multitude of other internal and external factors (see also Novak, 2007). A picture emerges of an iterative relationship between place and the activity which occurs in it; each is affected and in some sense made possible by the other (Bender, 1993; Cresswell, 2004; Hirsch & O'Hanlon, 1995).

In relation to Sudanese refugee place‐making activities in the Kiryandongo Refugee Settlement in Uganda, influential factors include the nature of the physical territory and its use by local populations, national refugee policy and its local implementation, the role of UNHCR and aid actors, refugee host relations, internal competition and collaboration within and between refugee groups and so on. Appadurai argues that ‘locality itself is a historical product and…the histories through which localities emerge are eventually subject to the dynamics of the global’ (1996, p. 18). At the micro level, locality is also a socio‐culturally‐specific product; meaningful contexts are produced from the conjunction of certain social practices in particular physical environments. The process is far from static, however; Ucko and Layton draw on Bourdieu and Giddens to make the point that ‘meaning is established and changed through usage’ (1999, p. 7). Physical territories may be capable of being ‘made’ and unmade' as sacred spaces, for example, because of the socio‐ritual activities enacted there.

Turton points out that organization and investment is required for the ‘work’ of ‘place‐making’, as he calls, it, to occur (Turton, 2005, pp. 267–268). He also emphasises the extent to which the world's disempowered peoples, including many refugees, are limited in the influence that they can bring to bear on the production of their localities, in comparison with the overbearing influences of states and other powerful actors (ibid., p. 286). The Mursi, for example, are ‘dependent, in fact and in aspiration, on things and ideas, the production and distribution of which are beyond their reach and control’ (ibid., p. 275).

Similarly, the ‘context‐producing activities’ (Appadurai, 1996, cited in Turton, 2005, p. 271) of the ‘nation‐state’ have a clear impact on the Kiryandongo refugees' capacity – imaginatively or literally – to assert control over their own lives. They are, of course, unable to choose where they should live in Uganda and are highly restricted in what kinds of livelihood activities they can pursue under the Uganda Government's refugee policy. They are restricted not only by the attitude towards them adopted by two states – their own and the host state – but are also heavily constrained by the activities of rebel movements in both Sudan and Uganda. Their disempowerment in this respect is thus extreme. Discourses of place are competitively generated – with specific political and policy objectives – by other actors such as the Government of Uganda and UNHCR.

Rodman's work on multilocality provides a reminder that contestation and variation should be expected within and between place‐makers; ‘a single physical landscape can be multilocal in the sense that it shapes and expresses polysemic meanings of place for different users’ (1991, p. 647). There is no good reason why we should expect consensus in an activity that defines the meaningful world. In the Kiryandongo case, it is evident that refugees' cultural construction of their environment is not limited to the physical world. It refers to their occupation and use of a multi‐faceted world with physical, social, spiritual and aesthetic dimensions in the here and now (in Uganda), but also to their ‘lost’ lands (in Sudan). The implication is that they more ‘fully’ inhabited the Sudanese territory because they knew it in a deeper and more intimate way than they now know the exilic landscape.

This paper relies on an analytical conceptualization of ‘landscape’ rather than ‘place’ in order to emphasize the diverse ways in which environment and action intersect as refugees attempt to re‐build their lives in unfamiliar places. The critical difference between the categories ‘place’ and ‘landscape’ is that the latter is understood to incorporate and include the socio‐ritual, symbolic, spiritual, political, economic and aspirational dimensions of the ‘places’ inhabited by refugees, and does not restrict itself to a narrow, territorially bound understanding of either the point of origin or the exilic location. Unlike some historical constructions of landscape as a mere backdrop to social action (for elaboration see Cresswell, 2004), this conceptualization of it, following the recent work of anthropologists such as Bender, highlights the fact that ‘engagement with landscape and time is historically particular, imbricated in social relations and deeply political’ (Bender, 2002, S104). There is an intense and subtle relationship between action and context; ‘the contours of the interacted‐with landscape – the materiality of social relationships – are dynamic. Human interventions are done not so much to the landscape as with the landscape, and what is done affects what can be done’ (ibid., S104). One advantage of the use of the notion of ‘landscape’ is that it again provides an opportunity to recognize the fluidity and interplay of the structural conditions or constraints of peoples' lives, and the space for the assertion of individual and collective agency.

This leads us to the wider point that when people are forced to leave their homes, the ways in which they have understood, interpreted, constructed and represented the landscapes in which they have lived – physical, socio‐cultural, political, spiritual – to themselves and to others is interrupted. While their culture is also challenged and undermined by the losses consequent to displacement, and efforts have to be made to re‐assert, replace or accommodate it in the new environment, it is significant that this has to be done in new landscapes, which are not known in the same way as the original ones were. Instead of asking in a general way how culture has been affected by displacement, or how changing physical places has had an impact on individuals and groups, we might rather ask in a more systematic way, how changes and challenges to the material, physical, socio‐cultural, political and spiritual landscapes have been negotiated and managed in the context of exile. This approach invites consideration both of the dynamic nature of inhabited landscapes as well as the ways in which experiences of different landscapes interrelate with each other. It should also be noted that this experience may be expected to be different for members of different sub‐groups within a differentiated population. The materiality of occupied landscapes remains constant, but locations change, and associated dimensions of social activities respond accordingly.

Relations Between Place, Landscape, Culture and Social Action – Kiryandongo Refugee Settlement

The discussion above implies strongly that it would be unproductive to seek to demonstrate the reproduction of ‘culture’ for refugees in exile in any linear, literal and exclusive way. One might rather think in terms of more fluid and open categories – Ferguson's concept of ‘cultural style’, perhaps, which searches for a logic of surface practices without necessarily mapping such practices onto a “total way of life” encompassing values, beliefs, attitudes, et cetera, as in the usual concept of culture' (Gupta & Ferguson, 1992, p. 19), or James's ‘archive of moral knowledge’ (James, 1988) which suggests a position based on non‐exclusive shared experience and understanding. While explicitly socio‐cultural activity is known to often increase in the immediate post‐flight period as individuals and groups take solace in the familiar and assert their cultural identities in the face of the evident challenges to them that flight has represented, it is equally clear that as time passes and situations change, something much more complicated than a straightforward reproduction or replication of a discrete and homogenous culture is being enacted. Rather, groups in new configurations may be in a position to engage in a much more dynamic process whereby they create a new socio‐cultural configuration that remembers their past, reflects their current situation, meets some of their needs, and protects what they aspire to for the future.

In what ways and to what extent does a new location determine what can happen there in these terms? There may be practical limitations but, as we shall see, the assumptions, beliefs and expectations of refugee groups will also define the freedom with which they operate within specific topographic and socio‐political landscapes. This section of the paper asks, what is the nature of the link between landscape, identity and socio‐cultural action in this case? It does so via an examination of some forms of social and ritual actions observed in Kiryandongo and by asking specifically, why and how are some socio‐ritual activities sustained and developed in exile, while others have been reduced, transformed or allowed to go into abeyance? It also considers how the physical location and the fact of exile play into these processes. Some of the literature on what ‘home’ means to the displaced and non‐displaced emphasizes that home is a location where one feels in control, partly because one is able to carry out familiar and socially important activities and practices (see e.g. Terkeneli, 1995) while other commentators are more sceptical, arguing that the concept is more ambivalent (Rapport & Dawson, 1998). In many cases, refugees in exile explicitly seek to reproduce familiar social forms, activities and relationships as a means of coping with the challenges of displacement (e.g. Dudley, 2000).

It is, of course, very common that refugees should find themselves in situations where, for a number of reasons, they are unable to carry out the full range of socio‐ritual activities and practices which have previously been considered indispensable. Nevertheless, evidence from a range of situations suggests that accommodations are frequently made, with a reduced number and kind of such activities being maintained as the minimum that people can expect to manage with and retain their sense of what it is to be who they are (e.g. Gow, 2002; see also Davis, 1992). In her discussion of strategies employed by Ethiopian returnees, for example, Hammond shows how social norms and conventions can be deliberately subverted or avoided by returnees so that social rules may be maintained even when they cannot be fulfilled (2004).

Although Acholi ancient history is a narrative of onward movement and the search for a suitable territory on which to settle, since the pre‐colonial period migration has in general been limited to small‐scale movement within recognizable Acholi sub regions‐which are themselves associated with pre‐colonial chieftainships. In these sub‐regions, now formalized by recognition as administrative units by the state, some land stewardship functions have been conflated with the activities of formal governance at the local level. Internal borders do exist and have social currency but what is significant to this account is that there is a clear sense that the land which has been temporarily lost is clan or family land on which clan members used to live communally to some extent. It should be noted, of course, that very many Sudanese Acholi were displaced during the first Sudanese civil war during the 1960s and that for some of the Kiryandongo refugees this is the second period of extended exile. On this occasion, as previously, some refugee families have left representatives behind in Sudan with the explicit intention of preserving their tenure on land held under customary law. In some cases, refugees based in Kiryandongo have travelled back to the home area in Sudan in order to bring land under cultivation to demonstrate their continued claim to unoccupied family plots.

Links Between Material and Non‐material Culture

Parkin (1999) places emphasis on the difficult and time‐pressured choices that fleeing refugees are forced to make about what material objects they are to carry with them when they leave their homes. In practice, many refugees – including the Kiryandongo refugees – have little time to collect more than their children and whatever is closest to hand when they leave home.3 In 2006, some 17 years after the initial flight from Sudan, most of the Kiryandongo refugee population no longer focus on the material objects they were able to bring with them or were forced to leave behind. However, some consequences of choices made do exist, not least in very practical terms. Many refugees have struggled endlessly, for example, with attempts to carry on with education, training or professional activity in the absence of documentation proving their level of qualification or training where this was left behind. Interestingly, many people do still keep much thumbed photograph albums that they have been able to carry and preserve, and these are still treated as important repositories which provoke and facilitate many reminiscences of life before flight. Some important cultural objects have been lost to the population in exile, and these are missed although their loss has by now merged with other factors to make them seem quite far removed from the refugees' everyday life. Such objects include rain‐making stones of the royal clans, some sacred drums and other important musical instruments, and robes and costumes associated with ritual activity.

A crucial factor here is that we recognize the actual and analytic continuities between different forms of socio‐cultural objects and action, and their aesthetic and material dimensions. Refugees' responses and social coping strategies may take material and aesthetic form in directly artistic, ritual or productive activity. This paper hypothesizes that such responses and activities are amenable to analysis in the same terms as wider processes of social creativity – for example the preservation or re‐configuration of social relations and practices – and makes no artificial separation between these related forms (Davis, 1994). As Parkin notes, ‘people in flight store, so to speak, their precluded social personhood within mementoes of mind and matter, including cherished small objects, songs, dances and rituals, which can, under favourable circumstances, be re‐articulated (even recreated) as the bases of social activity’ (1999, p. 315). I have elsewhere written about the importance of music and dance in Kiryandongo, in particular considering the way in which values and social norms are transmitted and transformed via these media (Kaiser, 2006a). Here, I want to emphasize that the production and use of material objects, aesthetic or artistic activities and social action can each be conceptualized by the Kiryandongo population as meaningful activities and suggest that in a materially impoverished community it is unsurprising that the latter may appear to be privileged over the former. The following sections therefore look in more detail at three kinds of important socio‐cultural or ritual practices in Kiryandongo, and consider their role and impact in terms of interaction with the socio‐physical landscapes of the settlement.

Third‐Stage Funeral Rituals

The burial of the bodies of the deceased is one of the first and most obviously pressing problems faced by refugees trying to manage life on foreign soil. For the majority Acholi group in Kiryandongo, the dilemmas which arose when this situation first presented itself were common ones. When at home in the Acholi area of Sudan, people had typically buried family members on the family compound, meanwhile maintaining a shrine (abila) to their family ancestors nearby. When the first deaths – especially of ritual leaders and other key community members – took place in exile, it was a shock to be obliged to bury them on land which families hoped to leave at the earliest opportunity. As the years passed in Kiryandongo, people became more and more used to burying their dead in the settlement, with more prosperous families going so far as to concrete over the top of the grave, a practice copied from Sudan. There have been notable exceptions in cases where, because of the ritual significance of deceased individuals, great efforts have been made to transport the body back to Sudan so that burial can take place on home ground and where shrines are expected to be maintained in perpetuity. Interestingly, for some refugees, as the time spent in Kiryandongo has lengthened, it has itself become an acceptable compromise burial location for those who have died elsewhere and are unable to manage a return to Sudan. One refugee who died in 2005 in Nairobi, for example, was carried back to Kiryandongo for burial, indicating that, if it has not become ‘home’, it has nevertheless gained some status in this respect.

For the Acholi population of Kiryandongo and as would also have been the case in the Acholi area of Sudan, the funeral rites or lyel which take place immediately after death are mainly limited to family members and members of the immediate community who gather to pay their respects. Burial normally takes place quickly after death and there is neither the time nor the inclination to organize large‐scale social activity at this time. Burial practices and their ritual accompaniments are carried out without fuss at the homestead of the deceased, and the burial itself nowadays takes place either there or in one the settlement's cemeteries. Approximately one year after the bereavement, however, the family of the deceased hold a much bigger and more public event; a ‘third‐stage funeral rite’. Perhaps unexpectedly, it is this event which has noticeably developed and been extended in Kiryandongo, despite the constraints of poverty, difficulty of communication, and the fact that burial has usually taken place away from the ancestral land in Sudan to which people insist they are still so attached.

The 24‐hour event includes a number of components, some of which have been newly included during the period of exile. The proceedings open usually on the allocated refugee plot of the deceased with prayers – often both a Christian service and indigenous practice – and include a number of speeches about the life and achievements of the deceased before a great feast is served to all attendees. This is accompanied by the consumption of quantities of locally brewed millet beer which has been contributed by and is redistributed to members of clans and other groups in any way connected to the deceased. The event closes with music and dancing which generally continues all night.

Family members may travel from far away to attend such occasions, which are typically held in the dry season for ease of access. It is not uncommon to have people arriving in the settlement from Kampala, Nairobi, other Ugandan settlements, Europe or the USA to attend the funeral ritual of a respected family member and most people agree that these events have become increasingly lavish, even excessive, with expenses mounting to as much as 1–2 million Uganda shillings (hundreds of pounds) in a cash‐poor environment.

It is in some respects surprising that while other comparable ritual markers of life‐cycle stages such as marriage ceremonials are agreed to have been severely curtailed during the period of exile due to poverty and the dispersal of family members, such great efforts have been made to sustain and develop a ritual activity which, more than most, would seem to be tied, at least ideationally, to a specific home territory due to the continued importance of ancestral land.In fact, the evidence is that in the Sudanese Acholi area itself during the period of the conflict, funeral rituals have continued where possible, very much as before. Simpler and less extravagant than the Kiryandongo events, they do not include – as the settlement events very noticeably do – amplified Western pop music. Rather, they reportedly rely on the culturally significant and appropriate dances of otole and bwole. Although these music and dance forms are also found in funerals in Kiryandongo, they face competition from loud disco music which tends to go on through the night, long after the performance of ‘traditional’ dancing has finished. In Juba, the capital of South Sudan in which many of Sudan's internally displaced people have congregated, comparable Acholi funeral rituals are even simpler, with some prayers and possibly some food reportedly sufficing as the centrepoint of activities. In Kiryandongo, apparently, the ritual has clearly expanded and developed, regardless or possibly even because of exile.

Agricultural Practices

The question of land use in exile is a complex and multi‐faceted one. When it comes to agricultural activity in the settlement context, the need for a multidimensional concept of landscape may become more apparent. It is far from being the case that refugees' agricultural capacity is simply a practical matter of the amount of good‐quality land to which they have access, labour and other material inputs. On the contrary, without an additional understanding of the socio‐cultural and spiritual associations people have with land, agricultural practices in the Kiryandongo settlement and peoples' attitudes to them may remain rather obscure.

In the first instance, there are a series of challenges associated with farming under the conditions found in the Kiryandongo settlement. Farmers in the part of Sudan from which the refugees come typically had access to several areas of land in different environmental zones, enabling them to exploit different kinds of crops in different gardens and to practice shifting cultivation in order to avoid soil exhaustion and erosion. In the settlement, on the other hand, each family was on arrival allocated a single plot of land which, in most cases, was not big enough to allow for shifting cultivation. As time has passed, via networks of social relations and local arrangements, farmers have where possible contrived to broaden their land holdings by borrowing or renting additional plots dispersed both inside and outside the settlement in order to be able to produce a range of crops and spread risk in case of drought or other environmental problems.

Clan‐based agricultural work groups, on which people had previously relied for heavy tasks such as land clearance (see Allen, 1987), have necessarily been transformed as the dispersal of family groups in flight and resettlement has meant that extended families are no longer usually able to live together and co‐operate in subsistence activities as they once did in Sudan. Exacerbating this fact, plots in the settlement have been allocated to nuclear families without reference to clan or sub clan membership, thus making the communal living on extended family compounds, which was the norm in Sudan, an impossibility in the settlement, even when entire family groups were present there. Agricultural work groups (oroma) have been adapted by refugees so that they may now consist of any convenient grouping of people; neighbours, church group members, student groups and so on. In effect, the social form of the work group has been preserved but the kinds or categories of people who inhabit it have been allowed to change to meet current needs.

Farmers, especially from the older generations, point to the differences of farming clan land which is intimately known and ‘owned’ at home, and making a living on land which has been allocated to them in the settlement. The issues are not practical ones: local farmers have, for example, been generous in sharing local knowledge about weather patterns, preferable crops and so on, but older refugee farmers feel strongly the absence of control over the spiritual forces which control agricultural land. In Sudan, each clan and family member was intimately familiar with the various territorial and ancestral spirits which were associated with specific people, places and activities in and around clan land. In refugee accounts of life and social practice in Sudan, no distinction is made between the world of the spirits and the land occupied by them and Acholi people. The implication is that they are naturally paired together and it is the case that with the exception of some domestic family spirits, they are not assumed to have left Sudan with the refugees. Some of these spirits are believed to have provided in the Sudanese Acholi area for success in subsistence activities including farming and hunting so long as the relevant and appropriate ritual activities were carried out by elders on and for the land. Spirits were understood to reside in specific places – often near water – on Sudanese Acholi territory and were inseparable from it. Among the Sudanese Acholi, the won or father/owner of the land was bound in a close and important relationship with such spirits who he was able to propitiate by carrying out rituals as required. This familiarity and facility is perceived to have been lost in Uganda. The refugees understand that the land they currently farm in Kiryandongo is also occupied by spirits of a similar kind, but they profess themselves to be ignorant of them and unable, therefore, to control either them or the land adequately.

In this sense, moving from their own physical territory and a familiar spiritual landscape even to an ostensibly welcoming foreign land, has had disproportionately great consequences for their sense of their own capacity to farm successfully. The refugees have little sense that they have taken ownership or control of the land in any other than a literal way. They cultivate it, but they do not claim or pretend to have a more profound engagement with it. Indeed, they positively reject this as a possibility, insisting that it is not in their power to acquire the spiritual knowledge that they currently lack and asserting that they do not know the true owners of the land and cannot, therefore, ask them for help in this respect. One group of women living in the settlement insisted that although they had been warned by local people on arrival that the land was occupied by jok or malign spirits, they had no choice under the Government of Uganda's regulations but to settle where they were told, even if this meant that the spirits would kill them there.

The inability of the elders to know and control the spirits and thus the land which they have been allocated to farm by the government and UNHCR also challenges their status within the refugee community itself. Their inability to continue to do in exile what had been one of their prime responsibilities in Sudan has clear implications for the maintenance of familiar patterns of authority and leadership structures. Whether this can be recouped later remains an open question.

Initiation Ceremonies (orwanga)

My final, brief example relates to a form of ritual activity that has never taken place in the settlement, since ritual elders who control such matters insist that it can only take place on ancestral ground in Sudan. Although some members of the Acholi, Latuko and Madi groups all practise male initiation ceremonies in Sudan, none has ever been held by any group in the settlement. While not all Acholi sub‐groups in Sudan continued to practice initiation rituals – at which boys were inducted as adult men, age sets formed and the social hierarchy established – it was still in existence in the pre‐flight period in some Acholi areas (including those of the Acholi sub‐groups of Parajok and Obbo), as well as in Latuko areas.

For the Acholi, holding an initiation ritual in Uganda is understood to be impossible for two main reasons. In the first instance the ceremony is meaningful in relation to groupings of people who are associated with specific social and geographical categories. For the people of the Acholi sub‐group of Parajok, for example, holding an initiation ritual outside that physical location is considered highly problematic. This is partly because the ritual has become associated over time with that particular territory, and partly because it requires all initiands to undergo the ritual together and in the same place, by implication in the home area. This has become impossible given the displacement of large numbers of the Acholi population. It has not been possible to reach agreement between the displaced and non‐displaced Acholi, as to the possibility of holding an initiation ritual in the home area in the absence of the displaced youths, or exclusively for them in the settlement in Uganda. A further complication in relation to the latter option lies in the fact that members of respective Acholi sub groups are all ‘mixed up’ in Kiryandongo and organization of the ritual on the basis of these identities is crucial to the purpose of the ritual which is very much to associate young men with their territory and sub group. The ritual includes various competitions of athleticism and fearlessness which allow for the appointment of an age group leader from one area which gains enormous prestige from this fact. It is clear that a social context such as a refugee settlement, with people settled without reference to social identity on foreign ground is not conducive to such an event.

In summary, it is clear that the Kiryandongo refugees feel that holding initiation rituals in exile is impossible, so intimately is the process associated with the physical and symbolic landscapes of home. Interestingly, neither has the ceremony been held there in their absence, indicating that these alone are not sufficient to ensure that it goes ahead; a critical mass of people is also clearly required. In any event, it is also the case that during most of the period of the refugees' exile, those who remained in the Sudanese Acholi area have struggled to survive in a conflict‐affected area in which insecurity, poverty and uncertainty have also combined to make large‐scale ritual activity difficult if not impossible. As Gupta & Ferguson note (citing Bhabha, 1989, p. 66): ‘it is not only the displaced who experience a displacement’ (Gupta & Ferguson, 1992, p. 10). No initiation has been held in the Sudanese Acholi area since 1986 and refugee elders are confident that as soon as a return to Sudan becomes possible, youths will begin to agitate to be ‘given a name’ as an age group, and arrangements will be made.

In the meantime, there are social costs and changes which have resulted from the impossibility of holding initiation and the consequent absence of discrete and defined age groups of men. Entitlement to the married state is no longer as it once was, necessarily constrained by having gone through initiation and membership of a particular age group. Since marriage was once the principal way of asserting adulthood, this makes the position of married but uninitiated men uncertain in the present context. James, a teacher, commented that he still thinks of men of his age as age mates, regardless of whether the initiation ceremony has taken place and this may be common.

In another context, for Parajok agricultural work or digging groups, membership of a particular age group used to define the kind of work that an individual would do. Members of the Ogwal age group cut trees, Bonyo dig, Rucu, the youngest, act as gofers. When digging groups have been redefined so that they may comprise men (and even women) all of the same age, and when the differentiation of formal age divisions is no longer put into practice, the ambiguity of each group's position is further complicated. The sometimes tense relationship between the elders and the youth, evidenced particularly by the elders' rejection of changing systems of marriage arrangements may also be complicated by the absence of the formal structure of age groups within which every man once knew his relative position. The authority of the elders may have been compromised by the awareness of the youth that as things stand, they will never be ‘promoted’ to positions of political prestige and respect unless they develop new ways of doing so. This in turn reduces the power and authority of the elders, who might be expected to resist it.

One of the most important fora in which uninitiated men's authority is tested or highlighted in the refugee setting lies in the fact that they are not considered entitled to participate in public discussions and decision making as fully fledged members of the adult community. Neither can they stand as community leaders, although this does not prevent them from holding government office. In 2006 the Chairman of the Kiryandongo Refugee Welfare Council was a young, uninitiated man. This posed no problems as far as ritual elders were concerned, they were happy to defer to him on bureaucratic matters, but he had no personal authority outside of this.

Compromise and Conflict: Substitution

With respect to some social activities, institutions and relationships, exile, poverty and uncertainty have combined to make their full performance effectively impossible in the context of exile. Instead, significant compromises or accommodations have been made in Kiryandongo so that in some cases even when a full event could not be held, a smaller or less expensive version was held to symbolize what should have taken place. Notably, there seems to have been no tendency to simply abandon the attempt to fulfil familiar social and cultural roles and activities.

In many of the socio‐cultural activities and rituals still performed in the settlement, not everyone and everything properly associated with the practice is available in the settlement. In these cases, refugees have been adept at accommodating absences and substituting people and objects where this is necessary for the ritual to go ahead. In the case of the funeral rituals discussed above, for example, elders' dance costumes are made up of a number of materials, some of which only ‘stand for’ others which are unavailable in the settlement. For example, black plastic bags stretched over a wooden frame substitute for the splendid ostrich feather headdresses once worn in Sudan, and strips of white paper flutter in imitation of white feathers. Importantly, elders wearing these compromised costumes are treated with as much respect and accorded as much dignity as if they were indeed wearing their valued headdresses of old (Figure 1).

Figure 1 Sudanese Acholi elders wearing ceremonial headdresses made of wooden frame and plastic carrier bags dance at the funeral of an important ritual leader, taken by the author at the Kiryandongo Refugee Settlement, Uganda October 1997.

Not everything is replaceable, however. More prosaically, refugee women in the settlement, unable to acquire the clay or grasses required for them to make pots and baskets to their own design and specification, resort to buying what they consider ‘less beautiful’ examples of the finished product from local markets. While gathering such materials freely from the countryside was considered a normal part of life lived in a landscape known and owned in Sudan, life as a refugee rarely allows the privilege of wandering outside the confines of one's plot or settlement in search of valuable natural resources. Refugees' political and material landscapes elide in the absence of the right to help themselves to materials conceptualized as owned by the host population rather than by the refugees. In the Acholi language, land use and moral authority are conflated in the name of the won ngom, the ‘father of the land’. The lack of such authority is sorely felt by refugees living in foreign landscapes for protracted periods if mechanisms are not created or invented which allow them to assert themselves over the environments in which they live.

Where the possibility of substitution does exist, it is not limited to the material world but also to the realm of social relations and action. A similar principle applies in some cases to the substitution of one person, or category of person, for another in a ritual activity in the settlement. For example, on one occasion, the father of a young woman who had given birth to twins in the settlement initiated the rut ritual which acknowledges and welcomes the arrival of twins into a family, whereas according to Acholi tradition, it should have been her husband who did so. Her husband, a soldier in the SPLA, was absent from the camp with his whereabouts unknown, and the substitution of the father for the husband made it possible for the ritual to go ahead, with all concerned accepting that the husband was so represented. In the formal and solemn distribution of locally brewed beer at the third‐stage funeral rituals mentioned above, the very precise and carefully calibrated social relations which would in Sudan have been asserted via the social groupings defined at lyel (funeral rituals) and the associated reciprocal giving and receiving of beer have been transformed in the landscape of exile. While ‘at home’ in the Sudanese Acholi area, the expectation would have been that guests would be very numerous from the immediate surrounding area comprising people related directly or indirectly to the deceased and his or her family. A smaller number of more distant social connections would also have been expected to attend. In the settlement, the presence in refugee neighbourhoods of members not only of different clans but even of different ‘tribes’ means that much broader and more inclusive groupings are allowed. This accommodates the inclusion as relatively close social connections, people who would have been considerably more distant in the ‘old system’ and the maintenance of what are – in the refugee settlement – important and mutually supportive relationships.

These processes of substitution and accommodation contribute both to the ‘place‐making project’, but also to an attempt to protect, preserve and develop a certain kind of cultural identity while responding to the challenges of modernity and wider social change in the context of conflict and displacement. They are not, however, always effective. Ritual elders in the settlement talk sadly about the failure of certain justice rituals, for example, precisely because the wrong people had attempted to enact them in the settlement in the absence of the correct clan ritual leaders. In one case, a refugee leader attempted to carry out a ritual designed to identify and punish the murderer of another refugee near the border of the settlement in the late 1990s. The failure of the ritual was later explained in terms of the fact that the ritual should have been carried out by a Ugandan leader. Some refugee women insisted that the same ritual performed by the same person would have been effective had it been carried out in Sudan, but that the elder's power was undermined by his being in a foreign place, or not in his own land where he did not control the spirits.

The ritual elders themselves sometimes insist that their interventions are still effective even in a foreign land, but complain that people are now reluctant to follow their advice and practices not because they are ineffective but because refugees are now more strongly following Christian church teachings, some of which instruct them that they have to leave aside such ‘traditional’ activity. While this may be true, in general there is a significant degree of accommodation between the practice of Africanized Christianity (the majority of the settlement's Christians are members of the Catholic Church) and belief in indigenous religion and that the two are rarely conceived of as incompatible. The same elders also note that for political reasons they choose not carry out some of the rituals that they would have performed in Sudan. They argue that they do not attempt to carry out rain‐making and hunting rituals as so doing might offend or alienate the local host population whose right it is to do so on their own territory. Their rights to engage with the landscape are limited, and understood as such. It is acknowledged by many of the Kiryandongo refugees that the elders re‐emerged from relative obscurity in the settlement to play a vital and active role in peace making after a serious incident of inter ethnic violence in the settlement in 2000.

Overall, it is clear that there are tensions within the refugee community over the source and location of moral authority in their protracted exile. Elders clearly feel that their authority over the youths in particular has been lost, and all agree that the changes in residence patterns and the mixing up of clans and sub‐clans, and the new social configuration that this has produced, has encouraged this trend. The fact that extended family groups are no longer resident together on land known and owned by them has minimized the amount of time available to elders to socialize and win the respect of youths. Numerous refugees – men, women, young and old – insist that, on return to Sudan, this pattern will be reversed with families settling together again on ancestral land and they expect a social reversal in which elders re‐assert clan and cultural authority and values in line with this. In the meantime, however, there evidently is less adherence to ‘traditional’ beliefs, especially among the younger generation who have different markers and frameworks to guide them. This has resulted to some extent in intergenerational tension with a strong feeling on the part of some elders that they have lost control over the youth, not least because they are unable to show their control over the spirit world as described above. The ‘jok’ spirits they can still expect to control, meanwhile, are out of range of the settlement, being located in the home area of Sudan, so the youth are understood to be free from any form of retribution as long as the refugees are out of their spiritual landscape too.

A group of secondary school students discussing these issues in January 2006 agreed that they have paid relatively little attention to the kinds of cultural activities that the elders control and furthermore have mixed views about whether they have value. Some expressed interest and support for selected aspects of cultural life (e.g. some dances and aesthetic practices), but reject what they consider to be its outdated components. They admit that their exposure to formal education has increased this tendency, but are clear about the fact that they do not rule out submitting to the authority of the elders on return to Sudan, at least in some respects. Whether this is enough to satisfy those elders in the event that a full‐scale return does take place, of course, remains to be seen.

Conclusions

There are clear limits to the extent to which the Kiryandongo refugees have sought to claim and occupy the physical territory on which they live in Uganda. While they use and transact land for subsistence purposes, they do not claim ‘ownership’ of it in spiritual or symbolic terms. There are probably several explanations for this. In the first instance, of course, the land on which the refugee settlement has been established is gazetted for the purpose by the Government of Uganda, which retains title to it. More importantly, however, refugee elders and farmers state categorically that they are unable to control the spiritual forces which control the territory, and this has an impact not only on the way that agricultural practices are managed, but on a whole range of social, spiritual and symbolic activities that can be carried out there. It is notable that although exile has been protracted, there appears to have been little change in the extent to which such control is considered feasible or even desirable by refugees and their leaders.

Unlike the Mursi people of Ethiopia, about whose place‐making activities David Turton has written so compellingly (2005), the Acholi of Kiryandongo made no positive choice to remove themselves to their current location. Rather, they were obliged to flee violence repeatedly and have landed in Kiryandongo which the majority of them consider to be no more than a temporary home. It is hard to escape the conclusion that the refugees have rejected the notion that they should decisively ‘make place’ in Kiryandongo, for instance by learning or accommodating the local spiritual order. Rather, they maintain their separation and continue to look to a time when they will return to Sudan and the physical and social landscapes – as well as the systems of social organization and authority – they know and own there. In the meantime, however, it has been necessary for them to take steps to engage with and exert some form of control over their temporary refuge, and this they have done by deploying the kinds of strategies described briefly in this article.

With respect to the creation, transaction and use of material objects, economic pressures and priorities of many of the refugee population have a clear bearing on the extent to which the generation of value through material forms has been a possibility in Kiryandongo. Some exceptions existed in the recent past but even these are now less easily found. For example, in the 1990s it was still fairly common to find the ubiquitous stringed adungu musical instrument being produced in the settlement, even when makers preferred to import animal skins for them from Sudan where this was possible. Ten years later, instruments are rarely made there, specialist craftspeople having moved away or turned their attention elsewhere. Women's pot making and basketry are hamstrung by the difficulty of accessing raw materials, either because familiar grasses or clay are not found in this environmental zone, or because refugees are not considered by the local population to have rights to use such natural resources freely. These limitations have clear implications for the extent to which a physical territory can be ‘made’ home by a population. Little wonder, then, that in this case refugees appear to have directed their aesthetic and creative efforts towards intangible socio‐cultural processes, activities and forms. The kinds of ‘substitution’ mentioned above, as in the replacement of one form of material culture for another or by a non‐material for a material form, can certainly be seen as part of a wider and possibly generic process of managing change. Even so, emphasis should be placed on the performative dimension of such activity, with an almost theatrical effort made by members of the refugee group to preserve and re‐present important aspects of culture – material or otherwise – in a context which is in other respects characterized by uncertainty and relative poverty.

While there is a clear causal relationship between being ‘out of place’ in terms of negative impacts on the enjoyment of rights and entitlements for refugees, this situation thus does not constitute the full extent of the challenges that they have to face. Related to legal, protection and livelihood challenges – indeed integrated and embedded in them – is another, related dimension of experience that refugees, often invisibly, engage with and respond to. This paper hopes to have shown that the challenges to socio‐cultural, ritual and political identities and activities are just as great as the more tangible, concrete challenges of protection and subsistence. Although they have been infrequently researched and discussed in the context of refugees in the developing world (much more so in relation to the ‘adaptation’ and ‘integration’ of refugees in industrialized countries), the strategies and accommodations employed by the refugees in Kiryandongo give an indication of how complex, subtle and sophisticated problems and responses in this area can be.

While such matters are, in my view at least, of interest in their own right, they are also crucial for their interaction with the wider political and economic forces that also impact on them. As Gupta & Ferguson (1992, p. 11) remind us, there is necessarily an interaction between the ‘conceptual processes of place making’ and ‘the changing global economic and political conditions of lived spaces’. In the present case what this point implies is an understanding of how refugee efforts in the realm of place‐ and life‐making in Kiryandongo intersect with efforts to support and protect them via the international refugee regime and the national asylum framework of Uganda. Unfortunately, current research, including much of my own, is not sanguine about the extent to which this has been fully successful.

Notes

1. Appadurai in Turton, 2004.

2. The current research project, entitled ‘Answering exile: how Sudanese refugees deal with displacement’, is funded by the AHRC under its Diaspora, Migration and Identities programme. My thanks to that programme, as well as to all the refugees and others who were willing to participate in the research. While my other recent writing has concentrated more directly on the legal, political and material status of refugees in Uganda, this paper will place more emphasis on the socio‐cultural, symbolic and spiritual dimensions of experience. For more on the former (see Kaiser, 2006b and Kaiser, 2007). I would also like to thank Paul Basu and Simon Coleman, co‐convenors of the ‘Migrant Worlds, Material Cultures’ seminar series at Sussex in 2006. It was a pleasure to engage with them through the seminar and I am grateful for their insightful and helpful comments and suggestions on the paper as well as for useful and interesting feedback from other members of the seminar, some of which I have attempted to incorporate here.

3. The Kiryandongo population left home relatively early in the conflict, fleeing the arrival of the SPLA in their villages. There is an assumption within this group that some related people (notably those previously living at the Achol‐pii Settlement) who left later than they (and who are more closely associated with the SPLA), benefited from what had been left behind by them.