‘A refugee is someone who refused to be oppressed’ (‘Siaka’, Kampala, 9 December, 2004).

This comment by a Congolese artist who runs programs for refugee children and young people in Kampala, Uganda, contrasts with official discourses about refugees, particularly young refugees, as being helpless victims of mass displacement. As I have shown elsewhere (Clark 2007a; Clark-Kazak 2009a), refugee agencies such as the United Nations High Commissioner for Refugees (UNHCR) typically approach young refugees within a vulnerability framework. This ‘vulnerables’ approach tends to essentialise their experiences and overlook local protection strategies employed by refugee children and young people as well as their families and communities.

Based on ethnographic fieldwork in Kampala and Kyaka II refugee settlement, and supplemented by other empirical studies in the broader literature on young people in migration situations, this paper explores individual and collective, and daily and long-term self-survival strategies employed by young people from the Democratic Republic of Congo (DRC). While ‘forced’ migration discourses imply that refugees have no choice but to flee generalized insecurity, the experiences of Congolese young people highlighted in this paper suggest that more complex decision-making processes are at play. Drawing on Ruth Lister’s conceptual approach to agency, I demonstrate how young Congolese in refugee contexts in Uganda ‘get by’, ‘get (back) at’, ‘get out’, and/or ‘get organized’. In the final section of this paper, I suggest ways in which outside actors can better recognize and support these self-survival strategies.

Conceptual and Theoretical Frameworks

In this paper, the notion of young people includes both chronological and social age (Clark-Kazak 2009b) definitions. Most organizations working with refugees employ the former to categorize children as human beings under the age of 18, as per the Convention on the Rights of the Child. In contrast, there is no recognized definition of young people in international law. As a result, governments, agencies, and researchers employ a variety of chronological age definitions of youth. For example, the United Nations and its agencies often use the chronological ages of 15–24 years old, based on the definition provided by the General Assembly for International Youth Year in 1985 (UNGA n.d.). However, an analysis of UNHCR annual reports and appeals over a ten-year timeframe revealed differential applications of the chronological age designated as ‘youth’ (Clark-Kazak 2009c).

This variation in definitions of young people highlights the arbitrariness of chronological age (Cohen 1980; Freeman 1983). Many scholars have further critiqued the chronological age definition as being homogenizing and ethnocentric (Boyden 1997; Burman 1994; Pupavac 2001; White 2002). Indeed, sociologists, anthropologists, historians, and geographers have long shown how childhood and youth are socially constructed and hence variable across time, place, and space.1

Building on this literature, I have developed the concept of social age to complement chronological age, similar to the distinction between gender and sex (Clark-Kazak 2009b). Social age refers to the socially constructed meanings and roles ascribed to different stages in the life course. It also highlights power relations within and between generations. Informed by this social age analysis, this paper defines young people as those who have past puberty, but who have not yet married. Chronologically, research subjects varied in age from 13 to 35 years.

In this paper, the term ‘refugee’ is used broadly to describe the circumstances of people who have come to Uganda in the context of conflict in the DRC. This categorisation is framed by historic migration and African practices of prima facie status, as well as research subjects’ self-identification as refugees primarily because they are Congolese. However, the experiences of these young Congolese refugees problematize the assumed binary between forced versus voluntary migration.2 While all fled generalized insecurity in the DRC, the research presented here shows that they took decisions about when to leave, where to go, and with whom to associate. Even in the face of similar levels of violence, young people and their families had differential perceptions of risks and benefits of flight. Hilary Evans Cameron has similarly argued that there is ‘well-documented variance in human response to danger’ (2008: 568). Second, once young people had fled the DRC, they took decisions about where to go within Uganda. Some settled in border areas of western Uganda, particularly where there were ethnic or familial ties to similar groups. Some moved to large urban centres like Fort Portal and Kampala. Others registered as refugees with the UNHCR and the Government of Uganda, and were then transported to designated refugee settlements. Finally, young people decided with whom they would live and associate, as detailed below.

In this way, young people exercised agency even within contexts of constrained choice. This paper interrogates young people’s actions and decisions as part of collective and individual survival strategies, using conceptual frameworks developed by Lister (2004) and further applied by Redmond (2009) and Jones & Sumner (2011). Ruth Lister has developed a widely cited taxonomy of agency in contexts of poverty. She starts from a broad definition of agency as the capacity to act: ‘Survival in the face of oppression and deprivation is helped by a belief in the ability to exercise some measure of control over one’s own life, however limited’ (2004: 126). Lister then develops a way of analysing how poor people take decisions and actions in both short- and long-term perspectives, and in both everyday and strategic ways. Figure 1 below summarizes her approach, which is further explained and applied in the following sections.

Figure 1 

Forms of agency exercised by people in poverty

Redmond (2009: 545–547) has applied this broad taxonomy of agency to thinking about children’s actions in contexts of poverty. Here, ‘getting by’ can be interpreted as the small actions and decisions that poor children take to cope with daily challenges. ‘Getting back at’ refers to rebellious behaviour that children exhibit in response to perceived unjust or unreasonable circumstances they encounter in everyday interactions. ‘Getting out’ is a longer-term strategic activity aimed at transforming the individual child’s circumstances. ‘Getting organized’ refers to collective action that children undertake with other people to lobby for structural changes.

Drawing on Redmond, Jones & Sumner have further demonstrated the application of Lister’s conceptual framework in thinking about poor children’s actions. They have ‘mapped’ the four aspects of agency outlined above across the three well-being domains they have identified as being instrumental to children and young people’s human development (2011: 19–20). The first domain is material; here, agency is related to needs, resources, and observable outcomes (2011: 14–15). The second is relational in which agency is related to connections to others, including both networks of support and obligation. Third, the subjective domain relates to ‘meanings people give to the goals they achieve and the processes they engage in’ (2011: 15).

Methodology and Research Context

In this paper, I would like to apply and extend this discussion of agency to analyse the ways in which young people in conflict situations make decisions about migration and within migration contexts as part of individual and collective self-protection strategies. I draw mostly on ethnographic data I collected from September 2004 to December 2005 with young people from the DRC living as refugees in Kampala as well as in the Kyaka II settlement in Uganda. This is supplemented with empirical evidence from other studies with young migrants in a variety of different contexts.

Uganda’s capital city, Kampala, is a large urban centre that has attracted migrants from across the country, as well as neighbouring states in the Great Lakes region. However, according to the settlement policy in Uganda, refugees are not officially supposed to live in the city and thus do not receive humanitarian assistance nor do they have access to social services. In contrast, Kyaka II is a designated refugee settlement in an isolated, rural area of western Uganda. Refugees arriving in the settlement are registered, documented, and allocated a plot of land where they are supposed to engage in subsistence farming under the Ugandan government’s self-reliance strategy.

I generated data using a variety of methods, including semi-structured interviews, focus group discussions, writing exercises, and observation. Over 400 research subjects were identified using snowball and purposive sampling. Snowball sampling can result in bias towards respondents who share certain characteristics and/or are more visible, thereby undermining representivity (Jacobsen & Landau 2003). Purposive sampling was thus also used to identify multiple entry points into communities and research subjects. Despite time and logistical constraints, the study sought to include young people of different ethnicity, sex, and age, and living in different circumstances. I carried out research, without interpretation, in French, English, and Swahili. However, due to my limited Swahili language skills, the research over-represents people who could speak some French or English and hence had completed some formal education. There is a consequent bias towards middle class research subjects. I have translated all direct quotations in French or Swahili into English. All names in this paper are pseudonyms to protect the confidentiality of research subjects.

‘Getting By’ in Contexts of Conflict and Migration

Lister situates ‘getting by’ kinds of agency in the ‘everyday-personal’ area of her taxonomy of agency. While these small, daily actions are too often invisible to policy-makers, practitioners, and scholars, they are important to refugee young people’s individual and collective survival strategies. Taken together, they also represent important patterns of behaviour that can interact with larger economic, social, and political structures and relationships.

Analysing the intersection of ‘getting by’ agency with considerations of material well-being (the first domain in Jones & Sumner’s framework outlined above) reveals the ways in which refugee young people are important economic actors. Most in this study undertook daily ‘reproductive roles’, such as childcare, cooking, cleaning, and fetching water. Some also engaged in remunerated work, often in the informal economy. Such productive labour included working in hair salons, restaurants, and internet cafes, undertaking domestic work for pay, and doing physical labour, such as farming and transporting items by foot or on bicycle. Individual and family migration decisions were sometimes informed by differential economic opportunities. For example, many young people in Kampala had deliberately migrated to the city in order to try to access employment that was not available in the DRC, in refugee settlements, in rural border areas, or in smaller towns.

In the relational domain, my research shows important ways in which young people exhibit agency when deciding how and when to contribute to collective livelihood strategies in inter-generational families and households. One of the most interesting findings is the fact that some young people choose to live in peer networks over inter-generational networks. Within the refugee literature and programming, there is an assumption that ‘unaccompanied minors’ - people under the age of 18 who have migrated without a parent or guardian - are among the most vulnerable in migrant populations (Clark 2007a; Orgocka 2012). Governments and organizations working with refugees therefore expect, and indeed promote, family reunification as a ‘solution’ to the ‘unaccompanied minors problem’. However, many Congolese young people in my study had carefully weighed the material and relational costs and benefits of living with adults versus living with peers. They concluded that they had greater freedom over their own labour and collective resources when they lived with peers than with adults.

While none of the young people in the research presented here had experienced inter-generational conflict, other studies indicate that children may actually be migrating away from their families as a survival strategy in the context of abuse and/or neglect. For example, Bossin & Demirdache (2012) analyse the case of a young woman from Mexico who successfully claimed refugee status in Canada based on her claim that her mother was the agent of persecution.

Finally, within the subjective domain, young refugees may cope with the challenges of their daily lives by positively interpreting their experiences and maintaining hope for the future. For example, ‘Bondeko’ migrated to Kampala with his younger sister, ‘Promesse’, who had been raped and suffered long-term health complications. He worked as a private French tutor and sold handicrafts in the street in order to make enough money to pay for Promesse’s medical expenses because they could not access public health services. He desperately wanted to be reunited with his parents and the rest of his siblings, with whom he had lost contact when the family fled the DRC. In the hope of finding them, Bondeko regularly visited refugee-serving organizations to ask if anyone had seen them. Eventually, he was reunited with his elder sister, who had been resettled to the United Kingdom. In this way, young people actively seek to achieve their hopes and dreams - even when they seem unattainable - through both daily activities and longer-term strategies.

‘Getting Back At’

At the other end of the personal-political continuum, but still within the ‘everyday’ quadrant in Lister’s taxonomy, is the notion of micro-level forms of resistance in which poor people engage. These include ‘destructive forms of agency against themselves, their families or neighbourhoods, or the wider society’, such as substance abuse and crime. Within this categorization, one would also find James C. Scott’s ‘weapons of the weak’: ‘foot dragging, dissimulation, false compliance, pilfering, feigned ignorance, slander, arsonage, sabotage, and so forth’ (1985: 29).

Applied to my data, in the material domain, some young people deliberately disobeyed and/or undermined economic structures that were deemed unfair or too restrictive. For example, many young people in Kyaka II contravened food distribution policies by selling or bartering their rations. Others registered in the settlement so that they could receive rations and other material benefits, but actually lived for the most part in Fort Portal or Kampala. In these ways, young people were deliberately engaging in everyday forms of resistance against authorities and economic structures.

Related to non-compliance with economic policies were ways in which young people resisted relational structures by appropriating and transforming them through daily actions and speech. For example, some young people engaged in ‘vulnerables’ discourse to leverage additional resources and access to political spaces. In Kyaka II, it was particularly important for young people to self-identify as a ‘vulnerable’ person in order to access UNHCR officials, who came irregularly to the camp to conduct protection interviews.3 Utas (2004) has coined the term ‘victimcy’ to describe the use of agency to present oneself as a helpless victim. Similarly, young refugees in my study used the local and international discourses of victimhood to gain access to decision-makers and decision-making processes.

At a subjective level, young refugees sometimes exercise what Redmond has termed ‘self-exclusion’ (2009: 545). For example, children and young people may not ask their parents to pay for certain activities or apply for subsidies ‘in order to protect both themselves and their parents from […] stigma […] as well as protect their parents from financial demands that they cannot easily accommodate’ (2009: 545). From my research, the experiences of ‘Lucie’ are illustrative. Coming from a wealthy family, Lucie attended private school in the DRC. When she fled with her step-mother and siblings to the Kyaka II refugee settlement, Lucie did not attend school because she was primarily responsible for reproductive tasks at home, and because she was disappointed with the poor quality of education in the settlement. The principal of the secondary school encouraged Lucie to write an exam for a scholarship to study outside the settlement. However, Lucie withdrew her application for the scholarship after her step-mother told her that a scholarship could jeopardize the family’s chances of resettlement.

‘Getting Out’

In the personal-strategic quadrant of Lister’s taxonomy are ways in which refugee young people exhibit agency by ‘getting out’ of their situation - or attempting to do so. At a material level, formal schooling was one strategy to which all young people in the study, including those who had never had access to formal education, aspired. When probed about the reasons for wanting to pursue further studies, young people linked education to white-collar employment, class, and socio-political status. Historically, under Belgian rule, Congolese with formal education occupied special status as ‘évolués’. After independence, évolués consolidated status as political elite. Young people believed in the continued causal relationship between education and leadership/status. For example, in individual writing exercises, students in Kyaka II wrote: ‘In the future, I want to be a doctor because someone who has an education has a right to speak in the community.’ And, ‘If you are talking in public, they will listen to you.’ Similarly, in a focus group discussion with male secondary students one remarked, ‘My plans are very big. I don’t know if they will come. I want to study. Then I can be a leader and do my best to explain the problems facing refugees in public…in the UN.’ Young women also saw education as a way to secure better marriages and negotiate daily power relations: ‘If you’re a career girl, you can win a man at the same level. Without education, you only get a village boy’ (Female secondary student, Focus group discussion, Bujabuli, Kyaka II).

This gendered difference in educational objectives speaks to other relational ways in which young people attempt to ‘get out’ of unequal power structures. As mentioned above, peer networks are one way in which young people challenge aged-based social norms and hierarchies. While historically in the DRC young people did not move out of inter-generational families until they married, youth-headed households are more common in contexts of conflict and displacement. By taking decisions for themselves at a household level, young people became empowered to demand more access to resources and decision-making at community and policy levels (Clark 2007b).

At a subjective level, many young people had long-term plans and ambitious hopes for the future, as demonstrated above in relation to education. Many young people also turned to faith and organized religion as a way to make sense of their experiences. For example, ‘Rose’ is a young Rwandophone Tutsi woman whose family was massacred in a refugee camp in Burundi. She migrated by herself to Kampala and lived precariously, exchanging domestic labour for shelter and food with different Congolese families, in one of which she was subject to sexual harassment. She worked long hours for no pay (because she was ‘in training’) as an apprentice in a Congolese hairdressing salon. The only break she had from her salon and domestic work was attending an all-night Pentecostal service on Fridays and singing in the choir at another church on Sunday mornings. In response to her daily struggles and long-term challenges, she always referred to her faith in God.

‘Getting Organized’

In the final quadrant of Lister’s taxonomy, she describes how people act at the collective, strategic level to ‘get organized’ to change their situations. Materially, young Congolese pool rations and supplies in Kyaka II. Over the course of my research there, I developed relationships with two groups of young people who had spontaneously come together in order to benefit from materials that the UNHCR would only give to ‘households’: plastic sheeting and cooking pots and utensils. The young people constructed a shelter together and took turns preparing collective meals, to which they all contributed rations and/or food they had bartered or bought.

At a relational level, the formation of organizations is important. In Kampala, the Refugee Youth Association (RYA) is a forum for collective action. Similarly, some young people in Kyaka II created a clandestine organization to document alleged abuses in the settlement. There was also an official ‘youth’ representative on the Refugee Welfare Committees within Kyaka II. These fora allow young people to raise their protection concerns to representatives of the government, the United Nations, and refugee agencies. Such opportunities available to groups of young people supersede individual capacity and interests. As RYA members explained, ‘God sends rain, but people must organize to collect water.’ Young refugees also spontaneously organize in response to particular issues and events. For example, several young people decided to stage a demonstration to protest school fees at Bujabuli Secondary School in Kyaka II.

At a subjective level, many young people in this study took pride in participating in cultural activities closely associated with their Congolese national identity. Music, in particular, was a way for young Congolese to make sense of their experiences, drawing on a long tradition of music as a means of both cultural and political expression in the DRC (Stewart 2000; Wheeler 2005; White 2008). As the leader of the RYA’s music club explained, ‘UNHCR tells us not to talk about politics. But everything that happens is political. We are here because of politics. Even if in the refugee convention it says that refugees can’t talk politics, it penalizes refugees. […] We want to send messages through music. Our songs can go even in areas where we can’t.’

Through music and other means, young refugees are able to act collectively to strategically position themselves vis-à-vis discourses and meanings of refugeehood and youth. As noted in the title of this paper, many young people redefine refugeehood from a passive status of victimhood to an active way to resist oppression. As ‘Précieux’ notes, ‘Being a refugee is not shameful, it is a protection you give to yourself.’ Similarly, they interpret ‘youth’ in different ways than adults. For example, one elder in the Kyaka II refugee settlement defined a child as ‘someone who doesn’t reason’ and a young person as ‘someone who doesn’t understand’. In contrast, young people self-identified in positive ways. For example, ‘Annette’ defined a young person as ‘someone who is strong, who is capable of doing everything’. By providing alternate discourses, young refugees are subjectively (re)appropriating - and potentially transforming or undermining - restrictive structures.

Conclusions and Recommendations

The research presented in this paper suggests several points of departure for academics, policy makers, and practitioners working with refugee young people. First, it shows that young people themselves are engaging in survival strategies in the face of both practical everyday needs and larger structural challenges. While the international community too often prescribes protection strategies for refugee children and young people with preconceived assumptions about vulnerabilities (Clark 2007a; Orgocka 2012), this research suggests that more attention should be paid to the multiple ways in which young people, their families, and communities exercise decision-making with respect to material, relational, and subjective well-being.

Second, we should not underestimate refugee young people’s political and strategic roles. In my view, Lister’s taxonomy erroneously separates out the ‘personal’ from the ‘political’ in the horizontal continuum of her conception of agency (see Figure 1). Feminists have long argued that the personal is political. Similarly, my research has shown how young people’s daily decisions at a micro level can have impacts on power relations within families, households, and communities. Moreover, patterns of everyday behaviour can undermine or transform broader political structures. Finally, young people who took decisions within their own households were more likely to be empowered to take on leadership roles at community and policy levels (Clark 2007b). Another conceptual problem with the agency taxonomy presented by Lister and applied by Redmond and Jones & Sumner is that it implicitly equates politics with resistance and ‘negative’ activities. Instead of pathologizing political activity, we need to better understand young people’s decision-making roles in families, households, communities, and policy spaces. In the examples above, I have attempted to demonstrate both destructive and constructive political activity undertaken by refugee young people.

Third, recognizing young people’s agency should not romanticize their experiences (cf. Salazar 1991; West 1990) nor allow governments to abdicate their responsibility to protect and provide for displaced populations. Here, there is a role for outside actors in identifying and attempting to change restrictive and oppressive structures and practices. For example, young Congolese with whom I worked were doubly legally disenfranchised. First, young people under the age of 18 do not have the right to vote in Uganda or the DRC. Second, Ugandan law at the time of research prohibited refugees from engaging in ‘political’ activities, but politics was not defined. This allowed authorities to use their discretionary power to prohibit activities they deemed ‘too political’. For example, in Kyaka II at the time of research, refugee women were prohibited from setting up a cooperative to pool agricultural products and sell them collectively to traders in nearby towns on the grounds that such a cooperative would be ‘political’. Moreover, a culture of impunity existed in the settlement. After some students peacefully demonstrated against school fees, settlement officials allegedly physically assaulted those who participated, imprisoned them without charge, and burned down their huts. It is within these contexts of asymmetric power relations that refugee agencies can advocate for increased attention to and respect for young people’s rights, experiences, and opinions.

A final recommendation relates to differential experiences within refugee populations. While social age does influence young people’s opportunities and constraints, it is not the only factor. Their experiences of violence and displacement are also shaped by other subject positions and power relations, particularly gender, class, and ethnicity. We need more data on, and analysis of, this intersectionality. As Redmond (2009) notes, we should also not overlook the exclusion and structural barriers that non-poor (and, in this case, non-refugee) children and young people may exhibit towards poor and refugee children and young people. Being of the same chronological age or cohort does not automatically make for solidarity. Moreover, scholars, policy-makers, and practitioners cannot view ‘the refugee child’ in isolation from their intra- and inter-generational networks, including families, households, and communities.

When identifying and supporting young refugees’ self-survival strategies – both individual and collective as well as in relation to material, relational, and subjective well-being – we need to recognize the complexity of individuals and migration processes. As the introduction to this volume has noted, the literature on civilian protection tends to focus either on normative and institutional frameworks, or on specific examples of civilian self-protection. This article falls mostly in the latter category. However, by analysing the empirical data through Lister’s agency framework, this paper could serve as a point of departure for further work on the interaction between individuals’ self-protection and migration structures institutionalized by states and humanitarian organizations. In particular, more attention to the ways in which refugees – individually and collectively – perceive, shape, and/or undermine norms and institutions that are ostensibly there to protect them could ultimately improve protection responses to forced migration.