Precarious Care across Migrant Generations in Tanzania

Abstract

Based on ethnographic fieldwork, this article is concerned with how undocumented refugees and migrants use invisibility strategies to navigate a hostile host environment in Western Tanzania. This article explores how the shifts in Tanzania’s refugee policy have affected different generations of refugees differently, and how older cohorts assist newer cohorts. This article argues that the challenges of migration are productive of ‘affective circuits’ and of generating new forms of kinship. It argues that it can be productive to bring together the different understandings of generations, as it was found that generations as cohorts can transform into generations as kin in situations of rupture and adversity.

1. Introduction

Refugees have been arriving in Kigoma, Tanzania, for over fifty years now, fleeing from violent conflicts in Burundi and the DRC. Some wind up in refugee camps, while others try their luck in and around the city.1 The Tanzanian state’s immigration policy has become increasingly restrictive, following a strict policy of encampment and making life in camps difficult, so as to encourage repatriation (Chaulia 2003; Milner 2019). These hostile policies have led to a sense of suspicion in the local community and a sense of fear amongst refugees and migrants. In the words of Hussein,2 who was born in Tanzania to Burundian migrant parents, “It doesn’t matter that someone has a Tanzanian citizenship [uraia], one must always make sure to hide one’s Burundian background by speaking a perfect Swahili […]” Regardless of their legal status, refugees and migrants go to great lengths to hide their national identity, through strategies of invisibility (Bjarnesen and Turner 2020) and try to ‘pass’ as Tanzanians (Daley et al. 2018; Sommers 2001; Weima 2021).

The study is part of a larger research project on Everyday Humanitarianism in Tanzania (https://www.everydayhumanitarianismintanzania.org/, (accessed on 29 June 2024)) where we examine how ordinary people in Kigoma engage in assisting refugees in need. In this paper, we explore how the current hostility and suspicion in the country, along with the disruptions to kinship structures caused by migration, have led to the formation of new kinship bonds. First, we argue that different generations—in Mannheim’s sense of cohorts—make use of different strategies because different options are open to them, depending on when they arrived in Tanzania. Those who arrived when Tanzania’s refugee policy was relatively welcoming have used other strategies than those who arrived later. Second, we suggest that the older cohorts respond to the hostility of the host environment by assisting newer arrivals. For instance, refugees who have lived in Tanzania for a while and have ‘papers’ of some kind or have established a position in the local community, creating a kind of ‘de facto citizenship’ (Miletzki 2020), are able to use this position to help newer migrants start businesses, find employment or buy property in their name. The first cohorts also use their positions in Tanzanian society to assist later cohorts in avoiding the scrutiny of the Tanzanian authorities.

The older cohorts’ solidarity towards the later cohorts points to another aspect of generations, namely generation as ‘a genealogical relationship of kinship’ (Alber et al. 2008, p. 3). By assisting later cohorts of refugees, they are effectively ‘making kinship’ (Andrikopoulos and Duyvendak 2020), ‘kinning’ (Suerbaum and Richter-Devroe 2022) or ‘regenerating kinship’ (Cole and Groes 2016). In other words, while kinship ties might have been broken during displacement, new ones are forged in the hostile host environment. The most concrete example of this is when migrants who arrived decades ago call new arrivals ‘our brothers and sisters’ as reasons for assisting them, and newcomers may refer to their benefactors as fathers.

Finally, we explore how concrete generational relations within families are reconfigured. For instance, the refugees who leave the camps to search for a better life in the city are often young adults, leaving their parents in the camp to take care of their infant children. The camps cannot offer them a future and let them ‘become someone’. Meanwhile, they maintain their links to the older parents, who also keep the door open to resettlement in the US or elsewhere in the Global North.

While we argue that the challenges of migration are productive of what Cole and Groes have called ‘affective circuits’ (Cole and Groes 2016) and of generating new forms of kinship, we have to keep in mind that such relations are not simply based on a positive affect such as love and trust. On the one hand, there is a feeling of empathy and solidarity towards fellow migrants and ethnic/national kin. On the other hand, the presence of new migrants who are not able to remain invisible, jeopardises the security of the old refugees, and any association with the newcomers puts the old cohorts at risk. The older generations are often suspicious of the newer arrivals, accusing them of being violent or just disrespectful. Conversely, new arrivals are jealous of the older generations. Such ‘dark sides’ of kinship (Geschiere 2020; Andrikopoulos and Duyvendak 2020) are, however, common to kinship anywhere (Bakuri et al. 2020) and are still productive of kinship.

Our main argument is that displacement and, in particular, state policies towards migrants in the receiving society at once challenge existing generational relations and ‘re-generate’ new ones. The hostile environment has meant the breakdown of some generational relations amongst the refugees, while their attempts to assist one another have created new kinships. In other words, large political structures like war in Burundi and the DRC and Tanzanian national refugee policies are entwined with the intimate relations of kinship. While there is a large literature on transnational kinship (Baldassar 2014; Cole and Groes 2016; Parreñas 2005), we focus here on the relationships that are created ‘in place’ through practices that might not be based on biological relations. Furthermore, we argue that it can be productive to bring together the different understandings of generation, as we found that generations as cohorts can transform into generations as kin in situations of rupture and adversity.

This article is organised as follows. In the following section, we outline anthropological debates on generation and kinship in relation to migration. This is followed by an introduction to the methods used. Next, we outline the refugee situation in and around Kigoma. What follows is a section on how migrants and refugees navigate this hostile environment. We explore how invisibility is actively pursued in order to blend in, and how both migrants and hosts are slipping in and out of invisibility to avoid suspicion. Then, we discuss how different cohorts apply different strategies of suspicion and invisibility. We explore how the early cohorts of Burundian and Congolese refugees use their invisibility to assist new arrivals from Burundi and Congo, respectively, thus creating new kinships. Finally, we explore how generation as a kinship relation is used as a strategy to overcome the challenges of refugee life.

2. Generations, Kinship and Migration

Generation can have both a passive and an active meaning. Individuals are generated; they are the product of a position in history and in society. However, to generate is also to create something new (Alber et al. 2008, p. 3). Generation may, in other words, help us understand how individuals are the product of history and social relations while at the same time being able to (re)create history and social relations. This is particularly important, we would argue, in times of rupture. In such times, generations as we know them might be challenged and take on new forms. In such times, generations are generated but not merely reproduced.

What do we mean, however, by generations being relational and historical? We may distinguish between three understandings of generation; kinship, age set and historical cohort. The classical anthropological approach is to focus on generations as genealogical relations of kinship where the main concern is the relationships between parents and children, or grandparents and grandchildren (Alber et al. 2008). This is also the core focus in newer migration literature that explores transnational relations of care in families (Baldassar 2014; Boris and Parreñas 2010; Cole and Groes 2016; Fog Olwig 2002). Much of this literature takes an active approach to the understanding of kinship relations, exploring how kinship is ‘made’ (Andrikopoulos and Duyvendak 2020) or ‘regenerated’ (Cole and Groes 2016).

Related to the genealogical understanding of generations as relational is an approach to generations as a principle for structuring societies (Alber et al. 2008, p. 4) where the focus is on socially defined age groups. From the late 1990s, we have witnessed a number of studies on ‘youth’ or ‘young men’ in Africa (Berckmoes 2014; Christiansen et al. 2006). However, while these studies had their eyes fixed on an age group, they also explored them in relation to larger historical and political contexts like globalisation and the economic crisis, and explained that it was these conditions that prevented these youths from becoming adults (Honwana and de Boeck 2005), making many of them decide to migrate to ‘find life’ (Vigh 2009; Berckmoes and White 2016). This understanding of generations as historically shaped cohorts is often ascribed to Karl Mannheim, who argues that each generation is shaped through ‘fresh contact’ with a different historical reality. It is this understanding of generations that is implicit in much mainstream migration research where terms like first-, second- and third-generation migrant are used at will and often inconsistently (Kertzer 1983; De Haas et al. 2019, p. 25). However, exploring generations as a Mannheimian cohort can be helpful to analyse the experience of refugees in Kigoma, as they had ‘fresh contact’ in Tanzania during very different historical contexts.

We explore the different experiences of the different cohorts and how they navigated them accordingly. Furthermore, we explore how different cohorts relate to one another. This means that we combine an understanding of generations as cohorts with an understanding of generations as relational kinship.

Carsten makes an important distinction between kinship as being and kinship as doing, where the former is the passive reproduction of kinship positions and relations, while the latter is about the effort of maintaining and remaking kinship (Carsten 1995, 2020). Andrikopoulus and Dyvendak relate this point to migration, claiming that kinship can assist and shape migration (as seen for instance in chain migration) but that migration can also shape kinship, in effect ‘re-generating’ kinship (Andrikopoulos and Duyvendak 2020). Such a regeneration of kinship in migration is related to care and affect, what Cole and Groes call ‘affective circuits’ (Cole and Groes 2016). They define affective circuits as ‘the myriad exchanges of goods, people, ideas, and money through which migrants negotiate their social relationships, drawing particular attention to the deeply held sentiments that ride alongside and become a part of these exchanges’. (Cole and Groes 2016, p. 2). Whereas the ‘care chain’ literature focuses on unequal power relations along a unilinear chain of exploitation of care (Boris and Parreñas 2010), affective circuits ‘capture the circular movements of people and resources both among migrants in Europe and between migrants in Europe and their kin or friends who remain in Africa’ (Cole and Groes 2016, p. 7). Baldassar similarly argues against the idea of one-way ‘care chains’ and proposes the idea of the circulation of care (Baldassar 2014).

When we discuss doing kinship and the ‘labour of love’ (Bakuri et al. 2020) that is invested in these affective circuits, we must keep in mind that ‘the circuit metaphor captures the potential for disconnection and conflict: it implies that the social networks through which objects, ideas, and people move are subject to regulation, slow downs, and blockage’ (Cole and Groes 2016, p. 7). Furthermore, we must not assume that all affective circuits are happy or positive. Kinship also has a ‘dark side’ linked to mistrust (Andrikopoulos and Duyvendak 2020), secrecy (Bakuri et al. 2020), witchcraft (Geschiere 2020), jealousy and competition (Boccagni and Baldassar 2015; Fog Olwig 2002). Whether the bright or the dark side, however, kinship is made and re-made through affective circuits. Affect can be negative or positive but it links to circuits of bodies, money, favours and goods. As Mauss has shown, the circulation of material objects creates affective relations between giver and receiver (Mauss [1954] 1990), and as such it becomes productive of relationships and circuits.

Finally, when we claim that the rupture of migration is productive of kinship—that it regenerates generations—we do not mean that migrants suddenly have ‘more’ kinship relations than before or kinship relations of a deeper or more significant quality. We simply mean that they change; old relations cease to make sense, while others emerge. Suerbaum and Richter-Devroe call this kinning and de-kinning (Suerbaum and Richter-Devroe 2022). Likewise, generations-as-cohorts emerge in Tanzania that would not emerge in the home country due to ‘fresh contact’ as migrants, while other generations-as-cohorts might be lacking in Tanzania.

In sum, we explore kinship as both cohorts and kinship relations, and show how kinning takes place through affective circuits in this situation of rupture.

3. Methods

Doing fieldwork among people who do not have papers and/or who are stigmatised by mainstream society poses a number of practical and ethical challenges. Practically, they might not be interested in being ‘found’. Ethically, we can ask ourselves, on the one hand, whether we ought to be revealing their existence. On the other hand, by studying such populations, we shed light on issues that are important and that otherwise would go ‘below the radar’ in research, politics and policy work.3 We have chosen to focus on this important group of invisibilised migrants and have taken a number of concrete precautions in order to protect them from harm. Apart from the concrete steps, outlined below, our main ethical thrust is to follow the anthropological and ethnographic search for a holistic understanding of these people’s practices—beyond the simple binaries of legal and illegal.

Both authors have done ethnographic fieldwork in and around Kigoma in Western Tanzania. Turner visited Kigoma for two weeks in 2022, when he managed to establish contact with a group of Congolese refugees who had established a church for Congolese refugees and with some Burundian fishermen, many of whom had been refugees since 1972. In 2023, he returned with Ruzibiza. They followed up on the networks established the previous year; especially, a young Congolese man led them to numerous interlocutors. Furthermore, Ruzibiza has a Burundian background and was able to address Burundians in shops, on the street, in taxis, etc. While sceptical at first, these contacts proved invaluable. Turner stayed only a few weeks, while Ruzibiza stayed for another month, allowing closer relationships with their interlocutors. In additional, both authors possess extensive experience with refugee populations, particularly in conducting research among Burundian refugees. A significant advantage is that Turner not only has experience with themes related to refugees and migration, but also with Tanzania as a research context. This familiarity allowed us to effectively draw on his past experiences and contextual knowledge.

Interviews took place in public spaces such as an empty beach on the lake, in church buildings, in empty bars and at our hotel. A few interviews took place in people’s homes. We conducted interviews with thirty-five Burundians, twenty-seven Congolese and nineteen Tanzanians. Our interlocutors spanned migrants and refugees with the permits to stay in Kigoma to newly arrived migrants who had no documents and who slept outside near the lake—and everything in between. Some had refugee status in the camps but spent most of their time in the city. Others had been expelled from Tanzania only to return via the ‘back-roads’. Some were born in Tanzania, while others had married Tanzanians. Interviews were conducted in a mixture of Kirundi, Swahili, French and English. The mixing of languages and the attempts to translate certain terms often created interesting points of reflection and common analysis.

Interviews were conducted, recorded, transcribed and translated, except in cases where the participants declined to be recorded due to concerns for their safety and apprehensions about the use of their recordings. Despite our efforts to clearly explain the academic nature and purpose of the research, some participants remained uneasy, fearing that the recordings might be used against them or reported to the authorities. In instances where recording was not permitted, both Turner and Ruzibiza maintained separate fieldnotes. These fieldnotes were also used to document their personal observations, informal discussions and details of formal interviews where recording was not allowed. The transcribed interviews were manually coded by the researchers and systematically organised into themes, leading to the identification of the theme discussed in this paper.

Information and citations from the interviews are referenced according to the location of the interview and the date of the interview, e.g., (Prince, 1 April 2023, Grano Beach), and when it is fieldnotes taken by the researcher that are used, they are referenced as such, followed by the date, e.g., (Fieldnotes, 11 March 2022).

4. Refugee Hosting in Tanzania

The 400.000 Hutu who fled Burundi in 1972 were mostly settled in settlements in the Central Tanzanian regions of Tabora and Rukwa, where they were given plots of land and soon became self-sufficient (Anthony 1990; Armstrong 1990; Christensen 1985; Gasarasi 1984; Malkki 1995). It has been argued that the refugees were framed by the Tanzanian state as resources for economic development, and that the early settlements were models of villagisation (Daley 2007; Rosenthal 2015; van Hoyweghen 2001). Other Burundians were given permission to stay in Kigoma. We met several of these refugees who have settled in the Kibilisi fishing village just outside Kigoma town. The majority of the Burundians from 1972 whom we met have papers of residence but not full citizenship. In 2007, the Tanzanian state offered the Burundian refugees from 1972 the choice to either repatriate or naturalise (Daley et al. 2018; Miletzki 2020). The process has been long and bumpy and has been stalled by the Tanzanian government on several occasions. By 2015, Tanzania had successfully granted naturalisation status to more than 160,000 individuals (Kuch 2018; Miletzki 2020). Many are still waiting for their cases to be processed, however.

Due in part to the massive influx of refugees from Burundi, Rwanda and the DRC in the 1990’s, Tanzanian refugee policy was tightened, resulting in a strict encampment policy (Boeyink 2019; Chaulia 2003; Fellesson 2021; Kamanga 2005; Kuch 2018; Milner 2019; Rutinwa 1999; van Hoyweghen 2001). While most Burundian refugees were repatriated after the peace agreements in the early 2000s, renewed political unrest and violence resulted in a new influx in 2015. In recent years, the Tanzanian state has made life in the camps much harder, with the explicit aim to force Burundians to repatriate (Boeyink 2021; Fellesson 2021; Milner 2019). Due to the harsh living conditions in the camps, many refugees leave the camps and work for Tanzanians in neighbouring villages (Msoka and Kweka 2022), settle in fishing communities along the coast of Lake Tanganyika or hide in Kigoma town.

Refugees continue to arrive from the DRC. They arrived in large numbers during what has come to be known as ‘the first Congo war’ in the late 1990s, when they were placed in refugee camps together with Burundian refugees. Since then, Congolese have been arriving according to the ups and downs of the various conflicts and the concomitant economic challenges in the Eastern DRC. Many of the Congolese living in Kigoma town seem to have papers from the IOM and manage quite well in the community. Others have left the camps clandestinely, and are in constant fear of being caught by the authorities.

Throughout our interviews, Tanzanians would express ambiguous perceptions of Burundians. One moment, they would tell us that Burundians are kind people, and the next moment, they would accuse them of being untrustworthy and murderous.4 Congolese, on the other hand, were most often perceived in a more positive light, although most Congolese we spoke to feared the Tanzanian authorities and feared that Tanzanians would report them to the authorities. Prince, a Congolese man who has lived in Tanzania for 13 years, expresses these relations well.

In short, we live a secret life here. We hide from everyone. You do not get mixed up in your neighbours’ lives to avoid attracting attention. You hide your identity of being a Congolese refugee in Kigoma. You can be a Congolese (with proper visa/documents allowing you to be in Tanzania), that is fine, but if you are a refugee, you must do whatever it takes to hide that because it is your neighbours who might report you to the migration officer or authorities. (…) we are living, but our hearts are afraid. (…) We are cautious and do our best not to disclose that we are refugees. First, you yourself cannot dare to say that you are a refugee. It can’t get out…

(Prince, 1 April 2023, Grano Beach)

Remaining invisible is vital for Prince. ‘We live a secret life’, he explains, ‘we hide’ and ‘avoid attracting attention’. The trick is not to avoid contact with Tanzanians but to hide certain sides of one’s identity—namely one’s status as refugee. Bjarnesen and Turner (2020) argue that migrants only try to be invisible to certain audiences in certain contexts. In other situations, it is important to be visible, like when migrants apply for protection from the IOM or for resettlement from the UNHCR. Boeyink has similarly argued that refugees who leave the camps, looking for work, want to be ‘visible enough’ to be seen by potential employers, while remaining ‘invisible enough’ to avoid being stopped by the authorities (Boeyink 2020).

In his seminal book Fear in Bongoland (2001), Marc Sommers observes how Burundians settled in Dar-es-Salaam (also known as Bongoland, because you need to be street-smart and use your brain—bongo in Swahili—to survive) live in constant fear. He attributes this fear to their inherited experience of ethnic violence and mistrust in Burundi. We agree with Daley et al. (2018) that this fear is due, rather, to the situation in Tanzania. This fear does not prevent them from engaging with Tanzanians, but it makes them cautious and creates what we call a ‘tense sociality’. Collette, a Congolese woman, whom we also met at the empty beach resort, Grano Beach, explained that she has plenty of Tanzanian acquaintances, ‘but I never fully trust them’. She has to be cautious around them and consider carefully what she reveals to them. We suggest that this ‘tense sociality’ between hosts and migrants is due to a widespread mistrust between neighbours in Tanzanian society in general and that this may be related to the Tanzanian surveillance state. The state has, since the time of Ujamaa, penetrated and controlled society right down to the sub-village level, which nurtures a sense of never feeling safe amongst Tanzanians (Abrahams 1987; Cross 2014, 2016; Fleisher 2000). In this context of hostility from the authorities and general mistrust amongst the local population, the Congolese and Burundian refugees must navigate their lives. In the following, we explore how the change in government policies has affected different cohorts differently.

5. Shifting Times and Generational Strategies

In March 2022, I (Turner) was visiting a group of elderly men who had arrived in Tanzania from Burundi in 1972. The host—the chairman of the local fishermen’s association—apologised that he had been suspicious towards me when we met a few days ago, but they are wary about talking about being Burundian. They explained that when they arrived, the Tanzanian villagers welcomed them, everything was good and they experienced no discrimination. Some of them had lived in camps, while others had lived in villages. Being fishermen, however, they preferred to live near the lake. At one point, the UNHCR created a special zone—which became Kibilisi and Katonga—for both refugees and Tanzanians, where the UNHCR established a primary school and a dispensary that could be used by refugees and Tanzanians alike. ‘We were happy with this’, they said. In 2010, they could opt for Tanzanian citizenship. They went for interviews, but nothing happened. In 2018, there was a follow up, but nothing has come of it. ‘We are still waiting’, they said. It was this uncertainty that was getting at them. They could build houses and send their children to school and even university. But it remained a challenge to get proper fishing licences and business licences. Also, they were not able to travel freely without the right papers. One of the old men explained that ‘we live in uncertainty’—moving his hands from side to side to demonstrate the lack of direction and stability. One of them took it to a more existential level, explaining that ‘we live being unsure who we are… It takes our confidence’ (Fieldnotes, 11 March 2022).

These old men lived Tanzanian lives; they had often married Tanzanian women, and their children spoke Swahili and had done all their schooling in Tanzania. While getting formal citizenship would not change their life options radically, the fact that they were still waiting after fifty years gave them a sense of being kept outside. This also meant that they felt that they could not become whole people and move forward in life. They were particularly concerned for the future of their children and grandchildren who were born in Tanzania. And while these concerns may seem minor in relation to other migrants’ challenges, they contributed to the sense of not being able to ‘move forward’ in life and across generations.

Abdul has lived in Kigoma since 1991. We met with him through a Congolese friend at a small bar in Kibilisi in 2023. Abdul is a quiet, modest man in his late fifties. He is a pastor and lives in a house that he has built in Mlolo on the outskirts of Kigoma. He explained to us that to be accepted by the Tanzanians, one has to ‘integrate well into their families’. We were not quite sure what he meant by this, until he explained that he had married a Tanzanian. Daley et al. also found that intermarriage was a strategy of invisibility among ‘1972’ refugees from Burundi (Daley et al. 2018, p. 30). He had also changed his name from the French, Christian name, Jean Paul, to a more Swahili, Muslim name, Abdul, something that Duchai and Ntihirageza also found in the case of Burundian refugees in Tanzania (Duchaj and Ntihirageza 2009).

The conversation shifted into stories about betrayal and complicated court cases with Tanzanians, and he became increasingly nervous and elusive. He suggested that it was not safe to continue the discussion there and insisted that we meet again so that he could present us with his documents. The following day, he arrived at our hotel, well dressed and clutching a number of papers and documents that he carefully laid on the table and started to explain. Some were hardly recognisable identity papers from Zaire, his original refugee papers from Tanzania and his residence permit. The other documents were from a court case. We gradually understood that the betrayal story was related to an NGO that he had established with a friend and which had become successful until he and his friend were forced out. He explained that his Tanzanian collaborators falsely accused him and his friend of fraud—basically so that they could pilfer the laptops and motorbikes that belonged to the NGO.

During his interviews, Abdul repeatedly highlighted the fact that refugees and migrants are constantly being monitored, no matter how hard they try to integrate. For instance, even his personal experience of marrying into Tanzanian society, changing his name and collaborating with local Tanzanians to start an NGO did not always have positive outcomes. Abdul had married his way into Tanzanian society, but when his Tanzanian wife died and he married a Congolese woman, he was again shunned from the local community, he claims. Abdul was falsely imprisoned twice and lived in constant fear of being wrongly accused and stripped of his rights.

Indeed, many of the respondents we interviewed highlighted this tense sociality. Hussein, whom we mentioned in the introduction, has Tanzanian papers as he was born in Tanzania, but he explained that even with authentic Tanzanian citizenship documents, further checks were conducted, including checking the location of one’s smallpox vaccine—what Reeves (2013) terms “the space beyond the document”.

[…] but being born in Tanzania has helped because the vaccine mark that they check when they are doubting your nationality and background have saved me at multiple occasions… it helped to confirm my Tanzanian citizenship whenever I was scrutinised”.

(Hussein, 2 April 2023, Katonga)

Apparently, it is easy to differentiate someone born in Tanzania from someone born in Burundi or the DRC by checking the location of the scar from the smallpox vaccine (Weima 2021).

The issue of not being allowed to be successful was prominent in their narratives.5 Abdul explained that ‘Tanzanians can accept us as long as we are suffering. If we are more successful than them, they will report us to the police’. A fisherman in Katonga told us that ‘the Tanzanians are jealous of our fishing skills’, and that is the reason why they report them to the authorities (Fishemen, Katonga, 9th March 2022).

We hide and can never disclose that we are refugees. Some Tanzanians are jealous, and when they see us doing well, they are not happy. We are doing the same work as them (e.g., I drive a bajaji like other Tanzanians) […] Some may report us to the migration authorities. Therefore, we refugees, here in Kigoma hide to avoid being reported.

(Kevin, 10 May 2023, at the hotel)

Abdul has lived in Tanzania for decades and has papers of some sort. He is also a respected pastor with a mostly Tanzanian congregation, which helps him navigate this cautious sociality with Tanzanian neighbours and authorities. We might claim that he has practical citizenship. Other migrants are in a more precarious situation than him. These are the migrants and refugees who have arrived later and who have no official papers. Their main preoccupation seems to be the fear of being caught by the police.

Papy and Heritier were two young men who were officially registered in Nyarugusu refugee camp but who spent most of their time in Kigoma, ‘finding a life’. They travelled regularly to the camp in order to remain registered, as they were also applying for resettlement in the US and Canada. Indeed, Papy’s resettlement case was successful, and he is currently in Canada. When we asked about the dangers of travelling the 150 km between the camp and Kigoma—along a road with many police checkpoints—they responded casually that they just have to pay the bribe or spend four days in prison. Others told more dramatic stories of police officers looking through the contacts on their mobile phone and asking them to get relatives in North America or Australia to transfer funds for their release. Their lack of papers made them vulnerable to such exploitation.

On our final visit to the fishing community in Katonga, we got to speak to some of the most precarious migrants in the Kigoma region. These young men flee poverty and a lack of opportunities in Burundi. If they are lucky, they are hired by older Burundians to work on fishing boats, where they work all night for three weeks and have one week off. They explained that they feel safe on the water, where they are not exposed to police stealing their money. Some do not have a place to sleep but say it is an advantage being a fisherman because they are out on the lake at night and during the day they can just rest outside in the open. The little money that they earn, they send to Burundi. They need to send the money in order not to have it taken by the police. In the meanwhile, they hide the money between two pairs of underpants (Fieldnotes, 6 April 2023).

The above stories illustrate the different ways in which refugees and migrants in Kigoma try to integrate and survive and how this depends on their time of arrival. The Burundians who arrived in 1972 have been able to settle and become Tanzanians to a large degree, due to a more open refugee policy in the 1970s and 1980s, even if they still feel as outsiders in some aspects of life. The Congolese and Burundians who arrived in the 1990s are more precariously positioned in terms of formal rights. However, through hard work and manoeuvring, some manage to establish a foothold—through intermarriage, through business relations and through becoming pastors in the local community. They remain vulnerable, however, to exposure by envious neighbours, business partners and others who see an interest in reporting them. In other words, the relations that they build with Tanzanians are at once essential for their survival and what makes them vulnerable. Finally, Burundians and Congolese who have arrived in recent years live even more vulnerable lives, constantly having to navigate police harassment and locals telling on them. They do their best to remain invisible. Meanwhile, in order to pursue livelihoods and possible futures elsewhere, they travel back and forth between the camp and the city, exposing themselves to the police. Others, like the desperately poor fishermen arriving from Burundi, prefer to remain invisible on the lake but have to rest on land, where they often simply sleep outside on the ground.

The generations that we have analysed here are to be considered in Mannheim’s sense of cohorts rather than age sets, as it is the hosting environment at time of arrival that is important to the migrant experience. Although the cohorts are not clearly delineated, we may distinguish between the Burundians who arrived in the 1970s during Tanzania’s open-door policy and Burundian and Congolese refugees who arrived in the 1990s after the policy shift. Finally, the latest cohort has arrived since 2015, when the government of Tanzania adopted an active policy to rid the country of refugees. Hence, although Abdul and the fishermen in Kibilisi might be the same age, they belong to different cohorts, as he arrived almost 20 years later than they. And although Abdul and Hussein belong to different age groups, they both have a long experience in Tanzania. Importantly, they are able to use this experience in their relationships with the cohorts of migrants who have arrived later.

6. Care across Generations

We found that members of the older cohorts were keen to help newcomers, arguing that they must help because they know what it is like to be a refugee. We also found that Burundians tend to assist Burundians, while Congolese mostly help Congolese. This was in part explained in terms of ethnic/national relatedness, expressed in terms like ‘All Burundians are my family’. It was also noted that it was simply easier to identify someone from your own country. Most often, however, the will to help would be explained by the fact that ‘we have also suffered’. In the latter case, Congolese would sometimes help Burundians and vice versa. A Congolese man explains why he assists: ‘The stigma we face outside the camp pushes us to feel that no one else should suffer whatever injustices we go through’. He gives an example of a fellow Congolese refugee who had been apprehended by migration officers and had reached out for help.

That night, despite the risk of being apprehended myself, I visited him in jail and brought food. It is to commit. It requires a strong heart. I helped out of compassion and because he was a refugee and a Congolese. They (the police) could have asked me how we met. I wouldn’t know how to explain our acquaintance. If you are unlucky, you are also caught. But I took heart, and luckily, I was not caught. I met someone I was familiar with, and I gave him the food, and I left.

(6 May 2023, Kigoma)

He explains how it takes commitment, a ‘strong heart’ and luck to overcome these risks. But he is ready to run this risk for someone with whom he feels related—due in part to a shared origin and to a shared experience of exclusion.

One of the major ways that migrants with papers—and with other claims to practical citizenship—can assist migrants in precarious situations is to help them avoid the authorities. A Burundian woman explained:

Burundians who came in 1972 and are now naturalised “have uraia”. They are well integrated into society, and when they meet us (fellow Burundian refugees/newcomers) and recognise us as their countrymen or as people who share similar migration backgrounds, they are usually willing to offer help. Similarly, integrated Congolese (mostly Wanyema and Babembe tribes) who arrived in TZ in 1969 and now have Tanzanian Nationality often extend a helping hand to Congolese newcomers in times of need and trouble. […] Because they have been here since 1972, you find that their children who are Tanzanians are working or have jobs in government institutions. […] they can maybe vouch for us by pretending that we are one of their children just coming from Burundi or Congo, and they then request that we get released.

(1 April 2023, Grano beach)

The care that is given across cohorts is couched in terms of ethnic solidarity and presented as kinship to the Tanzanian authorities, as the older generation ‘pretend’ that they are the parents of children visiting from Congo or Burundi.

Hussein proudly proclaimed that he only employs Burundians on his boat, despite the risks involved. The immediate reasons given are solidarity with co-nationals and because he also knows what it is like to suffer as an excluded outsider. He has been subject to suspicion by Tanzanians all his life and therefore employs Burundians to spite the Tanzanian authorities. However, in order to do so, he has to work on his Tanzanian relationships. For instance, he maintains a good relationship with the local Mwanangambo (local security officer) by regularly giving him fish as gifts, so that he may turn a blind eye. As an established Burundian/Tanzanian, he is playing on the ambiguity of ‘tense sociality’ with Tanzanian neighbours in order to assist his ‘Burundian brothers’.

Cole and Groes argue that affective circuits “emerge from people’s efforts to reposition themselves as nodal centers in wider networks of exchange, a process that is closely tied to migrants’ efforts to achieve valued forms of personhood”. (Cole and Groes 2016, p. 9). While Hussein might not be related to the fishermen by kin, his actions are similar to those mentioned by Cole and Groes. In other words, he creates affective circuits akin to kinship across the cohort generations, and we argue that this is due to the hostile bordering practices of the Tanzanian state. Hussein, we might argue, emerges as the benevolent and powerful protector of the newly arrived refugees, which in many ways is the figure of the father.

Claude was born in Burundi but has lived much of his life in Tanzania. He drives a bajaji to provide for himself, his wife and their young child. Although he is often harassed by the authorities, he knows the city well and has found strategies to successfully navigate the city. One evening, he spotted five young men who were obviously from the camp arriving outside the city. He knew that they would be stopped at the next police checkpoint if they continued, so he picked them up and offered to let them stay the night at his place before helping them safely into the city the following morning. He did not know the men, and although they were Hutu, like almost all the refugees in the camps, and he quite easily can be detected as Tutsi, he ran the risk of having them in his house all night. Claude seemed to gain nothing from assisting these young men, whom he never has met since. We witnessed many such everyday humanitarian acts that best can be understood as expressions of solidarity—either as fellow human beings or because they themselves have been through similar ordeals and might themselves have been helped by others.

We may see Claude’s acts as acts of social kinning beyond legal and co-sanguine kinship. We may further see his actions as an act of citizenship (Bloch 2022, p. 860) where he performs citizenship by making such affective attachments (Fortier 2016). Having practical citizenship himself, he is able to extend this to his co-patriots through such acts of helping, in this sense also defying the exclusionary politics of the Tanzanian state.

Just outside Kigoma, we visited a small church on a number of occasions. On a fenced plot where cassava and other crops were growing, there was a simple church building and a few other small buildings made of sunbaked mud bricks. Here, it turned out, a number of Congolese refugees were lodged in cramped rooms. The pastor, a Congolese who has lived in Tanzania since the 1990s and who arrives at the church in a four-wheel drive car, started this church, and he takes care of the poor refugees living on the compound. The members of the Congolese congregation and the refugees who were being assisted conflated the church and the pastor in their gratitude for the assistance that they were receiving.

We are grateful to our pastor. Even if you are incarcerated, our pastor will go to the police and claim you as his own child. He won’t just say you are a church member; he will simply say, ‘He is my child’. […] We appreciate him. If you have any issues, he won’t leave you without help. And if you have a child who is suffering, he will assist you. Everyone here calls him “Baba” and “Father”.

(Jafari, 4 April 2023)

The kinning practices are quite explicit in this case. The pastor’s benevolence towards less fortunate Congolese is interpreted literally as a father helping his children. In Burundi and the DRC, pastors and other ‘big men’ often take on the role of a ‘patron’, which is parental and paternal. Our point here is that such positions as ‘fathers’ emerge and are created in the migrant context. The pastor—like Abdul and Hussein—uses his position in the Tanzanian community to assist the new arrivals and protect them from the Tanzanian authorities. These key positions—with a foot in Tanzanian society and a foot in the migrant community—make them pivotal for the newcomers.

Christine, a young Burundian woman who had lived in Tanzania from 2005 to 2012 and again from 2015, explained why and how she helps:

[…] We help. For instance, we lodge newcomers […]. We are from the same family […] Here we are all strangers […]. It is risky, but we do it in hiding. You take him/her in and tell them to hide like you do. You help them learn the language and give them other streetwise information.

(Christine, 2 April 2023, Grano Beach)

The fact that they are all refugees makes them family, and hence she is obliged to help—as one is with family. The main task, she claims, is to teach them to ‘hide’ or be invisible in the right way. Invisibility is achieved through learning to be streetwise—a skill that can be taught. Teaching newcomers to be streetwise is risky, however. Several of our interlocutors mentioned that some migrants act too much like Congolese or Burundians and therefore stand out in public, and the visibility of the newcomers would make the older cohorts visible to scrutiny, as well. Dieudonné, a young Congolese man, born in Tanzania, explained that he helps lots of refugees, but he has refused to help Congolese who look and act too Congolese (4 April 2023). In other words, appearing to be non-Tanzanian jeopardises the security of the newcomers but also those who help, and while you might want to teach them to be ‘streetwise’, as the young, Burundian woman above says, there are times when it is best to withdraw.

The dangers of being associated with later generations of refugees could also lead to suspicion and direct animosity. We mentioned earlier that it was common knowledge in Kigoma that the Congolese were perceived as less problematic than the Burundians. However, it was also explained by the fact that the Burundians had experienced so much violence, and that these traumatic experiences had hardened them and given them a ‘heart of stone’. Interestingly, many of the older Burundian refugees shared this understanding. Thus, the Burundians who had arrived in 1972 claimed to be peaceful like Tanzanians, while the Burundians who had lived through the decade-long civil war in the 1990s were perceived to be hardened and violent.

During our first visit to the fishing community in Katonga in 2022, they were very keen to show us photos on their phones of what seemed to be dead bodies and a dozen outboard engines. They explained that ‘thieves’ had come from Burundi at night and stolen outboard engines but had been intercepted and shot by the Tanzanian coast guard. We were surprised how keen they were to show us these images and tell the story. Perhaps it was not just affirmation that many newcomers have bad intentions. Perhaps it was also a way to demonstrate to us—the first author and a Tanzanian research assistant—that they indeed did not support such individuals.

In sum, older cohorts assist newcomers in a number of ways, from food, shelter and jobs to helping them navigate the hostile environment. These relations may be seen as a kind of ‘kinning’, made through practices of care and solidarity. Kinship relations are forged when members of the older cohort act as parents for the younger cohort, providing them with food, shelter and protection. We found that both men and women engage in these relations. Men with positions of power—such as the Congolese pastors—can provide protection and become ‘patrons’, while women as well as men provide safety and protection. They will often go to great lengths to establish these affective circuits of goods, favours and protection. However, there is always a risk when assisting, and at times the newcomers pose too much of a danger to be part of the family.

7. Generations Upset

So far, we have discussed generations as cohorts and acts of kinning. In this final section, we explore how generations as an age set and as a kinship relation change in the Tanzanian context. The majority of our respondents explained that younger people tended to leave the camps and go to Kigoma to ‘search for life’, leaving their families behind. These families could consist of parents and siblings, or in some cases, their birth families, as well as their own families if they were married.

Malonga came to Tanzania from the DRC with his parents when he was 4 years old in 1996. He is still registered in the refugee camp but drives a bajaji in Kigoma town. When we met him, he was married and had a child. For him, being in Kigoma is to be able to help his parents, two siblings, wife, and one child.

That is why you see many youths like me scatter (go in different directions) [...]in the camps there are no programmes that help the youth to expand their horizons, think and develop themselves. […] so, we go out […]

He talks about ‘youths’ as an age set in the sense that Africanist scholars have studied them (Christiansen et al. 2006; Cole 2004; Honwana and de Boeck 2005; Turner 2004; Vigh 2006). However, youths are not just any age set: youths are in transition; ‘becoming rather than being’ (Christiansen et al. 2006). In the words of Malonga, youths need to ‘expand their horizons, think and develop themselves’. The camp does not offer such possibilities. It provides protection, health and food, but it offers no opportunities to become someone. In order to become someone and create a future, they need to leave the camps and try their luck in the city (see also Turner 2015, 2016).

However, the youths who leave the camp are not simply an age set. Rather, they relate to other generations in terms of kinship relations and reciprocity. In other words, even though they leave the camps in search of a future, they maintain close links to older generations in the camp. As Malonga explains,

We have family members whom we need to take care of; we have kids, and we must go out to find “Mboga” (vegetables) to balance our diet. We do our best to take care of our children and families […] We go out because it is needed, and it is our way of surviving […].

(Malonga, 31 March 2023, our hotel)

Leaving the camp to ‘expand his horizons’ is not in contradiction with assisting his family in the camp. On the contrary, it is expected that he leaves and seeks his fortune in order to assist his family.

A Burundian young woman explained that as the young person in her family, she was the one who moved to Kigoma to find a job, and she often sent money to her family in the camp. Another Congolese man explains,

I came to Tanzania with my family and we were welcomed in Nyarugusu refugee camp. […] As I grew older, I decided to leave the camp. Here in Kigoma, you can do whatever you want. The problem or troubles come when one is caught by migration authorities and identified as a refugee outside the camp. […] In Kigoma I am alone. My parents stayed in the camp. I am married and have children who are with my wife in the camp too. I send them any money I make, even if it’s just 200 shillings, to help with their needs. My family consists of my mother and five siblings. My sister and I also have children, which brings the total to 10.

(4 April 2023, Grano Beach)

As a response to the ruptures caused by war, flight and life in a refugee camp, families do not disintegrate but re-group. Intergenerational relations help refugees to be able to keep their future options open in precarious situations. Members of the young adult generation leave the camps to find incomes in the city, while their parents and their infant children remain in the camp. In the camp, they are provided for by the youth and protected by the UNHCR against police harassment. They also play an important role in the families’ strategies to apply for resettlement in the USA, Canada or Australia. In this sense, different generations within families play different roles in relation to strategies of survival and strategies for creating better futures.

8. Conclusions

We originally pondered how Burundian and Congolese migrants and refugees deal with the rupture of entering Tanzania, the hostile state policies and suspicion in the host populations. We found that different cohorts have had different experiences, due to shifts in Tanzanian refugee policies. Those who arrived in the 1970s experienced a more welcoming policy, allowing them to establish themselves as Tanzanians, while those who arrived in the 1990s have been struggling to establish some sort of practical citizenship. It appeared that even the most established refugees, however, still feel that they are not fully respected and fear their neighbours turning on them. The most recent refugees and migrants live very precarious lives, constantly in fear of their neighbours and authorities. This does not mean that they do not engage with their Tanzanian neighbours. On the contrary; they need to engage with Tanzanians for livelihoods and protection. However, they are constantly vigilant and suspicious and feel that they need to hide certain aspects of their identity.

We also found that older cohorts assist younger cohorts, giving them employment and shelter or protecting them from the authorities. Their motivations for engaging with the later cohorts—despite the dangers—emerge from ideas of either shared experiences of being excluded or shared national origins. These practices of care can be seen as kinning, where a parent–child relationship is created between the first and second cohort—irrespective of biological age. In some cases, kinship metaphors are used to describe national solidarity, where fellow Burundians or Congolese are seen as brothers or sisters. In other cases, the person who helps is referred to as a father. The members of the older cohorts use their positions as ‘semi’ insiders to assist the later cohorts, thereby creating ‘affective circuits’ in the manner described by Cole and Groes (2016) in their studies of transnational kinship relations. The most valuable asset that the older cohorts can offer the younger is protection—from the authorities and from exploitation. This affective flow of protection is central to kinship relations, placing the protector in the position of the parent/father.

In other words, we have argued that it can be productive to bring together the different understandings of generations, as we found that generations as cohorts can transform into generations as kin in situations of rupture and adversity.

While the hostile environment in Tanzania creates solidarity amongst co-nationals who have common experiences and common origins, it also creates tensions between cohorts. Members of the old cohort fear being associated with newcomers, as the latter would give away their origins. It is vital for them to remain invisible to the authorities—and being associated with noisy Congolese or peasant-like Burundians could blow their cover. Hence, they would constantly balance between wanting to help and wanting to distance themselves from the newly arrived refugees.

Finally, we explored the changing relationships between generations in families, where the youth often provide for elders. While this might seem like turning the relationships upside down, they are in fact cast in the classical narrative of youths leaving the village in search of opportunities. In these narratives, the youth are believed to be able to stand the dangers of the hostile environment, while potentially reaping the fruits that the city has to offer.

We further found that the families are investing in several parallel strategies, and that each generation contributes to these. While the youth are investing in life in Tanzania, their parents are keeping a gateway open for a life elsewhere.

As long as the Tanzanian state maintains its hostile policies against migrants and refugees, and as long as surveillance of the population remains so strong, the tense sociality between hosts and migrants will create the need for migrants to assist co-nationals, creating and reproducing kinship practices. Similarly, the nature of the camps will produce kinship patterns where the youth leave the camps, while the young children and grandparents remain the camps.