Berg-en-See street boys: merging street and family relations in Cape Town, South Africa

Abstract

Despite a wealth of research exploring street children's lives, this has tended to focus on the micro-scale, rarely drawing connections with wider society. Yet, it is rare for street children to sever all ties with home and this paper explores these connections by taking a relational approach to the production of street life. Drawing on in-depth qualitative research with 12 boys living on the streets in a coastal suburb of Cape Town, the paper identifies that street children are part of powerful inter- and intra-generational relations that connect them to their families: interdependent but sometimes forced and contested. The paper concludes by identifying that street children are not isolated on the street, but rather positioned relationally in between street and family life building relations within and across spatial boundaries. This has implications for the way in which we conceptualise street children's lives and adds to wider theoretical understandings of childhood as relational.

Keywords:

• family
• gangs
• relational perspective
• siblings
• South Africa
• street children

Introduction

Street children's situation has received much international policy-related attention since the 1980s and 1990s in parallel with the emergence of the ‘new’ social studies of childhood, which argued for a renewed focus on children's lives (see Tisdall and Punch 2012 for a comprehensive discussion), and the children's rights agenda. This shift in conceptual thinking about childhood resulted in a body of work, particularly in the UK (see Mayall 2012), highlighting children's social agency, celebrating their competencies and shaping an agenda for giving children ‘a voice’ in research. Research with street children has followed a similar path, the increase in research related to their lives producing studies that advance our understanding of the intricacies of street life. This includes work investigating street lifestyles (Aptekar 1988, Panter-Brick et al. 1996, Rizzini and Butler 2003); reasons for being on the streets (le Roux and Smith 1998, Matchinda 1999, Conticini and Hulme 2006); survival strategies (Hecht 1998, Lugalla and Mbwambo 1999, Sherman et al. 2005) and sub-cultures and identity (Beazley 2000a, 2002, 2003, van Blerk 2005, Herrera et al. 2009).

However, this focus on children's agency has often been narrowly applied, exploring children's competencies and contribution to social life, without offering the same attention to their socio-economic and political contexts, and perhaps to the detriment of positioning children within wider academic debates (Vanderbeck 2008). As Tisdall and Punch (2012) highlight, children's agency should be theorised more critically, as not always positive and often constrained by the influences and perspectives of others so as not to limit the wider field of power within which children's lives are lived (Ansell 2009). Recent critical theoretical reflection means that childhood studies are exploring children's lives in context: situating their lives within large-scale socio-economic processes (Horton and Kraftl 2006, Tisdall and Punch 2012), and drawing on sociological work on families (Jamieson and Milne 2012). Such a perspective understands children as relational and connected to wider influences in society (Hopkins and Pain 2007, Jamieson and Milne 2012). This is not to undermine the position of children as social agents but rather seeks to explore their everyday lives as part of friendships, families, communities and wider societal networks (Punch 2002, 2007, Hopkins and Pain 2007).

This is especially pertinent for investigating street children's lives, where work has generally taken place at the micro-scale, often focusing on aspects of street life that are considered deviant, and rarely drawing connections between street children's lives and wider society, including their interactions with family and community. Where this has occurred, research has focused on the breakdown of family ties and the problematic nature of street children's community relationships or on substitute ‘street’ families (Beazley 2000b, 2003, Hansson 2003, Rizzini and Butler 2003). Despite this, it is rare for street children to sever all ties with home as a few studies allude to (Jones 1997). Exceptionally, Hienonen (2000) has explored how street-working children have different, and sometimes strong, associations with their families based on differences in child-rearing practices, ethnicity and gender, and Ennew (1994) discussed street-working children's reciprocal relations with, and provision for, family members, as they move between the home and the street on an almost daily basis. In addition, there is evidence that homeless street-living children also retain connections with their biological families and ancestral communities but this is not well researched (Wells 2009). Beyond discussion of street(-living) children's1 understandings of ‘home’ (Beazley 2000b, Evans 2004, Ursin 2011), research has yet to fully explore their relations with family members and consider the implications for their fluid street and family lives.

This paper focuses on research with street children and their families in Cape Town, South Africa and is concerned with the ways in which street life and family life are integrated. The paper begins by highlighting the methodological approach before going on to outline street children's position within conceptualisations of family and what this means to them. The paper then explores the ways in which their lives are shaped by particular powerful relational processes which fluidly cross the spatial boundaries of street and home, resulting in family relations that are both connected and contested. The paper concludes by discussing the implications of this for the ways in which we theorise street children's lives as part of childhood studies and offers specific recommendations for policy and practice.

Methodological approach

A street researcher methodological approach (Bemak 1996) was employed in the research, integrating ethnographic and participatory research with the underlying philosophy of ‘novice researcher’ and ‘participant expert’. This participatory ethnography comprised biographical interviews, focus groups and participatory activities, which was supplemented with family and key informant interviews.2 This paper draws on the data produced working with one particular group of 12 boys,3 aged between 11 and 18, and living on the streets in Berg-en-See.

Aside from all the standard procedures for undertaking research with children that are outlined and discussed widely in the literature (Valentine 1999, van Blerk and Barker 2008, Beazley et al. 2009, Alderson and Morrow 2011), there are particular considerations when working with street children (Young and Barrett 2001). Anonymity and confidentiality must be carefully managed for street children as reporting what they are doing, where and with whom, could result in harm, particularly if activities are considered illegal. In this instance, I worked on the principle of doing no harm to participants but encouraged some who had experienced trauma to seek additional support from non-governmental organisation (NGO) staff or engage in de-briefing with the project psychologist. In addition, the issue of consent was constantly re-negotiated on various levels. Verbal informed consent was gained from all the boys, who regularly turned up to meetings and opted into the research process, and from organisations that they frequented, acting in loco parentis. Informed consent was also obtained from families and all the boys gave prior permission for their parents to be interviewed.

Positioning street children within conceptualisations of family relations

The childhood studies literature, which has framed the way in which children's lives have been researched, has mainly emerged from the Minority World, yet much can be gleaned about the wider relationality of children's lives from the Majority World context. Here traditional/cultural conceptions of family and belonging may rather be centred on maintaining and supporting kinship relations and networks stretched across time and space (Chant 1997, Ansell and van Blerk 2004, Payne 2009). For example, in southern Africa, families are fluid and significantly divergent from the western-centric view of family as a nuclear arrangement. Various arrangements of extended family and wider kinship networks are incorporated into understandings of family as units of economic survival, households, and incorporating patrilineal and matrilineal diversities (Foster 1997, Nyambedha et al. 2003). Family living is often multi-generational as young couples remain in the patrilineal homestead and grandchildren traditionally grow up in the proximity of their grandparents (Barbarin and Richter 2001, Russell 2003). This household/family nexus is based on mutual and/or powerful obligations to support each other through, for example, economic, education or labour transfers and is fluid, incorporating different members of the extended family in different times and places (Ansell and van Blerk 2004).

Key relations in children's lives then are also often intergenerational, based on obligations ‘or contracts’ of care (Kabeer 2000, van Blerk and Ansell 2007). Children may engage in, or receive, care and education and may work to support the household both externally and from within (Izzard 1985, Townsend 1997). Yet, the expectations placed on different family members are dynamic and may change for a number of reasons including traditional/cultural expectation, financial or social status and proximity. For example, children in southern Africa often move between households of extended family members, sometimes travelling long distances, to provide support or be supported (Bray 2003, Ansell and van Blerk 2004). Increasingly, this is encased within a variety of social and economic contexts such as the impacts of the AIDS pandemic, through which the nature, form and duration of extended family relations involving children has diversified (van Blerk and Ansell 2007, see also Payne 2012). Similarly, Punch (2002, 2007) explores the connectedness between Bolivian young people and their families through interdependencies that take place during periods of work away from home in neighbouring Argentina. Family relations can therefore be important for children even if they are distantly located to the Minority World conception of the ‘family home’. This has some useful messages for the way we investigate street children's lives.

Street children have always been positioned in relation to their families and communities but rather than focusing on the importance of such connections, research has tended to separate them out as different (Ennew and Swart-Kruger 2003). Minority World constructions of social organisation view the nuclear family unit as superior where children, as passive recipients, receive nurture, socialisation and protection from the corrupting influences of (adult) society. Childhood is portrayed as a period of innocence and children who deviate from this are considered ‘wicked’ (Valentine 1996). This popular dualism of childhood has influenced conceptualisations of street children in Majority World societies. As Ennew and Swart-Kruger (2003, p. 3) point out:

… morally-powerful social constructions of family, home, domesticity, and childhood could not exist without the construction of the ‘other’ – the danger of the street, the amorality of street life and above all, ‘street children’ who are outside the domestic sphere and challenge the order of social existence.

This has resulted in a widely held view that the breakdown of the (nuclear) family, stemming from conditions of poverty and un-/under-employment, has fuelled problematic lifestyles including drunkenness, violence and prostitution, as the principal reason for the emergence of street children (le Roux 1996, Williams 1996, Lugalla and Mbwambo 1999). Their position in public space depicts them as ‘isolated, cut off from social networks, alone in the world’ (Wells 2009, p. 80). In South Africa, the burgeoning research interest in addressing the impacts of AIDS has (problematically) supported this notion, where extended family networks are also being seen to breakdown (Bray 2003). Yet, Wells (2009) notes that this view is generated more by the image of the street as ‘other’ and its oppositional relation to home than by street children's actual everyday lives. In fact, there is evidence to suggest that street children do engage in lives that are highly relational and interdependent. For example, Richter (1991) suspects that street children are rather a product of the need for individuals to ‘pull together’ in situations of poverty, where they engage in income-generation to support the family to meet its basic needs, illustrating the possibility of close connections between street children and their families.

This assertion that street children are connected to others suggests that although our understanding of their lives has benefited immensely from the centrality of children's agency in childhood studies giving street children prominence in research and policy agendas, it is important to examine more closely their complex relationalities for fully understanding the production of street life. In taking a relational perspective to childhood studies, research must move beyond a narrow focus on street children's competencies in specific spaces and explore how children's lives are shaped through their connections to others. Jamieson and Milne (2012) further argue for taking an inclusive approach to relations to include those in their wider networks. This suggests that by subsuming children into families, we somehow miss out on the importance of children's other relations with, for example, peers and non-related adults. Yet, we must also continue to investigate the intricacies of family relations across contexts for fully understanding the ways they affect the choices children make. Street children provide a useful example through which to demonstrate this point. Research has connected street children in distinct ways to non-kin through their public existence on the street but we know little about how family relations impact on their lives. This paper draws on the conceptualisation of African families as fluid and stretched over space to explore the complex positive and negative ways family relations shape street life, despite street children's location outside the household.

Constructing street children's family relationships in Berg-en-See

In the Western Cape, the concept of the Rainbow Nation is heightened as the mix of racial and ethnic groupings in and around Cape Town are tri-fold with significant (formally – classified), White, Black and Coloured population groups, all with different nuanced understandings of family relations. The formerly Black and Coloured classified townships established mainly in the 1960s and 1970s housed those who were forcibly removed from the urban areas pronounced White by the Group Areas Act4 (Salo 2009), and halted those migrating in from the rural areas to access the city. Neo-liberal economic reform implemented in the post-Apartheid era (entitled growth, employment and redistribution), that transcends racial boundaries, although offering free market access and opportunity to the entrepreneurial New South Africans, for most Black African and Coloured residents this is not the case. Those who lived and grew up on the Cape Flats during the apartheid era, and their families, are generally still located there, experiencing the same spatial and economic marginality in relation to white populations (Lemanski 2004): the outcome has been deepening levels of impoverishment (Salo 2009).

Family relations in the formerly Coloured townships, despite having origins in the creolised religious and cultural backgrounds of the Cape Malay population, are based on current conceptualisations of respectability present in the Christian ideal of monogamous marriage and nuclear family units (Salo 2009). Yet, this is not void of connections to other networks that reach across blood and friendship lines, where the relative stability of the communities has produced dense extended family networks (Moses 2006). Despite this, communities and families are plagued with social problems including drugs, alcohol abuse, gangsterism, family breakups and violence. Shelmerdine (2006) highlights the ways in which such problematic issues permeate young people's relationships: a key aspect of which is tensions between young people seeking independence and parents attempting to protect them from engaging in behaviours such as dropping out of school or taking drugs. As both Shelmerdine (2006) and Salo (2009) point out, parental/guardian attempts to control the activities of young people becomes an integral part and purpose of their relations. Where significant adults are unable to engage positively with their children through fending off poverty with working long hours or because of drinking, violence and drug taking, other adults sometimes replace parents including grandparents, neighbours and even older siblings (Shelmerdine 2006). These relations are based on processes of protection and control, and display significant lines of power from adults to children within the family.

White Sands and Cosmos are typical of these Coloured townships. They are both densely populated communities whose residents are faced with problems related to poverty including unemployment, family breakdown, alcoholism and drugs. There is a high degree of gang culture in these locations, and they are home to one of Cape Town's notorious gangs. The proximal town of Berg-en-See is a formerly White area based several kilometres out of the city centre. It once was a popular holiday destination, with one of the city's best beaches. The area has become a little run down in recent years but is now going through a regeneration phase and still attracts numerous visitors at weekends and holidays. The street boys emerging from White Sands and Cosmos position themselves here.5

In discussions around ‘understanding family’, the boys highlighted the importance of protection in family relations; both intergenerational, between parents and children, but also intragenerational, between fathers and mothers (and between siblings as the next section shows), illustrating that the locus of power is held with the male adult household head:

Kevin (15)6: It's like a father to a mother, he will protect her so he will protect the children as well

Sker (16): Like if a father has made a bond with the mother that he will provide for the children and so he has made a bond with us, he will provide for us.

The complexity of community problems and relations mean that in many cases the protective aspect of parenting the boys highlighted had broken down, either because of changing family structures in the household such as step-parenting, absent parents or ‘missing’ parenting (due to drugs, alcohol and violence) and so they supplemented their family relationships with similar structures on the streets (as exemplified by Kevin, Kyle and Martin).

Kevin (15): Ja they've got houses. Some of us can go home, and you can … like the last time that I was at the house I worked for a merchant. Now my stepmother, my real mother is dead, it's almost as if she wants to go away from my father and she's after my father's money. Now she makes cases against us but I don't know what I've done. She's cunning with me, and took a case against my father as well …

Kyle (11): I don't want to go home often. It's not nice there. They [mother and step-father] hit me too much

Martin (15): They lock him in the toilet

Kyle (11): And they drink too much

When asked to provide concrete examples of their own families, the boys simultaneously talked about their biological family in the community and their ‘family’ on the street:

Arturo (12): My family is at home: my mother and father and sisters … but this is my family sitting here too.

Martin (15): Family's on the street. My brothers [biological siblings] and my friends are on the street.

Although they merge ‘home’ families and ‘street’ families in their accounts of family, these relations are also infused with power. It is evident from the discussions outlined that the boys do not respond well to the powerful relations they are part of at home (for example, where Kyle is beaten and locked in the toilet). These power-infused relations between adults and children are resonant of Foucault's (1976) disciplinary power. Here, intergenerational power is exercised within the space of the home to produce docile bodies (Foucualt 1995) – those that can be controlled (as noted by Shelmerdine 2006 in discussing parent–child interactions in township communities), and illustrates the negative impacts on children when this breaks down. Yet, children are subject to powerful family relations not only within the home but also outside of it.

It is helpful here to turn to a more nuanced Foucauldian notion of power for understanding how this influences street children's relations. By combining the notion of disciplinary power with that of governmentality, we emerge with the following: ‘A society is not a unitary body in which one power and one power only exercises itself, but in reality it is a juxtaposition, a liaising, a coordination, a hierarchy, too, of different powers which nonetheless retain their specificity’ (Foucault 1976, p. 156). Here, we can perceive a community of entangled power relations (Sharp et al. 2000) that cut across different attempts to influence street children's lives. Garmany (2010) demonstrates this well through a discussion of a favela community in Brazil, where despite very little state control over the population, particularly with reference to law enforcement, there are varying networks of social power existing in the community through a variety of smaller communities based on religion, long-term residents and criminal cultures, all of which influence the running of the slum community through their own hierarchies and surveillance.

This is easily transferable to the lived context of Berg-en-See. In street children's lives, there are several ‘communities’ based on varying relations that have their own power hierarchies, and which are played out within and across the micro-spaces of their daily lives. These include: intergenerational relations in the home where there is the controlling (or disciplining) power of parents who are seeking particular pathways for their children, powerful gang relations in the wider community (which street children are connected to in various ways not least through the involvement of other family members) and street relations; all of which are played out in the spaces where street children operate. The following section explores in more detail the ways in which the production of street life is relational. Although mention is made of other relations that are important in street children's lives (such as with community members, police, security, NGOs), the intentional focus on family relations seeks to explore the special influences of family relationality in the creation of self identities on the streets and contributes more broadly to a relational understanding of childhood.

The production of street and family life

The spatial proximity of the streets in Berg-en-See, to the communities of White Sands and Cosmos, means that children's street and family lives are particularly fluid, providing good evidence for the ways in which street children live their lives relationally. The varying spaces in their daily street lives – the main streets, back alleys, family homes and gang territories – are all infused with various networks and relations of power that together produce street life.

The context in which street life is lived presents the boys’ micro-geographies as fairly localised7 – they are mainly staying in and around Berg-en-See, White Sands and Cosmos, occasionally travelling to other nearby seaside locations when there is a market or to look for things to steal. They access income by begging, doing small tasks such as running errands and washing cars for people working around Berg-en-See and by stealing copper pipes and water tanks to sell as scrap in White Sands. One of the boys mentioned that some are ‘taking drives with uncles’: a pseudonym for transactional sex. They have distant relationships with the security police, mainly seeking to avoid too much contact and will therefore stay out of sight for most of the daylight hours. The boys also sometimes go home to eat and change their clothes but they mostly sleep rough in the central park area of the town. The following extracts illustrate some of these activities as part of the production of street life, but also show that these activities are not undertaken in isolation: as Kyle notes, sometimes the people they work for attempt to support the boys with money for food and advice, while Booti points out that it is important for the boys to trust each other.

Kyle (11): We do begging and stealing … Like the one guy says ‘Here's a R10 go buy something to eat’ … we get times to talk with them. There's times when they in a hurry then they talk to us next to the road. And they ask us did you have something to eat today, are you hungry or what?

Booti (14): You just do this with the burglar bars … Then we all go in … Off with the door, all of us go in then we just take the geyser, dvd, copper pipes … we take friends with us to look for the police and help carry the stuff … someone that can stick like one with you [someone that can be trusted]

The boys also talked at length about drugs as part of street life. They mentioned they mostly drink pap sak (cheap wine) and smoke dagga (marijuana), buttons and tik (methamphetamines), which they buy from Nigerian drug dealers in Berg-en-See or locals in White Sands. They also pointed out that they might act as go-betweens for students and tourists looking for dagga, forming part of a relational network between the street and the township, fluidly crossing spatial boundaries. Evenings are usually spent playing video games in the local shop and later smoking and drinking while they relax. The following discussion demonstrates how the boys develop relations with drug dealers in White Sands, sometimes combining visits home with buying dagga:

LvB: Where do you go to get dagga?

Mikey (16): We go to White Sands

LvB: Do you always get it from the same person?

Mikey (16): Ja, sometimes we buy it from the same person

Sker (16): He's got a lot of bodyguards, but I just stay a few houses away and sometimes I just go in there and I buy the stuff then I come out

LvB: So how do you feel about the people that you buy from?

Mikey (16): They like friends, ja, but …

Sker (16): We just come and we go and they give and we take. We just come there and go, and give and you take it and then you're gone again.

Martin (15): Sometimes we go fetch for free, Ja, for regular customers, and maybe clients from here, white people, then we take them there and get a discount.

This brief illustration of the lives of Berg-en-See street boys not only highlights the dynamic nature of street life, and the importance of street children's agency through the unique and resourceful ways in which they seek survival on the streets, but also presents street life as relational: merging home/community and street spaces. Further investigation revealed that these relations are highly complex, providing insights into how we theorise children's relationships with family members. In particular, the data revealed two important ways in which families influence the production of street life: through connected and contested relations, demonstrating powerful connections that children respond to in divergent and dynamic ways.

Connected relations: merging street and family life

Only two of the Berg-en-See boys were unrelated to any of the others. Of the remainder, there were three family groups present within the ‘street family’. Observing the group as a whole revealed some interesting insights into how they supported and protected each other, particularly through intragenerational relations, which appeared to be important on the street.

Intragenerational ‘sibling’ relations were especially important for younger boys in the group: the youngest, Kyle (11), often behaved badly in the group, stealing from others and teasing his own age peers. Martin, his older brother, was a particularly strong character having been on the street for several years. Although there was no formally designated leader among the boys, Martin had a strong presence in the group, often controlling what others would do: for example, sometimes sending the younger boys to beg at the traffic lights for him. Here, Kyle was instrumental in helping his brother, and Kyle's friends would tag along in return for Martin's protective influence. Martin's protectiveness of his brother seemed to give Kyle respect among his peers. However, Mal (12) was regularly the brunt of jokes, and although ‘street-smart’, was no match for the older boys. He had no familial ties with any of his peers and his age and size meant he could not assert himself to any significant degree. Yet, when necessary (for example, if harassed by others outside the group such as security or gang members) the boys would support and help him. In contrast, Mikey (16) was a smart, strong young man that could not be easily defeated in a fight. Although he also had no blood relative in the group, his physical abilities and mental sharpness meant he received greater respect than Mal from the others, who might look to him for help on occasions. This suggests that having relatives on the street was especially important for younger, smaller boys. Here, the transference of protection to intragenerational relations among the group indicates this. Birth order between siblings has been shown elsewhere to play an important role in children's relationships with older siblings taking on more adult roles such as looking after younger ones as well as offering advice (Punch 2001, Payne 2009) and the street environment appears to be no different. Special significance seemed to be placed on relationships that were based on blood lines with siblings/cousins, with younger siblings rallying their friends to work for their older brother, and older brothers/cousins protecting younger ones more valiantly. This suggests that family connections on the streets offer an additional layer of support and protection for some and may provide incentive to younger siblings also taking to the streets. This tells us something about the strength of family relations and the importance of family for the creation of self identities (Jamieson and Milne 2012) where boys, who appear to be disconnected from their families, are displaying stronger obligations towards family members.

In addition to these interdependent intragenerational relations on the streets, family relations also connected street children back into their home communities: creating fluid boundaries between street and home spaces. The boys talked about going into White Sands to buy drugs and to sell copper pipes to scrap merchants, combining these trips with visits to their families. Parents often mentioned that their children would come back for visits, to eat and change into clean clothes, but that they usually only stayed for a short time. Martin and Kyle's mother talked at length about her boys coming to get clean clothes and she regularly bought second-hand clothes to store in the house for whenever they appeared. These relational processes with adults were often also interdependent with the boys receiving clean clothes and food in return for bringing small items for younger siblings or money. Martin and Kyle never go home empty handed and their mother pointed out that they give her money to buy food: He said that day, I can have R5. Bread and Polony [processed meat] and eggs  …

However, the interdependencies are sometimes forced as the boys also go home at times when they know the house is empty to steal clothes or money. For instance, Kyle's mum regularly drinks in the local shabeen8 and recently, while she was out drinking, Kyle and his friends broke into her house and stole some money: Kyle, he and his mates cleaned out my house. I went to school with the little one. Then I drink a beer, not wine, beer sometimes if I've got the money  …  When I come back the money is gone but I know it is Kyle.

Sometimes there was no particular reason for the boys to return home. When they did not need anything the core reason for their visits was to check on younger siblings. During my visit to Kyle's home, he had accompanied me and brought some shoes for his young brother. As his brother was at school, and Kyle had only wanted to visit his sibling, he chose not to stay during the interview. Similarly, Lawrence's and Booti's mothers both noted that their boys came home to visit their younger siblings and would spend time playing with them.

LvB: What makes Lawrence come to visit?

Lawrence's mother: I don't know. Lawrence says sometimes he misses the house or he misses his [baby] sister and he come and just plays with her.

Booti's mother: Booti comes here often. Booti is one that will eat with me, but he's in and then out again. Almost as if he is embarrassed.

LvB: Why do you think he comes here?

Booti's mother: It's about the baby. He comes to visit the baby.

These relational processes connect street children to their families and are based on what Salo (2009) identifies as protective relations – those that the boys also drew out as significant for family life. In this instance, the spatial proximity of street and home, where the boys could easily move between environments without transport, more readily identifies three ways in which these relations produce interdependencies, demonstrating different expressions of power through their own micro-communities (Garmany 2010). These are: intergenerational relations between parents (particularly mothers) and children; intragenerational relations between siblings on the streets; and finally, intragenerational relations between street boys and their much younger siblings. Positioning within families by age, gender and birth order must be considered in an integrated relational understanding of childhood. Just as older siblings look after younger siblings in child-headed households (Payne 2009, Payne 2012), older street boys looked after their younger siblings both on the street and at home. These intragenerational relations, in part, assume some of the status of intergenerational parent–child relations. Therefore, despite the negative aspects of home life, family relations appear important for street children's understanding of who they are; this is demonstrated by the obligations they fulfil towards younger siblings and the interdependencies they share with parents (especially mothers). Therefore, the connected relations street children have with their families are based on intra- and intergenerational relations of power that connect across family/community and street spaces.

Contested relations: street as oppositional to family

Although we have ascertained that street children's lives are very much connected to their families, they are also contested aspects of their lives. The image of the street as public space, separate from the private sphere, is important here. The visibility of the street to public institutions including drop-in centres and the police, in fact, makes the street the ideal place for many street children. Despite living lives that contrast with Minority World ideas regarding where children should be, the street offers a haven from the negative powerful relations and forced interdependencies of family and community life. This section illustrates the ways in which family power relations create interdependencies that the children seek to subvert.

Life in White Sands was very much tied into gang culture and crime. In all cases, the boys talked about how they had gone to the streets to avoid being drawn into gang life in their community:

Booti (14): In White Sands you die quick. You just stand there and then they shoot you with a gun …

Des (18): … like there in White Sands is the [Gangs]. They want all of us to turn [Gang members] too but we didn't agree to become gangsters. And it's like every time when they get us they like want to rob us, take our stuff off. But now we [are on the streets we] don't let them take our stuff off. We don't give them the power to take our stuff off now.

Crime and drug dealing is a significant part of gang culture and children, particularly if they are family members, can participate in a supportive role carrying weapons or drugs for older gang members. Therefore, family relations have important interdependencies for the continuation of gang life and for preserving the status of particular members, who may be on probation or at home awaiting trial. Arturo's mother lamented that her children are on the streets because of their father:

Arturo's mother: It's because of him, he's shouting at them … That's why they sleeping on the street, it's not me … it's because of this man. That's why my children run around … He's in and out of jail, broken service [breaking parole conditions], and we must still worry with [support] each other.

The multiple and gendered power relations within, and through, the space of the home coupled with alcohol and drug addiction, demonstrate a lack of care and protection of children, particularly where adult male members of the household encourage the boys to engage in or support gang activity. In addition, locating these relations within the home makes it difficult for young people to opt out of gang involvement and still remain in the family house. The unwritten obligations of intergenerational contracts (Kabeer 2000) suggest that children will perform tasks asked of them by their parents in return for care and protection. However, in situations where divorce and single parenthood is common or where fathers are working away or in prison, other relations can assume parent–child relationship status. In many cases, mothers’ alcohol/drug use meant that step-fathers/boyfriends and adult male children living in the house take on a power role as household head (see also Payne 2009). However, this is not in the sense of protecting and providing for the family, but rather in controlling the home and what goes on there. For example, Mal (12) lived on the streets following the death of his father when his elder (adult) brother Nico became the household head:

Mal's Grandmother: Mal's father had died and his mother was working in the Spar to make some money to buy food, there was no food in the house. But the older boy was just doing what he wanted in the house – there was no control or discipline of him … He was drinking, bringing friends and doing drugs … This had started when the father was sick – it was a drug house.

Nico came out of jail during the research. He is involved in the gangs and had been in prison for stealing. The boys told me there is someone who is looking to kill him. While on the streets Nico turned up looking for Mal. Mal did not want to go home with his brother and Nico threw a brick at Mal's head, injuring him badly. Yet, because Mal had chosen to live on the streets and not at home, he could no longer be forced to participate in gang activity. By spatially locating on the streets, he was protected by outside observers from any significant danger or coercion. The public environment of the street removes the boys from being part of the power hierarchies in their communities. On the street, a different network of social power exists (Garmany 2010): state power regulated by security and police takes precedence over gang power. The following extract demonstrates how relations are spatial, disrupting intergenerational obligations in relations with gang members, through being on the streets.

Martin (15): They shoot you with a gun. Gangsters there … you stand by your gate then they shoot you

Mikey (16): The lifestyle of the people in that White Sands is just stealing, smoking buttons, tik, drinking and and they die on a Friday night

Mikey (16): They get killed by brothers [gang members]. There's a tavern, every Friday there's a gun shooting or a knife stabbing, then someone dies

Martin (15): Every Friday. It's horrible there now. We don't want to go there every Friday, come rather to Berg-en-See

Sker (16): Here in Berg-en-See, the gangsters don't come to you and tell you, you must stand [for] this gangster and that gangster …

Family relations also connect street children to criminal cultures indirectly, where in some instances the relational network stretched by association into business/work place encounters. For example, Des's mother worked in a shabeen owned by a gang leader. Her job was to sell alcohol to customers and sometimes she was also asked to supply drugs. In an interview she talked about how she took Des with her from an early age and how this resulted in him developing his own relations with other adults in her place of work. For Des, it was his relationship with his mother that had connected him to wider criminal cultures. His desire to remove himself from the gang resulted in Des going to the streets.

Des's mother: From a young age he wandered around Berg-en-See … From 6 years old he was already on drugs.

These examples demonstrate how street children's family relations are contested producing interdependencies that are forced as well as adverse home life experiences. The varying social networks that exist in the community related to gang culture mean that children can be drawn into gang life through powerful adults in their lives. As Des's example shows, this is not always part of a direct family relationship, where children are asked to undertake gang tasks, but can be connected through family into the community. Here, Des was being drawn into gang life as part of his mother's work and the family's economic survival was linked into his continued involvement. However, we must not forget that children are social agents and key players in producing their relations. Although they may be subject to powerful control over their lives, they can resist and subvert these hierarchical relationships (Jamieson and Milne 2012) and here space is significant. Although relations fluidly cross between street and home/community spaces, intricately connecting these spaces in street children's lives, they are governed by different power hierarchies, thereby enabling street children to contest negative relations by crossing spatial boundaries between the home and the street.

Conclusion: street life as relational

The complexity of the lives of Berg-en-See street boys demonstrates a highly relational existence that is very much bound up with family power relations both on and off the street. We can no longer consider street children's daily lives in isolation or merely connected to the streets and must look beyond their immediate environments to understand the ways in which street children are connected more broadly to family and community. This paper has demonstrated that street children in Berg-en-See live their lives at least partly through connected and contested relational processes that both encourage them to leave home and to remain part of it.

By taking a relational perspective, the paper advances the way in which we conceptualise children as social agents, given that their abilities to participate in their lives are subject to powerful dynamics within families/communities. Although childhood studies has done much to reinvigorate theoretical interest in the inter-relations between children's agency and society through positioning them as competent social beings, the often over-celebration of children's competencies in UK research has marginalised interest in the way in which children are connected to others. For street children, research about children's competencies positioned from their own perspectives has brought much needed awareness to their lives and included their voices in debates about their circumstances. However, at the same time their perceived autonomy, through their decision to go to the streets, has marginalised them to a position of being ‘alone’ on the streets, disconnected from family and community (Wells 2009), and spatially located in the street. This paper has argued for a broader understanding of street children that positions them in between the street and the home and most importantly connected to others in the wider family/community nexus.

Power and protection within families (such as noted by Salo 2009 for Cape Town Coloured communities) operates fluidly, based on particular gendered, birth order and age-related networks, and results in emerging interdependencies between parents and children, and between siblings, as they seek to support each other. In some contexts, these interdependencies are contested – not desired by the children, resulting in them placing themselves in the streets, where different power relations govern societal dynamics (Garmany 2010), in order to remove themselves, at least partly, from gang activity. However, they still remain connected to their families: returning to visit, eat, change their clothes, play with and look after siblings, and bringing gifts and money. Therefore, children are connected to various micro-communities of power relations that operate in the spaces they are part of (Foucault 1976). In many cases, street children, avoiding the outcomes of particular negative relational processes at home, fulfil the need for protection through relations on the street yet returning back home to maintain positive relations with mothers and young siblings. The power relations which affect children's lives (generally where power flows from adult/older sibling to child/younger sibling), and shape the contexts in which they participate, also operate across space. Therefore, in order to fully understand children's lives and self-identities, children should be considered as socio-spatial relational agents. Their lives cannot be spatially compartmentalised into the home or the street or other location, but they must be seen as building relations within and across spatial boundaries.

Taking such a socio-spatial relational perspective also has important implications for policy and practice with street children. The current body of literature detailing street children's resourcefulness, demonstrating their ability to influence their lives has high prominence in policy circles and, when coupled with the children's rights agenda, tends to position street children as alone in the street. This supports policy and practice that advocates for family reunification, shelter provision and individual work with children. For a significant time now, this has formed ‘best practice’ in working with street children. However, this paper identifies that street children should not be considered in isolation and suggests that a more holistic approach is required. Given the emergence of ‘new’ street geographies that may push more and more children on to the streets in their local communities (van Blerk in review), street children are increasingly likely to have relational lives that merge street and family/community together.

Therefore, policy and practice must deal more effectively with the underlying problems that are at the root of their presence on the streets. At the ‘blue skies’ level, this means reducing gangsterism and drug crime in communities coupled with targeting the root causes of many family problems such as unemployment, alcohol abuse, violence and poverty. On a smaller scale, policies used for supporting street children in communities may have to be re-addressed. Although not to detract from the desirable features of returning children to live with their parents, this research suggests that the social context including wider family and community relations beyond the immediate homespace should be taken into account to ensure safe futures for reunited children. This paper shows that family reunification may simply return children to situations where they are drawn into forced interdependencies with criminal and gang cultures. Working with the whole family, offering alternative living arrangements, counselling and work placements may mitigate against children returning to the streets. The paper also suggests work directly with sibling groups, given the close ties that exist between brothers and cousins on the street. Programmes that seek to support brothers/cousins together may help to reduce the incidence of children returning to the streets following intervention. The supportive relations between street siblings may reduce the desire for them to seek out others on the streets. In addition, practitioners should look beyond the street to support the younger siblings of street children, who will become subjected to similar powerful relations between themselves and adult family members involved in drug and gang crime. Support and monitoring at a young age may also help to avoid their future involvement in the street.

Acknowledgements

This research was funded by an ESRC small grant (RES-000-22-1286-A); however, the views contained in the paper, including any shortcomings, are entirely my own. I would like to thank the editors and two anonymous referees for their helpful comments on an earlier draft of this paper. I also gratefully acknowledge the support provided by The Homestead Project for Street Children, The Centre for Social Science Research at the University of Cape Town, and the assistance of Amber Abrams, Jac Jacobs and the children who participated in the research.

Notes

The problematic nature of the term ‘street children’ is acknowledged: it locates children in the street (Hecht 1998), which is static, excluding the capacity of children to move between different social and spatial environments and it associates the negative characteristics of street environments to childhood (Conticini 2004). It is, however, used here to highlight the special circumstances of children living in the streets. This distinguishes the children in this study from street-working children, who only spend part of their time earning money on the streets. The remainder of this article refers to children working and sleeping in the streets, when using the phrase ‘street children’.

Family interviews were mostly undertaken with mothers, although sometimes fathers, siblings and other relatives were also present. On a few occasions, where children had not been previously living with their mothers, the interviews were undertaken with other guardians including grandmothers, aunts and foster carers.

At the time of the research there were no girls living on the streets in Berg-en-See.

The Group Areas Act 1950 was the legislation that initiated race-based urban segregation and resulted in the forced removal of those classified as coloured or black from urban areas between the 1960s and 1980s.

Berg-en-See became an important area of study in the research as urban governance strategies in Cape Town city centre were producing ‘new’ street geographies, positioning street children out of the city centre into less central, albeit still highly public locations (van Blerk 2012, in review).

All quotations are translated from the colloquial mix of mainly Afrikaans and English the boys used. All names in the paper, including place names, are pseudonyms.

In part, this is due to urban governance tactics to reduce street children numbers in Cape Town city centre.

A shabeen is a local informally constructed bar.