Gender-based violence: Young women’s experiences in the slums and streets of three sub-Saharan African cities

Introduction

The Millennium Development Goals, with their constellation of concerns about how to alleviate poverty, effectively set the promotion of gender equality through schooling (Goal 3) alongside the ambition to have achieved by 2020 ‘a significant improvement in the lives of at least 100 million slum dwellers’ (Goal 7). As Davis’ (2007) famous book The Planet of Slums highlights, there is now a global concern about the increasing number of people living in urban slums in conditions of extreme poverty and overcrowding. Within this agenda lie worries about the effects of such living conditions on women’s lives and livelihoods. This article places the struggle to achieve gender equality within these environmental concerns about urbanization and slum life by focusing specifically on forms of gender-based violence. The United Nations (UN) (1992) Committee on the Elimination of all Forms of Discrimination against Women defined gender-based violence as:

Violence that is directed against a woman because she is a woman or that affects women disproportionately. It includes acts that inflict physical, mental or sexual harm or suffering, threats of such acts, coercion and other deprivations of liberty (United Nations, 1992: para. 6)

The UN commitment to reduce such violence against women suggests that those concerned with promoting girls’ schooling need to address not just the quality of schooling in deprived urban environments, but also the high levels of violence experienced by girls and young women living in such ‘habitats’. As Kiwala et al. (2009) point out, slum dwelling has particular consequences for women of all ages, making them even more vulnerable. It is now recognized globally that women are frequent victims of gender-based violence in contexts where there are chronic crowded housing conditions, a lack of security and privacy, anger and frustration over the lack of income, food and employment that spills over into marital violence, and violence connected with the use of drugs and alcohol. The Centre on Housing Rights and Evictions’ report on Women, Slums and Urbanisation (2008) indicated that women’s insecurity is aggravated by a lack of basic services such as proper streets, street lighting and police authorities to engage with crime and violence in the area. The lack of access to safe, alternative housing (or refuges) also keeps women trapped in violent domestic situations. Consequently, women living in slums are inadequately protected and are at high risk of experiencing personal violence, and of contracting illnesses such as HIV/AIDS.

In this article, we explore how young women living in three African slums experience violence and, on occasion, use violence themselves. In all three contexts, young women confront different types of violence – for example, domestic, sexual and street gang violence. Experiences of uncontrolled physical and sexual abuse as children or as school pupils (Dunne et al., 2006; Leach, 2003) can drive young women and men to leave home and school to live on the street, making them vulnerable to further violence. Further, such family-based or school-based violence can introduce young people to the use of non-premeditated violence against others of their own age. Schools may also provide lessons on where and when violence is an appropriate response (Dunne et al., 2006). Deep within such training in how to ‘do violence’ and the use of violence are concepts of masculinity and femininity, and the emotions associated with power and subservience. Gender equality programmes cannot easily isolate young people from such violent environments – gender relations, whether male–male, male–female or female–female, are the site within which violence occurs. Implicated in these relations are sexual relations and control over bodies, particularly young women’s.

Drawing on Bronfenbrenner’s (1979) social ecological systems theory, and on Hobsbawm’s (1998) concept of the ‘rules of violence’, we explore some of the findings from three different research projects in which young people living in acute urban poverty describe how and where gender-based violence is enacted. Such cameo data suggest the ways in which the contexts of young people living in townships, slums and cities circumscribe the role of violence in their lives. Within these various interlocking contexts, young people lead poverty-stricken, often desperate, lives in which their social, economic and political rights as young citizens are frequently denied and their chances of escape are minimal. However, we also demonstrate that the gender-based violence that is associated with such contexts is not without its own rules. As Hobsbawm (1998) has argued, it is essential to investigate these social uses and rules of violence. He argues not only that violence provides ‘subjective psychological relief’ (305), but that ‘for the weak and helpless poor, violence and cruelty . . . are the surrogate for private success and social power’ (304).

We begin by recommending the value of a social ecological approach to gender-based violence before exploring, through the voices of young South African, Ghanaian and Kenyan youth, the deeper connection between gender-based violence and poverty. These data expose the extreme forms of gender-based violence that lie at the heart of young people’s poverty and their social exclusion. Such violence represents bodily violation but in the ecology of townships and slums it is also a part and parcel of survival. The rules of violence re-order young people’s understandings of social norms such that violence is not challenged but is instead feared and coped with strategically.

A social ecological approach to gender-based violence

Bronfenbrenner’s (1974, 1979, 1992) ecological theory is particularly relevant to studies of gender-based youth violence in African cities such as Cape Town, Accra and Nairobi. Ecological theory focuses on the role of the social environment and context in people’s development, seeing people as significantly affected by the interactions of overlapping ecosystems. Swartz’s (2009) adaption of this model recommends a ‘social ecological’ approach to violence and the morality associated with violence. This approach is concerned with the web of human relationships within their environmental contexts using interdisciplinary and multidisciplinary perspectives and employing multiple levels and methods of analysis (Swanson et al., 2003). From this perspective, the study of young people’s experience, and use, of violence focuses both on the physical environmental and social contexts of people’s lives. It readily acknowledges that these contexts are complex and interact with each other as much as with the person located within the context. It is essentially an approach that recognizes the interdependence of multiple systems. According to Alcalay and Bell (2000: 23), the relationship between people and their environments is reciprocal:

People’s actions are affected by the environment . . . but the environment can also be shaped by the actions of individuals and communities . . . People have different capacities for action in varying environments because environments differ in the resources they provide to individuals.

Bronfenbrenner (1992: 226) proposed that a child’s development should be considered through a ‘hierarchy of systems at four levels moving from the most proximal to the most remote’, though he later added a fifth level. He described the most immediate context for each individual as the microsystem – largely the close ‘interpersonal relations experienced by a developing person in a given face-to-face setting’ (227). These relationships include those a young person has with peers, teachers and other staff (school); mothers, fathers, siblings, their own offspring, and extended and blended families (home); friends, peers, gangs, acquaintances and romantic or sexual partners (streets); and neighbours, unrelated older peers, unrelated younger children and religious establishments (community).

At the farthest extreme Bronfenbrenner describes various macrosystems – ‘pattern[s] . . . characteristic of a given culture, subculture, or other broader social context . . . [especially] belief systems, resources, hazards, life styles, opportunity structures, life course options, and patterns of social interchange’ (228, emphasis in original). For the slum and street youth described below, the various macrosystems in which they find themselves include the political contexts of neoliberal fiscal policies, poverty, their strong youth and traditional cultures, and various manifestations of structural injustice, including gender and racial discrimination.

Between these microsystems and macrosystems lie two further systems. The first is the mesosystem, which describes the interrelationships between microsystems of which young people are a part. For urban youth, the mesosystem consists of interrelationships between home, school, streets and community. In impoverished contexts, these interrelationships seldom exist. The second system Bronfenbrenner refers to that lies between microsystems and macrosystems is the exosystem. The exosystem comprises institutions and practices of which the young person is not directly a part, but whose consequences s/he experiences (227). Many young people experience the results of, for example, the criminal justice system, local government policies, and adults’ general behaviour in society (alcohol, drugs, work ethic) without necessarily being a part of these systems. Further examples of the exosystem include parents’ workplace conditions, the national economy, health, social and transportation services, and mass media, as well as:

[T]he fragmentation of the extended family. . . the disappearance of neighbourhoods . . . geographic and social mobility, child labour laws. . . welfare policies. . . the growth of single-parent families . . . the delegation of child care to specialists and others outside the home, urban renewal . . . an explicit national policy on children and families (Bronfenbrenner, 1974: 4).

Finally, Bronfenbrenner (1992) describes a fifth system, a chronosystem, as changes both in the environment and in the person that alter the relationship between the person and the environment. For these young people, besides obvious changes such as puberty and increasing cognitive sophistication, the chronosystem also includes the move from traditionalism, colonialism and Apartheid (in South Africa) to living in a nascent democracy (Swartz, 2009).

A social ecological approach to the issue of gender-based violence foregrounds the importance of young women’s or men’s multiple contexts when considering how young people face and experience violence against themselves or others. This framework moves us beyond the narrow confines of notions of individual choice or even the close influences of gender relations within the family, amongst peers, in schools and the community. It compels us to expand our vision beyond the purely psychological focus on violence to include sociologists’ macro focus on structural violence and the immediacy of social relations in poor urban settlements where daily survival is based on particular forms of violence and violations. These forms are neither random nor gender neutral. Rather, they are mediated through the rules of violence within the context of particular social relations and gender identities.

Within each social ecological layer are, following Hobsbawm’s recommendation, different types of violence. They are contextualized in time, space and within relational worlds. Tackling violence against girls, therefore, as our research shows, means addressing the relationship between gender identities and male power relations, as well as the lack of power these young people experience within marginalized degraded environments. By listening to the voices both of young men and women, we can begin to understand the connections between, specifically, urban poverty, gender and violence. Furthermore, we can begin to perceive the types and levels of work that schooling needs to do in shaping a secure environment for both sexes. As we demonstrate, gender-based violence by definition cannot be addressed entirely as a female or male issue. It is embedded in multiple ecosystems.

The Strategy Paper on Urban Youth in Africa developed by the United Nations (UN-Habitat, 2007) describes a range of different types of violence among citizens that supports the above view. Gang violence is among the forms of violence identified, and is especially relevant to the study of street youth. The formation of groups of young people into gangs who are engaged in criminal or delinquent behaviour is frequently seen as ‘a reaction to exclusion and marginalization in society’ (27). As this report notes, the majority of members of such gangs are male, often in the transition stage between childhood and adulthood. Young men find in such gangs ‘acceptance, status, identity and social and recreational opportunities, as well as economic gain’ (27). Yet membership can mean becoming involved in fights, drug dealing, violent crime, intimidation and, as we shall see from the data presented from our studies, rape. Although there is still insufficient information about how gangs work within African settings, there is considerable evidence of the levels of sexual assault and rape of young women. Sexual violence and rape are especially heinous given the prevalence of HIV and AIDS. ¹

Rates of rape and other forms of sexual violence are high in sub-Saharan Africa. So, for example, one in five women aged 15–49 were found to have experienced sexual violence in the year preceding the survey of the Kenya National Bureau of Statistics (2010), whilst gang rape in Nairobi constituted approximately 20% of reported sexual offences (UN-Habitat, 2007). The evidence from South Africa provides even more detailed information. It suggests that in 2007 there were 55,097 cases of vaginal rape reported (South African Police Services, 2010); one third of adolescent girls reported forced sexual initiation (Jewkes and Abrahams, 2002). Jewkes et al (2009) found that 40-50% of women reported having experienced violence at the hands of an intimate partner within the home, 28% of men reported having perpetrated rape, with three quarters of men who raped doing it for the first time before the age of 20, and 40% of victims who reported rate to the police were girls under 18. As alarming as these statistcis are, Jewkes and Abrahams (2002) show that there is a ten-fold under report of rape and attempted rape reported to the police compared with a representative community survey in South Africa. The situation of sexual violence (and under-reporting) is no different in Ghana, where 21% of female adolescents report experiencing forced sexual initiation (World Health Organization, 2002). Additionally, the Domestic Violence and Victim Support Unit (DOVVSU) of the Ghana Police Service reported a total of 109,784 cases of abuse between 1999 and 2010 (DOVVSU, 2010). ² Below we describe in more detail the types of physical and sexual gender-based violence which young African women face and how they respond to the poverty they experience within three African cities.

Researching the lives of urban youth

Although our research on youth in various urban contexts did not specifically focus on violence, the evidence we collected separately contribute to gender-based violence debates. It is difficult not to be confronted with evidence of violence when researching the lives of urban youth living in poverty. Sensitive listening to experiences of gender-based violence is exceptionally important to the study of schooling in such communities. In each of the three studies the structural violence represented by impoverished urban environments excludes young people from the benefits of citizenship. In such contexts, the different forms of urban violence impinge on their life chances and add to their disadvantage.

The cameos we present are drawn from three recent qualitative studies. We make no claims about generalizability of our data, preferring instead to highlight the importance of learning about violence from young people themselves. The research evidence was generated by three North–South research partnerships which explored the lives and experiences of male and female urban youth living in poverty in South Africa, Ghana and Kenya. Gender-based violence was uncovered, for example, in Oduro’s (2010) study of sexuality and protection from HIV/AIDS when she conducted in-depth interviews with a sample of 7 male and 17 female street youth (aged 14–19 years) living on the streets of Accra. Swartz’s (2009) ethnographic research of the moral lives of young people which included interviews and a range of activities with 37 male and female youth (aged 14–20) living in the Langa township in Cape Town exposed extensive and repeated examples of youth violence. Examples of how gender-based violence also infiltrates national/community politics were found in the interviews Arnot and her colleagues, in their study of Youth, Gender and Citizenship (YGC), conducted in poor households with 24 young men and women (aged 18–25) in Nairobi and 28 young men and women (aged 14–26) in Accra.

It is important to note that although these three qualitative projects do not draw upon the same samples, all these youth are, broadly defined, ‘street youth’. They are streetwise and in all cases experience the devastating effects of a degraded and impoverished urban habitat. In the case of Oduro’s research, the young people have no other home and sleep on central city streets; Swartz’s youth live in a sprawling segregated township comprising only black South Africans; and the YGC youth sample in Nairobi live in temporary shacks in one of the world’s largest slums. In all three studies, evidence of violence infiltrated the voices and statements of the youth. These data throw light on, though do not necessarily capture all, the violence nor all counter-violent strategies which female youth use to ensure their survival in such dangerous urban environments. It was not always easy to tap young women’s experiences, not least because of the silence around sexuality and sexual violence. Nevertheless, our data illuminate how gender-based violence permeates each level of life: from the macro- and exosystems of structural violence to that of the meso- and microsystems – the realm of personal, intimate relations.

Structural violence and youth violence: revenge, respect and power

At the macrosystem level, millions of young people living in urban habitats are already thought to be ‘in crisis’ (UN-Habitat, 2008). Not only has increasing urban migration led to the development of huge conglomerations, but these sites are often positioned without any rights to the land. They are mainly overcrowded, notoriously without potable water, sanitation, sewage, electricity, proper housing, and street lights. The lack of urban planning and governance makes many such slums lawless, with ‘no go’ areas for police authorities, and subject to mass slum evictions. Jones and Chant (2009) argue that the presence of such unsustainable degrading environments is a form of structural violence. The reduction of state welfare support, and the privatization of schooling within such environments further reduce the capacity of the poor to improve their lives and break the intergenerational cycles of disadvantage, especially in contexts without social security schemes, such as Ghana and Kenya. In the precarious life in such cities, young people have to navigate the violence imposed on them by the state as well as the violence within such communities where the fight is to stay alive more than it is to achieve schooling.

Swartz’s (2009) research in Langa township reflects how the effect of this macrosystem, specifically the political philosophy of Apartheid policies in South Africa, affects how young people address the meanings, rules and social uses of violence. Her research vividly demonstrates how violence is a core element in the lives of township youth inhabitants, whether during the running battles of Apartheid or in the crime and alcohol-fuelled clashes of everyday street life (Seekings, 1993). Young people’s explicit accounts of the social use of violence and its complex determinants uncovered their own rules for when violence was considered an appropriate response to daily experiences of confrontation and disturbance. Their personal stories of daily encounters with varied forms of violence revealed a textured portrait of the phenomenology of male and female violence. Macrosystemic violence infiltrated their lives in multiple ways. In the home, mothers were reported eking out a living by selling alcohol from the front rooms of their shacks, resulting in early exposure of young men and women to alcohol and its related drunken violence. Young people also spoke fluently of the territorial violence that occurs on the streets, from a group of girls protecting their sources of drinks and money from other girls in their street, to conflict between gang members in adjacent areas, or between members of different schools (or nearby communities). Indeed, school is a further site of violence. Many young women complained of being sexually harassed or molested at school by both peers and teachers. Corporal punishment, too, remained part of the experience of many youth, despite its legal ban in South African schools. ³

Apart from the violence of the macrosystem that marginalizes and stigmatizes such youth, the gendering of violence and the gendering of the causes of violence are shown to be complex elements of social interaction between family life, between youth (the mesosystem), and between young people and the laws of the state (the exosystem). Gender-based violence manifested itself in different forms, including physical, psychological, emotional and sexual ones, in different hierarchies and at different levels within their ecology – home, school, streets and community. Below we use examples of key moments to illustrate how gender dynamics were worked through using physical violence. Using Andiswa as an example, we show how she interpreted young women’s violent responses as a form of agency, a means of protecting oneself by becoming similar to the boys in the neighbourhood. Meanwhile, her boyfriend Andile legitimated his own violent responses and violated Andiswa in his desire to promote his masculinity.

Sexual violence on the street

The relationships between sexuality, forms of survival and the powerful manifestations of masculinity and gender impinge on the lives of girls living on the street in complex ways. We are reminded in Oduro’s (2010) study of the diversity of sexual violence in Ghana within different sites and with different consequences. She identified an overriding desire by young men for male dominance over the reins of seduction leading to sexual intercourse. Male school youth, for example, sought to affirm masculine strength, virility, physicality and power by seducing young women, using largely unprotected casual sex and demanding fidelity. When not involved in such dangerous casual relations with men, accompanied by their promises of marriage, some schoolgirls were also found to be at risk from the patronage they sought from older men who were expected to pay for school fees, to help find employment, or support young children. Similar findings in Zimbabwe, Malawi and in a separate study in Ghana (Leach, 2003) highlight the vulnerability of schoolgirls. Through transactional sex they risked the loss of not only their dignity, but also their health and educational opportunities if they contracted STIs, HIV/AIDS or became pregnant. Female sexuality was the only qualification that these young women could use as their agency or capital in exchange for education or a job.

Street youth and youth at school experience similar sexual interactions and violence around these interactions. The differences lie in the location of street youth on the margins of the exosystems. They are outside the scope of welfare institutions and family support and are directly exposed to aggressive street policing. Paradoxically, in this public space female street youth talk about the desire for ‘freedom’ and the exercise of female agency to gain control of their lives and indeed to survive the effects of social exclusion. Such ‘freedom’ places young women living on the streets of Accra in danger of extreme sexual violence and the loss of agency, health and the prospect of a better life. Male sexual violence dominates the accounts of girls who, having run away from or lost their home, find themselves living on the pavement exposed directly to it. These themes are illustrated below.

Street sexuality: risk from three sources

Male affirmation of sexuality was amplified on the street. Oduro’s exploration of what she called street sexuality, in no-man’s land between the exosystems with its institutions and the microsystem in which young men rule, revealed the street as their primary ecological system. Here young women were directly exposed to extreme male sexual violence (a violent interpretation of some Ghanaian men’s understanding of their rights over women). Frimpong-Nnuroh (2006), writing on sexuality among the Nzemas of Ghana, observed that forced sex was regarded as a taboo with heinous consequences, with the act of sex in public deemed demeaning and highly unacceptable. ⁵ However, the experiences of the vulnerable youth in Oduro’s study revealed how poverty, exclusion and other factors forced many of them into these unacceptable and dangerous public sexual practices. Such cultural traditions, whilst designed to protect women from rape, in this context prevented female street youth seeking help. Thus, bringing gender relations within the microsystem into the public domain provided little physical or moral protection.

The sexual experiences of street girls in Oduro’s study were such that the line between public sex (prostitution/multiple relationships) and private sex (sex with a boyfriend or regular partner) was blurred. This dangerous and volatile context meant that girls found themselves at the mercy of multiple forms of male sexual control. They encountered risk from three powerful sources, namely their regular boyfriends/minders, so-called ‘killers’ on the street, and men in the context of prostitution. ⁶ Lack of shelter and accommodation forced many of the street youth to sleep in open places, which exposed the girls to sexual violence in the form of frequent individual and gang rape. As 19-year-old Lizzie commented simply, ‘rape is so common here’. She elaborated, saying it was especially unsafe if boys felt that a girl was rejecting their efforts to seduce her. ‘They rape you to shame and quieten you.’ Like Adomako-Ampofo and Prah (2009), Oduro discovered that some boys intentionally gang-raped girls who were considered to be proud, arrogant, or who were perceived to feel ‘too good’ or ‘expensive’ for their male peers. To such boys, girls were supposed to be available to men and be flattered by their advances. If they behaved otherwise, then they had to be put in their place (Oduro, 2010, 2012).

Lizzie had stated that certain older boys known as ‘killers’ ‘rape girls in turns’ while the younger boys would get involved ‘on a few occasions’. Fifteen-year-old Blessing echoed these comments, saying: ‘everybody rapes here, both young and old’. Unfortunately, many girls were unable to report such attacks because of the stigma attached to rape. An added power dimension was that a victim of street rape was often perceived by her peers as being weak and having courted it: ‘the truth is, most girls feel shy to tell others that she has been raped so she keep quiet and suffer in silence’ (Happy, 17-year-old). There was also the fear of adverse effects on their prospects of getting a regular lover or even losing an existing one. A 14-year-old girl, Cilla, explained: ‘although it was not the fault of girls who were raped, some boys leave their girls when they are raped with the excuse that, I don’t want to be identified as the boy whose girl was raped’.

There were attempts by some girls to protect themselves against rapists by wearing very tight pairs of trousers when sleeping. As 15-year-old Golda explained:

It is better to sleep with your trousers or shorts on. If you wear a skirt or dress, you make it too easy for them. Enie, mbema no omo bϵ paapaa wo. Wo be sori no na omo aye wo paasaa. [Then the men will have easy access to you and tear you up. By the time you wake up then it’s too late.]

Both girls, however, said wearing tight trousers was not an assured security against rape. As a result of this constant fear, girls in the street hardly slept deeply during the night. Girls commented: ‘here you have to learn not to be a deep sleeper’, ‘can you imagine sleeping so deeply on the street with all the mosquitoes, insects and noise around? Yet, some girls sleep so heavily and [then] the boys get them.’

Ending violence?

In both case studies so far, young people demonstrated both understanding and agency when speaking of violence in their lives. It was therefore not surprising when the Youth Gender and Citizenship research team in Nairobi and Accra (Arnot et al., 2012) asked young people living in urban poverty about what needed to change in order to improve their lives, dealing with violence in their lives was high on their list of responses. Moreover, these youth were outspoken about the need for governments to listen to them and to address these issues. In Kibera, Nairobi, the site of considerable election violence in 2007, youth showed themselves to be deeply concerned about the violence in their community.

Asman described the sense of betrayal resulting from government’s failure to control violence not just amongst youth such as himself, but also surrounding political activities:

You see violence every day, tear gas, gang shots, many things come up. Neighbours fighting, that’s politics . . . So they will not care about us, so we will have to care about ourselves for them to know our rights. They should understand us.

Poverty, social exclusion and the perceived corruption of the government (part of the exosystem) militates against any chance of social improvement of young people’s environment. Asman argued that if government listened to youth (addressing in effect the needs of the exosystem), there would be more schooling and more employment, and as a result less theft and fewer theft-related muggings. Both female and male youth raised concerns about the lack of proper security, where young men talked about getting mugged and young women feared being raped as they went about their business. In Accra, 17-year-old Catherine asked the government to ‘make me safe’, adding ‘[politicians] should look into my safety’. Sexual violence, even rape, was linked to politics. Twenty-one-year-old Hana observed: ‘You know politics is a dirty game . . . the other day women went up there at X [a politician’s] rally. They were given money, after that, they were raped.’

In both Nairobi and Accra, young people described the benefits of schooling in relation to conflict reduction. The young men particularly spoke about needing to ensure peaceful co-existence as an outcome of school, not least by listening. Key to that peaceful co-existence was interaction with the opposite sex. Mixed secondary schooling, they reported, had had an effect of encouraging young men to see girls as their ‘sisters’ and for girls to redefine boys as their ‘brothers’. Also, gender socialization in mixed classrooms could, according to these youth, teach them that women have talents, that they can help men. As 18-year-old Tony from Kibera commented:

[Secondary schooling] has helped me because we were in a place where we dwelled with the girls, played and so on. So if not for the existence of school you would be seeing girls and have crazy thoughts [of a sexual nature].

There is some evidence here that schooling can also play a part in terms of redefining femininity, giving girls more self-respect and confidence. For example, it appeared to have helped young women in Accra to know about self-protection. They reported that they were beginning to ‘keep themselves as girls’ – describing a public image of a hygienic, healthier body and lifestyle. This raised their self-esteem, despite many leaving school because of pregnancy. Schooling in itself was not sufficient to counter the dangers within this community, yet both male and female youth were confident that their lives could improve through government’s provision of free schooling and scholarships for university education. However, these Accra youth also talked extensively about schools being violent places, and emphasized the need to stop violence against children in schools, so that learning can occur.

Discussion and conclusions

Globalization, economic liberalization and environmental destruction have led to the creation of global slums in sub-Saharan Africa, the majority of whose inhabitants are youth. In many African cities more than 50% of the urban population are under the age of 19 (UN-Habitat, 2007, 2008). This generation of youth have to survive, often on their own, without adequate adult role models and support systems within difficult contexts. Taking a social ecological approach, these macrosystems with their institutional manifestations contribute to the individual’s identity and general behaviour, which in this case links forms of structural violence to that of youth violence. In these marginalized and deprived habitats, it is widely recognized that youth suffer from isolation, marginalization, discrimination, rootlessness, vulnerability to crime, violence and sexual victimization. Most at risk of victimization are girls, street children, school dropouts, and those affected by HIV/AIDS (UN-Habitat, 2007). Slum/township dwellers have to navigate their survival in a context of dwindling family support, family breakdown, high levels of unemployment and few educational opportunities. As social institutions fail them, youth violence in this context becomes the means of self-protection and legitimated forms of self-expression and agency.

The UN report (UN-Habitat, 2007) on urban youth in Africa points to the important cultural and social differences which from birth offer boys far greater personal freedom, affording them increased opportunities for education, whilst girls are expected to sustain ‘household employment’ (25). The effect is a distinct difference in the ‘types and level of risks to which girls and boys are exposed’ (25). Significantly, violence, as we have seen, is based on the expressions of hegemonic masculinity not just poverty in the township and on the street.

Whilst such gender divisions reinforce the power of young men and their rights over women, they also implicate the use of rape as an instrument of oppression. Our data point to the use of rape by male politicians, male clients of young female prostitutes, sugar daddies and young men. When used by young men this gender-based violence represents, as we have seen, revenge and retribution but also the desire for respect, the expression of love of a mother, control over female sexuality and ultimately the assertion of collective notions of masculinity. This microsystem is in part a personalized expression of the structural violence of the macrosystem as much as it is a personal, intimate expression of complex physical, emotional, psychological, sexual and political beliefs. Gender-based violence and violation as it is used on the street, in the townships and sprawling urban settlements link space, power, conflict, identity and desire.

Urbanization as one landscape is associated with advantages such as a formidable workforce for industrial growth and a ready market for increased production and services. Women constitute a high percentage of slum dwellers in sub-Saharan Africa for reasons such as discrimination, the practice of wife inheritance in some African societies, domestic violence, HIV and AIDS stigma, and forced and early marriage (Kabajuni, 2009). As we have seen, young women in such contexts are highly vulnerable to male control and abuse. Kabajuni (2009) argues that, unless such female experiences within slums are addressed in slum improvement programmes, the Millennium Development Goal 3 on gender equality and the empowerment of women cannot be achieved (see also Jaeckel and van Geldermalsen, 2006).

We began by asking how learning of the gender-based violence which youth, (especially young women), experienced and used might shape our understanding of the types and levels of work that schooling needs to do in shaping a secure environment for both sexes. Adopting a social ecological approach to understanding violence is a useful tool since it links youth’s experiences of violence in school, community, street and family with the larger macrosystems of poverty, government policies, power relations and structural violence. If youth violence is to be reduced it therefore becomes essential that interventions occur at different system levels and at multiple points over time. While it is beyond the scope of this article to go into detail regarding what these interventions might look like, it is possible to offer examples of how a social ecological perspective might assist in identifying the multiple opportunities available within each system for violence reduction.

One starting point is to learn about the rules and roles of violence understood by different groups of youth – differently positioned within one space. Effective educational provision, legal protection and urban planning focused on women’s safety are urgently needed. Schooling also needs to restructure gender relations in a way that denormalizes such gender violence. If urban youth teach us anything it is that the task for educationists is now one of working with the experiences and values of young people who can identify what is needed to challenge the violence inflicted on them by poverty, the denial of their rights, and the violence they inflict on themselves and on others. Such violation and violence, however, must be contextualized within the habitats in which they live for it is these which shape their lives.

At the level of the chronosystem, violence interventions could take account of, for example, young people’s developing cognitive sophistication, and aim to ensure that this occurs optimally in the provision of stimulating and formative education – an outcome not always available in impoverished communities. Violence interventions might also build on the growing human rights cultures present in sub-Saharan Africa, and extant traditional philosophies and cultural practices of personhood (ubuntu) in which to frame violence prevention. At the level of the macrosystem with its focus on cultural beliefs, opportunities, options and social practices, there are numerous practices that might be strengthened and supported, while others require strong challenge and mitigation. As Jewkes et al. (2009: 2) recommends:

A national strategy for violence prevention must focus on the factors underlying the very high levels of violence . . . particularly poverty, youth unemployment, gender and other social inequity, dominant ideas about manhood, exposure of children to trauma and abuse, harmful levels of alcohol consumption, social norms on the use of violence, access to firearms, as well as the weak policing and legal responses.

These macrosystems overlap with the exosystem, and require applied thinking. Questions that need to be asked include: How might youth violence be dealt with in the criminal justice system so that it is more restorative; how might parents’ workplace conditions be made more amenable to better child supervision; and how might hyper-violent media portrayals be mitigated? At the level of the microsystem and mesosystem of streets, peers, school and community, and their interactions with each other, more creative engagement with multiple actors might provide new strategies for violence reduction. Might it be possible to place teachers, youth, neighbours, parents and peers in conversation with each other, especially since in impoverished spaces these groups seldom communicate, at least not as equal partners, in seeking solutions for intractable problems? So, for example, teachers could be trained in how to help young people resolve conflict with romantic partners as well as same-sex antagonists; and parents and neighbours might be helped to develop an informal ‘youth charter’ (Damon and Gregory, 1997) that agrees on how delinquent youth in a street or neighbourhood are treated.

With regard to specific forms of violence, for example violence in search of revenge and respect, a strategy might be to work with youth to examine alternative means to obtaining respect. And in the case of male domination over young women on the street, alternative ways to achieve protection and safety are required for young women, and new ways are needed to help young men experience mastery without resorting to domination. These interdisciplinary perspectives, along with multiple levels and methods of analysis, offer the opportunity for new thinking with regard to violence interventions, especially in the amelioration of gender-based violence in the slums and streets of impoverished African cities.