Boys’ and Young Men’s Perspectives on Violence in Northern Tanzania

Abstract

The challenge of violence for youth in low-income countries includes a range of experiences from witnessing to experiencing to participating in violence. Although boys and young men are often the perpetrators of such violence, they may also be its victims. Yet little evidence exists from the voiced experiences of boys themselves on perceptions and interpretations of the violence around them. Given the numerous negative health implications of violence for boys, for the girls and other boys with whom they interact, and for the health of their future partners and families, we conducted an in-depth study in rural and urban Tanzania with adolescent boys on the masculinity norms shaping their transitions through puberty that might be contributing to high-risk behaviours, including engagement in violence. The findings identified underlying societal gendered norms influencing the enactment of violence, and recommendations from the boys on how to diminish the violence around them. Additional research is needed with boys on the social norms and structural factors influencing their engagement in violence.

Introduction

The challenge of violence experienced by young people living in sub-Saharan Africa (SSA) today is increasingly documented in the literature. Although much of the focus is on gender-based violence with particular attention to girls and young women, more recent publications by the United Nations Children’s Fund (UNICEF), the World Health Organization (WHO) and an adolescent-focused series published by the Lancet have brought attention to the violence experienced by boys and young men as a significant risk factor for population health (UNICEF 2012; Blum et al 2012; WHO 2000). Although the high rates of violence experienced by girls are critically important to address, the violence experienced by boys and young men is in need of greater attention. Boys living in SSA and other low-income regions are at increased risk of physical and armed violence (UNICEF 2012; WHO 2000). Almost twice as many boys as girls in SSA report having been involved in physical fights; and larger numbers of in-school boys than girls aged 13–15 report having been the victims of bullying (UNICEF 2012). Although less well documented, likely for reasons of stigma and taboo, boys are also the victims of sexual violence. Dolan’s (2009) research from northern Uganda has highlighted the devastating consequences and continuing silence around male rape. A recent study of violence against children in Tanzania indicated that one in seven males under age 18 was the victim of sexual violence, with three-quarters of both men and women having experienced physical violence by 18 years of age (VAC 2009).

The increased rates of violence experienced or perpetrated by boys and young men are frequently attributed to high rates of urban growth, social, political and economic inequalities, and other potentially dislocating aspects of society (UNICEF 2012; Ricardo & Barker 2009). In SSA, a young man’s sense of his own manhood and role in society suffers when opportunities for income generation are not available, possibly leading to engagement in violence (Barker & Ricardo 2005). Changing economic and social realties in East African societies may be leading to men’s sense of disempowerment, and their subsequent increased use of violence, both physical and sexual, as a mechanism for displaying their manhood amongst their family, peers and larger community (Silberschmidt 2001). The predominant or hegemonic masculinity within a given culture may also encourage a manhood involving aggression and violence rather than other forms of conflict resolution (WHO 2000; Connell 1996). Heald’s (1999) long-term study in Uganda explores this reality, and the complex ways in which masculinity and violence are frequently interlinked and deeply embedded in meanings of manhood. Local norms and gendered behaviours, such as the acceptance of wife beating, may also contribute to intimate partner violence perpetrated by boys as they grow into men (Jewkes et al 2009; Rani et al 2004). Over half of young men aged 15–19 and a significant percentage of girls in a select number of SSA countries justified wife beating under certain circumstances, suggesting norms around acceptable violence run deep in many societies (Rani et al 2004; Demographic and Health Survey 2010). In some contexts there may exist a range of perceptions among boys and men about what constitutes acceptable forms of violence against women, with intimate partner violence perceived as normal (Barker & Ricardo 2005; Brear & Bessarab 2012). The risk of engaging in violence begins early for many boys, with the average age of entry into gang violence at 13 years (United Nations Children’s Fund 2012). In Tanzania, the location of the study described in this paper, there is existing, albeit limited, evidence of widespread acceptance of certain forms of violence against women, along with interpretations of manhood that may contribute to other forms of violence within society (DHS 2010; Violence Against Children 2011). The social norms underlying the engagement in violence, from the perspectives of boys, remain less well understood despite the critical work of regional researchers (Silberschmidt 2001; Heald 2005; Maman et al 2010).

Definition of Violence

There are numerous definitions of violence within the literature, but for the purposes of this study, a focus was given to interpersonal violence, with particular attention to gender based violence, family and peer violence. This allowed the boys who participated to conceptualise and discuss their experiences of violence in its varying forms. The focus was primarily on physical and sexual violence, with less attention given to eliciting interpretations of emotional violence.

Implications of violence for society

There are significant implications of boys’ and young men’s engagement in and experience of violence for the health of a given population. Boys themselves may suffer injuries or fatalities from the violence engaged in when proving, defending or enacting their perceived manhood. They may incur mental health-related anguish and stressors whether they are the victims or the perpetrators of violence. They may choose to drop out of school if violence in the educational setting makes attendance stressful or uncomfortable. They may also engage in violence that negatively affects the health and wellbeing of their future sexual and marital partners, children, and community members (Petersen et al 2005; Jewkes et al 2006). The cycle of violence, one that is learned from a combination of socialisation or environmental factors including modeled behaviour within the society (such as parents or images in the media) and the shifting social norms regarding masculinity, may perpetuate into future generations in the absence of interventions that present a different norm for conflict resolution or differing images of manhood (Mercy et al 1993; Heald 1999). These social norms, or gendered norms as they are explored in this article, can be defined as “socially constructed roles, identities and attitudes” (Ricardo et al, 2006, p. 61) within a given society. Such norms consist of a complex set of individual and environmental factors (e.g. family, peers community, structural factors) that are dynamic, and influenced by factors at multi-levels of society and by historical changes. While young people are influenced by such norms, they also serve a role in reconstructing the norms (Ricardo et al 2006).

Despite the growing evidence documenting violence within many SSA populations, such as the gender based violence reported extensively from South Africa, and gang-related violence reported in Nigeria (Jewkes et al 2009; Barker 2005), there is a significant gap in our understanding of the masculinity norms as understood and received by boys as they transition through puberty into young adulthood that may be contributing to the perpetration and experience of violence. Critically, this includes boys’ voiced perceptions of the socialisation process that leads to such violence. The study described in this paper aimed to better understand the masculinity norms leading boys to engage in high-risk behaviours. Violence emerged as a gendered norm potentially influencing boys’ perceptions of the enactment of manhood as they transition into young adulthood in one SSA context, that of Tanzania. A rural versus an urban context were examined, given the possibility that the enactment of manhood and conveyance of masculinity norms might differ in more rural (and traditional) compared to more urban (and modernising) settings of the country. Although additional research is needed on the socialisation processes leading boys and men to engage in violence across Tanzania, this paper reports on violence-related findings that emerged from this exploratory study with boys.

The conceptual framework guiding the study was based on social-ecological theory (Bronfenbrenner, 1979) and the social construction of gender, incorporating the observation that research on masculinity has generally lagged behind research on femininity. The application of social-ecological theory framed the exploration of multi-level influences impacting boys’ lives. These illustratively included the family, community, peers, education system, religious beliefs and macro-economic shifts impacting on local employment and household daily life. Examining the ways in which different levels of influence are shaping boys’ perceptions of manhood, and in turn the ways in which boys are enacting that manhood within the complexities of the larger society, including episodes of engaging in or experiencing violence, were critical to acquiring a holistic understanding of boys’ present days transitions into young adulthood. This included an exploration of how larger economic and social shifts may be impacting the experience of violence within the society, and related, the societal dynamic in which boys are growing up and becoming young men. The study was also guided by relevant gender theory, framing the shifting gendered dynamics between the generations, and between boys and their male and female peers in and out of school (Kimmel, 2000; Connell, 1996). Connell’s descriptions of a hegemonic masculinity existing within a given society, alongside of a plurality of masculinities, provided guidance for exploring what predominant masculinity norms are currently being conveyed to boys growing up in the local context, while also enabling insights into alternative masculinities that might create opportunities for healthier adolescent transitions for boys. As Barker and Ricardo emphasise (2005), it is important to recognise the multiple masculinities that exist across and within countries in SSA, and to explore the range of masculinity norms influencing boys transitioning into adulthood, including those that may lead to violent or non-violent models of behaviour. The two theoretical approaches were complementary in providing insights into boys’ transitions into adolescence in modernising Tanzania.

Exploring the influence of masculinity norms on boys is particularly important during the transition through adolescence given the increasing divergence in gendered experiences that occurs in many societies for girls and boys after reaching puberty (Ricardo et al 2006). Adolescence may also be a time of experimentation, of increasing autonomy, and of growing influence of peers on young people’s behaviours and self-identity. Numerous social, emotional, and physiological changes occur over the course of the transition, reinforcing the tendency to enact new roles and behaviours (Lerner & Steinberg 2004).

Research Setting

The research took place in Moshi (urban) and Rombo (rural) districts in Kilimanjaro, Tanzania. The region is predominantly Chagga, although other ethnic groups exist in the area. Moshi is a well-populated town of crowded streets filled with businesses, internet shops, and a growing number of hotels catering to tourists. There are health clinics, hospitals, and schools. Rombo includes the town of Mkuu, with a small central market area with tailors, groceries, and other small shops, and the dirt roads in town having scattered small bars and shops. There are a number of schools located in and around the town, which has a large hospital, some health clinics, minimal internet availability and almost no tourists. In Moshi, a peri-urban secondary boarding/day school served as the source for in-school participants, with a youth center in urban Moshi as the source for dropouts. In Rombo, a secondary day school served as the source for in-school participants, with a vocational training center as the source for dropouts.

Methods

The research design was a comparative case study (rural versus urban) including observation (classroom, school grounds, market), key informant interviews with adults interacting in boys’ lives (e.g. parents, teachers), in-depth interviews with boys (aged 16–19), and participatory activities with groups of boys in and out of school. Qualitative and participatory methodologies are known to be particularly useful and appropriate in the collection of sensitive data (Denzin& Lincoln 2000; Bernard 2002). Such methods are essential for capturing the numerous social, cultural and economic influences on boys’ lives as they transition into young adulthood, including those from the household, community, peer groups, school, and mass media, along with boys’ perceptions of and reactions to such influences. This paper will focus on the findings from two of the participatory activities: 1) violence stories: boys were asked to write anonymous stories about an episode of violence in which they participated or that they observed, and their recommendations for an alternative resolution to the experience; and 2) Gender based violence group work: boys were asked to comment on gender based violence following a group session in which statistics from the world and varying regions within Tanzania were presented and discussed. The use of participatory methods was important for gaining insights into boys’ lived experiences of violence in an empowering and safe manner (Minkler & Wallerstein 2000). The study was approved by the Columbia University Medical Center (CUMC) and the National Institute of Medical Research (NIMR) Institutional Research Boards.

The research team spent eight weeks conducting the research (four weeks per site). In each site, schoolboys were recruited through the school administration, and out of school boys through the administration at a youth center (Moshi) and a vocational training center (Rombo) (n=80 total per site, total two sites =160). The sampling aimed at capturing a diversity of experiences, with boys selected from differing ethnic, economic, family structure, and academic abilities. In the school sites, a group was gathered from three respective grade levels (Forms 2, 3, and 4) to capture a further diversity of experiences. Older boys (aged 16–19) were intentionally sampled to capture their perspectives on the various influences having shaped their transition to young adulthood. All group activities were conducted in a confidential location (meeting once per week for four weeks), and all participants provided informed consent. Given the sensitivity of the topic, that of adolescent risk-taking behaviours and violence experiences, group discussion was not tape-recorded to assure participants felt comfortable being open. Instead, the research team took in-depth notes on verbal responses and non-verbal behaviours. All the written data collected from the boys was anonymous. Each weekly session of group work lasted one and a half hours.

Data Analysis

We used grounded theory to analyse the multiple sources of data from the overall comparative case study, including fieldnotes, narrative notes, interview transcripts, analytic memos, and data collected through participatory verbal and written activities (Charmaz, 2006; Cresswell, 2007; Strauss and Corbin, 1998). Systematic analysis began with multiple readings of the fieldnotes to generate themes and hypotheses, followed by coding of the data collected. Open coding was used to identify appropriate categories, themes and issues that emerged from the data (Emerson, Fretz, and Shaw, 1995), with axial coding used to build connections within categories. Feedback on preliminary findings and emerging themes was sought from senior Tanzanian experts on gender and health. Triangulation of the data collection methods, and on-going analysis permitted the research team to integrate the local expert’s insights into the final selection of priority themes from the research. The major themes that emerged from the two participatory activities that will be discussed here included: (1) the modeling of violence; (2) rationales for violence and masculinity norms; and (3) boys’ solutions for diminishing the violence.

Results

The Modeling of Violence

There was revealed to be much modeling of violence observed by boys in their daily lives in the home, the school setting, and the larger community. This modeling included role modeling (such as parental resolution of conflict) or practices of the use of violence (such as corporal punishment) as observed by boys in their everyday lives, reinforcing the learning of gendered norms or socially sanctioned uses of violence. This included peers fighting over girls, intoxicated men fighting in bars, family violence including gender based violence, and to a lessor extent, corporal punishment in schools and sexual violence or intimate partner violence. The predominant reporting was about gender based violence in the home, although this may have been the result of gender based violence discussions being conducted prior to the assignment of writing violence stories.

In the episodes of household violence described, there were stories written about fathers (often intoxicated) beating their children, but more frequent mention was made of conflict between husbands and wives over gendered roles and expectations perceived as being unmet or transgressed. A man’s role was described as being responsible for providing family income, including shelter, food, school fees and other necessities; while a woman’s role was described as responsibility for the household, and included deference to the man. As John and Deo (aged 16 years old) two peri-urban students described, conflict around the expected (and deemed unmet) gendered roles within the household sometimes results in the observation of violence within the home:

Some of the fathers are irresponsible or don’t work – they leave home with no money for food so if he gets back and finds the woman cooking cheap food he yells at her. So she says he didn’t leave money so she cannot cook other food. So he beats her.

One day a father came from the job and found his wife in the kitchen cooking…After some minutes he got back in the kitchen and started beating his wife, saying why didn’t she come and take the bag when he was arriving. So he put the family in a hard situation, and the children start having fear.

The second quote refers to a cultural gender norm around a wife being expected to greet her husband when he returns to the home. Other reported domestic violence involved women initiating the violence, often in reaction to income disputes or alcohol intake by men. Such modeled incidents of violence were often observed to revolve around the power underlying household decision-making and use of income. As one peri-urban student, Mussa aged 17 years old, wrote:

One day I witnessed a husband slapped by his wife because the husband took the money, which the wife got from the daily work on the farm…and contributed to the wedding of the son of his friend who is his partner in drinking. The wife said she had saved the money for their son.

While there were many more stories of men initiating violence than women, it is important to note the use of violence in the household by both genders as a behavioural norm being role modeled by parents (intentionally or unintentionally), potentially socialising children’s perceptions of a normal resolution of conflict. As one peri-urban student James, aged 16 years old, described:

[My friend]…heard his father telling his mother ‘you stupid woman.’ The boy asked his father…why are you slapping Mom? The father did not answer and went away. [His mother] told him that his father had used all his salary for drinking. His father had made the mistake …but did not want to be corrected. That is why he took a step of punishing the mother.

The boy added that the behaviour was not appropriate and that “we should accept our responsibilities as men and not quarrel when we are corrected by women.” This suggests a potential shift in the younger generation’s notions of gendered engagement, along with an opportunity to explore with boys themselves what their ‘responsibilities’ entail within a changing society. The notion that women cannot correct men, particularly within the household, arose as a perceived trigger for numerous stories of modeled violence and in the gender based violence discussions.

Two models of violence arose within the school environment, and included the use of corporal punishment of students and peer violence amongst boys. While corporal punishment was described by many teachers and other adults as necessary for discipline and academic achievement, its usage is another representation of the enactment of control; this time adults controlling or exerting their power over young people. Although not necessarily gendered in its application (although young people did report different acceptable punishments for each gender), its usage potentially reinforces the socialisation of boys (and girls) in the belief that violence is an appropriate form of conflict resolution, in this case, conflict over expected versus enacted classroom behaviours and academic performance. Its usage was observed in punishing entire classes when a majority of students fared poorly on an exam, or in disciplining students arriving late to school or not adhering to the uniform code. Some boys suggested that corporal punishment (which is restricted by the government) can augment violence a boy is already experiencing and is a form of discipline that can negatively impact their academic performance. As Joseph, aged 18 years old and a rural student, wrote:

… sometimes you will find a father…when he comes home drunk, he tries to beat the mother. And when you see that, you feel bad. So you can even stop studying and start to cry. And when you come back to school, you may be punished because of poor performance. But it will be the family which caused it.

A number of teachers (as noted in the field notes) shared concerns over the new government limitations on the usage of corporal punishment, expressing the difficulty of disciplining students who cannot be adequately punished for wrongdoing. Despite the new restrictions, its usage was observed in the majority of schools visited over the course of the fieldwork.

The descriptions of modeled peer violence in school, often the result of disagreements over girls or money, were likely related to proving one’s manhood and status amongst one’s peers. The tendency towards violence in these circumstances may be influenced by boys’ observations of conflict resolution within society and social norms reinforcing aggressiveness as a manly trait. One peri-urban student, Baraka aged 19 years old, described an incident of peer violence in response to thieving in a dormitory that mirrored the violent (mob) response to thievery noted to exist in society:

It happened in school when one boy had the behaviour of stealing from his fellow students. One day he stole a camera from a friend and his friend realised it. So his friend went to another student …they decided to organise how to make him pay…they went where the boy sleeps and started beating him with pieces of wood. The boy was beaten to the point of fainting and bleeding…the teacher took the boy to the hospital. So my advice is to never take the law into your own hands.

In Tanzanian society, stealing can be perceived as a legitimate reason for others to beat up an apparent thief (Ng’walali & Kitinya 2006), a form of mob violence that spontaneously occurs. This was evident in an incident described by a rural school dropout:

I was in a certain place and heard voices say “he should be killed!” Then I saw someone was being beaten with logs and machetes…the man died. When police got to the place, the people said this person was a thief, he used to steal motorbikes. The police said, why did you take the law into your own hands? [The crowd] said, it’s a must for a thief to be killed.

Although not necessarily normative in terms of representing an ideal standard or behaviour, such incidents nevertheless may socialise boys into the notion that responding to an event with violence is acceptable. A recent study of homicide reports and morgues indicates that young men in fact appear to disproportionately suffer in terms of mortality from mob violence (Outwater et al 2008).

Other types of community modeling of violence were described, including the excess alcohol intake by men (and sometimes women) augmenting what might otherwise be a non-violent resolution of a disagreement. Many boys described observing such violence at home breweries, bars and discos. As the below peri-urban student Robert (aged 17 years old) describes:

There were some people in the bar and they started fighting. Some of the people started to solve the problem, and as you know, when people are drinking, if you want to know if the drunkard is serious, try to take away his drink…one of the men broke his leg. One of the drunks was teasing the other that he caught his wife with another man.

Along with the implication that heavy alcohol intake is related to unnecessary violence (or being modeled as conflict resolution), the above story links to norms related to a man’s sense of his own manhood, including his reputation and standing among his peers with regard to his virility and control over, and being respected by, the women in his life. These findings are similar to the analysis of how the changing socio-economic realities in Tanzania are shifting men’s enactments of masculinity as other forms of manhood become more elusive (Silberschmidt 2001). Declining land availability and diminished global coffee prices have negatively impacted men’s abilities to financially support their families in Kilimanjaro (Vavrus 2004).

Rationales for Violence and Masculinity Norms

There were numerous rationales provided by the boys for why boys and men engage in violence, with many, as noted, rooted in societal norms about gendered roles, and more deeply in issues of control, power, and dominance. Boys described poverty and income inequality contributing to the use of violence, including men’s need to feel respected, however they perceived such usages of violence as less normative and more the result of structural realities in men’s lives. They explained corporal punishment in school as teachers wielding unfair power over them. They described perceived unnecessary use of male peer violence in response to disagreements over money, and the use of violence in boarding school dorms, which can take the form of bullying by older boys over younger boys. As Ricardo et al (2006) have noted, young people may be able to explain the social pressures that lead to certain gendered roles and behaviours (in this case violence), however they often cannot perceive the complex socially constructed mandates creating such norms. In analysing the rationales provided, boys critiqued the use of violence in certain gendered scenarios, while in other instances explanations indicated a gendered socialisation supportive of the use of violence had already have occurred.

In the boys’ specific analyses of the observed gender based violence observed, violence appeared to result from women pushing back against their subordinate position, or taking on roles traditionally deemed to be for men. However in the range of the violence stories written, almost no boys wrote about their own home and parents. Given the extensive number of violence stories written about “a man and a woman” or a “father and a mother,” it is to be presumed that some of the boys were describing their own household but that using the third person made them feel less vulnerable. Other interpretations suggest that while violence in the home is widely seen as normal, it is still a perceived shame to admit it is occurring in one’s own home, or boys are not comfortable speaking about authority figures in their lives engaging in perceived to be shameful behaviour.

The reasons boys described for the witnessed household beatings often linked back to a perception that a man had been disrespected by his wife, which threatened his status in the home or community. Such perceptions arose for a range of reasons, including a woman not having food on the table when the man returned, a woman getting home late and perceived as having been cheating, or a father blaming a mother for not adequately caring for the children. A number of these stories were rooted, as noted, in the father not having provided sufficient income for the household, or a man coming home drunk and taking out his frustrations (through violence) on his wife or his children. A group of peri-urban students suggested men will understandably beat or slap their wives if they are perceived to be talking too much, which speaks to the ways in which men may attempt to exert control or demonstrate their manhood if they perceive themselves to be failing to meet the societal hegemonic model. Similar findings have been reported from Swaziland (Brear & Bessarab 2012). A few boys, both students and dropouts, suggested that lack of education was the reason behind domestic violence, although examples were also provided of university educated men who felt the need to physically punish their wives if they felt threatened by her pursuit of higher education.

The issue of perceived cheating, either in a marriage or a partnership, often arose as the excuse provided for witnessed violence. Such behaviour was likely interpreted as threatening to one’s status in the community. Although cheating was described as being practiced by both men and women, there appeared to be differing gendered norms around socially sanctioned responses to the discovery of sexual affairs. Both men and women were described as responding violently to such betrayal, although men much more so than women. As Wilson, a peri-urban student aged 18 years old, explained, “For the husband to cheat is normal but if the wife cheats, the violence is unstoppable.”

In terms of the rationales provided for the existence of sexual violence, men or boys were usually described as the aggressors, and girls or women as the victims. Various explanations were provided, with excess alcohol intake a primary excuse, or explanations of boys and men feeling the need to demonstrate their manhood among their peers. There was also a perception that girls and women could be to blame. As one peri-urban student, Charles, aged 19 years old, articulated:

Girls are the reason because they make an appointment at a guesthouse with a man, and then when they are in the room, they change their mind, so the man is physically and sexually violent.

This explanation of girls being to blame was revealing of underlying perceptions of sexual expectations among some boys. This is similar to findings in South Africa and elsewhere (Jewkes et al 2006). The source of these expectations was less evident, with some boys implicating globalisation and the media. One peri-urban student suggested that “…they see women on TV with short skirts and so in Tanzania the girls try to copy, so they are raped.” This same boy explained that short skirts are not the local culture and become a temptation for man. One rural student, Sam aged 16 years old explained:

Some of the other men, they take alcohol and get drunk, and then when they pass along the road, they meet with the girl, they rape her and injure her.

Such quotes indicate the need for more in-depth research on the pervasiveness of sexual violence and alcohol in Tanzania.

Solutions for Diminishing the Violence

The boys themselves had clear ideas about which aspects of society contributed to the violence around them. While they may not have explored the deeper issues of power and control that underlay much of the violence, many of the boys had recommendations for how to diminish the violence around them. More importantly, they expressed a wish to see less violence. Five overarching approaches emerged, although implementation suggestions were not necessarily delineated. There were also observations made that a diminishment of violence in the society, both child-focused violence and gender based violence, is already occurring.

In terms of household violence, the rural school dropouts suggested that although still prevalent, its occurrence is decreasing due to education, changing social norms, and parents not hitting their children as much. The boys felt that continuing domestic violence is related to a father feeling his status as the head of the household is being threatened. Some of the peri-urban students felt that increased education (of men) and an increased awareness and knowledge about the rights of men and women, would help to eliminate instances of household violence. These solutions do not necessarily get at the roots of domestic violence, however they may be starting points for intervention.

In addressing GBV, a rural student Emmanuel, aged 19 years old, had a more extensive vision of what needed to happen for its diminishment. His solution included governmental intervention and a shifting of the societal norms rooted in the traditional structures of the community. He described the following approach:

The government should create laws about people who are violating others in GBV, maybe to take them in jail for two years or to be fined not less than 300,000 TZSH. To create a group in the society who will be dealing with avoiding and eradicating violence between people of one gender and another. This group should involve members of the clan. People should participate in education and the place where education is provided like in church.

In terms of other recommendations for diminishing GBV, the rural students suggested three solutions, including there should be laws against such behaviour; excessive alcohol drinking needed to stop; and women should respect their husbands. The issue of respect and disrespect arose a number of times in the two sites, and suggests a holding onto traditional gendered norms by the boys. There were also recommendations related to sexual affairs, with a peri-urban student advocating that both men and women needed to change their ways for the violence to stop: “the solution is for the husband and wife to be honest, and not cheat.”

The issue of excessive alcohol intake was described as an individual and societal influence needing to be addressed. This related both to alcohol’s role in augmenting the level of violence, and the conflict arising when household income was spent excessively on alcohol. As one rural school boy, Frank aged 16 years old, suggested:

Most of the fathers are drinking and get home very drunk and when their wives ask them to stop drinking, they beat them – the solution is that men should stop drinking so there is peace and harmony in the family.

The easy availability, use, and socially acceptable norm of (excess) alcohol intake, particularly by men, was described as a primary cause – and hence focus for a solution to – the problem of violence.

The larger issue of control over household income (as a cause of violence) was suggested to have a solution as well, with one boy advising that men “should be giving money to spend to their wives, and not take the law into their own hands and start beating their wives.” The income and violence linkage was analysed more deeply by some of the boys as relating to larger structural issues of poverty within the society. As Daniel, an 18 year old rural school boy commented:

The violence happens because they don’t have employment in Tanzania today so most of the young men…decide to do some abnormal things like alcohol…[This] can help them to forget difficulties that they face in the society. So if there will be employment this kind of violence will tremendously decrease.

The final recommendation focused on discipline within schools, with a number of boys writing with dismay over their experiences of corporal punishment related to unsatisfactory academic performance. They had specific advice for teachers and parents, as expressed by Florian, a 17 year old rural student:

My advice to teachers is that they should stop punishing students because of poor performance. If a student usually fails, they should just teach him more so he can understand. And if it’s a person who was performing well, and then suddenly he performs badly, it’s also important to ask him why he failed and not just to come up and start to beat him.

The implication being that teachers need to both shift away from the use of corporal punishment as a response to poor academic performance, and move towards better understanding an unexpected drop in performance.

Discussion

This in-depth exploration of boys’ perceptions of and experiences of violence revealed a complex reality that has been observed by others in SSA (Barker & Ricardo 2005; Barker 2005; Jewkes et al 2006; Dunkle et al 2006), that of boys and young men’s simultaneous harboring of increased support for gender equality and more traditional gendered norms about a man and a woman’s ascribed roles within the society. The expectation of gendered behaviours was frequently described as the source of violence observed within the household, such as a man feeling disrespected by his wife. Other rationales provided for the existence of violence included excessive alcohol intake, perceived cheating, disagreements over money or women and household income disputes. While the research and findings from this study were primarily focused on boys from the Chagga ethnic group, boys from other ethnic groups participated in the study. This is important to note given the diverse range of ethnic groups in Tanzanian society (over 120), and the varying gendered dynamics around intake of alcohol, and other ethnic specific findings.

Although many boys described scenarios of violence linked to gendered norms within the society, ranging from husbands feeling they were owed a certain level of deference, to boys or men feeling the need to prove their manhood through displays of physical strength, the underlying roots of the violence appeared to be fundamentally related to issues of dominance, control and power, and changing economic realities in Tanzania that threaten the traditional enactment of manhood (Silberschmidt 2001). This was indicated through observations of men’s (and sometimes women’s) violent reactions to spending of household income, or perceptions over household roles being inadequately fulfilled. Such findings about the underlying causes of violence and specifically GBV have been documented in Tanzania and elsewhere in SSA (Ricardo & Barker 2009; Koenig et al 2003; Rani et al 2004). Although situations of peer violence and sexual violence committed by the boys were described, the predominant descriptions were of violence observed around them, or enacted against them. It is unclear if the methodologies utilized in this study limited the expression of interpersonal violence committed by boys themselves, or if in fact, the majority of the violence experienced does not include them as participants.

Almost all of the boys, when writing violence stories that referred to family violence, wrote in the third person, describing observations of violence rather than their own lived experiences. This may suggest there exists some level of shame or stigma around admitting violence exists within one’s own household. This analysis is supported by findings by Nduna and Jewkes (2011) in South Africa that discuss the silence around distress among young people, and the critical importance of showing respect for one’s guardians. Alternatively, a more reflexive interpretation might be that boys in the two sites perceived the research team as having preconceived negative views on any form of engagement of violence, and did not feel comfortable admitting to violence within their household. Other violence-related literature highlights the sensitivity of the topic of violence, particularly with vulnerable populations such as women or children, and the challenges of assuring valid information is collected without traumatising the participant (Jewkes et al 2000; Sweatt et al 2002).

A number of additional societal or structural factors appear to be contributing to the prevalence of violence. This includes the widespread availability and affordability of alcohol (home brew and commercial brands) and the socially acceptable nature of alcohol intake for men, although excessive alcohol intake appears to carry some level of stigma (Mitsunaga& Larsen 2008). Kilimanjaro has been documented to have one of the highest levels of alcohol intake in Tanzania (Henricson 1992). The link between alcohol and violence arose both in situations of GBV and in disputes between boys and men. Other structural factors include economic disparities, with boys suggesting societal economic challenges negatively impact on household relations, and lead to incidents of violence. An additional societal issue is the practice of corporal punishment in schools, although its use is the target of government eradication efforts.

Although less emerged about the prevalence of sexual violence, Ricardo & Barker (2009) suggest that its prevalence is widespread in many contexts in SSA, with reasons including its social sanctioning, the influence of media images denoting sexual violence as normal, and changing gender roles. Increasing research from the region describes and analyses the root causes of perceived increasing levels of sexual violence in many SSA countries (Sikweyiya&Jewkes 2009; Brear&Bassarab 2012; Dunkle et al 2006). Given comments that were made by some boys about girls having provoked the occasions in which they were the victims of sexual violence, additional studies with Tanzanian boys on this topic are warranted. It should also be noted that many boys commented on the problematic nature of rape, suggesting many would like to see an end to such sexual violence. Less was conveyed about sexual violence against boys and men, something identified in the Violence Against Children study but likely considered too taboo for boys to discuss in the study’s activities (VAC 2009).

Although numerous instances of gender based violence were described in the study, it is difficult to know the extent of this violence across society when the focus is only on one part of the gender equation, or girls and women. The most recent Tanzanian DHS (DHS 2010) contained a gender based violence module that focused solely on girls and women, with no mention of boys’ and men’s experiences as possible victims of these forms of violence. Recent findings from a gender based violence study in Tanzania also highlight the challenges of violence faced (solely) by girls and women (USAID 2008). Additional research is therefore needed that involves the physical and sexual violence experiences of boys and men.

Conclusion

An overarching recommendation from this study is the need for additional research on the masculinities and other social norms influencing boys’ and men’s engagement in violence in Tanzania. This would include specific attention to boys’ actual engagement in interpersonal violence, along with their experiences or observations of violence around them. Based on this study’s findings, there is also a need for structural and individual level interventions that will influence a shifting of local masculinity norms as they relate to violence, including economic and other factors that may be shaping enactments of manhood. It is essential to note that many boys expressed dismay over the violence around them, and had recommendations for how to diminish the forms of violence that continue to pervade the society. Their voices – as potential future enactors of violence– are the most important to capture and address when finding solutions that will create a more conflict free society in the future.