Boys and violence : the construction and display of masculinities in rural Burundi

(1)

BOYS AND VIOLENCE:

The Construction and Display

Of Masculinities in Rural Burundi

Master Thesis | Markéta Sošťáková

(2)

COLOPHON

Thesis submitted on the 19th_{of
November,
2014
in
part
fulfillment
of
a} M.Sc. in International Development Studies

Graduate School of Social Sciences University of Amsterdam

BOYS AND VIOLENCE: The Construction and Display of Masculinities in Rural Burundi Supervisor: Dr. Jacobijn Olthoff Supervisor: Dr. Lidewyde Berckmoes Second Reader: Dr. Ria Reis

Cover Pictures: Markéta Sošťáková

Date November 2014 Markéta Sošťáková 10701176 marketa.sostakova@gmail.com

(3)

ABSTRACT

Motivated to gain insight into the gender aspects of the ‘cycle of violence’ in Burundi, this study adopts an interdisciplinary multi-‐scalar approach to explore the links between the use of violence and the construction of masculinities among primary school boys. Applying Bronfenbrenner’s bioecological ‘Process-‐Person-‐Context-‐Time’ framework, the study offers a comprehensive understanding of micro, meso and macro factors, relationships and their dynamics shaping boys’ masculinity in relation to violence over the period of middle childhood and early adolescence.

Based on in-‐depth interviews, participatory sessions and a survey, the study searches for trajectories to non/violence particularly in the micro environments: it examines the boy’s individual resources, his socialization in the current gender regime and his outlooks on the context of (post-‐conflict) violence in the community, school and home. Findings of this research suggest the general need for more complex and contextual understanding of the links between African masculinities and violence. Firstly, in contrast to dominant academic literature on African hegemonic masculinities, the study offers evidence that masculinities in Burundian rural areas are not in the first place connected to violence but to the traditional provider’s role, the values of morality and hard work associated with that. Violence against women is not a sign of manhood but a weakness. Secondly, diverging from the portrayal of young men as danger to themselves and the society, the qualities of hegemonic masculinity in Nyabiraba are affected by the context of rural poverty and lack of economic opportunities. Education is a key to the prison of rural life. Hence, it is primarily the educational achievements to prove boy’s talent and masculinity. Nonviolence is either connected to their repulsion by violent behavior and/or reflection on its consequences for their educational chances. Boys lacking academic skills and talents are more likely to take advantage of their physique to prove their masculinity to reach the same recognition. Thirdly, the study analyzes the display of this complex concept (masculinity) in the moment of a conflict. How the situation unfolds, seems to depend on the nature of the conflict (connected to individual’s breaking point), the situational factors (mostly the presence of a peer group), and the individual construction of masculinity.

The study is an important contribution to the literature on gender regimes and hegemonic masculinity in Burundi. It also adds to the child development literature as it provides complex view on the construction of boys’ masculinities in the rural, post-‐ conflict areas.

Key words: construction of masculinities, non/violence, rural Burundi, children, bioecological framework, ‘masculinity in action’

(4)

ACKNOWLEDGEMENTS

Greeting me with big hugs and warm three-‐step handshakes, I would like to thank all boys and girls from Mayemba, Nyabibondo and Nyabiraba, who welcomed me in their lives and have been sharing their stories with me. Their hope and determination to improve their fates despite the conditions of deprivation had been motivating me day by day to find out more about their lives.

I would also like to thank the professionals in Burundi who have shared with me their opinions and perspectives on violence and its prevention. I admire their hard and exhausting work for Burundian children and communities. I hope I can contribute to their work with my findings, to repay the kind cooperation, time and attention they had given me.

My big gratitude goes to my supervisors. I am thankful for their aspiring guidance, insights into the research world and advice from the first meetings until the last checks. I would like to express my appreciation to Jacobijn Olthoff, for her stimulating suggestions, for keeping me on track and focused. Your work and knowledge on children are a great inspiration to me. I am thankful to Lidewyde Berckmoes, for her encouragements through the fieldwork but more importantly, for making the Burundi project happen and giving me the opportunity to be part of it. My acknowledgements also go to all my friends who accompanied me throughout my studies in the Netherlands and my fieldwork in Burundi.

Finally, I thank to my family: my parents and my brother who have been supporting me throughout my entire studies and on whom I can always rely on, no matter where life takes me.

(5)

LIST OF FEATURES

LIST OF BOXES

Box 1: Critical sympathy. Box 2: A girl ‘mzungu.’ Box 3: My translator is late.

Box 4: Claude’s opinion on gender roles. Box 5: The stick.

LIST OF FIGURES

Figure 1: Bronfenbrenner’s ‘ Process-‐Person-‐Context-‐Time’ model. Figure 2: Conceptual Diagram.

Figure 3: Structure of Bronfenbrenner’s bioecological layer (Person) in this study.

LIST OF GRAPHS

Graph 1: “Do you think the father should have greater say in making family decision than the mother (f.ex. what to buy with earned money)?”

Graph 2: “How many times a day your family eats?”

Graph 3: “Do you think other kids will think a boy is afraid if he refuses to fight?”

LIST OF MAPS Map 1: Burundi.

Map 2: Nyabiraba commune.

LIST OF PHOTOGRAPHS

Photo 1: Children running to morning school assembly in Nyabiraba. Photo 2: Learning, playing, observing.

Photo 3: Brainstorming, participatory session.

Photo 4: Drawing ‘The Real Man’/ ‘The Real Woman.’ Photo 5: Tea fields in Nyabiraba.

Photo 6: Transportation in rural areas can be challenging Photo 7: The cost of schooling is excessive

Photo 8: Brainstorming. Photo 9: Drawing.

Photo 10: Stakeholder analysis

LIST OF TABLES

Table 1: Themes, participants and location of participatory group sessions.

(6)

ABBREVIATIONS

CE Civic Education CFS Child Friendly School

CNDD-‐FDD Conseil National Pour la Défense de la Démocratie–Forces pour la Défense de la Démocratie (National Council for the Defense of Democracy–Forces

for the Defense of Democracy) CP Corporal Punishment

FNL Forces nationales de libération (The National Forces of Liberation) PBEA Peacebuilding, Education and Advocacy Programme

PPCT Process-‐Person-‐Context-‐Time (model)

UNESCO United Nations Educational, Scientific and Cultural Organization UNICEF United Nations Children's Fund

WHO World Health Organization

(7)

1 INTRODUCTION

1.1. MAKING THE CONNECTION: MASCULINITIES AND VIOLENCE

Since its independence in 1962, Burundi has been suffering from ‘cyclical violence’ (Baghdadli et al., 2008; Daley, 2008; Vandeginste, 2009). Experiencing outbursts in 1965, 1972, 1987 and 1993, the country is now left with a complicated legacy of ethnic and political disputes after the last devastating conflict ending in 2005. Violence can be triggered by various pretexts—nationalism, poverty, ideology, racism and other inequalities to name a few— but given the concentration of weapons and the practice of violence, “gender patterns appear to be strategic” as the dynamics of violence are shaped by masculinities (Connell, 2000: 29). In conflict situations, masculinities may act like ‘gasoline on fire’— escalating aggression and cause difficulties to peace resolutions (Holter, 2000: 61). Scaling down from group struggles to interpersonal violence, men are also the main perpetrators of violence against other men and against women. With the presence of violence at home and in the community, children in Burundi perceive violence as part of their daily life. They experience it, they witness it but also, they perpetrate it. For mistakes at school or at home parents and teachers physically punish children; poverty can motivate them to engage into exploitative relationships, or being abused by their peers and adults. Violence and the fear of it is indisputably affecting children’s wellbeing and perspectives as it has long-‐term consequences for pupils’ physical and mental health, development but also concentration, and participation.

Children are also perpetrators themselves as many conflicts among them are resolved with hurtful insults or fierce fights. Small arguments and conflicts where use of violence is the easiest, or the only option considered, are fueling a vicious circle where fighting brings on beatings, beatings bring on anger which brings back fighting. By some children, violence is understood as a primary way to negotiate in relationships. Even in such context, violence does not ‘just happen’ but the way events will turn out is strongly influenced by boys’ masculinity (Hamlall and Morrell,2012: 3).

The current context of normalized violence and the outbreaks of conflict on a generational basis in Burundi are standing behind the motivation to this research. Analyzing the predispositions of the Burundian state to its past genocides, Daley supports the importance of examining local masculinities, as she believes it is the outcome of the intersection of race, ethnicity, the patriarchal state, masculinity, the geopolitical economy and militarism (2008: 231). Such investigation includes a challenge, according to Belgian scholar Peter Uvin, who has been extensively working

(11)

in the region of Rwanda and Burundi. He presents field data contrasting the dominant view on young men usually portrayed as a danger to themselves and the society, concluding instead that dominant academic views on masculinity and violence “does not hold in Burundi” (2013: Kindle 2591-‐2597). In his view, Burundian masculinities are not primarily violent and the academic generalizations on African masculinities are far beyond acceptable for this particular local context (Ibid.). With the collected field data I am aiming to gain insight into this complex situation: first, by examining the factors and mechanisms behind the construction of violent or nonviolent masculinities of primary school children and second, through their display. The main research question guiding this research is therefore as follows: “In which ways, and by what personal, situational and sociocultural factors are boys’ masculinities constructed and displayed in the rural context of Nyabiraba commune, Bujumbura Rural?”

Applying Urie Bronfenbrenner’s bioecological model to this study, it allows me to capture the complex construction of one’s masculinity throughout the layers of multiple environments, proximal processes and personal characteristics in the timeframe of childhood. The dynamic research design of this study corresponds to the intentioned use of the model for “science in discovery mode” (Bronfenbrenner and Evans, 2000: 999), instead of the verification process.

Masculinities are fragile and fluid constructs, often remaining an abstract concept. For this reason I choose to take an approach of ‘masculinities in action’— therewith providing the evidence when gender identity is uncovered and masculinity is played out, either through the use of violence or nonviolence.

A central starting point of this research is that the character of one’s own masculinity is a choice, not a biological predisposition, therefore the promotion of non-‐ violent forms of masculinity could help to alter the character of much of the violence. As Connell argues “the task is not to abolish gender, but to reshape it; to disconnect…courage from violence, steadfastness from prejudice, ambition from exploitation (2000: 39).” Development practice around the world can support the initiative with evidence of several promising interventions engaging with boys and men, who reported significant change in their attitudes to the use of violence (Promundo, 2012). Adopting interdisciplinary and multi-‐scalar approach to understanding the local aspects of masculinities and violence is highly relevant to policy makers and INGOs as these actors contribute to the search for better practice in violence prevention.

With my study, I would like to contribute to a safer and welcoming environment, where children can grow up without fears of their parents, teachers or each other, and moreover to a deeper change towards more just and equitable society.

(12)

1.2. OBJECTIVES AND RELEVANCE OF THE STUDY

1.2.1 Objectives

Children, who currently constitute a majority in Burundi1 will become tomorrow’s adolescents, and finally Burundi’s men and women. Working with children and their socialization in their early age can be fruitful when preventing violence, considering that previous research in the trajectories of violence is predicting high levels of aggression among adolescents if it was already high during their childhood (Brame, 2001). There is no evidence of the ‘late onset’ from transition from a low physical aggression in childhood to high trajectory in adolescence (Ibid.).

I associate myself with the belief that children are important social actors in their own right in context, where traditionally they are denied those rights of participation and their voices remain unheard (Alderson in Christensen and James, 2008). Only by understanding their realities and strategies of dealing with conflicts, we can confront the issues connected to masculinities and violence. It is then integral to view boys as experts on their own lives and find a meaningful way to consult with them, not through their adult caretakers. For that, it is appropriate to adopt participatory, child-‐centered and action-‐oriented approach especially with regard to my research goals: to collect boys’ narratives and thick descriptions of life attitudes, perceptions and beliefs considering violence, examine all aspects of boy's socialization, context, factors and mechanism, influencing boy’s masculinity.

1.2.2. Academic relevance

This research is a relevant contribution to the academic literature on the development of children’s masculinities in rural and post-‐conflict context of Burundi, a country largely overlooked by donors and academia.

Despite the efforts of cross-‐cultural psychology, the studies on children’s construction of gender identities are still limited outside the Western context, the less for Africa, or Burundi. Even then, the variety of the sources is hardly living up to the academic literature on childhood that has been forming in the West for decades.2

As Niewenhuys pointed out more generally, studies of childhood in the South seems to come from “ ‘lesser’ disciplines such as social work and social anthropology” (2013). Instead by inventing separate specializations, this disproportion could be balanced by

1_{The
median
age
in
Burundi
is
17
years
(Index
Mundi).}

2_{Such
notions
apply
also
to
other
disciplines
that
this
study
is
concerned
with,
for
example
rural}

gender relations. Until now, it is has been connected only to developmental issues (health, income, etc.) in Africa. On the other hand some of the predominantly Western literature in that

(13)

the eclecticism of the post-‐colonialism approach (Viruru, 2005). This study aims to contribute to that by freely borrowing and mixing concepts (Bronfenbrenner’s model originally from psychology, the concept of hegemonic masculinity from gender studies), data (qualitative and quantitative) and methods (traditional and participative) in order to combine them in one study.

This study aspires to collect data by applying an advanced Western, bioecological model well known among psychologists, in its full strength (but culturally sensitive—Triandis and Brislin, 1984) to children’s development in Burundi. Such attempt aims to contribute to bridging the asymmetry between the way academics study childhood in the West and in the South.

The gendered examinations of schooling in the developing countries have focused in large measure on the unequal access or unequal conditions in the classroom, gender based violence of boys and male teachers against girls3_{.
All
of
these
are
very}

pressing and serious issues, however the literature on violence among and between boys (particularly at schools, not urban street gangs) is less developed though there is a large global literature on boys at school, which includes some reference to discussion of competitive and aggressive masculinities (Martino and Meyenn, 2001, Swain 2004, Hamlall and Morrell, 2012).

This research also seeks to address a void: the strong link between boys, masculinities and violence can obscure important moments when boys avoid violence. Previously quoted Uvin sees the scholarly literature on masculinity in Africa ‘exaggerated and miserabilizing’ (2013: Kindle 3186-‐3196). This study uncovering both choices of violence and nonviolence aims to add knowledge and perspective to the academic sources.

1.2.3. Contribution to practice

The research directly adds to the development practice, as primary data collected will be shared with the partners of the Peacebuilding, Education & Advocacy Programme (PBEA). Given the recurrent outbreaks of violence in Burundi, this research is embedded in a larger project of multi-‐systemic analysis of the intergenerational transmission of violence and resilience aiming to build knowledge and enhance the design and impact of UNICEF interventions in Burundi.

(14)

1.3. OUTLINE OF THE CHAPTERS

The study is organized into seven chapters, the first of which (i.e. the current chapter) introduces the connection between violence and masculinities and its possible use for the prevention of intergenerational, interpersonal and structural violence. Chapter 2 gives a theoretical base to the main concepts and frameworks in this study. First, it elaborates on the multiplicity, hierarchy and fluid formation of masculinities. Second, the chapter defines the concepts of violence to establish the extent of this research at school, in the community and at home. It also includes the explanation of Bronfenbrenner’s bioecological theory as a tool for analysis but also an organizing principle for the empirical data. Lastly, all concepts are set together into a conceptual scheme visually showing the masculinity in construction and ‘in action.’

Defining more precisely the scope of this study with a research question and subquestions, Chapter 3 contains the methodology of the conducted fieldwork. It encompasses the methodological and ethical concerns of working with children and the reflection on the choice of methods and their execution. It also introduces the location and the practical aspects of the research that are important for the understanding of the obtained data.

Framing the context of masculinities, Chapter 4 is focusing on the first layer of the of the bioecological framework— community in Nyabiraba, Bujumbura Rural. It presents two aspects of social relations in the community, land disputes and impunity. It goes on discussing the ideal of hegemonic masculinity as a fundamental part of gender regime in the community to grasp the understanding of boys’ ideas of manliness. Chapter 5 presents the remaining two environments of the framework— the layer of school and family. This chapter provides the main building blocks for understanding both violent and nonviolent masculinities as it elaborates on the dynamics of the proximal relationships but also the personal characteristics. Personal characteristics can significantly influence other elements of the framework, and shapes the development of child’s identity. The chapter highlights the complex connections between education, rural poverty, future prospects and the consequences of violence. Chapter 6 is placing the boys in the moment of conflict, describing how their masculinity is challenged in provocations, and which situational considerations boys make before responding to aggression. Bringing together all empirical data of the study, the final section of the chapter organizes the main patterns behind the construction and display of nonviolent and subsequently, violent masculinities (serving partially also as a conclusion).

Finally, Chapter 7 concludes the study by summarizing the primary findings, relevant for wider academic debate as well as giving recommendations for developing nonviolent masculinities for policy and practice in peacebuilding interventions.

(15)

2 THEORETICAL FRAMEWORK

The two main theoretical concepts of this study are masculinity and its intersection with violence. The following parts will depict masculinities as a fluid and dynamic construction in everyday situations as well as their hierarchies among men and boys. The chapter goes on to explain the elements of the masculine identity, its mechanisms and how these are constructed within key sites in diverse relationships. Secondly, this chapter introduces the concept of violence and how the moment of provocation confronts a boy’s masculinity. As a tool for analysis, the chapter elaborates on the elements of ‘Bronfenbrenner’s’ Person-‐Process-‐Context-‐Time framework. The chapter concludes with a diagram visualizing the interaction of all concepts.

Unfortunately, this theoretical section will be biased towards the general patterns of masculinities identified in the literature coming predominantly from the West. Such choice is driven by reasons explained in section 1.2.2. but also following Uvin’s criticism on overly generalized literature on African masculinities. Rather than preoccupy one with generalizations, I aim to employ an exploratory research with the blend of methods and theories, as post-‐colonialism and cross-‐cultural psychology suggest, gaining a locally specific view (Triandis and Brislin, 1984). The African context is not left out, though. The chapter includes contemporary academic discussions on the link between African masculinities and violence, highlighting several challenges for the analysis and interpretation of empirical data.

2. 1. MASCULINITIES

“Masculinity is what any given society accepts as features associated with the male gender and expressions of maleness” (Uchendu, 2008: 3).

2. 1. 1. ‘The gender politics’

In recent years, the theoretical framework on gender has moved beyond the biological determinism of ‘boys will be boys’ and beyond simplistic views of socialization. The most recent view on the concept claims that masculinities are neither biological nor rigid social structures. Instead of focusing on masculinity as a fixed characteristic or a norm, this approach examines the dynamic construction of masculinity in everyday interactions (Hearn and Seidler in Connell, 2005: 39). Masculinity is a diverse, contextual, and fluid concept— constantly changing according to where we are, what

(16)

we are doing and whom we are with. Men and boys have a multiplicity of masculinities employed and enacted at different times (Paechter, 2003: 69) and make situationally specific choices in responding to social practices, rather than confining to one particular masculinity (Wetherell and Edley, 1999).

Boys are not a homogenous group. Gender identity is defined by an intersection with other social structures, factors and life experiences. Even though they may often start from similar positions, men take unique paths that form their gender identity. Yet, acknowledging the multiplicity of masculinities is only the first step of understanding ‘the gender politics’ in the gender relations. That is, not all masculinities are appraised equally: depending on the cultural dominance in the society, different types of masculinities are either hegemonic or subordinate (Connell, 2005: 76).

In the concept of Australian sociologist Raewyn Connell, hegemonic masculinity occupies the dominant position in gender relations in specific cultures. It is the idea of the ‘Real Man’ to which any other masculinity is being compared. In practice, the ‘subordinate’ forms of masculinity are classified according to their closeness to the hegemonic form. At the end, it may only be a minor group of individuals practicing and embodying this model, as the expectations of this kind of masculinity are impossible to meet (Kaufman, 2000: 214). In societies where masculinity is valued more than femininity, it also entails a greater emphasis on proof (Paechter, 2007: 19).

Most men, boys especially, doubt their masculinity and fear ‘not being manly enough’ (Ibid.). A boy’s anxiety about being ‘weak’ or ‘weird’ means he is trying not to be too different from ‘the hegemonic ideal’ (Phoenix and Frosh, 2001: 34). Boys gain their understanding of hegemonic masculinity by observations and encounters with men in their community. This practice leads boys not only to understanding of masculinity but also embed them in the surrounding culture as they try to follow this model (Paechter, 2007:6).

In a boy’s life, this search for a hegemonic, acceptable form of masculinity is the equivalent to gaining popularity and status (Epstein and Johnson, 1998), which is one of the most important parts of school or community life (Adler and Adler, 1998). The concept of the peer group is crucial for the construction of masculinity, as this group carries the gender definition of boyhood and its collective meanings (Morrell, 1998). Through these groups, boys tend to comprehend what it means to be a boy in terms of their values and interests. One must adopt specific characteristics and behaviors to be accepted into certain groups by its members. Through ‘intricate and intense maneuvering’, boys can earn a certain status in their peer group, which needs to be further sustained almost on daily basis (Swain,2003). For instance, if a boy has gained and maintains his status with fighting, he needs to be alert of potential rivals as his status could devaluate if he gets beaten (Ibid.). Kaufman calls it an “enormous terror,” when boys are frightened of being laughed at by others or even target of violence from

(17)

the others (2000:220). Therefore, boys are pressured to constantly demonstrate/ prove their masculinity which can lead to direct violence or to domination in spaces and ways that are not considered violent (such as triumphs at sports or dominating class discussions etc., Ibid.).4

2. 1. 2. The ‘making of ’ masculinities

Key sites

Considering the age group in this study (primary school children), there are three key sites for their formation of masculinities: the family, the peer group and school (Paechter, 2007: 2). The importance and influence of each site changes over time. As this research is focusing on the stage of middle childhood and early adolescence, the formal institutional setting of the school and the peer groups become at least as influential as their parents in the transition of home to school (Harris, 1998; Gilbert & Gilbert, 1998). Schools are central arenas for developing masculinity, as they set ideas about what it means to be a boy/ a man in addition to what children already observe in their family environment.

Relationships

As adults mostly structure all three just listed environments, the child’s understanding of appropriate behavior comes from observation of specific men and women (Paechter, 2007:14). In their gender practice, children combine both conformity and resistance to construct their identity in relation to others within the wider social context of spoken an unspoken rules and norms. In a boy’s life, there are particular relationships that have greater influence upon his identity. In the families, children are able to observe patterns of gender behavior from their parents from as early as 2 years old which is the stage in their development when they are starting to develop their own gender identity (Martin et al. 2002).

In early childhood, a child learns in this environment about different forms of masculinity and will start to navigate between the already developed form and the acceptable one in his new environment. At school, teachers can explicitly or implicitly draw gender boundaries according to their own perceptions and employ institutional/organizational practices to regulate these boundaries. Also the official

4_{While
there
exist
various
sources
of
boy’s
masculinity,
not
all
of
them
lead
to
dominance.
The}

(18)

curriculum aims to lead students towards diversity but sameness, where extreme forms of masculinity are constrained by various sanctions and punishments. Some boys may bring to school masculinity that is in conflict with the ethos of the school and pupilhood (Warrington et al, 2003). Lastly in the section on relationships, children might adopt and follow the model of the most powerful members in their local community.

For children in middle childhood, school is the primary site of gender practices where they interact with other peers and older children. Boys in their early adolescence spend more time with their peers group than with their parents. Therefore, some theorists (see Piaget, 1997; Harris, 1998) believe, that peers can play equal or even more important role in that particular stage of a child’s development than parents. In comparison to the child-‐parent relationship, which is typified by its asymmetry of power, the relationships between peers should be more balanced as by definition they are ‘the social equals’ who operate on the same level of behavioral complexity (Lewis and Rosenblum, 1975). Yet, some (especially older) pupils know ‘the rules of the gender game’ and will point out errors in what is boyish and what is not (Davies, 1989).

Being accepted and liked by their own peers is one of the most important parts of a child’s social life. Boys, who search for peer acceptance or prove their loyalty to their own peer group, have to display a similar masculinity to one of the group, which in return validates their individual form of masculinity. In some groups aggression may be a way to recognition (Cairns, Neckerman, Gest, & Gareipy, 1988), the same way peace is for other peer groups (Morell, 2012).

Generally, whether a child will or will not be popular is dependent on many factors. However academic and social skills are consistently being main components of a child’s popularity while aggression (especially in combination with the lack of social skills) is negatively perceived, with the exception of early adolescence (Shaffer and Kipp, 2014: 574). As popularity changes with age, the period of early adolescence provides a special ‘window’ for ‘tough boys’ who use aggression as a tool to achieve popularity among some of their peers (Ibid.).

Factors to violence and nonviolence

A single factor cannot explain why some individuals adopt violence; however, heterosexual dominant masculinity should be ‘the foundation for any realistic theory of violence’ (Heise, 1998:263). In a study of South African schools, Hamlall and Morrell identified the key factors in rising up to a fight as a response to provocation: “the form, the social and physical setting, the school’s gender regime and, critically, the investment of individual boys in particular constructions of masculinity” (2012). Boys are willing to secure, advocate or obey to certain values as part of the process of constructing masculinity (Swain, 2003). In their socialization, they compete with (or

(19)

position themselves over) others as they feel ‘entitled’ to resources and status, which is legitimized and associated with maleness by the society as in Uchendu’s definition of masculinity (2008:3). Violence then, can be performed to gain or secure this power, or as a defense against losing face. It is more prevalent especially in the context where both perpetrators and their witnesses see it as a legitimate mean (R.W. Connell 2000). The role of family, intervening against the use of violence sends a strong signal against aggressive behavior of boys (Gorman-‐Smith et.al, 2004). Moreover, research has encountered higher educational attainment, having a skill, a realm of competency to be an important factor in shaping non/violent behavior (Connell, 2005: 131). Some men have constructed their nonviolent gender identity as a reaction to trauma—witnessing masculine violence in their environment (Ibid.). Self-‐reflection on the ‘costs’ of violent hegemonic masculinity may thus also lead to perceiving violence and such form of masculinity as unacceptable. Alternative gender role models from their family or friendship groups who support or inspire boys to nonviolence (whether male or female) is one of the ‘keys’ to nonviolence, as men must rely upon other men for emotional support (Kaufman, 2007:50).

While individual psychology certainly plays a role in constructing masculinities, Seagal emphasizes the circumstances and institutional dimensions of masculinity, as masculinities are modeled in response to changes in the attitudes of families, other men around them, community and society in general (1990). Having a meaningful connection to a mainstream social institution will be interesting dimension to explore especially in the Burundian context that is marked by “broken institutions and indeterminacy” (Berckmoes, 2014: 49).

2.1.3. African masculinities and violence

While studies have outlined several characteristics that are cross-‐culturally associated with manhood5, there is no universal masculine entity (Connell, 2005: 43). In Connell’s view, the Western ideas of masculinity are dominant throughout the world, yet the local patterns of masculinity in the developing world are shaped by globalization, imperialism, colonialism, and international migration (2005). The experience of colonization pressed African men to accept the status of subordinate masculinity (Daley, 2008: 29). As Daley argues, factoring in almost thirty years of military rule, Burundian masculinities have taken on more violent and oppressive forms that become

5_{Characteristics
such
as
being
a
provider,
achievement,
recognition
in
front
of
a
social
group}

(20)

locally hegemonic using modern weapons of war and becoming further entrenched in all areas of the society (2008: 123).

Most of the classical academic literature adopts the violent view of masculinities, which are triggered by frustration and deprivation in the context of economic crises. The hegemonic ideal in Africa is that men “should take risks, endure pain, be tough or stoic, or should have multiple sexual partners...to prove that they are “real men” (Barker and Ricardo, 2007: 8). Sources on African masculinities overwhelmingly ascribe men’s violence in the household to the inability to fulfill the widely expected role of providing for his family in the changing social, economic or cultural context, which synthesizes an ‘identity crisis’ (Silberschmidt 2000:11). In her work, Silberschmidt focuses on rural and urban East Africa, concluding that due to this identity crisis, men adopt sexually aggressive behavior to regain their sense of identity. Indifferently, similar messages of violence are present in some of the academic literature, and overwhelmingly in the policy literature on Burundi. A report on masculinities by the Peace Research Institute Oslo, summarizes the attitudes of Burundians who consider domestic violence “as a necessity and, to an extent, as a good thing (PRIO, 2012: 23).” In the same report, boys are said to be growing up “with a notion of manliness…as someone who has the power to follow his own whims and desires,…who has the power…to make others, primarily their wives, wait on them and fulfill their every wish” (Ibid.).

The dominant view in current social sciences6 largely sees African men as an ‘immanent danger’ (Sommers, 2007: 2), which overrides any investigation of academic literature into the complexities of men’s every day relations or their positive behaviors (Collier, 1998: 22). A recognition of false universalism in the main concept of hegemonic masculinity, has led to the sporadic allowance that some qualities presently associated with ‘manhood’ may not necessarily be negative in all instances (Ibid.).

On the other hand, Uvin holds less pessimistic (alternative) view on African masculinities considering violence. With his experience in Burundi, he explains that frustrations are not automatically translating into violence given the flexibility of Burundian society and the persisting traditional norms of the rural context (Uvin, 2013: Kindle 3251-‐3262). While it seems to be minority view within the field of masculinities, it needs to be seriously taken into account, as my research was conducted in rural Burundi.

6_{Works
of
Barker
(2005),
Barker
and
Ricardo
(2005,
2008),
Contreras
et
al.
(2012),
Grey
and}

(21)

2.2. VIOLENCE

To understand the intersection of masculinity and violence, this link will be examined through exploring how it is possible to see masculinity in action during provocation, reasoning and resolution of a conflict, which might take violent or nonviolent forms.

2.2.1. Provocation and conflict

In the context of this study on masculinities, conflict and provocation are the outcomes of competitive relationships among boys and their masculinities. Provocation (either verbal or nonverbal) is “a challenging act to elicit response” and is part of a complex set of acts that situate boys within a hierarchy of masculinities (Hamlall and Morrell, 2012: 496). It challenges other boys, who are perceived as subordinates to the masculinity of the provocateur, to demonstrate their gender identity. In this competitive process, the provocateur establishes the dominance of the hierarchy as he attempts to display the hegemonic values of maleness that are idealized in the particular community or society (Connell, 2005). Boys might use violence to achieve or defend their masculinity though coercion. Rather than describing the fight, the study will focus on the first phase of a conflict: the provocation and reasoning of a boy to engage in violence or not.7 I will identify the links between particular kinds of conflict and various situational factors that are influencing the choice of violence and nonviolence in that given moment and how these influence conflict resolution among children.

2.2.2. Defining violence

Violence among school children encompasses a wide spectrum of aggressive behaviors ranging from violent ways of solving disputes, unacceptable ways of venting anger or frustration to serious incidents. As an unsuitable classification of violence can exclude or overlook its significant acts in the lives of children, this study adopts the flexible definition appropriate to the researched unit of analysis and the local context. This research aims for an inclusive understanding of violence, not only including the purely physical acts but also its subtle and less detectable forms such as humiliation (as part of provocation) and punishment since these can be seriously harmful to the mental wellbeing of a child.

7_{While
the
knowledge
of
the
act
is
valuable
to
assess
the
motivations
for
the
resolution
of}

conflict (who loses/wins), it tells us little about the patterns behind boys fighting and not fighting.

Boys and violence : the construction and display of masculinities in rural Burundi