Introduction
In Latin America, the mainstream debate on public security, especially in countries with high levels of violence in urban centres, focuses on crime, youth gangs and the police, and, in some cases, on death squads and paramilitaries.Footnote 1 These phenomena are often collectively referred to as ‘new violence’, a term developed mainly by European-based scholars to describe the rise in violence that occurred during democratisation processes across Latin America. The concept captures not only the proliferation of such low-level violence, but also the dampening of hopes of an overall reduction in violence following an apparent end to political tensions.Footnote 2 However, the literature on new violence gives short shrift to gender analysis.
This article problematises the relative neglect of women's experiences within this new violence literature, which tends to address urban violence through an examination of male actors. It argues that the silencing of women's experiences limits our understanding of patterns of violence and how it is reproduced. It concentrates on Brazil, which was singled out by a recent UN-HABITAT report for its high levels of insecurity related to violent crime, drug-related violence, kidnappings, and violence between individuals, gangs, the police and suspected criminals.Footnote 3 Much of this violence is associated with favela communities, and acute death rates affecting principally young, mainly black, males from poor backgrounds. Such men are consequently classified as an ‘at-risk’ target group for government policy and NGO interventions, based on the disadvantage and exclusion they face in a variety of arenas, as well as finding themselves subject to criminalisation by the police and dehumanisation by the media. In this discursive field, however, women are consistently missing, rarely figuring as active agents in the picture of violent cities marked by no-go zones, police brutality, vigilante justice and a controlling criminal class. Alongside the sensationalist manifestations of urban violence hitting the headlines,Footnote 4 everyday acts of violence, often horrific in their banality, cover a spectrum from the mundane to the lethal, encompassing fights between neighbours and family members, bar brawls, domestic violence and violence against children, as well as violent criminality, gang punishment and police brutality. All of these acts of violence affect and/or are influenced by women's involvement, direct or otherwise.
Some recent work has begun to address the underexplored links between different forms of violence in Latin America, which range from political and social to public and private.Footnote 5 Moser and McIlwaine demonstrate how, in the context of civil war in Colombia and post-war Guatemala, communities identify multiple links between the causes and consequences of the many forms of everyday individual, community and institutional violence. Other authors have examined how specific forms of aggression, such as the violent context of forced displacement,Footnote 6 have affected women, and how public forms of violence encroach on interpersonal behaviour and relationships.Footnote 7 Moreover, there is a growing literature on the phenomenon of gang violence and the institutional response of police brutality that so often goes hand in hand with actual and perceived increases in criminal activity.Footnote 8 However, while this work increasingly acknowledges the role of masculinities in reproducing violent practices,Footnote 9 it rarely investigates or interrogates the role of women or of ‘femininities’ in these dynamics. Critical feminist writing has argued for the need to examine the socialisation processes in the home that render (masculine) aggression an acceptable response in certain situations. As Pearce explains, there is a ‘feedback effect’ that enables expressions of violence in the public and private spheres to feed off one another in processes that constitute the socialisation of violence.Footnote 10 Hume's most recent work on El Salvador highlights how women's experiences frequently lie at the interface between private and public expressions of violence.Footnote 11 In the Brazilian context, the pilot study by Moura and Galera investigates the multiple impacts of urban violence on women's lives, while a report by Amnesty International centres on human rights abuses against women by state and criminal actors.Footnote 12
Nevertheless, while the lack of recognition of women's involvement, coerced or intentional, in violent practices has been challenged by a number of feminist analystsFootnote 13 who have offered a more holistic understanding of violence, there is virtually no gendered discussion of women's roles and experiences in the more mainstream debates on violence. Indeed, the concern of much Brazilian literature with gang-related violence, police abuses of human rights and the increasing role of paramilitaries in controlling communities has all but squeezed out women's voices.Footnote 14
This article argues for a gendered approach that acknowledges both women's neglected roles and experiences in relation to violence – as gang members or supporters, as partners, as mothers, as messengers, as carers for the wounded – and their more subtle behaviours that attempt to undermine or condone violence, or challenge or conform to prevailing cultures. It demonstrates that the dynamics of urban violence are highly gendered. Un-gendered debate eclipses the whole range of violence that women face in urban spaces and downplays the forms that they encounter in the home and in private relationships. Analysts and policymakers tend to prioritise certain types of violence, such as police and gang violence, within an implicit hierarchy based on visibility and death rates.Footnote 15 Yet discussions of such phenomena rarely go beyond the largely male protagonists to consider secondary impacts on a broader range of actors, including women, and hidden and non-lethal forms of violence.
It is not enough to simply ‘add women and stir’, however; a systematic gendered approach is needed.Footnote 16 The wide range of women's social roles needs to be understood within a pattern of gender relations in which women are subordinated, relations that are manifest across the range of intersecting aspects of identity, including race, social class and age as well as gender. That gender is a key identity variable in how security is experienced is beginning to be acknowledged in the literature on human security, security sector reform and international relations and security.Footnote 17 Equally, in the analysis of urban violence, going beyond top-down views of security brings into the frame a broader range of everyday forms of violence. These shape both individual and collective security needsFootnote 18 and require an appropriately broad range of responses.
So, while this article focuses on previously neglected women's experiences, it also argues for a more generalised gender perspective, acknowledging the central role that gender plays in the reproduction of violence in marginalised communities.Footnote 19 The first section interrogates the usefulness of the term ‘new violence’ as a theoretical framework, given that it ignores women's experiences of community as well as personal violence. The following section explores three areas that illustrate the relevance of women's experiences to the study of urban violence and demonstrates how all forms of violence have gendered impacts for both men and women. The concluding section reinforces the importance of gender analysis in order to gain a deeper understanding of the complexity of violence and thus inform policies and responses aimed at reducing violent behaviour and its consequences.
What is ‘New’ about ‘New Violence’?
The term ‘new violence’ was first suggested by Briceño-León and was popularised by Koonings and Kruijt in particular.Footnote 20 Although it refers to the social and economic violence that emerged during the processes of democratic transition in Latin America in the late 1970s and 1980s, it has parallels in other regions and countries, such as post-apartheid South Africa, that have also experienced political transitions to more democratic forms of governance.Footnote 21 Authors writing on new violence argue that recent increases in violence differ from those of previous eras because violence is no longer limited to state security forces and non-state actors (militias) linked to political and business elites, or confined to interpersonal conflict, but is a practice enacted by a spectrum of state/non-state social actors, in rural as well as urban contexts and across social classes. This violence encompasses ‘everyday criminal and street violence, riots, social cleansing, private account settling, police arbitrariness, paramilitary activities, post-Cold War guerrillas, etcetera’.Footnote 22 Explanations vary as to why violence has proliferated despite anticipated reductions in authoritarian and political violence. Koonings, for example, argues that this reflects a breakdown of the state's monopoly over the legitimate use of violence.Footnote 23 Others have emphasised factors related to state failure, such as the impact of social exclusion,Footnote 24 the pull of easy money through the trade in narcotics and other illicit goods, and the lack of other economic opportunities.Footnote 25 Hume and Bourgois point out how, during conflict, repressive states have not only engaged in direct oppression, but have also utilised local populations to control, inform on and carry out violence against others.Footnote 26 Benson et al. link current violence to the patterns of violence practised by the former authoritarian states.Footnote 27 But, with some notable exceptions,Footnote 28 in focusing primarily on the state and state-sponsored institutions, crime and delinquency, and urban militias, the bulk of the new violence literature tends to treat these issues as if they were the result of un-gendered processes, carried out by un-gendered actors. Women and/or the gendered character of violence are either ignored completely or given little mention. The protagonists are implicitly assumed to be male, while women are acknowledged to constitute some of the victims.Footnote 29 This literature does not attempt any significant deconstruction of masculinities in relation to this violence, rendering women and their identities and roles invisible and failing to explore any alternative masculinities among the men involved.
Nonetheless, the ‘newness’ of this violence does not go uncontested in the literature. Whether a perceived shift in actors and motives (from predominantly political to predominantly criminal) reflects a significant shift in the lived experiences of violence and insecurity is debatable. Arguably, actors have mutated but not changed; in some instances uniformed police officers are less likely to be involved in overt violence, but the same individuals may be functioning under the remit of death squads or militia groups. In any case, state violence against particular social groups, including poor, marginalised communities, as a form or result of exclusion and oppression, is an enduring, rather than new, aspect of modern society,Footnote 30 although in contemporary times these groups are also victims of a proliferation of privately funded security services and militia activities. The ways in which particular sectors of society, such as youth, unemployed and migrants of colour, are vulnerable to specific forms of violence are shaped by current societal conditions, which in turn are the result of longer-term micro and macro processes, and result in clusters of suffering and discrimination.Footnote 31 One risk in labelling current violence as a ‘new’ phenomenon is that this view concentrates more on certain contextual and institutional factors, such as state effectiveness, and less on the historical socio-economic processes, enduring forms of injustice and victimisation, and structural factors and continuities of the political and economic system. Of course, the two perspectives are connected. According to Pearce, the fact that states have failed to effectively contain violence is profoundly political; persistent violence legitimises the state's authority to operate in the name of security while protecting elites that have ‘never abandoned the violence that ultimately protects their interests and acquiesce in state security acts which violently target categories of non-citizens’.Footnote 32 Pearce terms this ‘perverse state formation’, meaning a state that does not seek to monopolise violence but rather uses it to accumulate wealth and power for a small group, and in so doing excludes or even ‘sacrifices’ large parts of the population, such as young, poor black men, or ‘gang members’. So rather than new violence being an apolitical phenomenon resulting from a weak state, it is selectively fostered by states; in other words, what appears to be anarchic and diffuse is essentially connected to the ongoing political project of new state formations.
As Benson et al. indicate, it is easy to pass the blame for the current levels of insecurity onto delinquent young people under the catch-all of ‘gangs’. While youth gangs have a long history in Central America, since the 1980s the region has seen a change in the form and a rise in the number and ubiquity of criminal gangs due to social disorganisation, structural/economic exclusion and criminal opportunity.Footnote 33 The existence of these gangs provides a convenient scapegoat, and legitimises extensive and punitive policing.Footnote 34 While Benson et al. argue that historical forces have left racialised ethnic groups in a vulnerable position, they fail to address how ethnic identities intersect with other lines of discrimination. Yet the logical progression of their argument implies that, just as poor and marginalised populations suffer disproportionately due to the organisation of society, so do subordinate subgroups within them, and cross-cutting identity markers such as gender intensify risk and vulnerability to certain forms of violence.
Another risk with the label ‘new’ is that, by definition, it sidelines forms of violence that are apparently unrelated to the changing social dynamics that have accompanied political transition. It is arguably the persistence, as opposed to the novelty, of violence against women, particularly in the private sphere, that has excluded women from the debate on new violence. My critique of the new violence approach is that, in seeking to characterise the ‘new’ forms of violence, the approach ignores the constancy and the importance of gendered personal violence. Although there are no reliable statistical data that indicate that domestic violence or other forms of violence against women have increased since transition, the fact that such violence exists and persists is important for our analysis. Gender-based violence against women is interlinked with other forms of violence, and should be acknowledged as part of the gendered impact and dynamics of new violence.
Although violence against women is an age-old phenomenon found in most societies,Footnote 35 it inevitably responds to contemporary social structures, pressures and norms.Footnote 36 Moreover, the persistence of this violence does not lessen the need for urgent violence reduction interventions. Whether a victim of domestic violence has access to support services or belongs to a marginalised ethnic group, and whether a perpetrator has access to a social network with similar moral codes or access to firearms, are all factors that shape the social legitimacy and nature of violence within relationships. Therefore, given that all these elements may be shaped by the current climate of new violence, violence against women could legitimately be included in the debate, whether or not the moral claim for inclusion is accepted. This leads us to question the extent to which the actors in new violence in general are in fact new, or whether it is simply their more widespread access to weaponry and dominance in the public imagination that have changed. At a basic level, the inclusion of women in this debate is merited by the fact that the ‘new’ social actors with their increased access to arms also perpetrate violence against women, and that some women play key roles in these structures. This article aims to show, however, that it is necessary to go beyond these minimal justifications and to examine the multiple links between women's lives and urban violence, with the socialisation of violence as one clear example.
Without ‘pathologising’ certain cultures as more prone to violence than others,Footnote 37 understanding cultural values and socialisation processes can contribute to our analysis of violence. If causes and manifestations of violence are context-specific, normalisation and resistance also form part of the local context in which violence is both reproduced and contested.Footnote 38 Challenging specific social values that ignore or condone certain modes of behaviour has been key in the fight against domestic violence in many parts of the world,Footnote 39 as has revealing forms of hidden abuse such as incest or sibling violence. This questions the notion that the home and the community are sites of refuge, rather than the location for the reproduction of both identity and violence.Footnote 40 Acknowledging the shifting forces that shape social values and cultural norms prevents socialisation processes being presented as static and resistant to change.
New Violence: Capturing Women's Experiences?
Despite progress in areas of justice and public awareness, socialisation processes that legitimate violence against women are stubborn. Notwithstanding global efforts to increase awareness, improve services and introduce protective legislation in recent years, violence against women is all too often seen as inevitable and blamed on external forces, particularly when it forms part of everyday life and is not part of a sensational act of public violence. Domestic violence is frequently explained as resulting from the tensions and frustrations of unemployment, poverty and consequent drug and alcohol misuse, or even from evil spirits in some religious discourses. It is easier to see violence as a reaction to events rather than a reflection of the dominant social values that permit and perpetuate it.
In contrast, Pearce proposes that we examine gendered socialisation processes that transmit and reproduce violence across time and space, and across the ideological boundary between public and private.Footnote 41 She argues that socialisation processes tend to foster violence among men as a legitimate response to negative feelings of shame and disgrace and to attacks on honour and pride, while the same reaction in a woman would undermine her feminine identity. Although these processes are not learnt exclusively in the domestic sphere, behaviour learnt in childhood is reproduced across time (in later life) and space (in relations beyond the home). For example, the respondents interviewed for my research noted that aggressive behaviour learnt by children in the home was practised in the school and street. Moreover, claims that one form of violence may prompt another are supported by research linking military action and domestic violence,Footnote 42 domestic violence and child abuse,Footnote 43 football violence and domestic violenceFootnote 44 and so forth. If violence learnt in one sphere feeds back into the other, as Pearce suggests, this adds strength to the struggle for recognition of women's experiences of violence in the home and family. The following sections draw out some of the links between urban and private violence, and the way in which relationships and trajectories of influence cross and re-cross both spheres.
Including women's experiences in the existing analysis raises a number of fundamental, if difficult, conceptual and empirical issues. How do public forms of violence affect violence in the home, and vice versa? Given that specific forms of violence such as police sexual harassment, gang rape and vigilante controls on women's movement in the main affect women directly, does this amount to ‘more of the same’ familiar violence against women, or does it indicate that ‘new’ forms of violence affect women in new or different ways? Does the tendency in the new violence literature to focus on a more transparent and accountable state as a solutionFootnote 45 provide a sufficient answer to gendered forms of violence if attention is not paid to the broader socio-cultural causes of violence? And, most significantly, can the term ‘new violence’ facilitate efforts to make visible and conceptualise women's experiences, or, as claimed above, does it further marginalise their experiences? While this article can address only some of these questions, they form an agenda for further debate around the new violence literature. The following section therefore begins to unpack the term ‘new violence’ in order to include gender as an analytical category and stimulate more critical discussion of this concept and its implications.
Violence in Urban Brazil: The Context
Gang violence and police repression frequently dominate the news in Brazil, with headlines such as ‘Traficantes voltam a atacar militares a tiros’ (‘Traffickers Attack Military with Gunfire Again’).Footnote 46 Such sensational reporting tends to divide actors into distinct categories of legality and criminality, innocence and blame, that foreclose further interrogation of the dynamics of urban violence. In contrast, activists and researchers are striving to expose a more nuanced reality and the social inequality that underlies much of this violence. They point out that homicides are clustered in geographical locations characterised by low incomes and social deprivation;Footnote 47 that there is a lack of opportunities for young black men living in favelas;Footnote 48 and that young men (and women) face high risk of injury or death if they enter the drugs trade.Footnote 49 Youth, ethnicity and gender are decisive risk factors, with 15- to 24-year-old ‘frequently black’Footnote 50 men the most likely murder victims.Footnote 51
The panic around public security and urban violence in Brazil is associated with specific urban centres, and most commonly with Rio de Janeiro, which receives a disproportionate amount of coverage in the media.Footnote 52 Deaths attributed to the police amounted to 1,098 in 2005, more than 10 per cent of all homicides.Footnote 53 Homicide rates differ greatly across the city, however, and this research focused on a group of favelas (Complexo da Maré) situated within a policing district with the third-highest homicide rate. Relative to other districts, police reports show a high level of violent crimes and police killings of civilians allegedly ‘resisting arrest’, both of which are indicative of a high level of gang activity. It is estimated that approximately 1 per cent of the city's favela population are directly involved in the drug gangs,Footnote 54 which would imply around 1,200 members in Maré. Ultimately, gang members face a very high risk of death, whether they are shot by the police, caught in shoot-outs between rival drug gangs or killed in punishment by members of their own gang. One study of 230 gang members in Rio found that at least 20 per cent had been killed during the two-year period of research.Footnote 55 These figures may well underestimate the impact of gang violence, however. Since they focus on male gang members, they fail to capture other individuals caught up in the violence, including women, and other men who may play less direct roles in the violence or experience less direct consequences.
This article is based on data collected in 2006 and 2009, largely through interviews and group work with over 150 favela residents and 20 project workers and academics. Of these, two-thirds were women or girls; approximately half were between the ages of 12 and 24, and the other half were aged 25–64. Access to residents was gained through grassroots NGOs and government-funded social projects. Consequently, each interviewee had a particular perspective relating to the extent of their involvement in their NGO, often demonstrating a notable degree of assimilation of the organisation's moral rhetoric. These interviews are not representative of the whole community, therefore, nor are they from the most marginalised groups. Rather, their voices are used as an illustration of certain community views and perspectives, providing insight into how people deal with urban violence on a daily basis. Although participatory methods were used in data collection,Footnote 56 they did not go beyond the data collection stage and thus the research as a whole does not constitute participatory research per se.Footnote 57
Although both gangs and the police are responsible for the majority of the violence within favelas,Footnote 58 popular attitudes towards them differ. Interviewees from Maré demonstrated a relatively nuanced opinion of the local gang, acknowledging their community spirit, ability to arbitrate in conflict and consideration towards residents alongside their propensity for violence. In contrast, attitudes towards the police were more consistently negative, associating them primarily with aggression, abusive attitudes, arbitrary killings and corruption. Carlinhos Viera, an ex-resident of Maré, compared police behaviour to that of ‘another drug faction’ in the community, an attitude supported by reports of corruption and collusion with gangsFootnote 59 as well as the violence the police use in their operations. Consequently, the police are criticised both for their inability to guarantee security and for their part in reproducing insecurity. One young favela resident, Violeta (aged 19), highlighted the role of the police in reproducing criminality. When asked what problems the community faces, she responded, ‘Policing, which, instead of helping the residents, actually makes them take the traffickers’ side'.
Pécaut comes to similar conclusions on Colombia. Where the only point of contact with state authorities is in the form of military (or police) incursions that do not attempt to establish either territorial control or relationships with locals, residents appear more willing to tolerate the demands and excesses of other armed groups.Footnote 60 Thus it is not surprising that the local gangs present themselves as an alternative to policing, given the vacuum left by the police and the state.Footnote 61 Gangs provide parallel justice, as well as various social services. Each local gang has its own set of de facto laws, the violation of which may lead to violent punishment. Gang laws commonly prohibit stealing within the community, physical aggression and speaking to the police.Footnote 62 These laws vary from gang to gang, however, and the severity of the punishment is subject to the inclination of the current gang leader (dono), so the response is never truly predictable.Footnote 63 Violence, or the threat thereof, is key to a sense of order and control, as Dowdney points out:
Punishments are generally carried out by lower-ranking faction employees and range from expulsion from the community to forcibly shaving women's heads, being shot in the hands or feet, beatings and death. Like the rules, the imposition of punishments depends on the discretion of the local dono/gerente geral and also on the perceived seriousness of the infraction:
‘Depends on their mood. If they think you should be killed, you're killed’. [Favela resident]Footnote 64
Adolescents I interviewed did not necessarily view this form of justice as right, but they accepted it as the status quo. As Marcello (aged 15) explained, ‘If a guy robs something and the gang beats him, it isn't the best option. But, in a way, they know that they cannot rob.’ Several young interviewees said that turning to the gang was the only option open to them if they faced some form of conflict. For example, after having described a scene of domestic violence she witnessed on the street, Eliú (aged 13) claimed that the only source of help in such a situation was to go directly to the gang: ‘Yes, you can [ask for help]: go there, talk to them and explain that this man is beating a woman. If he is in the wrong they will make him pay.’ Violeta only viewed the gang as an option when all other possibilities were exhausted, however. When asked who she would turn to for help in the case of aggression, she said she would go ‘First to my family, but as a last resort, since the police don't resolve anything, it would have to be to the gang.’
In this way, the drug gangs are not only policing but also acting as judge, jury and executioner, deciding on guilt and punishment for members of their own gangs and for any members of the community who are seen to have stepped out of line. While young men may be the primary victims of this ‘justice’, women also suffer the consequences, or they may be targets by association. Punishment for a wide range of misdemeanours is the rationale behind much gang violence that occurs within favelas. Although such violence may encourage a feeling of protection, it actually increases the overall insecurity that it purports to control.
This section has outlined the type of urban violence that is the focus of much research and policy debate. Although male experiences of urban violence are undeniably important, alone they fail to provide a comprehensive picture. The following section looks at certain elements of these forms of visible violence in order to show how women are involved and affected, both within the home and in the more public arena of the street.
Gendering the Analysis of Urban Violence: Women's Complex Roles in Urban Violence
In the context of urban violence, despite the variety of roles that men and women assume, much of the time actors are categorised by the public and media into simplified groups of perpetrators (gang members or police) versus innocent bystanders and victims. Women are assumed to fall into the latter category, only really taking on a solid form when an act of violence brings their existence into relief either as accidental victims or as grieving partners or mothers. Occasionally, women are represented as exotic aberrations when their involvement in crime is ‘discovered’.Footnote 65 In contrast, Stanko cautions against categorising an actor as either offender or victim, since actors may find themselves in the position of offender in one instance and victim in another.Footnote 66 Similarly, Robben and Nordstrom argue against this ‘distorting dichotomy … as if one is, by definition, passive and the other active’, pointing out that the front lines are ‘much more volatile and inchoate, with violence being constructed, negotiated, reshaped, and resolved as perpetrators and victims try to define and control the world they find themselves in.’Footnote 67 Simplified roles are inadequate for explaining the complexities of social interaction in contexts of violence, but they also lend themselves too easily to gendered dichotomies. This section will look at two aspects of this complexity: firstly, the range of women's roles in relation to gangs, and secondly, the complicated role of motherhood. The latter point is then elaborated through an examination of some of the various impacts of urban violence in the home.
Women and/as armed actors in the public sphere
Women are not simply passive in their relationships with violent actors. Just as in wartime, interaction may range from active participation and supportive roles to passive sympathy or dissent.Footnote 68 Women play a range of roles in their relations and interaction with gang members: they protect their partners, support gang members in jail, incite violent takeovers, hide guns or carry drugs, or in some cases are directly involved as traffickers. Women's direct involvement in gang activity is reportedly on the increase, as recent prison data imply;Footnote 69 while accurate figures are scarce, however, it is estimated that women still only make up 10 per cent of gang members. Moreover, there are more subtle ways in which women engage with the concept and reality of gang membership. Moura points out that women's responses to male gang membership may range from opposition, imitation or rejection to admiration, dependency, autonomy or subordination.Footnote 70 Any apparently active response within this range may well be the result of gendered power imbalances, however, resulting from coercion or a perceived lack of choice rather than an autonomous decision. This wide-ranging understanding of women's roles is illustrated by Gay's anthropological study of a woman who at different times has been the girlfriend of gang members, a drug dealer herself, and the family's designated prison visitor.Footnote 71 Lucia describes a range of responses to different gang members who have played significant roles in her life, variously supporting, directly influencing and challenging these men. She tells of her reaction after Rui, the local dono, had allowed the murder of Rogério, the father of Lucia's child:
So Amanda [Lucia's daughter] was crying and Rui picked up a bunch of money and handed it to me and said, ‘Here, buy some things for her.’ So I threw the money back in his face and told him to save it for his funeral. At that point, I was ready to go down to the police station and denounce them all… and leave. Then one of the members of his gang said, ‘Kill her quickly, kill her quickly.’ So I said (and at this point she begins to cry), ‘Go on, kill me. Go on, kill me… and my daughter… finish the job…’… So then the next day I left. They told me that my daughter could have everything and that they'd give me money each week. I didn't want it. I told them, ‘You're going to pay for his death. The money you'd give me won't pay for his life.’Footnote 72
While this account provides a vivid illustration of the direct impact on her life of the violence associated with drug gangs, in the loss of the father of her child and the threat to her own life, it also demonstrates Lucia's anger and willingness to challenge these violent men and to some extent resist their attempts at dictating her life choices. I am not suggesting that she is free to make her own independent decisions; the book is also testimony to the extreme violence that women suffer at the hands of gang members, and the high levels of control imposed on their lives. However, to assume women are passive agents and victims in these contexts fails to acknowledge the nuances and contradictions within these relationships, the various roles women assume and the degree of their agency. As Moser and Clark warn, to essentialise men as perpetrators and women as victims denies women's agency and voice, however limited these may be.Footnote 73 Indeed, Lucia refers to incidents where women betray their boyfriends to the police, engineer takeovers by internal gang rivals, and lie to incriminate other women and men within these social networks. Girlfriends of gang members may also be punished for their boyfriends' ‘crimes’, or they may be kidnapped for extortion.Footnote 74 Making women victims of gang justice in this way may also serve to lay down the boundaries for ‘tolerable’ levels/forms of violence against women more generally.
As noted above, domestic and sexual violence is a regular facet of women's lives in Brazil, as it is globally, yet the vast majority of these crimes go unreported.Footnote 75 According to Datasus, there has been an overall increase in the use of firearms in violence against women in Brazil, reaching 42.4 per cent of all homicides against women in 2002.Footnote 76 Although the contexts surrounding these deaths are not specified, the data suggest that the nature of interpersonal violence may be influenced by the availability of small arms. A pilot research project among victims of violence in Rio de Janeiro reported that, in households with firearms, 68 per cent of the women had been threatened with guns.Footnote 77 Moreover, relationships with gang members are shaped by the power imbalance provided by their public status, as well as by their access to firearms and other modes of violence. For all women in favelas, the pervasive issues of stigma, economic and emotional dependence, and the cyclical nature of domestic violence are further aggravated by the control that gangs impose over residents' behaviour. These difficulties exacerbate the universal problem of underreporting of domestic violence,Footnote 78 rendering accurate estimates of rates among favela residents even harder to obtain. Thus the prevalence of arms complicates the way we understand women as victims of violence in favelas. It is too simplistic to see women merely as either supporting or opposing the use of firearms; they can be involved in a range of relationships and interaction with armed actors, which can be less or more visible and to a greater or lesser degree coerced. In reality, women's responses and actions depend on their positionality and perceived judgement about how best to secure their personal security and that of their family members.
Women and/as actors in the private sphere
As with other relationships, the context of urban violence complicates the already complex role of motherhood. In less extreme situations, women routinely try to encourage, protect, educate, control and punish their children, all in the practice of parenting. In Maré, women are frequently the primary or sole carers of children as well as the central income earner in their households, roles that are difficult to juggle. The key female figure in many children's lives is not necessarily the biological mother, however, but may be other relatives or neighbours who assume the role of primary carer. When asked to describe her life, one interviewee began with the words, ‘I struggle alone with my children; this is the most difficult problem, because they have a father, but he doesn't live with me’ (Inês, aged 48). Another woman similarly responded, ‘Well, I brought up my son alone, working, with a lot of sacrifice, in the middle of all the delinquency, but he's a good boy, thank God’ (Rita, aged 64). The fact that these women start their stories in this way suggests the importance of motherhood in their personal narratives, whether this is because they feel their struggle requires recognition or to highlight the absence of a caring, supportive father figure. The discourse of struggle is a consistent feature, irrespective of whether husbands are present, and is intensified by factors such as alcohol abuse that render men's behaviour problematic.Footnote 79
The educational aspect of rearing and instilling a sense of social and moral values in children (socializar os filhos) figured strongly in residents' discourse around motherhood. This reflects parents' attempts to protect their children from the violence and risks that children face in their daily lives. The nature of the causal links between broken homes, single parenting and delinquency is the subject of much debate,Footnote 80 and discourses about such links tends to shift the blame onto poor parenting and away from structural factors. The women interviewed saw the burden of keeping children on the ‘right path’, particularly for single mothers, as potentially overwhelming. Fernanda (aged 34) described these difficulties and, although not a single mother herself, she implied her fears for her own young son:
The majority of kids stay with the mother because they never had a father, or he died, or he works and isn't at home much. So they spend time with their mother, more or less. Because if she goes out to work they stay alone and end up on the streets, hanging out with whoever, eating anything … They end up getting involved in things they shouldn't because there isn't anyone to look after them.
Violence against children is an understood part of parenting practice in favelas. Thus women frequently play the dual role of protectors -- keeping children and other family members safe from the dangers of the street -- and of perpetrators, when physically reprimanding and punishing children. Depending on the severity of the violence, physical punishment is commonly constructed as ‘education’ rather than abuse.Footnote 81 As Hume points out in the context of El Salvador, violence against children is socially perceived to be a quick and effective means to achieve results – if, indeed, it is recognised as violence at all.Footnote 82
To what extent can such actions be explained by contextual factors, such as a lack of access to alternatives like psychoanalytical support, as Naira argues, or by the brutality of structural violence that parents themselves suffer, or by the extremes of violence on the street from which parents are trying to protect their children?Footnote 83 Although not all parents adopt violent practices, as long as there is some level of social and normative acceptance of the ‘functionality’Footnote 84 of violence as an effective method of parenting, it is hard to challenge the notion that violence is a natural and legitimate means to solve conflict and to exert power more widely. The discourse surrounding violence against women, for example, frequently mirrors that relating to violence against children, in that such violence is seen as an expression of love and duty and as a way to control ‘improper’ behaviour. While I do not wish to fall into the trap of determinism, socialisation processes evidently contribute to the reproduction of violence.Footnote 85 As mentioned above, men are permitted or even expected to use violence to counteract feelings of shame, disgrace and assaults on their pride.Footnote 86 But violence is not the prerogative only of men, and there are other contexts – not only in parenting but also, for example, in response to sexual jealousy – where violence by women is condoned or encouraged. Such lived experiences of violence inform future encounters with violence, as they reinforce or challenge the boundaries of acceptable/legitimate aggression, albeit subject to contestation and revision. While violence perpetrated in public may act as a conduit of socialisation at a community level, violence in private contributes to socialisation within the domestic sphere. Boundaries of acceptable violence in the private sphere also lay the ground for public violence, and vice versa.
The examples given here, of women's roles both in relation to gangs and within the context of mothering, serve to illustrate the variety of roles individuals play in relation to different, mutually reinforcing forms of violence. As people move through different social and geographical spaces, different aspects of their identity predominate and place them at different levels of vulnerability and risk, as well as mediating their propensity to perpetrate or challenge acts of violence.
Gendering the Analysis of Urban Violence: Impacts of Urban Violence on Women
Just as women's roles in violence are simplified, so too are the impacts of violence, and women's experiences of urban violence have mainly gone unanalysed. In reality, women suffer a wide range of direct and indirect impacts that may be severe but often not immediately visible. Firstly, although women are statistically less likely to be directly targeted by the police or gangs, they may still be deliberately targeted, or they may be the victims of stray bullets in shoot-outs, or caught up in arbitrary and aggressive policing tactics.Footnote 87 In Gay's study, Lucia describes one particular incident when police officers raided a house where she was staying, threatened her with electric shocks, hit her, stole her valuables and took the inhabitants of the house to the police station to demand a bribe for their release.Footnote 88 Not only do those implicated in illegal activity find themselves on the receiving end of aggressive tactics, but also innocent bystanders are caught when the police ‘come in shooting’,Footnote 89 such as the numerous children hit by stray bullets each year.Footnote 90 One male resident, Vito (aged 19), reported the death of a girl shot by the police in the street shortly before our interview.
Secondly, women frequently suffer the indirect effect of violence when those close to them are the direct victims of violence.Footnote 91 For example, in view of their caring roles and the more consistent contact they have with their children, women experience the pain caused by the death of a son, partner or family member, the all-consuming job of caring for a victim injured by gunfire, or being left as the sole provider for a family when a partner is killed. While this is not to suggest that men do not also suffer emotionally, statistically it is men who are likely to die, leaving women to deal with the aftermath. More women than men have been involved in justice campaigns following homicides.Footnote 92 The sociologist Vilma Reis (Federal University of Bahia) claims that poor women are routinely affected by the deaths of young men: ‘In any poor community [throughout Brazil], you get the same story: practically all the women have already lost a husband, a son, a father or a brother, killed by the police or by drug dealers.’Footnote 93
Thirdly, the persistent threat of violence and overall levels of stress and anxiety encroach on everyday practices and complicate social interaction, daily routines and mobility. This can have specific effects on women. Women worry not only about themselves but also on behalf of others, hoping to keep children, sons and partners out of the line of fire, away from dangerous individuals, and occupied. One 27-year-old mother, Noralí, described how she took herself and her children to sleep in the safest area in the house, on the bathroom floor, when gunfights began outside, a tactic echoed in media reports.Footnote 94 Another mother, 38-year-old Mafalda, expressed despair at her children's lack of reaction to the sight of policemen driving by in a patrol car brandishing guns, since she saw their failure to run away as her failure to instil a survival instinct in them. Mafalda's concern was to protect her children from stray bullets, instructing her children to ‘throw yourself on the floor; you have to take care of your own life, because they don't care’. These examples illustrate the close association between violence and the street, a place also associated with temptation (criminality, drugs, sex) and undesirable characters (gang members, delinquent children, abusive adults). The response to these perceived risks is often to restrict the amount of time spent outside, or to fill time with alternative activities, whether sports, work or social projects.
Fourthly, urban violence also impacts on relationships within the home, and vice versa. A practical example of how local power structures circumscribe parenting practices is illustrated by the difficulties interviewees faced in instilling a sense of principle and morals in children. Mothers reported how they were unable to give explanations freely about particular events or behaviour without putting themselves or their children at risk. For example, providing explanations about economic activities, ostentatious wealth or power relationships in the community proves difficult; thus, Fernanda reported feeling constrained in what she could say to her son:
It is difficult to bring up a kid in the community, because our kids grow up seeing these things and think they are normal. Every day they see everyone involved in trafficking, people using arms, using drugs. And I think that … because of this kind of thing they end up getting into crime. Because our kids see this a lot: the son of Joe Bloggs, whose parents don't work, yet he always goes round in expensive, fashionable, smart clothes… How am I going to explain this to [my son]? It's complicated; he might even go to the kid and say, ‘My mum said you are so well-dressed because your dad is a drugs dealer.’ So it's pretty complicated if I tell him this. So I give different explanations – it must be an uncle or a relative that gave him this.
As well as the desire to protect, women also expressed frustration at being unable to help others for fear of bringing problems into the home, at having to moderate what they say to whom, and at not inviting visitors to the house for fear of arousing suspicion. These are examples of women's curtailed freedoms in the context of urban violence. Such responses are a form of self-censorship, as they take on the rule of silence as a personal security strategy and thereby reduce the need for violent actors to enforce such behaviour, with fear acting as the intermediary.Footnote 95 This is only a small sample of the myriad ways in which violence influences people's lives, permeating social interaction and daily routines in the favela and defining social relations both within and outside the home.
Gendering the Analysis of Urban Violence: Men as Gendered Actors
The discussion so far has focused on the experiences of women and girls, given the neglect of these experiences in the mainstream debate on violence and security, but clearly the gendered nature of male-on-male violence should not be ignored. The fact that men and boys are the most vulnerable to participating in and being victims of violence is just as much a gendered issue. This link between men and violence is related to the negative factors of social exclusion and economic disadvantage as well as positive gains in terms of identity and other tangible benefits. Men interviewed and observed in this research exhibited different versions of masculinity, echoing the growing interest in the construction of masculinities in contemporary gang contexts and the motives behind involvement in gangs.Footnote 96 Scholars suggest that gang membership can be a means of establishing a sense of masculinity through the acquisition of goods, money, respect and women,Footnote 97 and of providing a sense of belonging, the promise of a glorified subculture and rewards for loyalty.Footnote 98 Nevertheless, the construction of dominant masculinities is the result of interpretation and negotiation by individuals so that, as Rodgers notes, these masculinities emerge from the dynamic interplay between structure, agency and practice, albeit working within given social and institutional frameworks.Footnote 99 Key to the version of hyper-masculinity that gang membership provides is the possession and use of firearms (intensified by the prohibition on arms ownership by other residents) and the status, power and hegemony that these weapons bestow. Moreover, as Moura argues, this model is dependent on the submission of women to a particular version of femininity, and on the submission of certain men.Footnote 100 Just as there is scant discussion of the construction of femininities in the context of urban communities dominated by armed violence, or of how male–female relations contribute to the reproduction of violent practices, so there is no significant analysis of how women might manipulate aspects of hegemonic masculinity in order to achieve certain goals in a male-dominated world – for example, by winning respect as an active gang member.Footnote 101 Yet dominant masculinities and femininities are built, at least in part, on differentiation from constructions of the other, and should be seen as relational rather than viewed in isolation as one end of the spectrum.
A powerful example of the link between violence and gendered identities is provided by the connection between drug gang members and violence against women. Anecdotal evidence from interviewees as well as other ethnographic studies suggests that male gang members are heavily implicated in domestic violence, which is sometimes justified as a form of alternative justice. Young people in Maré referred to gang members committing acts of interpersonal violence, frequently against women, as Elza (aged 20) illustrated when she said, ‘I have seen many people fighting, often getting beaten up … People mixed up in “them”, those that are addicted to drugs, for some reason … Some of the women, their girlfriends, are always getting beaten in the street’.
The visible abuse of partners and girlfriends by gang members demonstrates that the dominant group culture reproduces and encourages certain attitudes towards women, thereby validating particular notions of masculinity. Not only is there a certain normalisation of physical abuse, but it is understood that a bandido will have more than one girlfriend on the go at any one time, while a woman is expected to be faithful.Footnote 102 Moreover, once women are involved with gang members, it is not always easy to extricate themselves from these relationships. In one case that illustrates this very clearly, an adolescent girl told her gang-member boyfriend, who was in prison at the time, that she wished to leave him; from prison, he ordered the murder of her new partner and also threatened to kill her, or her family, if she left the neighbourhood. This attitude shows how dominant masculinities feed off particular femininities that rely on loyalty and submission. The legitimation of abusive treatment and control over women therefore provides only a partial picture. It is equally important to examine the ways in which women engage with models that require their submission, as well as their role in promoting violent masculinities. The socialisation of such norms and ideal models must be understood if the potential for a ‘feedback’ effect shaping acceptable levels of violence more generally is to be interrupted.
The treatment of men who are not part of the ‘hegemonic group’ also merits exploration. The fact that it is young men who make up the ranks of gang members means that other – uninvolved – young men are put at risk by virtue of common identity markers (young, male, balck), and not as a result of their own actions. This point underscores the importance of refusing to essentialise male and female roles, since men may just as easily fall victim to the stereotype of violent males. Young men face a greater risk when crossing gang boundaries, when a shoot-out occurs, when the police invade communities looking for suspected criminals, or when a turf war flares up between rival gangs. While some young men may wish to get involved in gang activity in order to ‘become men’, many others seek to avoid being implicated in covert activity, sometimes by deliberately rejecting behaviour associated with hegemonic models of masculinity. For example, they may avoid certain individuals – whether (rival) gang members or the police – or spaces and times considered to be flashpoints where covert activities take place.
A prime example of the risk young men face is when the police caveirão Footnote 103 enters the community. This armoured vehicle with blacked-out windows provides anonymity and impunity for the perpetrators of any indiscriminate aggression and causes widespread panic. Particularly with young men, therefore, the caveirão featured heavily in interviews as a source of violence and a producer of fear, as Marcello (aged 15) pointed out: ‘The residents are scared of the police. The caveirão enters at night and injures or even kills people. So when shooting starts, you have to stay in the house.’ This is particularly true for those who fear being mistaken for gang members if they are spotted in the wrong place at the wrong time. Milan (aged 18) explained that when the caveirão approaches, ‘everyone runs home’. He also pointed out the limitations that fear of the caveirão puts on socialising: ‘We go from here [the project] to school, so we don't see much – only at the weekend, when we want to hang out on the street chatting and we can't, because of the caveirão.’ This is a clear example of the restrictions young men feel in both the spaces and times at which they can freely move around. The general belief was that potential flashpoints are less risky for girls and women, which is vividly illustrated by female residents tailing the police during searches for suspected criminals. The women's presence is intended to prevent an extrajudicial killing by the police on the pretext of a suspect resisting arrest.Footnote 104 Police action like this causes particular concern, as its unpredictability does not observe any commonly understood framework of rules of engagement. This section demonstrates the importance of a gendered analysis of urban violence that not only acknowledges an understanding of women's roles in urban violence and the impact of such violence upon women, but also men's gendered experiences of this violence.
Conclusion
Towards the Gendering of New Violence: Including Women's Experiences
This article extends the gendered analysis of urban violence by demonstrating the intersecting links between different forms of violence in the context of social inequality and repressive policing in Latin America. Firstly, it critically appraised the literature on new violence which, it is argued, is flawed because of its gender blindness. Secondly, it detailed the different forms of violence in urban communities, in both the private (domestic) and public (community) spheres, highlighting both the roles of women in the perpetuation and reproduction of violence, and the impact on women of violence. Thirdly, using field research in a group of favelas in Rio de Janeiro in Brazil, this study highlighted the links between different forms of violence against women in the private sphere and the more commonly discussed urban violence found in marginalised communities. This research challenges the implicit assumption that private violence is less important than the more visible violence on the street. Demonstrating that all forms of violence intersect and overlap in terms of actors, impacts, processes and interrelatedness not only strengthens this argument but also adds to our understanding of the ways in which violence reproduces inequality and oppression along gender as well as class, ethnic and other axes.
At first glance, the concept of new violence appears to offer a useful theoretical framework for this discussion by highlighting the localised as well as state-linked actors who are engaged in the use of violence as an organising principle, and the state's lack of a monopoly on violence.Footnote 105 The implication of this framework is that a reformed, more responsive state with more control would significantly reduce levels of violence, but two questions arise from this assumption. Firstly, is the state best positioned to provide a solution when its own conduct suggests an active interest in (violently) maintaining the status quo and repressing marginalised communities? And secondly, is such a top-down view appropriate when attempting to tackle hidden, socially condoned and trivialised forms of violence, or is a greater focus on social power relations more useful for challenging norms and processes that facilitate the reproduction of violence? An alternative approach might give more weight to the social movements and organisations that advocate social justice and interventions which strengthen the rights of those living in communities with high levels of violence. For such an approach to confront the full range of violent activities that are endemic in marginalised communities, however, the debate needs to move beyond a focus on the violent actions and identities of men and boys and to understand these as part of a wider landscape of gendered patterns of dominance and behaviour. While there is an increasing awareness of the importance of masculinities in the literature on new violence, this has not progressed to a gendered analysis of how masculine behaviour impacts on – and is influenced by – women's behaviours, experiences and roles. Thus, if the concept of new violence is to be useful in practice and is to serve as a framework for policy intervention, it is necessary to broaden the term to include a wider spectrum of forms of violence, actors and impacts, rather than selecting and omitting specific forms of violence and/or actors according to a perceived hierarchy of violence based on visibility, mortality statistics or political pressures.
The inclusion of women's experiences in the debate about new violence serves several purposes. It highlights that it is not only state violence in communities that is problematic, but also the state's inability to protect women in favelas, which implicitly condones their oppression by masculine gang structures and increases their vulnerability to violence. Furthermore, women's experiences help illuminate the wider context in which the gang and drug cultures operate within communities. Given the evidence of increasing involvement of women and girls in these structures, it is important to better understand the wider dynamics that facilitate this and make gang involvement an attractive option. Finally, women's experiences of violence in the private and public spheres demonstrate the continuities between pre- and post-transition violence, as well as showing how manifestations of violence respond to changing social and economic conditions.
Women's experiences demonstrate how violence provokes a range of consequences that impact upon individuals differently according to their gendered identities and social positioning. In the same way that any conflict involves multiple gendered actors and produces multiple gendered impacts,Footnote 106 the wars that are played out on city streets affect men and women, young and old, in different ways. Women's experiences of urban violence are not limited to binary responses of rejection or support; rather, women have a complex relationship with urban violence and play a number of roles that do not fall neatly into victim/perpetrator categories, instead demonstrating varying degrees of involvement and agency. Equally, men's experiences of violence also need to be tackled from a gender perspective, acknowledging both the empowering and disempowering aspects of masculinities for different men in contexts of violence, which in turn affect the ways in which the boundaries between acceptable and unacceptable violence are constructed.
