Centre for the Study of Violence and Reconciliation

"Marriage is Like Sitting on Red Coals":
A case study of domestic violence in four villages of the Moretele District, Tshwane Metropole

by
Samantha Hargreaves, Lisa Vetten,
Vera Schneider, Lebo Malepe & Romi Fuller

Research report written for the Centre for the Study for Violence and Reconciliation, September 2006.

This Project has been made possible by the generous support of the Royal Danish Embassy.

Samantha Hargreaves is a researcher specialising in land and development issues.

Lisa Vetten is the former Manager of the Gender Programme at the Centre for the Study of Violence and Reconciliation.

Vera Schneider is a former intern at the Centre for the Study of Violence and Reconciliation.

Romi Fuller is a Researcher in the Gender Programme at the Centre for the Study of Violence and Reconciliation.

Contents

Chapter 1: Introduction

1.1 Abuse and Intimate Relationships in South Africa
1.2 The Domestic Violence Act (No. 116 of 1998)
1.3 Research Area and Sites
1.4 Research Methodology
   1.4.1 Community Perceptions of Domestic Violence
   1.4.2 Police Perceptions of Domestic Violence and Responses to Domestic Violence in the Moretele District
   1.4.3 The Implementation of the Domestic Violence Act at Temba Court

Chapter 2: Community Perceptions of Domestic Violence

2.1 Introduction
2.2 Violence in Intimate Relationships
   2.2.1 The Function of Social Roles and Rules: Authority Established
   2.2.2 Roles, Rules and Authority Prior to Marriage
   2.2.3 Roles, Rules and Authority in Marriage
   2.2.5 Why do Men Beat Women?
2.3 Recourse
   2.3.1 Domestic Violence – The Silent Scourge
   2.3.2 Recourse in Dating Relationships
   2.3.3 Recourse in Relationships in which Magadi has been Paid
   2.3.4 Recourse for Women in Cohabiting Relationships
   2.3.5 Recourse for Married Women
2.4 Conclusion

Chapter 3: Police Perceptions of and Responses to Domestic Violence

3.1 Introduction
3.2 The Duties of the Police under Section 2 of the Domestic Violence Act
3.3 Police Perceptions of Domestic Violence in the Moretele District
   3.3.1 Frequency of Domestic Violence
   3.3.2 Types of Relationships in which Domestic Violence Happens
   3.3.3 Police Perceptions of the Causes of Domestic Violence
3.4 Police Responses to Violence in Intimate Relationships
   3.4.1 Respondents' Understanding of the Domestic Violence Act and the Procedures to be followed in Terms of the Act
   3.4.2 Referrals
   3.4.3 Problems Typically Encountered when Dealing with Domestic Violence Cases
3.5 Impact on Police of Dealing with Domestic Violence Cases
3.6 Police Opinions on the Role of Other Structures in Dealing with Domestic Violence
   3.6.1 The Role of Family
   3.6.2 The Role of Community Structures
   3.6.3 The Role of the Metropolitan Police
3.7 Conclusion

Chapter 4: The Implementation of the Domestic Violence Act at Temba Court

4.1 Introduction
4.2 Findings
   4.2.1 Information about the Sample
   4.2.2 The Acts of Domestic Violence
   4.2.3 The Relationship between the Applicant and Respondent
   4.2.4 Who Else is Affected by the Abuse?
   4.2.5 Weapons Used in the Abuse
   4.2.6 Injuries Recorded
   4.2.7 Reasons for Urgency of Application
   4.2.8 The Interim and Final Protection Order
   4.2.9 The Return of Service
   4.2.10 The Final Hearing
   4.2.11 Breach of a Protection Order
4.3 Discussion and Conclusions

Chapter 5: Overall Conclusions in Brief

References

Chapter 1: Introduction

Men's violence and ill-treatment of their intimate female partners is widespread in South Africa. In an effort to curb such abuse and provide comprehensive protection to those subjected to it, the government passed the Domestic Violence Act (DVA) in 1998 (no 118 of 1998). While some studies have investigated aspects of the implementation of the Act, most notably in the Western Cape (Parenzee, Artz, and Moult, 2001; Mathews and Abrahams, 2001), there has been little documented exploration of the interaction between culture and tradition and women's recourse to the DVA. Certainly there are studies examining community perceptions of domestic violence (Jewkes, Penn-Kekana, Levin, Ratsaka and Schrieber, 1999; Blumberg, Swartz and Roper, 1996) but these have not explored how relationship status also mediates the nature of assistance available to women seeking help with domestic violence. These are important key omissions, as this case study shows. It explores how, in one district, men and women are embedded in a range of community beliefs and practices that shape not only responses to domestic violence, but also the nature of assistance provided. This context not only affects when, how and which women use the Act, but also highlights how a variety of institutions mediate women's access to the criminal justice system.

The first chapter of the report contextualises the study and describes the Moretele District. It also outlines the study methodology. The next chapter presents findings from the focus groups conducted in the district. These findings situate domestic violence within a complex set of roles and rules for men and women which not only prescribe how they should behave in relationships with one another, but also provide an understanding of the role these norms play in creating situations that allow for the emergence of abuse, as well as determining the nature of recourse available to women. Building on what community members said in the focus groups, chapters three and four then explore criminal justice system responses to domestic violence. Chapter three presents findings from interviews with police officers stationed in the district studied, while chapter four reports on the local court's response to domestic violence. This chapter is based on an analysis of applications for protection orders at Temba magistrates' court, the court servicing Moretele District. The report ends with a brief overall conclusion that offers some thoughts around addressing domestic violence in ways that engage with people's traditions and cultures.

1.1 Abuse and Intimate Relationships in South Africa

Violence is present in very many of South Africa's communities. One study surveying 1 306 women in three provinces found that 27% of women in the Eastern Cape, 28% of women in Mpumalanga and 19% of women in the Northern Province had been physically abused in their lifetimes by a current or ex-partner (Jewkes et al, 1999). The same study investigated the prevalence of emotional and financial abuse experienced by women in the year prior to the study and found that 51% of women in the Eastern Cape, 50% in Mpumalanga and 40% in Northern Province were subjected to these types of abuse (ibid). Another study, undertaken with a sample of 168 women drawn from 15 rural communities in the Southern Cape, estimated that on average 80% of rural women are victims of domestic violence (Artz, 1999). Interviews conducted with 1 394 men working for three Cape Town municipalities found that approximately 44% of the men were willing to admit that they abused their female partners (Abrahams, Jewkes, and Laubsher, 1999). National figures for intimate femicide (men's killing of their intimate female partners) suggest that this most lethal form of domestic violence is prevalent in South Africa. In 1999, 8,8 per 100 000 of the female population aged 14 years and older died at the hands of their partners - the highest rate ever reported in research anywhere in the world (Mathews, Abrahams, Martin, Vetten, Van der Merwe and Jewkes, 2004).

The national femicide study is also significant for how it highlights the particular risk faced by women in co-habiting relationships in comparison to women in other types of relationships. Fifty per cent of women in this study were killed by partners with whom they cohabited, 30% by men they were dating and 18% by the men they were married to (ibid). No other South African studies examine the effect of relationship status on domestic violence. Theories attempting to explain this association between relationship type and violence in intimate relationships have not been advanced either.

One may speculate that this vulnerability to violence is perhaps linked in part to the inferior legal status of cohabiting relationships, with women in such domestic partnerships enjoy few legal protections. Currently no duty of support is created between the parties, meaning that they are neither entitled to inherit upon the death of the other (unless specifically nominated as a beneficiary in the will) and nor does any proprietary claim against the parties' estate exist. The South African Law Reform Commission (SALRC), in its discussion paper on domestic partnerships, recognised that while cohabitation may be a matter of choice for the middle class, "it is a real problem out of the control of most poor women" (2001: 18). Apartheid law (such as the Group Areas Act, influx control and the pass law system) played a significant role in the disintegration of African families, which was also hastened along by migrant labour. Poverty and unemployment have also been cited as additional key reasons why women remain in relationships where men chose not to marry them (ibid: 18 – 20).

The need to legally recognise cohabiting relationships becomes all the more important in light of the fact that marriage may no longer represent the most common form in which men and women live together. South African statistics suggest that increasing numbers of people do not marry. The 1996 census data showed that approximately 1.2 million people were cohabiting with a partner. Other research has suggested that the rate of marriage is declining, with Statistics SA showing a decrease of 2 168 officially recorded marriages between 2003 to 2004 (the latest data available). The 2001 census data also shows that significant numbers of men and women cohabit rather than marry:

The 2001 census reveals that where women do marry, they tend to do so at younger ages than men across all race groups, for example, in the age group 20 – 24 years, 93.1% of African men had never been married, compared to 80.3% of African women. For Indians or Asians, the figures were 84.5% of men compared to 65.6% of women. It further shows that significant numbers of people do not marry at all. In the age category 35 – 39 years, 33.6% and 33.4% of African women and African men respectively, had never been married – the highest proportion amongst all race groups.

It is a significant oversight on the part of the law to ignore the growing number of South African women who are not marrying.

1.2 The Domestic Violence Act (No. 116 of 1998)

The DVA provides a civil remedy for persons affected by domestic violence and allows a protection order to be granted to an applicant ('the complainant') against the perpetrator of the abuse ('the respondent'). The DVA was passed in response to the inadequacy of the Prevention of Family Violence Act (PFVA) to address domestic violence. The shortcomings of the PFVA which the DVA sought to address include: lack of a definition of domestic violence; failure to explicitly make provision for a range of domestic violence relationships; lack of creative remedies; lack of clarity regarding the role of various stakeholders; and a lack of provision for sanction where various role players failed to perform their duties in relation to the Act.

The DVA provides for:

Failure to comply with a protection order issued in terms of the Act is a criminal offence. The police are mainly entrusted with the criminal aspects of the Act: securing the safety of victims, laying criminal charges, conducting criminal investigations and gathering evidence where a criminal offence was committed in the process of committing an act of domestic violence or where an abuser has breached a protection order.

1.3 Research Area and Sites

The field research was conducted in the Moretele municipal area, one of five local municipalities in the Bojanala District Municipality in the North-West Province. The Bojanala district is situated in the north-eastern corner of the North West Province and includes the towns of Rustenburg, Brits and Sun City. The Moretele Local Municipality consists of 22 wards and spans 1369.299210 square kilometres.

In 2001, the total population of the Moretele Local Municipality stood at 177,905. Poverty levels are high, with 134,358 individual members of the population reporting no source of monthly income, and a further 26,863 reporting a monthly income of R800 or less in 2001. In this same year, unemployment levels (excluding those members of the population classified as not economically active) stood at 42%. However, since a significant proportion of the unemployed are excluded in these more conservative estimates, it can be assumed that unemployment levels are substantially higher than 42%. There are no gender-disaggregated unemployment statistics readily available.

Of the population over the age of 20 years in 2001, 17% had no schooling, 8% had completed primary school, 18% had passed Grade 12, and only 4% had obtained a higher educational qualification.

Table 1, presented below, provides an indication of levels of particular crimes reported between April 2004 – March 2005 at three of the police stations servicing the Moretele District. (Data for the fourth station, Carousel, was not available from the SAPS website). These stations serve the four villages where the focus groups were conducted.

Crime category Temba Cyferskuil Makapanstad
Murder 61 2 9
Rape 354 20 65
Attempted murder 119 3 24
Assault with intent to inflict grievous bodily harm 1 279 35 182
Common assault 966 23 124
1.4 Research Methodology

This study comprised of three different research components, each with its own research methodology.

1.4.1 Community Perceptions of Domestic Violence

A team of three researchers undertook fieldwork during the period January to March 2005. In January, a series of orientation visits was undertaken and briefing meetings held with a range of stakeholders including the House of Traditional Leaders, councillors and staff of the Moretele municipality, the four traditional councils, district health services, high-ranking staff of four police stations servicing the locality, and representatives of local Community-Based and Non-Governmental Organisations.

The field researchers visited each of the research sites at least twice prior to conducting the field research: the first visit was made with the support of the traditional council and involved meetings with various local organisations and stakeholders at a village level. The second visit deepened the researchers' engagement with local organisations and enabled the finalisation of the field research arrangements.

Twenty focus group discussions were conducted in the four research sites (five focus groups per research site) in February and March 2005. For the qualitative element of the study, four research sites were randomly selected, one in each of the four traditional authority areas in the municipal area. The sites were Bollantlokwe village under Kgosi Makapan; Ratjiepane village under Kgosi Mathibe; Little village under Kgosi Nawa; and Greenside village under Kgosi Maubane.

A focus group discussion guide was developed by the research team – the project leader and the three field researchers – and served as a guide to the facilitation of the focus group discussions.

The following focus group discussions were conducted in each research site:

Demographic data is available for 12 of the 18 focus groups. The groups for which demographic information is available were all women, and therefore encompass the focus groups of women teachers, mothers of children and women organised by the traditional leaders. No demographic data is available for the focus groups of traditional leaders and the civil society groups, comprising a mix of men and women.

Just over 100 women were reached through the 12 focus groups for which demographic information is available. The youngest participant in the group discussions was 22 years of age and the eldest was 81 years. Women in the focus groups of traditional leaders were most likely to be middle-aged and elderly, with a median age of 65 years. Focus groups of teachers and mothers were usually comprised of women under the age of 50 years, with a median age for teachers being 43 years, and 42 years for mothers.

Of the eight focus groups of women not employed as teachers, only three women (4%) were employed, with two of these indicating that this employment was on a part-time or temporary basis.

In terms of educational levels achieved, all but one of the teachers had obtained post-matric qualifications. Of the balance of the research participants for which data is available, 29% had obtained some level of primary school education, 23% had obtained a Standard 6 (Grade 8) pass, 10% had passed their matric (Grade 12) and 6% had obtained a tertiary qualification.

With regard to marital status, a full 76% of the research participants were married either under civil (the majority) or customary law. Twenty-two participants (22%) and ten teachers (32% of teachers) were single.

Tapes of two of the focus group discussions unfortunately went missing, and the analysis therefore only draws on discussions held in eighteen of the focus group discussions. A list of the focus group discussions, and the acronyms used in the report when quoting from these discussions, is attached as Appendix A.

The taped discussions were transcribed by a team of two translators and written notes of the focus group discussions were used as the basis for analysis. Where basic demographic and background information was available for focus group participants, these were tabulated and analysed. A data coding system was developed to order and code hundreds of pages of data. Compilation sheets were usefully employed to summarise and compare data across the different categories of research participants.

1.4.2 Police Perceptions of Domestic Violence and Responses to Domestic Violence in the Moretele District

Semi-structured interviews were held with 8 members of the South African Police Service (SAPS) stationed at the Carousel, Cyferskuil, Makapanstad and Temba police stations situated within the Moretele Traditional Authority areas under study. An interview was also held with a member of the Tshwane Metropolitan Police Department whose area of jurisdiction overlaps with some of the Moretele Traditional Authority areas.

One of the shortcomings of this section of the study is that not all interview questions could be thoroughly explored during the interviews. Police authorities were somewhat reluctant to give us permission to interview SAPS members for this research and when permission was ultimately granted, our field workers were restricted to one hour per police station. Two police officers were interviewed at each station, and this limited the scope and depth of the interviews.

1.4.3 The Implementation of the Domestic Violence Act at Temba Court

A retrospective review of applications for protection orders registered over a two year period at Temba court was undertaken. Written permission to conduct the study was obtained from the Department of Justice and the Chief Magistrate of Temba court on condition that no identifying information of the applicants or respondents would be used in the report, and that these persons would not be contacted. A structured questionnaire was designed to document the contents of the court files in accordance with the information required on the application for a protection order. Data that was captured included particulars of the applicant and respondent, the acts of domestic violence, other persons affected by the abuse, the urgency of the application, details of orders requested and granted, the status of the interim and final protection order, and contraventions of the protection orders where applicable. Fieldworkers were trained to collect the data. The data was coded and entered in a computer programme, and appropriate statistics were used to analyse the data using SPSS version 10.

There were a number of limitations to this part of the study. Several applications for protection orders were lost or unaccounted for from the required time period. Reasons for this could not be conclusively established, but Mathews and Abrahams (2001), who encountered a similar problem in their Western Cape study, suggested that it may be due to cases being investigated at the time and hence files being with other personnel, that high caseloads created the possibility of misfiling, or that files were lost or destroyed on purpose through interference from perpetrators. What information was available was often scanty, and it is recognised that the findings may be an under-representation of usage and implementation of the DVA. The accuracy of the data was also questioned in a small proportion of cases (for example, in some forms the interim order was granted before the date of application), and to preserve the integrity of the data, any irregular information was captured as unknown. Finally, as information is reported from only one site, it cannot be assumed that the findings are generalisable to all semi-urban courts.

Chapter 2: Community Perceptions of Domestic Violence

2.1 Introduction

This chapter is an analysis of focus group discussions held with women organised by traditional structures, traditional leaders, mothers of children organised through a local school, women teachers and representatives of local civil society structures such as churches, health and development committees and some local service organisations, with a view to explore their understanding of and responses to domestic violence.

This chapter explores:

A brief conclusion is set out at the end of this chapter.

2.2 Violence in Intimate Relationships

All of the groups acknowledged that violence in all forms of intimate relationships is a feature of the villages they reside in. Most of the discussion centred on physical violence, although other forms of domestic violence such as financial abuse, sexual violence and emotional or psychological abuse were also raised in a few of the groups. This section will focus on physical violence, and will also include some discussion on sexual abuse (marital rape), since these were the substantive issues discussed in the groups. The group discussions on emotional/psychological and financial abuse were, unfortunately, too limited to draw any meaningful conclusions from and have therefore been excluded from this analysis.

2.2.1 The Function of Social Roles and Rules: Authority Established

Rules are the codification of a given set of gender roles and relations prescribed for men and women in a given context. Rules are key to the regulation of relations between and behaviours of men and women. Rules governing relationships and gender roles are important in fostering an understanding of practices, beliefs and traditions that shape perceptions of domestic violence, with the violation of rules and the failure to fulfil expected roles often used as justification for domestic violence.

The myriad rules and prescriptions relating to gender roles and relations are mainly applied to women, the work they must do, the role they must play in family and society, who they may associate with, their movements, and so on. These rules aim to keep women, and to a lesser extent men, in their place. The traditional structure places men as the head of the household with authority and power over wives and children; women perform their designated roles of wives and mothers and subsume themselves within these societal roles. This section explores views and perceptions of the gender roles assigned to men and women, the rules and practices that underpin these roles, and discusses the reported challenges to these.

Participants in ten of the focus groups conducted held a very traditional notion of womanhood, centring on the woman's role as wife and mother. A theme that ran through seven of the groups was the imperative to protect girls and women's sexual 'integrity' and regulate their sexual conduct throughout their life stages. In only three of the groups did participants emphasise the importance of young women prioritising their education, obtaining employment and saving their money. Some of the respondents were clearly quite critical of the differential and unequal treatment of boys/men and girls/women. These participants contested the myriad rules that girls and women are expected to adhere to:

Most of the time rules only apply to women and I don't think it is fair when the same rules are not told to men. [The elders] only tell [the man] that it is okay for him to have more than one partner, they would say to a woman she should let her husband be. I have never heard them saying to a man he should love his wife. (GT)

There is a layered development of rules over the various stages of a woman's life. The regulation of a woman's contact with members of the opposite sex – and related thereto her movements, dress and even her relationships with other women – are consolidated over these life phases. From childhood, a girl is groomed and prepared for marriage, and the role through which she – and her family – will be realised. A young woman who fulfils her domestic responsibilities and duties, and follows prescripts on movement and behaviour (being obedient, respectful, dressing modestly etc.) is respected, and similarly, is perceived to be showing respect for herself, her elders and her parents. The reputation and integrity of a family stands or falls according to the extent to which its members have schooled their daughter to be a good, obedient and respectful wife. This is because women are described as the property of the family, especially of the father as head of the family. Preserving their integrity by raising a respectable wife for a man brings status to a family, and 'protects magadi''. There is tremendous social pressure on women to fulfil the prescription for a 'good wife'. The behaviour of a woman through the phases of courtship and marriage is indicative of the status and reputation of the family. This focus on a woman's behaviour may account for the disproportionate responsibility women carry to ensure the success of their marriages and the reason why women often opt not to leave abusive relationships.

Enforcement, and the locus of this enforcement (i.e. the person who holds the authority to enforce the rules), is central to the notion of rules. This is critical to a discussion on domestic violence, since it is in the perceived or actual breach of laws that women are subject to discipline and, in most cases, to violence. Some women adhere to the rules on the surface, but subtly push the boundaries and manipulate the rules. Thus, the direct challenge to the rules may emerge from women's rebellion.

2.2.2 Roles, Rules and Authority Prior to Marriage

In the different stages of a woman's cycle of intimate relationships, the 'holder' of authority over the woman for the enforcement of social rules varies.

Dating relationships A young girl of dating age is expected to conform to certain rules related to conduct, work and relations with the opposite sex. A desirable daughter is described in general terms as one who is humble and respectful; attends school; does household duties; is home on time before sunset; does not spend a lot of time hanging around with friends ('naughty girls') in the streets; listens to her parents; wears modest clothing; and does not fall pregnant. A girl should behave respectably – be good and obedient – if she is to be a 'real woman'. A young woman will not be respected if her house is dirty: she must do household chores, washing and ironing, and keep herself and the house clean. The extent to which a young woman observes these rules determines the likelihood of her being a good wife and thus her candidacy for marriage.

Manners mean a lot and they would get married if they behave themselves. Who wants her son to marry some rough and uncultured young woman? We prepare our daughters for the future when we do that. (LT)

When a woman is of dating age, the imperative driving the rules to oversee and regulate women's relations with men is to ensure that a young woman does not fall pregnant outside of marriage. This is explained in different ways – for some participants, a young woman falling pregnant damages her prospects for marriage and the realisation of her most significant role in life. Young women who openly consort with young men and flaunt their relationships are characterised as 'bad', disrespectful and, at the extreme, whores. Their behaviour speaks badly of the family, who are condemned alongside their disrespectful daughters. A pregnant daughter, or daughter with child, is damaged goods/property and will not fetch 'good magadi'. At the extreme, respondents in one of the groups talked about 'their lobola' being damaged by a pregnancy outside of marriage.

As a parent you monitor your child (that is the girl child). Boys can easily trick and she can shame the family by falling pregnant. If she falls pregnant you will not get the magadi you deserve like other men. (BTL)

Some participants were of the view that a young woman falling pregnant damages her prospects for study and the possibility of building an independent life. In this view, women's role in society is perceived as encompassing a life and possibilities beyond or inclusive of marriage. The research participants who emphasised the importance of young women prioritising their education, obtaining employment and saving their money characterised marriage as onerous to women, and wished to see young women having a greater freedom of choice before them.

I will tell [my daughter] that education is important for her freedom and happiness. I will also tell her that she must work first before she can get married, that marriage is not easy. She must have money before she can commit herself. She must not have boyfriends before she has finished matric and is working. She must be afraid of men and stay away from them. (GM)

From when a woman is of dating age until her betrothal, the locus of authority over a woman lies with her parents (particularly her father). The parents are responsible for articulating and enforcing socially or traditionally-sanctioned rules for their children, with a special emphasis on their girl children. The parents (again, particularly the father) are also responsible for disciplining a woman or young girl when she has 'misbehaved'.

Across the groups, there seemed to be general acceptance that a boy dating a girl has no or minimal 'right' to exercise control over her behaviour, and physical violence as a mechanism for 'disciplining' girls and women that are either flouting, or are perceived to be flouting, rules is generally not tolerated. Because the locus of authority over a young girl is her father – and the daughter is the property of the family, and the father more specifically – the condemnation of the violence perpetrated by a boy dating a girl is related to the violation/damage of the property belonging to the father. The boy is disrespecting the father – overstepping his position and therefore challenging the powers of the legitimate holder/s of authority – if he beats the girl.

No man is allowed to touch my daughter unless she is married to that person. He is disrespectful of me if he beats her. He clearly shows that he will abuse her if he marries her. You can discipline your wife not someone you are not married to and did not pay magadi for. (BTL)

Some respondents rejected violence in dating relationships altogether as a matter of principle, and made no reference to the locus of authority. Where there is disagreement between and within groups on violence in dating relationships, the differences centred more on whether boys have a 'right' to discipline girls. Beating a woman is a 'right' or prerogative earned through marriage. Participants in seven of the focus groups condemned young men for beating their girlfriends. In five of the groups participants reasoned that the young men have not paid magadi and therefore have 'no right' to beat the girl, suggesting that some participants believe that married men are permitted to beat their wives. Their opposition to the violence principally relates to the question of authority. The locus of authority for disciplining a woman lies with her father/parents until she is married. In the balance of the groups, participants indicated that violence ('the hand') does not resolve matters of the heart, and that alternative ways need to be found to solve problems and conflicts arising in dating relationships.

Participants in six of the focus groups indicate that the young woman is generally held responsible for the violence resulting from a misdemeanour, usually related to some breach in the social rules governing young women's behaviour. Examples given here include a young woman dating other young men or being seen with other young men; shouting at her boyfriend or speaking to him disrespectfully; failing to take instruction from her boyfriend or inform him of her whereabouts.

While not conclusive, the focus group discussions suggested that beatings may be justified where a young woman is perceived to be or is actually 'playing the field' (seeing a number of boys simultaneously), a contravention of a most deeply held social rule.

If the boy found the girl with another boy he had the right to beat her. He must call the girl to order and teach her that she should not be going out with a lot of people. (GM)

Moreover, a few of the research participants equated beatings with love and care:

You see sometimes the boy hits the girl out of love and care. He is serious about her and has good intentions to marry her but she is messing him around. He is just threatening her to stop misbehaving. Maybe he does not want people to see that his girlfriend is misbehaving and people will lose respect for her. (BWTL)

Once magadi has been paid but before the woman leaves her parents' home
Once magadi is paid, a further layer of rules is added – a young woman is advised to 'choose her friends carefully'; inform her parents about her movements; change her dress (she is no longer attracting a potential husband); and is generally not permitted to socialise with the opposite sex (in settings that might open the possibility for sexual liaison, for example, parties or taverns). She is also required to increase her household duties to prepare herself for taking care of her husband and keeping him clean and respectable.

On the payment of magadi, control over a young woman's behaviour shifts towards a greater, but not necessarily equal, role for the male partner. There seems to be a tacit acceptance that he has a role to play in monitoring and regulating her behaviour – preserving her integrity and standing as his future wife. The research suggests some notion of joint custodianship if magadi has been paid for a woman. There is, however, some variation in understanding as to how a male partner may exercise this role as joint custodian. Participants in some of the groups indicated that a man who is engaged to a woman is permitted to verbally reprimand her and draw the attention of her parents, usually the father, to 'poor or bad behaviour'. The father is, however, principally responsible for checking the behaviour of the woman for whom magadi has been paid.

Cohabiting relationships
In two of the groups comprised of men and women associated with the institution of traditional leadership, some of the participants were deeply critical of the practice of cohabitation. The perceived increase in cohabitating relationships, for which women are principally blamed, is closely associated with the progressive disintegration of the institution of marriage. Some of the participants in the traditional leaders group characterised women in cohabitating relationships as lacking in self respect because they give themselves freely to men by moving in with them before the payment of magadi. They also perceive cohabitating women as manipulative and self-seeking, whose primary reason for entering into these relationships is to extract resources and financial support from powerless men. Women are accused of bringing shame on themselves and their families by entering into cohabitating relationships. A woman in a cohabiting relationship is also perceived to be causing problems for herself because the man's family does not know her and will not come to her assistance if she has marital problems. Her children will not bear her partner's name, which is considered a disgrace. They will also not be recognised by her husband's family.

Women who are in such unions have endless problems. The families of the man do not know their kids and in some instances they do not want them. The ancestors do not also know the kids or where the kids are. Women should not marry themselves; it is too shameful to their families. The people stop respecting them. (BTL)

Cohabitation is seen to undermine the essence of marriage as it has been practiced traditionally in a number of ways. Firstly, the union is not negotiated between the families. Participants talk about people 'marrying themselves'. Magadi is not paid and there is, therefore, no binding of the two families through marriage. Secondly, and most importantly for understanding the resistance to cohabitation, through marriage men are traditionally vested with the status of head of the household, which brings with it a number of privileges and, to a much lesser extent, obligations. Marriage is recognised by the families of the husband and wife, and accords the husband's family a particular role in the regulation of relations between the two parties. Women and men who cohabit therefore place themselves, to some extent, outside the box of marriage and the complex rules, prescriptions and codes that structure relations between men and women in the most primary social institution, that of marriage.

2.2.3 Roles, Rules and Authority in Marriage

Further rules are imposed on a woman entering into marriage. She is told to respect her husband as head of the household (she must realise that she is not his equal); be humble and respectful; never question a man; never shout back; keep her man clean and respectable; never question a man's whereabouts; endure his affairs; undertake domestic work for her in-laws; respect and adapt to the ways of her in-laws; have meals prepared on time; take care of the children; and always inform her husband of her whereabouts. Women take sole responsibility for household chores and once they have children, they take sole responsibility for rearing them. Men's responsibility is to work in order to provide for their family and to exercise authority over their wives and children.

Once the woman is living with her husband, the locus of authority shifts very clearly from her father to the husband. The authority to regulate a woman's behaviour, set rules for conduct and check 'unacceptable' behaviour is located with the husband, and to a lesser extent, with his parents if the couple resides in the home of the in-laws. The key issue related to the authority of a husband over his wife is whether he is permitted to discipline or chastise (i.e. beat) her.

A central theme running through the research, and articulated by the female participants specifically, is that of the hardship of marriage. This is related to the definition of women's central roles in life as wife and mother. Many of the female participants spoke of the difficulties they have to endure in marriage.

A woman can work six to six, twenty-four hours. (RM)
When he gets home from work he just sits there and I become his slave. (RM)

Two women provided a powerful analysis of the weave of institutions, idiomatic expressions and practices that enforce and reinforce men's power over women. In their analysis, these inequalities create the conditions for violence against women and keep women locked into abusive and unhappy relationships:

This treatment men give woman is brought about by the unequal powers and idioms given to women and men. The one idiom about men says 'monna ke thaka o nabile' … 'man is like a pumpkin leaves, he stretches as he likes' and the other says 'lebitla la monna le tseleng.' Man's grave is by the roadside' (figuratively, this means that a man is not bound by marriage and could end up anywhere). But then again it says about women 'lebitla la mosadi le bohadi' … 'woman's grave is at her in-laws' (figuratively, this means a woman may not leave her in-laws homestead under any circumstances). This is a very oppressive idiom. It locks women in abusive relationships while at the same time there are those idioms like the one mentioned before that men are free to explore their potentials. All this is said to give men power over women. Women are told to stay in marriages that do not work. Men can have affairs and make mistakes but a woman could not. In some cases a man who is having an affair would give all his salaries to his girlfriend and come home with no money for the children. When you ask him about the money he will say he was robbed on his way home. But then you wonder on which way home because he came home very late. (GCS)

While some women commented on the hardship of marriage, and offered a critique of this, other participants talked about hardship and difficulty as duties that married women must bear. On the payment of magadi and just prior to the traditional wedding ceremony, a woman is counselled by her father and the female elders in her family to prepare herself for the drudgery and hardship of married life.

(In the wedding ceremony) … the girl is told about the burden of being married and what is expected from her as a woman. The elderly women keep telling her about being a real woman and building a family. (BTL)

In the Little Teachers group, some of the women contested this prescription and indicated a shift, saying that women are now realising themselves and obtaining more meaning in their lives.

The division of labour
Many of the rules governing gender relations in marriage pertain to the division of labour. A fairly traditional division of labour was described by many of the participants. For some types of work, there is a very strict division of labour. Chores such as changing nappies or feeding the baby, ironing, washing the clothing of other family members, or polishing the floor is almost unquestionably women's work and forms part of the duties they must fulfil as 'good' wives. A woman can never (or very rarely) ask or tell a man to perform a domestic duty, although some participants did indicate that she can do so if she is occupied with another domestic task that requires her pressing attention. Some participants indicated that a woman could respectfully ask a man to perform a domestic duty if they are alone. But repeated requests may be interpreted as 'disrespectful or undermining' of the man and may result in a beating. Participants in seven of the focus groups indicated that it is forbidden for a woman to ask a man in the presence of others to perform a domestic task. Participants in six focus groups mentioned that a man may undertake certain domestic responsibilities if he volunteers.

Participants in five of the groups (interestingly, three were groups of traditional leaders, and one of women associated with the institution of traditional leadership) talked about the need for flexibility in determining the division of labour, and acknowledged that the division of labour might vary from one family to another depending on what has been negotiated between the parties. There was an acknowledgement in these groups that some men do help women at home. The practical circumstances of the family are very important in determining the division of labour. For example, a male participant in the Bollantlokwe Traditional Leaders group mentioned that he lives alone with his wife and hence assists with domestic duties. Participants in six of the focus groups mentioned a few common points of flexibility in the division of labour: when a woman is sick or is away visiting family/attending to social duties, such as funerals, then it is permissible for men to take on expanded domestic responsibilities. However, if a woman is sick or tired she is still expected to take primary responsibility for the care of babies; men may only assist.

Some of the research participants in the Greenside Teachers group commented, on an abstract level, that the traditional division of labour is unfair to women and should be challenged. Some of the positions articulated through this group may reflect their class location – employed women have more leverage because of their economic independence. An income also presents the woman with options, such as employing domestic assistance from outside the home.

A man is supposed to know his wife's weaknesses as much as a woman should know her husband's weaknesses. If he knows that his wife does not like doing the washing he should employ someone to come and do the washing, as much as if he knows that his wife cannot cook he should also employ someone who can cook in the house and he should do that in the name of love. I don't want to condone violence. (GT)

Some of the group participants talked about changing practices, or at least about a desire to change practice. In six of the groups, participants commented that household chores are taught to boys and that it is acceptable to have sons cleaning the house.

There is a young boy at home whom you won't find playing on the street. He stays at home and does his school work. If you can visit my place and enquire from my father about the person who is cooking he would tell you it is that boy. (LCS)

While changes to the division of labour are not traditionally sanctioned, some marriages, even amongst the older generation, may see men taking on more domestic responsibilities. Some men in the Little Traditional Leaders group reported that they do assist with childcare, such as taking babies to the clinic, changing nappies and assisting with feeding. Participants in other focus groups reported the following:

Well if you are busy and he is doing nothing I guess he should help you if he loves you.
My husband used to help me all the time. (LM)
I would make an example of myself, I have a five year old and her father used to wake up and feed him during the night and I would just sleep so that in the morning I would be the one who will take care of the child. (RWTL)

One woman commented that 'real men' assist their wives, while another alluded to the 'new rights' creating the conditions for a fairer division of labour. Women who work may obtain more assistance than non-working women (although this did not clearly emerge from the research), and the practice of husbands and wives establishing homes away from his in-laws may facilitate a different and more equitable division of labour:

Today marriages are not the same as it was. Today you will find a husband and wife who are both working and when they come home from work they have to help each other like when I'm busy with the baby he has to prepare the fire so that I can start cooking. It is not the same as women who got married many years ago because they used to stay at home with their mothers-in-law hundred percent. (GWTL)

Some participants talked about the need for a fairer division of labour, but did not necessarily refer to a change in existing practice. In general terms, respondents felt that a husband and wife should help one another in the home, with a particular emphasis on men assisting women with childcare and when women are busy with other domestic chores.

A man and a woman should help each other, yes, you married her but you should help each other. (GTL)
If he gets married he must be told that his wife is not a slave … (LM)
2.2.5 Why do Men Beat Women?

The research sought to uncover how people explain the abuse of women within intimate relationships. Domestic violence must be located and understood in relation to gender relations, which find concrete expression through the 'rules' that govern the day to day work, conduct and behaviour of men and women. Across the groups, explanations for domestic violence are located in, and responses sought through, the actual and perceived flouting of these rules and conventions by women.

People's views on violence, and whether it is justifiable or not, are deeply personal and influenced by the specific circumstances surrounding an incident. The factors that would seem to inform a person's response to violence are diverse and include:

Violence in marriage
Some informants rejected domestic violence in marriage outright no matter the circumstances. Participants in five of the groups specifically mentioned that there is no justification for a man to beat a woman - even if she is at fault for having breached established rules, 'he is wrong to hit her' (GM, LM).

Nowadays beating a woman does not work and should not be used. Some men still use it but they know that it is old fashioned and illegal. (RCS)

None of the research participants explicitly stated that physical abuse is an acceptable method of 'disciplining' women in the context of marriage. However, it can be inferred from the discussion on violence in dating relationships that on the payment of magadi or at the conclusion of marriage, physical violence may, under certain circumstances and in certain quarters, be regarded as a legitimate tool for disciplining women who overstep or challenge social rules and boundaries. This was the position expressed by participants in five of the focus group discussions.

We tell him that he will beat her when he has paid magadi for her and she is his wife. (LT)
A boy has no right to beat a girl up, especially if he did not even pay a cent for her. I will ask him if he is married to my daughter, and if not, I will then ask him what gave him the powers to beat her up. (BTL)

Even where there is a tacit acceptance of a man's authority to discipline his wife, some participants pointed to limitations on the exercise of this authority:

If the woman has gone back to her family, the man's family must follow her … The woman's family will then say yes they have the child there with them but she is complaining about the way the husband is treating her. Then the man's family will ask for forgiveness … The woman's family can ask money or cow for damages to their daughter. They then tell the man that when he married her they gave him her body but her head belongs to them. 'Re gofa mmele wa ngwana wa rona but tlhoho ya gae ke ya rona, o gopole seo ka metlha …' meaning we are giving you her body but her head belongs to us (that means that he can use her body for his pleasure and to carry his children but her life still belongs to her parents … he should not kill her if he does not want her any more. He does not have a right to hit her). (BWTL)

Some informants indicated that there must be a reason for the violence, and often attributed the problem to the woman. In seven of the focus groups, some participants indicated that men hit women for a good reason. She must have provoked it in some way or asked for it or have done something wrong – and the 'provoking' lies in the challenging (perceived or actual) of the rules governing relations between men and women. When a man's traditionally privileged role is challenged, then the woman will be disciplined (beaten). Beatings are often justified on this basis. In the Bollontlokwe Traditional Leaders group, participants assumed that women 'ask' to be beaten.

You don't have a right to hit your wife - you must hit her for a reason and not without a valid reason. (LTL)

Some participants adopted a more nuanced view on the beating of women who shout at or contradict their husbands. The participants who spoke to this issue condemned the use of violence, but simultaneously also criticised the behaviour of women who are 'disrespectful' towards their husbands. They advocated alternative remedies to the resolution of the problem. Women should take personal responsibility for fulfilling the social rule that women treat their husbands with respect and never raise their voices to them. Alternatively, traditional dispute resolution mechanisms should be activated, involving members of the husband's family in resolving the dispute. Some participants also indicated that the woman could be sent back to her family to be 'taught manners'. Despite opposing violence as a means for 'disciplining' women who misbehave, the intention of the alternative remedies offered up was to restore women to their traditional roles and ensure that they adhere to the given social rules ordering relations between men and women.

Some participants in nine of the focus groups argued that it is acceptable for men to beat women if they don't talk 'properly and respectfully' to their husbands. Violence is especially justified if the woman 'disrespects' her husband in public.

Some participants in nine of the focus groups indicated that domestic violence may be justified in instances where the woman has failed to perform her domestic duties and, more importantly, if she asks or tells a man to perform domestic work, particularly in the presence of others. This breach undermines and challenges the traditional roles of men and women in the household and is seen as illustrating a woman's disrespect for her husband. If a woman tells a man to do domestic work then 'she is asking for it' and will get exactly what she deserves. However, the specific agreements that have been struck between a husband and wife regarding the division of labour are also important in determining how people view domestic violence. In the Bollantlokwe Civil Society group, some participants acknowledged that if a husband has been performing domestic duties and then beats a woman for asking him to undertake a domestic chore, he would be in the wrong. Some participants in the Bollantlokwe Teachers group indicated that men who are married to working women need to take more responsibility for domestic work, and condemned a man for beating his wife for failing in her domestic duties if she had good reason.

Participants in ten of the focus groups indicated that domestic violence may also be justified if a woman is having an affair, or if her husband or male partner suspects that she may be having an affair. This suspicion may arise when a woman fails to inform her husband or partner of her movements, when she is seen in the company of other men, when she refuses him sex, or when she socialises ('goes to parties') without her male partner or husband. As previously discussed, women's movements are controlled to regulate her contact with the outside world and especially with other men. It may therefore also be justifiable for a man to beat a woman if she fails to seek his 'permission' to go out, or inform him of her movements. This prescription and justification for violence applies to marital relationships, and also to relationships in which magadi has already been paid for the woman but she may not be residing with the man.

The final major justification for domestic violence relates to the prescription that wives must service their husbands sexually. In the context of the fear around the transmission of HIV/AIDS, this prescription is coming under a degree of criticism and pressure, with some women asserting that men's conjugal rights must be limited. Read together with the social belief or prescription that men have affairs, women are fearful of contracting HIV/AIDS and some assert that women should have the counterbalancing right to insist that their husbands use condoms or that women be empowered to deny men their conjugal rights.

Women are also beaten for failing to perform their domestic duties. This was raised in eleven of the focus groups. Women are reportedly beaten if they do not take good care of the house, do not cook on time, loiter on the streets if the house is not clean, fail to do the washing and ironing, or do not take adequate care of the children.

It is the woman's responsibility to keep her man clean. That's why most men will beat their wives and if the elders are called in to help the woman will definitely be found guilty and negligent. She will be told that she provoked the man. (BTL)

In nine of the groups, participants mentioned that women are beaten if they ask men to perform domestic duties:

In our culture a baby belongs to its mother. The woman cannot ask a man to change the baby's nappy. Even if she is tired she is a mother and has to get up. That is why some men beat their wives - because they are disrespected and undermined. If Selina comes to me as my daughter-in-law I would tell her that she was asking for it. (BCS)

In six of the groups, participants mentioned that women are beaten for misusing household finances to gamble and drink. The Little Teachers group indicated that this is a primary reason for men beating women.

Participants in nine of the focus groups indicated that women are beaten for answering their husbands back, being 'cheeky' or disrespectful, and/or raising their voices to their husbands, especially in public. These behaviours are characterised as disrespectful and undermining of men.

A key reason given for violence in intimate relationships relates to women violating, or at least perceived to be violating, a set of rules that regulate and circumscribe women's relations with and interactions with other men. Women are beaten for having an affair, for suspicions that they might be having an affair, for failing to dress according to their husband's rules, and even for 'just looking beautiful'.

At some point when he suspects that a woman is having an affair or when she is having it the men get so angry they can even kill you for that. It is an insult on men for their women to be having affairs with other men. (GM)
(Girls will be beaten) … when a man will take money from his home and give it to [his girlfriend] and the girl is seeing other men … they say wa jewa (a man who is used for his money). (RWTL)
Some men would beat a woman because they are jealous. If the wife is always clean and she is beautiful … (LM)

A husband may beat a woman up during her pregnancy because of his suspicion that the child was not his. This illustrates the link between a perceived violation of a cardinal rule relating to fidelity and domestic violence.

He even beat me when I was pregnant. One time I was so ill I thought I was going to have the baby in 7 months … He said that he does not think that the baby is his. He has been suspecting it for a long time. Then I understood why he beat me like he did during that pregnancy. (GCS)

Women are also beaten for not informing their husband or boyfriend of their whereabouts, or not seeking their permission to go somewhere. This rule relates quite centrally to the regulation of women's movements and their interactions with other men. Women who fail to seek permission to move outside of the home are often suspected of having affairs. This reason was raised in four of the focus groups.

Sometimes a guy comes to your house and does not find you, he will beat you when he sees you because he thinks you were with other men or boyfriends. If you go to a party without telling him, it's going to be a problem; he will kill you. (LCS)

In ten of the focus groups, participants indicated that men beat women for refusing to have sex with them. A key rule that underpins a traditional concept of marriage is that the woman must service her husband sexually. Where a woman refuses to perform this duty, she may be accused of having an affair, and is often beaten. In the Little Traditional Leaders group, participants said that a man might kill a woman for refusing him sex. Participants in one of the groups of traditional leaders indicated that this is a primary reason for domestic violence.

Some men would beat their wives or girlfriends if they refuse to sleep with them. Some men think that you are having an affair if you refuse to sleep with him. (GM)

Participants in eight of the groups indicated that women are beaten for inquiring after men's sexual liaisons or affairs. As referred to above, women are strictly instructed at the time of marriage to accept men's affairs, to not 'follow their men around', and never inquire about a man's movements outside of the home. Women who inquire after men's affairs are, therefore, violating a key rule underpinning marriage.

Some men when they are guilty they become abusive and defensive. They use terror to avoid answering questions. He might beat her up terribly for asking him about his life. He will even accuse her of spying on him. (BCS)

Participants in five of the groups talked about men beating women who exercise independence because they earn their own money, refuse to submit to the authority of men by, for example, not handing over the earnings or grant money that they may have access to, or by asserting their rights against him. Mention was made that unemployed men beat women, suggesting that where there is a dislocation in traditional roles men might turn to violence to assert their authority over women. These reasons are the most explicit evidence of men beating women to maintain traditional relations of authority and assert traditional roles for men and women.

When a woman is working and the husband is not, he will become jealous of her and start accusing her of having an extra marital affair. Maybe when a woman comes home late from work the husband will say it is because she is having an affair. (GT)
There is also a case of the man who beats his wife for her money. The man does not work and the woman works. He demands money from her for liquor and if she refuses to give him money he beats her. Last time he broke her arm. (GCS)

The group discussions suggest that domestic violence may be more readily justified if it comes around the breach of rules that require women to: (a) treat and speak to their husbands with 'respect' (i.e. honour his position as head of household and take instruction from him); (b) undertake their domestic duties (cook, clean, wash and iron and take care of the children); (c) be sexually faithful to their husband/male partner (including the set of sub-rules concerning the regulation of her movements and interactions with other men); and (d) related to the previous point, service their husband sexually.

Participants also referred to the following as explanations for domestic violence:

Justice has a problem that is why he doesn't want Conny to have friends, he is jealous and where does jealousy come from - love - and that is what makes him to beat her. (RWTL)

Therefore, the underlying rationale for domestic violence is that because women are the property of men (whether father or husband, and in some instances, male partner) and under the control of men, men's status and standing in society is closely tied to women's behaviour. A woman whose actual or perceived behaviour does not conform to established social rules is showing disrespect to herself, her husband and her family, and should be punished. 'Disciplining' a woman – restoring or ensuring her compliance to rules and prescriptions relating to women's role and conduct in a given social context – therefore establishes a woman's respectability and status, and by association that of her husband, male partner and family as a whole.

2.3 Recourse

This section discusses research participants' views on the different options available to women seeking recourse for abuse in different types of intimate relationships. The discussion begins with a description of some of the social and economic obstacles women confront when seeking support for physical abuse in intimate relationships. The options available to women vary according to the type of intimate relationship, the history of abuse, and previous efforts to address the violence. In discussing the different options, the analysis reflects participants' views of both traditional mediation and the criminal justice system.

2.3.1 Domestic Violence – The Silent Scourge

Participants in seven of the focus groups mentioned that women, married and single, tend not to disclose abuse. The reasons given to explain women's silence are the following:

Other focus group participants suggested that the silence surrounding domestic violence, and the requirement that these be treated as private, family matters, is a practice of the past and may be changing. Some women implicitly critiqued the silence surrounding domestic violence and the related requirement that women endure abuse and mistreatment in marriage. However, a few participants offered an explicit critique of the silencing of women, and the effect that this has on them.

The women should also not keep secrets about how their men treat them. The majority of women are sitting on very hot coals. They do not want to tell in case they shame their husbands and themselves. They will not tell any one until the situation gets out of control. (BTL)
2.3.2 Recourse in Dating Relationships

Participants in eleven of the focus groups indicated that when a young woman has been beaten by her boyfriend, the family of the girl should seek a meeting with the boy's family to discuss the matter. It would, therefore, seem that a traditional approach of reconciliation is favoured over the more confrontational recourse through the criminal justice system. The support for this approach may, in part, be attributed to the more rural context in which the research was undertaken. It is likely that most young men and women that are dating reside in the same village, and there is thus an imperative to retain and build relationships between residents rather than undermine them by involving the police and courts.

The process usually followed is that the young woman's parents (and/or members of her family), accompanied by the young woman, visit the boyfriend's parents. The young man and woman are each given an opportunity to give their version of events. The focus of the mediation process is to build an understanding of the incident of abuse and the circumstances surrounding it; to locate the party at fault; and to advise the parties on how they should behave in future. The young man is typically instructed not to beat his girlfriend, while the young woman is advised to change her behaviour. The boy may also be advised to report any 'misbehaviour' on the part of the young woman to her parents, usually the father. The mediation is, therefore, focused on two outcomes: restoring the locus of authority over a young woman to its legitimate place, that of her parents or father; and restoring relationships and reconciling the parties by identifying the 'source' of the problem and instructing the parties to conform to given social rules. Hence, some participants indicated that the focus of the mediation may not necessarily be to end the relationship.

If he beats a girl we must go to his family to report the deed. The mother must ask some friends or relatives to escort her to the boy's home. She asks people to escort her as witnesses and normally as the parent she will be very angry and just shout and fight. The father can also go but also with relatives or friends. The girl must come along with them. When they get to the boy's home they report the issue then the mother or father of the boy will call him and ask him if it is true he did beat the girl up. The girl will be asked to relate what happened in front of every one and then the boy will be asked why he beat the girl. (BWTL)

Participants in seven of the focus groups indicated that they would report the abuse to the police. The following reasons were presented for seeking recourse through the criminal justice system:

Participants in five of the focus groups indicated a 'phased' approach to resolving the problem of abuse in dating relationships. They would first seek reconciliation through the traditional dispute resolution process, and would involve the police only if this route failed to resolve differences and deliver the results they had anticipated, either immediately or in the short to medium-term. An immediate expectation might be for the young man's family to attend to medical expenses and/or pay damages for the harm that the young woman has incurred.

We have a right to fine the parents a cow or some money for damages … If she is injured … the boy's family will have to pay for the medical costs and buy fruits for her until she is well again. I am not going to nurse her back to health on my own. If they refuse they know that I can lay a charge against their son … (BTL)

In the short to medium-term, some participants expected that traditional dispute resolution would end the violence. As such, three of the participants clearly indicated that if the young man beats his girlfriend again then they would lay charges against him.

(If we have already discussed with the boyfriend and he beats her again) - I will report him. We don't have time for him now. (GTL)

Participants in three of the focus groups flagged other forms of recourse for dealing with abuse in dating relationships. In one case, a young woman's brothers beat her boyfriend for abusing her, and in another, a mother went to the boy's parents and told them he should stop seeing her daughter. Another participant mentioned that she would take the matter of abuse to the lekgotla (traditional committee).

2.3.3 Recourse in Relationships in which Magadi has been Paid

The research suggests that once magadi is paid, but the woman is not yet residing with her husband/in-laws, a traditional mediation process is supported in cases of violence in a relationship. The payment of magadi, which binds the two families, is a critical element in the constitution of a customary marriage and there is a strong desire to resolve conflicts and differences rather than exacerbate them. Involving the police and the courts in such matters was perceived as inevitably leading to the breakdown of the marriage.

The different views held by participants as to whom the woman should approach for assistance in the case of domestic violence, reflects her 'in limbo' status. Some participants indicated that the woman should advise her parents and representatives of the woman's family – either her parents or the aunts and uncles (maditsela) that were negotiating the marriage – should visit the family of the fiancι to discuss the matter. The man and woman would be asked to explain what happened and why he beat her. The matter they fought over would be discussed and the two parties would be advised accordingly. Typically, the young man is told not to hit his fiancιe ('we address the issue of the hand first before addressing the reason why he did it', BCS), while the young woman is instructed to change her behaviour to that befitting a woman for whom magadi has been paid.

If my daughter comes home beaten by her fiancι I will want to know why she was beaten. I will then ask her aunts and uncles who were receiving her lobola to go to the boy's home and find out why he did this. If they find out she was wrong she would be told that she must listen to the rules and laws her husband lays down for her. The man has paid lobola and can put limitations on where she goes, with whom does she go around with, can choose friends for her and also tell her how to dress. (BWTL)

Another view is that the woman should not approach her family, but must rather speak to her mother-in-law or the man's uncles. The rationale is that the parents of the woman will be biased towards her and therefore cannot deal with the matter 'objectively'.

Participants in two focus groups indicated that it would be acceptable for the young woman to report a first incident of abuse to the police. Participants in four focus groups explicitly opposed the involvement of the police. Some participants in one of the focus groups indicated that the matter should be reported to the police if the traditional mediation process fails – if the man does not respect the process, if one of the parties is dissatisfied with the outcome of the traditional mediation process, and/or if the man assaults the woman again.

Participants in the Little Traditional Leaders focus group described a 'tiered' system of recourse. Women must 'visit' each of these levels sequentially in order to obtain support:

[He paid lobola, then finds her at a shebeen and hits her] In that case we go to their families and talk to them, if we don't reach an agreement we then go to the community council … if that doesn't help we go to the police forum … If we spoke to the family, the community council and police forum and [are] still not able to solve the case we will then say we are defeated. (LTL)
[The police] know how it works. They know that … it must be dealt with in a certain way, that's why they send her back home [if woman has not discussed the problem with his parents first]. (LTL)

Participants in two of the focus groups indicated that a woman who is abused by her fiancιe would be advised to break up the relationship. This opinion does not seem to be widely held.

2.3.4 Recourse for Women in Cohabiting Relationships

There was limited discussion on abuse in relationships in which a man and woman cohabit but are not married. Participants in two of the focus groups indicated that in such relationships the option open to women confronting abuse is to go to the police. Women may face particular constraints exercising this option. One participant mentioned that women might not go to the police for fear of the relationship ending and being left alone with the children.

She can only run to the police station and that destroys many marriages. The men in such relationships might choose to leave because they are afraid of being arrested. (BTL)
The women are also afraid to be left with children so they don't normally go to police, not unless the abuse is severe or in most cases when they have new boyfriends. (BTL)

A teacher from Greenside Village expressed the perception that 'when you go to the police they would say the woman should pack her things and go because they are not even married' (GT).

Women would not be in a position to solicit support from the parents of her partner as the couple are not traditionally married, and no magadi has been paid. The woman is therefore not regarded as a member of his family.

Both parents and the ancestors do not bless the marriage. The magadi that united the two families was never paid and that causes problems. In that case women from such unions cannot claim any help from his in laws. She has no right to run to them for help. (BTL)

Participants from two focus groups indicated that tribal structures do not attend to the complaints of women complaining of abuse in a cohabiting relationship.

2.3.5 Recourse for Married Women

Traditional mediation
Participants in eleven of the focus groups indicated that married women confronting problems of domestic violence must follow the traditional mediation route. The prescribed route is for a woman to report cases of domestic violence to her in-laws, or alternatively to the uncles (or aunts) of her husband who negotiated the marriage (maditsela). After magadi has been paid and a woman has been married traditionally, she is no longer regarded as a member of her family. She is therefore required to address the problems with her 'new' family.

Procedurally, members of his family (whoever has been approached with the woman's complaint) would convene a discussion involving the husband, wife and members of his family. Each of the parties would be given an opportunity to recount their version of events, and are then advised accordingly. It would seem that the man is generally instructed not to beat his wife, while the woman is advised to remedy the behaviour that 'provoked' the beating. Both parties would be counselled or guided on how to deal with domestic problems and conflicts in the future.

Culturally in cases of misunderstanding the woman should report to her in-laws and not go to her mother's house so as to have a strong case against the abusive/violent husband. The same applies to the man in the event of any problems with the wife he must also report to his in-laws or the elders who were present when they got married; they are the ones to resolve their problem. (LTL)

If the woman is not satisfied with the outcome of the mediation led by members of his family, if he beats her again or if his family is unable to resolve the problem, then the woman or his family may seek assistance from her parents. Participants in five of the focus groups reported this as the next step in a tiered system of options for recourse. The woman's family may play a variety of roles. Firstly, they may send a delegation of family members to his family to discuss the problem. Secondly, the man's family may invite the woman's family to a roundtable discussion. Finally, the woman may return to her family home, in which case his family is traditionally required to follow her and enter into negotiations with her family. Her family may require compensation and are also permitted to counsel the husband.

If her in-laws side with their son then she can go back to her home and explain to her father why she has returned. Her parents would then wait for her in laws to follow her. When they arrive her parents would state her complaint and demand a cow as a fine. They will then tell her to go back to her husband after they have talked with him. (BTL)

The research also elicited people's views on the traditional dispute resolution process married women are expected to follow when they are confronted with domestic violence. Women participants in five of the focus groups were critical of traditional dispute resolution as they feel it is biased towards men. Participants indicated that the elders will usually ask why he hit you and will normally conclude that the woman provoked the assault. Women in the Greenside Teachers group were particularly critical of the attitude of elders to men's affairs in the context of HIV/AIDS. They argue that the elders condone men's affairs and simultaneously insist on men's conjugal rights within marriage, which places women at risk of contracting STDs and HIV.

Male camaraderie can also leave a woman feeling that justice has not been done and can result in her further isolation. One woman recounted her experience of enlisting the support of her uncle to intervene after her husband had assaulted her:

I remember my situation when I just got married. He beat me up in the evening and I went to tell my uncle about it. I went to the uncle that I know was very harsh and strict. I knew he would be angry with him and try to confront him. When we arrived with my uncle at my place I left them alone to talk. A few minutes later my husband came back to the bedroom to take some money and asked me what does my uncle drink. He bought the beers and they discussed the matter while still drinking. My uncle was not angry with him any more. He did not shout at him or tell him that he must never repeat it again. He kept saying that marriage needs patience and understanding. He did not ask him why he beat and that he should never do it again. He then started counselling us about the challenges of the marriage. After that he just left, drunk. Thereafter my situation got worse and I had no one to go to now. I knew that if I brought my uncles he would buy them beer and they would not discipline him. (GCS)

There is a specific critique of the role of mothers-in-law to whom the woman might report the domestic violence. Women participants in three of the focus groups complained that the majority of mothers-in-law stand with their son against the wife, and fail to resolve the problem. They mentioned that some mothers-in-law actively foment conflict between their sons and their daughters-in-law.

Only about 20% of mothers in law genuinely try to resolve the problems but the majority of them stand with their children. They will not address the dispute fairly, they always say that a woman should respect her husband and must not complain too much. They don't want to let go of their children and interfere too much. (GM)

Community members
Participants in four of the focus groups mentioned that the Tribal Council, teachers, neighbours and social workers play a minor role in matters of domestic violence. Teachers, through changes registered in the performance and behaviour of their pupils, may come to realise that there is a problem within a family. Teachers in the Greenside Village indicated that they might visit the family and if domestic violence is identified as a problem, they would usually refer the parents to social workers.

And sometimes we have problems with husbands when we suggest Social Workers and you know how stubborn these men can be, they will say you want them to talk to a young girl (Social Worker) about their problems. (GT)

Participants in two of the focus groups indicated that the support of social workers should be sought in instances where the husband abuses his wife on an ongoing basis, and where traditional dispute resolution options have not achieved any impact.

The research was not able to clarify the role of the Tribal Council or court in matters of domestic violence. While the involvement of traditional leaders in local dispute resolution processes, usually led by the Community Policing Forum (CPF), was identified, there has been no specific reference to the role of Tribal Councils. One participant from the Bollontlokwe group of traditional leaders indicated that:

Most of the domestic violence cases go straight to the police stations but there are those that still go to the Chief. Depending on how the two parties involved feel about their matter, the headman can suggest that the case be referred to the Chief. The council does deal with domestic related cases. (BTL)

CPF intervention
Participants in three of the focus groups indicated a role for the CPF in solving problems of domestic violence in marriage. Two options were outlined in this regard. Firstly, if a woman has alerted her parents to the domestic violence and they have sought a meeting with the man's family but the process has not yielded a satisfactory outcome, the woman and her family may approach the CPF for a letter to go to the police station. Secondly, the CPF may undertake a dispute resolution process that would involve the families of the man and woman, CPF members and traditional leaders. Both parties are given the opportunity to recount their stories about the domestic violence, and the CPF is required to locate the problem (the party to blame). Participants in the Ratjiepane Civil Society focus group indicated that they usually find that the woman is 'in the wrong'.

The CPF may issue the woman complainant with a letter to go to the police if she is dissatisfied with 'the ruling' of the CPF, or if the husband does not arrive for the dispute resolution meeting. The Ratjiepane group indicated that the police usually consult with the CPF on the outcome of the mediation process.

Police intervention
Participants in only one of the focus groups indicated that a married woman could go directly to the police, without passing through the traditional dispute resolution process, in order to open a case. The participants who spoke of this option indicated that the in-laws usually take sides and complained that the parents 'drag out the issue'.

Participants in four focus groups indicated that a married woman should take her case directly to the police (two groups indicated that it should be done via the CPF) if she has been seriously injured; for example, if her arm is broken. No other injuries were mentioned so it is difficult to ascertain specifically when this option would apply.

A more general view, indicated in six of the focus groups, is that a married woman may seek assistance from the police once she has pursued or exhausted options within the system of traditional mediation, and the physical abuse continues.

Participants in eight focus groups were generally critical of involving the police or the courts in matters of domestic violence in marriage. Participants in four focus groups offered only general statements opposing the intervention of the police and the courts, so it was not clear exactly what they are opposed to. Their statements could be read as a blanket opposition indicating that women should never make contact with the police, even if it is to have a person 'with authority' warn the abuser, which seems to be the intention of some women who approach the police. These statements could also be interpreted to mean that a woman should not phone the police, or solicit the support of others to contact the police, to request intervention in instances where the woman's life may be under threat. A more narrow interpretation is that women should not open criminal cases or obtain protection orders against their husbands.

In cases of marital rape, participants in nine of the focus groups indicated that a woman should bypass traditional systems and 'go straight' to the police station. It was not always clear what research participants meant when they talked about going to the police. Participants in four of the focus groups indicated that, if their husband raped them, they would lodge a complaint with their in-laws. Other participants variously suggested that the social workers should be asked to intervene, that the woman should bring the matter to the Tribal Council, and one other woman participant indicated that she would 'pack her bags and go back to my home. The man is not interested in the marriage and building a home' (GCS).

A participant in the Little Civil Society group raised the point of the expectations of women vis-à-vis the police (and by association, the CPF):

That's why we avoid taking certain decisions because if … the CPF gets hold of the husband and he is arrested, the family starves then the woman comes back and says that the CPF arrested her husband, but all along she wanted him reprimanded. The kids are now starving and once the husband comes back from being arrested, I'm sure that the love is gone. (LCS)

Participants in two of the focus groups clarified that they do not necessarily wish to bring charges, have their husbands arrested, appear in court and be sentenced to time in jail. Their wish instead is for the police (people whom they perceive to carry a measure of authority) to warn their husbands.

How can they arrest him? I just want them to warn him and tell him not to force himself on me. (LM)

Other women were educated about the implications of bringing a charge of marital rape against their husbands. They wished to have their husbands arrested for the crime of marital rape and see him serve time in jail, and were prepared to risk losing their marriage.

Protection orders
Participants in six of the focus groups registered a specific opposition to protection orders. The major concern with protection orders was the perception that protection orders undermined given social relations between husbands and wives. Men served with protection orders no longer know how 'to handle their wives', and the orders are perceived to undermine the power of the head of the household. Sexual relations are interfered with, either because the man is not permitted to enter the bedroom, or because he is 'scared' to ask for sex. Moreover, there is a view that women become 'cheeky and disrespectful' to their husbands and no longer seek their husband's permission to move outside of the domestic sphere.

There is this thing called Protection Order. It is very bad and destroys a lot of families. It creates hatred and men tend to hate their wives because they do not know how to handle them any more. They withdraw and become distant, they come home late just to sleep and they drink a lot to cover the frustration. It is very difficult to even ask her to make love to you as a man if she has a protection order against him. (BTL)

Participants in six of the groups indicated that 'involving the police and the courts', whether mentioned more generally or with specific reference to protection orders, signals the end of the marriage. Protection orders create 'anger and hatred' between the husband, his family and the wife, and may lead some men to pursue affairs outside of the marriage and/or leave their wives.

If you apply for a protection order against your husband whilst living with your husband that means your marriage is over. They will tell him that he is not allowed to come close to you, that means he is not allowed to come to your bedroom … (GM)

Participants in two of the groups contrasted the criminal justice system against traditional methods of mediation and found the former wanting. These participants argued that men and women should deal with conflicts culturally, and that 'matters of the heart' cannot be dealt with through the law. They emphasised that traditional mediation helps to build families and address the underlying problems, while the criminal justice system destroys marriages. Importantly, a participant in the group of Bollantlokwe traditional leaders emphasised that the criminal justice system does not offer women the remedies they seek – an end to the violence, but not the destruction of their marriage.

Separation and/or divorce
There were very diverse views on the option of a woman separating from her husband, and seeking a divorce. Participants in five of the focus groups advised that women should leave husbands who are repeatedly abusive, and where all available options for local or traditional dispute resolution have been exhausted.

It is not right to tell women to stay in such marriages or with men who abuse them … she must just go back to her home. (BWTL)
There is nothing like 'till death do us part' … you know that every month you are going to have a blue eye; he will end up killing you. You must walk out. He will make sure that you die because you will go to the police station. (GCS)

There are however, numerous constraints to women leaving a marriage. The first major constraint is the prescription to women that they must make their marriages work by enduring hardship and difficulty. A failed marriage is a sign that 'you are not a real woman'.

Are you saying she must just go and see how she supports the kids? She cannot take the kids that belong to another surname and family back to her family … Her kids need their father and she should stay and make the marriage work. We did that, we worked on our marriages and they eventually worked. She will not be a real woman if she just packs her bags and goes back to her home. Then she was not ready for a marriage at all. She should have known that marriage comes with big responsibilities. Lenyalo ho tshwana le o dula mashala a hlenneng (meaning marriage is like sitting on red coals). If one has chosen to be a woman one must stay and work on the marriage. (BWTL)

The second major constraint confronting women is the dilemma of how they will support themselves and their children.

Some would rather divorce than stay in abusive marriages. Even divorcing is difficult for a woman who has children because the man can just choose to refuse to pay the child support. The financial burden of single parenting is locking the women in their unhappy marriages. His family too would support that he should not pay maintenance because the children belong with him and if you leave him you should leave him with his kids. (BCS)

One participant from a focus group of women in Ratjiepane indicated that some women fear that their husbands will kill them if they leave.

2.4 Conclusion

Most research participants hold a traditional notion of womanhood, centred on a woman's role as wife and mother, for which a girl is groomed from childhood. Characteristics of womanhood are attributes such as respect, humility, diligent performance of household duties, and obedience. On the contrary, there does not seem to be an equal emphasise on roles and rules to be observed by boys and men, nor does the transgression of rules by men attract sanction. Boys are groomed to provide for and protect their families and so become the head the household. As men, they are expected to provide for their families, head the household and exercise authority over women. This perpetuates a power imbalance between men and women reinforced and perpetuated by rules, gender roles and idioms, all of which are oppressive towards women and lock women into abusive relationships while giving men the freedom to explore their potential.

All focus groups believe that violence in all its forms is a feature in intimate relationships in the villages in which they reside. Research participants located domestic violence within the context of rules that govern relations between men and women, with the actual or perceived flouting of these rules given as explanations for domestic violence. Where violence follows the infraction of traditional rules by women (whether married or in a dating relationship), there is a tendency to blame women for it. Participants' views on domestic violence (whether they believe it is justified or whether they reject it), are deeply personal and seem to be influenced by factors such as the type of relationship, circumstances leading up to the incidence of violence, the history of abuse in the relationship, the nature of the violation, and whether traditional dispute resolution mechanisms were previously resorted to.

According to most research participants, factors such as type of intimate relationship, the history of violence in the relationship and previous efforts to address the violence tend to inform the recourse likely to be taken by woman in resolving the violence. Women, single or married, tend not to disclose the violence for the following reasons: domestic violence is regarded as a private matter to be managed within the family; fear of shaming themselves and damaging the reputation of their family; and fear that reporting the violence will lead to a breakdown of the relationship, leaving them (and in marriage, their children) without financial support. That young persons conceal dating relationships from their parents also means that they cannot report abuse when it occurs in these relationships.

Two approaches seem be followed to address the violence in both marriage and dating relationships. Favoured by virtually all participants in relation to marriage relationships, and by some participants in relation to dating relationships, is a conciliatory approach to ending violence whereby a mediation is held between the two families with a view to understand the abuse, the circumstances that led to it, locate the party at fault and advise the parties on how to behave in the future. The second is a confrontational approach which entails reporting the matter to the police with a view to ensure the punishment of the offender. Most participants believed that where domestic violence has resulted in serious injury, the conciliatory route would not be appropriate, even where the parties are married, and the matter should be reported to the police at once.

In relation to dating relationships, those who favour a conciliatory approach believe that it serves to preserve the relationship between the two families because, in most cases, they live in the same village. Others, however, believe that the matter should be reported to the police because magadi has not been paid and the two families have not been joined in marriage, therefore the woman's family has no basis for meeting with the man's family to discuss the abuse.

Disturbingly, the CPF seems to be playing a filtering role in domestic violence matters by attempting to resolve the issue through mediation and giving women permission to approach the police only when mediation has failed or when the abuse has continued after mediation. According to research participants, tribal councils, teachers and social workers play a very limited role in domestic violence matters.

When attempting to leave abusive marriages, women face serious constraints. They are bound by the prescriptions that it is their responsibility to make their marriages work by enduring hardship and difficulty, and that they bear responsibility for the abuse by flouting gender rules and not fulfilling their roles. They are concerned about how they will support their children and themselves, as most women are economically dependant on their husbands. Some women fear that their husbands will kill them if they leave.

Chapter 3: Police Perceptions of and Responses to Domestic Violence

3.1 Introduction

This chapter sets out the analysis of semi-structured interviews held with members of the South African Police Service (SAPS) stationed at police stations within the Moretele Traditional Authority areas under study, as well as an interview with a member of the Tshwane Metropolitan Police Service (TMPS) whose area of jurisdiction overlaps with some of the Moretele Traditional Authority areas. The chapter also integrates the focus group discussion on the role of the CPF in domestic violence cases.

The chapter explores the extent to which the police response to domestic violence is influenced by community perceptions of domestic violence. The chapter also explores the police perceptions of violence in intimate relationships, looking specifically at the prevalence of domestic violence in the villages under study, typical domestic violence cases handled by the police officers interviewed, types of relationships in which domestic violence occurs, police perceptions of causes of domestic violence, the nature of disputes underlying domestic violence cases reported to the police, and police opinions of the CPF and other community structures involved in domestic violence cases. The chapter examines the police response to violence in intimate relationships, looking specifically at police understanding of the Domestic Violence Act and procedures to be followed under the Act, referral networks and police collaboration with other role players, problems typically encountered by the police when dealing with domestic violence cases, and the impact of domestic violence cases on the personal lives of the police. A summary of findings and a brief conclusion is set out at the end of the chapter.

3.2 The Duties of the Police under Section 2 of the Domestic Violence Act

The DVA does not criminalise domestic violence. It only criminalises the breach of a protection order. However, certain types of domestic violence constitute recognised crimes, such as common assault, assault with intent do cause grievous bodily harm, attempted murder, damage to property, trespassing, rape, incest, crimen injuria, etc. Where such conduct has been committed, a woman may lay criminal charges in addition to applying for a protection order.

The police are a key role player in relation to the DVA. Out of concern for the casual manner in which the police responded to domestic violence incidents in the past, the legislature made provision for the duties of the police in the DVA. This provision is aimed at ensuring an adequate police response to domestic violence by making it obligatory for the police to assist complainants in domestic violence cases. Section 2 of the Act requires the police to:

In addition to the above duties, the National Instructions issued under the South African Police Service Act require the police to take reasonable steps to gain entry into a building under the following circumstances:

If a police officer fails to comply with his duties under the DVA or the National Instructions, the station commander where the police officer is stationed is required to:

The Area Commissioner is required to forward the record of complaints, hearings and recommendations to the National Commissioner. The National Commissioner is required to submit a report to parliament on a six-monthly basis, setting out:

3.3 Police Perceptions of Domestic Violence in the Moretele District
3.3.1 Frequency of Domestic Violence

According to police officers interviewed, domestic violence is highly prevalent in the Moretele area. Although the police outlined the collection of statistics on domestic violence as one of the roles played by the CPF, no statistics were provided to the research team. When asked about the prevalence of domestic violence in the Moretele area, some of the police officers interviewed said:

It is happening almost every day. The rate is very high. There isn't a day that goes by without us issuing a protection order or warning letter.1

Most of the domestic violence incidents described by the police in the interviews seem to be criminal cases involving serious injury or murder. These included a case in which a man murdered his ex-girlfriend with the help of his friends and secretly buried her; a man who threatened to shoot his family and then kill himself; a boyfriend who shot his girlfriend; a woman who requested a warning letter to be issued against her husband for economic and verbal abuse; and a case of assault with intend to cause grievous bodily harm described in the following statement:

It was a case of SM who was staying here at Majaneng, I opened two cases for her both of assault GBH then she came back and told us that she was withdrawing the case and was forgiving her husband. The last time she came here to open a case because her husband had assaulted her and bit her under her left breast and was admitted in hospital. The fight started when she went to attend the wedding and when she came back her husband demanded to know where she was and she told him that she was from a wedding and her husband started beating her because he said she did not tell him when she left for the wedding. When he bit her she screamed and told their child to boil water so that they can pour over him so that he can stop biting her … So when the child was bringing boiling water something tripped her and she fell that is when the woman grabbed his penis and when he felt the pain, he let go of her and that is when she escaped and ran to the police station. She was admitted to hospital for a couple of days and when I went to visit her in hospital I had no choice but to open a case of assault GBH.

It is unclear from the interviews whether these cases stand out due to their seriousness or whether they truly represent a typical domestic violence case handled by the police in the Moretele Traditional Authority areas. However, these cases are consistent with the community perception that married women should prefer informal family mechanisms to the confrontational criminal justice system and that domestic violence should only be reported to the police where serious injury is sustained or where a woman has been killed.

Domestic violence incidents reported to the police involving a boyfriend and girlfriend seemed to be consistent with the community perception that a boyfriend has no authority to discipline his girlfriend, and that if he beats her up the appropriate step would be to report the incident to the police because resolving the case between the two families would be inappropriate. On the other hand, the seriousness of the domestic violence incidents between boyfriend and girlfriend in these cases would have prompted recourse through the criminal justice system even where the woman's family would have preferred to maintain peaceful relations with the abuser's family.

Only one of the seven incidents related by the police involved a non-physical form of domestic violence. This is not necessarily out of order because non-physical forms of violence do not constitute a criminal offence unless they involve the breach of an existing protection order. However, none of the police offers made reference to incidents of breaches involving non-physical forms of violence. It is therefore unclear whether such a breach is ever reported to the police and, if so, what the police response to it is.

3.3.2 Types of Relationships in which Domestic Violence Happens

According to the respondents, domestic violence mainly occurs between people who are intimately involved. Only one incident of domestic violence between siblings was referred to during the police interviews. Police officers also indicated that in their experience, it is mostly women who lodge domestic violence cases against men, although there have been several incidents of domestic violence perpetrated by women against men.

Those police officers interviewed perceived a substantial number of domestic violence incidents to occur in dating relationships. This seems to correspond with the perception in some focus groups that domestic violence between boyfriend and girlfriend should be reported to the police. According to some of the police officers interviewed, a number of cases reported to them involve cohabitating partners because, as one of the respondents put it, 'once a man and woman have been living together for more than two years, he starts treating her like his wife.'

3.3.3 Police Perceptions of the Causes of Domestic Violence

Police perceptions of the causes of domestic violence echoed many of those provided by the focus group participants. Some of the reasons cited by police respondents as causes of domestic violence included:

One police officer stated that the community accepted a man beating his wife for refusing to have sex with him. Some forms of domestic violence were also tolerated by the community. The gist of the responses was that 'subtle' forms of domestic violence were acceptable to the community, such as slapping, emotional abuse and sexual abuse, particularly where a woman refuses to have sex with her husband. A man beating a woman to the extent that she sustains injuries is not considered reasonable chastisement and is found unacceptable.

Only one police respondent informed the field researcher that no matter what the reason, domestic violence is unacceptable,

No, whatever the reason is they don't like it. They know it is a serious problem that needs attention.
3.4 Police Responses to Violence in Intimate Relationships
3.4.1 Respondents' Understanding of the Domestic Violence Act and the Procedures to be followed in Terms of the Act

Training on domestic violence
Virtually all police officers interviewed have been trained on the provisions of the DVA. Most of them were trained at the Hammanskraal Police College. According to police respondents, the content of the training covered the causes of domestic violence, the rights of victims of domestic violence and the procedure to be followed by the police when receiving a domestic violence complaint in terms of the DVA.

We were taught about domestic violence and what it was and how does it happen and this is how we deal with the victim of domestic violence when she/he comes to the police station. If they come here we analyse the problem and advise the victim to apply for a protection order like if it is emotional abuse or financial abuse. And then if it is physical abuse we would organise a shelter for the victim and then go and arrest the suspect.

The police officers interviewed also demonstrated a basic understanding of the Act. However, it does not seem from the interviews that officers strictly followed the procedures set out in the Act when dealing with domestic violence complaints. During the interviews, none of the police officers made reference to their duties under the DVA. Indeed, is evident from the interviews that the police do not always comply with their duties when responding to domestic violence complaints. Whether action has ever been taken against members of the SAPS in the Moretele area who fail to comply with their duties in terms of the DVA or the National Instructions was not explored during the research.

Receiving complaints
It appears from the interviews that when lodging domestic violence complaints with the police, women are not always advised of their rights in terms of the DVA or the various remedies available to them under the Act. Only one police officer interviewed informed the field researchers that in addition to accepting the criminal complaint, he also gives women the option to apply for a protection order.

The first step that we do is ask the victim if she wants to open the case and if she doesn't want to open the case we then advise her to go and obtain a protection order. Again the victim can open the case and at the same time apply for a protection order. After that we go and arrest the suspect.

The other police officers interviewed for this research did not seem to advise women of their option to apply for a protection order in addition to laying a criminal charge.

Police conception/understanding of their role
While some of the police officers interviewed understood their role as law enforcement officers in domestic violence cases, others saw themselves as mediators attempting to end the violence in a manner that preserves the relationship.

It was an allegation … the child opened a case against his parents because they were always fighting in front of the children so we had to go there and we sat both of them down and we talked to them about what they were doing and how it affected the children.

The police officers interviewed mentioned one case where mediation was used and another that was referred to social workers.

He came to tell us that he has made up his mind and that was what he was going to do and I asked him to give me his wife's contact numbers. Fortunately when I called his wife she was in Room 12 applying for a protection order and I asked her to pass here at the police station. I sat them down here in this office and I asked his wife to tell me what she thinks the problem is and she told me that her husband has been suffering from splitting headaches and every time when he experiences headaches that is when he will start to get out of control and gets very angry and violent. I asked him if he was taking any medication and he said he was and I told him that his headaches are causing him to be violent and he should try to control it and he agreed.
It was a case of a person who came to the police station at Temba … in fact that person was called to come at the Trauma Centre office and L was assisting him, the case was of domestic violence and was very serious. This man came to the police station to report that he was going to get a gun with 14 bullets to kill all his family and himself. So M and I had to calmly talk to him … Yes, I was in fact called by L to come to the Trauma Centre and when we got there we found out later that this man had a serious problem and L had to involve the Social Workers …

One of the police officers interviewed seemed to have the perception that women deliberately pick fights with men because of the