Giving Women their Voice
Giving Women their Voice:
Domestic Violence and Restorative Justice in South Africa
Amanda Dissel & Kindiza Ngubeni
Paper presented at the XIth International Symposium on Victimology, Stellenbosch, July
Amanda Dissel is Manager of the Criminal Justice Programme at the Centre for the Study
of Violence and Reconciliation.
Kindiza Ngubeni is a Senior Community Facilitator in the Criminal Justice Programme at
the Centre for the Study of Violence and Reconciliation.
Restorative justice is becoming an increasingly familiar concept in South Africa. With early
roots in traditional African forms of conflict resolution, more recent applications relate to
dealing with criminal disputes between parties. It has been applied in several community
based projects dealing with criminal disputes, and it forms the basis of well established
practice in child diversion. The concept is also finding its way into legislation through the
Child Justice Bill which is currently before Parliament.
One community based project is that of the Restorative Justice Initiative (RJI),1 an NPO2
that has been implementing restorative justice in three areas in Gauteng and one in the
North West Province for over two years. The project aimed to provide victim offender
conferencing in criminal cases referred by the magistrate’s courts, as well as by the
communities in each area.
The Victim Offender Conferencing model sought to be community-based and applicable
within a South African context. However, it was still derivative of the models developed in
Canada and New Zealand, and mainly premised on the understanding that the crimes
referred would relate to one-off incidents of crime, primarily between strangers or people
not well known to each other. Instead, the project found that in over 90 percent of cases, the
parties were well known to one another, and in over 60 percent of cases the parties were
family members or intimate partners. 33 percent of cases referred to the project were
between people who were intimate partners, or who were divorced or separated from one
another (Dissel, 2000; Dissel, 2003). This raises important questions about the application
of restorative justice, and more particularly of Victim Offender Conferencing in the context
of domestic violence in South Africa. This paper seeks to look at some of these issues,
locating restorative justice within a context of the high prevalence of domestic violence in
South Africa. It aims to look at the inter-relationship between VOC and the objectives of
the Domestic Violence Act. It also looks at the effectiveness of VOC in cases of domestic
violence, particularly in relation to issues of safety, the needs of the victim, and also at the
extent to which VOC assists in preventing further violence. Finally, the paper aims to
highlight some challenges for future interventions of this kind.
Prevalence of domestic violence in South Africa
Statistics on domestic violence are notoriously difficult to obtain. Domestic violence is not
a crime as such, and police statistics reflect only reported crimes, such as assault, rape or
malicious damage to property, etc. There are many actions that constitute domestic violence
that are not defined as crimes, such as forced isolation, verbal abuse, stalking and economic
abuse. Domestic violence is often not just a one-off event, but a series of events, or a
continuous pattern of abuse. Domestic violence may also exist as a continuum with one
serious violent incident at one end, to ongoing and multiple abuse at the other. This pattern
of abuse is not recognised in South African criminal law, which tends to isolate individual
incidents. In addition, most crimes of domestic violence are not reported to the police for
fear of intimation, shame, fear of not being believed, self-blame, or fear of retaliation
(Bollen, Artz, Vetten and Louw, 1999).
That being said, it appears from studies and work by organisations offering support to
victims that the incidence of domestic violence is very high in South Africa. Community
based studies have revealed that in some areas almost a quarter of women report having
being abused in their life times by a current or ex-partner, and that up to a half are affected
by emotional or financial abuse (Jewkes, Penn-Kekana, Levin, Ratsaka and Schreiber,
1999). A survey of 269 women who had contacted helping agencies in three major cities
found emotional or physical abuse to be the most prevalent form of domestic violence
experienced by women, with 90 percent of the sample having experienced this abuse
(Bollen et al, 1999).
The Domestic Violence Act (Act 116) was introduced in 1998 with the aim of affording
women protection from domestic violence by creating obligations on law enforcement
bodies to protect women (victims) as far as is possible. The Act sets out a broad range of
behaviours that constitute domestic violence; including physical, sexual, verbal, emotional
and psychological abuse, stalking, intimidation, harassment, malicious damage to property,
unauthorized access to the complainant’s property, as well as other forms of controlling
behaviour which may cause harm to the safety, health or well being of the complainant. The
Act also extends the notion of ‘domestic relationship’, affording to protection to married
couples; same-sex relationships; couples who are (or were) in a dating, engagement or
customary relationship, including an actual or perceived relationship; any person in an
intimate relationship; parents of a child; and people who do or have recently shared a
The Domestic Violence Act (DVA) allows a complainant to apply to court for a protection
order asking for relief or protection. An interim order may be granted, and after an enquiry,
a final order may be granted against a respondent. The protection order may prohibit the
respondent from committing any act of domestic violence; from entering a specified place;
or from committing an act set out in the order. A warrant of arrest is issued simultaneously
with the order, and brought into effect if the complainant reports that the respondent has
breached any provision of the order. If he or she is found guilty of a contravention, then the
court may convict the respondent and sentence him or her to a fine or a term of
imprisonment. In terms of the Act, no prosecutor is permitted to refuse to institute an
action, or to withdraw a charge unless authorized by the Director of public prosecution.
While the Act criminalizes the breach of an order, it does not create an offence of domestic
violence. Where the offender commits an act that is recognised by the criminal law as a
criminal offence, a woman can report the case to the police, and it may proceed to a
The Act is an indication of the seriousness with which domestic violence is now viewed in
South Africa. It offers one form of protection to women who are the victims of domestic
violence, but it does not necessarily offer remedies to the problem. A criminal trial, or an
alternative process, such as a victim offender conference, or an enquiry by the traditional
authority, may be additional mechanisms that women may turn to for assistance.
Restorative Justice Application in Cases of Domestic Violence
It is fairly recently that restorative justice has been applied to cases of domestic violence
and family violence in South Africa, although its use seems to be better established
internationally. Victim offender mediation and other restorative approaches has been
applied to domestic violence cases in Canada (Edwards and Haslett, 2003), New Zealand
and Australia (Hooper and Busch, 1996) as well as parts of Europe ( Pelikan, 2002)
although its use appears to be fairly limited. There are several possible reasons for the slow
take-up in relation to domestic violence cases. Firstly, Edwards and Haslet talk about the
lack of resources and skilled personnel dedicated to dealing with the complex issues that it
involves. Secondly, they refer to studies which indicate that work in this area has caused
harm to victim or offender. Other writers are cautious of using mediation where there has
been violent abuse (Hossack, in Saskatoon Community Mediation Services, 2001). Some
women’s advocates argue that women are better served by traditional adversarial processes
than by mediation (Hooper and Busch, 1996).
Despite these concerns, there is a need for restorative justice processes, and increasing
support for it as long as some general principles are observed (Edwards and Haslett, 2003).
Two key principles are that mediation should not be attempted in any event unless the
female party, or the victim, is specifically requesting it; secondly only once the safety
concerns have been properly dealt with (Ibid).
Magistrates in the VOC project were particularly supportive of mediation in cases of
domestic violence. Families are often dependent on the income generated by the male
breadwinner, and if he is imprisoned either awaiting trial or afterwards, the family is likely
to suffer from the consequent loss of income. Even if the man is sentenced to a fine, the
woman is ultimately punished. Imprisonment may also result on the loss of a job and result
in economic insecurity, and the stigma of imprisonment may jeopardize future employment
prospects as well. They said that mediation presents the opportunity to look beyond the
crime at its causes. If the offence is linked to alcohol or drug abuse, then the offender can
be referred for counseling. It also enables the parties to come to the table and solve the
problems. It may be particularly effective when used with first time offenders (Dissel,
2003). One magistrate referred to the additional benefits of VOC:
I think that in the process of VOC we have kept the family together, and saved
the marriage, which to me is a more important form of Justice (Van der Merwe,
in Dissel, 2003).
Magistrates believed that mediation should be used for ‘first time’ offenders. If the woman
reported abuse on another occasion, then the case should proceed to trial as usual. However,
in their view, mediation was very successful and they rarely saw repeat cases of domestic
violence, which was more frequent if the case was dealt with by going through trial.
The Victim Offender Conferencing Project
In its second year of the Victim Offender Conference, the community organisations
attached to RJI held 384 conferences in the four sites that it was operating in: Odi
community Law Clinic in Odi; Conquest For Life in Westbury; West Rand Justice Centre in
Dobsonville, Soweto; and Alexandra Community Law Clinic in Alexandra and Wynberg.
Of these, 116 were held between intimate partners, of whom 7 were divorced or separated.
Most of these cases had been referred by the Magistrate’s courts after charges had been laid
by the victim3 (usually a woman) against the offender (usually a male). Cases were also
referred by community structures, the Domestic Violence Unit, and the Traditional
Authority in GaRankuwa. The cases dealt with a range of offences, including charges of
pointing a firearm, assault and assault GBH, malicious damage to property, crimen injuria,
The cases were mediated by trained mediators drawn from the communities serviced by
each site. In addition to being trained on mediation skills and restorative justice, they also
received training and awareness-raising on domestic violence.
The VOC process followed a simple procedure. Once the case was referred to the VOC site
by the court or other referral agency, the mediators would meet separately with both victim
and offender to establish willingness to participate and to prepare for the conference. They
also met with any support people identified by the parties. If the case was accepted, the trial
would be postponed to allow for mediation to occur. Mediation would be arranged at a later
date (although some occurred on the same day as the referral). The mediator would
facilitate a process of story-telling, and discussion of the issues, and facilitate them arriving
at an agreement that would be based on their discussion and the issues raised. The parties
would both sign the agreement, and this, together with a brief report to the mediator would
be handed into court. If the magistrate was satisfied with the arrangements, and if the
victim had agreed, the case would be postponed to allow for the completion of the terms of
the agreement, or would be immediately withdrawn.
A short-term follow-up occurred shortly after the mediation was completed to determine the
extent of compliance with agreements as well as satisfaction on the part of victims and
offenders. Participants to this process indicated a high level of satisfaction and compliance
(Dissel, 2003). However, this process was not an in-depth one, nor did it probe specific
types of cases, perceptions of the mediation, or long term impact of the mediation.
This research paper is concerned with determining the impact of mediation in domestic
violence cases,4 particularly from the viewpoint of the female victim. We were interested in
determining the extent to which the women’s needs had sufficiently been taken into
account, both during the mediation and afterwards, as well as looking at how the mediation
itself was handled. We contacted 21 women who had completed the mediation. In all cases
the offender was a male partner, except in one case where the offender was a woman – in a
same-sex relationship. A 45 minute telephonic interview was held with the women between
6 months and 18 months after the completion of the process.
Although our sample is by no means representative, the findings do shed light on the
experiences of those women, allowing us to understand the short to medium term impact of
the mediation on their lives, and the extent to which it has altered the behaviour of the
abuser towards her. It also gives us some understanding of the complexities of mediation,
and some of the issues that may be important to focus on in further mediations. What
follows is an outline of some of the findings of our study.
Victim Perspectives on Victim Offender Conferencing in Practice
Making the choice to go through mediation
Most of the cases had been referred to mediation by the courts. The parties were briefed
about the process and could decide whether they wanted to persist. Most of the women in
our sample had reported the case to the police. Their reasons for doing so differed. Some of
them wanted to send a warning to their man to show that they were serious and that police
action could be taken against them. On the other hand, there were women who wanted their
man to be sent away, or jailed for their protection. However, they were willing to go ahead
One woman was referred to VOC directly by a friend who had been assisted by the VOC
project. She said that her friend had been helped in the process, and her marriage had
changed as a result. So, she went into the VOC process knowing that she wished to
participate. Another woman was referred to VOC by a community based women’s
organisation. One woman had reported the case to the police and it had been referred to
court. She was dissatisfied that the trial was taking so long, mainly due to numerous
postponements, and when a friend advised her about VOC, she approached the Law Clinic
for help. She was satisfied that they were able to deal with her problem speedily so that she
could get on with her life.
When asked what the women would do if the abuser repeated the abusive behaviour, again
their responses differed. Some of the women said that they would go back to the VOC
project, while others indicated that they would take stronger action and report him to the
Dealing with the broad picture of abuse
Often harm caused and experienced in criminal cases is as a result of a particular incident.
However, in family/domestic violence, the harm caused is often the result of a pattern of
abuse or violence. In exploring the history of the incident leading to the offence, it is
important to look at the pattern of abuse that has occurred within the relationship, whether
there is ongoing abuse, and how this contributes to an understanding of the nature and
extent of violence in the relationship (Edwards and Haslett , 2003). This could assist the
participants to understand the cause of the violence, and should contribute to the offender
taking responsibility for his perpetration of abuse.
Hooper and Busch point out that one of the dangers of a conferencing model is that it
requires the participants, and especially the family or community, to ‘understand’ and
‘explain’ the causes of violence. Some of the explanations, they warn, attribute the
responsibility wholly or in part to the victim, or they may see a certain level of violence as
acceptable in a conflict situation. However, more of a concern, they argue is the potential of
explaining the violence as a problem in the relationship between the partners. If the cause
of the violence is identified as a problem in the relationship, then the solution is likely to
focus on relationship dynamics, rather than addressing the man’s problem of violent
In order to understand how the violence was understood in the mediation, the women were
asked why they believed their partners abused them. The majority of them (11) indicated
that alcohol or drugs was a factor in the abuse, and most said that their man became violent
when he was drunk. Others indicated that the man drank a lot and did not want to spend
money on household expenses. The respondents also mentioned that the abuser was
adversely influenced by his bad friends (4 respondents). Other women said that their men
were violent people, or short tempered, indicating a behavioral problem. They also
mentioned factors relating to their partners controlling behaviour and jealousy. Women also
felt that their man blamed them for certain things. One woman said that her husband
blamed her for not conceiving, and therefore felt entitled to sleep with other women who
might give him a child. One woman said that she did not count in the relationship, that her
husband despised her and thought she was ‘revolting’. Two women attributed the abuse to
particular circumstantial factors. In most cases then, the women understood that the
violence occurred as a result of actions, behaviour and attitudes of the man, and not as a
symptom of problems in their relationships. However, as will be seen in a later section of
the paper, the agreements reached were often indicative of an understanding of mutual
responsibility for the violence, as well as for solutions to it.
This indicates the range of issues that are implicit in a case of domestic violence, and points
to some of the factors that need to be dealt with in the mediation.
Safety of participants
The safety of the participants is another key concern. Harm has already occurred by the
time the case is referred to the police, but there is a possibility of further harm occurring to
the woman or members of her family during the mediation, or afterwards when she
(usually) returns to the home with her partner.
Not all women interviewed were concerned about their safety, as they had not felt that their
lives were in danger from the offender, alternatively they felt that they were able to fight
back against him. One woman was concerned for her safety of her health, and was fearful
that she may be infected with HIV/Aids due to her husband’s infidelity. For others though,
this was a life threatening issue. Some women felt that there safety was guaranteed by a
protection order that had been issue before the mediation; while others had experience that
this did not prevent further assaults or threats.
Most respondents felt that the mediation provided a safe space where their personal safety
was not threatened. The mediators were careful to assure the women that they were safe
during the mediation. In one case where the man had been charged with pointing a firearm
at the woman, the mediator ensured that the firearm was not allowed into the mediation.
Although this may have been effective in the short term, on the longer term this action may
not be sufficient. Since the criminal case was withdrawn from court, the usual enquiry into
the offender’s fitness to carry a firearm was not done, and the offender could continue to
carry his firearm (if indeed he had a license for it), possibly endangering her or others at a
One woman said that she thought that she would be safe during the mediation, and that
afterwards the offender would feel himself bound by the agreement.
One woman was particularly concerned about her safety. Her husband had hit her with an
iron on her head, and she felt sure that he would kill her one day. She reported him to the
police as she wanted him arrested. He was arrested, but later released and the case was
referred to VOC. The woman was initially reluctant to participate in VOC, but after being
assured that if it didn’t work she could refer the case back to court, she agreed. She was also
fearful of the community organisation that was implementing VOC because of the large
number of men there, but she was reassured by the male mediators assigned to the case.
The offender threatened her on a number of times during the mediation and had to be
warned by the mediator. The woman felt reassured sufficiently by this. She was also not
intending to return to the home, and thus the mediation provided a safe space to negotiate
the terms of their separation.
It is important to safeguard the participants, usually the women, from further abuse during
the intervention period, as well as to ensure her safety afterwards. The mediators would
advise the parties that should abuse continue, the woman had the right to report the case to
the police, to request a Protection Order, or to have the case proceed to trial. This seems to
have been sufficient in the cases referred to VOC, as all respondents indicated that the
offender had not abused them since the mediation. However, it may be necessary in other
cases to look at linking the woman up with resources in the community to ensure her
protection. In one case where the parties agreed to separate, the mediator accompanied the
woman home when the offender collected his belongings from the home to ensure that he
did not abuse her in that time.
Providing an opportunity to speak ones mind
One of the central tenants of the restorative justice philosophy is creating the opportunity
for the person affected by the violence to have an opportunity to engage in dialogue and to
explain how it impacted on them. Having a safe space to tell their story, be heard and
validated can be a powerful experience, especially when this is combined with hearing the
offender take responsibility for his actions (Edwards and Haslett, 2003).
However, the context of a domestic violence relationship may place restraints on the
woman’s ability to effectively participate in this process. Hooper and Busch argue that
domestic violence relationships are often characterized by power imbalances and dynamics
of control that suggest that the victims of the violence are not able to negotiate freely and
fairly with their abusers. The concern is that this trend which exists in the home might play
itself out in the mediation. A woman may have learned that in order to avoid further
violence, she has to be compliant with her partners own wishes and subvert her own needs
and desires. In a mediation situation, they argue that this may result in a woman settling for
what she thinks she can get, rather than for what she actually wants (Hooper and Busch,
1996). They add that perceived risks of harm may diminish the battered woman’s ability to
participate effectively and assertively in the mediation process.
Unsurprisingly then, one of the most important issues raised by the women was the
opportunity that VOC gave them to speak on an equal basis with their partners. Most often,
the women complained that their men did not give them the opportunity to speak at home.
VOC has helped us to talk to one another. He used to beat me up every time
that I tried to express my feelings before.
In contrast, they felt that the mediation provided a safe space to speak. Whereas in the
home women were silenced by the threat of violence should they speak, they were
protected by the mediator and the rules provided in the mediation.
The mediation enabled the women to tell their version of their story for perhaps the first
time. But they were also able to talk about how the actions of the abuser affected them
personally and emotionally. Not only was this liberating for the women, but it also appeared
to be the first time that some men actually listened to what the women were saying. One
said, “I was happy to be in mediation. I saw it as an opportunity to speak up against him. I
wanted him to know how I feel about the beating”. She said that she wanted an opportunity
to speak to her husband because at times he misunderstood her.
Another woman said that VOC provided her with an opportunity to have a say in her future,
indicating that for her the power dynamics had equalized to some extent. For another
woman it was important to be able to state what she wanted from her husband, which was
to separate, and the terms of the separation.
The respondents reported that it was important when the abuser acknowledged the impact
of his actions on the woman. One woman said that during the mediation it was the first time
that the offender realized how his actions had hurt the woman.
It is important to build ways of shifting the balance of power during the mediation. The
separate pre-mediation meeting with the victim was a crucial part of this aspect. Although
in many cases this meeting was short, it enabled the mediator to determine the woman’s
interest and willingness to pursue mediation, allowed her to tell her side of the story to an
impartial third party, as well as allowing her to outline any concerns she may have about
the process, or about her safety. It also provided the opportunity for the mediator to explain
her legal rights and her right to continue with the criminal prosecution.
Dealing with the dispute in private, versus involving family members
One of the rationales of many restorative justice processes is that it opens the possibility for
the involvement of other people central to the parties’ lives, who can help in solving the
dispute, as well as acting as witnesses in the agreement to make amends. The Victim
Offender Conferencing model allowed for the participation of people supportive to either or
both of the parties. An advantage of this approach is that it would remove some of the
secrecy usually associated with domestic violence, and that it would also involve the family
in the process of dealing with the problem. However, in the VOC project very few parties
actually took advantage of this, and most opted to proceed with only those people directly
involved in the dispute.5
Other people were seldom involved in the conference of the women we followed up. Most
of the respondents indicated that they were happy to have the matter resolved ‘in private’
without other people witnessing the process. They also preferred not to have other people
‘interfering’ in their domestic lives. Several women reported that they had experienced their
families as biased, critical of them, and more supportive towards the man. This was
particularly pronounced where women were criticized by their families (and usually by the
man’s family) of not acting in a sufficiently subservient manner as required by their culture.
One woman reflected her need to pacify the man by indicating that she did not request
support because she didn’t want her husband to think she and others were ganging up on
On the other hand, some respondents appreciated the support of family or friends. In one
case, the victim felt safe and supported in the mediation because her mother and sister were
present with her.
Addressing the problem: Behavioural changes
Restorative justice processes should have an open agenda with no predetermined outcome,
and no pressure on the parties to reconcile. However there was an expectation from the
courts that the parties would arrive at an agreement that would assist the court to decide on
whether or not to proceed with the criminal trial. Although there was some expectation that
the parties would arrive at an agreement,6 this did not mean that they had to reconcile with
one another, or even agree to stay together. The terms of the agreement would be
determined by the needs and interests of the parties themselves.
The biggest test on the efficacy of mediation is to determine the extent to which there has
been real behavioural change in the offender, and to what extent this has impacted on the
relationship. Relationships are complex entities, with many different subtleties defining
how the parties interact with one another. It is often difficult to quantity the extent of
change. It is also difficult to ascertain the success of an intervention of this sort, because
even if the man had not abused the women since the conclusion of the mediation one can
never be sure that his behaviour has changed for good. He may still continue with his
abusive behaviour at some later stage. However, an average of one year had lapsed by the
time that we interviewed women again. We asked the respondents how the abuser’s
behaviour had changed leaving it to them to identify any changes. In all cases, they
indicated that change had been positive, and that overall there was an improvement in his
behaviour and conduct towards them. Of those who had been assaulted or physically
abused by their partner, all reported that there had been no further assaults since the
mediation. It was also reported that the offender was no longer abusing the child of the
All the women who were still in their relationship indicated that the relationships had
improved following VOC. Women talked of the mediation having saved the marriage, or in
a few cases it led to marriage, or the parties moving in together. Of course, in some
relationships, the VOC helped to facilitate a separation from each other, setting out the
terms of their separation.
In most cases, respondents said that their communication had improved, and that their man
was able to listen to them. Others said that their husband talked to them more now. One
woman said that whereas in the past her husband used to come home and read the
newspaper or watch TV, he now talked to her. She indicated that “the mediation helped us
because I realized that the things that made him behave the way they did, and that I also
had to change myself.” This comment indicates that the woman took some responsibility
for the problems in the relationship, and for dealing with them. Sometimes the agreements
reflected a joint responsibility to change behaviour, usually reflecting an undertaking to
respect one another, or to talk to one another when they had problems.
Where verbal abuse had been a problem, the respondents reported that this had stopped.
The full extent of the agreement was not always fully adhered to, but this did not appear to
affect the perceptions of the success of the mediation. Sometimes the parties had agreed
between themselves to amend the terms of the agreement. In one case where the parties had
agreed to separate, and agreed on a division of property, the woman reported that a month
afterwards the offender came to her house and removed property that was supposed to stay
with her. She reported this to the Odi Community Law Clinic who insisted the man return
the household items. They threatened him with legal proceedings should he refuse to do so,
and they advised the woman to obtain a protection order to prevent him removing goods
from the house again. It is apparent from this that not only do many men consider the
agreement to be binding, but that when combined with a threat of re-initiating the criminal
action or instigating an inquiry for a protection order, he is more likely to abide by it. In one
case however, the agreement had lasted for a period of six months, but when the husband
again started to have affairs, the parties divorced.
Many of the women indicated that substance abuse was a factor in the abusive behaviour.
However, although the agreements often referred to the man giving up or reducing his use
of drugs or alcohol, this very seldom resulted in any real change. In one case the abuser
undertook to attend alcohol counseling, and did so for a sort while, but soon gave up. In
most cases he still continued to drink, although it was often reported that his consumption
levels had reduced, or his drinking patterns had changed. One man continued to drink, but
no longer went out with his friends and came home drunk. One man continued to drink, but
gave up the drugs. But despite the continued consumption of alcohol, their abusive
Following the mediation, women reported that the man had abided by the agreement in
contributing to the maintenance of the household and for the children. One woman reported
that her husband had found a job. In another case, the husband contributed to the household
by performing household tasks, such as cleaning and cooking.
Five of the cases in our study resulted in the parties agreeing to separate. The mediation
assisted them to make that decision, and helped the parties to negotiate the terms of the
separation, in particular, the division of property. In the majority of these cases the
separation had been amicable and the offender had not harassed or threatened the victim,
and had complied with the terms of the agreement. In one case, the woman’s partner was
killed by gang members before he could fulfill the whole agreement.
The role of culture
The VOC project sought to be culturally relevant to the African context. In the rural areas,
particularly in those areas serviced by the Odi Community Law Clinic, many of the
participants came from very traditional backgrounds still heavily influenced by their
African culture. Although living in more temporary accommodation in settlements in this
area, they were often recent arrivals from more rural areas. They also often had recourse to
the traditional authority, which is charged with assisting in the resolution of disputes. The
traditional authority would refer cases to VOC for mediation. It was interesting for us to
look at how the women perceived the VOC mediation in comparison with the traditional
structures, and also how culture affected the outcome of the case.
The four women in our study who had been referred to the Odi site did not support the
intervention of the traditional authority. One woman refused to go to the authority since her
partner was known by them and she feared that they would not be impartial. She reported
the case to the police who referred it onto VOC. In another case, the woman complainant
was referred to the traditional authority where she complained that even though her
complaint was about her husband, he spoke for her on her behalf. In contrast, the women
were satisfied that the VOC mediator were impartial, unbiased and helpful to her cause.
In two of these cases, the women indicated that their partner’s behaviour in the relationship
- the sleeping with other women, verbal and physical abuse, and unreasonable demands
placed on her – were seen as acceptable conduct for an African man in their culture. The
men had difficulty in accepting the concept of equal power, respect for one another, and the
need to allow the women to speak in the mediation. One man had felt compelled to accept
one of the terms of the conditions, and had even challenged this with the traditional
authority afterwards. They had advised him that it was an agreement that he had entered
into with his partner and that they could not change it. Although he had been unhappy, he
had continued to abide by the agreement. Another man who also felt so compelled had
negotiated afterwards with his partner to change some of the terms of the agreement.
Despite this apparent reluctance, the women reported that their partners had become
changed men in the relationships. The men had accepted the responsibility and were no
longer abusing them. In two cases, the women reported that their mother-in-laws were
unhappy with the apparent subservience of the man after the VOC, accusing the woman of
bewitching her partner, or of using korobela (a herbal medicine given by a woman to make
her husband soft and compliant, or to subdue him).
From these cases it appears that VOC is effective in cases where strong tradition exists, but
that it challenges the culture and can be instrumental in introducing new behaviours and
This small scale study illustrated that there was a high level of satisfaction with the VOC
process among the female victims of domestic violence. For most of the women, it afforded
a unique opportunity to make their voices heard, to tell their story, and to insist on changes
in their partners’ behaviour. The experiences of these women indicate that restorative justice
practices can be successfully used in domestic violence cases, and that it can result in
lasting and meaningful change. It also appears to be relevant and understood within African
culture. The study does not shed light on the abuser’s behaviour in respect of other acts of
criminality or violence towards other people.
This does not mean that domestic violence cases should be dealt with lightly. Before cases
are referred to mediation, there should be preparation with the victim so that her decision to
participate is voluntarily made based on a sound understanding of the process and its
consequences. In all cases there should be a full assessment of her safety concerns, and
steps taken to ensure that she and her family are protected. Where firearms are used or
threatened in domestic violence, it is our view that an inquiry should still be held as to
whether the offender is a fit and proper person to have a license. If the charge involves an
unlicensed firearm, it goes without saying that criminal proceedings should continue
Proper training is crucial to ensure that mediations are properly conducted. It is also
important that mediators are impartial and do not buy into the cultural perspectives on
women that are so prevalent in our society. Finally, we believe that it is important to
institute a process of monitoring the case for a period after the mediation. This is to ensure
that the agreement has been complied with and that no further abusive behaviour has
occurred. It is also important to stress to the woman that the organisation is available to
assist the woman after the mediation, or to make the necessary referrals for further
1 The RJI is a consortium body consisting of the following organisations: the Centre for the
Study of Violence and Reconciliation; Restorative Justice Centre; Mennonite Central
Committee; West Rand Justice Centre; Conquest for Life; Alexandra Community Law
Clinic; and the Odi Community Law Clinic.
2 Non Profit Organisation.
3 Note on the terminology used. During the victim offender conference the complainant
was referred to as the ‘victim’ and the accused as the ‘offender’. These terms are used here.
The offender is also sometimes referred to as the ‘abuser’. We are aware of the negative
connotations that are associated with the term ‘victim’ but it is also used to distinguish the
target of the abuse from the abuser.
4 We used the broader definition of domestic violence, and did not only look at cases
involving physical violence.
5 Support persons were present in 27 percent of victim offender conferences (Dissel, 2003).
6 In the VOC project, 91 percent of cases mediated resulted in an agreement (Ibid)
Bollen, S., Artz, L., Vetten, L., Louw, A. (1999). Violence Against Women in Metropolitan
South Africa: A study on impact and service delivery. Institute for Security Studies
Monograph Series No 41, September 1999. Also available at:
Dissel, A. (2000). Restoring the Harmony: A report on a victim offender conferencing
project. Johannesburg: Centre for the Study of Violence and Reconciliation and Restorative
Justice Initiative. Also available at: http://www.csvr.org.za/papers/papvoc.htm.
Dissel, A. (2003). Giving a Face to Crime: a report on phase II of the RJI (VOC) project.
Johannesburg: Centre for the Study of Violence and Reconciliation and Restorative Justice
Initiative. Also available at: http://www.csvr.org.za/papers/papdis11.htm.
Edwards, A. and Haslett, J. (2003) Domestic Violence and Restorative Justice: Advancing
the Dialogue, presented at the 6th International Conference on Restorative Justice.Available
Hooper, S. and Busch, R. (1996) Domestic Violence and Restorative Justice Initiatives: the
Risks of New Panacea. Available at
Jewkes, R., Penn-Kekana, L., Levin, J., Ratsaka, M., Schreiber, M. (1999). “He must give
me money, he mustn’t beat me”: Violence against women in three South African provinces.
Pretoria: CERSA (Women’s Health) Medical Research Council.
Pelikan, C. (2002) “Victim Offender Mediation in Domestic Violence Cases – a comparison
of the effects of Criminal Law Interventions: the Penal Process and mediation. Doing
Qualitative Research.” Forum Qualitative Social Research 3(1) 2002. also available at
Saskatoon Community Mediation Services (2001) Report on the Consultation on
Restorative Justice and Violence against Women, Saskatoon.