VICTIMS ’ EXPERIENCES OF RESTORATIVE MEDIATION : A DEVELOPMENTAL SOCIAL WORK PERSPECTIVE

INTRODUCTION Victim-offender mediation is an expression of restorative justice. The wellbeing of victims of crime is a central feature of restorative justice services. The process of restorative mediation entails all parties, i.e. the victim, the offender, their families and members of the community, voluntarily participating in face-to-face dialogue, where truth-telling enables the offender to take personal responsibility for his/her criminal behaviour (Rainford, 2010). An important outcome of restorative mediation is reaching an agreement regarding the strategies to remedy the wrongdoing. A developmental social work perspective counteracts retributive justice in shifting the intervention focus from individual defect and blame to reform and social change (Midgley, 2010). Developmental social work is embedded in investment strategies in restoring people to full functioning in society (Rainford, 2010). In fact, “social investment, economic participation, empowerment and human investment are relevant to all systems and forms of social work intervention” (Midgley, 2010:12).


INTRODUCTION
Victim-offender mediation is an expression of restorative justice.The wellbeing of victims of crime is a central feature of restorative justice services.The process of restorative mediation entails all parties, i.e. the victim, the offender, their families and members of the community, voluntarily participating in face-to-face dialogue, where truth-telling enables the offender to take personal responsibility for his/her criminal behaviour (Rainford, 2010).An important outcome of restorative mediation is reaching an agreement regarding the strategies to remedy the wrongdoing.A developmental social work perspective counteracts retributive justice in shifting the intervention focus from individual defect and blame to reform and social change (Midgley, 2010).Developmental social work is embedded in investment strategies in restoring people to full functioning in society (Rainford, 2010).In fact, "social investment, economic participation, empowerment and human investment are relevant to all systems and forms of social work intervention" (Midgley, 2010:12).
Evidence from abroad shows promising results in reducing rates of recidivism, even though the main part of restorative practice is healing relationships rather than preventing future crime (Rainford, 2010).Despite widespread policy and legislative support for restorative justice in South Africa, little empirical evidence is available about the experiences of beneficiaries of restorative practices.The aim of this article is twofold.Firstly, it articulates the experiences of victims with restorative mediation insofar as referral, preparation, meeting the offender and outcomes of victim-offender mediation are concerned.Secondly, the results are contextualised within a developmental social work perspective to identify strengths and shortfalls in the comprehensive delivery of victim-offender mediation services.Therefore, the questions we answer in this article are: How do victims of crime experience procedures of restorative mediation?What are the outcomes of mediation for victims?Do mediation practices further the ideals of developmental social work?

Conceptualising restorative justice
Restorative justice represents an important departure from the dominant goals of justice that have been followed over the past two centuries.Conventional systems of jurisprudence are essentially adversarial in nature.They are preoccupied with defence, where justice is structured as a contest between the state and the offender (Hargovan, 2007;Zehr, 1990).The passing of judgement implies that there is a winner and a looser (Pule, 2002;Schneider, 2000).Retributive justice believes that delivering pain, stigma and shame through severe punishment will somehow vindicate victims (Fivaz, 2002;Jacobs-du Preez, 2002).In retributive systems offenders do not participate in developing their treatment plans or strategies to prevent re-offending.Instead, these decisions are made by justice officials and human service professionals (Carrillo & Carter, 2001).Very little scope, if any, exists for personal apologies and reconciliation.It is assumed that retributive justice does not necessarily create an understanding in offenders of the human impact of their criminal behaviour.This hampers awareness about, and an internalised shift in attitudes away from, offending (Morris, 2002;Skelton, 2002).
Conventional justice largely ignored victims, since their interests are represented by either the state or legal practitioners (Hargovan, 2007;Mousourakis, 2004).Crime of any kind can have an immense impact on victims.Not only does victimisation often imply financial costs, but it leaves the victim with many unanswered questions (Zehr, 1990).Yet victims mostly become witnesses in the unfolding of justice, since most emphasis of legal proceedings is placed on the offender (Coates, Umbreit & Vo s, 2006).Also, victims generally have little direct say in the sentencing of offenders (Muntingh & Monaheng, 1998).As mentioned, a sense of closure on the victim's part depends on retributive principles.Under these circumstances, little empowerment of victims takes place (Dzur & Olson, 2004;Pule, 2002).
With the above in mind, restorative justice assumes that an offence causes social, emotional, physical and/or financial harm to victims, offenders, their families and the community.Therefore, it is considered that damaged relationships must be healed for the affected parties to move forward (Hargovan, 2010).In terms of the philosophy of restorative justice, this depends primarily on remorse by the offender and forgiveness by the victim (Shearar, 2005).

Development of restorative justice
Mediation has a rich cultural heritage and draws upon various indigenous problemsolving techniques (Baffour, 2006).Its formalisation in restorative justice in the 1980s stems from the traditional conflict-resolution techniques of the Maori in New Zealand and in Indian communities in North America (Zehr, 2004).Restorative justice was also backed by international moves toward the de-institutionalisation of punishment and empowerment of victims (Umbreit, Coates & Vos, 2007).
Elements of restorative justice have been part of African customs for hundreds of years.Traditional social control mechanisms were employed to address stress, anxiety and tension (Ovens & Prinsloo, 2009).Unlike modern legal systems, which are based o n individual responsibility, the collective was held responsible for a member's actions.The Igbo of Nigeria and the Songhai Empire of West Africa are revered as examples of indigenous social control and justice strategies that focus on forgiveness, communalism, healing and restitution instead of retribution (Agozino, 2005;Okafo, 2006).Western legal procedures attempt to obtain the truth in fact and in law, and is therefore called factual or forensic truth.Indigenous legal procedures take a broader understanding of truth into account, including personal and narrative truth, social or dialogue truth, and healing or restorative truth.A main task of legal proceedings is to ascertain that reconciliation has taken place.African legal procedures aim at resolving a matter appropriately and good-naturedly, reconciling the parties and coming to a decision that can be accepted by all.Apparatuses of justice are aimed primarily at peacemaking rather than allocating rights between disputants (Taylor, 2007).
Traditional models of conflict resolution in Africa included meetings by elders to deliberate matters such as war and domestic problems (Skelton & Frank, 2001).Prior to colonisation, African societies did not have prisons.Although this does not deny the existence of crime, it implies that people followed other mechanisms to resolve disputes, while still enabling the normal co-existence of the parties involved (Makhathini, 1996).Typically, if someone offended, the reputation and dignity of the whole family were at stake.They often had to compensate for the criminal act or risk having to leave the village.These customary norms ensured a high level of social control and responsibility for each member of society (Muntingh & Monaheng, 1998).
Another driving force behind the adoption of restorative justice in South Africa was the revitalisation of ubuntu, which is the principle of caring for each other's wellbeing with an attitude of mutual support (Inter-Ministerial Community on Young People at Risk, 1996).Mkhize (2004:50) defines ubuntu as "a person's knowledge of his or her duties and responsibilities within a community of other, interdependent human beings".In essence, it means that a person is a person through other people.Based on the principles of honesty and forgiveness, the Truth and Reconciliation Commission of the 1990s exposed the South African public to restorative practices, particularly by giving the victims of political crimes a voice while demanding accountability from perpetrators (Mousourakis, 2004).

Theoretical framework and intervention methods
The foundation of restorative mediation draws from belonging, systems and humanistic schools of thought.Building and maintaining relationships are human characteristics.
Most of what people do takes place within a group context.Although it is possible to live on the fringes of groups, social isolation and loneliness are regarded as problems in modern society (Gray, 2010).Attachment to others and a sense of belonging provide the structure through which personal relationships are entered into and maintained (Toseland & Rivas, 2009).In addition to social bonds, individuals constantly interact with multiple social systems such as the workplace, the community, friendship networks, and religious and recreational settings (Gray, 2010;Rose, 1998).Groups do not exist in a vacuum but make up the parts of a broader system that legitimises and influences their purpose (Toseland & Rivas, 2009).With this in mind, the ecological perspective appreciates the meaningful structures within a person's life and how he or she functions within that environment (Bronfenbrenner, 1977(Bronfenbrenner, , 1994)).It explains interrelationships within the system, which holds specific value for the assessment of problems and the level or sub-system which requires attention (Gray, 2010).It also provides for recapitulation, which Toseland and Rivas (2009) explain as the opportunity to address unsatisfactory relationships among family members or other affiliations.From a therapeutic framework, the strengths of, and existing capabilities within, the system play a pivotal role in promoting an individual's effective functioning.The key concepts of systems theory are (Zastrow, 2009):  wholeness, which implies that no system can be understood holistically once it has been reduced to its component parts;  relationship, which rejects simple cause-effect explanations, since the interactions between structures are as important as the elements within those structures;  homeostasis, as any system seeks a balance to maintain and preserve itself.Disequilibrium distorts harmony within the system.
From the above, it is evident that theorising about restorative mediation represents a shift from individual-oriented theories to those of collective existence.This has important implications for contemporary South Africa, given its diverse population and traditional notions of collectiveness, including ubuntu (Gray, 2010).A focal point in present-day South Africa revolves around mutual understanding and respect for fellow human beings, regardless of ethnic origin and personal preferences.From a humanistic approach, the values underlying the very nature of human existence, conflict and the search for healing include (Umbreit, 1997):  belief in the connectedness of a common humanity and the desire of most people to live peacefully and to grow through life experiences;  the capacity of all people to draw upon inner reserves to overcome adversity and to assist others in similar circumstances;  the inherent dignity and self-determination that arise from embracing conflict through dialogue and mutual aid.In the light of the underpinnings of restorative mediation, it is evident that mechanisms are needed to establish meaningful dialogue among the parties affected by the offence.Group work as intervention method meets this requirement.It is defined as goal-directed activities with two or more people aimed at achieving socio-emotional goals and accomplishing tasks (Toseland & Rivas, 2009).Activities take place within a system of service delivery and set out to reduce or eliminate blockages to socially desirable behaviour.In group work individuals remain the focus of concern, with the group serving as vehicle for growth and change (Farley, Smith & Boyle, 2003).Group work generally pursues the goals of improving social functioning, promoting feelings of belonging with others, stimulating positive personality traits, and assisting in solving problems (Strydom & Strydom, 2010).Group work consists primarily of three phases, the first of which entails the careful preparation of each participant, through a series of interviews, regarding the purpose of the intervention.The process phase represents the interaction among the participants where they are guided toward the best solution to the matter at hand.An important outcome of the process phase is the compilation of a contract which binds an individual to the obligations determined by the group (Farley et al., 2003).The final phase revolves around follow-up and mechanisms to monitor the implementation of the contract.

Evidence on the outcomes of restorative mediation
A search for published evidence about the effectiveness of restorative procedures showed the South African cupboard to be fairly empty.Two research initiatives are noteworthy, given their measureable assessment of restorative experiences and outcomes.In 2002 the Centre for the Study of Violence and Reconciliation (Dissel, 2002) investigated the outcomes of 224 cases which went through restorative intervention in Gauteng.Common assault (42%), assault with intent to inflict grievous bodily harm (31%) and damage to property (11%) made up the bulk of cases.The offenders apologised for the wrongdoing in 64% of the cases, while charges were withdrawn in 70% of cases.A third (33%) of cases entailed direct compensation to the victim.The average duration of mediation was two hours and twenty-five minutes.Research in KwaZulu-Natal (Hargovan, 2009) found that, following restorative intervention, 81% of victims reported changes in the behaviour of offenders.The cases largely amounted to crimen injuria (33%), common assault (23%) and assault with intent to inflict grievous bodily harm (10%).The majority of victims (83%) were of the view that justice has been done and 76% of cases were withdrawn.Most of the offenders (90%) appreciated and acknowledged the harm they caused by their criminal behaviour.In the light of local evidence, therefore, restorative practices appear to yield favourable outcomes.Still, it is not clear which types of restorative intervention, as well as which profiles of victims and offenders, are more prone to positive programme outcomes in South Africa.
Investigations from other countries, especially the United States, Australia and New Zealand, provide further indications of restorative outcomes and experiences.However, these studies are often marred by small sample sizes, lack of representation of study populations, absence of experimental controls, and non-comparable definitions of recidivism.In addition, many studies focus on process measures, in particular experiences of fairness and restoration, without considering outcomes such as recidivism (Baffour, 2006;Bergseth & Bouffard, 2007;Bonta, Wallace-Capretta, Rooney & McAnoy, 2002;McCold & Wachtel, 2000).It is also important to keep in mind that results are influenced by self-selection bias, since programmes are voluntary and participants may be more motivated to ensure positive outcomes (Hudson, 2002;Latimer, Dowden & Muise, 2005).
The greater part of investigations, including three meta-analyses, suggests that about a third (28-34%) of offenders recidivate following participation in restorative programmes; in experimental studies the majority of offenders were less likely to reoffend when compared to control groups (Bergseth & Bouffard, 2007;Bonta, Jesseman, Rugge & Cormier, 2006;Bonta et al., 2002;Bradshaw et al., 2006;Latimer et al., 2005;Rodriguez, 2007).In general, victims and offenders express high levels of satisfaction with the mediation process (Bonta et al., 2006;Bradshaw et al., 2006;McCold & Wachtel, 2000;Mutter & Dugmore, 2008;Umbreit & Fercello, 1997;Wemmers & Cyr, 2005).Re-offences also tend to be less serious compared to those committed by nonrestorative groups (Nugent & Paddock, 1995).On the negative side, restorative intervention appears to have little impact on the recidivist behaviour of higher-risk offenders (Bonta et al., 2006).An offender's previous contact with the police serves as an important predictor of poorer restorative outcomes (Baffour, 2006;Bergseth & Bouffard, 2007;De Beus & Rodriguez, 2007;Rodriguez, 2007).Female offenders demonstrate better outcomes than males (Baffour, 2006;De Beus & Rodriguez, 2007;Rodriguez, 2007).Also, longer periods of follow-up result in lower re-offending rates (Bradshaw et al., 2006).Younger participants show improved outcomes for restorative programming than older ones (De Beus & Rodriguez, 2007).A study involving 2 428 cases that went through restorative procedures found that participation in mediation is less likely for cases involving personal offences than for those involving property offences.Also, as time passes, the probability of participation increased for personal offences but decreased for property offences.The study further shows that some victims decide along racial and ethnic lines whether they want to participate in mediation (Wyrick & Costanzo, 1999).

Developmental social work
Developmental social work is practised within a developmental social welfare system (Gray, 2006).The White Paper for Social Welfare (RSA, 1997) proposes a developmental social welfare approach to social welfare in South Africa.This approach embraces a welfare system that is more just, equitable, participatory and appropriate in meeting the needs of all South Africans (Patel, 2005).The Department of Justice and Constitutional Development ( 2011) stipulates that restorative justice processes must comply with the rule of law, human rights principles and the rights provided in the South African Constitution.
Developmental social welfare is entrenched in the Bill of Rights of the South African Constitution, 1996, which enshrines the rights of all people in the country and affirms the democratic values of human dignity, equality and freedom (Lombard, 2008).Developmental social work affirms the commitment of the social work profession to social justice and human rights, and to the eradication of poverty and inequality (Lombard & Wairire, 2010).This commitment reflects a shift from the remedial and "maintenance" to the social change function of social work which is central to developmental social work (Midgley, 2010).The shift emphasises one of the key themes of developmental social work, namely bridging of the divide between micro and macro practice (Patel, 2005).Direct, micro practice from a developmental perspective includes personal growth and individual development, whilst macro social work's emphasis is on collective improvements to effect change in communities and policy in order to produce tangible improvements in standards of living (Midgley, 2010).
As mentioned earlier, restorative justice assumes that an offence causes social, emotional, physical and financial harm to victims, offenders, their families and the community.Acceptance of responsibility for the wrong-doing and forgiveness is the beginning of healing of relationships.Social workers understand that relationships between and among people are an important vehicle for change and seek to strengthen relationships among people in a purposeful attempt to promote, restore, maintain, and enhance the wellbeing of individuals, families, social groups, organisations and communities (United Nations, 1994).
Developmental social work maintains that the main goal of restorative justice, namely healing of relationships (Rainford, 2010), can only be achieved within a broader social development context which requires investment strategies.Social investment strategies in the correctional context include crime prevention, diversion strategies, drug treatment and counselling of offenders, with the emphasis on more successful re-entry of prisoners into community living (Rainford, 2010).Social investment strategies include interventions such as mobilising of human and social capital, facilitating employment and self-employment, promoting asset accumulation, and hence bringing about significant improvements in the material welfare of individuals, families and communities (Midgley, 2010).It is in the context of a need for material welfare that developmental social work incorporates social development as both an investment strategy to address structural injustices in society and to make a contribution to achieving national social development goals through socio-economic development programmes (Midgley, 2010).
Within the context of developmental social work, restorative justice is a social investment strategy which seeks social change through a range of community-based, asset-accumulation, strengths-based and empowerment strategies, including prevention, human and social capital development strategies, and partnerships.Patel (2005) affirms that the focus of developmental social services in the criminal justice field is on community-based strategies, which include crime prevention, crime reduction and rehabilitation of offenders from a restorative and a social and economic justice perspective.
Developmental social work utilises the strength-based approach in helping service users to recognise and use their inner resources, skills and capacity for growth.Social workers using the strengths perspective believe that both the victim and offender "are innately resilient and that their ability to identify and negotiate solutions should be supported" (Midgley, 2010:14).Empowerment is similar to strengths; however, it has "a stronger contextual connotation, suggestive of the relationship between individuals and the negating, disempowering, and oppressive environments in which they find themselves" (Midgley, 2010:14).In macro social work, practice strengths and empowerment are integral elements in bringing about change at the community level and are similar to Sen's notion of capabilities, which has been widely invoked in social development circles (Midgley, 2010).Sen (1999) argues that development is a process of expanding on human freedoms and that freedoms depend on determinants such as social and economic arrangements, for example, education, health care, political and civil rights.Restorative justice sees crime as an act against the victim and shifts the focus to repairing the harm that has been inflicted on the victim and community; it believes that the offender also needs assistance (Department of Justice and Constitutional Development, 2011).Developmental social work focuses on the capabilities of both victim and offender by facilitating human and social capital development programmes, which include providing all parties with complete information on the purpose of restorative justice process, their rights within the process and the possible outcomes of the process; encouraging both victims and offenders to bring support persons to the encounter, provided that this does not compromise the rights and safety of any other party; and engaging in training of role-players who are responsible for the facilitation of restorative justice processes and ensuring that these role players have experience to facilitate restorative justice processes (Department of Justice and Constitutional Development, 2011).Developmental social work includes victim empowerment and prevention programmes ranging from preventing recidivism to addressing risk factors that lead to crime and violence in the society, such as poor socio-economic conditions (Patel, 2005).To this end, social investment strategies include concrete investments in the form of resources and services (Midgley, 2010).
Developmental social work regards the concept of social integration and normalisation as central to the theory of developmental social work and, therefore, integrates service users into the community and promotes normal community living (Midgley, 2010).As such, community integration "requires significant investments that facilitate normal community living such as access to housing services, transportation, education, medical care, and recreation and cultural facilities.It is closely associated with the idea that those living in the community should be afforded opportunities to participate in the productive economy" (Midgley, 2010:15).
Developmental social workers build on micro skills such as interviewing, interaction and communication to a range of mezzo and macro skills which transform social welfare to social development, and promote social justice and human wellbeing, including skills in managing organisational change, project management, action research, mediation, conflict resolution, facilitation, development of partnerships, advocacy and antidiscriminatory practice (Patel, 2005).Midgley (2010) affirms the link between mezzo and macro skills in that many of the skills used by developmental social workers with groups also apply to community practice.These skills fit within the group processes of restorative justice and they link with the macro community context for social investment strategies and include facilitating decision-making in group meetings and educating group members to engage effectively in decision-making tasks; skills that empower local people through cooperative action among a variety of communities, women's, youth and other groups, and to link them to more established developmental agencies; skills to coordinate, network and mediate cooperation between different projects and agencies; and advocacy skills in order to facilitate linkages between local groups and those who have resources (Midgley, 2010).The investment of resources for social development requires brokering, "but it also involves wider advocacy for social justice such as when community members face entrenched inequalities in resources and power, or when they are exploited or discriminated against" (Midgley, 2010:20).Hoefer (2012:3) asserts that the key difference in advocacy practice by social workers and other professions is that social workers always "have the goal of securing or retaining social justice as the primary motivation for their advocacy".

RESEARCH METHODS
A cross-sectional survey was conducted among victims of crime who participated in a restorative mediation project in Atteridgeville, Pretoria.Within the quantitative approach, the researchers opted for an explorative and descriptive research design to articulate and illustrate the views and experiences of victims following participation in restorative mediation.A list containing the contact information of participants was obtained from the service provider.The time frame from which to select respondents was set for the period May 2010 to August 2011 so as to ensure that respondents had adequate memory of the mediation experience (data were gathered during October 2011).A total of 97 victims completed restorative mediation during this time period.No sampling strategy was employed as the aim was to interview the maximum number of participants.The study managed to gather data from 59 respondents (61% coverage of the study population) as the contact numbers of some participants changed while others had moved away.Some worked far from home and could not be interviewed.No direct refusals were encountered.Respondents were informed about the purpose of the study and their written consent was obtained prior to the interviews.The standard ethical considerations of confidentiality, no harm and voluntary participation were adhered to (Babbie & Mouton, 2001).
The research instrument was a structured questionnaire which was completed during face-to-face interviews.Data were captured in Microsoft Excel and transferred to the Statistical Package for the Social Sciences (Version 20) for analysis.The results are presented as frequencies and percentages, while tests of association were conducted to determine significant associations between variables.Depending on the nature of these variables, Pearson's chi-square, Fisher's exact and Mann-Whitney U tests were used.To provide a richer, more in-depth understanding of the limitations of restorative mediation, information from one completed questionnaire is summarised and presented as a deviant case study.

Background characteristics of respondents
The study comprised two thirds female (n=39; 66.1%) and a third male (n=20; 33.9%) respondents.Their ages varied from 19 to 45 with a mean of 29.39 years and a standard deviation of 6.2 years.Africans made up the vast majority of respondents (n=56; 94.9%), followed by only one White (1.7%) and two Coloured (3.4%) respondents.Nearly half were single (n=26; 44.1%), one in three was living with a partner (n=19; 32.2%) and one in five was married (n=12; 20.3%).Two respondents (1.7%) were divorced and another two (1.7%) were widowed.
Nearly two thirds of respondents (n=38; 64.4%) were victimised by someone they knew well.One in 5 (n=13; 22.0%) did not know the offender that well, and roughly 1 in 10 (13.6%) did not know the offender at all.Of the respondents who knew the offender well, roughly half (n=20; 52.6%) were victimised by a spouse or partner, 7 (18.4%)by a family member, 4 (10.5%) by a friend, 3 (7.9%)by a neighbour, 2 (5.3%) by someone they know in the community, and 1 each (2.6%) by an employer and former partner.The offences affected the respondents' relationship with the offender in various ways.Most reported that the relationship deteriorated (n=18; 47.4%), while 1 in 5 (n=7; 18.4%) ended their relationship with the offender.Six respondents (15.8%) stated that they were scared of the offender and 3 (5.1%)had to change residence because of the offence.Three respondents (5.1%) did not want to have any contact with the offender, while 1 refused her child's father to visit the child (1.7%).

Referral to restorative mediation
Three quarters (n=44; 74.6%) of respondents indicated that the court referred them to restorative mediation, while a quarter (n=15; 25.4%) indicated the referral agent as a social worker.Slightly more than half (n=31; 52.5%) stated that they were not consulted about the possibility of mediation before their cases were referred to restorative intervention.Asked about their first reaction upon learning that their cases have been referred for mediation, half of the respondents (n=28; 51.9%) expressed emotions ranging from shock, anger and disbelief to being scared and feeling intimidated.A quarter (n=14; 25.9%) wanted to know more about victim-offender mediation, while 10 (18.5%) had no problem with the referral.Two respondents (3.7%) reportedly felt relieved that their cases were referred for mediation.Most respondents (n=54; 91.5%) voluntarily agreed to participate in the intervention.

Preparation for restorative mediation
Two thirds of respondents (n=46; 78.0%) stated that they were prepared for restorative mediation, and sufficiently so (n=42; 91.3%).Preparation mostly took the form of premediation meetings with the facilitator (n=39; 84.8%) and participating in a victimempowerment programme (n=7; 13.0%).One respondent (1.7%) benefited from both types of preparation.Of the 4 respondents who stated that they were not sufficiently prepared for mediation, 2 complained about a lack of preparatory information, while another 2 indicated that there was no time for preparation as their mediation meetings took place on the same day as referral to the programme.

Implementation of restorative mediation
The time lapse between the offences and the mediation sessions varied between 1 and 11 months with a mean of 3.5 months and a standard deviation of 2.5 months.Most sessions took between 1.5 and 2 hours to complete (n=23; 39.0%), while 11 (18.6%)lasted between 1 and 1.5 hours, and 2 and 2.5 hours.Seven sessions (11.9%) lasted between 2.5 and 3 hours, 6 sessions (10.2%) between 3.5 and 4 hours, and 1 (1.7%) more than 4 hours.All respondents (n=59; 100%) stated that their mediation sessions were conducted in a language that they understood.Slightly more than half of respondents (n=33; 55.9%) had someone accompanying them to the mediation meeting.
These persons were mostly family members of the victim (n=29; 87.9%), while 3 (9.1%)were accompanied by a friend and 1 (3.0%) by a neighbour.
Half of the respondents (n=29; 49.2%) rated the offender's attitude during the mediation meeting as "cooperative", while the same proportion rated his/her attitude as "neutral".Only one respondent (1.7%) indicated the offender's attitude as "uncooperative".Most respondents (n=47; 79.7%) stated that the offender's behaviour during the meeting did not affect them.Twelve respondents (20.3%) felt intimidated by, and scared of, the offender during the mediation.Respondents' rating of the restorative mediation meetings are depicted in the following table.

Outcomes of restorative mediation for victims
Three quarters of respondents (n=46; 78.0%) stated that the offender showed remorse during the mediation meeting.The offender apologised for his/her wrongdoing in nearly all of the cases (n=56; 94.9%), with most respondents (n=41; 73.2%) experiencing the apology as honest.An agreement was reached in the majority of cases (n=51; 86.4%).Most respondents were satisfied with the agreement (n=47; 79.7%).The main reasons for their satisfaction were the offender promising to refrain from his/her problematic behaviour (n=12; 28.6%), having been apologised to (n=10; 23.8%), the offender being referred for treatment (n=9; 21.4%), and that peace was re-established (n=5; 11.9%).In 2 cases (4.8%) the victim's property was returned, while another 2 (4.8%) reported that the offender apologised and accepted going to prison.One respondent (2.4%) was relieved not to return to court, while another (2.4%) was happy to separate from the offender.
The majority of respondents felt empowered following participation in restorative mediation (n=46; 78.0%).Nearly two thirds (n=38; 64.4%) indicated that the process brought justice to the offence.Slightly more than half indicated that they had a better understanding of the offender's situation (n=35; 59.3%) and that the mediation benefited the offender (n=33; 55.9%).The bulk of respondents (n=53; 89.8%) stated that they would recommend restorative mediation to other victims of crime, because of the strengthening of relationships (n=21; 37.5%), having an opportunity to express oneself (n=19; 33.9%) and the matter being dealt with outside the formal justice system (n=9; 16.1%).Of the 6 respondents who would not recommend restorative mediation, 2 (3.6%) stated that the mediation meeting only made them angrier, it did not help in their situations, and that the intervention does not work for strangers.The vast majority of respondents (n=56; 94.9%) would not change anything to the restorative programmes they went through, while 3 (5.1%)recommended that more significant others be invited to the meetings.

Follow-up and aftercare
Slightly more than half of respondents (n=34; 57.6%) were informed of further support services following the mediation meeting.Three respondents (5.1%) stated that they require follow-up counselling.

Deviant case
One case not representing the experiences of the majority of respondents relates to a young man whose brother was murdered by a stranger.He initially did not want to participate in victim-offender mediation and later on regretted agreeing to the meeting.The meeting took place nine months after the offence.Despite the respondent indicating that he was adequately prepared for the meeting, he provided a poor rating in terms of healing the damage caused by the offence and the meeting mending damaged relationships.A good rating was given for the opportunity to ask the offender questions and having time to share his feelings about the offence.Nevertheless, he provided a poor rating for mediation in terms of achieving what it set out to achieve.The respondent did not experience the offender's apology as honest and the meeting did not result in an agreement.He indicated that he didn't feel empowered by the process, didn't think that the process brought justice to the offence, didn't have a better understanding of the offender's situation, and didn't believe that the mediation benefited the offender.As to whether he would recommend mediation to other victims, the respondent stated: "I regretted agreeing to take part in it.Seeing him made me angry.Not all crimes must be referred to victim-offender mediation.The process was useless and brought more pain than healing when you see the offender".

Tests of association
Tests of association found no significant relationships in terms of biographical information and the results of the study.However, a significant association existed between whether respondents experienced the offenders' apology as hones t and achieving what the mediation set out to achieve (Mann-Whitney U: p=0.038), the offender showing remorse (Fisher's exact: p<0.001), and reaching an agreement following mediation (Fisher's exact: p=0.038).In a similar vein, having reached an agreement showed a significant relationship with healing the damage caused by the offence (Mann-Whitney U: p=0.033), having been consulted about mediation prior to the referral (Fisher's exact: p=0.018), and the offender showing remorse about the offence (Fisher's exact: p=0.010).Significant associations also featured regarding the anticipated outcomes of restorative mediation (see Table 2).

DISCUSSION
Seen broadly, the results of this investigation confirm the anticipated positive influence of restorative mediation on victims of crime.The majority of respondents were victims of interpersonal offences, mostly by someone they know well.In most cases the relationship between the victim and the offender deteriorated, with some relationships even coming to an end.Following mediation, however, the majority of victims were apologised to, an agreement to remedy the situation was reached, and the damage caused by the offence was healed.The positive impact of restorative mediation is further reflected in the result that the vast majority of respondents would recommend mediation to other victims of crime.In addition, the significant association between important restorative justice variablesi.e.showing remorse, honest apology, reaching an agreement, healing and consultation in decision-makingattest to the positive outcomes of restorative mediation for victims of crime.Despite these encouraging results, important questions are raised from a developmental social work perspective as to restorative procedures and the broader potential influence of mediation on participants.This is particularly relevant in that a quarter of participants in this study were referred for restorative mediation by social workers.
The true voluntary nature of restorative mediation is not entirely clear.Although the majority of victims indicated that they participated voluntarily in mediation, indications are that about half were to some extent coerced into the intervention as they had not been consulted about the possibility of mediation prior to referral by the court.Participation in decisions that affect people is a central theme in developmental social work.For restorative justice procedures to accomplish their essential philosophical aims, the voluntary and non-coercive principles should be stated directly (Abrams, Umbreit & Gordon, 2006).With a clear introduction as a voluntary initiative, victims can be assured that their participation is not compulsory and that the case can still follow standard legal proceedings without any penalty levied against them.
Restorative mediation is not a once-off solution to crime and other social ills.Mediation is one step in a process of personal growth and change, which are the cornerstones of social investment programmes.The finding that only a quarter of respondents confirmed that mediation mended their relationship with the offender attests to the need for continued strengthening of relationships, whether in the family, friendship or community contexts.From a developmental social work perspective, this emphasises the need for human and social capital development.Social capital is embedded in trust, which is needed for new and strengthened relationships and networks (Larance, 2001).Mediation should instigate a series of intervention and support mechanisms, given that assaultpotentially as a result of substance abusewas the most frequent crime committed against victims.Entry into the justice system could open further doors for a continuum of care for people trapped in cycles of violence and crime.However, such avenues appear closed for many victims as only 57% were informed of further support services.
Given its holistic approach, developmental social work is an appropriate strategy to address the challenges of marginalised communities.Atteridgeville in Pretoria is a poor community characterised by numerous social ills and high unemployment.A major shortfall emanating from the present investigation is the absence of linkages to other forms of social intervention that could enhance the economic situation of beneficiaries of restorative programmes.Slightly more than half of respondents were informed about other social services, which limits avenues for material welfare and poverty alleviation.This important deficit of the mediation programme that was investigated is unfortunately not unique and has been recorded in other forms of restorative and non-custodial interventions as well (Steyn, 2008(Steyn, , 2012)).While the results strongly suggest a once-off mediation meeting to affect positive change at the personal level, it fails to address the broader economic and material needs of participants in mediation.Interventions that contribute to improvements in material wellbeing are important for developmental social work.Midgley (2010) elaborates that these interventions are community-based, utilise capabilities, foster empowerment and facilitate the integration of service users into community living by providing access to housing, medical services, education, medical care, and recreational and cultural facilities.
Social workers have a reform and rehabilitation role, which means that they are not only involved in providing rehabilitative services within correctional facilities but also with community-based services such as probation, parole and restorative justice (Rainford, 2010).From a holistic perspective, developmental social workers recognise the ecological perspective that everything is connected (Ife, 2012) and hence facilitate the inclusion of service users in all aspects of community living (Midgley, 2010).Being committed to social justice and human rights, developmental social workers should be "particularly concerned with the plight of excluded groups and can, as such, be agents of inclusion" (Lyons & Huegler, 2011:37).The result that nearly half of respondents had no one accompanying them to the restorative meeting creates an undesirable situation, as significant others, in particular family members, neighbours and friends, are not mobilised as human capital in the resolution and prevention of conflict.While it must be acknowledged that some victims may not wish others to know of their personal and domestic problems, it is equally important for the developmental social worker to facilitate adequate support systems for victims of crime.Restorative practitioners cannot overlook the important role which the community plays in either ignoring or condoning violent behaviour.Furthermore, family and community members could play a significant role in monitoring the good intentions to reform which offenders pledge in the mediation agreement.
In comparison with the deviant case, findings indicate that the nature of the crime, murder in the case in point, and the level of familiarity with the offender have an influence on the outcome of restorative mediation.The willingness to participate also plays a prominent role in restorative outcomes.The deviant case confirms the relationship between honesty, belief that justice has been done, whether the victim's rights are respected, and understanding of the offender's situation.This leaves the victim disempowered and in turn his or her perception that the offender has also not benefitted from the process.Evidence (Hargovan, 2010) supports the observation that mediation is inappropriate in cases where offenders are not remorseful for what they have done or where a history of violence is present.
Understanding reasons for the crime has a positive outcome for restorative justice, because it leads to specific promises to undertake specific actions intended to correct the factors which contributed to the wrong-doing.Victims were satisfied knowing the commitment of offenders to engage in correctional interventions, including treatment of substance dependency, anger management and family counselling.Studies emphasise the role that substance abuse plays in violence, drug offences and committing of other crimes (Rainford, 2010).Patel (2005) refers to the inextricable link between crime and violence with poverty, under-development and the history of repression and associated violence in South Africa.
Information and being well-prepared play an important role in the outcome of restorative mediation.They contribute to being in control of emotions, not being intimidated by the offender, and capacitated to participate in the decision making.Developmental social work also invokes the concepts of self-determination and participation (Midgley, 2010), which is in line with a human rights perspective (United Nations, 1994).In developmental social work the notion of social rights reflects the belief that those served by social workers not only have a right to make decisions but to benefit from services and supports.The value base of human rights recognises the inherent worth and dignity of all people (Straub-Bernasconi, 2011).This imperative in the context of restorative mediation is adequately supported by the result that 85% of respondents stated that they were treated with dignity and respect throughout the restorative justice process.
Findings reflect that it is not so much the level of being prepared or the opportunity to ask questions to the offender which influence a positive outcome, but rather the level of sincerity and honesty of the offender's remorse for the wrong done which may be, on a deeper level, also related to the willingness and readiness of the victim to forgive.The reaching of an agreement clearly supports the restoration of peace.However, given the lack of evidence that it will succeed, the sustainability of restorative mediation is uncertain.Developmental social work is committed to social investment strategies and hence "requires the use of cost-effective interventions and evidence-based methodologies that promote effective practice" (Midgley, 2010:13).However, re-offending usually takes place at a distant point from the restorative intervention and is more difficult to measure against victims' experiences of the process (Hargovan, 2010).
Developmental social work is embedded in investment strategies in restoring people to full functioning in society (Rainford, 2010).Important investment strategies which restorative mediation offer victims of crime include victim empowerment, insight into the origin of problems, participation in decisions that affect them, having their rights respected, and involvement in determining the outcomes of remedial action.Investment strategies for the offender include gaining insight into his or her wrong behaviour and its impact, opportunity for honesty and showing remorse, non-custodial measures in appropriate cases to continue with normal economic activities, personal growth through forgiveness, reintegration with the family and community, and referral to specialised intervention.

RECOMMENDATIONS FOR POLICY, PRACTICE AND TRAINING
Restorative justice programmes are linked with developmental services through social investment strategies.By bridging the micro and macro service delivery divide, developmental services create a platform for restorative justice programmes to integrate both the victim and the offender into community life.This shift is in accordance with Vision 2030 of the National Development Plan (NDP) (RSA, 2011).The key priorities for South Africa as outlined in the NDP are to eliminate poverty and reduce inequalities through a new approach which moves from a passive citizenry receiving service from the state to one that systematically includes the social and economically excluded, where people are active champions of their own development, and where government works effectively to develop people's capabilities to lead the lives they desire.The success of this approach includes, amongst other things, investing in skills development and employment, education, and strengthening the links between economic and social strategies.The NDP (RSA, 2011) acknowledges that it is harder for people to develop their capabilities, pursue their personal goals and take part in social and economic activity if they feel unsafe.South Africans, therefore, have to feel safe everywhere and have confidence in the criminal justice system to protect them.
A developmental approach provides a policy framework for guiding restorative justice practice and training programmes.To this end, it is recommended that restorative mediation programmes should be extended to:  embrace an integrated, holistic approach embedded in human rights and which values the dignity of both victim and offender;  show commitment to social justice and promotion of inclusiveness and equality;  take the structural causes of crime and violence into consideration and address the risk factors that lead to crime and violence in society, such as poor socio-economic conditions, poverty and unemployment and a lack of economic, educational and social opportunities;  prioritise social investment strategies that build human, social and economic capital development, including education, skills development, employment, microcredit and micro-enterprises, asset programmes, removal of barriers that inhibit economic participation, and integration of service users into community living by providing access to housing, medical services, education, and recreational facilities;  utilise strengths-based, empowerment, capability and advocacy approaches, strategies and interventions to facilitate social change;  facilitate participation of individuals, families, social groups, organisations and communities in active partnerships;  prioritise crime-prevention programmes targeting vulnerable groups, including women, children, youths and offenders;  monitor and evaluate the effectiveness of restorative justice programmes.

LIMITATIONS AND FUTURE RESEARCH
It should be kept in mind that this study was predominantly explorative in nature and conducted with a small number of respondents to determine the outcomes of restorative mediation for victims of crime.Moreover, the study investigated restorative mediation among a fairly homogenous population.Given the diverse profile of South Africa's citizenry, research is needed to determine the potential influences of culture, race, gender and age on restorative procedures and outcomes.Given the local realities of socio-economic inequality, it is equally important to investigate the possible impact of differential economic status between victims and offenders.In addition, longitudinal research is needed to evaluate the longer-term impact of restorative mediation, including outcomes such as recidivism and behavioural change on the part of the offender.