Perceptions of traditional health practitioners on violence in the Helderberg Municipal Area , Western Cape

South Africa has a layered, pluralistic system of health provision, in which the paradigms of African traditional healing, biomedicine and alternative medicine operate in parallel with each other. People who are unwell often use two or more of the systems available, sometimes simultaneously. In this process they move between different ways of interpreting and understanding illness and health and access different ways of finding relief and treatment (Digby 2006; Gqaleni et al. 2007; Levine 2012). An estimated 300 000−350 000 traditional health practitioners are active in South Africa, approximately 15 times higher than the number of medical practitioners (Liddell, Barrett & Bydawell 2005). According to a study by Nxumalo et al. (2011) underpriviliged people, as well as those who live in rural areas, are most likely to consult traditional health practitioners.


Introduction and background
South Africa has a layered, pluralistic system of health provision, in which the paradigms of African traditional healing, biomedicine and alternative medicine operate in parallel with each other.People who are unwell often use two or more of the systems available, sometimes simultaneously.In this process they move between different ways of interpreting and understanding illness and health and access different ways of finding relief and treatment (Digby 2006;Gqaleni et al. 2007;Levine 2012).An estimated 300 000−350 000 traditional health practitioners are active in South Africa, approximately 15 times higher than the number of medical practitioners (Liddell, Barrett & Bydawell 2005).According to a study by Nxumalo et al. (2011) underpriviliged people, as well as those who live in rural areas, are most likely to consult traditional health practitioners.
In the past decade there has been increased interest in the knowledge of traditional health practitioners, as well as in their approaches and roles in the management of health and illness in South Africa (Levine 2012;Richter 2003).There are several areas of enquiry in this regard.One This study on perceptions of violence was conducted with 56 traditional health practitioners (diviners: amagrirha) in the Helderberg Municipal Area of Cape Town Metro.It forms a subsection of a larger study on African medicine.This particular research focuses on how traditional health practitioners perceive violence, including gender-related violence.Individual, in-depth interviews were done with 21 traditional health practitioners and focus group discussions were held with 35 participants.The paper reports on their understanding of, as well as the meanings attached to, community and gender-based violence in an urban setting.The traditional health practitioners related violence to a range of disconnections in society, ranging from not adhering to traditional norms and practices, to breaks in relations between parents and children, within families and in marital-and sexual relations.They referred to a general sense of disjuncture between the living and the ancestral worlds.The accumulative effect of this sense of not being connected was seen as damaging and a precursor to violence.In two sites where there were high concentrations of violence, ceremonies were held to purify the areas by ritual.In addition to attending to the physical manifestations of illness, distress and violence, these traditional health practitioners attempted to enhance and restore proper social relationships between the living, as well as between the living and the dead.
A number of studies, especially in anthropology, have been done on the ways in which traditional health practitioners understand the aetiology of ill-health (Thornton 2004;Wreford 2008), particularly that of HIV (Henderson 2005(Henderson , 2011;;Leclerc-Madlala 2002;Wreford 2005aWreford , 2005b)).There is also a growing body of literature concerning the professionalisation of traditional health practitioners (Devenish 2005;Thornton 2009).Literature concerning traditional health practitioners often emphasises that they are able to attend to clients in their own language and share a similar cultural background, as well as having related reference points and world-view (Henderson 2011;Wreford 2005b).
Although studies in South Africa (Jewkes et al. 2010), more specifically, the Western Cape, show high levels of violence, especially community-, gender-related and domestic violence (Dinan, McCall & Gibson 2004;Dunkle et al. 2004;Jina et al. 2010;Joyner & Marsh 2011), little is known about how traditional health practitioners view this situation.Despite anecdotal evidence that men and women consult traditional health practitioners concerning experiences of violence, there is scant information on the issue.Little is known about how traditional health practitioners make sense of and try to deal with people who consult them regarding violence (Gibson, Dinan & McCall 2005).The goal of this study is to help fill this lacuna.

Problem statement
During the past five years the author has been involved in work and research with traditional health practitioners, including diviners (amagrirha: isangoma) and herbalists (ixwele: inyanga), on African medicine in the greater Cape Town area.The Traditional Health Practitioners Act 22 of 2007 defines a traditional health practitioner as one who is 'engaged in traditional health practice' and is so registered under this Act.Provision is made for categories of diviners, herbalists, traditional birth attendants and traditional surgeons (Republic of South Africa 2008:7).The participants in the study, are diviners or amagrirha (isiXhosa).As the Interim Traditional Health Practitioners Council was only instituted on 12 February 2013, they are not as yet registered under the Act (Department of Health 2013), but practise as amagrirha.During interaction with the author they often raised as a matter of concern the high prevalence of violence in the Western Cape.An earlier study in which the author was involved also indicated that survivors of sexual violence did not always seek help from formal health facilities, yet they sometimes turned to traditional health practitioners for this purpose (Gibson et al. 2005).The ways in which traditional health practitioners tried to make sense of violence, as well as what they perceived as violence, were not always evident.For example, when the participants talked about violence, they referred not only to physical assault and domestic abuse, but also to witchcraft, poverty and even accidents.The latter issue was highlighted by newspaper reports in 2011 of a discussion by the Nelson Mandela Bay Municipality about making funds available in order to organise a ritual cleansing ceremony of a road with a very high accident rate (Matavire 2011).This made sense to the participating amagrirha, who seemed to have diffuse, complex perceptions of violence as constituting multiple forms of harm.This in turn raised the question of how they, inter alia, understand accidents as a being a form of violence.A related question was how traditional health practitioners make sense of violence in the communities in which they live.

Aims of the study
The study attempts to gain insight into and describe the understanding of violence, as well as the meanings attached thereto, by amagrirha in the Helderberg Municipal Area in the Western Cape.Another aim is to ascertain what these participants identify as violence.Finally, the research aimed to gain insight into the factors that the traditional health practitioners perceive as contributing to community-, interpersonal-, gender-based and sexual violence.

Definition of key concepts
As indicated in the problem statement, the participants in the study are, broadly speaking, traditional health practitioners and, in accordance with the categories set out in the Traditional Health Practitioners Act, belong to the category of diviners (amagrirha).According to Henderson (2005:26), it is imperative to understand that diviners perceive health as not just involving the wellbeing of the human body but also the correct arrangement of diverse social relationships.In addition to healing the physical manifestations of illness, amagrirha endeavour to enhance and restore proper social relationships between the living, as well as between the living and the dead (Henderson 2005:26).

Significance of work
Many studies have been done in South Africa on violence (Ashforth 2005;Seedat et al. 2009) and on gender-related and domestic violence (Dunkle et al. 2004;Jewkes et al. 2009Jewkes et al. , 2010;;Joyner & Marsh 2011), but a review of the current literature on traditional health practitioners (Gqaleni et al. 2007;Henderson 2011;Thornton 2004;Wreford 2008) gives little indication of the meanings attached to, and understanding of, violence by such traditional health practitioners.This paper aims to shed some light on this important aspect.

Design and samples
The study is informed by ethnography and is qualitative, context-specific and explorative.Practising, fully-trained amagrirha were selected by means of a non-probability 'snowball' sample (Heckathorn 2002).Individual participants were asked to refer the author to four other traditional health practitioners.Since snowball samples are sensitive to affective bias, not all referrals were included in the study; instead, two out of every four referrals were selected randomly for inclusion.In-depth interviews were done with 21 traditional health practitioners.Two focus group discussions (FGDs) were arranged and 20 potential participants were invited to each.The expectation was that not more than 10 participants would be able to attend.Although only eight confirmed their attendance for the first session, 15 arrived.In the second focus group 11 confirmed and 20 arrived (making a total of 35 participants).In both instances the amagrirha were adamant that they should all participate, otherwise no-one would take part.The focus groups were thus larger than anticipated.Group interactions were managed so as to ensure that each participant had an opportunity to take part.Each participant was handed the itshoba (the beaded stick with oxtail used to ward off spirits and to disseminate knowledge during training) to enable him or her to speak without interruption.

Data collection
The data presented in this paper is based on in-depth individual interviews and FGDs done with traditional health practitioners who are categorised and practising as amagrirha [diviners] in the urban setting of the Helderberg Municipal area, Western Cape.One or more in-depth interviews were done with 14 females and seven males.A final-year medical student who speaks fluent Xhosa and had built up a good relationship with the traditional health practitioners, was trained in qualitative research methods and conducted the initial interviews.With the consent of the participants, the interviews were tape recorded, transcribed and translated by the interviewer.A doctoral candidate in Sociology (who is a mother-tongue Xhosa speaker) checked the translations for accuracy.Because of time constraints the interviews were not reverse-translated, but the author, a medical student, a doctoral student and two Xhosa-speaking Master's students discussed and reviewed the translations.One focus group was held with 15 participants (10 women, five men) and the other with 20 (13 women, seven men).The FGDs were done under the supervision of the author and facilitated by the doctoral candidate, assisted by the two candidates for Master's degrees in Anthropology, also mother-tongue Xhosa speakers.All had extensive training and experience in FGDs.To create a neutral setting, the FGDs were conducted in a secluded venue at the university.The sensitivity and potentially-distressing nature of a discussion on violence was discussed carefully with the participants beforehand and each was urged to call a halt to the proceedings if they felt uncomfortable.Participants were informed that the discussion would be about the community and traditional health practitioners in general and that they were not expected to reveal personal information or experiences.Although the focus groups were made up of men and women, the participants were trained amagrirha and this facilitated open discussions.
At the end of 2011, a feedback session on the research results was organised with the traditional health practitioners who had participated in the study.They indicated their satisfaction with the outcome and the way in which the resulting research material was presented.

Data analysis
Qualitative analysis is aimed at presenting the complexity of studied phenomena by attending to the structures underlying it and by presenting it in a way that makes sense.The author's task was to present an emic view of research participants whilst analysing the results in order to establish its broader significance and meanings (Green & Thorogood 2004:175).All transcribed material was scrutinised for patterns of meaning and material was coded to indicate similarities and/or differences therein.At the same time the texts were checked for saturation adequacy.The first round of analysis involved open-coding, followed by listing the analytical themes that emerged.In this way data were categorised, labelled, organised and placed into themes.The second round of analysis involved thematic coding.Finally, selective coding was utilised so as to give greater analytical coherence to the data (Auerbach & Silverstein 2003;Strauss & Corbin 1990).The selected codes, categories and themes were discussed with a colleague -an experienced ethnographic researcher.

Context of the study
The research was conducted in the Helderberg Municipal Area, which borders on a residential basin of more than 600 000 potential clients.It includes Strand, Somerset West, Gordon's Bay, Sir Lowry's Pass, Macassar, Faure, Firgrove, Nomzamo, Lwandle and Broadlands.There are a number of informal settlements, namely, Nomzamo Extensions B and C, Broadlands Extension and Bridgewater Extension 1.The areas with the highest percentage of informal dwellings (16.7%) are Broadlands, Nomzamo, Lwandle and Sir Lowry's Pass (Helderberg District Plan 2012).Unemployment levels (50% -60%) are the highest in Bridgewater extension 1, Nomzamo and Lwandle (Romanovsky & Gie 2006).Crime (18%), violent crime and domestic violence are highest in Bridgewater extension 1, Nomzamo and Lwandle (Western Cape Provincial Department of Local Government 2010).

Ethical considerations
Ethical clearance (reference number 10/1/43) was obtained from the Faculty of Arts Research and Ethics Committee, as well as the Senate Ethics Committee of the University of the Western Cape.The study adhered strictly to the Ethical Guidelines of Anthropology Southern Africa, as well as to international Ethical Guidelines for Social Science Research in Health.The aims of the research were set out carefully and discussed with amagrirha individually and in group settings.None of the participants wished to sign consent forms, but they gave their consent verbally (this process was tape recorded).They were assured that participation was voluntary and that they could withdraw at any time without penalty.In the case of in-depth interviews, anonymity was ensured and both participants and researchers signed agreements of confidentiality.In FGDs, as well as in the final feedback session, the traditional health practitioners agreed verbally not to divulge the identity of fellow participants or what had been said by individuals during the discussions.The author, facilitators and moderators also agreed to maintain the confidentiality of the focus groups.Each participant's verbal agreement was tape recorded.All participants were given numbers for purposes of recording and transcription.As agreed with the participants, all recordings were erased after transcription.The researchers and facilitators were trained in domestic violence and rape counselling and had relevant contact information for facilities for counselling, both on-and off campus, if required.

Trustworthiness
The author and postgraduate students have worked extensively with traditional health practitioners over the past three years and care was taken to approach interactions respectfully and with sensitivity to the rituals needed to make it possible for them to share their knowledge.The principal methods of enhancing the dependability of this qualitative study were triangulation through the use of indepth interviews, FGDs, extensive notes taken during FGDs and a feedback session to check with the traditional health practitioners whether they had been understood correctly (Shenton 2004;Silverman 2010).Because the main aim of the study was to understand the phenomenon of violence from the perspective of the traditional health practitioners, they were ultimately the people who evaluated the credibility of the results.The author and the interviewers discussed the questions and guidelines both before the interviews and again after the interviews and FGDs.The accuracy of the translation process was enhanced by the skill of the interviewers, who were trained in this beforehand.Interviews and FGDs were transcribed and translated in tandem by the medical and doctoral student and then checked for similarity and accuracy in discussion with the author and the Master's students (Squires 2009).

Discussion of results
The discussion of the results includes an overview of the way in which violence is understood in health research and the extent to which this understanding resonates with the perceptions of the amagrirha.Verbatim quotes are utilised so as to contextualise the data and to highlight the factors the participants identified as constituting and contributing to violence.

Broad understandings of interpersonal violence
In both public-and mental-health discourse, violence, especially of an interpersonal nature, is often analysed in relation to a socio-ecological framework (Dahlberg & Krug 2002;Ward et al. 2012) involving a complex intersection of factors in relation to experiencing and/or perpetrating violence.In this framework, causal relationships are drawn frequently between conditions in the past, for example, marginalisation, poverty, overcrowding, unemployment, a history of family violence and/or mental illness and the prevalence of violence or individual incidents in the present.The ecological framework accordingly does not offer a singlelevel explanation for violence but rather a need to interrogate the interaction of multiple variables from the individual to the society, across a backdrop of social inequality (Eisikovits & Edleson 1989;Ward et al. 2012).It examines violence involving interconnected individuals and also in relationships, inter alia, with wider family, friends and other social institutions such as schools, mass media and churches.It includes the beliefs and values of people in interactive and dynamic relationships on a variety of overlapping planes that impact on each other.Violence is viewed as being rooted in the social-, economic-, cultural-and political context in which it occurs.Central foci are frequently on the conditions under which violence occurs, how such conditions can be anticipated and prevented or changed, the physical and psychological outcomes of violence and how victims can be assisted (Jewkes et al. 2009;Seedat et al. 2009).
In relation to the perceived causes of violence, the traditional health practitioners referred to what is understood broadly as being structural violence, to the impact of certain rituallypolluted spaces that 'attract' violence, to the violent outcomes of spiritual disconnection from the ancestors, occult violence, as well as gender-related and sexual violence.A selection of excerpts from interviews and FGDs is presented in order to provide better insight into the ways in which traditional health practitioners expressed their understanding of the causes of violence.

Structural violence
Broadly speaking, structural violence is embedded in unequal access to economic, political, legal, religious, health and cultural structures and arrangements and so forth and has a negative and harmful impact on peoples' lives (Farmer 2004).The notion of structural violence approximates the ecological model of violence and, in a number of ways, resembled the manner in which traditional health practitioners explained its causes in a focus group discussion: 'Most of the problems of violence are about poverty.People are frustrated, they thought things will be different, but they find they cannot change their problems.There is no education, there is no food.There is apathy.People can't function, they are just severely depressed.They do stupid things.Violence becomes an outlet.People are unemployed and they rob and hurt other people in order to make a living.' (P1, female, 47 years of age, healer: FGD) In the above quote, the healer, like many others in this study, expressed concern about the lack of grass-roots transformation experienced by poor people against the backdrop of the larger political transformation of the country.As articulated by another healer, an array of circumstances contributed to exacerbate the possibility of violence: 'Some people are innately violent; they come from families that have a history of violence.There are people who are very disturbed and like to commit violence, but it is also true that in the townships you find more violence as compared to the suburbs.In the townships people are poor, they will not always get good mental care, there is much use of alcohol, and it is not so safe.There is a lot of fighting too.In the suburbs there is more robbery, muggings, theft.The Police cannot deal with it.'(P2, male, 51 years of age, healer: FGD) In the above excerpt, socio-economic circumstances and the potentially negative effect they have on individuals and on communities were represented as being potential causes for violence.The deterioration of family, intergenerational and interpersonal relationships was frequently raised as being a contributory factor.
'Our young are brought up in ways that are not acceptable.They cannot be disciplined and they behave in ways that are causing concern to the older people.There is no respect for each other, for old people or those with authority.'(P3, female, 49 years of age, healer: Interview) From the above it is apparent that violence is perceived as a complex interweaving of issues at multiple levels and in a variety of spheres.Legislative changes concerning the 'disciplining' of children was mentioned as contributing to increased community and family disorder and violence, as can be seen from selected excerpts: 'If you try to discipline your child these days, you will be told you are abusing your child.You don't have any way to discipline your children.So they just do what they want and parents just let things be because things have changed.'(P4, male, 55 years of age, healer: Interview) 'Teenagers these days drink a lot, they are at the shebeens [drinking places].If you try and get them from the shebeens and you want them to wash dishes and make food … by trying to get them home or by slapping them when they are too drunk … you will be told you are abusing them.As a parent you don't have any rights.You will be told you are abusing them.Even the boys are not the same as before, they are rude and disrespectful.
They take what they want by force.' (P5, female, 39 years of age, healer: Interview) For the traditional health practitioners, violence is thus complex and impacted on by an array of structural issues, including high rates of unemployment and sexuallytransmitted illness.At the same time, violence is linked to the spiritual world.

Spaces where violence 'collects'
During the course of this part of the wider study on African medicine, amagrirha referred to two ritual cleansing ceremonies held in spaces in informal settlements where a number of people had at some or other time during the past three years been killed, raped, assaulted and/or robbed.These sites were represented as spaces (kwelemimoya) that 'draw' or 'attract' violence, malevolence and troubled spirits and as places where violence and even deaths accumulate (Feldman 2002).In the FGDs, traditional health practitioners explained that such places are defiling and can make people 'dirty inside' (bangcole ngaphakathi) and cause them to act like animals (ilwanaya) or to 'become violent in a way nobody would understand' (P6, male, 60 years of age, healer: FGD).Such places can sicken people and they (or the spaces) need to be cleansed in a ritual manner.
Historically, for instance, jails, mines where accidents have happened and settings such as Vlakplaas and Robben Island, where political detainees were killed or made to suffer, are examples of current and former spaces of pollution that can make a person or group of people sick and/or produce more violence.According to Feldman (2002), mines in Gauteng are often cleansed ritually to prevent accidents from happening.Intersections where many vehicle accidents happen, for example, railway level crossings where people are killed and bushy areas where people are robbed and women are raped, are 'attractors' for more evil.A traditional health practitioner explained: 'As years go by, there are always added episodes of violence.You find that you always hold your chin when you learn about an incident, and the next time you get to know about other kinds of violence more serious than the preceding one.What you find then is that we have sick communities.'(P7, male, 46 years of age, healer: FGD) Another healer elaborated further that: 'Those [places] have demons which drive people mad.Sometimes the demons can be so destructive and powerful that nobody but the law can deal with the situation.Any evil spirit that is drawn there can just enter the body and a person ends up doing things you never imagined.It is not you, it is the evil spirits of that place.' (P8, male, 63 years of age, healer: FGD) Examples of such polluted spaces and of publicised efforts to end such cycles of violence in recent times in South Africa include the ritual slaughtering of a bull at Khayelitsha railway station on January 27th 2007 'to ask for divine intervention to reduce horrific train-related accidents and attacks' (Gophe 2007) and the ritual cleansing of Robben Island by over 100 traditional health practitioners in October 1997 (SAPA 1997).On 15 December 2001 about 700 traditional health practitioners attended a ritual cleansing of Vlakplaas, the notorious farm used for covert murder and torture by agents of the apartheid state (Butcher 2001;Portfolio Committee of Arts, Culture, Science and Technology 2002).As traditional health practitioners explained, future violence could be prevented by resanctifying the area or site, to placate the dead, as well as by seeking the support of the ancestors, perhaps by 'slaughtering a goat and cleaning it [the space], and then they will be at peace with themselves thereafter' (P9, female, 46 years old, healer: FGD).

The violence of disconnectedness from the ancestors and 'tradition'
closely -the ancestors are very much involved in the lives of their descendants and influence their behaviour.The ancestors are protectors but can also be wrathful and bring misfortune if their descendants are disrespectful, do not follow rituals or lapse in morality (Gumede 1990;Schapera & Comaroff 1991).All of the traditional health practitioners referred to 'loss of tradition' as being causal to violence, especially in urban areas and new settlements where people are isolated from previous social relations and 'forget' their ancestors.The quotes below give a better sense of the perception of participants: '… in the past, you could walk at night and only think about ghosts that may appear.But nowadays people want to rob and kill you.People do not work and then they want to get things from other people.Some people are violent because they are angry and jealous with everybody in the world.There is something they lack, and it is peace with their ancestors.'(P10, male, 43 years old, igrirha: Interview) 'Whenever you do something wrong, the bones of the ancestors have moved, or are disturbed.Every person in our culture is watched by his ancestors, they protect you in whatever is going on around you.Sometimes, in a dream, they will ask you to do something for them.In most cases this requires one to slaughter a goat or brew traditional beer [usile umqombothi]; if you fail to do that they will abandon you, and leave you to fend for yourself.In that instance you will not be protected from any evil spirits that come your way, you get into trouble and wayward behaviour, you get into violence.'(P11, male, 58 years old, igrirha: Interview) The disconnectedness from the ancestors and 'cultural traditions' is believed to have the potential to affect the whole community, a family or an individual because they are interlinked, both with each other and with the spiritual world.All the amagrirha in the study emphasised that when such relations are broken, diluted or weak it results in violence and/or ill-health and misfortune.This concern was expressed, inter alia, by traditional health practitioners in the following ways: 'Our communities are cursed, people do not do their rituals anymore, and they do not even know what they are doing.They do not even know or honour their clan-names.When somebody performs their rituals, they do them in the wrong way.That is why people steal, rape, kill and abuse other people.Their ancestors have forsaken them.Their ancestors are angry.'(P12, female, 47 years old, igrirha: Interview) 'Sometimes their ancestors are angry.People in townships no longer engage in traditional rites, they only drink and forget that their ancestors also want their share.When they [the ancestors]tell them that they are thirsty, they simply ignore that, or sometimes they do not know what to do.When people do not respect their ancestors, there is much violence.'(P13, female, 56 years old, igrirha: FGD) The detachment of people from their ancestors and from accepted societal interrelationships and correct behaviour is viewed not only as resulting in spiritual unease and violence but is also seen to be expressed in everyday behaviour and as general moral and physical malaise.As an igrirha said: 'People have forgotten traditional ways, for example, of dealing with sexuality.They have adopted Western ways and now there are problems.Things like when a child is born, it used to be introduced to the elders.Also, working men must give thanks to the ancestors and acknowledge them.But they don't do this.' (P14, female, 43 years old)

The violence of the occult
All of the participating traditional health practitioners expressed great concern about the practice of witchcraft, which is believed to be clandestine and malignant.According to the amagrirha, people are concerned that jealous neighbours or others are bewitching them in secret.Any adversity, such as the inability to find a job, illness, death in the family or trouble at work, could be viewed as the consequence of sorcery.When such difficulties arise and occult forces are suspected, the outcome involves affliction and even injury.Such malevolent harm and violence is caused by the meaningful actions of others.According to Tebbe (2006), commenting on the work of Ashforth (2005): For many Africans, spiritual harm constitutes an injustice in the same way as physical harm.They feel that the government ought to be protecting them from witchcraft in the same way that it punishes physical aggression using the criminal justice system.(Tebbe 2006) Traditional health practitioners perceived suffering from ilishwa (misfortune or bad luck) to be a serious malaise which can affect a person's health and quality of life or make someone very ill.It is a form of spiritual violence, as explained during a focus group discussion: 'Some jealous people may act against a progressive [affluent] family or an individual.To destabilise the family, a witch will send evil spirits to them.Maybe a boy is doing well at school, then the witch will be so jealous that he/she sends some evil spirits.Sometimes there is isichitho, something that is sent to destabilise the household.People in such a home will always be at each other's throats, and there will be fights.There are times when a person will inadvertently plant certain plants or shrubs that are known to be isichitho.Those plants are very dangerous and can break up families.'(P15, male, 40 years old, healer: FGD) As stressed above, the practice of the occult is thus viewed as being in itself a form of violence and all the traditional health practitioners stressed that they had a responsibility to establish whether a person had been afflicted because of it, to treat and remove it accordingly and subsequently to provide some form of protection against it.Whilst all amagrirha stressed that they did not get involved in any kind of witchcraft-related practices, they emphasised that the belief that someone had practised witchcraft could turn into violence against the accused.Witchcraft could also lead to violence or ill-health without the people involved realising what was happening (Ashforth 2005): 'Some people have bad luck.If someone hates you in the community she/he may cause these kinds of misfortunes.This is when some people have izothe.This happens when all people hate you to such an extent that people are angry when they see you.You are therefore prone to accidents and people can just attack you for no apparent reasons.' (P16, female, 69 years old, igrirha: Interview) 'Some people use black muthi for bad purposes.The medicine is not good or bad, your heart is bad and full of jealousy, [so] the medicine can be used to cause bad things, bad luck and sickness and violence for others.' (P17, female, 45 years old, healer: Interview) Since most suffering is believed to derive from negative and thus destabilising forces, treatment involves rituals to ward it off and to restore the equilibrium with other people, with the spirits and the ancestors.At the same time physical treatment might be needed (Hewson 1998).All of the above could potentially affect gender-related and sexual violence, which is explored next.

Gender and sexual violence
As indicated, the participants felt that a whole array of 'disconnections' had given rise to an ongoing communal experience of violence.They explained gender violence as being the manifestation of a dearth of 'tradition' and the intrusion of occult practices as described above, but were also keenly aware of the impact of socioeconomic and other factors, such as aggressive notions and practices of masculinity, excessive drinking by males, the relatively low status of women, improper policing, lack of resources and so forth.Gender violence was seen as being affected by changing values and practices, especially by changes in gender relations.Other types of physical violence were perceived as being equally on the rise.The traditional health practitioners had an almost apocalyptic view of the current state of society, as expressed in the following excerpt from a focus group discussion: 'They beat and kill their women.There are so many things in our communities these days that makes one think that the [entire] communities are bewitched.People these days do things you never thought they would be doing.They kill quite easily and rape their own [girl] children.I think we are approaching the end of the world.Things are not going right.' (P16, female, 69 years old, igrirha: Interview) Changes in wider social relationships, interlinked with changes in society at large, are perceived as having brought about shifts in cultural practices and beliefs, especially in gender relations and practices: 'One underlying reason behind violence is that women no longer respect their men.In the past, a man would beat a woman for whatever reason, and there was nothing big about it.Nowadays you will find that even slapping your wife, or punishing her for whatever she has done, will be treated as violence.'(P18, female, 38 years old, igrirha: Interview) At one level, the traditional health practitioner above was referring to the double standards that were prevalent in the past in relation to sexual relations for married men and women, but also indicating that such perceptions had not changed much since: 'Women are expected to respect their men.However, men take advantage of that and do things you would never think of.Our communities are full of men who take advantage of defenceless young women.' (P19, female, 38 years old, igrirha: FGD) The townships, especially the newer sections and informal settlements, are viewed as melting pots where social and spiritual relations are easily disturbed.This is because: 'Life in townships is about people who come from different areas of the land.They do not know each other.' (P20, female, 57 years old, igrirha: Interview) 'In the past there was more family support, and when people got into all kinds of trouble, their families would be supportive.Nowadays, things are different, the families are scattered all over the country and couples have to deal with their own problems.The parents and in-laws do not intervene when necessary anymore.' (P21, female, 36 years old, igrirha: Interview) 'Families are no longer like those in the past.In the past married couples lived in an extended kind of life where problems in the household would be shared and support and advice would come from all sections of the family.Nowadays people live in solitude and lack the support that was enjoyed by people in the past.That is where the problem lies.' (P22, male, 40 years old, igrirha: FGD) Young people were perceived by participating amagrirha as being particularly violent in sexual relationships.They ascribed it to a lack of parental control and discipline, increased involvement of young men in criminal activities, lack of sexual constraint of young men and women, young men's perceptions that they were sexually entitled and their practice of beating their girlfriends because they thought it made them seem strong and manly: 'Young men beat up their girlfriends and one girlfriend may be jealous of another and suspect her of jealousy or of causing trouble.It all makes [causes] more violence.'(P23, female, 39 years old, igrirha: Interview) Although male traditional health practitioners were more tolerant of the use of limited, but not excessive, physical violence against a married woman, all female traditional health practitioners mentioned that the laws in relation to domestic violence had changed and that they viewed domestic violence as being highly problematic.According to a female traditional health practitioner in a focus group discussion: 'Violence will remain violence, as long as it is directed at a person who is defenceless or who is unable to fight back.Although domestic violence has been treated as a light "fight" between members of the family, it remains a violation of one's rights.Sexual violence is the most serious of all forms of violence.'(P24, female, 46 years old, igrirha: FGD) Whilst the majority argued that most people would not necessarily perceive forced sexual intercourse within marriage as rape, especially when lobola was paid, they all perceived sexual violence against young women and children as being on the increase.It was even seen as being an apocalyptic sign that societal relations, as well as the connection with the spiritual world, had been harmed deeply.The participants all stressed the health risks in cases of rape (dlengulwa).The woman was not only abused physically but was also polluted by the act and had to be cleansed: 'When young women are raped, there is risk of getting pregnant and risk of HIV or of things like "drop" [STI] infection.What I advise is first to take care of those things.I usually give somebody plant medication to make sure they are not pregnant.I try to assess the amount of damage that is done to the victim.This is now how badly hurt she is, how troubled she is.In cases like this I like to include family, especially with counselling.I don't want to leave family behind.'(P25, female, 47 years old, igrirha: Interview) 'If somebody comes to you and they have been raped … I would take medicine and wash her with it: the leaves will take away the bad luck that is around her.The medicine will keep her out of trouble, it will protect her.' (P26, female, 43 years old, igrirha: FGD) Traditional health practitioners were most concerned about dealing with cases of sexual violence against children or young girls.One had been trained specifically as a counselor for rape survivors, but all stressed that they had a role to play in addressing violence in their communities: 'Healers can assist by counselling, [and] playing a supportive role.If people are hungry, give them food.Get involved in their lives.Try to make them feel better and happy.So they don't look so bleakly at the world.' (P27, female, 48 years old, igrirha: FGD)

Limitations of the study
Since the paper focuses on a small number of isiXhosaspeaking amagrirha in an urban area, the results of the study are not necessarily representative of traditional health practitioners in a wider context.The FGDs were larger than usual and whilst each participant was given an opportunity to speak, this was not an ideal situation.Language and translation, as well as the sensitivity of the topic, are all limiting factors that influence the research, even though the measurements described above were set in place to lessen this impact.Because of the small sample size, it was not possible to measure variations between amagrirha, for example, in relation to where they had been trained, their position within the hierarchies of traditional health practitioners as a group, gender, age or levels of education.These are all potential areas for future study.

Recommendations
With a restricted sample, only tentative recommendations could emerge from this research.Since there appears to be a strong link between poverty and violence for this group of participants, this issue should be studied more extensively by involving a more representative sample of traditional health practitioners.The authority accorded to traditional health practitioners within communities can be utilised in order to support and promote community interventions and campaigns against gender-based and sexual violence.There is also a need to train traditional health practitioners in gender-based violence prevention and interventions.The study focuses on understandings of violence but gives little indication of the ways in which traditional health practitioners deal with different forms of violence or the outcomes for their clients.Both issues require further investigation.

Conclusion
The traditional health practitioners in this study saw violence as being affected by an array of disconnections, with the detachment of people from their ancestors, societal interrelationships and correct behaviour being linked directly to spiritual unease, violence and general moral-and physical malaise.This is understood to be caused by ruptures within families and kinship relations, within communities, between rural-and urban beliefs and practices and, more ominously, by being dislocated from the ancestors and the spiritual world.The traditional health practitioners in the study attempted to detect such disruptions and deal with them individually and, if feasible, they involved the family members in the process, because it was important to restore individual balance as well as balance in the wider sense.This was made more difficult by the movement of people into areas where they did not know each other or were not necessarily related to each other and where there was less family support.
Violence is seen as resulting from a variety of breakdowns and disturbances in the proper arrangements of many kinds of social relationships.Structural issues, poverty, lack of employment, the disaffection of young people, a disregard for the law and for the authority of elders, all intersect at a variety of levels.In many ways, this situation heightens the propensity for violence and the likelihood of its occurring and spreading.Where there were particularly high concentrations of violence, attempts were made to resanctify the site or area by holding cleansing ceremonies.In addition to attending to the physical manifestations of illness, distress and violence, practitioners of traditional medicine strive to enhance and restore proper social relationships between the living, as well as between the living and the dead.Whilst the traditional health practitioners were able to deal with this and assist in restoring balance following an array of incidents of violence, they felt almost overwhelmed by what was perceived to be a deep, general and apocalyptic growth of its prevalence in an urban setting.