This chapter begins with a brief outline of the origins and development of vigilantism up to the early 1990s. Vigilante activity during this period is then contrasted to the post 1994 era when significant political changes occurred in South Africa. The reasons for vigilantism in the past decade are discussed. Finally, the chapter attempts to define vigilantism.

Vigilantism up to the early 1990s

African communities in South Africa have a long history of developing their own systems of policing, dispute resolution and punishing offensive behaviour. The South African government had as early as 1910 sought to control indigenous African systems of policing and justice, generally among African communities. This was done through the appointment of chiefs in rural areas and representatives of homelands in urban areas.4

In the early 1940s the state sanctioned the extension of civilian protection in the townships. The Civic Guard units for example, were established in several townships including Sophiatown, and were tolerated by the state.5 The units were banned in 1952 but due to continued public disorder and the lack of government protection, communities organised themselves against criminals.6

The origins of vigilantism in South Africa have also been linked to other community structures established in the 1970s under the concept of Makhotla.7 Although controversial because of their association with tribal or 'homeland' representatives in the townships, the purpose of Makhotla was to rebuild and care for the community. They functioned as alternative systems of township justice alongside the formal legal system.8 However, when they began transgressing boundaries of legal behaviour they were no longer tolerated by government or by the communities where they operated.9

At the same time as Makhotla were operating, the country began to experience political upheavals in the form of the Soweto riots and the period of resistance that was to follow. It is this period and the political context that shaped it, that some believe gave rise to vigilantism. According to Coleman, vigilante groups first made their appearance in South Africa around 1985 and have their origins in the support systems that were built up around the highly unpopular structures of homeland authorities and Black Local Authorities created under apartheid.10 During this period vigilante groups were, loosely explained, conservative groupings either formed by or supported by the state, to destroy resistance against apartheid.11

These arguments suggest that the emergence of vigilantism in both the 1970s and the 1980s was encouraged to support the aims of the apartheid government. Vigilantism became a form of social control over township residents who resisted the apartheid system, particularly the homeland system and the Black Local Authorities. Although these groups portrayed themselves as guardians of residents who aimed to "educate, serve, acquaint people with the requirements of the legal system, integrate and organise public opinion and develop areas", many township residents viewed them with suspicion.12

Examples of such groups include the Council for Brothers of Chiefs , Mamelodi Venda Urban Representative Board and Inkatha Yesizwe, which had close ties with unpopular state structures particularly the Black Local Authorities and so-called homeland leaders. Anti-apartheid activists believed these groups embraced the notion that African people belonged in the homelands and that they insisted on imposing apartheid laws on township residents. They have been explained as trying to educate people in the townships with the requirements imposed upon them as "visitors" to South Africa to ensure compliance with the formal legal system.13

During the township uprisings of the 1980s, structures such as the Black Local Authorities and the homelands came under increasing attack from the ANC-aligned United Democratic Front (UDF), youth and civic organisations. The homeland leaders were seen as puppets of the apartheid government, which inevitably created tensions between them and the liberation movements. Mangosuthu Buthelezi (former head of the KwaZulu homeland and leader of Inkatha Yesizwe), Oupa Qgozo (former head of the Ciskei) and Lucas Mangope (former head of Bophuthatswana) became embroiled in heavy conflicts with the UDF and its allies. These homeland leaders and the Black Local Authority councillors used a range of vigilante groups to destabilise communities and counter the attacks by the UDF aligned civic and youth groups (Table 1). Clashes involving these vigilante groups left thousands dead and others homeless.

Table 1: Vigilante groups who supported the former homeland leaders and Black Local Authorities

Province	Vigilante group
Mpumalanga (former KwaNdebele)	Mbokodo
Mpumalanga	Black Cats
Gautent	Toaster gang
KwaZulu-Natal (former KwaZulu)	Amasinyora and Amambutho
Western Cape	Amasolomozi, Witdoeke and Amadoda
Eastern Cape (mostly in the former Ciskei)	Kekanas, Mememsi and AmaAfrika
Free State	Three million gang, Mabangalala and A-Team (Phakathis)

Source: M Coleman, Crime Against Humanity, Analysing the Repression of the Apartheid State, David Philip, Cape Town, 1998.

Violence related to vigilante activity intensified during the 1980s and by October 1988, over 90% of unrest-related deaths were caused by vigilante and counter-vigilante violence in South Africa.14 In response to the ever-increasing numbers of attacks by vigilante groups and in an attempt to replace official state structures with ANC structures, many communities under the direction of their local civic associations formed street committees and people's courts.

Street committees were set up primarily to protect communities and to alert them to pending attacks from vigilante groups, particularly when these were aimed at leaders' homes. People's courts on the other hand were set up to try alleged police informers, criminal elements within the liberation movements commonly known as comtsotsis and common criminals. At times the people's courts also intervened in family disputes.

The forms of community justice that developed in response to the state sponsored vigilante groups as well as the lack of protection from official sources, led to the term vigilantism becoming synonymous with 'popular justice' in the early 1990s.15 In truth, by this stage many kinds of informal justice had begun operating, including irregularly constituted 'kangaroo courts', regularly constituted court structures, street committees or disciplinary committees, spontaneous street gatherings dispensing ad hoc and summary justice, and various informal policing and vigilante activities.16 The variation in types of informal justice operating during the 1990s probably encouraged the tendency to seek one all-encompassing term for these activities:

"These different forms of informal justice involve contrasting procedures and activities and different groups of people, but they are often conflated into the single category of people's courts." 17

In conclusion, the development of vigilantism up to the early 1990s was influenced by:

The repressive nature of the state during apartheid and the reactions and counter-reactions that this drew from various political and community based groups.



The fact that the state did not provide services, particularly in respect of safety and justice that the communities needed.



The fact that institutions of the state were inaccessible to local communities.18

After 1994: high crime and weak service delivery

In the post-1994 period vigilantism can be explained largely in terms of high crime levels, public perceptions that government is unable to respond, the poor delivery of services associated with safety and the inaccessibility of justice to most South Africans.

Two years after the 1994 elections, South Africa began to experience the re-emergence of community justice or vigilantism. Proponents of vigilantism argue that poor service delivery and the inaccessibility of the criminal justice are the key factors causing people to support these methods.19 South Africa is not unique in this regard.

These factors have been noted in other contexts as giving rise to vigilantism. In the UK for example, communities that believe the police have failed to deal with criminals advocate the "pursuit of criminal deviants, the righting of a criminal wrong by violent and informal means, the leaving of a warning…for others who might possess similar criminal dispositions." 20 In Northern Ireland, punishment, beatings and shootings of suspected criminals are believed to exist for three main reasons: the lack of a legitimate and adequate police service, the rising levels of petty crime, and the failure of the formal criminal justice system.21

In South Africa, supporters of Pagad, Mapogo-a-Mathamaga and Umfelandawonye—some known vigilante groups—all name weaknesses with the criminal justice system and the problem of police corruption as reasons for the establishment of these groups.22 A review of available indicators about crime levels, the performance of the criminal justice system, perceptions about safety and corruption, and access to justice indicate that many of these perceptions are accurate.

Crime trends23

According to official statistics of the SAPS, levels of recorded crime in South Africa began to increase in the mid-1980s—dramatically so in the early 1990s.24 Expectations that levels of violence would decrease after 1994 did not materialise. During the first three years after South Africa's political transition in 1994 overall crime levels stabilised, albeit at very high levels. The number of recorded crimes increased by 3% between 1994 and 1996. Over a similar period South Africa's population increased by an estimated 4%.

Since 1997 however, overall crime levels have increased steadily. While crime levels decreased fractionally between 1995 and 1996, and increased by only 1% between 1996-97, the increase was 5% in 1997-98, 7% in 1998-99 and 8% in 1999-2000.

Not all crimes increased or decreased at the same rate between 1994 and 2000. Overall, the 20 most serious crimes increased by 24% during this period. Residential burglary increased the most (by 33%). Assault with the intent to commit grievous bodily harm, robbery with aggravating circumstances and common assault all increased by more than the total. The number of recorded murders and car thefts decreased between 1994 and 2000.

Violent crimes are, given their nature and likelihood of attracting media coverage and public anger, of particular relevance to the issue of vigilantism. The statistics reveal significant trends in regard to the changing level of violent crime. Unlike overall recorded crime levels which decreased slightly in 1995-96, recorded violent crime has increased consistently since 1994.25 In 1994 some 618 000 incidents of violent crime were recorded, increasing to 825 501 incidents in 2000. Between 1999 and 2000 the number of recorded violent crimes increased by 10%—more than any other crime category. Between 1994 and 2000 violent crime increased by 34%. By comparison, property crime increased by 23% over the same period.

By global standards South Africa has high levels of violent crime. In 1999 a third of all crimes recorded by the police in South Africa were violent in nature. In the United States, which is considered to be a relatively violent society, 15% of recorded crimes were violent in that year, while about 6% of recorded crimes in the United Kingdom were violent in nature.

According to 1998 Interpol statistics, South Africa had the highest per capita rates of murder and rape, the second highest rate of 'robbery and violent theft', and the fourth highest rates of 'serious assault' and 'sexual offences', of the 110 countries whose crime levels are listed by Interpol.26 It is however worth noting that apart from murder, international comparisons of other crime categories can be misleading due to the variance in reporting tendencies from one country to the next, as well as in how crimes are recorded by law enforcement agencies. Nevertheless, the fact that South Africa's murder rate is the highest of all countries reporting to Interpol confirms the particular problem of violence that faces this country.

Indicators of criminal justice system performance

Measuring the performance of the criminal justice system is complex, particularly when certain statistics such as police arrest rates are not available. However several indicators, based on available data from the police, courts and prisons, indicate that the functioning of the system is weak in particular areas:

Many detectives are ill equipped to deal with their task and most are over-burdened. In 2000, the ratio of recorded crimes to detectives in the country was 118:1.27 Of the roughly 20 000 detectives in the SAPS at least a third have not received basic investigative training. This severely limits their ability not only to investigate cases to the point that an arrest can be made, but equally importantly, to provide follow-up to the victims about progress on their case. The provision of follow-up (contacting and informing the complainant about the state of their case and what to expect) has been identified in a recent survey as the most important complaint about police service delivery among people who reported a crime to the police.28



The number of cases resulting in a conviction as a proportion of the number of cases recorded by the police is low. In 1999 it ranged from just over 50% for drug related offences to 17% for murder, 8% for rape and 2% for car hijacking. This means that on average, only one out of every six recorded murders end in a conviction of the perpetrator. For rape the ratio is about one out of 13 and for car hijacking, one out of 63.29 This low conviction rate is probably the result of problems both with the quality of detective work and information provided in their case dockets, as it is with over burdened court rolls. Whatever the reasons for the low conviction rate, it strengthens the perception that the criminal justice system is weak.



Once the decision is made to institute a prosecution, the success rate is however high. On average, about three-quarters of all crimes that are prosecuted result in a conviction of the accused person. This is not surprising given that the prosecution service can choose those cases where there is a reasonable chance of securing a conviction.



The prosecution service has been taking on fewer and fewer cases. Although the number of crimes has increased over the past decade, the number of cases taken on by the prosecution service has declined over this period. In 1985/86 some 480 600 prosecutions and 373 980 convictions were recorded. This decreased to 257 390 prosecutions and 202 590 convictions in 1999.30 This decline is worrying when considering that the number of serious crimes recorded by the police increased by 481 427 between 1994 and 2000.



The increasing number of unsentenced prisoners in South Africa's prisons is another clear sign that the criminal justice system is struggling to process cases. The number of awaiting trial prisoners increased from 19 571 in June 1994 to 55 558 in December 2000—an increase of 184%. Over the same period the number of sentenced prisoners increased by 35%.31



The length of time unsentenced prisoners remain incarcerated until their case has been finalised also increased considerably between 1996 and 2000. In June 1996 the average time spent awaiting trial was 76 days. By December 2000 this had increased to 136 days.32



As a result of the increase in numbers of unsentenced prisoners, the prisons have become dangerously overcrowded. The South African prison system was designed to accommodate 100 668 inmates. Currently there are 172 271 prisoners behind bars. The problem of overcrowding varies by prison and by province. One of the worst cases is Johannesburg Medium A where 6 250 prisoners are kept in cells designed to hold 2 630. The most overcrowded prison in the country is the awaiting trial prison in Johannesburg, which is at 393% occupancy.33

These problems in the criminal justice system affect the relevant departments' ability to respond to the needs and expectations of victims who report crime, as well as the needs of victims and witnesses who must appear in court when their cases are being prosecuted. Under these circumstances, delivering a service to the public that can counter the fears and frustration brought about by rising crime levels is understandably difficult. The overburdened system thus fuels public perceptions that the criminal justice process is slow, complicated, favours the offender over the victim, and is skewed towards servicing the wealthier sections of society. These are all complaints that have been noted by supporters of vigilantism.

Perceptions about safety, policing and the criminal justice system

Several surveys provide evidence that public perceptions about personal safety and the ability of government to provide safety are low and have deteriorated since 1994.

The most comprehensive national survey covering these issues is that conducted annually by the Human Sciences Research Council (HSRC). These opinion polls, conducted since 1995, show a dramatic fall in the extent to which ordinary South Africans felt safe. In 1994, over two thirds of people felt safe (73%). This proportion dropped to its lowest level in 1998 when only 43% said the same. The decline has however since tapered off. In 1999 slightly more people (44%) said they felt safe. These trends must be seen in the context of the transition process:

"One of the most striking elements of the post-1994 South Africa is the way in which crime quickly emerged as a key governance issue. The hopes of the government and the South African public that the legitimacy of the first democratically elected government would serve to reduce crime rates, soon floundered…" 34

Although the more recent HSRC survey data (for November 1999) suggests that public perception might be improving, it is too soon to establish a clear trend in this regard.

The HSRC survey also asked people's opinion about the extent to which government had control over the crime situation. Less than one in ten people in South Africa (9%) believed that government had "full control". Nearly half (49%) said government had some control and 35% thought government was "not in control". The remaining 7% of the sample said they did not know.

Similar views were expressed by respondents in the national victims of crime survey conducted in South Africa in 1998. The survey showed that only 38% of all respondents (victims and non-victims) were satisfied with the police, 40% were not satisfied and 23% were neither satisfied nor dissatisfied.35 A more recent survey conducted by the ISS of 13 659 respondents in 45 police station areas across the country, found that nearly half (47%) of people said the quality of policing in their area had deteriorated over the past four years. A third (32%) said there had been no change and 21% said policing had improved.36

Other localised surveys on vigilantism and attitudes to punishment produced similar results. A survey about support for Pagad conducted by Idasa in 1996 in the Western Cape showed that 55% of all respondents disapproved of overall police performance. When victims who had laid a charge with the police were asked their opinion, 68% felt that overall police performance was poor. The police were not the only government agency whose service respondents were dissatisfied with. When asked about their views of the justice system, 65% said the overall performance of the courts was poor.37

In a study on attitudes to punishment conducted by the ISS in the Eastern Cape in 1999, nearly half of all respondents (47%) thought there had been no improvement in the functioning of the criminal justice system since 1994.38 A quarter believed the criminal justice system was performing well, while 50% stated that it was not. When asked whether they agreed with the statement that "the government has done a good job in fighting crime and lawlessness", 25% said they did, while the majority (58%) said they disagreed. In response to all these questions, black respondents were more inclined to respond positively than were those from other race groups.

It is however encouraging that according to the national HRSC survey, the perceptions recorded in late 1999 (noted above) were an improvement on the results for 1998. In 1998, 5% of South Africans said government had full control of the crime situation, 41% said it had some control and 51% said government was not in control. In addition the 1998 HSRC survey found that crime was identified as the major national priority for the country. In 1999, crime was replaced in this position by job creation.39

This improvement in the public assessment of government's ability to respond to crime could be attributed to the high profile and tough talking new Minister of Safety and Security, Steve Tshwete, who took up office at the beginning of 1999. While these perceptions are undoubtedly positive, public spokespersons should guard against leading the public to expect a quick fix to the high crime rate.40 If raised expectations are met by continued high crime levels, the results may be similar to those recorded after the euphoria of 1994 had worn off—despite what the politicians say.

Although several surveys, such as those described above, show that perceptions of the police and courts' performance is poor, few studies show a direct link between these perceptions on the one hand, and support for vigilantism, on the other. The one exception is the survey about support for Pagad conducted by Idasa in 1996 in the Western Cape.

The survey sought to understand why people in the Western Cape supported Pagad, and why they would join the organization. Using multiple regression, the analysis showed that "the most important factors that distinguished those who would support or join radical collective action against crime [through use of intimidation or violence against criminals]", were:

Fundamentalist values: those who believed in severe punishment and that society should punish wrongdoers that courts did not, and religious wars.



Gender: men were more willing to participate in citizen action than women.



Perceptions of safety: those who felt less safe at home or in their neighbourhood were more willing.



Legal values: those who were less supportive of the rule of law and the rights of the accused.



Religion: Muslims were more likely to get involved.



Perceptions of treatment by the police and courts: those who felt they were treated "unequally" by the police and courts.



Performance ratings of the police and courts: those who disapproved of the overall and specific job performance of the law enforcement institutions.



Age: younger people more likely than older people to become involved.41

However, the study found that most of these same factors also explained why people would participate in or support "peaceful methods of collective action" (our emphasis) such as signing petitions, boycotting businesses and marching on the local police station. The similarity in the factors that characterized those who would support and participate in radical action, with those who supported peaceful action suggests that this list of factors explains the propensity to engage in collective action, per se. However it does not explain why people support violent collective action, or vigilantism. In this regard, it is telling that the Idasa survey showed that the only two factors of the list above that did not characterize those who said they would support peaceful action were 'perceptions of safety' and 'performance ratings of the police and courts'.42

This suggests, as is argued throughout this monograph, that the key reasons why people support violent collective action against criminals are that they believe the police and courts are performing poorly and that they feel unsafe. Other data from the Pagad survey supported this. Respondents who were dissatisfied with the crime situation and those who disapproved of the performance of the police and courts, were more likely to say they would join Pagad:

28% of the respondents who were dissatisfied with crime said they would join Pagad, compared to 17% who were satisfied with the crime situation.



33% of those who disapproved of the police's performance said they would join Pagad, compared to 17% of those who approved.



30% of those who disapproved of the courts performance said they would join Pagad, compared to 15% of those who approved.43

Although the ISS survey on attitudes to punishment in the Eastern Cape did not directly link perceptions about criminal justice to support for vigilante action in the way that Idasa's Pagad survey did, it provided useful indications about why people say they support harsh punishment—one of the defining characteristics of vigilante activity.

The ISS attitudes to punishment survey found substantial support for non-formal methods of sanctioning criminals: just under half of the total sample and as many as 75% of rural respondents supported alternative or traditional forms of punishment.44 Most respondents were also in favour of harsh punishment. Three quarters thought the death penalty should be reintroduced for persons convicted of serious crimes, and just over half (51%) said the same about corporal punishment. However when presented with brief descriptions of a number of crimes and asked to impose a sentence on the perpetrators, respondents suggested sentences that were more lenient than those that are currently on the statute books under the minimum sentencing legislation (Figure 1). Respondents were particularly lenient when it came to punishing people who had been convicted of vigilante crimes. When asked what kind of sentence a person deserved who had murdered his daughter's rapist, 5% said no punishment at all and 38% chose a non-custodial sentence.

"Without corporal punishment, our government will never stop crime in South Africa." Montle Magolego, president of Mapogo, ISS seminar 8 June 2001.

Thus, although most respondents supported harsher punishment, when faced with more detailed scenarios of actual crimes, they were considerably less draconian. This questions the extent to which public demands for harsher punishment (and thus possibly vigilantism as well) are indeed about punishment and a rejection of the 'human rights culture' which is believed to afford criminals more rights than victims. It is possible that the calls for punishment are related to the need not only for swifter justice but also for more clarity on the process of criminal justice.

The survey indicated a clear need for the right kind of information about the criminal justice process, how it works and what people should expect from it (see the section on 'access to justice' below). It also showed—contrary to popular belief—that most respondents thought it was important that the criminal justice system respect the constitutional rights of accused persons. Three quarters thought it was important that the justice process ensures that no innocent people are convicted of crimes, even if this means some guilty people will go free. Almost two thirds said that juveniles should be treated more leniently than adults by the criminal justice system and most believed that convicted offenders should be treated fairly.45

Source: ISS survey on attitudes to punishment in the Eastern Cape, 1999

Furthermore even when people said they support harsh punishment and "alternative justice" they were unlikely to be willing to participate directly in punishing offenders themselves. For example, in the ISS attitudes to punishment survey, only 5% of respondents indicated that they had participated in vigilante activity. A further 7% said that members of their household or their friends had actually participated in vigilante activity.46

Perceptions about corruption

One of the complaints raised by proponents of vigilantism in relation to the poor performance of the criminal justice system was corruption. Survey evidence shows that corruption among government officials is widely regarded as a problem by the public. In the HSRC's November 1999 countrywide opinion poll over half (53%) of respondents said they thought corruption in state institutions was increasing.47 Just less than a quarter (22%) thought it was decreasing and 25% said they did not know. White (84%) and Indian respondents (81%) were much more likely to hold this view than coloureds (63%) and Africans (45%). The percentages who thought corruption was increasing in government were nevertheless high across race groups.

When asked whether they thought government was doing enough about the problem of corruption, respondents were most likely to believe that the priority government gives to fighting corruption is "sufficient" (34%). However, almost as many said the priority was too low (30%). Interestingly 17% said the priority given was "too high" and 19% were unsure. More importantly, most respondents were unaware that the actions of public officials are regulated and bound by a code of conduct: 61% said they were unaware of government's anti-corruption efforts in this regard. It is possible that if more people knew that government officials are bound by a code of conduct, they would assess the level of corruption and government's efforts to combat it, more critically.

Public concern about police corruption was confirmed in the ISS survey on community policing and police service delivery conducted in police stations across the country. When asked why policing had deteriorated, the most common response (by 29% of people) was that corruption was the cause.48 This is significant since in other similar surveys conducted between 1997 and 1998 in the country's major metropolitan areas, corruption ranked fifth or lower as the reason for dissatisfaction with policing.49

These perceptions are not restricted to the police. When asked their views on the involvement of police and justice officials in corruption, 75% of respondents in Idasa's Pagad survey in the Western Cape said there is at least some level of corruption in the police. More than half (52%) felt that at least some judges and prosecutors were corrupt.50

The available evidence about the extent of police corruption and police involvement in criminality generally, suggests that the problem is real and not just one of perception. Between 1994 and 1997 an average of 13 954 complaints or charges per annum were laid against SAPS members. Figures from 1998 to 2000 were not available. Based on the 14 600 cases investigated by the SAPS against its members during 2000, indications are that the number of complaints may be decreasing.51

It is however difficult to tell the extent to which these figures are influenced by reporting or recording practices, or budgetary changes in the police. The Independent Complaints Directorate for example, recorded a substantial increase in the number of complaints received against SAPS members in the financial year April 1999 to March 2000. It received 4 380 complaints during that period, representing a 52% increase in complaints compared to the previous year.52 Of the complaints received in the last financial year, the largest proportion (38%) were for misconduct. Over a quarter (28%) fell outside the ICD's mandate, 17% were for serious criminal offences, 15% were deaths in police custody and the remaining 2% were referrals from the Minister of Safety and Security or the MEC's in the provinces.

Access to justice

The perceptions of criminal justice are compounded by poor access to justice. Access to justice is impeded both by physical and resource related constraints which make the police and courts inaccessible, as well as the reduced participation in criminal justice processes as a result of a lack of understanding about the complex legal system.

Access to the police in South Africa is affected by the various challenges facing the SAPS, such as limited resources, an overly centralised and bureaucratic hierarchy, a lack of appropriate skills and training, a shortage of managerial expertise and a limited intelligence and investigative capacity.53 These problems are exacerbated by South Africa's history of racially biased policing, which meant that in 1994, 74% of the country's police stations were situated in the white suburbs or business districts.54 Although this imbalance is being addressed, the problem still limits the access of the majority of the population to the police.

These limitations make it difficult for the SAPS not only to detect crime but also to provide proactive policing, particularly in the form of patrols and other methods that increase their visibility to the public. The visible presence of police officers increases public feelings of safety and levels of confidence in the police.55 It can also help to deter particular types of crime such as opportunistic street and property crimes. These are key areas of policing that are perceived by supporters of vigilante activity as inadequate.

"This [visible policing] is about restoring confidence in our communities. We need to create the perception in the minds of the people that the police care and with that we can help in ensuring a safe environment. How can we do this if we cannot be visible?" Assistant commissioner, SAPS Durban, October 1999.

The lack of access to the police is compounded in the rural areas of the country where more than 18 million people, who make up 46% of South Africa's population, live.56 The geographic isolation and inaccessible topography of many of these areas, the lack of infrastructure and the resource and capacity constraints of the police as well as those they are meant to serve, all play a role (see box below).

Problems accessing the police: the case of Tugela Ferry The town of Tugela Ferry is situated on the banks of the Tugela river in the remote hilly region of central KwaZulu Natal. The tiny town serves a population of some 1.5 million people located in ten villages scattered across about 1 400 km2. The population is very poor, and most of the economically active people are unemployed. The site of substantial political violence from the mid-1890s to the early 1990s, the area is notorious for its violence and is still troubled by taxi violence, stock theft, murder, armed robbery and gun smuggling. Set to deal with these problems are only 30 police officers. This provides a police presence of one officer per 50 000 residents spread over 47 km2. Once the shift system, vacation, sick leave and absenteeism are factored in, the actual operational policing capacity is closer to one police officer per 75 000 residents. According to the SAPS own assessments, Tugela Ferry police station is 50% under-resourced. The station has only 10 vehicles, most of which are inappropriate for the mountainous terrain and just four officers have a driver's licence. In practice this means that one of these officers needs to be on duty at all times which affects standby and rest periods. It is not surprising that the station battles with morale and discipline issues evidenced by absenteeism and alcohol abuse among the members. E Pelser, A Louw and S Ntuli, Poor safety: crime and policing in South Africa's rural areas, ISS monograph No 47, May 2000.

According to a survey of crime and policing in deep rural areas conducted by the ISS in 1998, respondents were most likely to say they never saw the police on duty in their area: 34% said this was the case, while 32% saw a police officer less than once a month, 18% saw an officer more than once a month, 10% said at least once a week and only 7% saw a police officer once a day.57 The limited police presence was compounded by a general lack of infrastructure in these areas which inhibits access to police services. For example, only 9% of people said they have a telephone in their home, and only 7% said their telephone was in working order. This means that the vast majority of people in rural areas can only access the police by traveling to the nearest police station. However most of the respondents (64%) lived more than 11kms from the nearest police station, and a further 36% lived within 10kms from a police station.

Thus, the remoteness of some of the police stations and the lack of telephones and personal transportation meant that for many in the rural areas, accessing police services can be a long and expensive exercise.

Access to courts is limited by similar factors as those described above. Long distances and the costs involved in reaching a court have led many victims and witnesses to stop attending cases, with the result that the charges are dropped or cases are withdrawn in court. This in turn means that justice is not seen to be done, particularly for the poor and those in rural areas. These problems have been identified in a report by the South Africa Law Commission dealing with the issue of community courts.58

In addition to logistical problems that make reaching a court difficult, the legal system as a whole, its operation and the language used by the courts, have often proved to be obstacles:

"The inability to meet the needs of the ordinary citizens was however not merely due to content of the law, but also because the structure and procedural requirements of the courts meant that many people were denied access to the courts."59

This has the effect of alienating people from participating in the proceedings, not least because they rely on translations.

The new constitution and the array of new laws that have been drafted since 1994 have also left many confused. For example victims, witnesses and the public in general often do not understand why an accused person can be granted bail for an offence that is regarded as serious. Many feel that criminals are afforded too many rights. In Idasa's survey of Pagad in the Western Cape, 57% of the respondents felt that suspected criminals should not have the opportunity to be released on bail. More than two thirds (68%) thought that the police should be allowed to hold suspects until there is sufficient evidence to charge them.60

The ISS attitudes to punishment survey in the Eastern Cape similarly found that many respondents were ignorant of the role and purpose of the criminal justice system, and the different functions of its component parts (police, justice and prisons). There was for example confusion about who should mete out punishment. Only 59% of respondents thought that magistrates or judges should be responsible for deciding on punishment.61 As many as 28% said that the police should be responsible and 10% believed the community should mete out punishment. This situation limits people's interaction with the criminal justice system:

"If people are to have confidence in the criminal justice system, they need to have at least a rudimentary understanding of how the system works. Moreover, if people are to co-operate with the criminal justice system, they need to have the confidence for example, to provide the police and prosecution service with information and evidence against criminals and their activities. Such confidence can be fostered only by telling the public what the role and duties are of, for example, police officers and prosecutors." 62

Given that the police are usually the first point of contact between the public and the criminal justice system, the SAPS has an important role to play in providing victims and witnesses with information about how the system works.

The police in the charge office have an obligation to provide information to complainants who report a crime about the likely procedures that will follow. However, it is the detectives, more than the officers in the charge office, who provide the link between complainants and the rest of the criminal justice process. They will need to inform people about the progress of their case, the anticipated court processes and the outcome of their case.

"The problem we have here is that our members often do not give feedback to complainants." Station com-missioner, Nebo police station, Northern Province, June 2000. "Very little, if any information on the manner in which police are dealing with the reported crime is provided to the victim." E Pelser, A Louw and S Ntuli, Poor safety: crime and policing in South Africa's rural areas, ISS monograph No 47, May 2000.

This is referred to as the 'follow up process' and has been identified as one of the weakest areas in the delivery of police service to the public. According to the ISS' community policing and service delivery survey conducted for the SAPS in 2000, the key service requirement after reporting was that the police provide complainants with information about the status of their case.

This is also one of the main complaints about policing among those who support vigilantism. In the majority of interviews conducted for this monograph, people said they would consider reporting crimes to vigilante groups rather than the police because of the perception that once cases are reported to the police, victims receive no further information about the matter.

What is vigilantism?

"Vigilantes was the name given to self-appointed law enforcement groups which appeared…occasionally in older communities where law officers and courts were non-existent, inefficient or corrupt; where municipal institutions were disorganised; or where established authorities seemed unable to cope with lawlessness and disorder." Encyclopedia Americana, 1985. "Six elements that characterise vigilante activity are: Minimal planning, preparation or premeditation. Private agents acting in a voluntary capacity. Activity undertaken without the state's authority or support. Force is either used or threatened. A reaction to the real or perceived transgression of institutionalized norms. Aims to offer people the assurance that established order will prevail." L Johnston, What is Vigilantism?, British Journal of Criminology Volume 36, no 2, Spring 1996.

Various definitions for vigilantism have been posited—the box above provides just two. All of the elements noted in the definitions above apply to various of the vigilante groups currently operating in South Africa. Mapogo-a-Mathamga however presents an interesting exception in terms of one of the criteria—its services are not voluntary. For the purposes of this monograph, three main characteristics are regarded as important for an understanding of vigilantism:

Acts of severe violence including serious assault and murder of alleged criminals

For example, respondents in the ISS attitudes to punishment survey in the Eastern Cape said that vigilante activity in the areas where they lived was more likely to result in violence than in peaceful resolution of the conflict. In the majority of cases, the victims of vigilante activity were beaten (55%), killed or shot (18%). In 10% of the cases, the victims were banished from the area, or forced out of the village.63

A group of men allegedly attacked a number of homes in the township in Potgietersrus, claiming they were looking for various crimes (sic). The men allegedly kidnapped eight youths, who were assaulted with rifle butts and knobkierries…one of them, a schoolboy George Mabotje (19) was declared dead on arrival [at the hospital after being assaulted]. The Star, 29 July, 1999.

Punishment that often exceeds the crime allegedly committed and is meted out in the absence of any form of evidence

One of the defining features of vigilantism currently is its often harsh and swift punishment. Vigilante groups do not hesitate to punish a person severely, even in the absence of evidence. In fact punishment is often used to extract confessions.

"In February 2001, three youths were hacked to death when they tried to break into a church building in Pimville to steal chairs. They were caught after a member of the community raised an alarm." Sowetan, 21 February, 2001.

This is well illustrated by the Mapogo-a-Mathamaga's case in the Northern Province. Mapogo claims that its success in the fight against crime is due to administering sethlare (medicine) to criminals in the form of whipping (see box below). "Crime is out of hand, the government cannot cope, corporal punishment is the best approach" stresses the leader of Mapogo.64

Mapogo-a-Mathamaga's call for harsh and swift punishment is not unique to South Africa, but a characteristic of paramilitaries in the United Kingdom and lynch mobs in South America. For example, a study conducted in Brazil in 1980 by the Brazilian Institute of Public Opinion showed that 44% of the respondents supported lynching on the grounds that "if the courts do not act, the people must." 65

We find a friend of the suspect and plead with him to tell us where his friend is or the loot is. If he refuses, we give him a taste of the medicine. After tasting the medicine he will confess. Then we give him more lashes for lying or trying to hide a friend. When we find the friend and he confesses to the crime and hands over the loot, we give him lashes too." Thomas Moeng, Diepsloot branch of Mapogo-a-Mathamaga, February 2001.

Engaging in illegal acts such as kidnappings, crimen injuria, malicious damage to property, theft, robbery and sabotage

Between 1996 and 2000, many criminal charges were laid against members of Mapogo: 139 for common assault, 82 for serious assault, 23 for kidnapping, 19 for attempted murder, 13 for murder, 9 for theft and 6 for robbery. Other charges included crimen injuria, malicious damage to property, intimidation and public violence.66