The poor, as has been argued earlier, are more likely to be policed by the public police in a reactive manner than are the wealthier classes who, in South Africa, are increasingly more likely to be policed preventively by private security.



The problems of criminality in poor communities can often be traced to socio-economic circumstances which can be altered by interventions quite apart from policing. Yet the response of state institutions is more likely to be reactive in such cases rather than seeking innovative and preventive policy interventions.



Preventive strategies, given the arguments around the poor's limited ability to cushion themselves from the costs of crime, are also more appealing. No matter how efficient the system of criminal justice, it will do little in the form of restitution for victims of crime and those they support. For the poor, then, who bear the burden of crime, successful crime prevention is critically important.

Without an adequate focus on crime prevention, the state institutions of criminal justice will remain overburdened. More seriously, state policy towards crime control will hinge more heavily on the reactive as opposed to the preventive sphere. In this context, prevention refers to those activities that are intended to reduce or prevent the occurrence of specific types of crime (or the fear of them), either by altering the environment in which they occur or by intervening more broadly to alter the social or other conditions which are thought to have causal significance in regard to them.155

Most criminal events have many possible causes: some are remote or 'distal' (for example, abuse in childhood producing violent assaults in adulthood); others may be structural changes (the introduction, for example, of electronic banking) which create new opportunities for crime. Importantly, as Paul Ekblom notes, some causes are 'proximal': 'the presence of a motivated offender in a suitable crime situation immediately before the occurrence of the event'. It is at the point of these 'proximal' circumstances that the diverse structural, social, ecological and psychological causes of the criminal event inevitably converge.156 And it is here that crime prevention interventions are, in the short to medium term, likely to be most effective, and indeed most measurable.

The diagram157 overleaf illustrates the causal chain of a criminal event. The 'programmed potential' of the individual refers to past life circumstances which may impact on the likelihood of committing an offence. 'Current state' refers to the position of the potential offender at the time, such as being under the influence of alcohol or being in (or outside of) the company of particular people. 'Modulators' encompass a range of distinctive factors, such as the possibility of intervention while the offence is being committed, the degree to which the offender believes there will be a follow-up after the crime, and circumstances (such as leaving a car door unlocked) which may make the offence easier to commit. The 'environment' describes all those factors relating to the physical context in which an event occurs. These generally make it easier for an offence to be committed (such as visibility or the lack thereof). In sum, the commission of a crime reflects the conjunction of:

a ready, willing and able offender;



a vulnerable, attractive or provocative target;



a favourable environment; and



the absence of willing, able and credible modulators.

If the above model is accepted, preventive intervention would include a range of actions which aim to influence the chain of events at any one (or all) of these levels. Indeed, comparative research during the past decade suggests that single measures are rarely sufficient to reduce crime, criminality and fear of crime in high crime neighbourhoods.158 Crime prevention initiatives which rely on a general response to a general problem have been shown elsewhere to make little difference to levels of criminality. The most viable crime prevention strategies are those which concentrate on a particular crime (or group of crimes), and aim to put in place a customised, comprehensive programme of preventive measures.

THE PRACTICE OF PREVENTION

Crime prevention strategies therefore require a range of multifaceted programmes which aim to combat and prevent a single offence or category of offences. That means that in Northern Cape, for example, a multifaceted strategy is required to tackle a single crime such as rape. This requires programmes which consider education, alcohol distribution, environmental factors (such as lighting or the positions of shebeens in relation to schools), victim support, and finally policing (to ensure that there are regular patrols in high crime areas, and that laws relating to alcohol consumption are enforced).

Yet provincial crime prevention strategies have often failed to take on board this programmatic approach. They are more likely to be statements of intent than a vigorous array of programmes. And police strategies in relation to domestic violence or rape are seldom measured in the same basket as victim support and prevention strategies. The result is that, given that they are not assessed together, real co-ordination between departments seldom moves beyond the committee room.

In Northern Cape it has been shown that the problem of criminality is not necessarily one of too few police but relates, among other issues, to addressing high levels of alcohol consumption and gaining a better understanding of interpersonal violence.

In terms of the former, prevention will need to include enforcing controls, such as the issuing of liquor licenses to bottle stores situated near schools and the selling of alcohol to minors, as well as more far-reaching strategies. Interventions which treat demand must be balanced with those that police and legislate consumption.159 Several alcohol consumption patterns reported by offenders surveyed in prison could, for example, be modified through law enforcement measures: risky drug and alcohol consumption (taken in high quantities) was found to be high in the period before being arrested;160 distilled spirits (which have the highest absolute alcohol content) were not only found to be quite popular, but were consumed often and in large quantities; drug-taking was commonly a group activity occurring mostly with friends (sometimes gang members) and at places of traders and drug dealers such as shebeens and taverns; and offenders generally reported having easy access to drugs and alcohol.161

Alone, however, enforcing laws and controls cannot prevent abuse. And simply policing crimes such as public drunkenness, for example, may do little more than overburden the criminal justice system. Other drinking and drug-taking patterns reported by offenders require education and development-oriented strategies aimed at ameliorating the social conditions underlying drug abuse. 'Morning' drinking – indicative of long-term and heavy alcohol use – was common; offenders lived in environments conducive to and accommodating of drug abuse in which use by friends and family often influenced initial consumption; and direct social pressure to use alcohol was quite common. Added to these environmental pressures, there was a 'positive orientation towards drug use': offenders did not expect discrimination against such use and were personally attracted to the habit, believing in the rewards which it would bring.162

These kinds of strategies also need to be considered within the criminal justice system. Evidence suggests that involving offenders in community-based or corrections-based treatment can 'reduce drug use, criminal behaviour and recidivism, even when the (drug users) ... do not participate in the programme voluntarily'.163

Government responses to alcohol abuse in the past have fallen far short of achieving this balance, focusing on control measures which are largely the responsibility of the departments of Police, Justice, and Trade and Industry. The treatment of alcohol problems has been inadequately funded, and facilities have tended to be concentrated in large urban areas, with major disparities in resources spent on the different races under the apartheid government. Departments of Health and Education have hardly been involved, and few resources have been spent on prevention.164

While there have been some advances in policy-making with regard to alcohol and other drugs, interventions which directly address substance abuse have been found to be ad hoc and fragmented. Problems include a lack of political commitment nationally to address substance abuse and to provide resources for policy implementation, an absence of an intersectoral national drug control strategy, a failure to involve local structures in the process, and a lack of leadership at provincial and national levels to drive the strategy.165

Prevention should aim to shape the initial intake of drugs and alcohol, detect abuse and the risk of abuse, and act on this through treatment and controls. This requires a focus on the individual, on the environment in which the individual lives, and on the use of drugs.166 While initiatives are beginning to be formulated to address these problems, they remain fragmented and unco-ordinated, seldom rely on rigorous policy analysis, and lack an evaluative component.

As it is, the focus on crime prevention more generally has been undermined by the urgent need to reform South Africa's system of criminal justice. The system, stretching across the departments of Safety and Security, Justice and Correctional Services, has never been a unified one, and is characterised by multiple blockages causing delays throughout the 'pipeline'. Links between various departments are weak and the involvement of departments such as Welfare, Education and Health – key role players in crime prevention – is minimal.

Within the police, reform has concentrated on the visible component of the service and on community policing. And while processes of transformation here are far from complete, policy-makers have virtually ignored the issue of detecting crime. This is part of the reason for South Africa's low conviction rates, but most critical is the lack of co-ordination and co-operation between detectives and public prosecutors. Also, many public prosecutors are inexperienced, and magistrates' courts are badly managed.

At the end of the criminal justice pipeline, South Africa's prisons are also in dire need of reform. Staff shortages, prisoner and warder unrest and increasing corruption are bringing the crisis to a head. Rigid sentencing policies also contribute to high prisoner numbers and overcrowding, but again, part of the problem relates to other elements along the pipeline. About one quarter of all the country's 130 000 inmates are awaiting trial.167

The control of parole by the Department of Correctional Services provides an invitation for bribery and an easy (but inappropriate) mechanism for releasing pressure on the system. But probably the clearest sign that the system is failing is the fact that more than half of those imprisoned will again commit crimes upon their release. Rehabilitation is practically non-existent and the likelihood of improvements is slim, given that new budgetary allocations will be for more prisons and the staff to guard them.168

The failure of the criminal justice system to deter or punish offenders has been marked by a growing loss of confidence in the police and courts, and a concomitant tendency among people to take the law into their own hands. Forms of alternative protection vary: the wealthier sections of society can afford to buy their safety from the private security sector; the poor have to confront the problem more directly, often by taking action themselves. Neither of these security options necessarily alleviates the burden on the public police. Indeed, as was argued earlier, the growth of the private security industry may encourage uneven policing, to the disadvantage of the country's poor.

PREVENTION IN PRACTICE

The weaknesses of the criminal justice system have not escaped the government: the National Crime Prevention Strategy (NCPS) released in 1996 aims to unify and co-ordinate the activities of the departments involved in crime control and prevention.169

Six core government departments – Correctional Services, Defence, Intelligence, Justice, Safety and Security, and Welfare – were involved in developing a more holistic approach to crime prevention. At a different level, but in keeping with these goals, the 'community policing' concept (which is barely mentioned in the NCPS document) is superseded by the concept of 'crime prevention', entailing the building of partnerships between government agencies and outside organisations and civil society. Despite this shift in the NCPS, community policing is still a priority for the police. The change of emphasis may, however, reduce pressure on the police – emanating from those responsible for the NCPS's implementation – to pursue community policing.

The NCPS's greatest strength, its inclusive and comprehensive nature, is also potentially its greatest weakness. The key to its success will be co-ordination and measurement, something government has shown little sign of tackling. While the document allows for monitoring at departmental and programme level, there is no national means of review, and virtually no time frames for implementation. A substantial literature is being developed elsewhere for the measurement of crime prevention and control initiatives, and comparative experience suggests that unless crime prevention initiatives are properly evaluated, the longer-term success of any programme will be jeopardised.170

At the outset, the focus of the NCPS is in fact more reactive than preventive, and contains several oversights. For example, it expresses a commitment to victims, particularly women and children, but apart from acknowledging that crimes against women require a special focus the document says little else. Initiatives taken by government and the police so far – such as installing one-way glass screens for identifying perpetrators, and 'victim packages' for women to use after medical examinations – have been ad hoc.171 The police are the first point of contact with the criminal justice system, and as such play a crucial role in determining the success of initiatives aimed at responding to violence against women.

As is the case with reforming the juvenile justice system (see below), much remains to be done at a national level to reform the state's response to the problem of domestic violence. The police have yet to formulate co-ordinated policies and procedures for dealing with women who are victims of violence, and particularly domestic violence. Several special police units have been created in this regard, but the SAPS's response has largely been inadequate.172 Domestic violence is not regarded as a serious crime, and South African criminal law does not enable people to lay charges of specific violations in this regard.

This is evidenced in a recent SAPS Police Plan (1996/97) in which crimes against property are prioritised above crimes against people. In fact, the only reference to any aspect of violence against women is a passing reference to loss of life as a result of 'violent crimes, domestic violence and witchcraft'.173 Also, within the police no specialised units deal with crimes such as rape (despite the country's high rates), with property crimes being singled out for special attention instead.

Training the police to deal more appropriately with crimes against women has been identified as important, and several initiatives have begun in this regard. But a review of these programmes has found them to be largely inadequate. The very low attendance rates under one voluntary programme reflects the negative attitude of individual police stations or station commanders towards gender sensitivity training.174 Police often regard handling women victims as better suited to female police officers, but in reality few charge offices and flying squads are staffed by women.

Even with improved police training, problems currently reported at the local level will further frustrate responses to gender violence. The new emphasis of the police on an 'integrated and multi-agency approach' which involves civil society and NGOs in the field has not had visible effects. And there is also a lack of co-ordination between welfare services and the police, with police responses to victims often being frustrated by bureaucratic procedures.175

Apart from weaknesses in relation to specific issues such as domestic violence, those prevention programmes which are contained in the NCPS, of which there are 18, are nationally driven and are far from being implemented – in the short to medium term they will have little effect on crime levels. The programmes are divided into four pillars: criminal justice process, community values and education, environmental design, and transnational crime.

Thus, under Pillar 1, the NCPS contains a large number of programmes which aim to repair and reform the operation of the reactive components of the criminal justice system. For example, one key pillar of the NCPS, which deals with repairing the blockages in the criminal justice system, concentrates largely on improving the efficiency of the state's response once a crime has already been committed. A number of business plans have been completed and strategies devised to, for example, improve court management, the handling of offenders between prisons and courts, and the management of convicted criminals in prisons and places of safety.



Pillar 2, which focuses on community values and education, aims at increasing community pressure and public participation in crime prevention. While the aims of the programme are laudable, they have little impact on the poor. Interventions which, to date, have been conducted through the labour movement and co-ordinated at the National Economic, Development and Labour Council (Nedlac) have pulled in only those in full employment in the urban sector, and have had little contact with poor rural people.



Under Pillar 3, environmental design, little work has been done. Programmes are mostly to be led by the Department of Safety and Security, involving a broad spectrum of public and private sector role players such as the departments of Sport and Recreation, Welfare, Housing, and Home Affairs, as well as architects, town planners and development agencies. The danger is that programmes to 'design out crime' will be seen as a panacea when comparative experience suggests that this approach may only have a limited effect on crime levels. Importantly also, environmental design strategies are more likely to be successful in more developed urban areas (where the wealthy live) than in underdeveloped areas where there is little infrastructure around which to formulate policies. It is also possible that environmental design strategies will encourage inappropriate physical design measures which will encourage greater social distance between the rich and the poor – walled suburban complexes and fenced-off suburbs are just two illustrations of this. These measures may also result in the displacement of crime to other, less protected areas (often where poorer people live) rather than its prevention. In so far as these strategies work better in wealthier areas, the risks for people living in lower income neighbourhoods are more likely to increase.



Pillar 4 looks at countering the question of organised criminal activity. This concerns the very poor only indirectly. Indeed, in some cases in Western Cape organised crime groups and local gangs constitute a social welfare net for many of the destitute. In Western Cape, support for People Against Gangsterism and Drugs (Pagad) has seldom been from poor communities, with members more likely to be drawn from the comparatively wealthier sectors.

The preventive components of the NCPS and their effect on the poor may be limited. In turn, given the weak assessment component of the Strategy, evaluation of the impacts of the NCPS on any component of society is unlikely. Thus the greatest weakness of the NCPS is that it does not lay out clear guidelines and an appropriate strategy for how prevention programmes can be implemented at the local level within the context of a national framework. What is required is an understanding of how local government and SAPS commanders can initiate innovative crime prevention programmes. This is a critical oversight, and has had the unfortunate consequence of concentrating all the pressure for crime prevention at the national level. In any event, the NCPS with its (short- to medium-term) reactive focus does little to address this issue.

Proposals around reshaping juvenile justice, for example, deal with young people who are already in trouble. But for the poor in particular, interventions must be aimed at preventing young people from coming into contact with the system at all. Youths tend to become involved in crime due to economic need, family problems, educational difficulties, boredom, lack of vocational opportunities, peer pressure, and a lack of conflict resolution skills.176 This not only places young people from poorer backgrounds at greater risk of entering the criminal justice system; they are also less able (as has been argued above) to withstand its impact. Indeed, a case study of state reform in the area of juvenile justice usefully illustrates the (slow) pace at which the transformation of the criminal justice system is taking place.

(POOR) YOUNG OFFENDERS

The criminal justice system currently processes juveniles in much the same way as it does adults. With its emphasis on retributive punishment, which neither curbs crime nor helps people not to offend again, many of these young offenders simply end up back in the system – often more brutalised by their experiences than before entering it. Documented cases of assault and murder in prison cells have led campaigners to state that 'locking a young person up with an adult, even for five minutes, may be fatal'.177

The need for prevention has been recognised by the Inter-Ministerial Committee on Young People at Risk (IMC). Set up in mid-1995, the IMC – consisting of representatives of the Ministries of Welfare, Justice, Safety and Security, Correctional Services, Education, Health, and the RDP Office – was intended to respond to the crisis brought about by the release in 1994 of hundreds of awaiting trial juveniles. It also sought to transform the child and youth care system.178

While the IMC states that prevention services cannot be too specifically targeted, special attention to poor children is called for. Strategies focus on education and the family and community as arenas for intervention. Suggested programmes include teaching parenting skills at school, support services such as child care for working parents, training teachers to identify youth at risk, life skills training, either at school or through community structures, and involving the media in prevention. Importantly, preventive programmes should also be available to reintegrate young offenders into their families and communities, upon completing early intervention programmes.

Once a crime has been committed, early interventions are aimed at diverting young people away from the criminal justice system.179 This requires the substantial retraining of police as well as setting guidelines around arresting juveniles, contacting parents or guardians, and liaising with reception centres. Ideally, probation officers should assess children at these centres to determine how cases are handled and whether they should be referred to diversion programmes. Young people not accused of committing serious crimes and those not prosecuted more than twice are targeted for diversion.

This process provides 'secondary prevention'180 by intervening to stop youngsters from reoffending. Diversion is key, but the capacity of these programmes is limited. Most are offered by branches of the National Institute of Crime and Rehabilitation of Offenders (Nicro) in 20 areas of the country, with probation services running programmes in some areas. Diversion also seldom reaches young offenders guilty of violent and sexual crimes, or street children. Recognised programmes are not a prerequisite for diversion – prosecutors can withdraw cases on condition of referral to a treatment programme. Many community groups and non-governmental organisations already offer programmes which can be expanded to cater for diversion. Still, the shortages must be addressed: estimates in 1995 were that community-based programmes could cater for only about 2 000 of the approximately 35 000 perpetrators of minor offences.181 This secondary intervention – particularly where diversion programmes are concerned – is the domain of the departments of Welfare and Justice. The co-operation of the police is also, however, crucial in the initial stages of arrest and assessment.

Two national programmes of the NCPS are aimed at addressing these issues: diversion programmes for minor offenders, and secure care for juveniles. A programme team was to have been established to 'manage the extension of diversion and rationalisation of diversion policy in the criminal justice system'.182 In both cases Welfare is the assigned lead department, in conjunction with the IMC.

Several pilot projects under the auspices of the IMC are currently testing programmes aimed at the assessment and diversion of juveniles in conflict with the law, as well as identifying and responding to those at risk. While some of these projects have yet to start functioning, notable successes have been reported by the arrest, reception and referral project in Durban and the Family Group Conferencing pilot in Pretoria. The former confirms that the majority of young offenders do not commit violent crimes: only 4 per cent of the 2 500 children seen during the past 11 months were admitted for violent offences. Problems experienced included a lack of co-operation by the police, the exclusiveness of Nicro's diversion programmes, which need to be expanded to accommodate more youths, and the lack of capacity to follow up cases after young people are assessed and released back into their communities.

Of the 23 family group conferences (FGCs) conducted by the latter pilot, none of the youths reoffended, and FGCs were found to cost a fraction of the adversarial process of trial and prison. Problems reported were that the police were found to be reluctant to refer cases after arrest to the project, and prosecutors were not referring more serious cases (youths attending FGCs were referred for assault, housebreaking and theft).

Despite the high costs involved, the pilot projects are considered necessary to test whether proposed policies around juvenile justice will work.183 But national-level changes recommended in 1994 around assessment, sentencing and diversion procedures have yet to be embraced by government.184 No mechanisms have been put in place to ensure uniformity and sensitivity in the handling of young offenders, and the police in particular have not developed mandates for processing juveniles.185 This is confirmed by the reports from several of the IMC pilot projects, and one reason is probably that while the NCPS identifies child offenders as a priority, provincial police plans do not.

In terms of sentencing procedures, community-based sentencing is not being exercised, and juveniles are still processed through the system as before. There has been little progress in reducing the numbers of youths being detained to await trial. Although 10 times more young offenders now attend Nicro diversion programmes, this still represents less than 5 per cent of those being arrested.186 Legislative changes – without the implementation of alternative diversion and sentencing options – have warped the system still further: incarceration as a sentence for juveniles has increased by 30 per cent (rather than decreased as proposed) since whipping was abolished in 1995.187

The NCPS has not begun its work on reforming juvenile justice,188 and implementation has been slowed by problems around logistics and budgeting. There are not enough probation officers to undertake the assessment and referral of young offenders. The Department of Welfare is currently training social workers in this area, although budget limitations are slowing the process and preventing the employment of youth 'assessors' of the juvenile justice process in particular.189 All the departments involved in reforming the justice system for young offenders are reportedly not ready to implement the necessary programmes. Welfare, the lead department, is overwhelmed with general transformation requirements, of which juvenile justice is just one part.190

The impetus for improving police responses to juveniles, promoting diversion, reforming sentencing procedures, and monitoring existing (unofficial) programmes in order to avoid duplication needs to come from the national level. Currently this falls far short of expectations. Implementation of the NCPS programmes and of diversion in particular will, however, most effectively occur at the local level, since this requires the involvement of local authorities.191 Indeed, as is argued below, the distribution of responsibility for crime prevention among various tiers of government is a critical success factor for intervention.

GOVERNANCE AND CRIME PREVENTION

A fuller understanding of the nature and distribution of crime in South Africa has profound policy implications: crime differs markedly from locality to locality. This suggests that Pretoria-centric crime prevention initiatives will have little effect on problems specific to different areas across the country, pointing to an urgent need to decentralise crime prevention policy initiatives.

At the same time this cannot be done without guidance from national actors, and the development of clear frameworks to ensure that crime prevention policies are in line with national frameworks. Equally true, however, is the fact that effective crime prevention requires detailed assessment of the problems of any area before adequate solutions can be implemented.

These problems point to a number of possible policy alternatives. At the outset, given the different crime prevention challenges across the country, it may be advisable to site the issue of crime prevention more firmly in the office of MECs for Safety and Security. Currently, crime prevention is concentrated at the national level under the NCPS. The staging of provincial crime prevention summits and the formulation of provincial crime prevention strategies emphasise the importance of this initiative. In essence, this should entail a greater responsibility (and political accountability) for co-ordinating crime prevention initiatives at a provincial and local level within the context of national frameworks. Crime prevention programmes should also be specifically budgeted for at the provincial level, although financial resources will still be allocated from the national level.

Such an option also offers an appropriate outlet for the frustrations of MECs who feel that, given the constitutional provisions of a national police agency, they have been sidelined. One consequence of this has been a growing (and in many cases public) tension between MECs and their provincial commissioners.192 The current arrangement distances MECs from political accountability for crime control at the same time as weakening their responsibility for crime prevention.What is also lacking is a comprehensive evaluative framework for provincial crime prevention plans – unless the provincial strategies are constructed according to a range of guidelines, and operate according to clear objectives, they run the risk of simply being statements of intent.

In the framework of both national and provincial crime prevention programmes, much greater attention needs to be given to specific crime prevention strategies for the poor. As has been argued above, the wealthier components of society are often in a far better position, given where they live and the resources they possess, to prevent crime. The growth of private security in the suburbs is a case in point. The poor, on the other hand, are seldom the focus of crime prevention strategies partly because they are not in a position to lobby on the issue, and partly because crime prevention programmes are easier to implement and monitor in wealthier urban areas. Those in government charged with crime prevention need to specifically build the needs of the poor into any evaluative framework for crime prevention programmes.

For poor communities, it must be emphasised that the police retain a key role in crime prevention initiatives. In wealthier metropolitan areas with better developed forms of local government and larger amounts of resources, the police cannot be the only players in crime prevention initiatives. Indeed, in these circumstances the police, given that they are largely designed to react to crime once it has occurred, are seldom well placed to lead or in some cases even participate in crime prevention strategies.

This is not the case in, among others, rural areas and informal settlements fringing the cities, where many of the poor live. Apartheid forms of governance have resulted in uneven criminal justice and policing responses to the problem – often bearing little relation to the forms and extent of criminality. Rural areas, particularly the former homeland regions, are especially underresourced, requiring policy responses quite different from those in the urban centres which are currently enjoying the attention of planners.

Much of the debate on crime and crime prevention in South Africa has concentrated on the country's urban centres. Here the transformation of policing to achieve a more equitable, efficient and service-oriented approach remains the key challenge. This is no less true of rural policing. Indeed, in these areas the challenges facing the new order may be greater, and (in the short term) the gains more palpable. Transforming the former bantustan policing agencies – which are central to the policing of many of the rural and urban poor – has been little studied. Yet evidence suggests that the short-term gains of better policing in poor rural areas such as the former Transkei may be significant, given a history of almost non-existent (or alternatively intermittent and brutal) policing interventions.

Service-oriented and community-centred forms of policing may also go some way towards undermining growing levels of insecurity among poor rural inhabitants. While it is difficult to determine, given current evidence, whether high levels of insecurity and crime impact negatively on development, there is substantial evidence to suggest that for many of the most vulnerable rural residents in the case study area, achieving a safer living environment was a central concern. And while effective policing is key to achieving this, the police agencies in rural areas are often, as has been argued, the 'front end of the state', delivering a wider degree of service functions than simply responding to criminal incidents.

Transformation of the police in these circumstances is critical for more effective crime prevention – hence the concentration of this study on policing. The case study of the former Transkei suggests that innovative and proactive forms of policing could have greatly reduced conflict in Tsolo and Qumbu. The transformation of policing at this level and the building of policing partnerships and greater accountability for community needs are urgently required. In rural areas in particular, given the potentially wide focus of their activities, the police can play a useful role in co-ordinating prevention initiatives. Yet, given high levels of crime, poor training and a lack of innovative management, the police remain essentially reactive – spending time and resources on problems (as in Qumbu and Tsolo) which could have been prevented with appropriate actions.

Questions of crime prevention then often hinge on the development of appropriate forms of governance. In areas such as Northern Cape, where local government institutions are arguably better developed and the causes of crime not necessarily linked to achieving better policing (while this is of course both desirable and possible), crime prevention requires the co-ordination of a range of actors. In areas such as the rural Transkei, given the changing nature of local governance and the weakness of local institutions, the police have a key role to play in crime prevention through appropriate interventions.

The implications of this argument for the poor are critical. At the outset, poorer communities are seldom seen (at least by the rich and powerful) as the predominant victims of crime. Yet the poor form not only the majority of victims, but also have fewer resources with which to cushion the costs of crime than the wealthy. That makes a focus on prevention strategies for the poor not only logical but essential: improving the reactive system of criminal justice, while important, does not necessarily assist poorer victims of crime. Unfortunately, the debate on crime prevention centres on the interests of the rich, who are in any event policed more preventively than the poor.

The danger is that government policy will only focus on improving the efficiency of the (reactive) criminal justice system. Also, government strategies such as the NCPS hold few benefits for the poor and destitute and, in any event, are more reactive than preventive in focus. In respect of security questions, the poor have always been the focus of reactive strategies. Without an adequate focus on crime prevention designed specifically for the poor, the interests of those who bear the burden of crime will continue to be ignored.