Marais, E. (1992). Report of the Community Policing Working Group. Paper presented to CCJ conference on policing in the New South Africa, Durban, September.


Etienne Marais

Paper presented to the Centre for Criminal Justice conference Policing in the New South Africa, Durban, September 1992.

Etienne Marais is a former Researcher at the Centre for the Study of Violence and Reconciliation.

The Community Policing Working Group is a co-operative effort involving a number of institutions. It was established under the auspices of the Centres for Criminal Justice, University of Natal, Pietermaritzburg and Harvard University at the first CCJ conference on Policing in the New South Africa, June 1991.


The working group arose to take forward the initial enthusiasm for community policing which was expressed at the first CCJ conference held in June 1991.

This enthusiasm was the result of the concepts apparent compatibility with particular South African concerns such as community empowerment, community control and community conflict resolution. In addition the concept seemed to have growing support among leading police agencies internationally. This credibility was seen as a positive factor - particularly in engaging with the South African Police1 in a constructive manner.

After our initial discussions it became clear that the implementation of community policing in South Africa would not be a simple matter. Our concerns are summed up in the following questions:

  • Is community policing practical, indeed possible in the present SA context, characterised by deep divisions and a militarised policing philosophy?

  • Is community policing such a good idea, considering it's inherent sophistication as a means of social control, in the absence of a representative state, or of a negotiated social accord?

  • Can community policing be pursued, indeed promoted separate from a concern for and reality of, the absence of real community empowerment in SA?

  • What should/could the relationship be between community policing - as a negotiated form of social control and other negotiations around local government structures which continue in all corners of the country?

  • What concrete recommendations can be made which will further the development of a "community policing" approach, if not full implementation of genuine community policing?

The activities of the Group have centred on in-depth consideration of the issue of community Policing, and wide ranging consultation, primarily with experts and practitioners close to the field of community policing.

This process of consultation has, in short, raised more problems than solutions in terms of our original objective - namely to consider how community policing might be implemented in South Africa. However this process of clarification and discussion has been invaluable.

From the deliberations two central issues have emerged:

Firstly it must be noted that "Community Policing" is itself a slippery concept. Both internationally and locally it has a variety of different meanings, which make it difficult to meaningfully promote "community policing" without substantial definition of precisely what one is talking about. Debates about community policing invariably involve a contest over the meaning of the term. This report attempts to clarify the concept and looks at the underlying principles of community policing and their significance for the South African context.

The second issue is that the implementation of community policing does not lend itself to a uniform process in South Africa. If community policing is defined as occurring through an interaction between the formal police agency and communities, it becomes clear that the enormous power disparities and differential attitudes to formal policing require a great deal of flexibility in the way that community policing is pursued. Community Policing must be participative if it is to succeed and the existing political culture within particular communities will determine to a substantial degree the forms which such "participation" can take.

Defining Community Policing

While the term "community policing" is quite extensively used in South Africa it has come to mean quite different things to different people.

Some Usages of the Term "Community Policing"

At the one end of the spectrum, community policing refers to the complete exclusion of the formal policing agency. For communities where there is a history of alternative systems of justice, community policing means policing by the community with the exclusion of the formal police agency. "Community Policing" therefore refers to what are popularly known as self-defence committees, or peoples courts. This notion of community policing, which is explicitly opposed to the formal police agency, arises from a perception that the police are primarily agents of oppression rather than protectors of the community.

However the degree to which these forms of "community policing" are directly opposed to the formal police agency varies, particularly since the signing of the National Peace Accord in September 1991. This is true of the various paramilitary groups associated with particular political or group interests, (which may not see themselves as a form of community policing). In this category are many of the so-called "private armies" - paramilitary units involved in the "defence" of township communities, hostel communities, tribal structures and traditions, farming communities and particular political interests such as the right-wing. Such structures may also be linked or associated with "informal" systems of justice such as Civic anti-crime committees, tribal courts or hostel committees, which may, to varying degrees, make use of the formal policing and criminal justice systems.

These structures can be described as a form of community policing in that they are involved with the protection or defence of particular communities - which are often defined in political or ethnic terms. There exists an uneasy or conflictual relationship between most of them and the formal police agency when they adopt the use of coercive force and step across the bounds of "formal legality". On the other hand some of these structures (such as farmers "defence committees" and tribal courts) enjoy positive and co-operative relationships with the police force.

The other end of the spectrum of usages of the term is that commonly used by the police. Here "community policing" refers to policing in the community. Community policing is not seen to be a fundamentally different style of policing from that currently in use, but simply means that the police structures are closer to the people in the community. This use of the term does however imply greater visibility and person to person contact in the actual practice of policing. The police speak of the "Bobby on the beat", satellite stations; or bicycle and mounted patrols as community policing. Police-community liaison forums and local dispute resolution committees are seen by the police to be forms of community policing. In general there is little distinction in the police discourse between police-community relations and community policing.

The third way in which the term community policing is used is to describe the activities of the community in preventing crime which occur alongside the policing of the formal police agency. Neighbourhood watches, and various "block watch" initiatives are structures which rely on the building of community support on a neighbourhood basis and rely on awareness, communication and crime prevention to police the area. Neighbourhood watches are not based on the use of coercive force and depend heavily on the formal policing agencies. However some such community initiatives are vigilante-like in their activities (such as anti-squatter committees), being dependant on a show of strength and often armed.

In many areas these are initiated by the police themselves or with the support of the police. Similar to this, but not reliant on actual policing activities on the part of community members is the phenomenon of private policing - a fast growing industry of home security, based on rapid armed response and an array of hi-tech security gadgetry.

The phenomenon of Business Watch is a police supported crime prevention initiative where perhaps the most equal partnership exists - both the business "community" and the police having active and complementary roles.

Problems with Wide Usage

This very wide range of uses of the term "Community Policing" poses a number of problems. These different usages are reflective of the variety of contrasting and often conflicting ways in which "community order" is seen - and the best ways to achieve this community order. The wide range of usages thus reflects large differences between communities in their attitudes to the formal policing agency and its role.

It has been argued that the wide variety of usages mean that it is not very useful in the South African context to talk about community policing. One should rather talk about the specific characteristics of policing which one wants to achieve - such as community sensitivity, accountability and representivity. However, the term community policing has caught on and is being used more and more in South Africa, particularly by the police themselves, to describe recent developments in "police reform". Like the term democracy, community policing is invariably a value-laden concept - and one which is usually held to be positive. The development of a working definition in the South African context is thus particularly important, as a reference point against which the range of "community policing" initiatives - both informal and formal, can be measured. It does seem to be important that greater clarity is reached among policing academics and policy makers as to what community policing actually entails.

Towards a Working Definition

In formulating a working definition of community policing, the group chose to approach the issue from an examination of the nature of formal policing. Many of those whom we consulted felt that the definition should be broader, and encompass some or all of the forms of "community policing" described above. (para 2.1) We have however limited our examination to formal policing. This was, after all, the central concern at the outset, and there are strong arguments to deal with informal policing initiatives at another level (see "The cases against formal community policing" below) To place all forms of policing at the same level seems to obfuscate the debates and is not particularly useful. Our working definition of community policing is thus centred on the formal policing agency.

A working definition is important as a yardstick against which to assess formal policing initiatives, as well as a basis for proposals for "South African" forms of community policing. Thus while the SAP's own use of the term is sometimes claimed to be based on models elsewhere in the world, this reflects the use of isolated components or policing techniques borrowed from community policing rather than a holistic "system" of community policing. In our view these forms of policing fall far short of community policing in the sense in which we believe it should be used.

The working definition which we propose is based on an examination of community policing elsewhere as well as of the specific features of South Africa which an "indigenous" form of community policing will have to take account of. These specific features of policing, and of communities in South Africa are vital in the development of community policing in the South African context. This is not to say that the experiences and ideas of community policing in other parts of the world are not useful, but rather that they cannot uncritically be imported into the South African context.

International Perspectives on Community Policing

Internationally there is no single definition of Community Policing. It is rather a general philosophy of policing which has taken the notions of partnership and service to the community to their full extent. In essence however there are three main features of community policing:

  1. It aims to create a real partnership between police and community, with the view to more effective protection of the community and a better quality of life.
  2. There is an emphasis on solving the underlying problems which lead to crime rather than simply arresting offenders.
  3. The police are more visible and accessible through beat patrol. This is achieved in conjunction with the assignment of staff to a specific, small, geographical area.

These broad strategic objectives have a range of implications for the organisation and management of policing. Rather than attempting to describe in detail all the features of community policing in different places, a range of strategies and principles are listed in order to provide a sense of the complexity and breadth of community policing. In the Canadian case,2 the following are some of the features of community policing:

  • Consultation with the community to identify the short term priorities for addressing crime, as well as to obtain a more enduring long-term mandate from the public.

  • Proactive policing centred on the systematic identification of crime and disorder problems, rather than simply reacting to calls for help.

  • An emphasis on co-operative strategies with other service delivery agencies.

  • An emphasis on the exchange of information between police and community. Police officers are thus seen as "information managers".

  • Personnel are permitted to become career generalists rather than specialists. Officers who work directly at community level are regarded as highly trained and are relatively well-paid professionals who have the respect of their colleagues and the local community.

  • Greater responsibility and autonomy for front line officers is facilitated by decentralised police management.

  • The hierarchical paramilitary organisational model of the police is surrendered for an organisation based on participatory management and greater delegation of responsibility to all levels.

  • Contact with, and knowledge of, specific neighbourhoods is attained through an organisational structure based on small geographical areas in which staff work on a long term basis. The notion of "area specialists" and crime generalists is applied to a range of policing functions.

In addition, the New York Police strategic plan3 includes the following features and emphasis:

  • A concern with alignment of the values of the police and their constituents. Some of the key values are a concern for equality of service regardless of ethnic or social background and cultural, religious and sexual orientation; a concern to place the highest possible value on human life and a concern for the "quality of human life".

  • A concern to adapt organisational mechanisms and procedures so that the values of community policing are encouraged and realised. This includes the encouragement of creativity and initiative in the resolution of problems; the establishment of new measures of departmental performance, new mechanisms of reward and a new focus for training of police recruits and officers.

  • The focus of community policing is to build neighbourhood capacities to reduce crime and disorder. This may necessitate the organisation of neighbourhoods.

  • Rationalisation or improved classification of calls for help (911), so that only those calls requiring rapid response will be dealt with in this way.

  • The enhancement of mechanisms for integrity control to minimise the chances of corruption within the police force.

  • Community Policing is not just a component of a conventional policing philosophy, but affects the entire organisation.

Within community policing circles, it is also quite widely accepted (through the impact of research) that institutional discrimination on the basis of race, gender and socio-economic status has been a problem for police agencies. Community policing is often concerned to establish mechanisms and programmes to deal with such discrimination and puts special effort into establishing relations with so-called "minority groups".

Specific "South African" Features to Take into Account in the Definition of Community Policing

The above description of the strategies and methods of community policing are drawn from societies where social order has a quite different basis from that in South Africa. In a sense "western" descriptions of community policing are based on a number of assumptions and features particular to these societies. These include the following:

  • democracy, including a tradition of civil liberties and the inviolability of individual human rights.

  • a tradition of public accountability and a relatively long standing acceptance of civilian scrutiny of police policy and practices. Civil society has fought for, and achieved the right of civilian oversight of police agencies, and the principle of freedom of information is relatively well established (compared to South Africa).

  • the vast majority of citizens assume that the police force is - at very least in its organisational goals - professional and accountable to the society as a whole and to the values that hold the nation together. It is assumed that the police will not be used to further any sectional interests.

  • the existence of an enduring "social accord" among the citizenry on which the broader social order is based. This is a set of values, principles and symbols which are the foundation of nationhood. In other words, there is a relatively high degree of consensus about how conflicts and "disorder" problems should be resolved, and on the rules which determine how political power should be exercised.

  • decentralised, or relatively decentralised control of police agencies.

While South Africa is undergoing a transitional process towards democracy, a number of these features of democracies and of policing, which determine the context within which policing takes place, are absent or in an embryonic form in South Africa.

  • Contested state legitimacy: Perhaps the most important single feature of South Africa is the contested nature of government authority. Despite the process at CODESA and the signing of the National Peace Accord in September 1991, the legitimacy of the state and hence the authority of the police force, remains contested.

  • Lack of accountability: This is related to the lack of direct accountability of the police force to the population as a whole. This lack of accountability is reflected by allegations of deaths in detention, continued doubts about the present role of the "security branch", a lack of confidence in investigations where police are alleged to have been involved in crime and widespread perceptions that the police have been directly involved in "political" violence. Perhaps most significant are the absence of effective mechanisms whereby ordinary members of the community can take up grievances against the police.

  • Lack of direct engagement with issues of racial discrimination. In the era of the "New South Africa" there is strong defensiveness about the issue of racial discrimination within institutions such as the Police. This is combined with the weakness of civil pressure around racial discrimination. However policing remains profoundly affected by our apartheid past and racial discrimination within the police force and in relation to the community, remains considerable.

For these reasons one cannot assume that the adoption of community policing by the formal policing agency will have the same legitimacy and support as the process has had elsewhere. The process whereby it is negotiated and introduced - as well as the prevailing political context will be of vital importance.

Perhaps the most important factor to be addressed in the implementation of community policing is that of accountability and our definition of community policing in South Africa needs to pay special attention to this issue. While definitions of community policing elsewhere seldom explicitly refer to accountability, it is implicit in these agencies understanding of their relationship to the communities they serve. In our context accountability is an essential ingredient of any attempts to implement community policing.

A Working Definition of Formal Community Policing in South Africa

This definition provides a vision of the manner in which the police agency should relate to society, particularly at a local level. This definition is covered in terms of principles, objectives and policing strategies.

Community Policing in the current South African context is policing which is based on the following:


(i) Acceptance of the principle of public oversight and accountability at an operational level. Key areas of concern here are the establishment of mechanisms for the credible handling of public complaints and involvement of independent persons in the investigation of police abuses and police criminality.

(ii) Police organisation should subject its planning and strategy development to public scrutiny. This "transparency" of the police organisation is particularly important in relation to training, promotions, internal disciplinary mechanisms and the mechanisms for control over undercover and surveillance work.

(iii) Accountability either at a regional or national level to a broadly representative civilian structure.

(iv) That organic or traditional systems of local policing such as anti-crime committees should be facilitated and supported insofar as they are broadly accountable and representative of the community and operate according to accepted criteria of political tolerance.4 Where conflict around such structures exists it is the role of the police to ascertain the nature of such conflict and to deal with this in a constructive way.

(v) The diversity of communities is recognised and accepted. This means that the maintenance of social order must, as far as possible, occur according to the values and norms in a particular community.5

(vi) That special attempts should be made by the police force to address the needs and concerns of special interest groups which are, or have been, discriminated against; or by virtue of some other factor, are disadvantaged in their ability to demand specific police service.

(vii) All communities should have access to the same quality of service and resources should be fairly allocated. Visible mechanisms must be established to ensure that powerful and influential groups are not allocated an unfair proportion of police resources.

(viii) Community policing requires changes to the police organisation to enable local operational flexibility, enhance the status of patrol work and the rewards for community problem solving rather than the arrest of suspects.


(i) Community police has as its objective the establishment of an enduring partnership between police and all communities, with the view to more effective protection of the community and a better quality of life. A particular focus of community policing are those communities where trust for, and co-operation with, the police have been particularly low.

(ii) Through research and the gathering of information on a systematic basis, to deal with problems of crime and violence in a proactive and "problem solving" manner.

(iii) Through input and information from the community, to fashion the nature of the policing service provided so that it, as far as possible, addresses the primary concerns and fears of citizens. Community policing thus aims to make the police agency more accountable to the community being served.

(iv) To provide a visible and accessible presence which enhances confidence in the police and improves crime prevention.

(v) To align the values of the police organisation to those of the emergent new South Africa, and, in addition, to produce police officers who can interact with their communities sensitively and in a way which respects local norms and cultural values.

Policing Strategies

(i) The establishment of liaison forums at community level which are aimed at establishing equal relations with all sectors of the community with a view to providing input into the management of local policing.

(ii) The use of needs assessment and research to ascertain the concerns of the community.

(iii) A process of thorough consultation around changes in policing practices at a local level.

(iv) The integration of the philosophy and information derived from community policing into all aspects of police-work.

(v) The development of special programmes to serve the needs of marginalised, minority and other special interest groups and to foster sensitivity to their concerns and values. In our context, considerable attention has to be applied to the needs of most black communities.

(vi) The assignment of personnel to relatively small geographical areas on a long term basis and the use of beat patrol in communities as a means of establishing contact with these communities.

Key Issues to be Confronted in the Implementation of Community Policing

In the course of our consultations during the last year, a number of reservations and criticisms about the notion of community policing arose.

These arguments focus on the key obstacles to the implementation of community policing, and need to be examined in some detail. The debate around these issues is a useful one, as it serves to give substance to our working definition of community policing. It also serves to highlight the key obstacles to effective community policing at present.

In this section these objections or reservations are described and then discussed in more detail. This section also deals with the "special problems" of the transitional process and of policing in rural and tribal contexts.

The Readiness of the SAP to Implement Community Policing - The Organisation and Culture of Policing in South Africa
South African police agencies are not ready for community policing. Changes to South African Policing are necessary for three reasons:
(i) One cannot have community policing in the absence of police accountability and hence the emphasis should be on creating mechanisms of control and accountability.
(ii) Community Policing cannot be implemented by an hierarchical paramilitary organisation, with little operational flexibility.
(iii) Regional or municipal police forces are necessary for community policing to become a reality. A single national police force cannot adapt, or be sensitive to the needs of specific communities.

Our working definition has placed key emphasis on accountability. This is recognition of the fact that workable and acceptable mechanisms of accountability are a precondition for community policing. Some of the steps which have recently been taken by the SAP, such as a new structure for the investigation of complaints of abuse on the part of the police are important steps forward. However civilian oversight and an independent complaints mechanism are areas requiring attention.

The paramilitary nature of the South African Police Force is a product of our colonial history as well as the more recent involvement of the police force in a counter-insurgency role. This is closely related to the centralised nature of policing in South Africa. Despite increased regionalisation of the policing function since 1989, there is little real or operational autonomy at a regional or a local level. This is borne out by the fact that the commanding officers of the 10 police regions are junior in rank to the chiefs of the 5 specialised Head Office divisions.

Although police-community relations and the "partnership in policing" is often emphasised by the South African Police, the paramilitary character of policing, as well as the highly centralised and "exclusive" form of police accountability, makes truly positive relationships at a local level hard to realise.

The New Police-Community Relations Initiative

The SAP sees the police-community relationship as important, and in the last year significant efforts have been made to improve such relations.

The South African Police has recently announced the establishment of a division of police-community relations to be headed by a Lt General.6 This is a positive sign in that it reflects the need on the part of the SAP to take police-community relations more seriously. The division will include four sub-divisions: Police-Community relations, Public Relations, Internal Liaison and NPA Coordination. This division will also be responsible for a strategic planning component, conflict evaluation and the co-ordination of the SAP's participation in commissions of enquiry.

It was also suggested that the division would be headed by one of the new "black generals". However there are no black Major-Generals and therefore no suitable candidates exist.7 Two potential problems with the move can be identified. Firstly, although a special division of police-community relations is needed, the organisational structure of the SAP does not facilitate easy interchange between the different divisions. "Area specialisation" as opposed to "task specialisation" is a key component of community policing and suggests that the current police structure of six parallel "specialist" divisions is inappropriate for community policing. This problem has already been demonstrated in relation to the Internal Stability Division. Good relations which exist between local station commanders and communities have often been ignored when the ISD has been sent into the area - leading to a breakdown in trust and co-operation.

The second weakness is that the police generally understand police community relations in the context of the authority of the police. The police is the "dominant partner" in the police-community relationship and the community is not seen to have the right (or indeed the capability) to co-determine the nature of policing practice. An example of the effects of this in practice is that the local level "forums", established as part of a police-community relations programme, have been hampered by the insistence on the part of the police that all meetings are chaired by the police. Indeed the emphasis for police-community forums appears to be on quantity rather than quality. There is little evidence that the police are generally aware of the issues of representivity, or that forums lead to substantive input and positive responses on the part of the police. The practice of meeting on a bilateral basis with a vast range of community groups is of dubious value, particularly as an alternative to properly representative community forums.

This is borne out by the experiences of those involved in the Local Dispute Resolution Committees of the National Peace Accord. As with many liaison forums the police are often unwilling (or unable because of organisational policy) to regard the views of the "community representatives" as necessarily relevant or deserving of an organisational response. Indications are, for instance that investigations which LDRC's regard as vital for the preservation of the peace are not necessarily prioritised. In addition, the style of policing has been predominantly reactive in the past, and although there is a trend towards pro-active policing, crime prevention remains underdeveloped.

Our working definition of community policing implies that community policing requires a flexibility of police management and response. In South Africa (as elsewhere) this requires substantial changes to almost all aspects of the police organisation. Some progress has been achieved through the establishment of the Police Board and the acceptance of the valuable role of outside experts in some areas of organisational development. However this tends to remain at the level of "image management", and a great deal has still to be done.

The history of the police and its relation with the black communities is such that the establishment of a truly co-operative relationship will be a difficult one. However, in the context of the negative attitude towards the police, it is likely that even a small change in policing styles may lead to a large swing in public opinion. The increase in public participation is, in the context of wider political change, likely to lead to greater support from within the police organisation for community policing strategies.

Despite the nature of the police organisation and the dominant attitudes to community liaison, the group has been encouraged by the support for community policing from individual officers in the police force. Indeed the paradoxical nature of power in the police community relationship seems to suggest that the police may not be opposed to the notion of community policing, even in its more "radical" form. Perhaps this is the reason for its acceptance by a large number of previously autocratic police agencies.

While the process of organisational change is undoubtedly a long and difficult one, this does not in our view mean that Community Policing should not be put on the agenda. Indeed, it is on the agenda already, and the debates around community policing in the South African context, coupled with lessons learnt in other parts of the world, can only stimulate the process of organisational renewal. It is our view that police interaction with communities, despite its current shortcomings, has an influence on the police organisation which, in the longer term, can only be positive.

Community Control, Regional and National Policing

It is clear that a centralised national police force, in its present form, is an obstacle to true community policing. Recently the government has announced plans to regionalise the police forces, while the issue of "metropolitan" police forces has been on the agenda in Cape Town and Johannesburg for some time.

There are several advantages to a more decentralised policing structure. The first is that it would, to some extent, reduce the effects of a huge, top heavy paramilitary bureaucracy through the existence of smaller units and a more manageable organisational scale. This in turn may contribute to greater sensitivity to local concerns.

It has also been suggested that by disempowering head office, regionalisation may lead to a reduction of central political control, even if police accountability continues to operate primarily at a national level.

The idea of regions or provinces is firmly entrenched in the South African Political debate, and one must assume that either in the federal vision of the IFP and the National Party, or in the centralised vision of the ANC, such "regional" units will have some degree of autonomy in terms of governance. However this is one of the central political controversies currently bedevilling the negotiation process, and it is unlikely that viable new structures of regional government (other than existing administrative structures) will be feasible until after elections.

In the absence of representative regional government (or credible forms of police accountability) the regionalisation of the police force does not necessarily equate with increased police accountability. At the same time, one must not assume that the diversity present at a national level will not also be present at a regional level. One of the apparent strengths of "Homeland" Police Forces is their ability to reflect the specific ethnic and cultural characteristics of the region they serve. In reality however, the overwhelming influence of the SAP means that all these police forces have maintained much of the character of the SAP, and are at best only marginally better accepted by local people than the SAP (if at all). The geographical areas which define these ethnic homelands have been widely discredited as a basis for "new South African" regions, and, in any event, the cultural homogeneity of these areas is often more symbolic than real.8 The new geographical regions are likely to be as socially and politically diverse as the current South African provinces. Political allegiances are likely to be diverse in all regions. The "problem" of policing within a diverse society is therefore not substantially affected by whether the police force is a national or regional one.

In effect South Africa has 21 police regions (10 Homeland police forces and 11 SAP regions). Most of these are likely to be increased in size to cater for the 7 to 11 provincial regions which will most likely form the basis of the new constitution. Such police regions will either be geographically vast (in the rural districts) or cater for several million people (in the urban areas). While the potentially greater sensitivity of such regional police forces is not disputed, this form of decentralisation does not necessarily bring us closer to our definition of community policing.

Emergent Forms of Local Government, Community Policing and Negotiations 9

As highlighted in para. 2.5 above, South African policing presently suffers from an absence of an enduring social accord (or set of rules of the political game) on which the legitimacy of the state and institutions such as the police force are based.

In the absence of such a social accord, the authority of the state and its institutions is the subject of ongoing contestation, and therefore cannot serve as a basis for governance. This has also been the case in the arena of local government where widespread challenges to the legitimacy of local government structures during the 80s and 90s has led to a situation where new relations are being negotiated. The first stage of such new forms of interaction is the establishment of exchange relations. These are relatively short term exchanges which are mutually beneficial and rewarding. Once ongoing modes of interaction have been established, these relationships can develop into relations of reciprocity. Actual examples of these are where township and other interest groups have, through ongoing negotiations and the establishment of working relations, reached specific agreements which confer mutual benefits and obligations.

The experience of the establishment of exchange relations and relations of reciprocity have important implications for our conceptual approach to community policing. They serve as real experiences of negotiated forms of co-operation and embryonic new modes of social order, in the absence of a generally legitimate authority. It is clear that the contestation of authority during the transitional phase remains one of the central problems in governance. However, even with the arrival of an interim government of national unity - and a new constitution, the legacy of apartheid policing would suggest that the authority of the policing agency should be "built" through a process of negotiations with communities rather than "imposed" on the basis of the acceptability of the newly negotiated state structure in place.

In a sense, the National Peace Accord provides the infrastructure for these forms of trade-off. Indeed, many of the agreements reached between police and communities (through structures such as LDRC's) constitute a form of exchange relation. They also create ad-hoc or temporary legitimacy for the state, but make it conditional on the police acting in certain ways.

The dynamic state of flux in local government negotiations is vitally important for another reason. This is because such "local government" structures or agreements cannot be equated with the metropolitan government or municipal government of other democracies, which, in the case of the US, form the most important locus of Police accountability and control. Local government has in the US context evolved a range of "rules of the political game", and an administrative infrastructure which provides stability to the process of governance despite changes in the elected political leadership. In South Africa, the "rules" themselves are subject to constant debate and contestation. Those in office were elected on the basis of rules which are regarded by substantial groupings as illegitimate. Nowhere are new local government structures, which are the result of accepted local social accords, in place. In this context the idea of basing police accountability on local government structures is perhaps premature.

A further point to note is the social diversity of urban areas. Local government negotiations in for example the metropolitan area of greater Johannesburg are a clear illustration of the diversity of communities within urban centres - a diversity which is reflected in all South African towns and cities.

It appears that the idea of basing community policing on formal police accountability to either local or regional government, is not feasible at this point. In the present such policing would be no more democratic (and thus accountable) to citizens, than the national police force. In addition the restructuring process - particularly at a municipal level - would require a substantial "change over" period during which time changes to policing philosophy would be rather difficult.

A Multi-level Approach to Community Policing

The forgoing discussion on local and regional options in the short to medium term, would suggest that they do not necessarily offer substantial hope for greater police accountability - and hence for community policing. The issue on which the potential for community policing exists is rather the philosophy of policing within whatever organisational units of policing exist.

This would suggest that at present, community policing could be as successfully carried out by South African Police on a national basis (with the necessary structures of accountability) as by any other unit of the same police force. What needs to be pursued are the principles, strategies and organisational implications of community policing for the South African Police - and we recommend that community policing is pursued within the present national framework of policing At the same time local initiatives (where organisational policy allows them) may be important in demonstrating the efficacy of community policing.

The Dilemma of Community Control
Community control of policing is impractical (or dangerous) in the context of community conflict.
Community policing is unrealistic where communities are deeply divided and fraught with conflict. Community policing implies some degree of consensus within a defined community as to what the main "disorder" problems are. Deep antagonisms and political intolerance cannot be dealt with in police-community forums. Because of the prevalence and nature of community conflict and the lack of a tradition (or the difficulty in the context of repression) of accountability, community control over policing poses dangers. Complete local autonomy of policing in the South African context raises the spectre of the police being drawn into intergroup conflicts as participants rather than mediators and "problem solvers". This concern arises from the experience of organic "community courts" and civic controlled self-defence units or anti-crime initiatives which have often represented the interests of small and powerful sectors of the community through coercive and violent means to the detriment of human rights, increasing fear and intimidation. The assumption here is that community policing entails community control without mechanisms of accountability to a broader "social accord" or nationally determined principles. The argument concludes that what is needed in order to deal with the violence arising from community conflict is a national (centralised) police force, but one which is accountable, credible and effective.

The question of the extent and modes of community control or input into policing is a complex one. It is also at this point that the nature and definition of "community" becomes vital.

The Nature of Communities

There are several senses in which the word "community" is used:10

  • the territorial entity defined by geographical boundaries;

  • the specific interest community defined by common interests;

  • communities as defined by a sense of belonging to relationships or places.

In the South African context "community" has increasingly come to refer to "the black community" and, in terms of the political polarisation, is defined both around the common interest (of liberation), and the sense of belonging to, or identification with, the oppressed people. This sense of "community" as reflective of political divisions is also apparent in the residence patterns which present a stark dichotomy in the lifestyles and perceptions of policing in the different South African communities.

However even specific geographical "communities" are divided into a range of sub-communities with differing interests, values and needs. These groups differ in the amount of power which they exercise in the community - some being more marginal than others. In terms of policing, it is functional to examine communities at the level of the smallest group with identifiable common interests or common identity. Thus a larger "community" is made up of other communities: Women, men, youth, the unemployed, particular political and cultural allegiances, which may or may not intersect. Women and men are part of the same community, but in terms of the social order, have different interests and are treated in different ways by the police.

In reality the status quo, as far as the social order in a "community" is concerned, is defined at any particular point by an equilibrium in the power relations between different sub-communities. Such power relations are generally dynamic, and change leads to a re-alignment of power between the different sub-communities - in effect, a challenge to the existing status quo as far as social order is concerned.

For a range of complex reasons, South African communities are often deeply polarised. The introduction of community policing rests on a crucial assumption that a consensus can be reached about what is to be policed.

In some areas, such as parts of Natal, communities tend to be categorised as either IFP areas or ANC areas. In fact, all these areas have a dominance of one party and a silent acquiescence of those supporting the other party. Labelling an area as legitimately "IFP" or "ANC" means that the police only relate to the structures of the dominant group. While this may be seen as an acceptable relation of political accountability it has the danger of simply legitimising the prevailing political status quo at a local level.

These same communities are also the ones where civil society is weakest and where the criminal lifestyle/discourse is strongest. Thus although such communities need community policing the most, it is likely to be more difficult to achieve than elsewhere. The reason for this is that in the absence of effective formal policing other forms of social control and new power relations establish themselves. Examples of this are "peoples courts", defence committees, or even powerful politically aligned civic structures. In the context of the white right wing the defence structure of the Transvaal Agricultural Union is a case in point. Such new power structures create vested interests in the status quo - further weakening community trust in the formal police agency. Thus the refusal of a civic organisations in an area like Natal to support a community policing project can partly be understood in terms of the threat it may have appeared to pose to incumbent civic leaders. While this is not the only possible explanation it does raise the difficult question of the attitudes of various political elites and stakeholders to a system which appears to challenge their exclusive power.

A Balance between Community Control and Broader Accountability

As a means of addressing the dangers of community control over local policing in the context of severe conflicts it is proposed that a dynamic relationship between local police autonomy and more centralised accountability of the police is required. This central accountability should be conceptualised as multi-directional, being exercised through "independent governing bodies", professional codes as well as elected political leadership.

This should go together with a concern to make the process of police-community liaison and consultation as representative as possible. In the context of violence, this calls for a far wider range of "contacts" and a detailed understanding at the local level, of the specific community dynamics at play.

In the context of the importance of strategies to address the violence, the "problem solving" aspect of community policing has a vital role to play. Members of non-governmental organisations such as monitoring groups have often played an invaluable "problem solving role" role in relation to the prevention of community violence. This role involves facilitation of meetings between opposing groups, assisting with negotiations, developing strategies to build trust between groups and the identification of the core issues at stake. In the case of an initiative such as ILEFO11 such "problem solving" has also been directed at facilitating a more meaningful response from the police force. These roles may, in other contexts, be described as community policing. Two key issues are highlighted by the roles of such facilitators:

  • Accountability to a community and its interests does not mean that policing needs to be conflictual, particularly within the "problem solving" mode of policing. In addition nationally determined modes of dispute resolution, codes of conduct and principles (such as those of the National Peace Accord) can provide a framework within which local diversity is accommodated.

  • The roles describe provide a valuable pointer to the potential of community policing in relation to the violence. They also suggest that contrary to our initial doubts, there exists a substantial body of expertise in the practise of "community policing" in these contexts.

Community Policing as a Threat to Social Order and Police Effectiveness
Community policing will lead to too much community input which will prevent the police from upholding law and order.
This argument regards community policing as the "tail wagging the dog". This relates to the dominant discourse around the relationship between police, society and social order. This view of social order is very important because it defines the "authoritarian" view of the police mandate from the state as being inviolate and unproblematic. This view of social order is dealt with at some length because of its importance in debates around community diversity, community policing and accountability. This view (which is described as "the consensus view") tends to assume that the police are impartial and apolitical, even in the South African context where police accountability has been exclusively to the National Party.

The concern that community policing (consultation with communities about policing) inherently weakens the capability of the police to enforce law and order is one of the conclusions of a particular discourse on policing and social order. This perspective is particularly important because to a great degree it determines police attitudes to community policing. It is also vital in understanding how the police see community liaison forums, accountability and community norms. Because of its importance, this discourse (and the alternative vision of policing) is examined in more detail.

The Dominant View of Policing : The consensus view of social order

The dominant philosophy of policing (simply put) suggests that it is the notion of "policing by consent" which allows the tension between democracy and policing to be accommodated. According to this view, the police are delegated authority and power from the state on the basis of a broad consensus about the nature of the social order to be policed. The process of parliamentary democracy allows citizens to express themselves on the values and norms to be protected. The fact that the police are delegated authority by society means that they are accountable to society for the use of those powers. It is thus the delegation of authority from the citizenry which underlies the police-community relationship.

Transgressions of the individual's freedom can thus only be justified within the context of public support for the methods and practice of policing.

However this approach views the main channel of police accountability as to the state, and to the law, which is generally assumed to be fair and unproblematic. This approach also places strong emphasis on the "independence" and "professionalism" of the police.

The problem of this view is that "independence" tends to be equated with impartiality. In fact independence of the police does not mean that the police are not tied to political interests.12 Police forces, far from being "inherently impartial" generally reflect the dominant interest groups within society. This is particularly obvious in the South African context, where the police claim to be operationally independent, despite the overtly political history of policing.

What must be recognised is that value systems which help determine the way in which the police make decisions are closely tied to those of the social group to which the police officer belongs.13

The influence of the "consensus view" of policing on thinking about police-community relations can also be seen in a number of related concepts and the way they are used in the conventional discourse of police-community relations. The concern with a police image tends to imply that there is a homogeneity in the experience of policing (and the social order) throughout society. The notion of the police as operationally independent strengthens the idea, which runs through the policing tradition, that the police are the experts in the field of crime and that their work is beyond the legitimate reach of public influence. The tacit assumption or effect of this philosophy of policing is that the police actively pursue a relationship with the public on their own terms. The unquestioned doctrine of police practice based on impartiality and minimum force is presented to the public as the unquestionable and logical outcome of democratic government and law and order. As van Heerden (1986) puts it:

"The favour and approval of the public must be sought at all times, not by pandering to public opinion, but by enforcing the laws with constant and absolute impartiality, giving prompt, individual and friendly service to all members of society regardless of status, social position or national affiliation, being courteous and friendly at all times and being ready to make personal sacrifices in order to save lives."14

A further important feature of this notion of police-community relations is that "the law" (and thus social order) is seen as something external to the community. In American policing, for example, concerns with corruption in the 60s lead to a trend where the police "relate impersonally with communities" and that the source of police authority was to found in "criminal law and police professionalism" rather than in the "political will of the community".15

A central feature of the "consensus" notion of policing is that accountability is primarily to "the law". In addition, the "independence" of the police helps to ensure that they are indeed impartial in the way that they relate to the public - and in the process of investigating crime. This emphasis on independence from political or other undue influence can be related to the emphasis on professional expertise. This, however, means that the police regard themselves as having the exclusive right to determine the nature of policing. The values and laws of central government (the state) are also regarded as having a higher moral standing than the views and customs of specific communities.

The "consensus ideal" tends to place the police on a moral pedestal. The nature of policing tends to be unquestioned and their role is to enforce a universally unproblematic legal system impartially. The emphasis of this view for community policing is on increasing the effectiveness of the police, through enhanced public co-operation on terms defined by the police. In essence this involves decentralisation of policing without diffusion of control.

The concern that community policing (and the loss of operational "independence" on the part of the police) will weaken the police force's ability to uphold law and order arises from the view that the state (as a source of authority for the police) is on a higher moral level than individual communities. For obvious reasons, this view is expressed in its strongest form in totalitarian societies.

Beyond Consensus - The "problem" of community diversity

The alternative view of policing is based on the acceptance of the diversity of communities and hence of social order. Society is made up of diverse communities with contrasting and often conflicting interests. This makes the nature of the relationship between police and society much more complex.

The "consensus" on which social order is assumed to be based on is inherently problematic as it tends to stigmatise or even criminalise differences in values and norms which exist. This has particularly been the case in so-called "divided" societies. Brewer suggests that it is only as far as the white community is concerned that policing in South Africa is actually "similar to the consensus mode associated with policing in liberal democracies".16

This analysis is borne out by the "crisis of policing" in much of the western world over the last two decades. In reality, one of the main reasons for this crisis is the way in which police forces have reflected, and acted in the interests of, the dominant groups in society - to the detriment of their relations and credibility among so called "minority" and special interest groups.

What is important here is the recognition that police culture serves as a conduit for the dominant culture. It serves to enable the interests of the dominant groups to be represented in the actual operational practice of policework. This is further exacerbated in the South African context by the centralised and exclusive nature of police accountability.

The emergent new tradition in policing is based on a concern for the following areas:

(i) A history of the police which recognises the partisan origins and role of the police in protecting certain power relations.

(ii) The consensus at a parliamentary level which leads to law-making is the product of the dominance of certain interest groups and the law tends to reflect these dominant interests. There therefore needs to be a recognition that the law may be perceived to be at odds with community norms.

(iii) A concern for police culture as determinant of policing styles, methods and the focus for favourable or discriminatory policing. The notion of the independence of the police does not necessarily imply that the police are impartial. There is thus an active concern for the representative nature of policing. The police must be made more representative of the community and its values. There also needs to be an evaluation of the values of the police organisation itself.

(iv) Impartiality is relative to the different values and norms within which policing operates. What is impartial in one community will be perceived to be discriminatory in another. The police have to be attuned to the specific values of each community.

(v) Police accountability should include a degree of accountability to the particular community being policed.

These concerns do not mean that the central state, the constitution and the law are undermined in the interests of moral diversity. In the context of a heterogenous democracy it must, however, be recognised that the values and rules under which the state operates are not absolute. This does not mean that they are not important. Indeed the symbols, principles and values of the society as a whole are of vital importance, particularly during the process of nation building. Community policing is in essence an attempt to reconcile national values with the freedom of communities to have their own identities and to propagate different values from those which define the nation.

State Policing, Surveillance and Power Issues
Community Policing involves an increase in State Policing, rather than an increase in community control.
Ultimately the role of state policing should be reduced and not increased. More democratic social order is facilitated by a move towards "privatised policing" whereby communities increasingly take control of their own policing.
This is also related to the concern with the potential of community policing to be used as a more sophisticated form of surveillance and hence of state control. In the absence of a representative democracy this concern is a very real obstacle to community policing. In this view community policing involves an increase in the power of the state and a decrease in the ability of the community to act independently. The alternative of "privatised policing", on the other hand, reduces surveillance by the central state.

Because of the sophistication of community policing, it raises the spectre of increased state power through more advanced "surveillance" and control over the lives of people. This is a particular concern for those of us for whom the notion of coercive social order maintenance is difficult to reconcile with the ideal of democracy. In the South African context where suspicion of the state is justifiable and the commitment of the police to democratic policing is questionable, community policing appears to provide the armoury for increased legitimacy and control without necessarily democratising police-community relations.

Community Policing and Power Relations

In relation to different visions of community policing, it seems that it is the "consensus" model which really poses the problem of increased state control, because it offers decentralised policing without diffusion of state power. The "diversity" approach to community policing premised on a lesser role for the state in the determination of social order appears to offer the potential for such "diffusion" of state power to the local level. The may of course still constitute an increase in state power but this is not necessarily a bad thing if we consider that the power relations along the police-community axis are essentially paradoxical.

This "power paradox" derives from the fact that a real increase community participation in the determination of policing priorities results in an increase in the power of the community. At the same time however this leads to greater police effectiveness and legitimacy which essentially increases the power of the police agency.

The power paradox also works in reverse. The more effective "surveillance" associated with community policing may be seen to constitute a threat to the freedom of the community. The police also have greater accountability to the community which implies a reduction in autonomy - the freedom to use their power unilaterally is therefore curtailed.

Problems of Community Empowerment

Voteless and poor black communities are, in many senses, disempowered. Organisations representing these communities often have a lot of grassroots support, but are weak in terms of resources and skills at their disposal. In order for Community Policing to meet the needs of such communities (in addition to the needs of the Police), the communities need to be able to define and articulate their needs and then negotiate them effectively. What is striking is the enormous power imbalance between many such communities and the SAP. This imbalance makes it unlikely that meaningful community participation will be easily achieved. At the same time the problem of community empowerment is exacerbated by the general weakness of the civil consciousness on policing.

Although policing is a high profile issue, civil society is quite clear on what it opposes, but not so clear on what to replace it with. The exclusivity of the political process has meant that civil society is, in general, quite ignorant of policing issues. Ideas about policing have been largely informed by the policing of the past, and a concept like "community policing" is almost unknown except in academic circles.

This highlights two key issues. The first is the need for public education around policing. The second is the necessity for community policing to be genuinely consultative and aimed at establishing relations of accountability with local communities. It also highlights the inseparability of community empowerment and organisation, from community policing.

The Privatisation of Policing

It has been suggested that a better route to follow is that of privatised policing, whereby various community and corporate interests take responsibility for their own policing. There are two problems with this view.

The first is an economic one. While privatised policing is a viable option for more affluent communities, it is doubtful whether it holds much promise for communities of lower economic status. Indeed the extensive protection of affluent (mainly white) communities by private security firms has possibly served to intensify dissatisfaction with the protection that the formal policing agency provides to the poorer (mainly black) communities. This "relative deprivation" around security feeds into political polarisation and stereotypes and hinders, rather than helps, the process of reconciliation. This is linked to a particularly South African problem relating to the history of the "corporate private policing sector". This sector has historically been closely integrated with the state security system and has often been regarded by black communities as part of the problem of political oppression.17

The second major problem is the question of accountability - beyond accountability to the paying client. Although not given the same authority by society, private security firms have been criticised for acting as though they have the same powers. The use of lethal force by private security guards, for instance, is not adequately controlled. The accountability issue is equally important in relation to private armies and other politically aligned forms of "peoples police". These forms of policing reflect the absence of an enduring social accord. They are narrowly (or not at all) accountable, and tend to represent only particular interests within communities. Because they take so many forms and are usually strongest where formal state policing is weakest, they will prove extremely difficult to control, even if (as we suggest) the formal police agency negotiates working relations with such indigenous forms of "community policing". Their ultimate accountability and control remains tied to the credibility and accountability of the national police force itself.

This is related to a more fundamental political argument against a reliance on private policing at this stage of our history, which is simply that policing is of vital importance to the success of the transition process. A more representative police force has become one of the central political demands of the transitional process. Although not yet with us, an embryonic representative police agency will be an important factor in the resolution of political conflict and a symbol of the new emergent nation. While privatised policing may serve the needs of some communities, it cannot take centre stage as the solution to our policing problem.

The Transitional Context

In the present context, various forms of negotiated agreement involving a wide range of groupings may provide the basis on which community policing can be initiated. The most obvious and important such agreement is the National Peace Accord. While the Accord has recently suffered from an overdose of scepticism, it has, in fact, achieved a great deal.

The National Peace Accord is important because it is the de facto channel through which most practical negotiations between broadly representative community structures and the Police occur. These negotiations, while arduous and difficult, carry the seeds of an emergent community relationship to policing.

It seems that the transition from a consensus idealism model of policing to a philosophy which recognises and acts according to the diversity of community social control is a non-starter without substantial pressure and input from civil society. The lack of effect which public pressure has generally had in South Africa was highlighted by the dramatic response to the "Waddington Report".18 In the last year however the effects of public pressure has started to be increasingly seen - albeit mainly at a national level. However the response to such interventions still takes the form of political "damage control" on the part of the National Party leadership. At the same time, however, the massive pressure for "reform" and improved police-community relations provides the ideal entry point for alternative models and solutions to the problems of policing.

Special Communities - Rural areas

Community policing generally conjures up visions of urban communities. However South Africa is far more rural than the countries whose models we draw on. Rural communities pose special problems of their own in terms of any notion of community policing, although the general features of police organisation are similar to those elsewhere.

In terms of large areas outside of the homelands, the polarisation of people is extreme. Conservative farmers have large groups of labour tenants living on their land who are in a position of extreme powerlessness. In this context violence and fear appear to be increasing.

Tribal areas in the homelands and elsewhere have their own special features. In many areas, tribal structures are breaking down and traditional authority is being challenged. Where these structures are intact and legitimate, they often play a role not far removed from community policing. Tribal police arrest perpetrators, provide the backup to tribal courts and inform chiefs or headmen about social order problems. These systems of indigenous policing offer some valuable insights. One example is the emphasis on reparation as the mode of punishment, so that the victim is empowered through the process.

But what of the future of these structures? The general trend of modernisation is challenging tribal authority in all parts of the country, often under the banner of the ANC. The main reason for their demise is the way in which apartheid policy co-opted the tribal system and thus discredited it. Its order maintenance function has been a source of conflict because of the perceived lack of objectivity on the part of the chiefs. One of the main contributory factors to spiralling violence is the absence of an alternative form of social control. However, it seems that where tribal structures remain intact and continue to be broadly legitimate, the "community aspects" of tribal ordering should be built on and worked with rather than trying to invent new forms of community policing in their place.

Conclusions and Recommendations

The initial enthusiasm felt by the participants at the CCJ conference for the concept of community policing remains valid. However the implementation of community policing in South Africa is no simple task. The key problems to be confronted centre around the fact that the formal police agencies are not accountable to democratically elected government bodies. This means that there is justifiable suspicion on the part of many communities about the strategies adopted to make policing more acceptable. This lack of political accountability also means that other channels of accountability lack general legitimacy. These factors and the central place of policing on the political stage mean that community policing has to be implemented in a far more participative way than that elsewhere.

This report has argued that the concept of community policing has value for a number of reasons:

  • It emphasised the importance of community empowerment;
  • It provides a set of principles against which to measure policing;
  • It is a challenge to South African Policing agencies;
  • It provides an opportunity for facilitators from outside the police to assist in breaking down mistrust and establishing a new relationship between police and community.

The ideas that the group has debated around community policing offer a new and challenging approach to policing in the South African context. In the current context these ideas are likely to have wide appeal and will serve to generate a greater interest in, and awareness of, policing issues. This is precisely what is needed if attempts by formal policing to co-opt and own the notion of community policing are to be countered by communities.

There are 4 different areas in which recommendations are made:

General Organisational Change on the Part of the SAP

A thorough process of organisational change must be embarked upon if the police force is to adapt itself to the needs and strategies of community policing. Such organisational change must take place in the public eye and preferably with the active involvement of a range of outside experts and community representatives, including members of other police agencies.

This process is centrally dependent on the willingness on the part of the police force to relinquish the relationship of exclusivity which exists between them and the National Party.19

Key issues which should be focused on are:

Civilian oversight and accountability - the opening up of the organisation and its plans

Management style and practice
Human resources, particularly promotion and assessment criteria
Culture change

Education and Empowerment of Communities to Engage Constructively with the Police; and Facilitation of Engagement between Police and Community

Community organisation is in a sense a prerequisite to effective community policing. As we have identified, communities are generally disempowered in their relations with the police and their ability to negotiate as equals. Because of the existing attitudes to policing in South Africa, independent agencies have a vital role to play in public education and the provision of skills to community organisations.

It is proposed that a national "Network on community policing" be established independent of the police force. This network would serve as an institute and provide independent capacity to empower communities and educate the broader community about community policing. The network would include all non governmental organisations with a concern for and expertise in issues of policing, community development and community law. The network would serve the following functions:

  • train community leaders;
  • promote debates through publications and public events;
  • provide resources and expertise to local communities which require assistance in their engagement with the police;
  • network and lobby with key political groups and other influential bodies such as the National Peace Accord, to promote acceptance of and support for the principles of community policing;
  • run workshops where local police and community can be taken through a process of discussing community policing together;20
  • provide resources for members of the Police force and local or regional forces, in order that the formal policing agencies may more rapidly come to terms with community policing.21
More Specific Steps to Enhance the Community Accountability and Sensitivity of the SAP

In our context community policing should not be seen as something which follows some form of regional decentralisation of policing, but should be pursued within the current framework.

  • The immediate implementation of a national "needs assessment" programme by the SAP. This might also be termed a detailed environmental scan. This should be done in conjunction with social researchers from outside the police and community policing specialists. The aim would be to provide the SAP with a thorough body of information on the nature of specific communities, different sub-communities and interests and concerns and specific obstacles to co-operative relations.

  • A concerted effort on the part of the police to establish open lines of communication and consultation with major interest groups and political parties in South Africa. This should preferably be done in conjunction with the national needs assessment. The "interest groups" should be prioritised in terms of the community policing strategies outlined in para. 2.6 (particularly objective [I])

  • The new division of Police-community relations, should be redesignated a division of Community Policing. It should be a service division which is tasked with monitoring and facilitating "community policing" approaches in all other divisions. It should also be functionally separate from structures such as the Public Relations Division. Under an interim government such a structure may be directly accountable to a multi-party committee responsible for law and order.

  • Liaison Forums as presently constituted must be reviewed and new criteria for assessment instituted. Criteria should include:

    • the representivity of the forum;
    • the actual responses of the local police unit to suggestions - which lead to the resolution of problems;
    • stability of membership and awareness in the community of issues discussed and responses made;
  • a programme for assessing and establishing working relations with all forms of "organic policing". This should be done in conjunction with experts in the area of these organic community structures;

  • the criteria of assessment of the success of the police force as a whole must be reviewed and altered to reflect the principles of community policing;

  • A discrimination and equal service audit must be conducted in conjunction with independent agencies. This audit will look at criteria for prioritisation, current resource allocation (to regions districts and neighbourhoods) as well as surveying racial attitudes and discriminatory policing within the police force.

  • a publicly visible programme of racial awareness training (and gender awareness) should be implemented across the board.

The Role of Commissions, Expert Panels and International Support

The role of experts, academic and otherwise, both locally and internationally in developing appropriate models and strategies in South Africa should not be underestimated. There are several ways in which such experts could make a contribution:

  • As members of multi-national panels convened under credible local bodies (such as the Police Board or the Goldstone Commission);
  • as advisory groups to Police Agencies;
  • as participants in the independent "network" outlined above;
  • In addition, foreign police agencies themselves have a role to play in making their experience and expertise available, so long as this takes account of the specific South African features described in this report.

Members of the Working Group on Community Policing

Local Members:
  • Etienne Marais, Centre for the Study of Violence and Reconciliation
  • Doreen Atkinson, Centre for Policy Studies, University of the Witwatersrand
  • Chris Botha, Police Science Dept, University of South Africa
  • Michael Cowling, Centre for Criminal Justice, University of Natal (Pietermaritzburg)
  • Saras Jagwanth, Centre for Criminal Justice, University of Natal, Pietermaritzburg
  • Mabel Makibelo, Criminology Dept, University of Venda
  • Janine Rauch, Centre for the Study of Violence and Reconciliation
  • Wilfred Scharf, Institute of Criminology, University of Cape Town
International Members:

Cape Town Meeting 31 July 1992

At this meeting the draft report and a discussion paper by Doreen Atkinson were presented and discussed. The following people attended this meeting:

  • Doreen Atkinson, Centre for Policy Studies, University of the Witwatersrand
  • Chris Botha, Department of Police Science, University of South Africa
  • Robert Cameron, Politics Department, University of Cape Town
  • Beattie Hofmeyr, Institute of Criminology, University of Cape Town
  • Saras Jagwanth, Centre for Criminal Justice, University of Natal (Pietermaritzburg)
  • Mabel Makibelo, Criminology, University of Venda
  • Etienne Marais, Centre for the Study of Violence and Reconciliation (Convener)
  • Janine Rauch, Centre for the Study of Violence and Reconciliation
  • Jeremy Seekings, Department of Sociology, University of Stellenbosch
  • Annette Seegers, Department of Politics, University of Cape Town
  • Clifford Shearing, Community Law Centre, University of the Western Cape
  • Graeme Simpson, Centre for the Study of Violence and Reconciliation
  • Steph Snel, Urban Monitoring Action Committee
  • Elrena van der Spuy, Department of Sociology, University of Stellenbosch


1 This report covers policing in South Africa. This includes the TBVC states and self-governing territories. In general we consider the features of policing to be similar in all these police agencies, and the report is applicable to all 11 police agencies. Where this is not the case it is specifically stated.

2 Solicitor General 1991.

3 Brown, L. 1991.

4 The Code of Conduct for political organisations in the National Peace Accord provides useful principles against which such structures could be measured. (National Peace Accord 1991).

5 This would of course be greatly facilitated through a process of "decriminalisation" of certain activities which are considered in communities to be quite acceptable. This proposal was widely supported at the HSRC conference on Crime Management in the New South Africa (Pretoria 4-6 August 1992).

6 Minister of Law and Order 27 August 1992.

7 For a black person as head of this new division, they would have to be promoted twice - from Brigadier to Major General; then from Major General to Lt-General.

8 Most homeland areas include people from a range of tribal and ethnic origins: Transkei, for example has significant populations of Sotho and Zulu speaking people.

9 This section is based primarily on a discussion paper by Doreen Atkinson, presented to the Community Policing Working Group, 31 July 1992.

10 From Wilmot (1987) cited in Shapland et al.

11 Independent Law Enforcement Facilitation Office - established in Natal on the initiative of IDASA.

12 Jefferson (1991) for instance argues (in the British context) that the "impartial history" of the police is in fact a myth. Impartiality and the process of exercising discretion, combined with the shortage of resources, mean that the police, at both management and street level are forced to make decisions about where they should place their energies which are of necessity subjective and hence "partial".

13 In the case of Police members who are drawn from "minority groups" (or in the SA context, the oppressed group), the dominant organisational culture tends to exercise a profound influence on the specific identity of the individual. Police from such minority groups (or in South Africa black police) do not necessarily behave in ways which are significantly different from their culturally dominant colleagues.

14 Van Heerden, 1985:78

15 Kelling 1986:12

16 Brewer 1990:84

17 A mine security firm for instance patented its own rubber bullet.

18 Dr P.A.J. Waddinton: "Boipatong Enquiry".

19 This also applies to the police forces of the TBVC states, self governing territories and their ruling parties.

20 Members of the Community Policing Group have already initiated such projects. These include workshops in Venda, Pietermaritzburg and a national workshop together with the Institute for Multi-party Democracy.

21 The Centre for the Study of Violence and Reconciliation recently ran a workshop on Community Policing for the Bophuthatswana Police.


Atkinson, D. 1992. Problems of Governance and Public Participation: Issues for the South African Transition. Centre For Policy Studies - Discussion paper.

Brewer, J. 1990 Policing In The Elusive Search for Peace: South Africa, Israel, Northern Ireland (eds): H Giliomee & J. Gagiano. Oxford University Press, Cape Town, 1990.

Brown, L.P. 1991. Policing New York City in the 1990s: The Strategy For Community Policing, January, 1991.

Jefferson, T. 1990. The Case Against Paramilitary Policing, Open University Press, London, 1990.

Kelling, G. 1985 The Changing Function of Urban Policing. In Community Policing in the 1980s - Recent advances in police programs, Solicitor General, Canada, 1985.

National Peace Secretariat. 1991. National Peace Accord (14 September 1991).

Shapland, J. & Vagg, J. 1988. Policing by the Public Routledge, London, 1988.

Solicitor General, Ontario. 1991. Community Policing - An introduction to the community policing philosophy and principles, (From Community Policing Manual).

Van Heerden, T. J. 1986. Introduction to Police Science, University of South Africa, Pretoria.

Waddington, P.A.J 1992. Report of the Enquiry into the Police Response to, and investigation of, events in Boipatong on 17 June 1992. Goldstone Commission, Pretoria.

© Centre for the Study of Violence and Reconciliation

Copyright © 2020 CSVR. All Rights Reserved.