Centre for the Study of Violence and Reconciliation

Provincial Government Oversight of the Police

by
Gareth Newham & David Bruce

Research Report written for the Gauteng Legislature, April 2004

Gareth Newham is a former Project Manager in the Criminal Justice Programme at the Centre for the Study of Violence and Reconciliation.

David Bruce is a Senior Researcher in the Criminal Justice Programme at the Centre for the Study of Violence and Reconciliation.

Abstract

This report, commissioned by the Gauteng legislature, seeks to add clarity and enrich debate as to role of the executive and legislative structures at provincial level with respect to oversight of the police. Using international literature, it sets a conceptual framework for understanding how police agencies are held accountable in democracies, and why oversight of the police service is substantially different to that over other government agencies. Towards this end, an argument is made that police have to be held accountable for both their performance (what they do) and their conduct (how they do things). Key elements of policing as a particular profession are presented so as to highlight specific challenges for external civilian oversight structures tasked with holding the police accountable.

The report then reflects on the executive and legislative oversight structures in the following countries: England, Wales, Australia, Kenya and Nigeria. In doing so, the report reveals that there is not a single model to be followed. By comparison with international examples, provincial governments in South Africa are unique in that their formal authority over the police is exercised indirectly through the National Minister. The report then closely examines the constitutional framework for police accountability at provincial level in South Africa. In particular, the report highlights the legal and political authority of the Member of the Provincial Executive government (MEC) and his or her Secretariat tasked with police oversight.

Focusing on Gauteng the report then turns its focus to the role of the provincial legislature in overseeing the police and secretariat. Particular attention is paid to three key oversight activities that the legislature's standing committee on Public Safety plays with regards to police oversight. These include passing the Provincial Secretariats budget, the posing of oral and written questions to the MEC, and the briefings the committee receives from the South African Police Services (SAPS) provincial management. The report concludes that the most useful information the legislature standing committee receives with regards to its oversight role comes from the SAPS itself. In contrast, the information received from the Secretariat about its role in monitoring the police and the answers from questions posed to the MEC about the police contain inadequate detail to allow for effective police oversight and accountability.

In conclusion the report presents six recommendations that could enhance the oversight role of the provincial legislature holding both the provincial secretariat, and the police agencies working in the province, to account.

Contents

1. Introduction

2. Towards Effective Police Accountability in Democracies
   2.1 The Focus on Police Performance and Police Conduct
   2.2 Working with Internal Systems for Police Control
   2.3 The Multi-Agency Approach to Police Accountability
   2.4 Operational Independence
   2.5 The Need for Objective Indicators

3. Government Oversight of the Police: International Case Studies
   3.1 England and Wales
   3.2 Australia
   3.3 Kenya
   3.4 Case Study Conclusions

4. The Legislative Framework for Police Oversight at Provincial Level
   4.1 The Constitutional and Legal Framework
   4.2 The Political Accountability of the MEC
   4.3 The Legal Accountability of the MEC
   4.4 Obligation of the MEC to Answer Questions
   4.5 Accountability of the MEC regarding Municipal Police Services
   4.6 Oversight of the Metropolitan Policing Services

5. Role of the Legislature and Police Accountability
   5.1 The MEC and the Provincial Secretariat for Safety and Liaison
   5.2 Legislature Questions to the MEC
   5.3 Reports from the SAPS Provincial Commissioner

6. Recommendations for Strengthening Police Oversight
   6.1 Clarifying the Purpose and Nature of Legislative Oversight of the Police
   6.2 Working with Internal Police Structures
   6.3 Working with Other Structures and Agencies
   6.4 Using Key Performance Indicators
   6.5 Detailed Oversight of the MEC and Provincial Department
   6.6 The Effective Use of Research

7. References

1. Introduction

With a decade of democratic governance under the belt, this report seeks to reflect on the role of provincial government structures in holding South African policing agencies accountable. Mindful of the police excesses of the past, the drafters of the South Africa's new democratic constitution ensured that the police service could be held accountable through a range of different structures. Importantly, the new South African Police Service (SAPS) had to be accountable not only to the law, but also to elected leaders through the executive and legislature structures of government. As a single national agency, the SAPS is firstly accountable to the National Minister of Safety and Security through the SAPS National Police Commissioner. A parliamentary committee of the National Assembly is responsible for overseeing both the ministry and the police service.

While a similar arrangement has been designed at provincial level, these executive and legislative structures do not have the same formal power over the police as is exercised at national level. Consequently, many questions have been raised as to how best provincial governments can exercise oversight of the police agencies operating in their field of jurisdiction.

This report was written in response to a request from the Gauteng Provincial Legislature to undertake research that assists in answering the question: 'To what extent can the Legislature exercise control in respect of policing matters?' The brief in this regard, was to explore two principal issues.

Firstly, the Legislature wished to clarify the responsibilities of the MEC for Safety and Liaison regarding police in Gauteng province. In this regard, the Legislature specifically queried:

Secondly, the Legislature wished to clarify the extent and nature of its own authority and responsibilities in relation to the MEC for Safety and Liaison, and policing generally. In this regard the Legislature asked:

The Legislature has also specifically asked for these issues to be illuminated by means of relevant and comparative examples from other countries. A request was also made for recommendations that could assist the Legislature in enhancing its oversight role of the police.

In light of these directives, this report investigates how democratic governments exercise control over the police through their executive and legislative arms. To this end the report includes:

Furthermore, a brief analysis of the current roles played by the Provincial Executive and Legislature in terms of police oversight will be presented. This serves to contextualise the current work of these structures and assists in providing practical recommendations as to how oversight of the police can be enhanced. This report is structured as follows:

Section One

A conceptual model for oversight of the police is briefly presented. This draws on work conducted over the past few decades by international analysts focusing on police accountability in democracies. The section presents four key principles that are relevant to the effectiveness of oversight.

Section Two

The second section focuses on structures and systems of executive and legislative oversight of police. The countries from which comparisons were drawn include England and Wales, Australia and Kenya. While South Africa is fairly unique in terms of national and provincial oversight of its national police service,1 it does embrace aspects of Commonwealth traditions. Insights from these countries can assist South Africa in reflecting on the strengths and weaknesses of its oversight systems.

Section Three

The third section of this report presents the constitutional and legislative framework guiding police oversight by provincial government structures. This section draws on a legal opinion, which addresses some of this report's key questions.

Section Four

The fourth section provides an analysis of the Provincial Department of Safety and Liaison's and the Gauteng Legislature's approaches to oversight of the police. This moves the focus from the legal framework to the current realities of police oversight and reflects on the role played by these structures.

Section Five

The final section presents a number of key recommendations as to how the Legislature can practically enhance its oversight function of the police in the Gauteng Province. It draws on the above analysis, as well as an understanding of the constitutional and legal frameworks.

In this report the focus is on the authority exercised by the Provincial Government over the South African Police Service (SAPS) and the Metropolitan Police Services (MPS). Municipal police services currently operating in Gauteng are the Tshwane, Johannesburg and Ekurhuleni Metropolitan Police services. It should be noted, however, that there are a wide range of other bodies in Gauteng and in South Africa that are involved in performing what could be considered policing functions.2

2. Towards Effective Police Accountability in Democracies

Given the complexities related to police work and the management of police agencies, oversight bodies can easily become overwhelmed by the number of issues they could give consideration to. It is therefore important that oversight bodies are clear as to the reasons why they do such work and where the key focus areas should be. This section provides a conceptual framework for guiding the work of oversight structures in democracies. Furthermore, some key principles are highlighted in order to guide the Gauteng Legislature in terms of its role in police oversight.

Ensuring police accountability is a complex and difficult undertaking in any society, but perhaps even more so in a period of intensive police reform. During the process of societal transition, oversight bodies can play a crucial role in monitoring, evaluating and guiding the transformation of the police. However, even in societies that have ended their transition phases, oversight of the police remains a key concern. There is much evidence that even in stable, modern democracies the police are prone to abusing their powers. In many instances abuse leads to major public scandals resulting in increased mistrust of the police.3

While South Africa has come a long way since the birth of its democracy ten years ago, it still confronts substantial challenges when it comes to safety and security. It may be that effective crime prevention and reduction requires more than simply adequate policing. The police can only ever react to the consequences of crime, as opposed to tackling the multitude of important social fabric factors that drive criminality, e.g. poverty, unemployment, inequality, education levels and family cohesion. Despite this, the police are the most public face of Government's efforts to ensure law and order and to provide services that enhance the safety and security of society. Strengths and shortcomings of the police, therefore, have direct implications for the way citizens perceive and respond to state authority. It is for this reason that, since the 1960s, so much work has been done on understanding and strengthening police accountability in other countries.

2.1 The Focus on Police Performance and Police Conduct

As democracies grow and develop, public expectations of the police are likely to shift. In South Africa, a key focus of government shortly after the first democratic elections was to shift public attitudes towards accepting the general credibility and legitimacy of the national police service (Rauch, 2000). Ultimately however, the credibility of the police will depend on the extent to which people believe they are able to deal effectively with crime and disorder. This does not mean that the police can 'do whatever it takes' to achieve these objectives. People will also be concerned with how the police behave while they attempt to achieve these objectives. If the police are incompetent, brutal or corrupt, the general public will lose respect for them and the extent to which the police can control crime (or achieve any other objectives) will be severely compromised. It is as important for ordinary people that the police do their jobs professionally, within the law, and with respect for those they interact with. These two components of police accountability in democracies have been termed the 'double demand', as 'Citizens demand that police protect them but do so only legally and respectfully' (Stone, 2004, p. 1).

For oversight structures then, the primary concern is that police should be held accountable for both their performance and their conduct.4

This 'double demand' presents democracies with a number of key challenges in relation to oversight of their police agencies (Stone, 2004).6 These challenges arise partly out of the nature of police work and partly out of the organisational culture of most police agencies. The key challenges can be summarised as follows:

  1. The link between police activities and crime prevention is not always clear or straightforward. Following an assessment of scientific evaluations of various policing crime prevention strategies and tactics in the USA, it was found that there was little evidence to show that some common policing tactics have much impact on crime. One study for instance, found that cutting police travel time from 5 to 2.5 minutes would have almost no impact on violent crime arrests, partly due to the fact that victims took an average of 41 minutes to report the offences, nullifying the potential benefits of rapid response (Sherman, 1998, p. 8).It is therefore important that police agencies are able to collect data on their activities (key performance indicators) and demonstrate as much as possible the link to their core objectives (key impact indicators).

  2. Many police officials do not believe that ordinary people really care about how they act while they are doing the job of policing (Stone, 2004). Often their personal experience will be that many victims of crime and even some of their supervisors are quite comfortable that they break a few regulations in the course of showing a criminal suspect (or anybody who challenges them for that matter) 'who's boss'. As long as this type of behaviour is not confronted it will become part of the culture of the way the police do their work. Research into police sub-culture demonstrates that members of police agencies will develop their own norms and standards that are not necessarily constrained by the formal rules and regulations (Reiner, 1985). Without an external structure to mediate police sub-culture, very problematic abuses of police power can emerge, such as those experienced during apartheid and which, in one form or another; have continued into our new democracy (e.g. police corruption).

  3. Policing professionally, within the law and regulations is more difficult than policing without such constraints (Stone, 2004). This does not mean that policing outside of the law is more effective because of the better results that are achieved. Rather, policing outside of the law may mean that certain results are achieved more quickly. Simply put, it may be easier for a police official to use or threaten violence to get people to respond to police demands rather than enter into a discussion with them. Similarly, it may be easier to locate a criminal suspect through torturing a close friend of his as opposed to organising a stake-out of his home. As a result there is constant pressure amongst police members to subvert the law and police regulations to achieve quick results. However, over the course of time policing outside of the law causes problems that undermine the effectiveness of the police. For example, ordinary people may start to fear or mistrust the police, as they do criminals, and will therefore be less likely to cooperate with the police to tackle crime.

  4. It is difficult for most people and government officials to know much about how police officials are doing their work (Stone, 2004). Police members usually work in small groups and out of sight of their senior commanders. It is therefore easy for police officials to give poor service or become involved in corruption as they are unlikely to be held immediately or directly accountable. The nature of police work as described above also makes it easy for commanders to 'turn a blind eye' to problems that occur. This is the reason why police commanders often state that they 'didn't know what was going on', or that an incident was an aberration as a result of 'bad apples' if problems suddenly come into the public eye.

It is the role of external oversight bodies, wherever they are placed, to solve these dilemmas and ensure that inadequate performance, poor service delivery, and abuses of police power are kept to a minimum. This should be borne in mind by any structure that has a role to play in overseeing and ensuring the accountability of police agencies.

2.2 Working with Internal Systems for Police Control

As police agencies undergo reform there are real dangers that levels of discipline and control may break down. As the old culture dissipates often it is not immediately replaced with a new sense of professionalism and pride. During this period police corruption and abuse of power may well continue and even increase. If these are not directly dealt with by the police agency, its ability to effectively enforce the law and respond to victims will invariably be undermined, as will police-community relations. It has therefore been strongly argued that,

'the early establishment of functioning internal controls is all the more important for police reform processes where a large number of personnel are retained from the former [security structures]. In these cases, there is clearly an increased risk that old and abusive practices may continue. It is undoubtedly important, if the reform process is to be at all credible, that police leadership clearly demonstrates that abuse will not be tolerated by developing the mechanisms to confront it' (Neild,1998, p.1).

Certainly current international trends suggest that increased focus must be placed on internal disciplinary and 'early warning' management systems. These systems must be established in such a way so as to alert senior managers and supervisors to potentially problematic behaviour that requires focused management intervention.

If small issues of discipline are overlooked in police agencies, serious misconduct and corruption will soon follow.7 International experience and research on police corruption and criminality also indicates that strong and well-resourced police internal investigation units are critical in order to ensure that the more serious cases of abuse are dealt with quickly and thoroughly.

Civilian oversight bodies, such as the Independent Complaints Directorate (ICD) can never be responsible for handling all complaints against the police. Not only would they require much more capacity, but experience shows that if an outside agency takes all the responsibility, then the police agency ceases to take responsibility. Nevertheless, outside agencies have an important role to play in reassuring the public that internal investigations against the police are carried out thoroughly and competently. However, arguments have been made that external structures can never be as effective as internal systems. As Punch (1983, p. 249) argues, 'no external body could possibly fully understand and penetrate the entrails of the [police] organisation'. Similarly, Kersetter (1985, p. 178) emphasises that, 'The [police] departmental administrative structure has by far the greatest potential for effective action to prevent, investigate, to adjudicate or to punish police misconduct'.

It is therefore critical that any oversight structures carefully engage with and monitor the police systems for identifying, investigating and dealing with misconduct and other problematic behaviour. Only if there is confidence that there are effective internal police systems in place to deal with problematic police officers and to deter others from abusing their policing powers, will the public begin to trust and improve its relationship with the police. As Reiner (1985, p. 178) proposes, 'Accountability institutions will only be truly efficacious in affecting police practices if they win over and work in conjunction with internal disciplinary and self controlling processes. They cannot be forced by a heavy hand.' Where possible, oversight structures should ensure that internal bodies are well resourced and capacitated to take the necessary action against problematic police officials, and should hold them accountable for doing so.

2.3 The Multi-Agency Approach to Police Accountability

In modern democracies such as South Africa, the police are held accountable by a number of different structures. According to Bayley (1997), democratic police agencies are expected to respond to the needs of individuals, groups and government. As a result, a democratic police agency is by necessity accountable to multiple audiences through multiple mechanisms.

In democracies, the police are typically directly accountable to state executive and legislative structures. These structures consist of elected representatives whom have been given a mandate to govern on the behalf of the majority of the citizens. Such structures are responsible for developing policy, passing legislation and approving the budget of the police agency. However, in democracies there is also a range of other state and civil-society structures that are involved in police accountability.

Within South Africa these include:

Similarly, outside of the state there is a wide range of institutions, organisations and structures that contribute to the oversight of the police. These include:

While it is a positive feature of democracies that there is such a wide variety of external structures concerned with police accountability, their existence does not necessarily imply that they play an effective role.

Analysts such as Stone and Ward (2000, p. 11) argue that while there may be no single correct approach to ensuring police accountability, better coordination and stronger linkages of the different oversight structures will assure more 'robust and effective accountability'. It is therefore important that provincial level police oversight structures see themselves as part of a larger tapestry of oversight mechanisms, and consider ways of improving liaison and coordination between them.

2.4 Operational Independence

One of the key concerns relating to policing in democracies is that of 'operational independence.'8 In South Africa this principle is relevant to ensuring that government does not use its authority over the police to promote political partisan or other interests that might be contrary to the spirit of the Constitution. Examples include using the police for illegitimate ends, such as protecting political leaders from investigation and prosecution, or directing the police to monitor and harass political opponents.

Misuse of the police is a common reality in dictatorial regimes (such as South Africa under apartheid), where the police are used to promote political party objectives as opposed to providing equitable services to all communities. However, the misuse of police can also easily occur in democracies. As Walker (2001, p. 9) highlights, for most of the past two centuries, many local level elected officials in the United States of America "… thought about the police primarily in terms of potential opportunities for graft or patronage."

South Africa's Constitution emphasises that the police and other security organs should not favour political party interests. Section 199(7) states that:

'Neither the security services, nor any of their members, may, in the performance of their functions –
(a) prejudice a political party interest that is legitimate in terms of the Constitution; or
(b) further, in a partisan manner, any interest of a political party.'

The need for all public servants to act impartially is also dealt with in Section 195(1) of the Constitution, which deals with the basic values and principles governing public administration. However, while this seems relatively simple, in practice it is not.

For instance, a taxi strike in Johannesburg is an example of industrial action that has the potential to substantially disrupt the city's economy. One of the demands of the taxi owners may be that police stop 'harassment' of the taxi industry which they perceive as the police frequently pulling taxis over at road blocks and issuing fines. However, if the taxi industry is responsible for many road deaths due to poor vehicle conditions and driving habits, and has been associated with internecine violence, it might make sense that they are given greater attention by the police than other road users. Questions then arise about who should make decision about law enforcement against taxi-drivers – the police or politicians? Clearly this becomes a question of public policy that goes beyond simple 'operational matters'.

At the most general level the 'Operation Crackdown' policy adopted by police since 1999 has been implemented partly in relation to the Government's drive to be seen to be acting decisively against crime at a time when the publics attitude towards the SAPS was generally negative. The focus on violence against women has also not been purely determined by the police themselves, but reflects the concerns of government, as well as the general public. Decisions to focus on, or ignore, specific crime problems – such as brothels and prostitution, drug laws, firearms, or the policing of white-collar crime – are therefore often influenced by the demands of government and the general public.

It seems that while on the one hand the police need to be protected from inappropriate political interference, on the other hand they need to be responsive to legitimate government concerns. Democratic governments therefore have a right to guide what the police do in some ways, but not in others. Typically this means that government will develop policies that the police will then operationalise through training and resource allocation. However, direct instructions from politicians to arrest certain individuals or to focus resources towards certain communities over others could easily be seen as political interference.

One way of addressing this tension is to ensure greater transparency. Bayley (1994, p. 155) offers that, 'Accountability is best obtained through open processes of evaluation not through directed policies.' Government, it is suggested, should evaluate the effectiveness of the police in achieving broad policy objectives as a means to accountability rather than whether issuing specific instructions regarding 'operational' matters. Furthermore, to prevent the types of problems associated with political interference, the objectives and demands made by the Government of the police have to be transparent and publicly known. All specific instructions issued by the Minister or MECs should be presented in a report before the relevant parliamentary committee. The failure to notify Parliament of executive instructions given to the police should be recognised as a breach of Government ethics.

In democracies, the police are supposed to be professional public servants who develop and maintain a high level of expertise in their work. Consequently, politicians should not issue direct instructions relating to the day-to-day work of the police. Rather, democratically elected structures should set broad policy objectives and leave the details of how best to deploy resources and strategies to achieve these objectives, up to the police commanders. The work of state oversight structures becomes then to ensure that police management demonstrates, in an informed and clear manner, how the objectives have or have not been achieved. If these objectives have not been met, they should be expected to submit reports clearly explaining why this is the case and how they intend dealing with the shortcomings.

2.5 The Need for Objective Indicators

If it is accepted that oversight of the police requires performance and conduct to be monitored, then, where a police agency is undergoing reform (as is the case in South Africa), oversight bodies should be involved in tracking developments and changes on a regular basis. Rather than relying on verbal feedback from police commanders, stating that, 'changes are happening' or 'things are improving', empirical indicators are needed against which objective assessments of fact can be made. Then, if police management were to say that service delivery, police discipline or community-police relations were showing signs of improvement objective measures should be presented to support such assertions.

Executive and legislative bodies would then be in a position to independently assess whether the information or data supplied by police management was valid and reliable. Furthermore, in identifying and agreeing on key performance indicators, oversight structures could then monitor these indicators over time to assess whether policies adopted by the government to affect police reform were:

a) being implemented and;
b) if implemented, establish whether or not they are having the desired effect.

The existence of such indicators would also assist both police management and oversight structures to identify where particular problems were being experienced. One of the realities of reform in large organisations such as the SAPS is that change seldom, if ever, occurs consistently across the entire organisation. There will always be 'pockets of resistance' and the quicker that these can be identified and rectified, the better.

A hypothetical example of this type of interaction follows. The police states in a presentation or report before the Executive or the relevant Legislature Portfolio Committee, that the conduct of police officials is improving. To support this assertion, they may point to a decrease in the number of disciplinary hearings that were held during the previous year. Oversight structures should be in a position to determine firstly, the reliability of this statement (i.e. whether indeed the numbers of disciplinary hearings actually decreased from the previous year); and then, if this is the case, to establish the validity of the statement (i.e. is this change in numbers actually an indicator of improved police conduct?). Through calling for further information or commissioning independent research, the Legislature could establish answers to these questions.

It may emerge that while the number of disciplinary hearings decreased, the numbers of complaints made against the police have actually remained the same. The findings may point to poor quality internal investigations, resulting in fewer of these cases resulting in a hearing which is the actual reason behind the decrease in numbers of disciplinary hearings (as opposed to police officials behaving better). It could also mean that the system is improving so that fewer unsubstantiated complaints are ending up in a hearing. Research could also point to specific stations where problems exist. For instance it could be that in most stations there has been a decrease in complaints leading to fewer disciplinary hearings, but in specific precincts the numbers of complaints and disciplinary hearings have actually increased. An analysis of objective indicators (i.e. the charges presented at these hearings) could point to a particular trend of problematic behaviour by officials at these stations.

The approach that needs to be adopted by oversight bodies is the identification and use of specific and measurable performance indicators that can objectively reveal key changes in organisational performance and conduct over time. Without objective data, it is impossible for oversight bodies to determine whether policies are being effectively implemented, and if so whether they are achieving the desired results. Moreover, it becomes very difficult to tell whether or not the budget and resources allocated to the police are being utilised effectively and efficiently and what the precise effect of an increase or decrease in the budget on resource allocation will be.

3. Government Oversight of the Police: International Case Studies

This section provides comparative examples of legislative and executive oversight of police in selected democracies. While an attempt was made to identify and gather information on countries with similar constitutional arrangements to South Africa, it appears that South Africa is fairly unique in that the formal authority of provincial governments over the police is exercised indirectly through the National Minister.9 Nevertheless, as much of the country's policing policy has been influenced by Commonwealth countries, valuable insights can be gained by comparing South Africa to some of these countries.

3.1 England and Wales
'Tripartite structure'

Unlike South Africa, which has a single national police force, England and Wales have 43 police forces or 'Constabularies'. Governance of most of the constabularies (a separate system applies to the police in London) is exercised through what is known as the 'tripartite structure'. This is divided among:

(i) the chief constable responsible for 'direction and control',
(ii) the local police authority responsible for 'maintaining an adequate and efficient' force, and;
(iii) the Home Secretary, with overall concern for the efficiency of policing throughout the country (cited in Reiner, 1996, p. xxi).
Doctrine of 'constabulary independence'

The idea of operational independence is expressed in its strongest form in England and Wales through the doctrine of 'constabulary independence'. In terms of the doctrine 'each constable enjoys independent authority derived from direct appointment under the Crown' (ibid, p. xviii).

The doctrine was interpreted to mean, 'that while the police may be paid by the local authority (and central government) this did not imply a 'master-servant' relationship. In other words, the financing authority could not instruct the police as to how to exercise its powers' (ibid, p. xxi). This was because, as the police were the upholders and enforcers of the law, it was only the law that should provide direction.

Court judgments affirmed the doctrine indicating that, 'in carrying out their legal duties, the police were accountable to no-one but the law; short of a complete failure to enforce the law, the courts would not question the professional exercise of police judgment' (ibid, p. xxii).

The local police authorities

Not only did the doctrine of 'constabulary independence' have powerful legal and ideological force, but historically other factors also detracted from the effectiveness of local police authorities in holding police accountable. 'For the most part, police authorities habitually deferred to what they saw as the professional expertise of the chief constable. They took their lead from him without question; in effect, a gentleman's agreement prevailed whereby they routinely accepted the police view on operational issues' (ibid, p. xxii).

In the 1980s radical Labour local authorities began to challenge the 'gentlemen's agreement' and demand stronger models of police accountability. As opposed to the 'explanatory and co-operative' type of accountability being practiced by local police authorities, these more radical authorities motivated for an approach in terms of which police were 'subordinate and obedient' to democratically elected authorities. However, these efforts were defeated by the Thatcher government, which disbanded the Metropolitan authorities that formed the main political base for these efforts.

The 1994 Police and Magistrates' Courts Act shifted the balance further away from elected local representation. Up until this point, elected councillors comprised two-thirds of the Police Authority membership. Currently however, elected representatives are in a majority by only one person (O'Rawe and Moore, 1997, p. 139).10 The Act also gave these oversight structures the new responsibility of publishing an annual local police plan containing policing objectives that take into account the views of the public following consultation.11

However, while the power of the Local Police Authorities to define local policing objectives gives them some amount of direct authority over the police, this is entirely subject to the authority of the Home Secretary. As defined by the Police Act 1996, the powers of the Home Secretary include:

The local police authorities can therefore exercise some level of direct authority over their local police agencies only as long as it is consistent with frameworks set down by the Secretary of State, and within the parameters set by the concept of constabulary independence.12 The local police authority may therefore, subject to the approval of the Secretary of State, appoint the chief constable, and call upon the chief constable to retire in the interests of efficiency or effectiveness (S11). At the end of each financial year, the chief constable's report is submitted to the police authority (S22).

The Police Act of 1996 can be seen as the result of a strong trend which emerged during the 1980s and 1990s, towards bringing the 'disciplines of the market to bear on the conduct of policing', through fixed-term appointments for all police officers and performance related pay. Powerful economic incentives started to pressurise the formally autonomous police to pursue the objectives specified by the authorities who controlled their job tenure and pay (Reiner, 1996, p. xxiii).

As a result of these developments Reiner (1996) indicates that power over policing can be seen as primarily divided between the police themselves and central Government. The disempowerment of local authorities was further accentuated by the development of forms of high-level international police cooperation that served to transfer 'power away from the nation-state' to transnational bodies (Reiner, 1996, p. xxiii).

3.2 Australia

Along with the Australian Federal Police, there are seven other police services each covering a different geographical jurisdiction, namely New South Wales, Queensland, South Australia, Victoria, Western Australia, Tasmania and the Northern Territories. With the exception of the federal police these police agencies fall under the governments of the respective states or territories.

Australia, as a colonial power, initially adopted the English model of policing and embraced the doctrine of 'police independence'.. Incorporated into the Australian legal system via common law the doctrine of 'operational independence' was generally understood to cover:

The entire range of policing decisions including what crimes to concentrate on and what crimes to turn a blind eye to; in which areas to deploy police; how to deploy them; how to handle order maintenance problems, and so on (Pitman, p.4-5).

In terms of this understanding of 'operational independence' the task of Government as represented by the police minister, was initially defined as primarily ensuring that police departments were sufficiently resourced. However, in terms of concepts of governmental and ministerial accountability, police ministers were accountable for policing performance and thus sought control over the administrative and operational use of resources within their departments.

While state governments had tended to observe self-imposed restrictions out of deference to the 'operational independence' of the police, in recent decades these limits began to be increasingly tested, with a series of conflicts emerging between police commissioners and governments (Pitman, 1998; Goldsmith, 1993). Consequently, efforts have been made to regulate, through legislation, the relationship between the police minister and police commissioner.

In Queensland, for example, Section 4.6 of the Police Service Administration Act 1990 provides that the Minister after first consulting with the commissioner, can give directions to the commissioner in writing, concerning:

(a) the overall administration, management, and superintendence of, or in the police service;
(b) policy and priorities to be pursued in performing the functions of the police service; and
(c) the number and deployment of officers and staff members and the number and location of police establishments and police stations.

The Minister is required to keep a register of all directions given to commissioner and is required to submit the register on an annual basis to the Crime and Misconduct Commission and Parliamentary Crime and Misconduct Committee.13 The Minister can also require the commissioner to submit reports to him/her.

The responsibilities of the police commissioner are set out in Section 4.8 of the Act, though the section states that it is not intended to limit the extent of the commissioner's responsibility.

In a similar manner, Section 6 of the Police Act 1998 in South Australia provides that, 'Subject to this Act and any written directions of the Minister, the commissioner is responsible for the control and management of the police', while Section 8 provides that the Minister, 'must cause a copy of any direction given to the commissioner to be –

(a) published in the Gazette within eight days of the date of the direction; and
(b) laid before each House of Parliament within six sitting days of the date of the direction if Parliament is then in session, or, if not, within six sitting days after the commencement of the next session of Parliament.'

Section 7 of the Act prohibits the Minister from issuing directions to the commissioner 'in relation to the appointment, transfer, remuneration, discipline or termination of a particular person'.

In addition to defining the scope of ministerial direction on police matters, setting out the responsibilities of the commissioner, and setting out the terms of parliamentary or other oversight of police/government relationships, legislation in various Australian states also regulates the terms of appointment and dismissal of the police commissioner.

While Australian legislatures have worked towards regulating the relationship between police ministers and commissioners, Pitman (1998) argues that attempts at regulating the relationship are inherently limited in nature. Pitman (1998, p. 59) cites Finnane who, 'believes that police commissioners have always been partners with ministers in the development of policy', and that as much as policing decisions have been influenced by government ministers, the police themselves have also 'played a role in policy making and law making'.

The quality of the relationship, and of the contribution of each of the parties, will also depend on a number of factors. For instance, 'one factor in determining the minister's capacity is the knowledge he might have of his portfolio before he got the job' (Weller & Grattan, cited in Pitman,1998, p. 66). Pitman therefore argues that, rather than the commissioner being seen as directly subordinate to, or independent of the minister, the best type of policing system will be the result of a, 'combined effort of cooperative achievement' using both parties' talents and abilities' (1998, p. 87), 'where the minister/commissioner relationship is defined in terms of shared or mutual accountability' (p. 90), and 'which is based on the transparency of a decision-making process which is open to public scrutiny' (p. 100).

While transparency is the best way to ensure accountability, there will always be ways in which the minister/commissioner relationship can limit itself to public scrutiny. Pitman (1998, p. 67) highlights that, 'whatever the statute prescribes, there will always be police commissioners who will not insist on securing such a [written] clarification, or ministers seeking greater informal control by not defining responsibilities and accountabilities for policy and administrative decisions.'

3.3 Kenya

The draft Kenyan Constitution, which was approved on 15 March 2004 by the National Constitutional Conference, is clearly influenced by the South African Constitution. In a clause which borrows directly from Section 199(7) of the South African Constitution the draft Kenyan Constitution s273(3) states that:

In the performance of their functions, the national security organs and every member of the national security organs, shall not –
(a) act in a partisan manner;
(b) further any interest of a political party or cause; or
(c) prejudice a political interest or political cause that is legitimate under this Constitution.

In other ways however, the draft departs significantly from the approach taken in South Africa. Rather than integrating its police services, the draft provides for Kenya to retain its dual police service system, consisting of both the national Kenya Police Service and the Administration Police Service established as 'a separate service'. The Administration Police Service was established under the former colonial system and is primarily used, 'in bandit-prone areas in controlling incidents of banditry and cattle rustling', and which receives 'specialised training for combat situations and for the protection of VIPs'. (Commonwealth Human Rights Initiative, 2003, p. 40)

The Inspector-General of the Kenya Police Service and the Commandant-General of the Administration Police are both appointed by the President with the approval of the National Assembly, for a single term of five years to 'exercise command' over their respective police services and perform any other duties that Parliament may prescribe.

The Inspector-General's independence is emphasised by Section 282(4), which provides that while exercising his or her functions, shall, 'not be subject to the direction or control of any person or authority'. However, the Constitution contains no similar provision relating to the Administration Police. As envisaged by the draft Constitution, the key institution responsible for oversight of both police services is the Police Service Commission (PSC). According to the draft (s823) the functions of the Police Service Commission are:

(a) recruit and appoint persons to hold or act in an office, and to determine promotions in the services;
(b) keep under review all matters relating to the standards or qualifications required of members of the services;
(c) keep under review all matters relating to salaries, allowances and other terms and conditions of service;
(d) exercise disciplinary control, including hearing and disposal of appeals by persons in the services; and
(e) exercise such other functions as are provided for by this Constitution or an Act of Parliament.

The PSC is based on the Nigerian model and represents a fairly dramatic departure from the approach adopted in other Commonwealth countries. In effect, a Police Service Commission is an external structure intended to control recruitment, promotion, conditions of service and disciplinary functions that are normally regarded as internal management responsibilities within many other police services.

The Kenyan Police Service Commission is expected to consist of eight persons including, a person who is qualified to be a high court judge; two retired senior police officers, one from each of the services; 'three persons of integrity who have served the public with distinction'; the inspector-general of the Kenya Police Service; and the commandant-general of the Administration Police Service.

The Nigerian Police Service Commission also consists of eight members including: a Chairperson, a former police officer 'not below the rank of commissioner', a retired judge, representatives from a 'women's interest organisation', the Nigerian press, a non-governmental human rights organisation and the organised private sector, as well as the secretary to the commission. All members are appointed by the President subject to confirmation by the Senate, and are required to be 'persons of proven integrity and ability'. The Act provides that the commission 'shall not be subject to the direction, control or supervision of any other authority or person in the performance of its functions other than as is prescribed by the Act' (S6 (2)). However, the President has the authority to issue the commission with 'directives of a general nature or relating generally to matters of policy'.

As in the case of the Kenya Police Service Commission the core functions of the Nigerian system also concern appointment, promotion and discipline. While the Kenya Police Service Commission is responsible for these functions in relation to two separate, but relatively small police services, the Nigerian Police Service Commission is responsible for a single police service numbering over 200 000 members.

3.4 Case Study Conclusions

South Africa resembles other democratic countries such as England, Wales, Australia and Kenya, in that the constitutional and legal frameworks confirm the 'operational independence' of the police, but also put forward a framework for executive and broad governmental control of the police which is consistent with this.

The position in which South African provincial governments find themselves (as further discussed in the following section) compares interestingly with that of the police local authorities in England and Wales in that the authority of central Government is given primacy. The local police authority's power is subordinate to that of the national Government and has to be exercised within frameworks set by national Government. Nevertheless, both the police local authorities in England and Wales and the Provincial Governments do have some level of direct authority over the police. Importantly:

The role of the Provincial Executive regarding municipal police services is in some ways more restricted than those relating to the SAPS in that local authorities exercise direct control over them. However, in South Africa there are also significant powers in relation to the right to monitor whether police services conform to conditions which are set down by the Provincial Executive (as well as standards set by the national commissioner) when the municipal police service is established. In this sense, the Provincial Executive has some level of direct authority over the municipal police.

Australian state governments have direct authority and control over their police services. In some of the Australian states one of the concerns that has emerged is for there to be full transparency and accountability by the police minister for directives given to the police. Laws have been set down which have attempted to define the ambit of ministerial policy directives, the procedure in terms of which these are issued, and to ensure that full disclosure is made.

The Kenyan Constitutional provisions give major emphasis to the independence of the Inspector-General, a response to the political misuse of the police that has been a strong feature of political life in Kenya under previous governments. Kenya has followed the Nigerian example in that a Police Service Commission is to be established to oversee recruitment, promotions and discipline. While this framework is entirely different from that in other Commonwealth states, it is significant that this enables an external agency to have some control over aspects such as the calibre of recruits, the grounds for promotion, and the disciplinary system. This suggests that these are regarded as key concerns and legitimate areas for external intervention, despite having traditionally been regarded as internal police responsibilities.

4. The Legislative Framework for Police Oversight at Provincial Level

Effective police oversight is not merely about the legal framework that provides for it. The dynamics of policing and accountability are often determined more by political than legal factors. Nevertheless, in a constitutional democracy, the final arbitrator of disputes, over interpretation of the law, are the courts.

This section presents an analysis of the authority and liability of the provincial government in response to the questions asked by the legislature. It makes reference to a legal opinion and response to queries from Professor Hennie Strydom from the Faculty of Law, Department of Public Law, Rand Afrikaans University.

4.1 The Constitutional and Legal Framework

In democracies police accountability is exercised through internal control systems as well as through state and civilian agencies. As indicated previously, the Provincial Legislature and Executive are involved in holding police accountable, along with a wide range of other official and civil society bodies.

Schedule 4 of the Constitution lists the 'Police' as one of the functional areas over which Parliament and the Provincial Legislatures have concurrent legislative competence. However, Schedule 4 indicates that this is 'to the extent that the provisions of Chapter 11 of the Constitution confer upon the provincial legislature legislative competence'.

In relation to legislation Section 205 (2) in Chapter 11 says that:

National legislation must establish the powers and functions of the police service and must enable the police service to discharge its responsibilities effectively, taking into account the requirements of the provinces.

The first two clauses of Section 206 of the Constitution deal with governmental policy directives to the police, saying that:

(1) A member of the Cabinet must be responsible for policing and must determine national policing policy after consulting the provincial governments and taking into account the policing needs and priorities of the provinces as determined by the provincial executives.
(2) 'The national policing policy may make provision for different policies in respect of different provinces after taking into account the policing needs and priorities of these provinces.14

Related to this section, Section 206(8) states that:

A committee composed of the Cabinet member and the members of the Executive Councils responsible for policing must be established to ensure effective co-ordination of the police service and effective co-operation among the spheres of Government.

The Constitution therefore not only requires that the concerns of provinces be taken into account in formulating policy, but also attempts to institutionalise 'effective co-operation' between the Minister of Safety and Security and MECs through the above committee.

The authority of the Provincial Government over policing is therefore limited in the following manner. The national Minister has the authority to 'determine' national policing policy, though the Minister must consult the Provincial Governments and 'take into account' their needs in doing so. In addition the Section 206(8) committee is intended to institutionalise co-operation between the provincial and national executives on policing matters, though whether or not this occurs will depend on how the committee carries out its work in practice.

The Provincial Legislature does not have the power to pass legislation that is binding on the police. At the same time the Provincial Executive does not itself have the authority to directly issue policing policy.

This is not to say, however, that Provincial Governments have no authority over the SAPS. The Constitution provides a range of indirect avenues by means of which the MEC can impact on policing policy. In addition to the authority to input on the national policing policy in terms of Section 206(1) and (2), and the authority to participate in the Section 206(8) committee, Section 206(4) states that:

A provincial executive is responsible for policing functions -

a. vested in it by this Chapter;
b. assigned to it in terms of national legislation; and
c. allocated to it in the national policing policy.

Further functions vested in the provinces, and by implication the Provincial Executives, are the powers:

(a) In terms of Section 206(3) of the Constitution, in terms of which 'each province is entitled;
(i) to monitor police conduct;
(ii) to oversee the effectiveness and efficiency of the police service, including receiving reports on the police service;
(iii) to promote good relations between the police and the community;
(iv) to assess the effectiveness of visible policing; and
(v) to liaise with the Cabinet member responsible for policing with respect to crime and policing in the province.
(b) In terms of Section 206(5) of the Constitution, provinces 'may investigate, or appoint a Commission of Inquiry into, any complaints of police inefficiency or a breakdown in relations between the police and any community', though it 'must make recommendations' to the Minister of Safety and Security;
(c) The authority provided for in Section 206(6) of the Constitution – to lodge complaints with the Independent Complaints Directorate regarding 'any alleged misconduct, or offence committed by, a member of the police service in the province'.

The concerns of the Provincial Executive are also relevant to the appointment and dismissal of the provincial commissioner:

(d) Section 207(3) requires that the national commissioner appoint the provincial commissioner 'with the concurrence of the Provincial Executive' and provides for the Minister of Safety and Security to mediate if they cannot reach agreement.
(e) The Provincial Executive has the authority in terms of Section 207(6) of the Constitution to institute proceedings for the removal etc of the provincial commissioner. In terms of Section 8 of the SAPS Act, 68 of 1995, the Provincial Executive would do this by notifying the Minister of Safety and Security that the provincial commissioner has lost the confidence of the Provincial Executive.

As indicated, Section 206(4) also specifies that in addition to policing functions assigned to it in the Constitution, the Provincial Executive is also responsible for policing functions 'assigned to it in terms of national legislation' and 'allocated to it in the national policing policy'.

The most important of these would appear to be the powers of the MEC regarding municipal police services. These are highlighted below in section 4.5 of this report.

4.2 The Political Accountability of the MEC

While the MEC does not have direct authority over the SAPS, the MEC nevertheless has significant powers and needs to be able to demonstrate that her/his office has exercised its powers in a responsible way. In the words of the Constitution these include the powers:

The broad areas that are of key concern in holding police accountable are firstly, standards of service delivery, which may be understood as being questions of the 'effectiveness and efficiency of the police service'; and secondly, questions of police conduct. The Constitution may therefore be understood to indicate that the Provincial Executive has responsibilities relating to all aspects of service delivery and police conduct as well as specific additional issues. These issues include 'good relations between the police and community' and the 'effectiveness of visible policing'.

The Constitution appears to differentiate between how the Provincial Executive is supposed to deal with different types of problems. For 'complaints of police inefficiency or a breakdown in relations between the police and any community', the Constitution indicates that the Executive may itself investigate, or appoint its own Commission of Inquiry to investigate the reasons for this. However, when it comes to cases of 'alleged misconduct or offence', the Constitution indicates that these should be referred to the ICD. It appears that the Provincial Executive has powers to 'monitor police conduct', but that if it wants alleged cases of misconduct or offences to be investigated, it should refer these to the ICD.

Professor Strydom of Rand Afrikaans University responds saying that, whatever the action taken, the MEC, whether informed about police inefficiency or a breakdown of relations between the police and any community, or an allegation of police misconduct, will:

Not really have discretion to act or not to act. The principles and values underlying the Constitution and to which I have referred in my opinion prohibit this. Another reason for this is 'The Promotion of Administrative Justice Act 3 of 2000'. In terms of Section 1 of this Act the decision to act or not to act qualifies as an "administrative action" and is justiciable. Also note that a decision taken in terms of Section 206 is specifically not included under those decisions that are excluded from the definition of justiciable administrative actions. It is further safe to say that a matter regarding the inefficiency of the police or the breakdown of relations between the police and the community, is a matter which can affect the public in the way envisaged in Section 4 of the Act. Should this be the case, the relevant organ of state must decide on which action provided for in Section 4 to follow. (Response to query, 15 March 2004)

In deciding what course of action to take, the Provincial Executive may consider conducting an investigation, or appointing a Commission of Inquiry in terms of Section 206(5):

It would appear correct to say that Section 206(5) grants discretion to a provincial organ whether or not to investigate a matter mentioned in the said provision. The use of the word "may", after all, seems to be indicative thereof. However, I am of the opinion that this provision be understood as merely providing for one possible way of performing the functions in section 206(3). In other words the province may decide to consider other kinds of action depending on the circumstances and what would be most appropriate in those circumstances. (Prof. Strydom, response to query, 15 March 2004)

While the Provincial Executive does not exercise direct control over the SAPS:

The matters referred to fall under the jurisdiction of the Executive and … the Executive has a certain responsibility to apply its mind to those matters.
The "direct" responsibility of the MEC may be construed on the basis of Section 207(4)-(6). In exercising his or her functions in terms of these provisions, the provincial commissioner is directly under the [jurisdiction] of the Provincial Executive (and Legislature) and these two organs will be first in [taking action] if improper control has been exercised. (Prof. Strydom, response to query, 15 March 2004)

It would appear, therefore, that MECs can be held politically accountable in so far as they need to be able to demonstrate that they have taken action within their powers regarding questions of 'police conduct', 'the effectiveness and efficiency of the police', (including 'the effectiveness of visible policing'), and 'relations between the police and community'.

4.3 The Legal Accountability of the MEC

In relation to the legal accountability of 'policing operations which have gone wrong and have resulted in undesirable consequences', Professor Strydom is of the opinion that:

Unlawful conduct by police officers will therefore render the commissioner as well as the Provincial MEC for Safety and Security liable under the principle of vicarious liability. (Prof. Strydom , opinion, p. 8)

and that,

The Provincial MEC for Safety and Security can be a party to proceedings in a delictual action against the state and is therefore in this regard not in a position different from that of a Minister in the Cabinet.' (Prof Strydom, opinion, p. 9)

Nevertheless, it needs to be borne in mind that the MEC only really has authority over the police indirectly and through the mechanisms that make it possible for the MEC to influence the National Minister. While state liability is vicarious, if an action was instituted directly against the MEC, there would be a chance that the court would hold that the MEC has no direct power in relation to the actions of the police and that the MEC cannot be held directly liable on this basis.

However, if the MEC was cited as second or third respondent in an action against the national or provincial commissioner or the National Minister, the MEC could not escape being regarded as a respondent. (See also section 57 of the SAPS Act, which appears to envisage that actions might be instituted against the national or provincial commissioner or the Minister, but makes no reference to actions being instituted against the MEC).

4.4 Obligation of the MEC to Answer Questions

The Gauteng Legislature has asked whether the MEC can be expected to answer questions by the Legislature regarding the day-to-day operations of the police in the province.

As indicated above the functions of the MECs office appear to encompass monitoring, oversight, and the ability to institute investigations, or lodge complaints with the ICD, concerning all aspects of service delivery (the effectiveness and efficiency of the police service), relationships between the police and community, as well as police conduct. It would therefore appear that, and as stated in the legal opinion:

The inference is justified that the oversight role of the Provincial Legislature relates to all aspects of policing in the province and that the MEC can be expected to be accountable to the Legislature with regard to the day-to-day operations of the police in the province. (Prof Strydom, opinion, p. 12)

In addition to requesting the MEC to answer questions, the Legislature also has the power in terms of Section 206(9) of the Constitution to require the provincial commissioner to appear before it and answer questions. This links to the obligation of the provincial commissioner to 'report to the Provincial Legislature annually on policing in the province (provided for in Section 207(5)).

4.5 Accountability of the MEC regarding Municipal Police Services

The powers of the MEC with regard to municipal police services (MPS) include the power:

Beyond the above powers:

The MEC has no other say over the functioning of a municipal police service and is therefore not accountable for the way in which the service otherwise functions. Here it must be emphasised that the CEO of a municipality and the executive head of the municipal police service assume responsibility for the functioning and day-to-day operation of a municipal police service and is accountable to the municipal council for the conduct of the service (Section 64B – C of the Act). (Prof Strydom, opinion, 15)
Section 64I(1) of the SAPS Act is also explicit that legal proceedings against a municipal police service 'shall be instituted against the municipal council in question.
4.6 Oversight of the Metropolitan Policing Services

Related to this, therefore:

The extent of the Legislature's monitoring and oversight functions is determined by the extent of the Executive's involvement in municipal police services as determined by section 64 of the Police Service Act of 1995. (Prof Strydom, opinion, 13 – as above)

5. Role of the Legislature and Police Accountability

This section briefly presents the processes and kinds of information through which the Legislature Standing Committee on Public Safety and Community Liaison oversees the executive department and the SAPS. The aim of this section is to present an analysis of the way in which the Gauteng Legislature's Standing Committee on Public Safety undertakes its oversight role. Typically, the legislature exercises oversight of the executive in the following manner:

Each of the above oversight activities will be considered in further detail in this section. Select examples have been chosen against which to illustrate some of the challenges being faced in this regards. The above activities will be analysed through the oversight framework presented earlier in the report. To refresh, it was emphasised that oversight of the police should primarily focus on:

  1. Police performance (i.e. service delivery);
  2. Police conduct (i.e. misconduct and abuse of police powers).

Furthermore, the following key principles should guide endeavours to enhance police accountability:

While, unlike the local police authorities in England and Wales, the Provincial Executive does not have the authority to directly define police objectives and plans, the MEC nevertheless has certain powers to monitor the police. The Legislature should scrutinise the Provincial Executive use of these powers, partly in relation to their overall performance, but also specifically in relation to the risk of inappropriate interference with police, something which has been entrenched by legislation in some Australian states.

5.1 The MEC and the Provincial Secretariat for Safety and Liaison

One of the primary ways that the Legislature contributes to the oversight of the police is through holding the Provincial Executive and Department of Safety and Liaison accountable for fulfilling its mandate. The Legislature passes the annual budget of the department on the recommendation of the standing committee. The committee also receives regular reports on the objectives and activities of the department. Through holding the Provincial Secretariat accountable, the Legislature contributes to the quality of the MEC's work, thereby impacting indirectly on the extent and effectiveness of police accountability.

While the Provincial Executives monitoring role appears to constitute the core of the MEC's (or Secretariat's) powers regarding both the SAPS and municipal police departments, the MEC also other powers. Of these perhaps the most important is the power to input on national policing policy. The MEC is therefore intended to serve as the voice of the province in ensuring that the priorities of the province are reflected in national policy. Other significant powers include the power:

The legal opinion is that the MEC needs to be able to demonstrate that he/she through the Secretariat has exercised these powers in a responsible way. However, evidence suggests that the Secretariat is not primarily focused on activities that enhance its ability to exercise these powers.

A 1998 report looked at the policy framework in order to provide the Gauteng Department of Safety and Liaison with strategic direction (Rauch, 1998). Based on the then activities of the department as well as a review of the Constitution; the South African Police Service Act of 1995; the 1996 National Crime Prevention Strategy of 1996; the White Paper on Safety and Security of 1998; and the SAPS Amendment Act of 1998 concerning Municipal Policing, the report recommended that the Provincial Secretariat focus on the following five key strategic functions:

  1. monitoring the police service;
  2. improving police-community relations;
  3. coordinating crime prevention in the province;
  4. contributing to national police policy;
  5. overseeing municipal policing in the province.

Subsequently, the department presented the following five objectives to the committee as part of the budget approval process for 2003-2004:

As is illustrated, the key objectives of the Gauteng Department of Safety and Liaison have not remained static and there is evidence of slight change reflecting the developments in national policy. Most significantly, initiating and coordinating 'social crime prevention activities' and coordinating the efforts of other departments to reduce crime, have emerged as a new focus area of the department.

There is a shift, therefore, from the initial constitutional provisions, all of which spoke directly to oversight of the police, to the current situation whereby only two objectives focus directly on overseeing the police. While this can be seen as a reflection of the shifts in national policy relating to a greater focus on 'crime prevention' through coordinating a 'multi-agency approach', it appears related to a downgrading of the role of the provincial department with regards to police oversight. The constitutional provision for 'monitoring police conduct' has been broadened to the current objective of 'monitoring, evaluating and assisting' the police to ensure quality services are delivered.

In the next section, a closer examination of the budget allocations and activities of the department will be undertaken to determine the nature and extent of the department's oversight role of police agencies in Gauteng.

Oversight of the Departmental Budget

According to the departmental budget request tabled before the Legislature committee on Public Safety at the beginning of 2003, the total allocation to the Department of Public Safety and Community Liaison is R33, 9 million. This is an overall increase of R7, 9 million from the R26 million allocated for the previous year, reflecting a 30 percent increase in the department's budget.

Annual change in programme share
Programme 2002/03 2003/04 % Change
Management and administration 34% 27% - 7%
Crime prevention 17% 34% + 17%
Monitoring and evaluation 22% 19% - 3%
Communication 27% 20% - 7%

As can be seen from the budget allocations, monitoring and evaluation now receives the smallest share, with this declining relative to the overall budget over the financial years 2002/3-2003/4.

According to the 'Budget Statement Book' of the Gauteng Legislature, a number of sub-programmes were identified. Along with each sub-programme, an objective was stated in the form of an outcome'. Measurable indicators were added and a budget estimate was provided. An overview of the contents of the 'budget book' and objectives reveals that with respect to the police, the department is primarily concerned with the effectiveness of police case and docket management, as well as with the provision of 'victim empowerment services'.

The table below is an example from the 'Budget Statement Book' on the 'Monitoring and Evaluation' programme of the department.

Sub-programme Service delivery: prevention of violence and abuse of women and children.
Stated actions and measurable indicators
  • Monitoring the efficiency and evaluating the effectiveness of the Family Violence and Sexual Abuse Units
  • % increase in reporting of crimes against women and children in targeted areas.
  • % reduction in incidents of violence against women and children in targeted areas.
Budget Cross cutting within directorates

As is reflected in the above table this sub-programme focuses directly on the SAPS Family Violence and Sexual Abuse Units. While measurable indicators are presented, an estimated budget is not provided in this example. Important detail necessary for the committee to effectively monitoring budget allocation and expenditure is generally not presented. Such information would include resources allocated to such a programme (i.e. budget, staff, vehicles etc), clear activity objectives and baseline figures against which to assess performance indicators presented.

The consequence of this is that from the information provided during the budgetary process, the legislative oversight committee would be unable to tell the following:

The indicators in the above example are also potentially contradictory in that apart from victimisation surveys (known for poor validity in capturing sensitive crimes such as domestic violence or child abuse), the only way of measuring numbers of incidents is through the cases reported to the police. As no budget is allocated for a expensive victimisation survey as part of this sub-programme, that indicator becomes meaningless. Consequently, only one indicator is being relied on so that for this programme an increase in cases reported to the police could be billed as a success (i.e. there is an increase in reporting of crimes) or failure (i.e. there is an increase in incidents of crimes). There are similar challenges in relation to many of the performance indicators presented by the department.

Oversight of Departmental Activities

In September 2003, a presentation of the Department of Safety and Liaison's Annual report for 2002-2003 before the Gauteng Legislature's Standing Committee for Public Safety and Liaison provided insight into the department's oversight role of the police. After highlighting the 'departmental objectives', the following four 'departmental outputs' were presented:

From these four 'outputs', the second one directly emphasises oversight of policing agencies in the province (through a monitoring and evaluating programme). The third output focuses on evaluating backlogs in 'crime intelligence' which is one specific aspect of policing. However, these 'outputs' are defined very broadly. There is no indication as to what exactly about policing will be monitored and evaluated; against what; or in what manner. The documented presentation provides no clear link between these outputs and other departmental activities mentioned in the presentation that can be empirically measured or assessed in terms of efficient budget expenditure by the standing committee.

All the activities relating to police oversight by the department were presented under the section headed 'Key Achievements of 2002-2003: monitoring and evaluation.' According to the presentation, the following 'service delivery improvements' were achieved (as presented to the committee):

Among the 'key achievements' were the following:

Under the heading 'Complaints management' it was stated that 863 complaints were received of which, '80 percent' were 'successfully dealt with'. The number of cases opened per area was presented and divided into the following broad categories of 'poor service delivery', 'lack of feedback on investigations', 'corruption' and 'referrals to other departments'. While 'referrals to other departments' might have included referrals to the Independent Complaints Directorate (ICD), no specific reference was made to such cases.

The other part of the presentation relating to police oversight presented the committee with statistical data in graph form as to the personnel, vehicles and docket load of the Family, Child and Sexual (FCS) offences units per SAPS Area.

Analysis

From the documentation provided to the Legislature it certainly appears as if there is a significant amount of work being undertaken by the Provincial Department of Safety and Liaison.

In terms of police oversight, against a backdrop of a 30 percent increase in the department's budget, there was a small decrease in the relative allocation for the 'monitoring and evaluation' programme within which most police oversight activities are located – though the allocation to oversight did increase in real terms.

It is evident that the department does to some extent focus on both 'police performance' and 'police conduct'. However, what exactly is being achieved through these activities is not clearly reflected in the documentation presented to the parliamentary standing committee. Clear and measurable performance indicators were not presented making it impossible for the committee to track changes over time with respect to the police. Moreover, there is no way that the committee can make an assessment of how effectively or efficiently the departmental budget is used, as none of the activities presented by the department were directly linked to programme spending (e.g. how much did the monitoring of 35 police stations cost, what specific activities were undertaken and what were the measurable achievements as a result). Similarly, no information is provided as to the corresponding improvements in the performance and service delivery of the Family, Child and Sexual (FCS) offences units in relation to improved resource allocations.

While the department has established its own complaints recording system to receive information on the performance and conduct of police officials, there is no analysis of how this information relates to other data gathered by either the internal systems of the SAPS or the Independent Complaints Directorate. For instance, how many of the complaints received by the departments were referred to the police for further criminal or disciplinary action and what were the outcomes. What cases were referred to the ICD, and what was the outcome of these cases? Furthermore, apart from broad categorisation, no details were provided as to why complaints were made and if any systemic interventions were needed to address the complaints. While it is understood that the documentation presented was an overview of the annual activities of the department for 2002- 2003, the committee should be able to ask for detailed briefings and written reports in relation to departmental monitoring activities.

In terms of operational independence, some of the activities presented in the report suggest that there may not be clear demarcation between the policy work of the executive and the direct management of the SAPS. This is reflected in claims by the department to have undertaken responsibility for 'implementing a plan' that transformed a police station and led to new managers being deployed. While, operational independence is a complex issue, it would be beneficial for the parliamentary committee to be informed as to exactly what instructions were being issued to which police commanders, as well as how the results of these instructions were being monitored or assessed.

Moreover, the documentation presented to the committee did not reveal that any monitoring of municipal police services had been undertaken. Nor was there any indication that the MEC had made any submissions to the Minister of Safety and Security regarding provincial priorities in relation to national policing policy.

5.2 Legislature Questions to the MEC

In order to further ascertain the nature of police oversight by the Legislature, this section briefly analyses the oral and written questions presented by members of the Gauteng Legislature to the MEC for Safety and Liaison.

Oral Questions

During 2003 oral questions were addressed to the MEC 12 times. African National Congress (ANC) members asked six (50%) of the questions and Democratic Alliance (DA) members asked the remaining six questions. Of the questions asked eight were directly related to crime and policing, while the remaining four were about other issues. Of the eight questions asked about policing, two were related to police conduct (corruption and racism), five were about policing resources or strategies to combat crime, and one was about the Police Orphans, Widows and Widowers fund.

The questions were all broad in nature as were the answers. For instance, questions about steps to address performance problems in the police were responded to very generally and failed to provide any meaningful details (e.g. the following sentence is an example of an answer to a question about performance problems at a particular station - 'the close working relationship between management with full time shop stewards'). There was no detail given as to how these interventions were to be monitored and assessed. There were also no 'follow-up' questions by members of the Legislature to determine whether interventions or solutions referred to in previous questions had led to the desired results. All but one of the questions were about the SAPS (apart from one question dealing with the disbandment of the commando system), and none focused on any of the municipal police services operating in the province.

Written Questions

A total of 160 written questions were submitted by members of the Legislature to the MEC between 29 January and 14 November 2003. All these questions were submitted by opposition parties with a vast majority coming from the DA, (141) and the remainder (19) from the Freedom Alliance. Of these questions 106 were related to the crime trends and human and vehicle resources at specific police precincts or areas. However, due to the National Ministers decision to only release crime statistics once a year, precise crime figures and performance measures (e.g. estimated values of items or number of cases leading to arrests and successful prosecutions per year), were not forthcoming.

The remaining questions were about a wide range of policies and specific incidents, developments or issues. There were only six questions related to police conduct in general and these were about allegations regarding an area head of detectives in Pretoria; the number of domestic violence cases against members of the SAPS reported to the ICD; general complaints received by the ICD; reports of three drunk police members at a crime scene; and an alleged irregular shooting incident involving two police members.

A document consisting of a matrix was presented of complaints and disciplinary action taken at one police station (Linden), which provided some insight into the systems in place to respond to allegations of misconduct at that station. This document claimed that in each case, whether or not disciplinary action was taken against the member, the complainant was satisfied with the outcome of the action taken. This needs to be treated with caution as a large amount of research suggests that it would be highly unusual for all complainants to be satisfied with the outcome of their complaints against the police. It would be prudent for the committee to further explore the extent to which the information provided in this document was valid and reliable.

As was the case with the oral questions, the answers were rarely detailed. However, there were a few questions that received vast amounts of detailed information (examples include a set of questions about vehicle distribution and a set of questions about unclaimed bodies at state mortuaries.) All questions relating to the police were about the SAPS and none about municipal police services.

Analysis

The vast majority of the questions put to the MEC were related to the SAPS's challenges and performance, or alternatively incidents involving the SAPS. Questions which were answered were mostly forwarded by the Secretariat to the relevant police commander who responded back to the Secretariat, who in turn then forwarded the answers to the Legislature. It is interesting to note that all the questions asked were from two opposition parties in the Legislature. Given both the nature of the questions asked and the answers given, it can be argued that this type of parliamentary process is not very effective for exercising oversight of the police.

While detailed information about human and logistical resources was presented with respect to some questions, generally the information provided would be of little use to make empirical assessments about improvements about police performance and service delivery. With regards to police conduct, very few questions were asked. However, none of the questions would assist the legislature in identifying trends or patterns of police misconduct and abuses of police power. For example, there were no questions as to the number, nature or extent of citizen complaints against the police and whether there had been any changes over the past years. Consequently, the legislature is not in a position to establish whether the Secretariat is focusing on the relevant issues, whether specific police managerial interventions are needed, or if police systems for internal control are working effectively or efficiently. For example, there were no questions regarding the trends relating to the outcomes of internal police disciplinary or criminal investigations.

5.3 Reports from the SAPS Provincial Commissioner

On 2 September 2003, the Standing Committee on Public Safety received a report on policing in the Gauteng Province by the SAPS Provincial management team headed by the Provincial Commissioner, Perumal Naidoo. The report included the following aspects relating to the SAPS in the province:

Analysis

Out of the three sources of information relating to the SAPS in the province, the report by the SAPS Provincial Commissioners office was the most insightful and useful. Information was presented in relation to both performance and conduct, with stations being identified as falling into one of three ranking categories (roughly defined as, 'above', 'in line with' or 'below acceptable' standards). Detailed information on some of the performance criteria was provided in the presentation, which implied that further information could be made available to the committee if required. The criteria could be further scrutinised by the committee in order to evaluate their reliability and validity as an assessment framework for police performance at a station level.

Furthermore, the committee was made aware of internal performance management systems that were to be implemented by the SAPS. This is very useful for the committee with regards to focusing its oversight role(i.e. monitoring the implementation and effectiveness of these systems.) With detailed data and information that could be received from the SAPS, the committee would be in a position to independently verify and analyse the data for its own needs. In so doing it could give considered and constructive feedback to the SAPS Provincial commissioner that could assist in improving their internal management systems. An attitude and working relationship of mutual respect and cooperation between police management and independent oversight structures is considered to be one of the preferable situations with regards to civilian oversight of the police (Walker, 2001, p. 151).

6. Recommendations for Strengthening Police Oversight

This final section presents a number of key recommendations that aim to assist the Gauteng Provincial Government in optimising and regularising its police oversight role.

6.1 Clarifying the Purpose and Nature of Legislative Oversight of the Police

As proposed earlier in the report, police in democratic societies should be held accountable both in relation to service delivery and conduct. This accountability has a bearing on police performance, as well as on occupational hazards such as brutality and abuse of powers. The nature of police accountability should be clearly understood by the Provincial Legislature in order to formulate its purpose and strategic objectives in terms of police oversight. Once the underpinnings are in place they can guide the Legislature towards identifying exactly which aspects of police performance and conduct requires monitoring and input.

It is unlikely that the standing committee will have the time, capacity, or resources to exercise thorough oversight of the entire provincial structure of the SAPS in Gauteng.15 It is therefore not useful for the committee to focus on specific incidents or stations, unless a particular crisis emerges. Rather, the committee should focus on key strategic issues that it wants to monitor, and use its powers to obtain regular and detailed information on those issues.

When it comes to police oversight with regards to monitoring, the Secretariat is the primary executive structure to do so. Therefore, the legislature should ensure that this department is undertaking its monitoring activities in line with clearly identified strategic objectives. Moreover, as part of the budgetary process, the committee should focus on ensuring that the Secretariat is providing value for money. This can only be ascertained with the assistance of empirical performance indicators that demonstrate both the precise activities that have been undertaken by the department and the resultant impact. In this way the committee will ensure that the executive body tasked with an oversight role is performing effectively and efficiently. Moreover, the focus of the committee should be on the extent to which the MEC exercises his or her powers. Towards this end the committee should call for detailed annual reports that detail:

Where the committee feels that it would like to address specific problems with any of the policing agencies, it can ask for detailed briefing from the relevant police commanders in the province through the provincial commissioner. If it is not satisfied with the response it can take the matter up directly with the Provincial Commissioner or the MEC.

6.2 Working with Internal Police Structures

In order to oversee police performance in relation to service delivery, there are a wide number of issues that can be focused upon. Police agencies are structured in a number of tiers and geographic regions. At provincial level, SAPS headquarters are divided into components involved in the management of specific operational services (e.g. crime prevention, detective services, crime intelligence, organised crime task teams) and support services (e.g. evaluation services). There is a similar structure at each of the seven Area Offices that oversee all police stations in the province.

At station level the SAPS is usually structured into the following components:

Each component develops its 'operational' plan within the broad guidelines and is responsible for developing its own performance indicators. Once the committee is familiar with these structures and functions, it will be easier to identify what types of policing performance it wants to monitor and oversee.

Similarly, with police conduct there are particular internal systems and structures that are in place to deal with police misconduct and criminality. The Legislature committee needs to be familiar with these internal structures if it is to start exercising oversight in a strategic and focused manner. For example, if the committee is concerned about the effectiveness of the SAPS in tackling and preventing police corruption, it can ask for the policies, objectives and performance indicators used by the police to evaluate this. Then it can start to request information about the various structures and initiatives which each of the provincial, area and station levels have in place to deal with the problem. This enables the committee, through comparing performance indicators, to establish how successful the various structures are.

Reflecting on what was presented by the provincial commissioner, it would make sense for the committee to begin by receiving regular (quarterly or bi-annual) reports on the provincial performance measurement system. This would provide concise, but useful, information on a number of priority stations in the province. Detailed attention could be given to problematic stations regarding appropriate interventions and their outcomes.

6.3 Working with Other Structures and Agencies

As mentioned in the first section of this report, there are many state and non-state structures that are engaged with policing activities and play a formal or informal role in police oversight. It would be useful for the committee to develop a database of such structures per SAPS Area and invite them to make presentations to the committee as to their concerns and experiences. This would give the representatives on the committee better insight into the range of public perspectives on issues relating to the police.

It would also be useful to compare information from a variety of sources. If the committee was concerned about the nature of complaints against the police in certain areas or stations, presentations could be invited from structures such as the ICD, Human Rights Commission, relevant CPFs and other local structures. Such information would inform the committee as to the nature of the problems being experienced and appropriate responses required.

After an agreed amount of time, these same structures could be asked to make further representations to the committee as part of an evaluation process to assess whether or not the interventions from police management had the desired impact where it matters most – at police station level. The coordination of a range of structures could result in more effective police oversight.

6.4 Using Key Performance Indicators

The Legislature's Standing Committee should start gathering information in the form of provincial or area level performance indicators. These indicators could relate either to the service delivery performance or conduct of the police. As police agencies generally collect huge amounts of data on a range of indicators, the committee could narrow its focus to what it perceives to be the most important for its oversight purposes.

Once key indicators are identified and agreed upon, the relevant information could be collected on a quarterly basis and Legislature researchers could analyse the data for trends and patterns. Such analysis would start to provide the Legislature with insights into trends and patterns of policing in the province. Specific questions could then be asked of the MEC or the SAPS Provincial management as to how they are tackling identified problems. Changes in the indicators would start to provide objective measures of the success, or otherwise, of policy developments and management interventions.

Examples of performance indicators that could be used by the committee are as follows:

Examples of conduct issues that the committee could obtain information and indicators on are:

There may also be other indicators that are not strictly performance or conduct indicators, but highlight key issues of concern which the committee might want to track. These could include for example, indicators regarding killing of police members while on and off duty.

Some of the above indicators would be easy to track as the SAPS already collects relevant data for management purposes. For instance, the SAPS Provincial Office has already developed performance charts for all of its stations in the province based on 32 performance indicators. Similarly, the ICD collects data on police action/ custody deaths and misconduct, which would be relevant information for the committee. The committee could start by requesting available data in order to assess the work of the Secretariat in relation to these stations. Other indicators may prove more difficult to gather information against and it may require further research. Essentially, it would be necessary for the committee to collect empirical data so as to be able to objectively track changes with regards to police performance and conduct over time.

6.5 Detailed Oversight of the MEC and Provincial Department

The Provincial Executive has a range of powers and responsibilities. The core of these powers concerns monitoring and oversight of the police. However, these respective powers extend to the authority to input on frameworks developed at a national level; to institute certain types of investigations; as well as specific powers relating to municipal police services. The ICD is also mandated to investigate cases lodged with it by the Provincial Executive. The MEC and the department should be asked to explain how they understand these powers; how they are used in practice; and suggest optimal ways of using them.

The committee should also focus on the effectiveness and efficiency of departmental spending. Currently however, the department does not give the committee details that relate expenditure to particular activities or programme outputs. Without such information it is impossible for the committee to reach a judgement as to how effectively and efficiently the budget for monitoring and evaluating the policing agencies, is being used. Therefore, there can be no clear standard against which to assess the need for increased budgets or whether the current budget is being adequately used.

The committee should seek to assist the department in improving the effectiveness of its work. The committee is the only structure directly overseeing the department and should be in a position to provide constructive feedback. This would also be a way of improving the oversight of the police in Gauteng. Done in collaboration with the MEC and the department it is more likely to be successful, and could become a model of legislative oversight throughout the county.

6.6 The Effective Use of Research

While indicators potentially have a lot of value there will still be unanswered questions about police performance or conduct, or the effectiveness of specific interventions. Good quality research and analysis can substantially enhance the ability of oversight structures to carry out their functions. It can also ensure that the feedback and nature of the oversight role is one that is based on sound analysis and can thereby also contribute to oversight being practiced in a constructive manner. The British Home Office, for instance, regularly commissions research, with research reports being published on the Home Office website.

Oversight practices that are not based on hard information and insightful analysis can lead to resistance and resentment on the side of the police as they may feel that the relevant structures do not fully appreciate the difficulty of their work and the challenges that they face. Research that is able to contextualise the key challenges and provide practical and useful recommendations could be of benefit, not only to the Legislature but also to the management of the police agency. This may particularly be the case where the police do not have the capacity or budgets themselves to undertake research that may be beneficial to them.

While one option is for the Legislature to commission independent research on any topic relating to police performance (e.g. evaluate the success of operation crackdown in Hillbrow), or conduct (e.g. whether there are user friendly systems in place for complaints against the police and how these complaints are dealt with), the Legislature could also contract or undertake independent analysis on data already held by the police.

Notes:

1 It is important to note, however, that the Preamble to the South African Police Service Act, 1995 states that the SAPS 'shall be structured at both national and provincial levels and shall function under the direction of the national government as well as the various provincial governments'.

2 These include traffic police, municipal security guards, and the private security industry (members of private security firms do not have full policing powers). Government has also established the Directorate Special Operations (the 'Scorpions') as one of a number of special investigating units that falls under the Director of Public Prosecutions. Particularly in rural, but also in urban areas, both Permanent Force as well as Commando units of the South African National Defence Force (SANDF) are also involved in policing activities. Members of the public may also be involved in state 'policing' systems as SAPS Police Reservists or through the above- mentioned SANDF Commando system.

3 While there are thousands of examples, two high profile cases stand out. Firstly, the 1991 Rodney King beating by police leading to massive rioting and destruction of property in Los Angeles, USA; and secondly the failure of the police to adequately investigate the 1993 racially motivated murder of Stephen Lawrence in the United Kingdom, resulting in a judicial inquiry into the racial attitudes and culture of the London Metropolitan Police Service.

4 Experts such as David H. Bayley, author of (1994) Police for the Future. Oxford University press: New York, states simply that police need to be held accountable for 'what they do and how they do it.'

5 The Government's White Paper (No. 18340) on Batho Pele was published by the Department of Public Service and Administration on 18 September 1997 and spells out eight principles (such as consultation, having clear service standards, ensuring access to service, courtesy, accurate information, transparency and redress for poor service delivery) which when applied will lead to a high level of customer service.

6 While Stone focuses primarily on three key dilemmas this paper adds a forth – abo