Centre for the Study of Violence and Reconciliation

Staggering under the burden

ICD policies on the receipt of complaints and on investigations and their negative impact on the ICD, on public confidence, and on police discipline
by
David Bruce

Edited version of a discussion paper initially presented at the Independent Complaints Directorate workshop on "Establishing a cordial relationship between the police, ICD and civil society: problems and prospects", Burgers Park Hotel, Pretoria, 6-7 February 2006.

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

Introduction

This paper is not intended to discuss the legal mandate of the Independent Complaints Directorate (ICD) but to rather discuss how it has interpreted this mandate. The ICD has increasingly emphasised its own role as a body for receiving complaints and in recent years the ICD has adopted an approach in terms of which it aims to investigate all cases which it receives. ICD policy seems to give no priority to how to strengthen the South African Police Service (SAPS) and municipal police systems for receiving and investigating complaints. The result is that ICD policy has, more and more, shifted the burden of dealing with such cases onto the ICD itself, with no consideration being given to the impact of these policy choices in relation to the effectiveness and efficiency of investigations, and ultimately in relation to improving police conduct.

As will be explained in this paper, the approach is not legally necessary, it is not justified in terms of practical or theoretical considerations (even if it was justified at an earlier point in the ICD's history), and it is self-defeating, contributing to inadequate investigations of complaints against the police. As a result the ICD has a limited effect on police discipline, and little ability to win public confidence or generate public satisfaction as it is ineffective in dealing with the bulk of cases which it receives.

Background to the ICD

The ICD started operating in April 1997. In the eight years up to the end of March 2005 it had, according to ICD annual reports, received 30 083 cases falling within its mandate.1 These have included 5340 cases involving deaths in police custody or as a result of police action, which, as will be discussed below, the ICD itself is obliged to investigate.

The ICD has therefore been operating for just under nine years. During most of this time the ICD has employed a staff complement in the region of 120-130 people, though the most recent annual report indicates that this has now increased to over 160.

The leaders and many of the staff of the ICD have often been highly dedicated to their work. If one was able to go through the case files of the ICD on a case by case basis one would find many instances where cases have been thoroughly investigated, or intensively monitored.

Most complaints against police in other countries do not result in a complaint being sustained against a police officer and the fact that this is also the case in South Africa does not in itself reflect shortcomings on the part of the ICD (see e.g. Walker, 2001:120). Similarly there is no reason to assume that the bulk of deaths in custody and as a result of police action are linked to criminal actions by the police and thus the fact that most of these investigations do not result in prosecutions or convictions of police members is not itself a reflection of the failure of the ICD.

However it was apparent last year that there was some level of dissatisfaction, particularly at provincial level, about the management of the ICD.2 While there were a wide range of issues around which this dissatisfaction was articulated it would seem that a central underlying factor which has contributed to this dissatisfaction has been that the manner in which the ICD has defined its mandate places an unmanageable case burden on the ICD, and particularly on the ICD provincial offices.

The fact that the ICD is not managing its case load is also reflected in the latest annual report which states that 'the backlog in respect of cases under investigation far exceeded the number of new cases received during 2004/2005 (p. 6). This view has been reinforced by recent research which suggests that very few cases are finalised promptly by the ICD.3

The core constitutional and legislative mandate

The legislation which provides for the establishment of the ICD, Chapter 10 of the South African Police Service Act, 68 of 1995, provides at section 53(1) that

'The principal function of the directorate shall be the achievement of the object contemplated in section 222 of the Constitution'.

Section 222 of the interim Constitution is the original provision in terms of which the ICD was founded, and despite the fact that the interim Constitution is no longer in operation, still of force as a result of section 53(1). We may recall that Section 222 provided that the object of the envisaged independent complaints mechanism would be that of

'ensuring that complaints in respect of offences and misconduct allegedly committed by members of the Service4 are investigated in an effective and efficient manner'.

It must be emphasized that these key founding provisions which define the objects and purpose of the ICD do not require that the ICD carry out any investigations itself nor even that that the ICD itself should necessarily be the principal vehicle for receiving complaints.

What these provisions do impose, and this should be seen as the central defining point of departure for the ICD, is an obligation for the ICD, to ensure that complaints in respect of offences and misconduct allegedly committed by members of the SAPS and MPS are investigated in an effective and efficient manner.

However what is principally important in terms of this paper is that these core provisions are essentially flexible regarding questions of which agency is responsible for (i) receiving complaints; and (ii) investigating them.

In theory this means that the core objective of the ICD should be to ensure that the optimal system is in place to ensure that complaints are investigated in an effective and efficient manner. It is not obliged to receive or investigate these cases itself.

Other legal obligations of the ICD

Section 53(2) of the SAPS Act states that 'In order to achieve its object, the directorate -

(a) May mero motu or upon receipt of a complaint, investigate any misconduct or offence allegedly committed by a member, and may, where appropriate refer such investigations to the Commissioner concerned;
(b) Shall mero motu or upon receipt of a complaint, investigate any death in police custody or as a result of police action; and
(c) May investigate any matter referred to the directorate by the Minister or the member of the executive council.

As is known by those who are familiar with the ICD, and presumably would be apparent to anyone trained in the interpretation of legislation, the presence of the word 'shall' in Section 53(2)(b) imposes on the ICD an obligation to investigate all deaths in police custody or as a result of police action.5

As indicated above the total number of these cases came to 5340 imposing a significant investigative burden on the ICD. The location of ICD offices means that ICD investigators often have to travel large distances to get to the scene of these cases, and the investigative work involved in investigating these deaths is often substantial.

Section 53(2)(b) is however the only legislative provision which imposes an explicit obligation on the ICD to investigate cases.6 Apart from this, as indicated, all of the legal provisions which give direction to the ICD in relation to the receipt and investigation of cases essentially allow the ICD a high degree of flexibility in relation to how the actual reception and investigation should be carried out, and who this should be done by.

ICD policy on the receipt of complaints and the investigation thereof

Receipt of complaints

The ICD policy on the lodging of complaints is implied rather than being spelled out in a document. For instance under 'Lodging Complaint against the SAPS' the ICD website provides details of how complaints may be emailed or faxed to the ICD offices. The webpage also provides, for instance that

'Should you have a serious complaint or one that calls for a high level of confidentiality, it is recommended that you approach the ICD office nearest to you to arrange an appointment'.

There are in the region of 1126 police stations around the country, as opposed to 10 ICD offices. Ignoring municipal police services the SAPS alone have in the region of 148 000 members of staff as opposed to the 166 employed by the ICD. SAPS standing order 101 makes provision for the SAPS to receive and investigate complaints from the public against SAPS members.

Despite these facts it would appear reasonable to say that, whatever the nature of their complaint against the police, ICD policy is to encourage complainants to lodge complaints at ICD offices. In addition the ICD has no policy of informing complainants that they can, or even that they should in certain circumstances, lodge complaints directly with the SAPS or municipal police department.7

The investigation of complaints

Policy on the investigation of complaints is sometimes more explicitly defined.

The 2002-2003 annual report for instance indicates (p. 31) that the ICD Investigation Programme has the following objectives:

The annual reports after this one indicate that the ICD has become more and more ambitious. No longer, the policies indicate, will the ICD monitor a certain proportion of the cases which it receives, but in future it will itself investigate all such cases.

For instance the ICD 2003-2004 annual report states that

'The target to investigate 50% complaints of misconduct was achieved and in some offices … even exceeded. The remainder of these complaints are subject to monitoring. Our target is that by 2005/2006, all complaints which are reported to the ICD, will be investigated by the ICD' (p. 35).

The 2004-2005 report indicates under a list of 'Main services provided and standards' that its 'Main services' include

'The investigation of all deaths in police custody and as a result of police action and, misconduct or offences committed by SAPS and MPS members' (p. 19).

While this and other information provided in both of the reports in relation to these targets and standards raises many questions what is of key concern here is simply this: During recent years the ICD has aimed to expand its investigative role so that it investigates more and more of the cases that it receives.

Underpinning ICD investigative policy over recent years seems to be an understanding that the ICD has some type of obligation to investigate as many of the cases which it receives as possible. The key consequence of this must inevitably be that the ICD spreads its investigative resources more and more thinly and that, rather than becoming more effective and efficient, the quality of investigations gets worse and worse.9

This focus seemingly disregards the primary object of the ICD as referred to above which is to ensure that complaints in respect of offences and misconduct allegedly committed by police are investigated in an effective and efficient manner.

The motivation for these policies

The view that 'police cannot investigate themselves'

The motivation for the establishment of bodies like the ICD has been experience worldwide that police are often resistant to receiving complaints against themselves, and that the investigation of such complaints often leaves a lot to be desired.

The above evidence is often taken to be proof that 'the police cannot investigate themselves'. While it is true that police services often have a poor record with regard to the receipt and investigation of complaints against the police it is basically not correct to say that the police cannot investigate themselves. The conclusion that it is necessary for another agency (such as the ICD in South Africa) to receive all such complaints and to investigate all such complaints basically therefore derives from an uninformed point of view.

However international experience indicates that it is possible for police to establish effective systems for the public to lodge complaints against them, and for internal investigations. These systems are potentially more effective than civilian oversight bodies. Factors which encourage this are:

In fact in the international world of civilian oversight it is virtually unheard of to have an oversight body which carries responsibility for investigating all complaints against the police. This is even in countries which are able to provide far better resources to their oversight bodies than South Africa.

Societies in transition to democracy

South Africa is a society which has recently emerged from an authoritarian form of government. Typically at the early stages of transition from authoritarian government the existing mechanisms of internal discipline in the police, and mechanisms for receiving complaints are very weak. In addition, as a result of the role played by police in enforcing repressive polices, there is a high level of mistrust and fear of the police.

At the same time that the police internal mechanisms are likely to be weak in the period immediately after the establishment of democracy, newly established complaints bodies are also likely to be quite weak, and suffer from problems such as the absence of proper systems, and shortages of experienced staff.

Despite this, due to the high level of public fear and mistrust of the police, immediately after the transition to democracy, it may be appropriate for a newly established independent body, such as the ICD, to adopt the type of policies which the ICD has adopted. As former ICD Executive Director Adv Karen McKenzie argued, it would be unreasonable in such a context to turn people away and tell them to go to the police.

Such policies should however be seen as short term policies and there is no longer a need for such policies to be followed in South Africa. Particularly if efforts at police reform are reasonably effective, in the medium and long term such policies are counterproductive, for a number of reasons, as will be explained below.

No SAPS focus on improving internal systems

It should be noted that ICD policies have in some ways been complemented by the policies of the ICD which have placed very little emphasis on strengthening systems for receiving complaints and carrying out internal investigations. In late 2002 for instance the SAPS even went as far as to close down its own internal Anti-Corruption Unit as if there was no need for such a unit. There is little evidence that the SAPS has attempted to compensate for this loss in the period since then (Newham, 2005).

The weaknesses of policy in relation to systems for receiving and investigating complaints against police are therefore not only weaknesses of the ICD but are weaknesses which are shared by the SAPS.

Mobilising resources

A third reason which may help to explain why the ICD has adopted these types of policies has been that the ICD felt that it needed to be able to persuade the Parliamentary Committee on Safety and Security and other relevant bodies to increase the resources provided to it. It may also have been the case that the ICD believed that the Parliamentary Committee expected it to investigate all cases which it received, rather than having a more nuanced view. The ICD may have felt that it needed to be in line with the thinking of the Parliamentary Committee in order to be able to win its support.

Levels of complaints against the police in South Africa

The most recent ICD annual reports indicate that the ICD has received the following number of cases falling within its mandate over the last few years (see Table 1)

Table 1: Cases received by the ICD in recent years
Year Class I (Deaths in custody and as a result of police action) Class III and IV complaints Total cases received by the ICD
2002-2003 528 3 915 4 443
2003-2004 714 5 189 5 903
2004-2005 652 5 138 5 790

The ICD is informed about most deaths through notification by the SAPS and therefore it is not correct to refer to these in general as 'complaints'. Most complaints which are received by the ICD relate to other matter and are generally recorded by the ICD as Class III and Class IV cases. As is apparent from the above table in recent years the ICD has usually received in the region of 3900 to 5200 complaints in addition to the cases relating to deaths which it has received.10

These figures do not however reflect the total number of complaints and other cases involving the police which are lodged each year. Unfortunately the SAPS itself does not report in a consistent manner on complaints against the police lodged by members of the public.

Under the heading 'complaints mechanisms' the 2003-2004 report for instance indicates (p. 106) that:

By comparison the 2004- 2005 annual report indicates that for the National Complaints Line (08600 10111) 1 528 cases were dealt with for the period 'April 2003-March 2005' (p 160). Assuming this is a error and the statistic only refers to a period of a single year, it is still unclear why the latter statistics should be less than 8% of those recorded in the previous year. During the same period the ICD case levels remained roughly consistent.11

These two reports indicate that the SAPS receives somewhere between 1 500 and 21 000 complaints against SAPS members each year simply through the special national complaints lines. Many complaints are also lodged against police at the individual police stations. Many of these complaints received by the police from members of the public are probably dealt with informally rather than leading to a disciplinary hearing, and some of them may not be dealt with at all.

We therefore do not have reliable information on the total number of complaints against police which are lodged each year.

Factors affecting the level of complaints

The number of complaints recorded against the police in any society is a product of a number of factors which include:

If the body responsible for investigating complaints against the police is effective and user friendly public confidence in this body will grow and more and more people will report complaints to this body (in the absence of evidence to the contrary it might be concluded that this has been the main reason for the increasing number of cases received by the ICD – though note that the latest figures indicate that this was leveling off).

Therefore effectiveness on the part of the body responsible for receiving and investigating complaints may lead to more and more complaints being received, and may feed into an increasing level of complaints, and this is not simply dependent on whether police misconduct is going up or down (this is similar to the experience with crime statistics).

The price of improving systems for receiving and investigating complaints against the police may therefore be that an increasing number of these cases is received, and not necessarily that the number of cases will go down.

Potentially the levels of complaints against the police in South Africa may still grow substantially if we strengthen our systems for receiving such complaints, even if police service delivery and conduct improves simultaneously.

Taking into account the statistics in the previous section it would therefore appear that:

Considering that the ICD is not coping with its existing case load simply improving the resources of the ICD and improving internal systems and resource levels may simply result in the ICD receiving more cases, rather than helping the ICD to better ensure that the investigation of cases is effective and efficient.

Arguments against current ICD policy

There is evidence that the ICD is not coping with its current case load. This is not surprising since the SAPS is one of the biggest police services in the world and thus generates a large number of complaints.

Part of the manner in which organisations, including police organisations, regulate themselves is by promoting good performance and discouraging ill-discipline. Organisations may do this through a variety of mechanisms including statements of values or codes of conduct, performance management, promoting positive morale, and through disciplinary systems and measures.

Police organisations also face special risks in regard to the misconduct of their members, and appear to be prone to attracting complaints from members of the public even where they conduct themselves with a certain level of professionalism. There is therefore a need to establish special systems for receiving complaints against their members and for investigating these complaints.

But the ICD is not ensuring that the SAPS, and municipal police services are doing this. Instead of ensuring that the these police services have proper systems for dealing with complaints and ensuring that these complains are effectively and efficiently investigated the ICD has instead set itself up to substitute or replace proper internal police systems. Whilst setting itself up as a substitute for the internal systems of the police, the ICD is also not ensuring that complaints are in general effectively and efficiently investigated.

The result of this is that complaints which often reflect shortcomings of service delivery and misconduct by SAPS and MPS members come to be problems which are lodged with the ICD. When these complaints are not properly investigated, rather than helping to ensure that complaints are efficiently and effectively investigated, the ICD instead becomes an additional source of frustration and dissatisfaction.

Making the ICD into the key place where complaints are to be lodged and into the key organisation responsible for investigations is therefore essentially counterproductive. Responsibility and accountability for how complaints, and related issues of service delivery and discipline, are dealt with, is then displaced to the ICD rather than the police. When cases are not dealt with properly the ICD becomes the focus of criticism even if this is as a result of shortcomings on the part of the police. Not only is there little benefit in terms of improved police discipline, but this is also a sure recipe for increasing public dissatisfaction with both the police and the ICD.

The issue of resources

The approach taken by the ICD in accounting for the difficulties which it has in managing its case load has been to argue that the resources provided to the organisation are inadequate (see for instance the 2004-2005 annual report, p. 6).

The ICD has also drawn attention to the fact that while the number of police officers is being increased, and that its mandate has been extended to cover municipal police services, it has not received a commensurate increment in its budget (ICD Annual Report 2003-2004, p 2).

But, within current policy, there is no reason to believe that simply improving the resources, systems and capacity of the ICD will significantly contribute to the effectiveness of the ICD. Any improvements in the ICD's effectiveness may simply increase the ICD's case load with an increasing proportion of complaints, many of which are currently lodged with the SAPS, being reported instead to the ICD.

It cannot be disputed that it would be beneficial for resourcing of the ICD to be improved. But even if this is true, this in itself will not necessarily lead to the ICD being able to better manage its case load. Improved resourcing should be linked to the development of far more coherent policies regarding the role which the ICD is to play.

Establishing a synergy between the ICD and the SAPS

If one was to, for example, triple the number of ICD offices, and increase the ICDs budget and staffing in line with this, this would make ICD offices more accessible than they are now. But the SAPS stations would still be a far more accessible locality at which to report complaints.

Clearly an effective system should seek to make optimum use not only of the ICDs resources but also of the capacity of the police themselves to receive and investigate complaints.

A different system which would be far more productive in leading to public satisfaction with the resolution of complaints, in promoting police discipline, and in strengthening the effectiveness and credibility of the ICD, would not necessarily involve changing current legislation but would include the following elements:

Such a system would hopefully create a synergy between the ICD and police internal investigative systems. The ICD would be able to be more selective about which cases it investigates and thus focus its investigative resources on a narrower number of cases thus ensuring high quality ICD investigations. The ICD would 'interfere' in cases when there is reason to believe that the police have not dealt with cases properly. The police would thus want to ensure that complainants are satisfied that cases have been dealt with properly. ICD investigations would consistently be of a higher standard with the result that the police would want to avoid having the ICD come in to investigate cases as this would result in their own investigations being exposed as deficient. The police would therefore be motivated to conduct high quality internal investigations.

Ultimately therefore the issue is about developing a comprehensive policy regarding the lodging and investigation of complaints against the police that binds police services and the ICD.

Notes:

1 In addition it would appear that the ICD has also over the years received many cases which do not fall within its mandate and which it has referred to other bodies (see for instance Bruce, 2003).

2 See for instance the 'Report of the Portfolio Committee on Safety and Security on the Independent Complaints Directorate, 24 August 2005.

3 The reference here is to an untitled draft research report on case backlogs compiled by the ICD research unit and circulated in January 2006.

4 The responsibility of the ICD for oversight in relation to municipal police services is defined in terms of the Regulations for Municipal Police Services. These are issued in terms of section 64O of South African Police Service Act, 68 of 1995, as amended in terms of the South African Police Service Amendment Act, 83 of 1998, which provides that 'The Minister may prescribe which other provisions of this Act shall apply to a municipal police service…'.

5 There are also other obligations imposed on the ICD. Section 18 of the Domestic Violence Act, 116 of 1998 imposes obligations on the ICD to report to Parliament on cases dealing with failure by SAPS members to comply with the act, which have been reported to it by the SAPS. Like all other government departments the ICD has reporting obligations under the Public Finance Management Act and treasury regulations, obligations to its staff under employment legislation, and obligations to the public under Chapter 10 of the Constitution and under other legislation such as that governing access to information.

6 Related to this section 53(8) imposes an obligation on the police to report all such cases to the ICD.

7 There is also a question about whether certain complaints, particularly those relating to matters of 'service delivery', should be dealt with by the provincial secretariats. See for instance Mistry and Klipin (200:16).

8 This policy would appear to contradict section 53(2)(b) of the SAPS Act which requires that the ICD investigate all cases in this category.

9 As will be illustrated below it is possible that the bulk of 'complaints in respect of offences and misconduct allegedly committed by members of the Service' are received by the SAPS itself rather than the ICD. In addition to undermining its own ability to conduct good quality investigations, the ICD policy also gives no attention to ensuring that these complaints are properly investigated. Arguably the ICD is also failing to deliver on its key mandate by failing to deal with questions to do with the quality of the investigation of complaints which are received directly by the SAPS and municipal police services.

10 On the categorization of complaints by the ICD see ICD annual reports and Bruce (2003).

11 The reports also indicates that the SAPS finalised 9117 disciplinary hearings during 2003-2004 (p. 122) and 8 252 during 2004-2005 (p. 180). However these cases are not necessarily generated by complaints from members of the public so they cannot in themselves be used as a measure of the level of such complaints.

12 There have recently been indications that the SAPS is doing away with, or scaling down, its area command structures and that these may be replaced by a more decentralised 'zone' command system.

Bibliography

Bruce, D. (2003) Gripes or Grievances? What the Independent Complaints Directorate statistics tell us (or not). SA Crime Quarterly No. 4, June 25–30, 2003.

Centre for the Study of Violence and Reconciliation. (1999). Submission Regarding the Proposed Independent Complaints Directorate White Paper, October.

Centre for the Study of Violence and Reconciliation. (1999). A New Day? Controlling Police Violence and Corruption. In Crime and Conflict, No. 18, Summer.

Independent Complaints Directorate. 1998. 1997-98 Annual Report. Pretoria: Independent Complaints Directorate.

Independent Complaints Directorate. 1999. 1998-99 Annual Report. Pretoria: Independent Complaints Directorate.

Independent Complaints Directorate. 2000. 1999-2000 Annual Report. Pretoria: Independent Complaints Directorate.

Independent Complaints Directorate. 2001. 2000-01 Annual Report. Pretoria: Independent Complaints Directorate.

Independent Complaints Directorate. 2002. 2001-02 Annual Report. Pretoria: Independent Complaints Directorate.

Independent Complaints Directorate. 2003. 2002-03 Annual Report. Pretoria: Independent Complaints Directorate.

Independent Complaints Directorate. 2004. 2003-04 Annual Report. Pretoria: Independent Complaints Directorate.

Independent Complaints Directorate. 2005. 2004-05 Annual Report. Pretoria: Independent Complaints Directorate.

Masuku, T. (2005) Strengthening Democratic Policing in South Africa: Enhancing and Coordinating the Internal and External Accountability Systems of the South African Police Service. Research paper written for the Centre for the Study of Violence and Reconciliation.

Masuku, T. (2004) "Numbers that count: National monitoring of police conduct". SA Crime Quarterly, No 8. June 2004.

Mistry, D and Klipin, J (2004) Strengthening Civilian Oversight over the police in South Africa: The national and provincial secretariats for safety and security. Institute for Security Studies, Paper 91.

Newham, G. (2005) Strengthening Democratic Policing in South Africa through Internal Systems for Officer Control. South African Review of Sociology, 36(2), 160-177.

Portfolio Committee on Safety and Security (2005) Report of the Portfolio Committee on Safety and Security on the Independent Complaints Directorate. 24 August.

Walker, S (2001) Police Accountability: The Role of Citizen Oversight. Belmont: Wadsworth.

Centre for the Study of Violence and Reconciliation papers are usually available on the CSVR website. ICD annual reports are on the ICD website. Papers from the Institute for Security Studies and the journal SA Crime Quarterly are available on the ISS website.

© Centre for the Study of Violence and Reconciliation

 
Home | Programmes | Research | Consultancy | Coming Events |
Staff | Internships | Annual Report | Background | Aims |
African Transitional Justice Research Network |
Justice in Perspective | Links

eXTReMe Tracker