Centre for the Study of Violence and Reconciliation

Testifying Without Fear:
A Report on Witness Management and the National Witness Protection Programme in South Africa

by
Jenny Irish, Wilson Magadhla, Kevin Qhobosheane & Gareth Newham

Research report written for the Centre for the Study of Violence and Reconciliation, October 2000.

Jenny Irish, Wilson Magadhla & Kevin Qhobosheane are Research and Investigation Consultants at Injobo ne Blandla.

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

Acknowledgements

Firstly, special thanks must be given to the Irish Embassy without whose funding this research would not have been possible. The role of the Irish Embassy in funding civil society initiatives such as this is hugely appreciated. Thank you for your support.

Special mention must also be made to the team making up Injobo ne Blandla, Research and Investigation Consultants who were responsible for conducting the research and writing up the report. Thanks goes to Jenny Irish, Wilson Magadhla and Kevin Qhobosheane for all the hard work that they put into this project.

Appreciation must be extended to the members of the Criminal Justice Policy Unit at the Centre for the Study of Violence and Reconciliation, who raised the funds for and managed this project. In particular thanks goes to Gareth Newham for designing and managing the project. Also to David Bruce and Amanda Dissel for their assistance with the research, development and editing of this report.

Thanks must be given to all the people who gave of their time and expertise towards this report. The participant interviewees in this project were very forthcoming and provided the researchers with valuable insights into the functioning of the National Witness Protection Programme.

In particular, special thanks must go to Director Kleynhans who has the difficult job of heading the protection programme. His cooperation with this project is highly appreciated and proved invaluable for the compilation of this report. Appreciation must also be expressed towards the various staff members of the programme who were willing to give of their time and experience to the researcher.

We would also like to express appreciation to the members of the South African Police Service Special Investigation Units (SIUs), members of the Directorate of Special Operations popularly known as 'the Scorpions' and staff of National Prosecuting Authority. Interviews with the relevant individuals of these structures provided valuable insight into their work with witnesses and witness protection.

Thanks must also go to those witnesses who are presently on, or have been on the National Witness Protection Programme who participated in the research. While for obvious reasons, the names of these witnesses have been excluded from the report to protect their identities, appreciation must be expressed to them for sharing their experiences with the researcher.

Further mention must also be made of the participation of the representatives of Non-Governmental-Organisations (NGOs) with this project. As individuals who interface with witnesses on a regular basis their contribution to the research proved insightful and very useful.

ANC African National Congress
CIS Crime Intelligence Service of the South African Police Service
CSVR Centre for the Study of Violence and Reconciliation
DPP Director of Public Prosecutions (Previously known as the Attorney General)
ICD Independent Complaints Directorate
IFP Inkatha Freedom Party
ITU Investigation Task Unit
LHR Lawyers for Human Rights
LRC Legal Resource Centre
MI Military Intelligence
NIA National Intelligence Agency
NIM Network of Independent Monitors
NPA National Prosecuting Authority
NWPP National Witness Protection Programme
SAPS South African Police Service
SANDF South African National Defence Force
SB Security Branch (a notorious wing of the old South African Police that operated during the years of apartheid)
TRC Truth and Reconciliation Commission of South Africa
UMAC Urban Monitoring Action Committee
 

Contents

Executive Summary

1. Introduction

2. The Report
   2.1 Methodology in Compiling this Report.

3. Different Types of Witnesses

4. Intimidation of Witnesses

5. The Approach by Government to Witness Protection
   5.1 Problems in Policy Approach to Witnesses
   5.2 Practical Problems in Dealing with Witnesses

6. Background to Witness Protection
   6.1 History of Witness Protection in South Africa
   6.2 Lessons Drawn from these Programmes

7. The Current Witness Protection Programme
   7.1 Admission to the Programme
   7.2 Agreement Between the Witness and the Programme

8. The Witness Protection Act

9. Challenges Facing the Programme
   9.1 The Existence of a Single Programme
   9.2 Administration and Operation
   9.3 Resources
   9.4 Role of the Director
   9.5 Current Staffing of the Programme
   9.6 Screening and Evaluation Processes
   9.7 Physical Security
   9.8 Police Investigations
   9.9 Expectations of Witnesses
   9.10 Lack of Support Structures and Services
   9.11 Inflexibility in Dealing with Money Allocated to Witnesses
   9.12 Lack of an Effective Internal Complaints Mechanism
   9.13 Accommodation of Witnesses
   9.14 Cultural Diversity
   9.15 Rehabilitation of Criminals
   9.16 Abuse by Defence Counsel
   9.17 General Societal Attitudes

10. Relocation and Resettlement

11. Recommendations

12. Conclusion

References

Notes

Executive Summary

The focus of this report is on the National Witness Protection Programme (NWPP). Ultimately it is hoped that this report will be able to assist the relevant role-players with improving the support that the NWPP provides witnesses whose lives are at risk.

The report emphasises the importance of witnesses for an effective Criminal Justice System (CJS). Witnesses are critical for the agents of the CJS to successfully identify, apprehend, gather evidence and prosecute a vast majority of the criminal offenders that come into contact with it. Although key government policy documents tend to recognise the important role witnesses' play in the CJS, this report argues that the overall government approach has been inadequate. While special focus is placed on witnesses who are victims, or belong to particularly vulnerable groups, there is a failure to provide specified minimum standards of service to all witnesses of criminal incidents by both the police and the courts.

Against this background the report focuses on the National Witness Protection Programme. To date very little, if any independent research has been conducted on the NWPP. Much of the public information circulating about the programme emerges from media stories usually reflecting negative incidents that have occurred. A key objective of this report therefore is to explore and evaluate the functioning of the NWPP both in relation to other witness protection initiatives that have occurred over the past five years, and in terms of its own operations.

This report is in no way definitive or broadly representative of the entire NWPP. It is rather exploratory in nature and aims to provide better insight into the structure and functioning of the programme. The report is primarily based on 35 in-depth interviews with a range of individuals around the country who have first hand experiences of the functioning of the programme. These individuals include witnesses who were under protection, or who had previously served on the programme, government officials who work with or on the programme, police members who work for, or rely on the programme and members of Non Governmental Organisations (NGOs) who assist witnesses on the programme. Access to witnesses themselves was difficult to gain, and was dependent largely on NGOs and staff of the National Witness Protection Programme. However, despite these limitations, this report represents the first independent assessment of the programme to have been made since its inception in 1995.

The report presents both positive aspects and current challenges facing the NWPP at the present time. It also attempts to contextualise the programme against the backdrop of previous approaches to witness protection initiatives that were in operation prior to the implementation of the current programme. The report explains and describes the way that the NWPP is structured nationally and provincially. An explanation is given as to how a witness in danger gets admitted onto the programme and what they can expect, and under what circumstances they can be removed. The key provisions of the Witness Protection Act are also presented to provide an overview of the legal framework within which the NWPP operates. It is notable that the NWPP was in operation for a number of years before the Act was brought into operation. The report briefly explores some of the changes that accompanied the implementation of the Act in March of 2000.

The report further attempts to take into account the concerns of both witnesses and officials connected to the programme. These are presented under section nine headed "Challenges facing the programme." The aim of this section is to provide a list of issues that emerged during the interviews with respondents as key challenges that needed to be tackled to improve the programme. One of the key challenges identified is that of relocating witnesses who may still be in danger after testifying. This issue points to the need to conceptualise witness protection 'holistically'. A brief description of the approach towards witness protection taken by the Directorate for Special Operations popularly known as the 'Scorpions', provides content to such a conceptualisation.

Finally, the report presents a number of recommendations that it is believed will assist in providing a focused approach to the challenges highlighted. The recommendations provide a starting point for discussions on where to prioritise and what could practically be implemented. This report recognises the fact that the design and operation of an extensive National Witness Protection Programme is a highly complex endeavour fraught with difficulties. Any one of a diverse range of issues may emerge at some point with which any witness protection programme will have to engage. It is therefore important to emphasise that while there are priority areas that need more immediate consideration, there are no 'quick fix' remedies for improving the programme overall. It will take much trial and error to ensure that such a programme can adequately address the needs of a majority of protected witnesses. In doing so the programme will have to engage with the diverse needs and characteristics of South Africa's population.

This points to an important ability that a witness protection programme needs to develop. The ability to recognise problem areas and shortcomings timeously, be able to solve them speedily and then be able to make the necessary adjustments so as to minimize the potential for problems to re-emerge at a later point. It is towards achieving such an ability that this research hopes to contribute.

1. Introduction

Crime, and the development of effective mechanisms to fight crime, is one of the most crucial issues facing not only the government but also the country as a whole. "Crime affects the quality of life of every South African"1 which ultimately will have a detrimental impact on our emerging democracy.

There are a number of initiatives that can be developed by both the government and civil society to address crime. However, these initiatives can only be effective if they run parallel to the development of a more effective criminal justice system.

Within the criminal justice system, one area that requires immediate attention is the unacceptably low levels of arrest and subsequent conviction of people involved in crime.

According to the famous Rand Study of Detectives (Chaiken et al, 1991), conducted in the United States of America, police success in solving a crime is dependent on the existence of three factors:

The most important of the three is the existence of a witness. Physical evidence is often dependent on a suspect being identified. For example, forensic evidence is strengthened where a suspect already exists. Without the existence of a suspect, the evidence is not necessarily very helpful. The evidence provided by a witness can itself lead to a suspect confessing, particularly when he or she is confronted with the evidence of more than one witness.

The above mentioned study in the United States showed that 70% of reported crimes were solved because the person who had reported the crime or a member of the public could identify the perpetrator and was prepared to co-operate with the police in providing information with regard to the crime (Chaiken et al, 1991, p.172).

In the United Kingdom, studies which looked at the processes by which criminal cases were solved conclusively showed that, "… prime determinant of success was information immediately supplied by members of the public (usually the victim) to patrol officers or detectives" (Reiner, 1992, pp. 150 - 152).

The existence of a witness is not only important to the police in solving the case but also to the Justice Department in later securing a successful conviction.

A survey in the United Kingdom found that in three-quarters of the cases where the prosecution was unable to proceed, the reason given was that the key witness was missing or had refused to give evidence. In 88% of the cases where the witness had come forward and given evidence, a conviction was secured. In the United States, in 89% of the cases where witnesses were involved, the State was able to secure a successful conviction (Newham, 1995, p.6). In South Africa, a study by Bruce, Newham and Reddy (1999) clearly shows that many criminal cases fail as a direct result of witness related matters.

During the apartheid era, the criminal justice system and the police relied heavily on confessions, more often than not obtained through force. However, currently with the protections afforded by the Constitution and enshrined in the Bill of Rights, the use of force to obtain a confession is no longer acceptable. In this context South Africa, like most democratic countries, has become largely dependent on information supplied by members of the public and the testimony of witnesses which is not always easily obtained.

In the 1995 case of Zuma and Others v State,2 the court ruled that the State is now obliged to prove that if a confession is to be used as evidence, it must have been given voluntarily. The implication of this judgment is that the courts are aware that in a significant number of cases the confessions have been obtained by force. This judgment represents a move, by the courts, away from a confession-based system to a more evidence-based system. Within the evidence-based system, witnesses are critically important to the prosecution and the criminal justice system as a whole. The implication of this judgment and the move to a more evidence-based system leads Bruce (2000) to argue that far higher and more rigorous standards have to be applied than in the past in demonstrating that confessions presented before court have been given voluntarily.

According to Bruce (2000), then, although the confession remains a fairly important source of evidence, state as represented by the prosecutor is generally compelled to rely on evidence other than a confession in a much greater proportion of cases. Consequently, ensuring the co-operation of witnesses and the effective management of witnesses are crucially important aspects of the criminal justice system.

2. The Report

Every month, the criminal justice system has approximately 200 000 new victims and witnesses. Apart from these new witnesses, there are approximately 400 000 victims and witnesses involved in court cases and approximately 100 000 linked to cases under investigation. This means that at any one point the criminal justice system is dealing with approximately 700 000 witnesses and victims.3 The management and protection of these witnesses and victims must be seen not only as important obligations of the criminal justice system but also as effective means of fighting crime.

This report attempts to evaluate the management of witnesses and, in particular, focuses on the functioning of the National Witness Protection Programme over the last four years. The report is in no way definitive and is rather exploratory. It is based on a limited number of interviews with witnesses, officials of the programme, police officers and Non Governmental Organisations (NGOs). Access to witnesses themselves was difficult to gain, and was dependent on NGOs, staff of the National Witness Protection Programme and the Media. However, despite these limitations, the report is the first independent assessment of the programme to have been made since its inception in 1995.

The report looks at both the positive and negative aspects of the current witness protection programme. It also attempts to contextualise the current programme against the backdrop of previous approaches to witness protection programmes that were in operation prior to the implementation of the current programme.

The report attempts to take into account the concerns of both witnesses and officials with the aim of looking at shortcomings in the management of witnesses and of recommending possible means of addressing these problems.

2.1 Methodology in Compiling this Report.

The report is based on the following sources of information:

Whilst the focus of the report is on national approaches to witnesses and witness protection, a large percentage of the interviews and experiences were drawn from KwaZulu-Natal as it is the province with the highest number of people on the programme.

3. Different Types of Witnesses

There are a number of different types of people involved in supplying information to authorities on criminal matters:

All three types of people play a role in the combating of crime and where necessary should be supported and protected by the authorities.

In most cases, the individuals operating as informants or as sources should not be exposed to criminal elements. They can be protected through informal processes such as the use of false names or by holding meetings in "safe circumstances" which are unlikely to expose the identity of these individuals to the general public or to criminals.

On the other hand, witnesses are normally expected to testify in a court or some other form of legal proceedings. As a result, the witness will, more often than not, be exposed to the person he or she is testifying against. In this situation, the witness often requires a more formal protection process to that of an informant or source as described above.

There are also differences in the type of people who become witnesses and require protection. Some of these different types of people include:

These witnesses have often been involved with the very criminals they have agreed to testify against. The relocation of this type of witness poses particular problems for any witness protection programme as the witness can pose a threat to the community to which they have been relocated. Experience in other countries has shown that such witnesses are also prone to committing future crimes.

This type of witness is often traumatised by their experience at the hands of the criminals they are testifying against, and may require special counselling and support services to deal with these experiences.

This type of witness often has the most to lose and the least to gain by testifying against the criminals. In most cases, a criminal may choose to testify against fellow criminals to protect himself or herself. They will often have a personal interest in seeing the criminal prosecuted. A victim also will have a personal interest in seeing the criminal/s prosecuted. However, in the case of innocent bystanders, they often have no personal investment in testifying against the criminal.

4. Intimidation of Witnesses

One aspect that often hinders the State's ability to secure witnesses is the high level of intimidation of people who may be prepared to provide information related to a crime. As one police officer stated, "It (referring to intimidation) is one of the main reasons for the reluctance of witnesses to provide information" (cited in Bruce et al, 1999, p.134).

As a result of this intimidation, fear of reprisal has become a powerful tool in preventing people from supplying information about crime. This fear is not felt only by victims and bystanders, it often binds people to continued involvement with criminal elements and provides a powerful deterrent to such people from speaking to authorities about their own and fellow criminals' involvement in illegal operations. The more organised the criminal operations, the greater the danger posed in supplying information to the authorities. The National Institute for Crime and Rehabilitation of Offenders (NICRO), cites intimidation of witnesses by gang members as a primary problem in the quest to curb crime in the Western Cape (Newham, 1995, p.3).

Furthermore, it is extremely difficult to prove intimidation, and often the threat is made in such a way as to make it impossible to prove legally. There are also numerous occasions where witnesses or other citizens have been blatantly threatened in front of police officers and no action has been taken. Rape victims have been openly threatened by the rapist whilst attempting to lay charges. During mediation between conflicting groups there have been instances when intimidation has taken place and people have even been threatened with guns. Documentation has been made of instances where people speaking at public meetings have made threats from their public platform without action having been taken by police officers present (Newham, 1995).

It is crucial that police officers take decisive action where possible to break the cycle of intimidation. In addition to taking action against people involved in intimidation, it is also of vital importance that people prepared to give evidence against individuals involved in criminal activity are protected and properly managed.

In the British Home Office, Speaking up for Justice (1998, p.12), it is stated that threats experienced by witnesses could be classified into the following three categories:

This consists of the most serious cases where intimidation is a life and death issue and where high levels of protection are required.

This consists of witnesses who have been intimidated but where the intimidation is not life threatening.

This consists of people discouraged from co-operating based on the perceived risk of threat or harm.

The existence of a witness protection programme is a major contributing strategy for dealing with the small inner core of witnesses who face life-threatening situations. However, the situation regarding the middle ring and the outer ring also needs to be addressed by generally improving the witness management system.

5. The Approach by Government to Witness Protection

Prior to 1992, no witness protection programme existed. However, legislation made provision for witnesses to be placed in "protective custody" under Section 185 of the Criminal Procedure Act.4 This section made provision for the Attorney-General to place a witness in protective custody if it was deemed necessary. Under this section, protection was not voluntary and the witness could be kept incommunicado. In such a situation, the witness was only allowed access to lawyers with the permission of the Attorney-General. The Act essentially allowed for the authorities to hold witnesses in detention. Witnesses were regularly detained involuntarily, largely to be used as state witnesses in cases involving political trials. The system was based on coercing people into giving evidence and focused on getting suspects to turn state witness in order to secure a conviction. Normally once a conviction was achieved, the witnesses were released and left to fend for themselves regardless of the consequences of having given evidence.

This legislation and the attitude of the authorities were restrictive and repressive in nature and did little to instill confidence and belief in the existing justice system.

As part of transforming the criminal justice system, it has been, and continues to be, necessary for this system to be entirely overhauled. The changes are important for the following two reasons:

The Government has introduced particular measures to address the protection of people prepared to supply information on criminal activities. One such measure is the introduction of legislation related to Whistle-Blowers. Another is the introduction of a Witness Protection Programme.

The key policy papers for both the Department of Justice and the Department of Safety and Security refer to witnesses in their policy papers. In the White Paper on Safety and Security entitled, In Service of Safety 1998 – 2003, it is argued that victims and witnesses play an important role in assisting both the police and the prosecution. The chapter goes on to list eight points intended to detail "good practice for servicing and protecting victims" (1998, p. 19). Four of these points refer exclusively to victims and the other four to both victims and witnesses.

The Department of Justice in its policy document entitled, Justice Vision 2000, also makes reference to the importance of witnesses and states:

We cannot try cases properly without witnesses. At present, we often have to withdraw criminal cases, or postpone them indefinitely because we have no witnesses. Witnesses are reluctant and afraid to come to court because they are afraid of being victimised by the alleged criminals or because they have little confidence in the criminal justice system. We need to make sure our witnesses are safe at court proceedings, so that justice can be done swiftly. The criminal justice system is like a chain whose strength is only as good as its weakest link. Our lack of witness protection is often our weak link. (1997, p.42)

Justice Vision 2000, discusses a strategy to address the protection of witnesses which particularly refers to the creation of an effective witness protection programme and a need to address vulnerable witnesses such as women and children.

5.1 Problems in Policy Approach to Witnesses

Some serious shortcomings have been identified regarding the policy approach of both the Ministry of Safety and Security and the Department of Justice towards servicing witness throughout the criminal justice system (Bruce et al, 2000; Newham, 2000). The White paper on Safety and Security can be criticised on two levels. Firstly, it is considered too general and vague, particularly with regard to witness protection. Bruce et al (2000, P.5), points to this vagueness which, "… is apparent for instance in its treatment of witness protection. With existing approaches to witness protection generally being either too costly and disruptive, or ineffective, the document fails to clarify how effective witness protection is to be provided." Secondly, the White Paper fails to distinguish between victims and witnesses. In some situations witnesses are not in fact victims and as a result cannot be placed into the same group. The White Paper is also criticised for not paying sufficient attention to other issues regarding witnesses who are not victims (Bruce, 2000; Newham, 2000).

As discussed above, Justice Vision 2000 recognises the importance of witnesses generally and the need to establish a witness protection programme. However, it deals only with the issues facing witnesses who are women or children and those witnesses requiring serious protection though a witness protection programme. The needs of a broad group of witnesses to criminal incidents who do not fit into either of these categories are not specifically addressed by the Justice Department's policy document. It is for this reason that arguments are made that criminal justice policy documents are incomplete in addressing the needs of a majority of witnesses.

As can be seen, there is a great need for the development of a general witness management approach within the criminal justice system. This must include witnesses who are victims as well as those who are not. Government policy fails as a starting point to place the witness in the centre of the system to deal with criminal offenders. An appropriate policy framework would seek to ensure that the different structures of the CJS, the police and courts in particular, are as user friendly and accessible as possible to any individuals wishing to report a crime, assist investigations and/or testify in court. Such a policy initiative should then go hand-in-hand with initiatives to address the special needs of vulnerable categories of witnesses such as children, or people whose lives are at risk.

5.2 Practical Problems in Dealing with Witnesses

On a policing level, cases may very often not be solved because police officers either fail to recognise the importance of the witnesses or they lack the appropriate skills to deal with witnesses. According to Bruce et al (1999, p.135), "evidence suggests that witnesses are often under utilised".

The police often from the start fail to effectively identify the witness in a case. Furthermore, the witness may not come forward because of lack of confidence in the treatment he or she expects to receive from the police. One detective told researchers "if you are a witness there is no security. There is some mistrust between the police and the community as people have experienced the police telling perpetrators who the witnesses are" (Bruce et al, 1999, p.135). In other situations, a lack of effective interviewing or failure to take a proper statement may result in important information and evidence being lost.

One problem that faces the police is that whilst the judicial system has moved from a confession-based system to a more evidence-based system where witnesses are crucial to the system, many police are unable to operate within this new context as they lack skills to deal with witnesses and to build the trust of such witnesses. Many of the older detectives who relied on confessions to solve their cases have no training or experience in dealing with witnesses and in encouraging witness and community co-operation and confidence in the system.

Even if witnesses agree to co-operate with the police, very often the neglect of the witnesses or long delays in the court process may lead to the case failing once it gets to the court stage. The witness may be badly managed, not only by the investigating officer but also the prosecutor. Furthermore, the witness may fail to come to court because they have not received proper notice or because they have no means of getting to court. Long drawn out court cases may also result in frustration on the part of the witnesses or could open the witness to potential intimidation. According to one prosecutor, "court cases can take a long time and the longer the case draws out the more chance there is that the witness may be intimidated."5

In order to ensure the criminal justice system is made more effective, it is essential that the area of witness management receive greater attention from both the South African Police Service and the Department of Justice. This should include convincing witnesses to co-operate with the police and then establishing effective means and approaches to managing the witnesses during subsequent investigation and prosecution.

In order to ensure the co-operation of witnesses and to address problems of intimidation, the SAPS and Justice Department will need to develop different approaches depending on the level of intimidation being experienced by a witness. An effective protection programme can protect the small inner core who require high level protection, and alternative measures need to be looked at with regard to addressing the middle ring and outer ring. In this regard, effective systems need to be developed that will not only seek to maximise the use of witnesses but incorporate a greater responsiveness to the needs of witnesses by both the police and the Justice Department.

In addition, while it is crucial that key witnesses from the inner core who require high levels of protection have access to a protection programme, there will be some inner core cases which are not suited to the programme and alternatives will need to be sought for these cases.

6. Background to Witness Protection

The Witness Protection Act came into operation on 31 March 2000. While the Act has only been in operation for less than four months, the current programme has existed since 1995 when the Minister of Justice created a Directorate of Witness Protection. This witness protection programme was not the first to operate in the country. After 1992, a number of different witness protection projects and programmes were established, all of which have had some influence on the structure and operation of the current programme.

The programmes that existed between 1992 and 1995 were all relatively limited in nature and the current programme itself is still relatively young, having only been in existence for four years. There are witness protection programmes, some of which have been in existence for a long time, in a number of other countries, which may offer some insights into certain aspects of the programme. However, the vast differences between South Africa and many of these other countries make it impossible to see these programmes as blue prints for our own situation.

The result is that while there are lessons that can be learned both from South Africa's past experience and from programmes elsewhere, at the end of the day these lessons remain relatively limited and new solutions need to be found that are relevant to the circumstances in South Africa. In 1995, it was noted that it was likely that an effective programme would only be achieved by trial and error over a long period of time. There is also a need for constant re-evaluation as well as regular review and improvements of the programme which would lead to the development of a more effective programme (Newham, 1995, p.13).

6.1 History of Witness Protection in South Africa

After 1992, a number of different programmes were introduced to guarantee the protection of witnesses in specific situations, and were linked to specific investigations or Commissions of Inquiry. However, it was not until 1995, when the Minister of Justice announced the appointment of a Director of Witness Protection that a comprehensive National Witness Protection Programme came into existence. This indicated a real attempt by the Department of Justice to establish an acceptable form of protection for witnesses.

The current programme has had the advantage of being able to draw certain lessons from the specific programmes that had been used introduced between 1992 and 1995, although as stated earlier, these programmes themselves were relatively limited in nature. These programmes included:

The Goldstone Commission was established in 1991 and was chaired by Judge Richard Goldstone. The Commission was established under the Prevention of Public Violence and Intimidation Act of 1991. The Commission operated from 1991 until the elections in 1994. The main brief of the Commission was to inquire into the factors behind public violence and intimidation and to research the prevention thereof. The Commission dealt with cases of political violence and the existence of hit squads.

The Goldstone Commission Witness Protection Programme only came into existence a year after the Commission started. New legislation was passed in terms of the Prevention of Public Violence and Intimidation Act, which allowed for a voluntary protection programme for witnesses appearing before the Commission. The witness programme was not really planned for and was largely a crisis response to serious security problems certain key witnesses were experiencing. In most instances the witnesses were accommodated in hotels under false names and at the expense of the Commission.

The programme was run by staff from the Commission, most of whom had no real experience in witness protection. The Commission's programme can be considered as the first official voluntary witness protection programme. It protected 37 witnesses and 40 family members of witnesses.

After the Goldstone Commission had received evidence of the existence of hit squads and covert operations operating within the South African Police, a special unit was established under the Transvaal Attorney-General, Jan D'Oliviera. The D'Oliviera Unit, as it became known, was involved in some high profile investigations into hit squad operatives and operations.

The D'Oliviera Unit protected three witnesses who were testifying in the trial of Eugene De Kock (a commander of one of the Police hit squad units). The witnesses had come through the Goldstone Commission and it was Judge Goldstone who organised for the witnesses to be protected. The witnesses, who were police officers, were protected in Denmark due to a fear that they could be intimidated by fellow police officers if they remained in this country. They remained in Denmark for a period of approximately 18 months.

The protection of the three witnesses was organised through personal contacts Judge Goldstone had in the Danish Embassy. The programme was funded by the Danish Government and had the support of a number of government departments in Denmark including Foreign Affairs and the Welfare Department. This programme also accommodated the families of the witnesses. The cost of the programme was approximately R1 million.

According to one of the people centrally involved in securing the testimony of the three witnesses,6 the programme was generally successful. In addition, most of the people involved in the prosecution team agree that had the three witnesses stayed in South Africa, it could have jeopardised the subsequent conviction of De Kock.

The Investigation Task Unit (ITU) was established at the end of 1994 to look at the existence of hit squads in the KwaZulu-Natal Police and the relationship between these hit squads and the South African Police and Military Intelligence. Due to the sensitivity of the matters being investigated, the ITU was made up of a group of handpicked police officers. The police officers were accountable to a civilian board of attorneys who were appointed by the Minister of Safety and Security and who reported directly to him.

The witnesses protected were linked to the case involving the ex-Minister of Defence, Magnus Malan, a senior ex-member of the KwaZulu-Natal Government and other high-ranking ex-apartheid security officials. The charges laid against these people were related to their involvement in hit squad activity and the massacre of a family in KwaMakutha outside Durban.

The ITU established its own protection programme based on the experiences of the Goldstone Commission. The programme was run and administered from within the unit and was responsible for the protection of certain high profile witnesses. Between 1995 and 1997, the programme protected only a limited number of witnesses (eight witnesses in total), although most of these were in serious danger. Most of the witnesses protected were taken out of the province of KwaZulu-Natal and protected in other provinces.

At the end of 1994, one of the Deputy Attorney-Generals, Chris MacAdam, established a mini-witness protection programme. The programme was specific to the Durban area and was responsible for the protection of 20 people. This was the first official programme not linked to a Commission or specialised investigation unit and which was voluntary. The programme did not place witnesses in prisons or police cells but instead sought "safe houses" or venues to house its witnesses. The KwaZulu-Natal programme was also the first programme established and run by an office of the Department of Justice.

The project was established after key witnesses to a massacre in KwaMashu agreed to testify against key "warlords" as long as they were guaranteed protection.

The Durban Deputy Attorney-General approached businesses to donate property and resources to the project. Money from rewards offered in the case was collected and then used for the project. The project was administered through the Durban Clerk of the Court.

Despite the small size of the project, it was able to impact on a lot more than six cases and served as an important example of the effect a witness protection programme can have. Through the co-operation of protected witnesses, the main perpetrators of violent crimes were convicted and sentenced. As a result of these convictions, there was no need for further protection programmes in certain areas because the key people involved in intimidation were behind bars.7

Based on the need to secure people prepared to give information related to political conflict that had emerged during the Goldstone Commission, and drawing on the experiences of both the Investigation Task Unit and that of the Durban run programme, the Truth and Reconciliation Commission (TRC) established its own witness protection programme. The legislation governing the establishment of a limited protection programme was set out in the Promotion of National Unity and Reconciliation Act 34 of 1995.8

Through this legislation, the TRC established an internal programme. The programme fell under the investigative arm of the TRC. The programme attempted to recruit people who had expertise in this regard. These experts specifically recruited to the programme included:

The TRC made use of government facilities where possible to house witnesses at relatively limited cost to the Commission. The programme also made use of social workers and psychological staff from the TRC's Reparation Committee to offer counselling to witnesses.

The TRC programme offered skills training to witnesses by making use of the services of NGOs who offered training and development. Witnesses under the TRC protection programme were protected for the duration of the TRC hearings, which meant that most witnesses spent between 8 and 12 months on the programme.

The TRC had a limited budget for witness protection (the budget was approximately R3 million to protect 600 witnesses), and as a result, the Commission only placed witnesses on the programme as a last resort. Rigorous admission criteria were used. A witness was only admitted to the programme after an evaluation was conducted by one of the police officials seconded to the programme to protect witnesses. The relevant commissioner in the TRC Investigation Unit then confirmed the recommendations of this evaluation. According to the TRC Final Report, "… this procedure protected the programme from abuse by persons who were either offering untruthful evidence or were not in danger" (1998, p.388).

Once it was established that there was a need for the TRC to play a role in regard to protecting the witnesses, a further evaluation was conducted in order to determine the nature of the risk. According to the TRC Final Report, "Persons assessed as low risk were placed in community-based projects, and only persons assessed as medium or high risk were placed in safe houses" (1998, p.388).

According to the TRC, requests for protection came from four different sources:

The TRC report states that confidence tricksters were identified by the staff of the programme without compromising the programme itself. The Final Report states that, "In one extreme case, the culprit was prosecuted on a charge of making a false statement to the Commission and sentenced to a year's imprisonment" (1998, p.399). According to the head of the TRC protection programme, there were only a handful of confidence tricksters who tried to enter the protection programme on this basis.

Once witnesses had been evaluated and ranked according to the risk they faced, those deemed to face low or medium risk were not placed on the programme. The TRC witness protection staff attempted to see if such witnesses could stay with family members or friends who lived outside the area where the danger existed. When this was not possible, the staff of the programme liaised with the police and asked them to increase patrols in the areas where the witnesses resided.

The final report of the TRC claims that in situations where witnesses were deemed to face a low or medium risk, the Witness Protection Programme made use of NGOs involved in combating crime in the communities in which they worked. They included violence monitors, Community Police Forums (CPFs) and visible policing structures who visited witnesses to discuss security and to keep regular contact with the witnesses. The TRC Final Report (1998, p.389), states:

This method of protecting witnesses, which had not been previously attempted in South Africa, proved highly successful and had the following advantages:
a) The witness' life was not disrupted and the attendant problems of loneliness, boredom, alienation and potential loss of employment were avoided;
b) The police, previously viewed as enforcers of the apartheid system, now became the protectors of victims, thus helping place the relationship between communities and the police on a better footing;
c)Witness protection officials were free to devote their attention to cases which warranted protection;
d) The notion of the need to protect witnesses was promoted in communities.

Although the TRC report states that these methods appeared to be successful, it was impossible to assess how effective this alternate type of protection was and whether it was effectively implemented. According to one investigator from KwaZulu-Natal, these initiatives were never implemented in KwaZulu-Natal or Gauteng, and as a result people classified as medium or low risk had to take responsibility for their own safety.9

The TRC protection programme made use of intelligence when protecting witnesses, and in its Final Report " (1998, p.389), refers to "unofficial nodal points being established to assist in gathering of intelligence. These nodal points included the National Intelligence Agency (NIA), the D'Oliviera Investigation Unit, the Investigation Task Unit (ITU) and the Justice Department Witness Protection Programme.

The TRC programme offers some important lessons that are of value to the current programme, in particular, the creative use of both personnel and resources. It also attempted to address the general needs of witnesses and not only their security needs.

However, there have also been concerns raised about the programme. The main concern being that some of the staff members had special security backgrounds, for example, they had been members of the Security Branch. This was not appropriate or safe given that a number of the witnesses were testifying against people from these security structures. For example, one witness who was a member of the previous SADF and who had been stationed in a Military Intelligence Unit in one of the provinces, alleged that one of his protectors, himself a former SAP security branch member, had telephoned the SANDF in that province and asked if they knew the witness. The protection officer had then told the SANDF person contacted the nature of the allegations the witness was making regarding their involvement in crimes during apartheid. The witness alleged that after this he no longer felt secure within the programme.10

6.2 Lessons Drawn from these Programmes

Although all the programmes listed above, with the exception of the TRC programme, protected a small number of witnesses, there are nevertheless a number of lessons that can be drawn from these programmes. Some of these lessons include:

The experience of the D'Oliviera unit in protecting witnesses outside South Africa, with the help of the government of the country offering sanctuary to the witnesses, is an important one. This practice may well need to be built on and developed further, particularly in cases involving organised crime and high political trials.

The Truth and Reconciliation Commission made use of intelligence structures in the protection of witnesses. Through the establishment of regular contact with NIA and other intelligence structures, the TRC was able to gather important information regarding the threat facing the protected witness. This type of co-operation will be important in the future and could assist the process of witnesses' protection by:

Both the TRC and Goldstone Commission were protecting witnesses testifying before a commission and not judicial court cases. It is important to recognise that it is not only witnesses who are appearing before formal judicial courts that require protection but also witnesses appearing before commissions of inquiry.

The TRC was able to protect 600 witnesses for more than a year at a cost of approximately R3 million. The TRC was able to reduce the number of resources required to protect witnesses largely through the use of State facilities to house witnesses. Accommodation is often one of the biggest expenses witness protection programmes face, and there is a need to look at what State facilities can be used in this regard.

The TRC programme and the KwaZulu-Natal programme of the Deputy Attorney General made use of resources outside the State to assist in the process of protecting witnesses. In the case of the TRC, use was made of Non-Governmental Organisations to provide skills training for witnesses. This helped in providing witnesses with opportunities to make a living after the protection came to an end. The KwaZulu-Natal programme tried to develop partnerships with certain businesses that agreed to make available accommodation to house witnesses.

While it is primarily the responsibility of the State to protect witnesses, given the limited resources available for witness protection in South Africa, it will be important for the Witness Protection Programme to form partnerships which could assist the programme.

The TRC process of assessing witnesses prior to accepting them onto the programme was an important one not only in assessing the threat facing the witness but also in evaluating the possible needs of the witness.

7. The Current Witness Protection Programme

The current National Witness Protection Programme has now been in existence for just over four years. The programme falls under the Department of Justice, which has created a Directorate of Witness Protection. Although the programme operated under the old Section 185 of the Criminal Procedure Act,11 until the new legislation came into existence on 31 March 2000, admission to the programme has been voluntary.

The Head Office of the Witness Protection is run by the Director of Witness Protection who has a small office in Pretoria. The Director is an employee of the Department of Justice and is appointed by the Minister of Justice.

Each province has a small team of SAPS officers who are responsible for carrying out the actual protection of witnesses. All SAPS members who are part of the programme are appointed at a provincial level and are accountable to SAPS Provincial Management through the Provincial Commissioner of Police in that office. The number of SAPS members allocated to each province varies according the number of staff allocated by the Provincial Commissioner's office in each province. The total number of police officers working on the programme nationally in April 2000 was 67.

The role of the police officer with regard to the programme is:

The salaries of the police officers come from the provincial police budget. However all expenses incurred by the witnesses are paid by the Head Office of the Witness Protection Programme and form part of the Department of Justice budget.

Since its inception in 1995, there has been a consistent increase in the number of witnesses being protected.

Table 7.1: Number of Witnesses Protected per Year
Date 1995 1996 1997
Number of Witnesses 40 200 550

The Witness Protection Programme in April 2000 protected 756 witnesses involved in 330 criminal cases.12 Over the last four years the majority of witnesses on the programme have come from KwaZulu-Natal. In April 2000, KwaZulu-Natal had 100 more witnesses under protection than any other province in the country.

There are two possible reasons that the demand for witness protection is greatest in KwaZulu-Natal. The first is that unlike other provinces in the country, KwaZulu-Natal has continued to experience political violence after 1994. A significant number of these witnesses have provided evidence related to ongoing political conflict in the province; currently 12 of the witnesses on the Witness Protection Programme come from the Scorpion Unit's investigation into hit squads in Richmond, KwaZulu-Natal. The second reason KwaZulu-Natal has more witnesses on the programme than other provinces could well relate to the fact that KwaZulu-Natal has more experience of witness protection than any other province. Prior to 1995, KwaZulu-Natal not only had experience regarding the Goldstone and TRC programmes, but also with the ITU and Deputy Attorney-General's protection programmes both of which were specific to KwaZulu-Natal. As a result of these experiences, both the police and the general public have greater knowledge of the programme as a possible option for witnesses.

The programme had a budget of approximately R14,5 million for the year 2000. Some 160 properties are rented on a monthly basis. There are also some witnesses accommodated in hotels. The programme spends just over half a million per month on rent and close to R2 million per year on transport.13

7.1 Admission to the Programme

To qualify for admission to the programme, a person should be a potential witness (or a dependant or family member of a witness) to a Schedule One offence or the offence of bribery, blackmail, perjury or defeating the ends of Justice. The witness and/or family members must also be "in danger" because of the fact that the witness will testify.

Over the last four years, it has been the Attorney-Generals (now called the Directors of Public Prosecutions - DPP) who have made the decisions about who should be protected on the programme. The protection has also come to include family members of the witness where this is required. The DPP has relied largely on information supplied by the investigating officer when deciding whether the witness requires protection.

The person applying for protection does so by filling out a prescribed form and handing it to any one of the following people:

The prescribed form is accompanied by a statement from the investigating officer of the case the witness is to testify in. The statement includes the nature of the case, the offences involved, the nature of evidence the witness could furnish and the source and nature of the danger to the witness or family member. This form is immediately submitted to the Witness Protection Programme Head Office.

Arrangements are then made by the officials of the programme to place the witness in a safe environment for 14 days. The Director of Public Prosecutions concerned is then approached by the investigating officer or a witness protection co-ordinator for a decision on whether the person should be kept on the programme. The Director of Public Prosecution's decision is then communicated to an office of the Witness Protection Programme. If the Director of Public Prosecutions confirms that the witness requires protection, this will take immediate effect. If the Director does not confirm the need for protection, the witness will be removed from the programme with immediate effect.

7.2 Agreement Between the Witness and the Programme

If the witness and/or family members are kept on the programme, they are required to sign a memorandum of understanding. This memorandum sets out the rights and obligations of both the Witness Protection Programme and the witness themselves.

The Justice Department does not pay a witness to testify. The Department may, however, reimburse the witness for any proven loss of income forfeited as a result of the witness joining the programme.

After checking the memorandum of understanding, the head office of the programme will determine what allowance will be paid to the witness. If the witness was unemployed prior to entering the programme and is staying in a catered facility, they will receive an income of R120 per month. If the witness is staying in self-catering accommodation, the amount received will be R570. This amount will be increased depending on the number of people in one family under protection.

If the witness was employed prior to the joining the programme but cannot provide proof of employment, the basic allowance will be R300 in catered accommodation or R750 if in self-catering accommodation.

If the witness is able to provide proof of previous income, the witness will receive their actual income less any expenses no longer incurred (eg rent) because the Witness Protection Programme now covers these.

The programme aims to remove the witness from the area where they have been living by hiding them in another area. According to an information bulletin issued by the Justice Department in June 1997, the protection will come to an end on finalisation of the case. However, the bulletin also states that the programme will not abandon a witness at the end of the case. The bulletin states: "If possible, the question of resettlement should be attended to even before the conclusion of the criminal case."14

The Witness Protection Programme tries to accommodate witnesses in a different province from the one in which they normally reside. The memorandum of understanding states that the standard of living prior to admission to the programme will be taken into account but will not be decisive in determining the place of safety where the witness will be kept, and accommodation may vary from case to case. The memorandum states that in appropriate cases the witness will remain in his or her house but guards will be furnished to protect the witness. The memorandum also states that in exceptional cases the witness could be kept outside the borders of South Africa if it is shown that the witness will not be safe if kept inside the country. The memorandum further states that as a general rule witnesses should not be kept in hotels or other places to which the public has access.

If a witness chooses to leave the programme prior to being discharged, the witness must sign a waiver and the Witness Protection Programme will no longer be responsible for their safety. According to the Director of the programme, once a person leaves the programme either voluntarily or after being discharged by witness protection officials, the witness will be transported to any place in the country the witness chooses. However, a witness who abuses a place of safety or commits criminal acts while on the programme will be charged and removed from the programme.

During the period in which the witness is being protected, when the family of the witness is not on the programme, arrangements can be made to visit family members under controlled conditions that will ensure the safety of the witness. Requests to visit families should be made in writing to the Director and, if possible, the witness or family member making the request should suggest how best they think the visit should occur. According to the Director, the programme has limited resources, and visits from family members need to take this into account. An example of this is with regard to a witness in North West Province whose family is not on the programme. Witness protection officials arrange three monthly visits between this witness and his family.

In all provinces, with the exception of KwaZulu-Natal, all funds to be used by the programme are channeled through Pretoria and the police officers assigned to the programme are then responsible for administering these funds. However, in KwaZulu-Natal the Clerk of the Durban Magistrates Court performs this function and the police officers are relieved of most administrative duties.

Although there are no national figures available regarding the number of successful convictions secured using witnesses on the programme, the Director of the programme states that over the last three years, for Gauteng alone, 42 people have been sentenced as a result of testimony offered by witnesses on the programme. Of the 42 sentenced, 27 of these people were given life sentences. According to the Director, the 42 were all convicted of serious offences including murder, armed robbery and taxi violence. The Director goes one step further to say that virtually all serious prosecutions currently underway in the country place reliance on testimony of people under protection by the programme.15

Section 13 of the new Act provides for procedures to be followed when witnesses get removed from the programme. Witnesses are given six weeks trial as a matter of course within which to make arrangements to leave the programme and return home following the conclusion of the trial with which they are involved. In special cases where this time is insufficient, witnesses may request in writing for a further six week period.

8. The Witness Protection Act

The Witness Protection Act was passed by Parliament in November 1998 and Gazetted on 27 November 1998. The Act was a result of extensive discussion within the Justice Department and with prosecution authority as well as public hearings, and was based on lessons learned from previous experiences of limited witness protection programmes introduced after 1992 and on submission by Non Governmental Organisations, legal associations and practitioners. The Act only came into existence in March 2000. One of the reasons for the delay was the need to establish whether structure and resources were available for the implementation of the Act.

The new Witness Protection Act,16 whilst providing a central base for the running of the programme, provides for the creation of regional branches with witness protection offices in the other five provinces. Prior to the implementation of this Act, the programme existed in different provinces but the only people working in these provinces were the police officers made available to the programme. With the implementation of the Act and the establishment of regional branches in six of the 9 provinces, administrative personnel from the Department of Justice will be appointed to the six provinces where regional branches are to be established. The administration of the other three offices will continue to be controlled by the Pretoria office. This change is unlikely to impact on witnesses under protection but is rather an attempt to introduce more effective administrative control over the programme.

Furthermore, prior to the implementation of the Act, the role of the Director was purely administrative. However, under the new Act, the Director's role is more proactive in that it includes considering applications to the programme and the evaluation of continued accommodation of witnesses on the programme. This role that was previously performed by the Attorneys-General.

Previously, protection of witnesses was only available to witnesses testifying in criminal court proceedings. The Act now provides for the protection of witnesses giving evidence outside the scope of ordinary criminal cases. These situations include witnesses in commissions of inquiry, inquest proceedings, proceedings of special tribunals and investigations of the Independent Complaints Directorate (ICD).17

The new Act also places a duty on other State departments to render reasonable assistance to the Witness Protection Programme. This means that the Witness Protection Programme now has the capacity and authority to request the Public Works Department to make facilities available to the programme or the Social Welfare Department for assistance with the programme.

Section 13 of the Act states that a person on the programme must be given written notice of the intention of the programme to discharge him or her and this notice must give reasons for why the witness is to be discharged. The witness must be notified that he or she is entitled to take this decision to the Minister of Justice on review. If the witness decides to take this course, they must submit written representation within five days of receiving notice of the Director's intention to discharge the witness.

The Act lists the following reasons why a witness may be discharged from the programme:

The Act also allows for a witness who is refused entry to the programme to take the matter up on review to the Minister of Justice. Pending the outcome of this review the witness will be placed on the programme.

Section 17 (8) of the Act makes it an offence to disclose any information that may endanger the security of a witness, and Section 22 of the Act makes it an offence to allow unauthorised access to a witness or to disclose information and the identity of a witness on the programme. According to section 22, the offence can result in a penalty of a fine or a maximum of 30 years imprisonment.

Over the last four years, the police officers serving on the original programme have come from different police units appointed by the provincial police management. In November 1999, the South African Police Service conducted a work-study of the programme. The study proposed the creation of special witness protection units in each province with a unit management structure based in Pretoria. This unit would be a national unit and the members appointed to this unit would be accountable to the national management of the SAPS.

However, according to Section 6 of the Act, the Director General of Justice may request the following people to second staff to act as security officers for the programme:

The security staff will be accountable to the Director of Witness Protection during their secondment and not the Provincial Commissioners of Police. The implementation of this section of the Act has also scuttled police initiatives to establish an SAPS witness protection unit.

There have been experiences in the past where for the safety of the witnesses involved it was necessary to protect them outside of South Africa. Section 21 of the Act allows the Minister of Justice to enter into agreements with international bodies, institutions, organisations or foreign countries regarding protection of a witness. According to a representative of the National Public Prosecutor's office:

Supplementing the existing capacity for witness protection cannot be done by South Africa alone. We need an international support base similar to what was done by the D'Oliviera Unit. This will be a quid-quo-pro agreement and we are assisting other countries with sanctuary so that we can relocate some of our complex cases. In terms of these complex cases we are only talking about 3 or 4 of the most complex cases.18

A number of issues are not dealt with but are referred to the Minister to pass regulations to govern certain areas regarding the protection of witnesses on the programme. One crucial area where the Minister is entitled to pass regulations on is the protection of the identity of a witness. Under the Witness Protection Programme the identity of a witness may be changed but the programme has had no legal means of verifying the previous qualifications of a witness. The Minister is in the process of passing a regulation that will allow him to issue a certificate specifying the qualifications of the witness under his or her new identity as long as sufficient proof is given.

9. Challenges Facing the Programme

Over the last four years a number of problems have surfaced regarding the Witness Protection Programme which have affected perceptions of the programme. Some of these problems relate to the structure and scope of the programme as well as the implementation and functioning of the programme. There are also those factors outside the programme's control but which nevertheless impact on it.

9.1 The Existence of a Single Programme

There are essentially two types of cases where witnesses require protection:

Despite the fact that witnesses from the two different types of cases require different levels of protection, the current witness protection programme does not distinguish between the two. As a result a good deal of resources are unnecessarily spent protecting witnesses who may require a less sophisticated form of protection, and not enough is spent on those requiring a more sophisticated programme. Speaking at a recent Centre for the Study of Violence seminar, a member of the National Public Prosecutor's office said, "The current witness protection programme is okay for approximately 80% of the cases but it can't manage the exception."19 The twenty percent he referred to were cases where organised crime was involved.

A lesson that can be learnt from the American system is that of the existence of dual programmes. The Marshall programme offers a sophisticated witness protection programme used when dealing with organised crime, whilst the local programme is run by local authorities and is used to protect witnesses from local criminals.

9.2 Administration and Operation

While the administration of the programme falls under the Department of Justice, the operational side of the programme is handled by police officers appointed by the provincial police management. The budget for these police officers comes from the provincial police budget. This has led to a number of problems within the programme:

Table 9.1: Number of Witnesses and Police officers allocated to the Programme per Province in April 200020
Province No. of Police No. of Witnesses Ratio
Northern Province 4 16 1:4
Gauteng 9 50 1:5,5
North West 7 86 1:12
Western Cape 9 90 1:10
Eastern Cape 14 100 1:7
Kwa-Zulu-Natal 8 208 1:26

The implementation of the Act will take this discretion away from the Provincial Commissioners, and it will be essential that the ratio of police officers to witnesses be addressed as a matter of urgency.

9.3 Resources

One of the major problems experienced is a lack of resources available to the programme. A contributing factor to this lack of resources is the fact that the programme is having to rent in excess of 160 properties each month at a cost of approximately R500 000 a month. A number of government departments have properties that could be used for this purpose but it is difficult to trace such properties.22

As the programme develops and expands so will the resource requirements. It is clearly going to be necessary to look closely at the best use of the limited resources available. A comparison of the money spent by the TRC of approximately R3 million to protect 600 witnesses as opposed to the amount spent during 1997/98 by the current programme of approximately R16 million to protect 700 people could offer some important lessons on ways of creatively using existing resources and thereby reducing costs. In particular, army bases, rather than hotels, could be used to temporarily house people who join the programme. When a person enters the programme while they are being evaluated, and a decision is taken on whether they will be placed on the programme permanently, they are often accommodated in hotels or bed-and-breakfast facilities. It would be a lot cheaper and safer if such witnesses could be accommodated on army bases anonymously during this period.

Subsection (4) of the new Act places a duty on other State departments to render reasonable assistance to the programme, and this would assist with providing much needed resources to the programme. In particular, it was recently revealed that the Public Works Department has approximately 20 000 properties they did not even know existed. Some of these are farms are being rented out.23 It is likely that some of these properties could to be used to house witnesses.

9.4 Role of the Director

Under the new legislation, the Director's role is changed from being purely an administrator of the programme to taking over the function previously performed by the Attorney-Generals, namely that of determining who should enter the programme and of evaluating their continued stay on the programme. There are a number of problems relating to this change in role:

9.5 Current Staffing of the Programme

During the four years the programme has been in operation, some serious problems have arisen with regard to the way staff in the programme have conducted themselves. The problems can be broken down into three different categories:

In a separate incident another witness alleges he was intimidated, threatened with death and sworn at by one of the police officers working on the programme. The incident occurred when the police officer visited the flat where the witnesses were being accommodated. According to this witness, the police officer, after swearing and insulting the witnesses, pulled out a gun and threatened the witnesses.25

There are serious problems with this line of argument: firstly, in this incident the Director of Public Prosecutions felt there was sufficient evidence to take the case to trial and, secondly, the brushing aside of what are extremely serious allegations indicates a lack of sensitivity towards witnesses. It is necessary that all allegation are taken seriously and that an established process is available to deal with such allegations as they arise. Where allegations are found to have substance they must be acted on immediately and where they are unfounded sanctions need to be applied to the accuser. This will go a long way in preventing witnesses from making vexatious allegations while at the same time holding programme staff accountable for their behaviour.

Prior to the alleged rape, a circular had been circulated to staff of the programme from the Justice Department saying that male police officers were not to be alone with female witnesses and that a female police officer should be present at all time. However according to the Director of the programme in KwaZulu-Natal, the police staffing the programme are under-resourced and therefore it is impossible to implement the instructions set out in the circular.

Further abuse of power in the form of intimidation has also been recorded. One witness claims that one of his protectors intimidated him while transporting him back from a court case. In an affidavit the witness stated, "The police officer (referring to one of his protectors) told my brother and I that we were talking too much, and that one day we would open our mouths but not close them. I took this to mean that he was threatening us for acting as witnesses."27

This incident occurred a few days before the witness was shot outside his hotel room. According to the affidavit a few weeks later, the same police officer issued the same threat to the witness again only this time he added, "After I left witness protection he would finish the story". The witness claims that shortly after this he was visited by two people from the programme in Pretoria who removed him from the programme despite the fact that he was still to testify in court.28

At the time of writing the Independent Complaints Directorate (ICD) in KwaZulu-Natal is currently investigating 11 cases of complaints made against the staff of the programme in KwaZulu-Natal. Most of these complaints relate to abuse of power by officers, sexual assault, rape and misuse of resources.

Any allegation made by a witness, or involving any criminal action by is taken seriously and reported to the SAPS for investigation. Allegations have been investigated against SAPS members working on the programme and independent investigations conducted and found to be without foundation. During investigations members have been suspended.

9.6 Screening and Evaluation Processes

There is currently no formal screening process to assess witnesses who apply to enter the programme. According to the head of the Scorpions Unit in KwaZulu-Natal, this has resulted in people entering the witness programme by lying or exaggerating facts. In some cases the intention of such people has been merely to access facilities offered by the programme and to escape obligations they may face. However, of greater concern is that some of these people have entered the programme to gain information and access to other witnesses on the programme.

A number of other countries have introduced an evaluation process to assess witnesses. In both Australia and the United States witnesses are screened before being allowed complete access to the programme.

Presently, there is an informal process of screening witnesses. Witnesses are initially only taken onto the programme on a temporary basis while their situations are investigated to determine whether or not they qualify for protection afforded by the programme. During these investigations a general profile of the witnesses is developed with the investigating officer.

According to the director of the Witness Protection Programme however, the need for improved screening of witnesses has been identified as being extremely important. As a result the programme is in the process of developing a pilot project in this regards which will be launched in the Eastern Cape. This project will entail that witnesses entering the programme, will initially be taken to an undisclosed venue where they will spend 14 days being carefully screened. During this time they will be evaluated medically and psychologically and potentially required to undergo a lie detector test. According to the Director, at the end of the period the programme should be able to assess whether witnesses are genuine or not. Furthermore the programme will gain a greater understanding of each witness's particular situation and personal traits, thereby being able to respond better to their individual needs. This should also include taking language and cultural experience into account in relocating witnesses. If the pilot project is successful, the programme will develop an admission centre to which all witnesses will be sent for two weeks.

The main aim of protecting witnesses however is in order to ensure a successful conviction in a court of law. Therefore, whilst a general assessment and evaluation may be useful for the effective running of the programme, any polygraphing and assessment of the honesty of the witness should occur prior to entry onto the programme.

9.7 Physical Security

With any witness protection programme the security of witnesses is always a major concern. According to public reports by the Director of the programme, Mr. Piet Kleynhans, the programme has had a 100% safety record.30

However, there have been certain incidents that have compromised the safety of witnesses. A witness based in KwaZulu-Natal complained that he had been approached by police officers at his safe house. These police officers, who were not part of the protection programme, allegedly offered him money to leave the programme.

In 1997, a witness was shot and injured while being accommodated at a hotel in Durban. The staff of the programme at the time claimed the shooting had occurred as a result of a drunken brawl. However, both the witness and the Network of Independent Monitors (NIM) dispute this version and claim that an unknown assailant in the corridor outside the witness's room perpetrated the shooting.31

The Director of the programme has made it clear that there are simply not sufficient resources to ensure that all witnesses are guarded on a 24-hour basis. "Just to cover the 160 rented properties you would need at least 480 police officers, and this figure does not take into account additional personnel required to cover rest days."32

While it may not be possible to guard all witnesses on a 24-hour basis, there is clearly a need to tighten security surrounding witnesses. There appear to be two areas where lax security could endanger witnesses:

9.8 Police Investigations

A witness protection programme alone cannot address the problems within the criminal justice system nor can it serve as a prop to disguise poor policing. The programme can only work if it is supported with effective policing.

According to one police officer there is a tendency for the police to use the programme incorrectly by using the programme to recruit informants. These informants are then told they will be accommodated on the Witness Protection Programme if they agree to become an informant.

Furthermore, in other situations the insensitivity on the part of police officers has exposed their informants to danger and has led to the need for these informants to seek protection on the Witness Protection Programme.33

The SAPS' approach to witnesses often opens witnesses to danger or causes witnesses to take fright and disappear. In one incident, violence monitors agreed to facilitate contact between the police and a witness in the Eastern Cape. However, the SAPS' insensitive manner of approaching the witness not only exposed the witness but also caused the witness to flee and refuse to co-operate. This situation resulted because the police had refused to allow the violence monitor to go to the village to collect the witness and had insisted on accompanying the monitor. When it had become apparent that the witness wasn't at home, the police had begun asking the neighbours questions about the witness.34

A number of ways of ensuring that a witness's identity is not exposed until it is absolutely necessary could be used. Some of these methods could include:

9.9 Expectations of Witnesses

When a witness enters the programme, a memorandum of understanding is signed between the witness and the programme administrators. This memorandum outlines what has been agreed between the two parties regarding protection.

The expectations of witnesses vary depending on the witness involved. It would appear that some witnesses have exaggerated expectations of the programme. However, there also appear to be very real problems, which indicate that the general expectations of witnesses are not being effectively addressed.

This is supported by feedback from all the witnesses interviewed that the programme has not lived up to their expectations and that promises made have not been kept. Some witnesses have gone public about their complaints about the programme and have stated that they feel betrayed and misled by the programme. Such publicity does little to instill public confidence in the programme and to ensure co-operation of future witnesses.

One of the witnesses interviewed said:

the programme is flawed in many ways starting with the promises made prior to protection being given. Witnesses under protection end up feeling like prisoners instead of witnesses. Even if they had to complain at times, the attitude of their handlers was such that we were left wondering whether we had done the right thing.
My experiences of the progamme were so traumatic that only commitment and loyalty to the country was able to suppress a strong urge to speak to the press about my experiences.35

The witness said that from initial discussions with the programme he had believed that he and his family would be kept secretly and that they would be adequately provided for by the programme. Certainly, he did not expect to face additional financial burdens as a result of entering the project. According to this witness, people outside the programme who were friends of staff of the programme, got to know where he was staying. Furthermore, the money allocated for the witnesses was insufficient and was misused by the protectors themselves. When he left the programme he was forced to pay the costs incurred in storing his furniture whilst on the programme.36

It is difficult to say categorically who is right or wrong with regard to the allegations made but it is clear from interviews with witnesses and from newspaper reports that there is evidence that dissatisfaction is growing among witnesses towards the programme. While some of this dissatisfaction could be influenced by the stress witnesses are forced to live under, it would also appear that the dissatisfaction is significant enough to warrant closer inspection and that the introduction of measures is needed to address this dissatisfaction. The agreements reached between the witness and the programme may need to be more specific and perhaps more attention should be paid to ensuring the witness understands the content of the memorandum of understanding.

While it is clear that the programme cannot in any way be seen to be buying witnesses' testimonies, it is also clear that there is a lot of controversy surrounding financial matters. There is an obvious need for a more formal and structured approach to determining the remuneration of witnesses, which will protect both witnesses and the programme. Many witnesses are not briefed on the criteria used in deciding the allowances, and when some witnesses receive more money than others do, it becomes a source of conflict.

There have also been allegations that police officers on the programme have been pocketing money meant for witnesses. The Independent Complaints Directorate in KwaZulu-Natal say they have received allegations in this regard but they have been unable to substantiate these allegations as yet.37

Each time a witness resorts to using the media to raise their complaints it undermines the programme. It is essential that a mechanism be put in place to address complaints. This mechanism needs to have the confidence of both witnesses and the programme. The Public Protector's office may well be able to fulfill this role.

9.10 Lack of Support Structures and Services

From the information available, there are indications that witnesses placed on witness protection programmes suffer from serious stress and that this increases the longer they remain on the programme (Newham, 1995, p.6). Added to this, uncertainty about what the future holds can lead to serious emotional conditions developing in witnesses. This stress and tension can play itself out in a variety of ways including alcoholism, drug dependency, the development of problems within the family and the witnesses becoming anti-social and unco-operative. A witness who in addition is also a victim is likely to suffer additional stress. A serious shortcoming of the programme is its lack of support services to deal with these problems.

Aligned to these problems is that of boredom suffered by the witnesses in the programme. Often witnesses are removed from their community and environment where they have support structures in place and are put into a totally foreign environment with no support. According to one senior police officer interviewed, "Witnesses are told to sit at home and are not given any assistance or skills training. Many get bored and start going out wandering the streets."38 Other investigators, members of the Directorate of Public Prosecutions and NGOs, support this contention. At times, witnesses ignore the dangers they face in a quest to overcome boredom and retain bonds with their families and they sometimes return home. In 1997, in the Western Cape, three witnesses who were central to cases of taxi violence left the programme to return home. The witnesses were followed and killed at their homes. As a result of their deaths all charges against the accused were dropped and the cases collapsed.39

This situation of removing witnesses from a familiar environment is exacerbated when witnesses from a rural environment, in which they have lived all their lives, are relocated to an urban environment to which they do not relate at all. There are no support systems to assist in translocations of this nature.

Lack of supervision of witnesses is another problem the programme faces. According to one witness it can be six weeks between visits from members of the programme. One investigating officer complained that he had even been contacted by a manager of one of the venues where witnesses were staying to complain about witnesses. The manager had contacted the investigator because the programme staff were not available to be contacted.40

This lack of support applies particularly to young people placed on the programme. In one incident, youth that were witnesses were placed in a safe house without any adult supervision. After some time tensions began to develop and the youth began to victimise each other and to indulge in alcohol. When the situation reached crisis proportions and an emergency security situation developed, staff of the protection programme were unreachable. The witnesses were under the age of 15 years at the time they were on the programme. According to a report by the Network of Independent Monitors, some of the youths misused their allowances, spending them on alcohol rather than food and transport. According to the report:

Inadequate supervision also meant that unavoidable tensions escalated into conflict that made their cohabitation impossible and threatened their security. At one stage it appeared that one of the witnesses, at his wit's end due to victimisation by the other youth, threatened to call the accused they were testifying against and request the assistance of the accused.41

NIM had to step in and remove this witness when the other youth threatened to kill him. NIM then had to seek accommodation for the witness for two days because it was a weekend and none of the witnesses' protection staff in KwaZulu-Natal could be contacted.42 In subsequent discussions, the Director of the programme expressed concern because it was normal procedure for the Director to be informed immediately when a minor is admitted to the programme and in this case he was not aware of the situation in KwaZulu-Natal.

One police officer said it was important to involve the Department of Welfare in the Programme to offer support to witnesses and ensure professional social workers or psychologists debrief witnesses. The police officer said that one witness on the programme had been responsible for the killing of 37 people and needed the assistance of a social worker or psychologist.43

9.11 Inflexibility in Dealing with Money Allocated to Witnesses

The greatest source of dissatisfaction among witnesses is money and financial reimbursement. The Witness Protection Programme has a policy of not rewarding witnesses for testifying. The Witness Protection Programme has a clear and structured financial policy in terms of reimbursing witnesses but one of the problems with this policy is that it lacks flexibility.

One example of this is with regard to people who own houses in tribal areas or on communal land. Such witnesses cannot resell their homes and, therefore, cannot be compensated by the programme even if they are never able to return to this home. One example of this applies to a family from Richmond in KwaZulu-Natal who owns a home in one of the tribal areas surrounding Richmond. The family is under witness protection because one of the family members is to testify against political hit squads operating in the area. The family will never be able to return to their home but neither can they sell their house because it is on tribal land. The family will not be reimbursed for the loss of their home.

In addition, if a witness sells a home in South Africa and is forced to relocate to a country outside South Africa, they will only receive what the South African monetary value of their house is, even if the cost of purchasing a similar house outside South Africa is substantially more.

Similarly a person may have been unemployed but able to survive because family members the person lived with were employed. Once taken away from this family support the person will not survive, a factor that the Witness Protection Programme does not take into account.

While one recognises the need for the Witness Protection Programme not to be seen to be rewarding witnesses and buying their testimony, there is also a need for witnesses not to find themselves in a worse position than they were in because they agreed to testify. This situation is unlikely to instill public confidence in the programme. The issue of not buying testimony needs to be balanced against the need not to prejudice witnesses prepared to testify. This can be done by using the existing approach but also by introducing a level of flexibility within this approach.

9.12 Lack of an Effective Internal Complaints Mechanism

Within the Act there are no prescribed procedures for complaints. Establishing an effective complaints mechanism for witnesses is crucial to the effective functioning of the programme. The lack of an effective complaints mechanism for witnesses has resulted in many of these witnesses going to the media with their problems and complaints. Not only is this an unsatisfactory means of addressing problems but it also has a detrimental effect on the credibility of the programme. This in turn jeopardises the potential for future witnesses to co-operate with the justice system and to see the programme as viable. According to the Director of the programme, witnesses can approach his office with complaints about the programme but this does not allow for an independent assessment of the problem.44 The new act also enables a witness to take any decision made by the programme on review to the Minister. However, what is really required is the establishment of an effective complaints channel that will have the ability to investigate complaints.

The Independent Complaints Directorate is currently investigating a number of complaints against the programme in KwaZulu-Natal. However, the ICD have complained that when they want information, staff of the programme refer them to the Justice Department who in turn refers them back to staff of the Witness Protection Programme. In addition, during an interview with a senior member of the ICD staff in Durban, concern was raised about the ICD's capacity to thoroughly conduct investigations into the programme.45

9.13 Accommodation of Witnesses

The Witness Protection briefing document issued by the Department of Justice states that witnesses should be kept away from other witnesses. However, in a number of incidents this is not the case. One police investigator said the practice of accommodating witnesses in one place or in close proximity to each other is a serious problem. He cited one example where a witness testifying in the Richmond investigation who was linked to armed robberies and had contacts with armed robbery syndicates was kept in close proximity to a witness testifying against armed robbers. Such a situation has the potential of seriously compromising the witness testifying against the armed robbers.46

9.14 Cultural Diversity

The witnesses on the programme come from a range of different cultural backgrounds. There has been criticism that the programme does not always take this factor into account.

One Zulu speaking youth who was being protected under the programme complained that he had been sent to an Afrikaans medium school despite the fact that the youth could not speak Afrikaans. In other situations there have been complaints that the programme has disregarded the needs of the extended family. According to the Director, the decision about what school a witness will attend will normally be made jointly by the provincial co-ordinator and the parents of the witness and, where possible, the parents' preference is taken into account. However, this appears not to have happened on this occasion and the witness faced serious problems as a result.

It is crucially important that the programme is flexible enough to take cultural diversities into account, and that it is flexible and sensitive in its handling of these diversities.

9.15 Rehabilitation of Criminals

A significant number of witnesses who are part of the programme have themselves been involved in criminal operations. Many of them will not be charged for the activities they were involved in because they have agreed to co-operate with the State. The protection programme has no rehabilitation process, and as a result these people will be free to go back to such activities once they leave the programme. Experience in other countries indicates that such witnesses, unless they become part of special rehabilitation programmes, will once again resort to criminal behaviour at some point in the future.47

According to the Director of the programme, if a witness on the programme commits a crime he or she will be immediately jailed. At present about eight witnesses from the programme are in prison following their involvement in criminal activities while on the programme. It is difficult to get information on people who may have involved themselves in crime once they were discharged from the programme. However, in KwaZulu-Natal, there is an individual currently on the programme who is a key state witness in the murder of controversial United Democratic Movement leader, Sifiso Nkabinde.48 This is not this persons first time on the programme. In 1998, the witness was on the programme when he testified against fellow criminals in a case involving a bank robbery.

There is a need to look at a rehabilitation programme for people on the programme so that they have alternatives to crime on leaving the programme.

9.16 Abuse by Defence Counsel

Another major problem faced by the Witness Protection Programme is the length of time it takes for the case to be finalised. The longer the witness is forced to remain on the programme, the higher the stress and the more likely the witness is to break down and return to his or her home. According to one prosecutor, once some defence attorneys are aware that a key witness or witnesses are part of the Witness Protection Programme, they attempt to delay the case and draw it out, in the hope the witness will break down. There needs to be a mechanism at court level to deal with this problem.

9.17 General Societal Attitudes

One of the problems that affects not only witnesses but anyone who provides information to authorities, including informants and sources, is the culture of our society which disregards anyone who gives information and sees such people as "breaking ranks". As a result, such people are considered worthless even if their information is important. Because of this attitude, the source or witness is seen as a tool to be used and then thrown away. There is a belief that the source or witness is not deserving of respect. The stigma attached to such people is one of being a 'sell-out' or "Impimpi" (as known in local vernacular).

This attitude to witnesses even exists within the criminal justice system. At the trial of Sifiso Nkabinde,49 the presiding judge forced a witness, who was part of the programme, to reveal to the open court the venue where he was staying. While the defence had on numerous occasions referred to witnesses being rewarded through the programme, the judge's order for the witness to reveal sensitive information regarding the safe house where he was being housed indicates complete disregard for the safety of the witness. This also sent a clear message to other witnesses that they should not necessarily expect judges to be sensitive to or respect their circumstances. If the judge felt that it was necessary to establish the circumstances surrounding the witness's accommodation, it should have been done in a manner which did not place the witness in jeopardy.

Fortunately, section 19 of the New Act seeks to prevent such incidents and stipulates that no protected person is obliged to give any information about their circumstances in a court of law. Witnesses can now refuse to answer questions relating to their whereabouts with impunity. However, attitudes need to be changed generally about witnesses. The programme could assist in educating the public about the plight of witnesses and the valuable role they play in promoting criminal justice in South Africa. Information about convictions as a result of protected witnesses testimony should be made publicly available in the programmes yearly report.

10. Relocation and Resettlement

One of the biggest problems identified by witnesses, investigators, prosecutors and NGOs, is the programmes lack of effective long-term relocation processes. In most cases, once the trial is completed, the witness is left to his or her own devices. The staff of the Witness Protection Programme clearly feel resettlement is something the programme is not required to address. Captain Mhlongo, head of the programme in KwaZulu-Natal claims, "In most cases once witnesses have testified the accused never go after them".50 Unfortunately, there are a number of cases where witnesses have been killed after giving their testimony, which appears to dispute the view of the Captain. The death of two young witnesses in Donnybrook, KwaZulu-Natal in 1999, is a clear example of this. The two youths were killed when they alighted from the bus they were travelling on, shortly after having given evidence in a murder trial.

The aspect of relocation and resettlement is not a problem unique to South Africa. A report on vulnerable witnesses written for the British Home office in June 1998 stated:

Often the criminal justice system's role is seen to have ended after the trial. However the potential for intimidation does not end there. Even if the offender is convicted and imprisoned, there may be a risk of retaliation by friends and family or by the offender upon release. This stage however has been neglected by Literature.51

In South Africa a system has been put in place to allow witnesses some time following the conclusion of the trial to prepare for leaving the programme. Although in the new Act there is no specific time period within which witnesses are expected to leave the programme, witnesses are usually given a six-week period within which to re-establish themselves. This time period may be increased following a risk assessment. A formal system is in place whereby both the witness and the Director of Public Prosecutions are given written letters informing them of the date by which the witness is expected to vacate the programme. Section 14 of the Act allows witnesses on the programme to request the Justice Minister to review any such decision taken by the Director of the programme.

However, many witnesses still feel insecure about leaving the programme. In some cases the witnesses feel unable to return to their homes because of fear of attacks from the person or associate they had supplied information about or testified against. Such feelings of insecurity have lead to complaints being made to the media about the witness protection programme.

In a case involvin