Monitoring Police Violence and Torture in South Africa

by

Piers Pigou

Paper presented at the International Seminar on Indicators and Diagnosis on Human Rights: The Case of Torture in Mexico, convened by the Mexican National Commission for Human Rights, April 2002.

Piers Pigou is a former Researcher at the Centre for the Study of Violence and Reconciliation.

Introduction

Quantitative monitoring of police violence and torture in South Africa in both apartheid and post-apartheid eras has been, and to a large extent remains, an under utilised area of research, and as such has not been used to inform or advocate for required changes in South African policing. Whilst much has been written on human rights violations committed by the police in South Africa, including aspects that have utilised some quantitative data, especially in relation to political detainees, the bulk of these violations perpetrated by the South African Police have not been subject to effective monitoring.

Available Data/Access to Information

Apartheid era

As in many repressive states, South Africa during the apartheid era was subject to widespread censorship and counter-propaganda initiatives around the issue of repression and human rights violations. In this context, the government attempted to retain some semblance of legal legitimacy by publishing details of political detainees arrested under specific clauses of emergency regulations and security legislation. Of course, many others – exactly how many is unclear – were also arrested and detained during this time, and consequently a detailed picture has not emerged.

In general, the government and policing/security agencies did not release reports on shooting incidents, or responses to allegations of torture involving members of the police. When forced to do so, illegal behaviour was routinely denied, and in the few cases where irrefutable evidence was established, such incidents were categorised as isolated – the work of (infamous) 'rotten apples'. This issue did not present any serious dilemmas for government, South Africa's white electorate remained largely ignorant or indifferent, and consequently, no proactive attempts were made by the government to ascertain the nature or extent of the phenomenon.

The extent to which relevant data may have actually been collected, collated and analysed by the State is not public knowledge. No attempt has yet been made to establish what, if any systems were in place (or efforts made) to record and monitor such allegations. It is assumed, however, to have been very limited, at best. Acts of torture, assault and extra-judicial execution were officially outlawed, and this provided an adequate legal veneer for a culture of denial to permeate.

It is now firmly established, however, that such abuses were in fact widespread. The Truth and Reconciliation Commission (TRC) received over 21,000 submissions relating to violations, thousands of which related to security force violations. In addition, over 300 former members of the security forces provided the Commission's Amnesty Committee with graphic detail of their involvement in assassinations, torture and other abuses.

In context of apartheid policing and its heavy emphasis on counter-insurgency measures, there was no political will to put an end to these practices. Indeed, available information from the Truth & Reconciliation Commission (TRC) shows that they were actively encouraged, even at the highest level of government. (Of course, this is still denied by the bulk of apartheid-era politicians, but a combination of disclosures made during various TRC hearings, coupled with documentation from political security structures that endorsed the 'elimination' of political resistance, clearly shows that the political leadership was, at best, aware of the various methodologies employed by the security forces.)

As in other countries under the yoke of repressive regimes, in South Africa there was a heavy reliance on non-governmental organisations - NGOs - (both domestic and international) to document, collate and report on incidents of violence and abuse by the security forces. These organisations were dependent on a range of sources, including the media,¹ legal fraternity, medical practitioners and other organs of civil society, as well as their own direct contact with victims themselves. Several domestic NGOs focused exclusively in this field, such as the Detainee Parents Support Committee (DPSC) in the 1980s and Peace Action, the Network of Independent Monitors in the early 1990s. Although they played an important role ion bringing a range of violations to the public's attention, they remained largely urban based with limited access to many rural and peri-urban areas. As such, our insight into abuses in many parts of the country (both in terms of qualitative, as well as quantitative data) remains very sketchy.

The DPSC, based in Johannesburg, had a membership that was spread across the country, and kept track of hundreds, sometimes thousands of political detainees during the 1980s. The Human Rights Committee of South Africa² estimates (conservatively) that there were 78,000 politically related detentions between 1960 and 1990. Between 1986 and 1987 alone it was estimated that 25,000 were detained.³ Many detainees were interviewed following their release and a record of those subjected to torture and other abuses was compiled by several NGOs and individuals. Although this was not a comprehensive record and no detailed quantitative analysis was attempted, it broadly established how reports of abuse increased in accordance with increases in the number of people detained, and that torture was "widespread and systematic".⁴ This was subsequently confirmed by the number of torture allegations received by the TRC for the five-period 1985-1989 (the period of most detentions), which was than from pervious 25 years combined, supported these findings.⁵

The early 1990s

During the negotiation period in the early 1990s, certain parts of the country were plagued by widespread violence and killing. In fact, many more people were killed and injured during this four-year period than the previous 20 years of conflict and repression. In 1992, the Community Agency for Social Enquiry, an NGO, also based in Johannesburg, compiled a database of alleged perpetrators and victims as reported in the print media over an 18-month period. Despite the limitations of relying on media reports that were often inconsistent and contradictory, as well as lacking in verifiable detail, the statistics generated from this data showed that in cases where the identity of attacker could be established, the Inkatha Freedom Party (IFP) and the South African Police (SAP) were the primary agents responsible for acts of violence.⁶ Although these findings were vehemently denied, it established that there was a clear pattern of security force collusion with the key protagonist, the IFP, as well as the culpability of African National Congress (ANC) supporters in the violence, albeit at considerably lower levels.

Violence monitoring was also hampered by allegations of political bias. Whilst the use of violence appeared to be integral to certain parties negotiation strategies, these statistics, along with other generally qualitative data generated by other NGOs working in the field, was used somewhat selectively for political capital, primarily by the ANC. This led some analysts and NGOs to accuse several violence monitoring NGOs of spreading dis-information.⁷

Despite the importance of such information, attempts to expand and improve NGO monitoring capacity remained insufficient, and activities consequently remained ad hoc. In terms of quantitative monitoring, the best that could be achieved were random compilations of figures drawn from a range of field and media resources. With regards to police abuses, monitoring organisations relied on descriptions of incidents that they argued represented typologies of abuse, through acts of omission as well as commission. Despite endemic levels of violence in some quarters, civil society involvement in such initiatives was sparse, and funding to develop capacity wanting. In this context (and given their own complicity in the violence) the authorities, including the police, were unwilling to work with, and share information with NGOs trying to identify patterns and trends of abuse, as well as those responsible for the violence.

During the negotiation era, the National Peace Secretariat, with representation from all the main political bodies, as well as security force representation, was the primary agency tasked with brokering and maintaining peace through an infrastructure of provincial and local dispute resolution committees. Little or no attempt was made to access and/or utilise available data on abuses to hold those responsible in check. Although the Secretariat played an invaluable role in stemming and preventing violence in many parts of the country, with only one or two exceptions this was done largely at the expense of any sort of forensic accountability.

Under the auspices of the CSVR's Human Rights Documentation Programme, the data generated by several (mainly Johannesburg-based) human rights and violence monitoring NGOs was subsequently collated using the HURIDOCS system into a larger database with [with over 5000 individual cases], which was subsequently made available to the TRC. Regrettably, and without explanation, the Commission made no use of this material.

Of course the TRC itself has also developed its own important database of violations. Although this has left us with an impressive archive and provided an unprecedented insight into a range of violations committed by members on all sides of the conflict (including the police), in terms of quantitative data regarding police abuses, the picture is limited on a number of fronts. First and foremost, access to the Commission's database is prohibited, and it remains to be seen what admission will be given to these and other archival material generated through the process. On a more substantive level, it is unlikely the quantitative data will provide us with an accurate overview of levels of abuse during this period. Although many people reported abuses at the hands of the police, it is suspected that many more did not. Many activists who were subjected to police brutality did not engage with the TRC, for one reason or another. In addition, and despite the important disclosure regarding police abuses made during the amnesty process, less than 5% of the 7,100 applications received for amnesty came from police officers.⁸ Most of those who came forward were from certain sections of the SAP's internal security wing (aka. The Security Police). Most appeared to have applied for amnesty as a result of revelations and progress made during post-1994 criminal investigations and prosecutions.

On the whole, available data on policing abuses is geographically biased according certain areas that received better coverage from and access to the Commission. In addition, certain time periods were not adequately represented. This includes, for example, for the period around the Soweto uprisings in 1976, or the Vaal Uprisings, eight years later, both of which experienced widespread police repression and violence.

As we can see, quantitative data on abuses by the police for this period has its limitations. Much of the available information is qualitative and also almost exclusively focused on politically related incidents. While the NGO and TRC databases are both likely to have to a certain extent helped to rectify this situation, although access to their database currently remains restricted.

Our picture of policing abuses during this period is also severely curtailed by the (almost) complete absence of data regarding the treatment of criminal suspects, including those who died in police custody or as a result of police action. Nor are there available statistics on allegations of torture and assault made by those picked up and/or accused as criminal suspects against police members. Despite this, in the light of subsequent research and interventions it is believed (albeit retrospectively) that torture was also routinely practiced by other policing units, especially certain detective and specialist units, and was widely employed against criminal suspects. Before the early 1990s, however, there were no detailed studies on this phenomenon and gauging levels (as well as locations) of abuse remains speculative. Given the widespread practice of such violations in terms of political policing, and the apparent effectiveness of using these methods to extract information, it is reasonable to suspect that the torture of criminal suspects for similar purposes was also systemic. This investigative methodology was also encouraged by a legal framework that put the responsibility of proving a confession was extracted illegally on the accused.⁹

In 1991, research conducted at the University of Cape Town based on a series of interviews with prisoners concluded that ordinary criminal suspects have also been victims of the same and similar treatment that was meted out to many political detainees.¹⁰ These findings were supported by findings in a report compiled by three non-governmental organisations in March and April 1995. This report documented 380 cases of alleged torture, extra-judicial executions, deaths in custody that had been recorded by human rights organisations in the provinces of KwaZulu Natal, Western Cape and Gauteng between 1990 and 1995.¹¹

The Gauteng component of this research, which made up the bulk of the allegations, was compiled by the Independent Board of Inquiry (IBI), which had developed a close working relationship with the appointed civilian responsible for police oversight in the province.¹² Although this official was prohibited from proactively investigating allegations of torture under his jurisdiction, he was able to pass information to the IBI, which subsequently conducted its own preliminary inquiries, and subsequently lodged official complaints with his office. In this way, during 1994 the IBI was able to submit a dossier of over 150 allegations of primarily torture, which subsequently became the subject of an unprecedented investigation launched by the first post-apartheid provincial (Gauteng) police minister, involving both international police officers and NGO participation.

Post-apartheid era

With the new political dispensation there has been a clear development of policy towards ensuring that policing in South Africa is conducted in a manner consistent with human rights and democratic values. This process has been multi-faceted in nature and has been underpinned by the adoption of a Bill of Rights, the establishment of a South African Police Service (SAPS) in 1995 and the implementation of a human rights training curriculum in basic police training. A civilian-controlled monitoring and investigative body known as the Independent Complaints Directorate (ICD), tasked with investigating allegations of police abuse, was also established. In addition, and largely in response to ongoing allegations of abuse, the SAPS introduced a "Prevention of Torture" policy by the SAPS in 1998/99.

The ICD remains the central official monitoring and investigative body of alleged police abuses. Both its investigative and monitoring capacity of police abuses, however, remains constrained in several areas and for a number of reasons. The ICD has 10 offices (nine regional and one national) spread across the country based in key urban centers. Access to these offices and is further limited by widespread ignorance about the ICD from the general public. With the exception of incidents of deaths in custody or as a result of police action, the ICD's picture of alleged police abuse is entirely governed by what matters are referred to them, from a range of sources, from members of the general public to the Minister responsible for policing. Proactive efforts to access other relevant information are largely ad hoc, and recent developments have seen such initiatives officially discouraged.¹³ Consequently, the ICD, by its own admission, does not have a comprehensive overview of reported allegations of police abuses.¹⁴

Despite this, since 1994 data purporting to provide an overview of alleged and proven criminal activity involving the police is periodically provided by the government. According to the Institute for Security Studies (and NGO based in Pretoria), between 1994 and 1997 an average of 13,954 complaints or charges per annum were laid against SAPS members. In 1997, 17526 charges were laid, including several thousand complaints of assault. Statistics for 1998 and 1999 were not available. In 2000 over 14600 charges were laid, including several thousand allegations of assault.¹⁵ On average, 1200 officers were convicted of criminal offences every year between 1995 and 1999.¹⁶ Almost a quarter (23%) of all convictions relate to cases of assault with the intention to cause grievous bodily harm (Assault GBH), and common assault, and a further 13% of cases for violations of the Firearm and Ammunition Act and for pointing firearms. During the same period approximately 170 SAPS member have been convicted for murder, and a further 220 for attempted murder – it is not clear how many of these cases involved incidents that occurred during the course and scope of policing duties.¹⁷

It is clear that there is a wide discrepancy between the number of charges laid and the number of convictions secured. This is influenced by a number of factors, such as: abusive practices by the police are typically under-reported by the public (especially in situations where there remains a fundamental mistrust of policing agencies); investigations are generally under-resourced and not given adequate priority; incidents of alleged abuse often take place in contexts where supporting evidence cannot be secured; court cases against police officials do not receive priority attention.

Deaths

The number of deaths in police custody or as a result of police action remains high. Since 1997, when the first statistics were produced, between 550 and 700 people die annually in custody, and as a result of (or in relation to) police action. Post-apartheid legislation governing the police requires that all deaths either in custody, or during the course of police action be referred to the ICD. Although there is some evidence that some cases have not been brought to the ICD's attention, since 1997 (when the Directorate was established), quantitative monitoring of these deaths has been relatively comprehensive.

Although these figures appear very high, this should not be regarded as an accurate indicator of abusive practice. With the exception of one targeted evaluation conducted by the CSVR on behalf of the ICD in 1998, there is no detailed analysis of these cases. The CSVR research referred to involved an analysis of 168 deaths in custody or as a result of police action that were recorded in the Gauteng Province between April and December 1997. This research highlighted some of the difficulties and complexities associated with determining whether or not abusive behaviour was employed, and touched on a range of issues, including; the consistency and quality of information recorded, the competency of specific investigations and the categorisation employed by the ICD.¹⁸

The ICD does provide its own breakdown in terms of cases dealing with deaths. In terms of deaths in custody, these are broken down into five categories; natural causes, suicide, injuries in custody, injuries prior to custody, and possible negligence. While this provides us with some insights into what may have happened and whether or not the police are responsible, the categories employed are not definitive. In terms of negligence, for example, although there is a specific category dealing with this, it is quite possible that negligence (by the police) also contributed to the deaths in other categories). In terms of abusive behaviour, it is also possible the injuries sustained "prior to custody" were in fact sustained at the hands of arresting or investigating officers, or that suicide was in fact induced.

Turning to deaths as a result of police action, the ICD breaks these incidents down into seven categories. These are: shooting during the course of arrest, shooting during the course of a crime, shooting during the course of an investigation, other intentional shootings, possible negligence, negligent handling of a firearm, and 'other'. Once again, this categorization does not provide us with any detailed insight into abusive practices. It appears, therefore, that we cannot place too much reliance on the ICD's own statistics in terms of determining levels of abusive behaviour resulting deaths. Although the bulk of the ICD's investigative resources are focused on this work, anecdotal evidence suggests that investigations are not necessarily thorough, which reflects the limited skills and experience of many ICD investigators.

Shootings

Recent research by the CSVR, on shooting incidents involving the police between 1996 and 1998, utilised several data sources from the SAPS (i.e. data on shooting incidents, data on criminal and internal disciplinary cases, and data on civil claims brought against the police). The report highlighted a number of problems, especially with regards to the breakdown in the SAPS monitoring systems – which consequently had serious implications for assessing levels of abuse.¹⁹

Indeed, as with the research on deaths in custody, this research provided some insight into the difficulties determining what cases might fall within the parameters of abuse. The research concluded that the level of fatalities as a result of police action was not exceptional in relation to the rates recorded in many US cities. It also found that in the context of the high levels of societal violence in South Africa, the police's use of force was not necessarily high, although this does not mean that this force was used "appropriately, effectively or justifiably".

Torture

Allegations of torture against the 'new' SAPS have continued to surface on a regular basis in the post-94 dispensation. The cases themselves have almost exclusively involved criminal suspects. As in the era of political repression, torture is primarily used as an "investigative" method, in other words, a means of extracting information and confessions.

Determining what cases may or may not fall within the definition of torture has a direct bearing on how levels of torture are determined (statistically and otherwise). Despite ratification of the United Nations Convention Against Torture (CAT), South African law itself has no provisions relating to torture, and such cases are dealt with in terms of the Criminal Procedures Act, under the broad category of assault with the intention to cause grievous bodily harm (Assault GBH). Although the SAPS has adopted a Prevention of Torture Policy which includes an even more expansive definition of torture than that contained in the CAT, the SAPS does not record or categorise complaints of torture and relies on the ICD for the provision of statistics in this regard. The ICD, which has recently adopted the SAPS definition of torture does not have, by its own admittance an accurate picture of the torture, and available statistics provide little insight into rates of prevalence. Indeed, the ICD's own reports on torture²⁰ indicate that the monitoring body has thus far utilized a very narrow definition of torture, in which certain methods (i.e. electric shocks, suffocation, suspension) are the only cases regarded as torture. As such, it's own reports to not correspond with either the CAT or SAPS definitions of torture. Consequently, this excludes a large number of other cases that would fall within these definitions.

Unlike cases of police-related deaths, there is no legal obligation on the police to refer cases of torture/assault GBH to the ICD. Consequently the ICD is dependent on what is brought to its attention. Many more cases of assault are reported to the SAPS itself every year. Some of these are likely to have fallen into the category of torture, as defined by the SAPS policy, and /or the CAT. In addition, given the absence of available or legitimate alternative remedial mechanisms it is likely that many others do not lodge complaints.

Available statistics

Statistics from the ICD's Complaint's Registry are contained in the Directorate's annual reports (available on the web – http://www.icd.gov.za). According to their latest report, for the period 2000 / 2001, they received an almost 20% increase in the total number of complaints received (from 4380 to 5225). In the same period, however, the number of serious cases declined by 27,4% - although the report warns that this does not mean the actual number of incidents has declined.²¹

As we have seen, the ICD's responsibility for effective monitoring of police abuses and the production of reliable and meaningful statistics is constrained on a number of fronts. ICD reports on torture and other abuses are in most cases descriptive Possibilities for rectifying this situation are, however, apparent. Official statistics on police issues are compiled by the SAPS' statistics arm, the Crime Information Analysis Centre (CIAC). Several categories of crime are constantly monitored, although the process of improving the data gathering process is an ongoing project – competing for funds amongst a range of other policing (and other) priorities. The existing infrastructure is clearly insufficient, and is compounded by human resource problems, such as insufficient training, inadequate resources and computer support at station and unit level, as well as a high turnover of skilled personnel.²² This situation reflects the legacy of a policing culture still struggling to come to terms with the practicalities of basic information technology.²³ The result is that quantitative monitoring of general crime for the purpose of informing crime management strategies remains very much in its formative stages.

Although the CIAC compiles statistics on several categories of crime, they are not involved in the monitoring or systematic co-ordination of available data on allegations of police abuse. Monitoring of these issues is undertaken by the SAPS department responsible 'discipline management', (with the exception of allegations of corruption that are handled by a separate department – the Anti Corruption Unit. The ACU has developed its own database, as well as its own web site on which its statistics are published and updated regularly.)²⁴

Every year the Minister responsible for the police responds to Parliamentary questions on the number of police personnel charged and convicted each year. These statistics are compiled by the discipline management department, and are drawn together from statistics that should be provided by each of the provinces (and area levels within the provinces) on a monthly basis. The department, however, only records matters that relate to disciplinary action. In theory, all criminal allegations against SAPS members, which are denoted as "serious misconduct", must also be subject to an internal disciplinary inquiry. Consequently statistics on internal discipline should provide a relatively accurate overview of criminal charges that are of a serious nature. The provincial statistics are broken down into categories of crime, but no detail on which stations or units are worst affected is generated. The absence of a proper database, means that available statistics on alleged misconduct are limited to very few variables.

Information is not shared between or effectively utilised by relevant departments. The Discipline Management Department, for example, has little to do with the CIAC, and only deals with the ICD in terms of cases it receives from them (for specific monitoring purposes). In other words, the ICD does not receive the monthly figures generated by each of the provinces in terms of internal disciplinary cases opened against specific members. This is remarkable, given the ICD's general oversight mandate, and means that between all the civilian oversight bodies not one has a detailed overview of allegations made against the police. The head of Discipline Management was keen to point out that the Department's responsibility was not with regards to criminal matters, but for internal disciplinary issues. It did not, for example, monitor the progress of criminal matters, although it was pointed out this would be possible on the basis of available information. Apparently, there are also plans in the pipeline to develop a workable database, which in turn will require improvements and standardization in processes of data gathering and capturing.²⁵

Despite the irregular production of aggregate figures on police criminality, police management does not have a detailed insight into the information compiled, and as such does not use it for the purpose of management and performance appraisals. No apparent plans have been made to rectify this situation, and criminality within the police, or at least reported human rights violations are unlikely to be subject to rigorous monitoring by the police.

The ICD is well positioned to develop a closer working relationship with relevant SAPS components, such as the SAPS own Internal Investigation Units (IIUs), which operate at area and provincial levels and are responsible for the bulk of criminal investigations into allegations against SAPS members.²⁶ Although the IIU liaises with Discipline management, there is no formal correlation of their cases. As far as we are aware, the IIU does not provide statistics based on its own investigations. The ICD could certainly play a role in assisting to improve and standardise their own monitoring and reporting systems. In the absence of a proper regulatory framework, however, the ICD is reliant on police co-operation in sharing information – something that has not always been forthcoming. The ICD is hoping to rectify this situation, and draft legislation before cabinet calls for clarification on these kinds of issues.

Monitoring of police abuses by civil society agencies has all but disappeared in the post-apartheid dispensation. Although some victims of abuse have found legal and other remedies to tackle violaitons and the media continues to report on certain (especially sensational) cases, NGO monitoring of police abuses is virtually non-existent.

This lack of involvement is paralleled by the limited attention given to the subject of policing abuses by the South African Human Rights Commission (SAHRC). The Commission has addressed a number of individual incidents, especially those that contain a distinct racial component. In addition, the Commission has tackled the police (and others) on the treatment of migrants and refugees. Since its inception in 1996, however, it has not provided a focus on the subject of systemic abuse and organizational culture, but has preferred to refer matters on a case-by-case basis to the ICD. No attempt has been made to address the systemic problems regarding police abuses as they relate to the broader transformation goals of building integrity, professionalism and accountability.

Apartheid era/Post-Apartheid Era – differences and continuities

The important developments in South Africa's legal and regulatory framework have not been complimented with an adequate monitoring and oversight capacity of abusive police practices, either from within the SAPS or the external independent monitoring and investigative body, the ICD. While there have been important strides in transparency and accountability in some areas, there remains considerable room for improvement in terms of developing internal SAPS monitoring mechanisms of criminal and unprofessional behaviour, as well as the consolidation of civilian oversight.

Of particular concern is the failure to place criminality and corrupt practices within the police, of which illegal use of force is but one component, higher up the list of priorities in the general campaign to tackle crime, which remains a central concern of nearly all South Africans. Dealing with abusive practices is almost entirely reactive, and generally limited to very public cases. Despite an important shift towards addressing internal discipline as a management function, the internal disciplinary system remains in disarray.²⁷ Consequently, police violence and criminality has not been factored in, in terms of police management's priority objectives, or even as indicators to achieve objectives within the disciplinary arena.

Energies are instead focused on other (important) priorities in relation to other areas of crime fighting. The National Crime Prevention Strategy and annually published policing priorities make little or no mention of the issue of police criminality. The lack of adequate data allows the authorities to continue treating the issue as a low-priority concern which is dealt with through long-term preventative measures, such as the human rights training programme. Shock is expressed and harsh action promised when cases of abuse are brought into the public arena, but no action has yet been taken to tackle police criminality as a priority management function. Although there is no detailed insight into the different circumstances in which abuses arise, there are clearly patterns of abuse emerging from particular investigation units and police stations. Why then is no meaningful action taken?

Since 1994, there has been a general concern that a concerted and rigorous effort to address police abuse would be counter-productive. Consequently a 'hands-off' approach has been adopted, with the intention that such 'problems' would best be dealt with by police management itself. Despite the police's central role in apartheid violations, abusive police practices do not command significant attention. Crime fighting is a national priority, for both the government and the general public, and there is widespread public sympathy for harsh treatment of criminal suspects. A national survey conducted in 1998, for example, found that a third of South Africans supported the police's right to use force to extract information from a criminal suspect. (A further 25% were indifferent on the subject). Physical abuse has been a traditional methodology of policing in terms of its various functions of protection, enforcement and investigation. As such, there is a widespread reliance on police who use or tolerate such practices.

Police criminality and corruption have, however, become serious obstacles to the SAPS' transformation, which in turn undermines the longer-term prospects for tackling crime. Despite this, police violence and torture are not viewed within the broader ambit of criminality, either within the police itself, or more generally. Indeed, the focus in terms of police criminality appears to rest on cases of corruption, although no detailed insights into the relationship between corruption and other offenses have emerged. Nor is violence and torture treated in terms of discipline and accountability as a professional and managerial concern. Instead, an ad hoc largely reactive approach has been retained, denials of systemic problems permeate, and an age-old explanation that abuses are the work of miscreant individuals endures. Whilst allegations of abuse do not necessarily meet with the fierce denials of previous administrations, actions and sentiment of police management remain largely protective.

In terms of police shootings, efforts to bring legislation in line with international standards have been fraught with complications, as well as fierce opposition from within the SAPS itself. Although the current situation remains legally confusing and unclear, it is extremely unlikely that there will be a return to the kind of impunity that existed within the SAP.²⁸ Whatever the precise detail of the finalised legislation, South Africa's constitution demands higher levels of accountability, which will help to curb the number of abusive incidents.

In terms of torture and physical assaults, these practices continue and are dealt with on a case-by-case basis with varying degrees of sanction – but typically not harsh. There are, for example, cases of police officers who have been found guilty of electric shock torture, for example, receiving fines from the courts and returning to work in the same units. A failure to take appropriate action is likely to feed into a culture of impunity that appears to permeate within many parts of the SAPS.

Conclusion

Certainly, a distinction must be drawn between South Africa's democratically elected political leaders and their predecessors, and it is evident that the current leadership has been relatively vocal in terms of calling for the protection of human rights. Although some of these words have been translated into concrete actions, in the context of ongoing policing abuses, pertinent questions remain as to whether enough has been done to finally put an end to police violence and torture in the new South Africa?

South Africa is touted as having one of the most progressive Constitutions in the world with a powerful human rights regime. One of the core criticisms of South Africa's transition has been the gap that exists between the development of policy and its implementation. The SAPS must put in place an effective information regime to assist in the campaign to root out criminality within the police service. Such a regime requires action at a number of levels, including an effective statistical monitoring component, and a willingness to use this information as an effective management tool for improving service and accountability. With no effective lobby within civil society to champion this issue at present, the necessary developments are dependent on the efforts of the under-capacitated ICD and a handful of other individuals. The opportunities for such a development clearly remain – but it remains critical that the appropriate advocacy is undertaken.

At the same time, efforts must also be made to galvanise relevant civil society structures to continue recording and collating qualitative data, relating to alleged violations committed by police members. Such records can be subsequently developed into statistical data. In 1994, one of the most important factors in prompting immediate government intervention to secure an investigation into torture in the Vaal Triangle was the sheer number of detailed allegations placed in front of the provincial policing Minister.

In 1995, as a member of the Police Reporting Officers Board, an interim structure that facilitated the shift from the Police Reporting Officer system²⁹ to the ICD, I suggested that if it was possible for the Minister to provide Parliament with an annual figure of alleged criminality within the SAPS, surely it would be possible to develop a monitoring system that would allow a more detailed overview of what the problems were, which in turn could facilitate targeted interventions – which was of particular importance in the context of limited resources. A senior SAPS representative informed me that this was already "in hand", and that this data would be available to the ICD when it was up and running. Almost seven years later, however, there has still been no action in this regard, and concerns remain that police criminality, including human rights violations, remains widespread. The opportunity to build an effective monitoring system remains, but without a policy shift that places the issue of police criminality within the broader parameters of crime fighting, and the associated political will that can make this a reality, it seems likely that police abuse will remain part of South Africa's policing landscape for the foreseeable future.

Notes:

¹ The media was subjected to draconian restrictions that severely limited their ability to independently monitor, verify and report on much of the violence and abuse. Although a number of reports were written on these issues, the State was able to coerce and co-opt and when necessary simply ban and censor.

² The DPSC was renamed the Human Rights Commission in 1990, and subsequently the Human Rights Committee of South Africa in 1995.

³ A Crime Against Humanity - reports of the Human Rights Committee of South Africa, edited by Max Coleman, 1998, p.48 & p.56

⁴ ibid p.53

⁵ TRC Report, Vol. 2, Ch.3, Para 103

⁶ Community Agency for Social Enquiry, Inkatha/ANC Responsibility for Reef Violence, May 1991 Community Agency for Social Enquiry, Who is murdering the Peace? C.A.S.E. Research Statistics, October 1991

⁷ S.A.I.I.R - South African Institute of Race Relations – Spotlight on disinformation about violence in South Africa – Anthea Jeffrey , Spotlight, No 8/92, October 1992 - Most reports, including those produced by international agencies such as Amnesty International and Africa Watch, did report on abuses committed against the Inkatha Freedom Party, although details remained largely sketchy. The IFP did produce its own report of officials and supporters killed during the 1990s, as evidence of the ANC's objective to eliminate political opposition. This list of several hundred incidents was submitted to the TRC, where efforts to corroborate individual cases were undertaken.

⁸ CSVR's (currently unpublished) research on publicly available information relating to amnesty shows that approximately 310 members of the apartheid security forces applied for amnesty from a total of over 7000 applications.

⁹ This situation has subsequently been changed, and the onus for proving confessions were made freely and voluntarily now lies on the state.

¹⁰ Fernandez, L: Police abuses of non-political criminal suspects: A survey of practices in the Cape Peninsula - Research Report - 1-91, Institute of Criminology, University of Cape Town, 1991.

¹¹ Breaking with the past? Reports of alleged human rights violations by the South African Police: 1990-1995 - Network of Independent Monitors, The Trauma Centre for Victims of Violence and Torture, the Independent Board of Inquiry, May 1995.

¹² The Police Reporting Officers (PRO), as they were known, were civilian monitors operating under the auspices of the 1992 National Peace Accord. Allegations of police abuse were reported to these individuals who were tasked to oversee the investigations of the Complaints Investigation Unit (CIU), specially established during the negotiation era. With the exception of the PRO for the Witwatersrand, and to a lesser degree the PRO for Natal, in most instances the PRO/CIU proved to be largely ineffective, as it was believed by some they were intended to be.

¹³ During 2000 and 2001, the ICD instituted a number of 'spot checks' on police stations about which they may have received complaints. This prompted a strong reaction from the police department and Ministry who effectively sought to bar the ICD from access to police stations, and in particular to holding cells. A memorandum from the SAPS was circulated to police stations effectively informing them that the ICD had no authority to instigate 'fishing expeditions' for complaints in this way. The ICD suspended further station visits, but continues to experience problems from some stations who have interpreted the aforementioned memorandum as authorisation for a blanket ban on ICD access.

¹⁴ Consultation with Shadrack Mahlangu, Head of Complaints Registry, ICD – 19 March 2002

¹⁵ Policing the Police: SAPS Members Charged and Convicted of Crime, Nedbank / Institute of Security Studies Crime Index, Volume 5, 2001, Number 2, March – April. http://www.iss.co.za/PUBS/CRIMEINDEX/01VOL5NO2/policing.html

¹⁶ Statistics provided by the Minister of Safety & Security in June 1998 show that 3767 officers were convicted over a three-year period, for a range of crimes ranging from murder and rape to petty offences. This included 252 convictions for assault (GBH) and a further 556 convictions for assault. – Nearly 4000 police guilty of a range of crimes – Cape Argus, 24 June 1998.

¹⁷ Policing the Police: SAPS Members Charged and Convicted of Crime – op cit.

¹⁸ Bruce, D. (1998). Towards a Strategy for Prevention: The occurrence of deaths in custody or as a result of police action in Gauteng, April - December 1997. This report was produced at the request of The Independent Complaints Directorate, 2 July.

¹⁹ Bruce, D. & O'Malley, G. (2001). In the Line of Duty? Shooting incidents reports and other indicators of the use and abuse of force by members of the SAPS. Research Report written for the Independent Complaints Directorate, October 2001.

²⁰ i.e. Report on torture cases investigated by the ICD, 1 January, 2001 - http://www.icd.gov.za/reports/torture.htm

²¹ http://www.icd.gov.za/reports/2001/areport01.pdf

²² Figuring it out: Getting to know police crime statistics - , Nedbank/Institute of Security Studies Crime Index, Volume 5, 2001, Number 3, May - June. http://www.iss.co.za/PUBS/CRIMEINDEX/01VOL5NO2/policing.html

²³ While South Africa may boast a capacity to conduct DNA forensic investigations in stock theft cases, for example, it still struggles to generate and maintain relatively basic data systems, such as the Case Administration System (CAS).

²⁴ http://www.saps.co.za/16_divstat/anticor/index.html

²⁵ Telephonic consultation, Senior Superintendent A Khan, Discipline Management Department, 28 March 2002

²⁶ Despite this intention, many policing areas in the country are not serviced by the IIU, and investigations against police members are conducted by other members of the SAPS. In some cases this results in investigations being conducted by someone from the same police station.

²⁷ Newham, G. (2000). Transformation and the Internal Disciplinary System of the South African Police Service. Occasional paper written for the Centre for the Study of Violence and Reconciliation, November.

²⁸ For a discussion on the key unresolved legal issues relating to the use of legal force, see Bruce, D. (2002). The Legal Framework on the Use of Lethal Force in Effecting Arrest - a new Section 49? Memorandum produced by the Centre for the Study of Violence and Reconciliation, March.

²⁹ The Police Reporting Officers (PRO), as they were known, were civilian monitors operating under the auspices of the 1992 National Peace Accord. Allegations of police abuse were reported to these individuals who were tasked to oversee the investigations of the Complaints Investigation Unit (CIU), specially established during the negotiation era. With the exception of the PRO for the Witwatersrand, and to a lesser degree the PRO for Natal, in most instances the PRO/CIU proved to be largely ineffective, as it was believed by some they were intended to be.

© Centre for the Study of Violence and Reconciliation