Police Brutality in South Africa
by
David Bruce
From the report Police Brutality in Southern Africa – A Human Rights Perspective by
Mwanajiti, N., Mhlanga, P., Sifuniso, M. Nachali-Kambikambi, Y., Muuba, M and
Mwananyanda, M (eds) Published by: Inter-African Network for Human Rights and
Development (Afronet), 2002.
David Bruce is a Senior Researcher in the Criminal Justice Programme at the Centre for the
Study of Violence and Reconciliation.
In addition to this case study of South Africa the full report Police Brutality in Southern
Africa – A Human Rights Perspective’ contains case studies of Botswana, Malawi,
Muaritius, Mozambique, Namibia, Swaziland, Tanzania, Zambia, Zimbabwe. Queries
regarding copies of the full report may be directed to Afronet at: afronet@zamnet.zm.
Contents
Abbreviations
1. Introduction
2. Defining police brutality
a. Police in South Africa
b. Defining police brutality
3. Problems with the use of statistics and press reports as indicators of the extent
and nature of police brutality
a. Statistics as measures of police brutality
b. Comparing rates between different countries and localities
c. Press reports as a source of information on police brutality
4. Indicators and evidence regarding the extent and nature of police brutality in
South Africa
a. Statistical indicators relating to the use of force and unlawful violent actions by
the police
b. Some reported cases of alleged police brutality
c. Some aspects of police brutality in South Africa
d. The profile of victims and perpetrators in terms of race and other factors
5. Variables impacting on occurrence of police brutality
a. Root causes
b. Legal frameworks regulating policing
c. International Human Rights Standards
d. Case law
e. Existing institutions regulating policing
f. Initiatives intended to address brutality
6. Limitations of current measures
7. Recommendations on the way forward
8. Notes
Abbreviations
ICD Independent Complaints Directorate
ICW Ixopo Community Watch
POP Public Order Police
SAHRC South African Human Rights Commission
SAP South African Police
SAPS South African Police Service
SIRS Shooting Incident Reports
1. Introduction
The issue of police brutality has been in the spotlight recently1 in South Africa particularly
following the screening of a video in November 2000 on the TV programme Special
Assignment depicting members of the North East Rand Dog Unit setting their dogs on,
assaulting and racially abusing three alleged illegal immigrants from Mozambique (the
incident is described in more detail in Case 1 in Section 4 below).
More recently a newspaper report quoted Mary Rayner, an Amnesty International (AI)
Researcher, as saying that “there are at least 20 to 30 severe cases of torture a year that are
reported” in South Africa “but ther are many more incidents than that that are not reported”.
As recently as September, she said, we received a report that the Brixton Murder and
Robbery Squad – which has been involved in systematic torture for years – continues to act
with impunity.2
As this report demonstrates however in section 4 the problem of police brutality is not
restricted to a problem of torture but also includes other problems of a serious nature
including apparent execution type killings in custody and at the point of arrest, deaths
apparently linked to torture or other assaults in custody and outside of custody, other deaths
related to apparently unjustified, reckless or accidental shootings including killings during
demonstrations, killings relating to arguments and domestic disputes, off duty killings of
criminal suspects involved in petty offences, other deaths linked to assaults of persons not
in custody and other non-fatal assaults of persons at the point of arrest or otherwise outside
of custody, cases of rape and other assaults on persons in custody and even on persons
reporting cases to the police.
This report is therefore an attempt to understand the phenomenon of police brutality in
South Africa. As such it forms one of a series of reports commissioned by the Southern
African Human Rights NGO Network (SAHRINGON). These reports are intended to
document information in the 11 SADC countries on the factors contributing to police
brutality, on the form or nature and extent of police brutality, and on the institutional
responses to the problem. The reports are intended to enable SAHRINGON to do the
following:
In the light of the history of brutality by the South African Police during apartheid there
would appear to be considerable justification for the exploration of questions to do with
police brutality in contemporary South Africa. This report attempts to provide an overview
of the issue in South African focusing on the period subsequent to the transition to
democracy in 1994. In doing so it also points to some of the difficulties in coming to a
definitive understanding of the nature of the problem.
2. Defining police brutality
a. Police in South Africa
Discussions of “the police” in the post-apartheid (post 1994) period in South Africa are
usually focused on the South African Police Service (SAPS) and its key predecessor the
South African Police (SAP).3 There are however a diverse range of police agencies in
South Africa. These include traffic police, metropolitan police services and municipal
security guards which mainly fall under the jurisdiction of local government. There is also
an extensive private security industry – which in fact far outnumbers the members of the
SAPS. Recently government has established the Directorate Special Operations (“the
Scorpions”) one of a number of special investigating units which falls under the Office of
the Director of Public Prosecutions. Particularly in rural, but also in urban areas both
Permanent Force as well as Commando units of the SANDF, are also in varying ways
involved in policing activities.
State agencies which are involved in policing functions make provision for the involvement
of members of the public in varying ways. These include the above mentioned SANDF
Commando system, as well as the “Police Reservist” system of the SAPS. Outside of these
formal structures forms of vigilantism which have received significant public attention in
recent years, such as the organisation Mapogo a Mathamaga, may also be seen to represent
forms of (extra-legal) policing by the public.
Whilst acknowledging the diversity of “policing” in South Africa, this report will focus on
the SAPS. However a few of the cases of alleged brutality which are documented in section
4(a) below do involve members of the organizations, other than the SAPS, which are
involved in policing functions. This issue will also be discussed again briefly in the
discussion of these cases in section 4(b).
b. Defining police brutality
According to Bittner the central thread that runs through police work is that it frequently
“consists of coping with problems in which force may have to be used”.4 While wherever
possible police should avoid unnecessary force, force inevitably has to be used in some
situations. In situations where members of the police use force this is therefore necessary
for them to carry out their responsibilities and is not unlawful.
Police brutality however occurs when members of a police service use force unlawfully.
Essentially therefore brutality is the (unlawful) abuse of the capacity to use force. By
implication, police brutality is generally deliberate unlawful violence but actions which
amount to criminally negligent uses of force should also be considered as acts of police
brutality.
Some issues which are raised by this definition of police brutality include the following:
• Popular usages of the term – the term police brutality is used more loosely on a
popular level, sometimes referring to uses of force by the police without a concern
to establish whether they are lawful or not, while sometimes it is used to refer to any
apparently objectionable behaviour by the police. In the sense in which it is used in
this report the term police brutality is therefore used more narrowly.
• Abuses of other powers – police have a range of powers including the power of
arrest, powers of search and seizure, and others. All of these powers may be abused.
5 Effectively therefore as brutality, as defined in this report, involves abuse of the
capacity to use force, abuse of other policing powers are not dealt with here.
• The distinction between brutality and corruption – in recent years particularly
there has been widespread evidence of corrupt activities involving the police in
South Africa.6 It should be noted that police corruption can be defined as the abuse
of police powers for gain.7 While in some instances police brutality is linked to
other corrupt behaviour, brutality and corruption to a significant extent represent
different forms of police abuse of power8 particularly in that corruption is primarily
motivated by greed (or personal or other gain) while brutality has diverse
motivations including a desire to assert authority, sadism, as well as frequently a
type of over zealousness in pursuing the “crime-fighting” goals of the police
organisation. This report is therefore focused on “police brutality” and not on
“police corruption” and therefore deals with corruption only in so far as it is related
to brutality.
• Unlawful violent acts outside of the police role – one area of difficulty in applying
the concept “police brutality” concerns violence by members of a police service
outside of the police occupational role. There is some evidence for instance that
male members of police services may be prone to involvement in domestic violence.
9 However while there may be an element of abuse of power involved (for instance
police members may use their position in the police service to prevent any action
being taken against them), in terms of the approach taken here, unlawful violence by
police members outside of the police role is not strictly “police brutality”.
Nevertheless a propensity to unlawful violence outside of the police role may be
seen as part of the same problem as that of police brutality in that factors which
contribute to the one may contribute to the other, members who are prone to the one
may be prone to the other (though the latter is not necessarily the case), and finally
it may be argued that both problems need to be identified as related phenomena by
police managers and need to be dealt with as part of an overall problem of “police
violence”.10 For the purposes of this paper therefore non occupational violence by
members of the police service is regarded as part of the problem of police brutality.
• Unnecessary force – a further important distinction concerns that between
“brutality” and “unnecessary force” where the latter occurs when “well meaning
officers prove incapable of dealing with the situations they encounter without
needless or too hasty resort to force”.11 It should be noted that the distinction
between “brutality” and “unnecessary force” is not watertight. Thus for instance the
question may arise whether a police officer who has used force “too hastily” may be
judged to have been criminally negligent in his or her use of force and thus to have
been involved in police brutality. Despite the fact that the distinction between the
two problems is not watertight they can in many ways be regarded as separate
problems attached to the use of force by police. However while they are in some
ways separate, measures by police management to address the problem of
unnecessary force are arguably a key measure in addressing police brutality.
3. Problems with the use of statistics and press reports as indicators of the extent and
nature of police brutality
a. Statistics as measures of police brutality
There are numerous problems associated with measuring police brutality:
• Police action often takes place in situations of low visibility in terms of the presence
of witnesses other than the persons who are targets of police action. It is believed
that the presence of third party witnesses tends to discourage unlawful behaviour on
the part of the police. In so far as police brutality is a problem it is therefore likely to
occur in circumstances where there are no (particularly non-police) uninvolved third
parties present.
• Most obviously this means that particularly in fatal incidents of police brutality the
only surviving witnesses are likely to be members of the police. However where it
happens that people who have been victims of police action make allegations of
police brutality (i.e. that the police acted unlawfully), it is frequently difficult to
ascertain whether their allegations should be treated as credible or not. Thus in
many cases where an arrested person has bruises or injuries it is common for the
police to assert that these were sustained as a result of the person having resisted
arrest, and as a result of the police having to forcibly restrain him or her. In such
cases it is frequently not possible to ascertain who is actually telling the truth with a
reasonable degree of certainty. The problem is aggravated, not only because there
are frequently two competing accounts of what happened, but because the people
making the allegations are often of low social status, and might in general have
difficulty in presenting their version of events in a manner which is more credible
than that presented by the police.
• Systems for the lodging and investigation of complaints are usually controlled by
the police. There is potentially a wide degree of variation in how such systems
function in different localities. In some police stations for instance the practise may
be that those wishing to lodge complaints of police brutality are discouraged by one
or other means (e.g. threats, having to wait a long time) from doing so. The fact that
larger numbers of such complaints are recorded in a specific place may therefore be
more of a reflection of the relative willingness of such an agency to record such
complaints. Even if complaints are recorded the investigation may not be very
effective. However, at the end of the day, as indicated above, even an effective
investigation may simply end up with two competing versions of events and may
not be able to make a conclusive finding.
• Reporting systems may also for instance record the complaints as complaints under
specific crime categories. Thus for instance if one understands brutality as unlawful
violence by the police in the occupational role should one include charges of rape
against members of a police service within the category? The problem is aggravated
by the fact that most reporting systems do not distinguish between cases which were
alleged to have occurred in occupational and non-occupational capacities.
Particularly if one defines police brutality as it is defined here it is difficult therefore
to know which recorded cases should be regarded as allegations of police brutality.
• More specifically a specific form of police brutality which is of particular concern is
that of torture. However there is no particular offence of torture in South Africa (and
many other countries) and torture can only be prosecuted under offence categories
such as assault or assault GBH or sometimes murder or attempted murder. As a
result it becomes difficult to evaluate the extent of a particularly serious problems of
police brutality such as torture.12
• In so far as incidents where the police are involved in using force/violence are
recorded in one form or another, a major question still arises as to what criteria to
apply in deciding whether action amounts to police brutality. Thus whether the
incident is recorded in police shooting incident data (whether as legal or not) or
whether the case is recorded as an alleged criminal offence by a member of a police
service the basic question still remains as to whether this should be regarded as a
case of police brutality or not. Thus civilian witnesses to an incident may judge the
police to have been brutal but if the incident is evaluated according to legal or
professional criteria the same conclusion may not be reached. Similarly a court may
come to the conclusion that the use of force in a certain situation was justified but
this does not necessarilly imply that a professional police officer would come to the
same conclusion. Particularly if one is looking at cases against police officers which
have gone to court it must be emphasised that the issues here concern not only the
criteria applied by the courts in deciding whether actions are legally justified or not
but the issue of standards of proof. What is important to bear in mind is that the
courts apply such criteria in terms of the standard of proof of “beyond a reasonable
doubt” in criminal cases. On the other hand in civil cases where the standard of
proof is somewhat lower, it is standard practise in many police agencies to settle,
out of court, cases where they believe there is likely to be a finding against them,
(whether they believe the police action was unlawful or not). However even in such
cases part of the settlement involves an agreement that settlement of the matter does
not amount to an admission of any of the facts in question.13 Even if actions of
alleged brutality are recorded in one way or another (which may not occur in many
cases) a fundamental question arises as to the procedures and criteria to apply in
deciding whether such actions amount to brutality or not.
Effectively therefore there is unlikely to be any single source of statistical information
which directly records police brutality in a reliable way and all statistics may at best be
used as “indicators” of police brutality. In addition however there may be basic questions
about the reliability of the systems which are used to record the above types of incidents. In
using official statistics therefore one primary question always relates to the basic reliability
of the system in question. While it appears that ICD statistics are a reasonably reliable
indicator as to the number of police action and custody deaths, there are major problems
with other systems, particularly those for recording shooting incidents, and those which
record criminal charges against SAPS members.14
b. Comparing rates between different countries and localities
If statistics on uses of force by the police, and particularly of police brutality, have such
great limitations it would appear obvious that attempts at comparison between different
jurisdictions and countries should be approached with enormous caution. Perhaps most
significantly any jurisdiction with a more reliable system for recording information relating
to the use of force by the police, and which records such information more consistently,
will be prone to registering much higher “rates” of the particular behaviour being recorded.
Thus while it appears likely that the number of deaths as a result of police action and in
police custody recorded in South Africa is relatively high by world standards it should also
be borne in mind that the system for reporting such deaths to the ICD and the ICD’s system
for recording such deaths is relatively unique particularly in the developing world but even
in relation to the United States where there is no centralised national system for recording
deaths at the hands of the police.
What also needs to be borne in mind is that meaningful comparisons of differences in rates,
for instance in the use of lethal force by police, may become more meaningful if they take
into account other variables which may impact on the frequency with which such force is
used. For instance in calculating rates of “deaths as a result of police action” it might be
appropriate to calculate these relative to any one or more of the following on which data is
available:15
• Number of police (possibly excluding those having primarily administrative or
managerial responsibilities);
• Crime rates or calls for service;
• Rates of police-civilian encounters;
• Arrests;
• Potentially violent police-civilian encounters;
• Attacks on or killings of police;
• Characteristic suspect behaviour – it is arguable that if suspects perceive police to
be constrained in their use of force they may be more likely to offer resistance
thereby increasing the need for police to use greater force (however the converse
may also apply in that if suspects expect the police to use force even where they
offer no resistance, suspects may be inclined to see force as a necessary form of
self-defence);
• Population (entire, adult, “police client”);16
• Presence of firearms or other weapons in the community.
However it should be borne in mind that even quantifiable variables such as the above
might only give a partial representation of the true story. Thus one factor which may be
likely to have a particularly strong impact on the number of incidents of use of force, and
potentially incidents of alleged brutality which a particular police grouping is involved in is
the units “workstyle”. Thus an “active” police member, unit, or organisation who/which is
involved in frequent encounters with suspects and other citizens might generate a far higher
number of incidents of the use of force and potentially incidents of police brutality than one
which is relatively inactive. However the latter member, unit or department might in fact
use brutality more systematically. The question might therefore be asked as to which of the
two departments, for instance, should be said to have a greater brutality problem?
Finally in using statistics to calculate the extent of a problem it should be borne in mind
that different systems may effectively be measuring different aspects of the problems. Thus
statistics may record the number of victims, events, charges or alleged perpetrators.
However what might actually be most meaningful is to identify those police members who
are involved in the use of force most frequently. Many police theorists consider this as
potentially the most productive place to focus in identifying members who are prone to
police brutality.17
c. Press reports as a source of information on police brutality
Finally it should be borne in mind that not only statistics but also the degree to which
alleged incidents of brutality are reported on in the press or other media, and the degree to
which the press may be regarded as a reliable source of information on such incidents, may
depend not only on the relative prevalence of police brutality but also on a wide range of
other factors including:
• The degree of censorship;
• Relationships between the media and the police – in particular journalists who cover
“crime” related stories often depend on the police in order to be effective and thus
may be reluctant to give too much attention to incidents of police brutality;
• News values in the media – whether a case is regarded as “newsworthy” may impact
on whether it is reported or not. Thus police practises which are regarded as
common-place or not cause for serious concern are unlikely to receive media
attention.
• Public attitudes – where there is large scale public anger about crime, extensive
press coverage of police brutality may lead to public anger against the press, and
potentially to declining newspaper sales;
• The social class or status of the victim, as well as other factors such as race and
gender or sexual orientation may also impact on whether particular types of
incidents receive coverage or not;
• The degree to which the allegation is regarded as credible by the newspaper
concerned and the degree to which the newspaper is concerned to, and is able to,
obtain verification of the incident or has another reason for regarding the report as
reliable.
In particular it must always be borne in mind that press reports will be likely to focus on
incidents which are regarded as the most shocking and thus sensational and thus focus on
those incidents which are more serious in nature. While press reports are therefore not
necessarily meaningful as an indicator of trends they may nevertheless provide accounts of
particular incidents which in one way or another give meaningful incite as to the nature of
police brutality.
4. Indicators and evidence regarding the extent and nature of police brutality in South
Africa
a. Statistical indicators relating to the use of force and unlawful violent actions by the police.
Some the statistical indicators which are currently available relating to uses of force and
unlawful violent actions by members of the SAPS are the following:
• During the three year period April 1997 – March 2000, 2174 people died as a result
of police action or in police custody in South Africa. The number of people who
died as a result of police action was 1548 while 626 people died in police custody.18
It should be noted that roughly 70% of the deaths (95% of the deaths as a result of
police action and 12-16% of deaths in custody) are the result of the use of force by
the police. However there is no clear evidence as to what proportion of these deaths
may be the result of unlawful actions by the police and therefore what proportion
may be said to be the result of police brutality.
• According to police statistics derived from shooting incidents reports (SIRs)
covering incidents involving members of the SAPS over the three year period 1996
to 1998 in three South African provinces (the Western Cape, Eastern Cape and Free
State) 358 people were killed by members of the SAPS while a further 1002 were
wounded in shooting incidents.19 Using these figures it is possible to estimate that
for every 10 people killed by the police in shooting incidents, roughly 28 are
wounded (a ratio of 1:2.8). Applying this ratio to the figure of 1548 deaths as a
result of police action (most of which are deaths in shooting incidents) it is possible
to estimate that roughly 4332 people were injured in shooting incidents involving
members of the SAPS during the three year period April 1997 – March 2000.
• Police Standing Order 251 indicates that the SIR must indicate whether, in the
supervisor’s opinion, the police action (shooting) was legal or not.20 SAPS data on
SIRs from the Western Cape, Eastern Cape and Free State between 1996 and 1998
indicate the following:21
• Of 4754 shootings, 14% were evaluated as not legal in the SIRs.
• Among fatal shootings, the proportion deemed to be not lawful increases
dramatically to 41% (127 of 311 fatal incidents).
• While 16% of on-duty fatal shootings (28 out of 178) were evaluated to be
not lawful, 74% of off-duty fatal shootings (99 out of 133) were deemed not
lawful. This suggests that the problem of unlawful shootings is much greater
where police use firearms off- duty and also potentially that off duty
shootings tend to be evaluated more critically (perhaps because they are
often not “occupational” in nature) than is the case with on duty-shootings.
• During the two year period April 1998 – March 1999 the ICD recorded 1051 cases
of deaths as a result of police action, 468 cases of attempted murder and assault with
intent to do grievous bodily harm, 128 cases of torture, and 736 cases of common
assault. On the other hand according to statistics provided by the Minister of Safety
and Security, during the four year period 1994- 1997, members of the SAPS faced
256 charges of murder, 125 charges of culpable homicide, 630 charges of attempted
murder, 1119 charges of assault with intent to do grievous bodily harm, 3564
charges of common assault and 660 charges of point a firearm.
• For the four year period 1995 to 1998 the SAPS paid out roughly R50 316 000 in
relation to approximately 1489 civil claims apparently relating to the use of force by
SAPS members.22 This constituted more than 50% of the total of R97 372 paid out
by the SAPS in civil claims and ex gratia payments during the period in question.
Perhaps most telling in relation to these statistics is the fact that:
As soon as a civil claim is settled and the compensation paid out the state
attorney has to determine whether the SAPS members whose actions were the
subject of the claim should enjoy state protection (ESP) or forfeit it (FSP). The
guidelines are set out in treasury instructions and address issues such as
whether the person acted within his or her duties and acted in good faith (bona
fide). If it is decided that the member acted outside of these standards he or she
forfeits state protection and the state may exercise its right to recover the
amount paid out from him or her. As is indicated … in 96% of the cases of
common assault and 98% of the shooting incidents for which the SAPS was
held liable it was decided that the SAPS members concerned should enjoy state
protection (ESP).23
Taking into account the points made in section 3(a) above about the use of statistics as
measures of police brutality it is also important to note, in relation to the statistics quoted
here, that:
• ICD statistics on deaths as a result of police action and in custody do not measure
police brutality. May of these deaths are not the result of unalwful actions by the
police. While these deaths are investigated by the ICD and not the SAPS there are
very few convictions of police members in relation to these deaths. This is partly
because many of the actions are lawful but also due to other factors of the kind
outlined above. Even where there is evidence of unlawful police action there is
unlikely to be a prosecution, and even less likely to be a conviction.
• Sources of statistical information on criminal charges against the police usually do
not distinguish between cases of an occupational and those of a non-occupational
nature. Furthermore most of these cases do not result in convictions either.
• Even where members of the SAPS indicate that a shooting may have been unlawful
in a shooting incident report it is rare for such incidents to lead to the conviction of
the police member involved.
• Sources of statistical information on civil claims against the police are also of
limited use as an indicator of trends or patterns in police behaviour partly because of
the potential variations in time between the incident and settlement of the claim.
They are however in some ways interesting as an indicator regarding the response of
police management to alleged brutality by police members, particularly in that, the
indication appears to be that in the vast majority of cases the SAPS pays the claim
on the individuals behalf, rather than holding the individual responsible, despite the
implicit admission that there has been some wrong-doing.
Finally it should be noted that the existing systems only record information on four types of
incidents (deaths as a result of police action and in police custody; shooting incidents;
alleged criminal acts; civil claims) which relate to the use of force, and possible cases of
brutality, involving SAPS members.
However it can be assumed that the bulk of uses of force by members of the SAPS do not
result in deaths, criminal charges, or civil claims and do not involve the use of firearms and
therefore are not recorded in existing systems of statistics.24 Of these uses of force many
(though not necessarily the majority) may be unlawful. Existing data systems therefore
ultimately cannot be regarded as providing a reliable indication as to the extent of either the
use of force by police, or of police brutality.
b. Some reported cases of alleged police brutality
While many cases of alleged or possible police brutality are recorded by the ICD and the
SAPS the details of most of these incidents do not necessarily become public information.
However some cases are reported on in reports produced by the ICD, or by other
organisations such as Amnesty International or in reports in the press. A selection of cases
of this kind (an exhaustive list, even of the relatively small number, of cases which have
entered the public domain would be much longer) is presented in this section. These cases
have been selected in order to attempt to illustrate some of the diverse forms that police
brutality takes in South Africa.
As noted in section 3 it is often difficult to find conclusive evidence about whether a
particular incident should be regarded as a case of police brutality or not, partly because
there are often conflicting versions of events. While in some of the cases reported below
people have been convicted or the evidence (such as video evidence) is conclusive, in many
of the cases, even where the evidence of unlawful police violence is quite strong it is not
conclusive. Many of the cases below therefore need to be recognised as alleged or possible
cases of police brutality and cannot be taken to have been proven.
Case 1
On the 7th of November this year a video was shown on the South African television
programme Special Assignment which horrified most of those who watched it. The video
depicts an incident, which took place in January 1998 and which contains roughly 14
minutes of coverage of an incident which is believed to have taken place over as much as
an hour. The video showed members of the North East Rand Dog Unit involved in setting
their dogs on three Mozambican illegal immigrants in what was characterized as a type of
“training” exercise. The savaging by the dogs is combined with racial abuse of and physical
assaults on the three immigrants by the dog unit members. In the words of the Star the
video showed scenes of
six policemen setting their dogs on three suspected illegal immigrants, and
assaulting their hapless, screaming victims on a mine dumpt near Springs on the
East Rand. .. The victims pleaded for their lives with the policemen, who hurled
racial abuse at the men as they punched, kicked and slapped them. The officers
stamped on the men’s necks and faces as the animals tore into their flesh. In one
piece it appeared that a piece of a dog’s tooth had come loose and was
embedded in the flesh of one of the victims. Two Alsatians, which were joined
by another large dog, which appeared to be a cross breed, attacked the men with
such ferocity and strength that they were able to lift one of the men up in their
jaws from his crouching position in the veld. The dogs were also able to drag
the men along the ground. When the animals were finally called off, the police
officers lined up the three injured men and assaulted them further. One of the
officers asked a victim: “Is jy ‘n kaffir? Is jy ‘n kaffir” (Are you a kaffir? Are
you a kaffir?). The man was then hit across the side of his head [and] then
punched to to the ground and stamped on the face. Other officers were heard
laughing in the background. Before the men were kicked again and then
bundled into a minibus with the dogs that had mauled them, one of the
policemen threw stones at the men, and another pulled his weapon from his
holster [and pointed it at the men]; but did not fire.25
Prior to the screening of the video six white members of the North East Rand Dog unit,
Inspectors Christo Koch (32) and Eugene Truter (28) and Sergeants Kobus Smith (31),
Dino Guitto (27), Robert Henzen (32) and Nicolaas Laubser (27) were arrested. Special
Assignment identified the two police who led the assault as Sergeant Smith and Inspector
Koch.26 The policemen, who have been suspended without pay, appeared in court on
November 10 and were finally released on bail of R2000 each on 22 November.27
The three victims Gilbert and Alexander Ntimane and Sylvester Khosa contacted police in
the week in which the video was shown and were placed in a witness protection
programme.28 Against a background of widespread public anger and condemnation of the
incident by the Cabinet,29 the National Commissioner of the SAPS, The minister of Safety
and Security, various political parties,30 newspaper editorials, NGOs and the Mozambican
government the Independent Complaints Directorate (ICD) also announced that it was
investigating a total of 22 other allegations of incidents of police brutality involving the use
of police dogs.31
Case 2
Hout Bay police officer, Andre Ferreira, was arrested after he allegedly shot dead suspected
robber Dumisane Zwane on 11 May 1999 while he was lying injured on the ground.
Ferreira was one of four policemen who responded to a call to apprehend the heist suspects
who had stolen R1million in pension payouts. Testifying at the bail hearing, ICD
investigator Gavin Meyer said that a statement alleged Ferreira had shot Zwane six times
after he had fallen on his back. It is alleged that Ferreira fired several shots at Zwane as he
lay on the ground. After returning to his vehicle, he went back to Zwane and fired more
shots at him. Two of the suspects were apprehended by the police while Zwane attempted
to flee.
The DPP ordered the officer, who had been released on bail of R1000 to be prosecuted for
murder and defeating the ends of justice. Two other officer police members, Christopher
Wesso and Stephen Vermeulen were also charged as accessories with an alternative charge
of defeating the ends of justice or obstructing the course of justice. In August 2000 Ferreira
failed to attend court for trial as he allegedly was “under psychiatric treatment for stress and
depressive illness”.32
Case 3
Andries Ndou was shot and killed by two white police officers, aged 33 and 34, in the
Messina area in Northern Province on 27 April 2000.33 According to a press report a post
mortem revealed that Ndou had a bullet wound in the back of his head and two bullet
wounds in his shoulders after he was allegedly shot at close range by the police officers.
ICD Northern Province director Thobeka Jozi said the two accused officers ahd told
investigators they had fired a warning shot before shooting and killing Ndou, who was
handcuffed at the time of his death. Jozi said a lack of witnesses had severely hampered the
investigation … . Some black police officers had however alleged that their colleagues had
killed the suspect to avenge the recent murder of a white farmer, Hannes Aldem, and the
wounding of his wife. The white officers had allegedly prevented black colleagues from
accompanying the suspect. Ndou had expressed a willingness to lead the police
investigators to the crime scene, if he was accompanied by black officers, but his request
was turned down. According to one of the sources quoted Ndou was taken by the two white
policemen to point out where the murder weapon was hidden while according to another
source Ndou was taken to the scene by six white officers from the local murder and robbery
unit, dog unit and crime prevention unit.34
The ICD has been involved in investigating the case and it is expected that a formal inquest
will be held.35
Case 4
On the 16 May 2000, a second year student Michael Makhabane, was shot dead by the
Durban Public Order Police Unit on the campus of the University of Durban Westville
during protest action regarding the deregistration of 540 students who had not paid
university registration fees. The bullet that killed the student was fired by a member of the
Durban Public Order Police unit who was using live ammunition. According to a report in a
Durban newspaper “UDW staff members said that police had fired on students as they were
running away, that students had pleaded with the police not to shoot”.36
Case 5
On Sunday 30 July 2000, Bheki Mkhize (40), an ANC Member of Parliament was shot
dead by members of the Ulundi Public Order Policing (POP) unit. Mkhize was shot in the
head with an R5 rifle during a police search for illegal weapons in the area. Thirteen
members of the unit were busy searching for weapons in the area when they allegedly
received a tip-off that there were weapons at Mkhize’s home.37
According to the police inspector who allegedly fired the fatal shot, Mkhize tried to grab
his firearm during a scuffle.38 According to Prince Zeblon Zulu, Mkhize’s wife said that the
police had demanded to see Mkhize. When he emerged from his bedroom, he pleaded with
the police to talk instead of being violent, but they didn’t listen and instead dragged him
outside. Mkhize fell on the stairway as he was being dragged outside by the police. They
then beat him although he had his hands high indicating that he was not fighting with them.
39 According to ICD executive director Karen McKenzie, Mkhize was kicked and was on
the ground in a semi-seated position and was being held by two police members. He
managed to free his right hand, and was protecting his face with his hand from a rifle that
was being placed in his face. This was when he was shot with a single shot which entered
his right hand and went through his left nostril lodged at the base of his skull.40
The investigation was conducted by the ICD assisted by a senior SAPS detective and
forensic expert.41 The investigation revealed that the POP members had tampered with the
crime scene.
Those who have been charged include Inspector Samson Buthelezi who allegedly fired the
fatal shot, and Bafana Magwaza who allegedly held Mkhize while he was shot. Capt
Zeblon Dlamini, charged with defeating the ends of justice, had tried to plant an AK-47,
confiscated from another home during the police search, at Mkhize’s house. Dlamini also
removed Mkhize’s hand-gun from the scene and returned it later with and empty cartridge.
Buthelezi was granted bail of R2000 while his two co-accused were given bail of R1000
each.42 A fourth suspect, reported to have gone into hiding,43 has also been arrested and
charged and is out on bail.44 The killing gave rise to accusations of “police assassinations”
and “hit squad activities” but KwaZulu-Natal provincial safety and security MEC warned
that these statements were reckless and could increase tension in an area where political
violence had already claimed many lives.45
Case 6
Thami Zulu was killed by members of the SAPS on 2 April 1998 in Tsakane in Gauteng
province. According to a friend of his, Simon Tshowa, Zulu was watching football with him
on television. Zulu left his house between 9.30 and 10 in a Cressida. Eddie Maseko who
lives nearby had also briefly visited Simon Tshowa at about 8 and saw Zulu arriving there.
Eddie Maseko then went to his home in Khumalo St which is the next parallel street to
where Simon Tshowa lives.
Some time after returning home Eddie Maseko saw a minibus dropping some passengers
across the road in Khumalo st. Two hijackers appeared and some shots were fired (one
person was injured). The hijackers then sped of in the mini-bus. A number of people
gathered in the street. Efforts were made to call an ambulance and the police. No
ambulance arrived but after a while a police van arrived accompanied by a Brakpan taxi
association patrol car.
Some of the people gathered in the street mentioned that a Cressida had been connected to
the hijacking. The block in Khumalo st where the hijacking occurred is bordered by Matse
st on the one end and Madonsela st on the other. According to Eddie Maseko, Zulu (who
had apparently just left Simon Tshowa’s home) came driving down Matse st and then about
5 minutes later down Madonsela St.
At the prompting of some of the people gathered on Khumalo st the police then chased after
Thami Zulu in the Cressida. Soon after there were gunshots. Eddie Maseko ran though
some houses towards the shooting, Thami Zulu staggered out the car shouting to the police
“What have I done my brother?”. He died soon thereafter.
According to an ICD investigator the police version is that Zulu sped away inhis car. An
attorney representing the family also says that the police claim that Zulu was accompanied
by another person in the car and that this person fired a number of shots at the police and
then fled from the car. However according to Eddie Maseko TZ was not driving quickly.
Furthermore no firearm was found in the car with Zulu.46
The ICD states that it had been monitoring the investigation conducted by the SAPS.47 A
formal inquest is to be held and is currently scheduled for February 2001.
Case 7
The ICD is investigating the death of Nkhagweni Mundalamo (15) who was allegedly shot
by police in Khuhvi in the Northern Province on 15 June 2000. The incident occurred at a
shebeen in Khuhvi in Thohoyandou, where police were responding to an armed robbery
complaint from a taxi owner who indicated that suspects were seen at Benjie’s shebeen. A
young boy who was not a suspect was shot dead, allegedly by the police. One suspect was
also arrested in the process. The investigation is continuing.48
Case 8
In this incident, 2 members of the Pretoria Dog Unit were charged with murder after the
fatal shooting of Mr BJ Booysen when they responded to a housebreaking call. It
subsequently turned out that it was Mr Booysen who had in fact summoned the police to
the scene. The ICD took charge of the investigation, and after finalisation thereof handed
the case docket to the DPP. The DPP agreed with the findings of the ICD investigation and
ordered the prosecution of the 2 members on charges of murder in the Pretoria High Court.
On 4 September 2000 one of the accused was acquitted, but the other was found guilty of
culpable homicide.49
Case 9
On 27 June 2000 Sgt. C.C. Martin was found guilty of murder and sentenced to 18 months’
correctional supervision with a further 18 months suspended for five years in the Cape
Town Regional Court. On 23 August 1997 Sgt Martin was with his wife, when the deceased
grabbed Mrs Martin’s handbag and ran away. Sgt Martin shot at the fleeing suspect. The
deceased was fatally wounded on the back of his head.50
Case 10
On 24 February 2000, Sgt. Neo Christopher Athibeng was sentenced to life imprisonment
in the Mmabatho High Court after being found guilty of the murder of Julias Modise in a
trial before Mr Justice Nkabinde. The incident occurred at the Motimalenyora High Tavern
in Ganyesa in the North West Province on 3 February 1999, where the two started a quarrel
which ended with Sgt. Athibeng fatally shooting Modise twice with his service firearm.51
Case 11
Moses Ngubeni (21) and three of his friends were arrested by Soweto police officers and
taken into custody at the Jabulani police station in July 1998. According to police
representative Superintendent Govindsamy Mariemuthoo the four youths tried to escape
from their cells at the station. The youths grabbed a police officer’s revolver. In the ensuing
struggle two policemen were critically wounded and Moses Ngubeni killed. However
according to Matthews Ngubeni, the father of the deceased, who along with his wife
identified the body. Moses Ngubeni had been shot five times, including shots in the
forehead, chin and leg. His fingers and toes were severely cut and bruised Matthews
Ngubeni believed are signs of torture.52
Case 12
In this incident, the deceased, Gert Baardman, was arrested for trespassing and stock-theft
on 15 July 2000 and detained at Fraserburg Police Station in the Northern Cape. On 16 July
2000 he was found hanging in his cell. A post-mortem conducted on 21 July 2000 revealed
that the deceased had sustained multiple injuries prior to his death, but according to the
doctor, who conducted the post-mortem, those injuries were not the cause of death. The
investigation into the death of Mr Baardman by the ICD is still continuing.53
According to a press report one of the problems which has faced the ICD during the course
of their investigation has been that the police involved in Baardman’s arrest have refused to
give statements to ICD investigators. Apparently when the family were informed about the
death and went to the police station to inquire about a death certificate they were allegedly
told that they had to bury their son immediately. Baartman’s parents said the believed they
were forced to rush the burial in a bid to destroy evidence. They claim they saw a gash at
the back of his head, which was allegedly not revealed by the post mortem. The family said
they were threatened with arrest if they did not go through with the burial.54
Case 13
According to an ICD report “the incident occurred on 2 July 1999 at Steelpoort Diesel
Garage, Burgersfort in Mpumalanga Province. A SAPS Inspector, together with 5 civilians,
allegedly handcuffed the deceased, Mr Benjamin Mabelane, to a steel door and tortured
him with a welding machine on his genitals. Mr Mabelane died of his injuries. Another
victim, who managed to survive, was also allegedly tortured and assaulted. The ICD took
over the investigation of this matter and arrested all 6 suspects. After finalisation of the
investigation, the docket was handed over to the Director of Public Prosecutions (DPP)
with a recommendation that 3 of the suspects be charged with murder and attempted
murder and the remaining 3 with defeating the ends of justice. The DPP accepted our
recommendation and instituted criminal proceedings against all accused. The case was
recently finalised in Lydenburg Magistrates court. The SAPS member involved, as well as 3
other accused were found not guilty and acquitted. The remaining 2 accused were
sentenced to 4 and 5 years imprisonment respectively. The ICD has further recommended
that disciplinary proceedings be instituted against the SAPS member.”55
Case 14
This incident occurred in Barkly East, Eastern Cape, on 10 April 2000, when SAPS
members arrested 6 youths who had allegedly committed a burglary and culminated in the
death of one of the boys Siphiwe Zide (16). According to ICD Eastern Cape Director,
Alfredeen Jenneker, the one sergeant involved saw a boy suspected of housebreaking and
arrested him. “The sergeant and two colleagues then arrested two more boys and assaulted
them. They then arrested a further two boys and took all five to a dam, pointed a gun at
them and forced them to hold their heads underwater. They then threw stones at the boys
while they were in the water. After taking them out, they arrested a sixth youth, and took all
six to a farmhouse and allegedly forced them to break in … . After the break-in, the
policemen allegedly bundled four of the boys into the back of the van. The one driving the
van tied Siphiwe’s arm to the driver’s side with rope while another one tied the second boy
to the passenger’s side, and forced them to run alongside the moving bakkie. While they
were running, Siphiwe’s rope broke and the driver stopped the van. He then held the boy’s
arm as he ran alongside the moving vehicle. The boy eventually got tired and fell, and the
bakkie ran over Siphiwe’s head. The vehicle also ran over the second 15 year old boy’s leg,
fracturing it.56
Following investigation by the ICD, the DPP accepted the recommendation of the ICD and
decided that all the members be prosecuted on charges of culpable homicide, assault with
intent to do grievous bodily harm, theft and attempting to defeat the ends of justice. The
members appeared in the Barkly East Regional Court on 7 September 200057 and again at
the the Elliot Regional Court on 17 October 200058 and have been suspended without pay.
59 The members involved have not been named but are identified as a white sergeant and
two black constables in one report.60
Case 15
In August 1999 police at Bayview station in Durban arrested and severely assaulted four
teenage boys in connection with a robbery. One died within 24 hours of his arrest. The
police allegedly hit them with broom handles and kicked and punched them in the station
parking lot, before transferring them to Chatsworth police station where they were locked
in a freezing, filthy cell. Although the detainees pleaded for medical treatment, the police
allegedly refused.61
Case 16
The case of Zweli Kenneth Ndlozi is detailed in a report by Amnesty International.
According to the SAPS Zweli Kenneth Ndlozi was found hanging in a cell at Germiston
police station on 7 September 1998. Prior to his death 22-year-old Ndlozi had been a
member of the SANDF from Dunnotar military base, Benoni.
On the night of 5-6 September SANDF Military Police officers came to Ndlozi’s home in
Meadowlands Soweto and assaulted him severely in the presence of eye-witnesses. They
accused him of involvement in the theft of firearms. They searched the house and also dug
up the yard, apparently looking for weapons, but found nothing. They then left, taking
Ndlozi with them. His family went to police stations and prisons looking for him but could
not find him. On 8 September his family received a call from a member of the SAPS who
told them that Ndlozi had been found dead in a cell at Germiston police station, hanging by
a nylon cord around his neck. According to the information provided to the pathologists by
the police, he was found dead at 16h49 on 7 September 1998.
According to the family they were asked to go to the station to identify the body. At that
stage the family realized that a post-mortem examination had already been conducted but
refused to take the body for burial until an independent examination was done. Prompted
by a request from Amnesty International and with the family’s permission, the Independent
Medico-Legal Unit (IMLU) based in Durban arranged for an independent forensic
pathologist to conduct a second post-mortem.
This examination took place on 16 September. The resulting report documented numerous
abrasions on Zweli Kenneth Ndlozi’s legs, feet, ankles, arms, hands, shoulders, back, chest
and head, as well as lesions due to thermal injury, probably caused by cigarette burns.
Forensic tests indicated that these injuries were recent and possibly inflicted close to the
time of death. The report also noted a deep bruise on the front of the scalp accompanied by
haemorrhaging of the brain tissue, though the skull was not fractured. The forensic
pathologist found the cause of death to be “Consistent with hanging -torture not excluded.”
Although the first post-mortem report noted that a red and white nylon cord was around the
neck of the deceased, this cord was not with the body when the second post-mortem was
performed.
There are indications that the Military Police may have handed over Zweli Kenneth Ndlozi
to a specialist SAPS unit, the East Rand Firearms Unit and that members of this unit
eventually took him to Germiston police station. When he was booked in at the police
station no record appears to have been made of his physical or mental condition and no
record appears to exist that he was medically examined. Under the police regulations then
in force, this responsibility lay at the discretion of the officer in charge at the time when the
prisoner was booked in to the police station.
According to a press report, Zweli Kenneth Ndlozi was taken to Germiston magistrate’s
court on Monday 7 September but enquiries made by Amnesty International revealed no
information about any charges which may have been laid against him in court before he
was returned to his cell.
Until early April 1999, the detective branch of Germiston police station were in charge of
the inquiry into the prisoner’s death. By early April 1999 no-one had been charged in
connection with his death nor had an inquest been held into the surrounding circumstances
and cause of his death. However, in mid-April the ICD took over responsibility for the
investigation from the Germiston police.62
Case 17
In October [1999] government lawyers conceded that Brixton Murder and Robbery Unit
members had tortured 54-year-old Lucy Themba and 24-year-old Charlotte Pharamela in
June 1996. Both women, who were being interrogated about the whereabouts of Lucy
Themba’s son, had been assaulted during arrest and subjected to electric shocks and
suffocation torture while tied by their arms and legs to chairs.63
Case 18
Shaheed Cajee was arrested in October [1999] in connection with possession of stolen
goods and was allegedly subjected to electric shocks and smothered with a wet bag while
tied naked and blindfolded to a chair at Brixton Murder and Robbery Unit headquarters. He
signed a statement under duress and was transferred to Diepkloof prison pending his trial.64
Case 19
At least 16 cases of assault are now being investigated against the police and soldiers in the
Ixopo area (and the surrounding areas of Creighton, Donnybrook and Highflats) in the
KwaZulu-Ntal midlands. Some of these cases are documented in a report provided to the
author by Cheryl Goodenough. Members of the Ixopo Community Watch (ICW) are
implicated in many of the cases.65 Some of the incidents allegedly involve ICW as well as
SANDF commando members. These include:
• An incident on 30 May 2000 where men claiming to be police and army members
searched the home of the Zulu family at eHlani. According to violence monitor
Mary de Haas other houses were also searched that night, and some men were
beaten and have opened cases at the local station.
• During a three week operation an SADF unit beat and terrorised locals in a number
of different incidents. In one early morning incident they demanded of one of the
local residents that he hand over an AK47. They then dragged him outside, beat him
and kicked him, and put a rubber tyre tube over his face. The soldiers subsequently
assaulted the same individual on two other occasions. On the third occasion they
tied his hands behind his back and put his head into a bag filled with water.
• The case of Basil Jaca who died on July 2 2000, the day after he was allegedly
assaulted during a raid. The assault included having a rifle inserted in his anus. After
the assault Jaca’s wife, who was kept outside the house during the assault, found
him bleeding from his motuh and anus. Six members of the SANDF and Ixopo
resident John Arkley, a police reservist and ICW employee were arrested after the
attack.
• The case of Gqomoza Mbhele who died on September 5 2000 after allegedly being
assaulted by a police reservist employed by the ICW. Police claimed that Mbhele
and others had intervened in an attempt to prevent the arrest of a housebreaking
suspect who was being put into a police vehicle and that Mbhele had a fit, fell and
hit his head.66
Case 20
This case relates to two incidents which were captured on a video which was first shown on
the BBC’s Newsnight on the 19th of April 1999. The video was filmed by a BBC camera
crew in early 1999 and showed two incidents, which allegedly occurred on January 15 and
24 in which members of the Johannesburg Flying Squad were involved in brutality against
alleged hijackers following vehicle chases. In the first incident, after the alleged hijackers
are arrested and are lying handcuffed on the ground they are punched and kicked and a
police dog is set on one of them. One of them is also taken into some thick grass and a
lighted cigarette is stubbed out on his head. In the second incident the vehicle in which two
hijackers were fleeing had crashed. After pulling the semi conscious hijackers from the car,
one of the alleged hijackers is hit in the stomach and on the head with a rifle butt.
According to the BBC camera crew the two hijackers were left on the roadside for several
hours before an ambulance arrived to fetch them.67 One of the hijackers, Sebastian Lewis
died in hospital a few days after the incident as a result of head injuries sustained n the
crash. The other suspect arrested in the latter incident, Frans Mudau, was sentenced to 15
years imprisonment for his involvement in the vehicle hijacking.68 However the persons
assaulted in the first incident were not charged.69
Press reports immediately after the incident announced that six members of the
Johannesburg SAPS Flying Squad had been suspended without pay.70 After an
investigation conducted by the ICD, the Director of Public Prosecutions (DPP) eventually
decided to prosecute two Flying Squad members Inspector Johannes Rautenbach (29) and
Sgt. Daniel Liebenberg (26). Rautenbach was found guilty on 2 counts of common assault
and was fined R2 000.00 on each count or 6 months’ imprisonment with a further 6 months
suspended for 5 years. Liebenberg was convicted on one count of assault with intent to
cause grievous bodily harm and was sentenced to a fine of R6 000.00 or 12 months
imprisonment with a further 18 months suspended for 5 years. Following on from
recommendations made to the ICD, SAPS have agreed to institute disciplinary proceedings
against the members concerned.71
Case 21
A senior police officer in Northern Province was arrested in December 1999 for allegedly
raping a female prisoner awaiting trial at Phalaborwa Police Station at the weekend.the 36-
year old inspector was arrested after an identification parade took place at the station.
According to the Provincial police spokesperson Motlafela Mojapelo the officer is alleged
to have come to the station at the bread of day and offered to help the woman make a
crucial telephone call to her home. When the woman was to return to the cell, the police
inspector allegedly grabbed and dragged her into a kitchen where he allegedly raped her.
He said the woman had been arrested for shoplifitng in Phalaborwa. According to the report
another police inspector attached to another Northern Province Police station was arrested
in October 1999 for allegedly taking part in a robbery that culminated in the murder of
three people and and rape of a woman who was later killed. A police sergeant was also
suspended from duty for allegedly raping a woman who had come to lay charges at
Sekgosese Police Station, while another sergeant was dismissed at the same station for
raping the wife of a suspect.72
Case 22
The case concerns journalist Thabo Mabaso, who lost the sight in his left eye as a result of
an assault on him by SAPS members at the Gugulethu Police Station. The incident occurred
after Mabaso, two friends, a taxi driver and a witness went to report a minor car accident on
27 June 1998. Mabaso was allegedly assaulted by the police after an argument.73 He was
allegedly carried to the parking lot at the back of the police station where he was kicked
and beaten by several policemen. After the attack Mabaso spent the night in police cells,
bleeding and without medical attention. Doctors were eventually forced to remove his left
eye.74
The DPP accepted the recommendation of the ICD and decided to prosecute nine police
members on charges of assault with intent to do grievous bodily harm and defeating the
ends of justice. On 30 August 2000 three police officers were found guilty of assault with
intent to do grievous bodily harm while the other six were acquitted. Sergeant M A Duda
and Sergeant L Mlondleni were both sentenced to five years’ imprisonment with two years
suspended for five years while Sergeant T Ginyigazi was sentenced to three years’
imprisonment with 18 months suspended for five years.75
Case 23
A police officer from Brixton police station verbally abused, assaulted and unlawfully
detained Dr Frank Nyame, a Ghanaian research scientist, claiming that he was an “illegal
immigrant”. On 18 April [1999] Dr Nyame was accosted in the street by two white men in
plain clothes, one of whom demanded to see his immigration papers and attempted to force
him into a nearby police vehicle. When he later complained at Brixton police station, one of
the two men, now in police uniform, told him he was under arrest and knocked him
unconscious. After he had recovered consciousness he was locked in a cell and, despite
repeated requests to see a doctor, was given no medical treatment. He was released after
some hours.76
Case 24
A report in the Star of September 8 1999 documented a number of alleged acts of brualtiy
by the police against illegal immigrants. These included:
• Paulina Cossa (24), a Mozambican citizen granted permanent residence in South
Africa, said she and her brother Sergio Cossa (20) went to fetch water in Bekkersdal
on August 8 when police approached them. Pauline said “my brother ran because he
was illegal. Three policemen caught him and beat him until he fell down and died
on the spot”.
• Illegal immigrant Sergio Machava (22) was arrested by police, held at Lindela
Repatriation Camp outside Krugerdorp and loaded onto a train for deprtation on
August 25. He said “We were hit by belts by the policemen. They then asked if
those with money did not want to be taken to the border. Some agreed and paid over
R100 and they were released”. Machava said that police on the train started
searching the people, robbing those who had money. He said those who were robbed
were then told to jump from the train which was moving at speed, and those who
were reluctant or scared were simply thrown off. “I had R250 hidden in my collar.
They took it and pushed me off the train. I hit my head and face and I was coughing
blood” he said.77
Case 25
An Angolan Rastafarian “Zion” (he asked the journalist to withhold his real name) said that
he was selling tickets at a restaurant in Rosettenville, one of the southern suburbs of
Johannesburg, when a fight broke out. The police arrived and sorted out the problem and
left but then two other police vehicles arrived. According to Zion “I was talking to someone
when one of the policemen asked me where my documents were. Someone else was talking
to me, so I turned my back on the policeman when then jumped from his car and grabbed
me by my shirt and threw me in their car. Zion said the uniformed policemen drove him to
a secluded place and a white policeman started to punch and kick him. “I was afraid to
move. I knew I was innocent, so it would be unwise to run away. But the white policeman
kept telling me he was going to kill me”. Zion said he sat in the car as they drove around,
stopping twice to beat up people in cars. .. Eventually one policeman asked Zion if he
wanted to get out of the car alive or dead. Zion said “it depends on you” but also asked the
police to please leave him in peace. They told me not to beg, but to give them money. All I
had was R40 and I gave it to them, he said. They eventually dropped him off at the
restaurant, where he managed to take down the car’s registration number. He then went
home to fetch his girlfriend. They went to report the incident to the police. According to his
girlfriend, “we went to the Booysens police station, where we got a chilly reception. They
looked at us in horror as if to say how dare we report such a thing. They gave us medical
papers and said they would not talk to us until we had been to the doctor. We went to the
nearby hospital but, after waiting a while, we decided it was not worth waiting because the
police would not do anything anyway”.78
Case 26
In February [1999], in the Johannesburg area of Townsview, members of a security firm,
BBR, beat and kicked 19-year-old Archie Ngubalane, whom they had handcuffed to a
railing and accused of an attempted stabbing. They threatened to shoot a relative who tried
to intervene. Police failed to arrest the BBR members for assault, but arrested Archie
Ngubalane and detained him at Booysens police station, where he was denied proper
medical care. He was subsequently charged with attempted murder, denied bail and
transferred to Diepkloof Prison, where he was again denied proper medical care, including
for epilepsy. He was released on bail in August and acquitted of the charge in September.
No progress had been made in the police investigation of his complaint against BBR for
assault.79
Case 27
A policeman, Mr Thabo Khobotlo, was arrested on the weekend of December 9/10 2000
after he allegedly killed four people as well as wounding an 18 month old baby. He is
alleged to have shot dead his pregnant former wife, Hazel, and the parents and sister of her
new lover, Samuel, Rebecca and Ingrid Mgangane. Neighbours told the Sowetan that
Khobotlo and his wife were no longer involved. Apparently Khobotlo was told that her
proposed future in-laws were going to open negotiations about marriage plans and “went to
their house and found her with her future in-laws”.80
Case 28
The Sowetan of 20 December 2000 reported three separate incidents, all of which
apparently happened on December 19, of policemen who committed suicide after shooting
their lovers. One of the women died while the other two were taken to hospital in a critical
condition with head injuries. The incidents were:
• Inspector Johnson Buys (28), who was stationed at Galeshewe near Kimberly in the
Northern Cape, killed his girlfriend Milicent Matobe before killing himself.
• Inspector Edmund Sekatane (33) of the Protea Crime Intelligence unit in Soweto
wounded his lover Nthabiseng Sekhute and later shot himself in the head. The
couple were apparently experiencing problems after Sekhute opened an assault case
against her lover earlier in December.
• In Secunda in Mpumalana province Sergeant MC Mfusi shot and injured a
colleague, Constable CN Sukazi, before committing suicide.81
Case 29
A police commander at Jericho in North West has been arrested and suspended without pay
for allegedly raping a female awaiting-trial prisoner who was held at the local police cells
last week. Police spokeswoman Inspector Erica Roos sadi yesterday the 44-year old
inspector was arrested on December 13 for allegedly repeatedly raping the prisoner at the
storeroom at Jericho police station. She was being held for alleged theft. The inspector, a
community service centre commander, who was in charge of the overall shift on December
12, allegedly fetched the 30-year-old prisoner from the police cells. He allegedly raped her
at the storeroom and later took her back to her cell,” Roos said. Shed said he allegedly went
back to fetch the prisoner in the early hours and again raped her before returning her to her
cell. The prisoner reported the incident to policmen who reported for their shift the next
morning, Roos said.The inspector, who has been in the police force for 15 years, was then
arrested. He is expected to appear in the Garankuwa Maginstrate’s court soon”.82
c. Some aspects of police brutality in South Africa
Of the cases listed, 12 took place in 2000, 9 took place in 1999, 6 took place in 1998 and 1
took place in 1996.83
The roughly 2 dozen deaths referred to in the above list of cases represent roughly 1 percent
of the total number of recorded deaths as a result of police action and in police custody in
the last four years in South Africa.84 While we may assume that many (even the majority)
of theoverall number of deaths are not caused by unlawful actions by the police it is
nevertheless possible that the number of deaths caused by police brutality in South Africa in
the recent period includes many which are not listed here. Amongst the deaths listed in the
cases above are:
• Apparent execution type killings in custody (Case 3) and at the point of arrest (Case
2) – in these cases criminal suspects appear to have been killed as a form of
“punishment” in circumstances where they posed no threat to life and where the use
of force appears no longer to have been necessary to secure their arrest;85
• Deaths apparently linked to torture or other assaults in custody (cases 11, 12, 13, 14,
15, 16) and outside of custody (case 24)
• Other deaths related to apparently unjustified, reckless or accidental shootings
(Cases 6, 7, 8, 9) including killings during demonstrations (case 4)
• Killings relating to arguments and domestic disputes (Cases 10, 27 and 28).
• Off duty killings of criminal suspects involved in petty offences (Case 9).
• Other deaths linked to assaults of persons not in custody (Case 19).
Amongst the incidents which are non-fatal in nature are:
• Other instances of torture (Cases 17 and 18) including the use of police dogs for this
purpose (Case 1)
• Non-fatal assaults of persons at point of arrest or otherwise outside of custody (Case
20)
• Case of rape86 (Case 21 and 29) and other assaults of persons in custody (Case 23,
Case 24, Case 25, Case 26) and reporting cases to the police (Case 22).
It should be noted that the cases include some which occurred off-duty and which are not
related to the performance of a policing function (Cases 10, 27 and 28)87 which in terms of
our definition do not strictly qualify as cases of police brutality but which nevertheless have
been included here due to the fact that such “non-occupational violence” appears to
constitute an important part of the problem of violence by members of police services in
South Africa.
It should also be noted that the cases listed are in some ways fairly unique in that, in a
relatively high proportion of the cases listed, some form of action in the form of arrest,
prosecution and even conviction, has been taken. In this respect the cases break down as
follows
• In four cases (19, 21, 27, 29) there is at least an indication that the SAPS (or other
organisations) members allegedly involved have been arrested.
• In another four cases (1,2,5, 4) the information provided indicates that prosecutions
have been instituted.
• In five cases (8, 9, 10, 20, 22) it is indicated that convictions were obtained against
SAPS members.88
• In two cases (3,6) involving the deaths of a person formal inquests are being held.
• In three cases (7, 12, 16) it is stated that the investigation is continuing while in one
case (26) it is stated that “no progress” has been made with the investigation;
• In a number of cases (4, 11, 15, 17, 18, 23, 24) it is not clear what steps have or are
being taken while in one cases (25) it is specifically stated that nothing is being
done partly as the victim was discouraged from reporting the case by the attitude of
the police and obstructed from doing so by being told that he first had to obtain a
medical certificate.
• In one case (28) the three different perpetrators all committed suicide and therefore
it is not possible to hold them responsible.
In relation to the five cases where it is stated that convictions were obtained the police
members were sentenced to jail sentence in two (10, 22) and to non-custodial sentences in
two cases (9, 20) while in one case (8) it is not stated what form the sentence actually took.
In one other case (17) it is stated that government lawyers conceded that torture had
occurred and therefore presumably this case has resulted in a successful civil claim though
it is not clear whether it has also led to any type of criminal prosecution or, as with all the
other cases listed where the members have not been sentenced to terms of imprisonment,
(SAPS members face automatic dismissal when they are sentenced to a term of
imprisonment), whether any disciplinary proceedings have been instituted against the
members involved, and the outcome of such disciplinary proceedings.
It must be emphasised therefore that these cases, while they may suggest something about
the nature of police brutality in South Africa are not representative of police brutality cases,
particularly in that, in a relatively high proportion of the cases listed arrests have been
made, prosecutions have been instituted, or convictions have even been obtained.
As shown “police brutality” in South Africa is not restricted to the main formal police
agency, the South African Police Service, but also involves other agencies involved in
performing a policing function including, the SANDF, private security agencies and
community based structures such as the Creighton Community Watch. However, even
where it is other structures that are involved in brutality it often appears that these
structures operate in cooperation with the SAPS and that SAPS members may be unlikely
to intervene or take appropriate investigative action (as is the case in terms of case 26)
when such brutality comes to their attention.
It should also be noted that, considering the degree to which the police in South Africa were
used in the past for political purposes the cases listed in particular do not provide much
evidence of the police being used for political purposes or that the police are politically
motivated in terms of their actions. One aspect of this is that there have been what seems to
be a relatively small number of killings during demonstrations in South Africa in the period
since 1994 and particularly in the last few years reflecting positively on the measures taken
to improve public order policing in South Africa.89
One case which many assumed to be politically motivated was the killing of Bheki Mkhize,
an ANC member of parliament by the Ulundi Public Order Police (POP) Unit.90 However
the facts that have emerged thus far have not clearly indicated what the motive for the
killing was and whether this was in any way political in nature.
In the case involving the Ixopo Community Watch, on the other hand there is some
evidence that those who have been targeted have been people who, at least at one point,
have been linked to the ANC. However it appears that in some ways the major motivation
for the actions of the ICW has been the killings of farmers which occurred in the area, and
the targeting of persons who are supposedly ANC aligned may reflect a perception that they
are behind the farm killings partly based on a suspicion that those conducting the killings
are using weapons which they acquired in the times of political conflict. Thus in these cases
perceptions relating to particular political groupings appear to influence the form of
brutality but this is different from saying that brutality forms part of police conduct which
is essentially politically motivated.91
This is not to say that members of the SAPS are no longer ever involved in brutality which
is politically motivated. It is likely that some elements within the police service continue to
strongly identify with particular political parties or concerns. However it appears that the
primary motives for police brutality resemble those which motivate police brutality in other
countries where the police are not primarily engaged in suppressing political opponents.92
d. The profile of victims and perpetrators in terms of race and other factors.
South Africa is a society which only very recently has emerged from apartheid rule which
itself was predated by many centuries of racial-colonial domination of black people by
whites. Within South African society therefore race and racism continue to be powerful
influences. Thus it is not surprising that six of the incidents listed (1, 2, 3, 4, 20 and 25)
point to the involvement of white police officers in acts of brutality against black persons.93
However this statement itself points to some of the complexities involved in analysing
issues of race or racism and brutality in South Africa. Thus while in so far as there are clues
to the racial identity of the victims, these appear to be virtually exclusively black (in case 8
the victim’s name is apparently that of a white person. It is assumed that the victim in case
12 is a coloured person while the victim in case 18 appears to be an “Indian” person),94 in a
further five cases (5,6, 10, 22, 27) it appears that the police members involved in these acts
of brutality are black while in two cases (14, 19) the persons involved in these acts of
brutality appear to include both black and white persons.95
Thus the selection of cases provided here suggests that while black people are the victims
of a high proportion of police brutality cases, in a significant number of cases where
brutality occurs this is as a result of the actions of black police, or even groups of police
which include both black and white persons.96 However at the same time it should be noted
that even in relation to those incidents where white police where involved in alleged
brutality against black persons it is not clear in most of these incidents to what degree they
were specifically acting out racist attitudes, or engaged in explicitly racist behaviour, while
carrying out these acts of brutality.
The one incident which represents an exception to this is that involving the North East
Rand Dog Unit (Case 1) where the video recording clearly captures the police involved
directing explicitly racist abuse at their victims. This incident, which was strongly
reminiscent of police behaviour during the apartheid period has served as a strong reminder
of the racist origins of the SAPS and suggests that intensely racist attitudes persist amongst
certain members. However as is the case with police brutality itself, racism is mostly
disguised and hidden and therefore it is not clear quite how prevalent it actually is.
However what also appears to be a disturbing trend (reflected in cases 1, 23, 24, and 25) is
the apparent high level of victimization of black foreigners by the police. While in the two
cases presented here where evidence of the race of the police involved is provided (Cases 1
and 23)97 the police members are both white, many cases of abuse against foreigners
involve black police. At the same time the police in Case 1 used abusive language which
was typically directed against black South Africans by racist whites during the apartheid
period. Thus racist and xenophobic attitudes and practices, though to some extent distinct,
may also be intertwined, and both feed into brutality.
While the victims of police brutality are generally male adults it should also be noted that in
addition to the rape cases (Cases 21 and 29) and killings in domestic circumstances (Case
28) where women are the victims, cases of torture and other acts of brutality by the police
also include amongst their number cases where women (Case 17) and children (Cases 7, 14,
15) are the victim.
Furthermore the victims of police brutality include not only persons suspected of being
involved in criminal activity but also immigrants both legal (Cases 23 and 25) and illegal
(Cases 1 and 24), members of the public involved in demonstrations (Case 4), persons who
are apparently wrongly identified as suspects (Case 6), innocent bystanders (Case 7),
victims of crime (Case 8) or other persons reporting matters to the police (case 22), persons
believed to be able to provide information relating to the whereabouts of criminal suspects
(Case 17) and others who were victimised by reason of personal relationships or other
circumstances (Case 10, 27 and 28).
5. Variables impacting on occurrence of police brutality
a. Root causes
Several factors may be seen as root causes of police brutality in South Africa some of
which are specific to the South African context but some of which are characteristic of
policing itself:
• Nature of policing – policing, as pointed out, is intimately involved with the use of
force Thus brutality is in some ways an inevitable by product of policing, the special
powers accorded to members of police services, and the arming of police service
members, and will inevitably proliferate unless concerted steps are taken to
discourage it.
• Historical factors – the South African Police Service was formed by integrating the
South African Police Service and 10 other police organisations all of which had a
reputation for brutality.98 Thus in South Africa policing can be said to reflect a
“habit of brutality” and transformation of the police requires that police be taught
(or assisted to learn) how to carry out their role effectively without relying on
brutality.
• High levels of violent crime – it is clear that the high levels of crime, which is often
of a particularly violent nature, is itself a major contributing factor to brutality. One
aspect of this is the widespread proliferation of firearms which in some ways
necessitates and legitimates the arming of the police. Furthermore the high levels of
attacks against the police appear to necessitate that they are also armed off duty.
More broadly however major exposure to violent crime and high levels of fear
contribute to desensitising and dehumanising members of the police service. Finally
the high crime environment also contributes to legitimating police brutality such
that public pressure is towards increasing “toughness” from government and the
police in dealing with crime and there is little public pressure or support for
measures intended to discourage brutality.
• Declines in CJS effectiveness – in an environment of proliferating violent crime
there have been major declines in the effectiveness of the Criminal Justice System
in South Africa during the 1990s. This has two consequences. On the one hand these
declines contribute to a perception that relying on the mechanisms of the criminal
justice system is likely to be ineffective and therefore that “self-help” measures are
called for. Thus it is likely that the apparent growth of vigilantism in South African
society generally has been paralleled by a growth within the police of “police
vigilantism”. On the other hand this decline is also manifested in the breakdown of
internal systems of control within police organizations. Thus it may be expected that
internal systems which are intended to discourage brutality have had limited impact
within the SAPS.
• Proliferation of agencies involved in policing – one further by-product of the
major crime wave affecting South Africa has been the proliferation of the private
security industry and of the involvement of other agencies in policing. As suggested
by the cases presented above the problem of brutality is not restricted to the SAPS
but extends to other agencies involved in policing. However in so far as measures
are implemented to discourage or prevent brutality these tend to be focused on the
SAPS with the consequence that there is limited scrutiny of, for instance, the private
security industry.
b. Legal frameworks regulating policing
Components of the legal and regulatory framework which are relevant to the use of force
and to preventing or sanctioning police brutality include the following:
• The common law -The core provisions of law which justify the use of force are
provisions of the common law which in combination define the circumstances in
which the use of force in “private defence”99 (commonly known as ‘self defence”)
may be justified. The law on private defence applies generally to everyone in South
Africa including members of the police service.
• The Constitution of the Republic of South Africa, Act 108 of 1996 – The use of
force by the police is not addressed directly anywhere in the Constitution. The
section of the Constitution which has profound and wide ranging, though not
necessarily straightforward implications in terms of the use of force is Chapter 2, the
Bill of Rights. Provisions which may be seen as relevant to the use of force include
provisions regarding: human dignity; life; freedom and security of the person;
privacy; assembly, demonstration, picket and petition; freedom of movement and
residence; labour relations; property; children; just administrative action; access to
courts; arrested, detained and accused persons, limitations of rights; states of
emergency. Provisions which also impacts on the investigation of incidents
involving the use of force are those which provide, or support, a right to remain
silent to persons including persons “arrested for allegedly committing an offence”.
100 It would appear that Section 36(1) of the Bill of Rights, which provides for
rights to be limited by a “law of general application to the extent that the limitation
is reasonable and justifiable in an open and democratic society based on human
dignity, equality and freedom” is also relevant to the interpreting provisions of the
Constitution and legislation which impact on the use of force.101
• The Arms and Ammunition Act, 75 of 1969 and associated regulations – The Act
inter alia defines the circumstances in which a person may be disqualified from
owning a firearm.
• The Criminal Procedure Act, 51 of 1977 – Provisions of the Criminal Procedure
Act which relate directly to the use of force include sections 27 and section 49.
Section 27(1) empowers a police official who is legally entitled to search any person
or any premises to use “such force as is reasonably necessary to overcome any
resistance against such search or against entry of the premises, including the
breaking of any door or window of such premises”. Section 27(2) requires that the
police official first audibly demand admission except in particular circumstances
which are outlined.102 Issues concerning arrest, including when arrest is authorised,
are dealt with in the Criminal Procedure Act primarily in section 39 -52. According
to Section 39(1) “Unless the arrestee submits to custody, an arrest is effected by
actually touching his person or, if the circumstances so require, by forcibly
confining him”.103
The prevailing position in South African, with respect to Section 49, the provision
governing the use of force in effecting arrest is one of uncertainty. Early in 1997 the
SAPS issued a Special Service Order which in some ways effectively modified the
fairly permissive provisions of Section 49 particularly by restricting the categories
of offences in relation to which lethal force was authorised. An amendment to
section 49 was passed late in 1998.104 Subsequently however the Minister of Safety
and Security refused to allow the provision to come into operation.105
• The Regulation of Gatherings Act, No. 205 of 1993 – Most notably in sub-section
9(2) the Act provides a framework for the use of force by the police in situations,
inter alia, where a gathering poses a “danger to persons or property” which cannot
be averted by other steps (detailed at 9(1)), or where the gathering is prohibited.
• The South African Police Services Act, 68 of 1995 – The powers of members of
the police service are to some extent defined by section 13 of the SAPS Act. The
issue of the use of force is referred to explicitly in subsection 13(3)(b) which
provides that “Where a member who performs an official duty is authorised by law
to use force, he or she may use only the minimum force which is reasonable in the
circumstances”.
• Policy on the prevention of Torture and the Treatment of Persons in Custody of
the South African Police Service – Known as “the Anti-torture policy” at Section 2
it makes it explicit that “No member may torture any person, permit anyone else to
do so, or tolerate the torture of another by anyone”. Torture is prohibited by the
South African Constitution (at 12(1)) and, particularly where it involves physical
force, constitutes a criminal offence. It is also, as the policy indicates contrary to the
United Nations Convention against Torture and other Cruel, Inhuman or Degrading
Treatment or Punishment of 1984. The provisions of the policy finally came fully
into effect on July 1st 1999.
• SAPS Standing Order as amended from time to time – Many of the standing
orders of the South African Policy Service are of relevance in one way or another to
the use of force. Most notable is Standing Order 251 which is discussed below in
the section on Accountability and Review Mechanisms. Also of particular relevance
is Standing Order 108 in terms of which members are issued with firearms as part of
their personal issue (discussed in the section on systems for issueing firearms and
other weapons to police officers). Other relevant standing orders would include, for
instance, those relating to the operation of the disciplinary system.
• SAPS Discipline Regulations – The regulations outline how disciplinary
procedures are to be dealt with. In section 18 the regulations define the different
forms of misconduct. In terms of the misuse of force 18(3) which indicates that an
employee commits misconduct when she or he “performs an act or fails to perform
an act which constitutes an offence” may be the most relevant.
c. International Human Rights Standards
South African has ratified many of the international Human Rights Standards including:
• The International Covenant on Civil and Political Rights which South Africa ratified
on 10 December 1998.
• The Convention Against Torture and Other Cruel, Inhuman or Degrading Treatment
or Punishment which South Africa ratified on the same date.
• The African Charter on Human and People’s Rights which South Africa acceded to
on 9 July 1996)
• The Convention on the Rights of the Child which South Africa ratified on 16 June
1995.
d. Case law
In Raloso v Wilson & others 1998 (1) BCLR 26 (NC), 1998 (2) SACR 313 (C) an
application to refer the constitutionality of s49(2) to the Constitutional Court was refused in
the light of the intended amendment of the section. The court therefore did not address
questions to do with the constitutionality of lethal force in any detail. Apparently however
the court did recognise “that there was a reasonable possibility that the section was
unconstitutional” (Fedler, 1999, 15-8, note 1) and in some ways endorse counsel’s
contention that ‘a manifestly unconstitutional statute remain[ed] on the statue books
purporting to give legal authority for the killing of persons in circumstances which [could]
not be countenanced by the Constitution” by terming it ‘indeed a sorry state of affairs’ (at
35J–36A) (Snyckers, 27-24A at note 1).106
e. Existing institutions regulating policing
• The judiciary – The period subsequent to the transition to democracy has seen
Increased emphasis being placed on judicial independence. Thus there may be an
increasing willingness on the part of the judiciary to condemn brutality by the
police. At the same time in rural areas particularly problems still persist of close
relating to the “small town’ phenomenon of close personal relationships between
members of the police and prosecution service and the magistracy. Furthermore
there may be a reluctance on the part of magistrates particularly to treat instances of
police brutality, and particularly torture, with the appropriate severity. Furthermore
the ability of the judiciary to deal with cases appropriately depends ultimately on the
quality of investigations by the ICD or SAPS as well as the effectiveness of the
prosecution service, and there are a number of problems relating to all three of the
above, and their capacities and motivation to investigate and prosecute cases of
police misconduct effectively. Finally it must be emphasizes that an approach which
relies on the judiciary as a mechanism for preventing police brutality is ultimately
likely to be deeply limited for a number of reasons, some of which have been
outlined above, to do with the difficulties of proving cases of police brutality, as
well as other reasons which have been outlined in major texts on the matter.107
• The executive – There has been a fairly substantial contrast between the role played
by the executive during the first democratic government (May 1994 – June 1999)
and the second (June 1999 to the present). During the first period the executive
appeared to observe a fairly consistent commitment to human rights though the
commitment, and their vision of the role of the police, was never articulated
particularly effectively or with much commitment. As a result the impression was
created within the public domain that the country suffered from a lack of leadership
in tackling the crime problem, an impression which appears to have contributed to
the growth of vigilantism as public and police disillusionment with government
particularly in relation to its willingness to tackle the crime problem, grew. The
appointment of Steve Tshwete as Minister of Safety and Security under the Mbeki
government in mid-1999 was initially marked by a totally contrasting approach.
One of his first steps in office was to obstruct the implementation of legislation
intended to restrict the use of lethal force by the police (and members of the public)
to circumstances of threat to life or threat of serious injury. Tshwete was not only
keen to boost police morale and demonstrate governments commitment to fighting
crime but also made a number of public statements which amounted to an
incitement of police brutality.108 His period in office has also been associated with
the downscaling of the National Secretariat for Safety and Security. Since the
beginning of this year Tshwete has become more restrained in his public
pronouncements though there continues to be little sign of a real commitment to
opposing police brutality. Overall the first six years of democratic government in
South Africa have been marked by a lack of insight and sophistication on the part of
government in understanding how to combine opposition to brutality with
promoting police safety and effectiveness.
• Independent Complaints Directorate – while it was first provided for in the 1993
interim constitution, the ICD only became fully operational in April 1997. While
some observers have expressed a degree of frustration with the body others have
been inclined to, at least until quite recently, give it the benefit of the
acknowledgement that it had been set up only relatively recently and thus that it
might be premature to criticise it. However whether inclined to criticise or defend
the body all commentators suffer from a lack of information as no full evaluation
has been conducted relating to the relative effectiveness of the ICD in appropriately
finalising cases which it is involved in. Its overall role requires that the ICD
combine a tenacious investigative approach with promotion of improvements in
practise within the SAPS in dealing with misconduct. In addition to issues of the
resources (budget) provided to the ICD, the powers afforded to its members, and the
effective management of the organisation one aspect which has been the subject of
some concern relates to the issue of the “independence” of the ICD and the fact that
the ICD falls under the Minister of Safety and Security. One final aspect which is a
definite weakness of the ICD relates to its ability to communicate effectively in
order to enable the issues which it is engaging with to be understood and to be
clearly addressed.
• The South African Human Rights Commission and other Chapter 9 institutions
– Chapter 9 of the South African constitution provides for the existence of six “state
institutions” intended to strengthen Constitutional Democracy. Of these the South
African Human Rights Commission has played the most prominent role in raising
issues of police brutality (the Public Protector also deals with complaints against the
SAPS but these relate to problems with “service delivery”). Thus during the latter
part of 1999 the head of the SAHRC, Barney Pityane, was one of the few high
profile personalities to publicly challenge The Minister of Safety and Security in
relation to his efforts to obstruct the implementation of the amendment to section 49
of the Criminal Procedure Act. The SAHRC has also engaged with issues relating to
the treatment by the police of immigrants, racism in the SAPS, and conducted a
brief tour of police custody facilities in relation to the reports of the high number of
deaths in custody. However the primary external agency which is responsibility for
engaging with issues is the ICD and the SAHRC, whilst it has engaged with these
issues seeks to avoid an overlap with and the duplication of functions which are
performed by the ICD. Their position as a Chapter 9 body therefore allows them
greater independence and they have used this position at times, though not
necessarily with much effect.
f. Initiatives intended to address brutality
Subsequent to the introduction of democratic government in South Africa various measures
have been implemented which are intended to prevent and discourage brutality. These
measures have included both “external” measures in the form of oversight mechanisms as
well as measures “internal” to the police.
In the transition period the emphasis can be seen to have been primarily placed on
“external” mechanisms. In addition to being accountable to parliament and potentially
answerable to the relevant parliamentary committee provision was also made for:
• A national ministry or “Secretariat” of Safety and Security as well as similar
secretariats in each of the nine provinces;
• The Independent Complaints Directorate which, in terms of the Interim Constitution
(Act 200 of 1993) was intended to be “an independent mechanism under civilian
control, with the object of ensuring that complaints in respect of offences and
misconduct allegedly committed by members of the [SAPS] are investigated in an
effective and efficient manner;
• Community Police Forums which were intended to promote the accountability of
the police service to local communities and “improve transparency in the
service” (South African Police Service Act, section 18(1)(f)).
While emphasis was placed on the creation of these external mechanisms, some initiatives
have been taken which can be described as more “internal” in nature. These have included:
• The introduction of a rigorous selection systems as part of recruitment procedures
for new members;
• Revision of the entire basic training curriculum;
• The development and introduction of a human rights training curriculum for
existing members of the service;
• The introduction of a code of conduct;
• The implementation of a Special Service Order on the use of force in affecting
arrest, intended to bring the regulatory framework relating to the use of lethal force
more in line with the Constitution;
• The development and introduction of an anti-torture policy;
• The reorganization and retraining of public order police;
• The introduction of the Tonfa baton as an intermediate weapon;
• The introduction of a programme of “critical incident debriefing” as a means of
providing support to police officers suffering the consequences of stress and
exposure to traumatic incidents.109
6. Limitations of current measures
As noted above the primary focus of approaches taken to the transformation of the SAPS
during the transition period was on the creation of a number of “external” mechanisms of
control. Some of the limitations of the measures implemented relate to these mechanisms.
Thus for instance neither the powers, nor the resources provided to the Independent
Complaints Directorate would appear to be adequate to the task which it has to fulfill.
Similarly it appears that the National Secretariat for Safety and Security has also been
downgraded thereby reducing its ability to play a significant oversight function. Finally it
would appear that community liaison bodies such as the CPF’s are inherently unlikely to
play a significant role in addressing the issue of police brutality.
The more “internally” focused measures have also been in many ways limited in their
impact. While no formal evaluation has been conducted it would appear that the anti-torture
policy might have had a generally beneficial impact. At the same time the primary impact
of the policy is likely to be on the treatment of arrested persons after they have been taken
to the police station to be placed in custody. As is perhaps suggested by the incidents
reflected in the BBC (Case 20) and Dog Unit (Case 1) videos, the introduction of measures
which are focused on preventing torture in custody may be associated with a
“displacement” of brutality to the period immediately after arrest and to localities other than
the actual police station and custody facilities.
More generally the measures which appear to be most rigorous are those, such as the
selection system and basic training, which impact on new recruits (as well as those focused
on the public order policing units). However new recruits are only being introduced into the
SAPS relatively slowly and thus the new selection and recruitment systems are only likely
to have an impact on the SAPS in the long term.
In terms of training more generally it is in relation to the measures which are focused on
members already in the service that there are the most problems. Apart from human rights
training, none of the measures implemented would appear to have any real potential to
contribute to behaviour change in relation to the problem of brutality. However the human
rights training is not sufficiently integrated with functional aspects, and with other
measures intended to support changes in police practise, with the consequence that
members are inclined to denigrate or otherwise dismiss it. Similarly the introduction of the
Tonfa has been obstructed due to the need for members to have a certain standard of
proficiency in the use of the weapon in order to use it effectively. However the realities in
relation to the provision of in-service training mitigate against such training being
implemented.
Further measures which are intended to address the values of members of the service are
entirely insubstantial. The code of conduct is loaded with jargon but does not provide a
clear framework of values and in particular says nothing about the use of force. Thus,
effectively there is no single guiding document which speaks clearly to members of the
SAPS in relation to issues of the use of force and brutality. Without any clear guidance in
terms of policy it is not surprising that commanders are often inconsistent or even entirely
neglectful in addressing issues of the values of the service with the consequence that issues
to do with police conduct, particularly in relation to questions of the use of force and
discouraging brutality, are rarely properly addressed.
Other problems, particularly those relating to the functioning of the internal disciplinary
system, have the consequence that key internal control mechanism play an exceptionally
limited role within the SAPS in general. Within a context of a general breakdown of
discipline it is unlikely that the problem of brutality will be addressed.
Within this context it is perhaps not surprising that the Minister and National Commissioner
have opposed, on the grounds of a concern for police safety, the amendment of section 49
of the Criminal Procedure Act intended to introduce a more restrictive framework for the
use of lethal force.110 Effectively therefore, almost six years after the April 1994, the South
African government has not as yet implemented an amendment to this notorious provisions
of the Criminal Procedure Act. The current position regarding the use of lethal force in
affecting arrest can therefore only be deduced from an informed reading of the
Constitution, the Criminal Procedure Act and the SAPS Special Service Order, and a
number of court judgements relating to the issue.
Overall the current approach taken to issues of brutality may be summarized as one of nonengagement.
Thus what we have is a situation where issues of the use of force, and thus of
police brutality are addressed primarily by default. In relation to the dog unit video for
instance the response of the Minister and the police has been to see the incident as a dog
unit problem without acknowledging the broader problem of brutality.
But what the police need is not that issues relating to the use of force are ignored
(supposedly in the interests of police safety) but that that there is an active engagement with
the difficult issues which face them in engaging with issues of the use of force in our
violent democracy.
7. Recommendations on the way forward
Due to the high levels of crime in South African currently the situation is not highly
conducive to public or government support or attention being given to measures to combat
crime. Efforts to tackle the problem of brutality therefore need to engage with the need for
improvement in police effectiveness in tackling crime and be formulated in such a way as
to support the police in meeting this objective. Furthermore efforts to engage with the
problem of brutality need also to be combined and integrated with, measures to address the
widespread problem of police corruption. Some of the types of measures which would be
appropriate would include:
• An active engagement with issues relating to the use of force. Such an approach
should combine and integrated measures to discourage excessive and unnecessary
force with measures to improve police safety, and effectiveness in using force where
such force is required.
• Improved powers and resourcing for the Independent Complaints Directorate. It is
clear that the ICD is severely under-funded and is lacking the necessary powers it
needs to enable it to fulfil its mandate adequately.
• The development of effective internal SAPS systems for the lodging of complaints
against police members and improvements in the functioning and effectiveness of
the internal investigative and disciplinary systems within the SAPS.
• Measures to clearly articulate the basic values of the SAPS and address the
responsibilities of the leadership of the SAPS in promoting these values and in
addressing issues of police integrity.
• Any attempt to clarify the values of the SAPS in relation to the use of force is
however likely to be in vain unless this is combined with an attempt to resolve the
current conundrum relating to the legislative framework regarding Section 49 of the
Criminal Procedure Act and the use of lethal force in affecting arrest.
• Furthermore measures to tackle brutality also need to be combined with efforts to
address problems of racism and xenophobia within the SAPS.
Overall however what is most clearly needed is a demonstration of greater commitment on
the part of government and police leadership to addressing issues of brutality, whilst
combining this with efforts to promote police morale and effectiveness.
8. Notes:
1 Note that this report was completed at the end of 2000. While he larger report on Police
Brutality in Southern Africa of which it is part was published by Afronet in 2002 this
specific report is only up-to-date as at the end of 2000.
2 The Star, 9 December 2000. Another two incident of alleged torture involving members of
the Brixton Murder and Robbery Squad are recorded below (see Case 17 and 18 in Section
4).
3 The SAPS was in fact formed (in terms of the South African Police Service Act, 58 of
1995) by amalgamating 11 police agencies which included the SAP and the police agencies
of the four “independent homelands” and six “self governing homelands”.
4 Bittner, E. (1974) Florence Nightingale in Pursuit of Willie Sutton. In Bittner, E. (1990)
Aspects of Police Work. (pp. 233-268) Boston: Northeastern University Press. p. 256.
5 One example concerns the power to arrest in relation to which the requirement is that a
person may be held for a maximum of 48 hours before being brought to court (section
35.1(d)(i) of the South African Constitution). However if the period of 48 hours expires
over a weekend, or other day “which is not a court day” the provision is that that arrested
persons must be brought to court not later than “the end of the first court day after the
expiry of the 48 hours. Police members may therefore deliberately postpone arrests to, for
instance, Thursdays, so that they can make use of the additional time over the weekend to
question a person. This is arguably an abuse of power.
6 See for example Syed, T. & Bruce, D. (1998). Inside and Outside the Boundaries of
Police Corruption, in African Security Review, Vol. 7, No. 2 and Newham, G. (2000).
Towards Understanding and Combating Police Corruption. In Crime and Conflict, No. 19,
Autumn.
7 See further Syed, T. & Bruce, D. (1998). Police Corruption: Towards A Working
Definition. In African Security Review, Vol. 7, No. 1.
8 Centre for the Study of Violence and Reconciliation. (1999). A New Day? Controlling
Police Violence and Corruption. In Crime and Conflict, No. 18, Summer.
9 See for instance Nix, J. (1998). To Protect and Abuse: An exploratory study discussing
intimate partners of police as victims of domestic abuse. Paper presented at the Centre for
the Study of Violence and Reconciliation, Seminar No. 4, 4 June.
10 Bruce, D. (2000) The Management of the Use of Force in the South African Police
Service. Unpublished research report submitted to the Faculty of Management, University
of the Witwatersrand.
11 Alpert, G.P. & Smith, W.C. (1994) How Reasonable is the Reasonable Man?: Police and
Excessive Force. Journal of Criminal Law & Criminology, Vol. 85, No. 2, pp. 481-501 at
489.
12 In South Africa the ICD has provision for complaints to be classified either as complaints
of assault, assault GBH, attempted murder (all criminal offence categories) or torture.
However it is not clear what criteria are applied in distinguishing e.g. cases of torture from
cases of assault GBH or whether such criteria are consistently applied by the ICD in its
different regions.
13 A further problem with civil claims is that these often take a number of years before they
are finally settled. A claim settled in any particular year may therefore relate to an incident
which occurred two or three years previously.
14 Bruce, D. and G. O’Malley (1999) In the Line of Duty? Shooting incident reports and
other indicators of the use and abuse of force by members of the SAPS. Unpublished draft
report for the Independent Complaints Directorate. Centre for the Study of Violence and
Reconciliation.
15 The points listed here are based extensively on those made in Adams, K. (1996)
Measuring the Prevalence of Police Abuse of Force. In Geller, W.A. & Toch, H. (Eds.)
Police Violence: Understanding and Controlling Police Abuse of Force. (pp. 52-93) New
Haven: Yale University Press.
16 See Adams (footnote 14), at p. 59;
17 See for instance Adams 1996 (see note 14) and Toch, H. (1996) The Violence-Prone
Police Officer. In Geller, W.A. & Toch, H. (Eds.) Police Violence: Understanding and
Controlling Police Abuse of Force. (pp. 94-112) New Haven: Yale University Press.
18 Bruce, D (2000) The Prevention of Police Action and Custody Deaths. Briefing
document prepared for the workshop on the prevention of police action and custody deaths.
Centre for the Study of Violence and Reconciliation.
19 Bruce and O’Malley (op cit, note 13).
20 The evaluation contained in a SIR is the initial opinion of the investigating officer and
does not have the status of a court judgement.
21 Bruce and O’Malley (op cit, note 13).
22 The claims cover categories designated as “Common assault” (427 claims paid out),
assault (unrest related) (29 claims paid out), shooting incidents (378 claims paid out),
shooting incidents (unrest related0 (9 claims paid out), injuries bystanders (19 claims paid
out), loss of support: deaths and disabilities (24 claims paid out), damage to property (603
claims paid out).
23 Bruce and O’Malley (note 13 above), 1999, p. 49–50.
24 This was confirmed by a recent American study (Garner, G. and C. Maxwell (1999)
Measuring the amount of force used by and against the police in Six Jurisdictions. In U.S.
Department of Justice (1999) Use of Force By Police: Overview of National and Local
Data. Publication No. NCJ 176330.) but there appears little reason to doubt that the
generalisation also applies in South Africa.
25 The Star, November 8, 2000
26 The Star, November 8, 2000
27 Beeld, 23 November 2000.
28 The Star, November 10, 2000
29 The Citizen, 9 November 2000.
30 The Citizen 10 November 2000.
31 Sunday World 19 November 2000. See also City Press, November 12.
32 Cape Times, 26 May 1999 and Citizen, 15 August 2000. (The Amnesty International
Annual Report 2000 says that three officers were charged in addition to Ferreira).
33 ICD Report on Outcomes or progress of some ICD cases, 14 September 2000 and Star, 8
May 2000.
34 Star, 8 May 2000.
35 ICD Report on Outcomes or progress of some ICD cases, 14 September 2000.
36 Press reports and other information on the incidents summarised in Marks, M
“Transforming Robocops? A Case Study of Police Organisation Change”. Paper presented
at SASA Annual Congress, July 2000, Cape Town.
37 The Sowetan, 2 August 2000.
38 The Star, 31 July 2000.
39 City Press, 6 August 2000.
40 The Star, 12 August 2000.
41 The Star, 1 August 2000.
42 Business Day, 15 August 2000.
43 Star, 12 August 2000.
44 ICD Presentation to the Parliamentary Portfolio Committee for Safety and Security, 11
October 2000.
45 The Star, 1 August 2000.
46 Compiled from statements by Simon Tshowa and Eddie Maseko and conversations with
ICD investigator Robbie Raburapu and Wits Law Clinic attorney Sebastian McKay.
47 ICD Report on Outcomes or progress of some ICD cases, 14 September 2000.
48 ICD Report on Outcomes or progress of some ICD cases, 14 September 2000.
49 ICD Presentation to the Parliamentary Portfolio Committee for Safety and Security, 11
October 2000.
50 ICD Report on Outcomes or progress of some ICD cases, 14 September 2000.
51 ICD Report on Outcomes or progress of some ICD cases, 14 September 2000.
52 Aaron Nicodemus, “Another Dry White Season” in Mail and Guardian, August 6 1999.
53 ICD Presentation to the Parliamentary Portfolio Committee for Safety and Security, 11
October 2000.
54 Business Day, 18 August 2000. Note that the latter report indicates that Baartman was
both arrested and died on July 16.
55 ICD Presentation to the Parliamentary Portfolio Committee for Safety and Security, 11
October 2000.
56 The Star, 18 April 2000, Mail and Guardian 20 April 2000.
57 ICD Presentation to the Parliamentary Portfolio Committee for Safety and Security, 11
October 2000
58 ICD Report on Outcomes or progress of some ICD cases, 14 September 2000.
59 Mail and Guardian, 11 August 2000.
60 Mail and Guardian, 20 April 2000.
61 AI annual report 2000.
62 Amnesty International SOUTH AFRICA: Establishing a culture of accountability for
human rights violations. CASE STUDY: The “disappearance” and death of Zweli Kenneth
Ndlozi, September 1998, 12/05/1999
63 AI annual report 2000.
64 AI annual report 2000.
65 According to Goodenough’s report the Ixopo Community Watch (ICW) is a section 21
(non-profit making) company whose employees include police officers and ex-policemen.
All of the civilian employee of the ICW are police reservists which gives them full powers
as policemen. The ICW was previously known as the Ixopo Farm Watch but the name
change is largely cosmetic. Despite its change of name the ICW operates from fees paid by
farmers and is not inclusive of community structures. The ICW is effectively a private
police force. Ensuring that its members are registered as police reservists enables it to both
be a private security force and exercise the full powers of members of the police. On
operations with members of the SANDF, these police reservists have therefore replaced
full-time police officers who would usually accompany soldiers on policing type
operations.
66 Cheryl Goodenough, Report on Assaults in the Greater Ixopo Area.
67 The Star, 20 April 1999.
68 The Star, 11 July 2000, Sowetan 22 April 1999, Star, 22 April 1999.
69 Beeld, 22 April 1999.
70 Sowetan, 21 April 2000. (Later reports indicated that the SAPS was considering the
suspension of a further 8 members. – Sunday Times, 25 April 1999; Rapport, 25 April
1999.)
71 ICD Presentation to the Parliamentary Portfolio Committee for Safety and Security, 11
October 2000.
72 The Sowetan, 2 December 1999.
73 The Sowetan, 3 August 2000.
74 The Star, 8 August 2000.
75 ICD Report on Outcomes or progress of some ICD cases, 14 September 2000.
76 Amnesty International Report 2000 – Country Reports (South Africa)
77 The Star, September 8, 1999
78 Saturday Star, 19 December 1998.
79 AI Annual Report 2000.
80 Sowetan, 11 and 12 December 2000.
81 Sowetan, 20 December 2000.
82 Sowetan, 21 December 2000.
83 There is no date for incident 12.
84 Since the ICD started operating nearly four years ago in April 1997 it has recorded in the
region of 2 700 deaths. However the deaths recorded in the cases listed would not
necessarily be recorded in ICD statistics. Thus in particular deaths which are linked to
“police reservists” (e.g. Case 19), or some of the alleged deaths of illegal immigrants who
were thrown from trains (case 24) might not have been reported to the ICD or recorded by
them.
85 The attempts to classify these deaths and other incidents needs to be treated with caution.
Some of the deaths are to some extent unexplained. Thus the full facts of the Bheki Mkhize
case (case 5) are not as yet known to this writer and it may possibly be an execution type
killing or an accidental killing related to reckless firearm use.
86 It should be noted that the rape of a suspect may arguably be regarded as a form of
corruption (if sexual gratification is seen as a form of “personal gain”) as much as it
represents a form of brutality or other form of “abuse of power”. The fact that it is not
amenable to easy categorisation points to the difficulties and limitations of attempts to
define and categorise forms of police criminality or misconduct.
87 It is debatable as to whether or not Case 9 should be included here.
88 Note that in case 13 the persons convicted were not SAPs members.
89 In addition to the incident listed in case 4 there are only a few other incidents which are
known to the author of this report in which people have been killed in demonstrations. One
of these happened on 8 January 1999. Twenty-two-year-old Yusuf Jacobs was shot by
police who were attempting to disperse a demonstration by the Muslim organisation
PAGAD against a visit to Cape Town by the British Prime Minister. A number of others
were injured, including a journalist. Yusuf Jacobs, who was dragged from the scene by the
police, died in hospital four days later. The ICD had not concluded its investigation by the
end of 1999. (AI Annual Report 2000) In the second incident on 30 May 2000 in Abel
Phetla (17), a student at Realogile High School, Alexandra, was shot and killed by members
of the SAPS who were attempting to disperse a group of angry students. The students were
marching in protest against the death of another student Andrew Radebe, who was allegedly
shot and killed the previous day by a relative of a local shop owner. The students were
angry that the police had not arrested the suspect in that matter. The ICD conducted a full
investigation into the matter, and after finalisation thereof handed the docket to the DPP for
his decision. The DPP is still to decide. (ICD Presentation to the Parliamentary Portfolio
Committee for Safety and Security, 11 October 2000) The latter incident is, as far as is
known to this author, incorrectly attributed to the Public Order Police in the ICD report
quoted.
90 While there has been a major improvement in the quality of policing in relation to
demonstrations this does not necessarily mean that there are not other problems with public
order policing units, particularly where they are involved in other policing functions. In
addition to the Bheki Mkhize killing POP members have been linked to acts of brutality in
Umtata in the Eastern Cape as well for example (Mail and Guardian, 20 April 2000).
91 These observations are based on this author’s interpretation of the Goodenough report.
92 Brutality against immigrants (discussed further in the next section) may be seen by many
as political in nature, depending in part on how the word “political” is used or defined.
However while brutality here may in some ways be related to the objective of discouraging
immigration to South Africa there is not evidence that police who are involved in abuses
against immigrants tend to see themselves as promoting the cause of a particular political
party or movement.
93 In a couple of cases (4 and 6) the information about race is not contained in the specific
case study but is known from other information received by the author. Unless specific
reference is made to the race of the person, the persons surname is used as a guide to their
“race”. While this needs to be acknowledged as a fairly risky approach to dealing with these
issues the approach is used here (as well as in Bruce, 1998) in an attempt to shed some light
on issues to do with race and policing in South Africa. At the same time it is acknowledged
that the method is problematic firstly because of the type of stereotyping which it uses
(persons with certain names are identified as belonging to certain race groups) but also
because of its use of racial categories which are themselves problematic.
94 In five cases (15, 21, 23, 24, 29) there is no information which suggests the possible
racial identity of either the police member or the victim and in two other cases there are no
clues to the racial identity of either the victim (9) or the police member (8).
95 In case 28 it would appear that one of the police (Buys) involved might have been a
coloured person while the other two police involved are apparently black. In case 14 the
available information suggests that it was the white Sergeant who was in some ways
“directing” the brutality but it is not at all clear to what extent the black constables were
active or passive accomplices.
96 Similar findings emerged in a 1998 study of deaths as a result of police action and in
police custody (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. Report
for the Independent Complaints Directorate. Centre for the Study of Violence and
Reconciliation).
97 In cases 24 and 25 the racial identity of the police is not apparent.
98 Rauch, J. (1998). Police Reform and South Africa’s Transition. Case study written for the
Northern Ireland Programme at the Kennedy School of Government, Harvard University,
USA. p. 2. The others were the police agencies of the four “independent homelands” and
six “self-governing” homelands. The process of amalgamation was consolidated with the
passing of the South African Police Service Act in late 1995.
99 Snyman, C (1995) Criminal Law (3rd ed.) Durban, Butterworths. p 97.
100 Bruce, D, Savage, K and De Waal, J (2000) A Duty to Answer Questions? The Police,
The Independent Complaints Directorate and the Right to Remain Silent. SAJHR, Vol 16,
Part 1.
101 Chapter 11 of the Constitution deals with security services. In so far as it deals with the
SAPS it largely deals with issues relating to the distribution of national and provincial
powers in relation to political and executive control over the police service. One of the
general provisions relating to security services which may be seen as of direct relevance to
the use of force is section 199(6) which provides that “no member of any security service
may obey a manifestly illegal order”.
102 Joubert, J. J. (Ed) (1998) Criminal Procedure Handbook, 3rd ed. Republic of South
Africa: Juta & Co. Ltd. p. 123.
103 Op cit, p. 87.
104 Section 7 of the Judicial Matters Second Amendment Act, No 122 of 1998.
105 In relation to the amendment of section 49 see Proudlock, P. (1999) License to Kill:
Police Use of Force. Crime and Conflict, No.15, pp.28-32 as well as Bruce, D (1999) Why
we can’t give police more firepower. Sunday Times, 19 September.
106 Some of the other case law relating to the use of force is discussed in Burchell, J (2000)
Deadly force and fugitive justice in the balance: The old and new face of section 49 of the
Criminal Procedure Act. In SACJ, 13, 200 – 213 as well as in Joubert (see footnote 102
above).
107 See for instance, Skolnick, J. and Fyfe, J. (1993) Above the Law: Polcie and the
Excessive Use of Force. New York: The Free Press. and Klockars, C.B. (1996) A Theory of
Excessive Force and Its Control. In Geller, W.A. & Toch, H. (Eds.) Police Violence:
Understanding and Controlling Police Abuse of Force. New Haven: Yale University Press.
108 See for instance the Mail and Guardian, 10 November 2000.
109 Some of the other “human rights culture” orientated policies introduced include the
“anti- discrimination drive, sexual harassment policy, diversity programme and new change
management focus” (Mail and Guardian, 10-16 November 2000).
110 It should be noted that the evidence from the US is that restrictions similar to that
proposed in terms of the amendment to section 49 have not had a negative impact on police
safety and may even have contributed to an improvement in such safety (see Geller, W. &
Scott, S. (1992) Deadly Force: What We Know. Washington DC: Police Executive
Research Forum).
© Centre for the Study of Violence and Reconciliation