WHEN CAN I FIRE? Use of lethal force to defend property · of lethal force to protect property was...

The 2003 ISS National Victims of Crime surveyconcluded that South Africans are much morefearful of crime today than they were in 1998

(see article by D Mistry in this issue). This growingpanic has prompted a wide range of self-protectivemeasures, including many people arming themselvesin anticipation of a criminal encounter. There havebeen a number of recent, well-publicised incidents ofthe use of lethal force in defending property. Thesehave been accompanied by media statements to theeffect that killing in defence of property is acceptableunder South African law.1

This situation poses dangers of its own. Ever since thedebate surrounding the changing law on use of forcein effecting an arrest hit the headlines, South Africanshave been confused about when they can and cannotuse their guns to defend themselves. If they err on theside of caution, they could lose their lives. If they erron the side of violence, they could lose their liberty.

This article focuses on just one aspect of this debate:the use of lethal force to defend property. While thecase law remains unclear, the guiding principlestoday suggest that killing another person in order toretain property is unlikely to be deemed lawful bythe courts.

The right to defend yourselfIn common law, the controlling principle on the rightto use force to defend one’s self or one’s property is

proportionality: the defensive act may not be moreharmful than necessary to ward off the attack.Although there are no hard and fast rules, courtsweigh up the interests protected by the defensive actagainst the interests infringed by the unlawful attack.

In determining whether a crime victim actedreasonably, the courts judge each case on its ownmerits. Certainly, an owner who is confronted by arobber is not expected to abandon his property. Heis entitled to protect it, and the court will considerall the circumstances2 when deciding whether themeans of defending the property were reasonable.

This right to self-protection can provide a defence toa charge of assault or even, in some cases, murder.Our law allows you to defend yourself, anotherperson, your property or the property of anotheragainst a current or imminent unlawful attack.3

When a person pleads private defence, his claim isthat the injury he caused was, in the circumstances,lawful and permissible.

This common law defence is often confused withthe statutory provision contained in Section 49 ofthe Criminal Procedure Act as amended,4 whichallows for the use of force when effecting an arrest.Despite certain similarities, these defences shouldnot be conflated with each other as they are usedfor different purposes and have differentrequirements.

SA CRIME QUARTERLY No 8 JUNE 2004 1

Anton du Plessis, Institute for Security [email protected]

WHEN CAN I FIRE?

Use of lethal force todefend property

Can you use lethal force to protect your property, and if so, when? South Africans are confused about how

much force they can use in defending themselves from crime, and mistakes in this area could have disastrous

consequences. While the law remains unclear, the constitutional right to life is likely to be given precedence

over the right to protect property.

Various requirements must be met before thedefensive act will be considered lawful. The attackmust be:• commenced or imminent;• against a legally recognised interest;5 and• unlawful.

The action made in defence must be:• necessary to avert the attack;• reasonable in terms of the amount of force used;

and• directed against the attacker.

Thus, the action taken must be in response to acurrently pending aggressive action, and the lawspecifically rules out any action being taken, on theone hand, pre-emptively or, on the other, in ‘revenge’.

What does the case law say?The first authoritative decision that dealt with the useof lethal force to protect property was Ex ParteMinister of Justice: In re S v Van Wyk.6 In this case, ashopkeeper whose shop had been repeatedly brokeninto took desperate measures to protect hisbelongings and rigged a shotgun in such a way thatthe intruder would trigger the device upon breakinginto the store. One night an intruder broke in, set offthe device and received a fatal wound. On a chargeof murder, the shopkeeper invoked private defenceand the court upheld his defence, acquitting him onall charges. The court reasoned that a person may, inexceptional circumstances, use lethal force to protecthis property when there is no other way in which thegoods can be retained. The only limit the courtimposed was that the value of the goods should notbe of a trivial nature. This decision was later followedin S v Mogohlwane.7

In terms of these two decisions, killing in defence ofproperty could be justified in situations wherevaluable property was being stolen. However, thesejudgements were handed down almost 40 years ago –long before South Africa’s shift to a human rightsdemocracy. If faced with similar facts today, the courtswould undoubtedly arrive at a different decision.

Changes under the new constitution South Africa’s new constitutional democracy turnedour legal system on its head. The Bill of Rights

SA CRIME QUARTERLY No 8 JUNE 2004 2 DU PLESSIS

protects various fundamental human rights, includingthe right to life8 and the right to property.9 In cases ofprivate defence, it is inevitable that these rights willneed to be weighed against each other. The court’sbalancing act would have to comply with therequirements as set out in section 36 of the Bill ofRights: was the infringement reasonable and justifiablein an open and democratic society based on principleson human dignity, equality and freedom? Applying thistest, it is unlikely that any reasonable court wouldconsider it justifiable to take another person’s life indefence of property.

Consider the following scenario. You are woken in themiddle of the night by the sound of breaking glass. Youlook out of your bedroom window and see a thiefstealthily driving your new sports car down thedriveway. You shout at the thief to get away from yourcar, but he ignores you and continues to drive away. Indesperation, you grab your gun and fire at the thief,killing him.

Your defence is that you were protecting your valuableproperty and that there was no other way ofpreventing the thief from stealing the vehicle. Also, thetheft was still in progress, so your defence wouldcomply with the requirements that the defensive actshould be aimed at an attack that is not yet completed.

In terms of the Van Wyk decision, you would almostcertainly succeed with this defence. However, in lightof the constitutional changes noted above, it is verypossible that you would find yourself in danger ofbeing convicted of murder.

On the other hand, you could argue that the Bill ofRights also protects your right to your property, andthat the constitution does not provide for a hierarchyof rights.10 This is perhaps so, but recent decisions haveindicated that the right to life cannot be arbitrarilyinfringed, allowing for lethal force only in situationswhere lives of innocent persons require protection.11

The landmark decision in S v Makwanyane12

entrenched the right to human life by abolishing theimposition of the death penalty in South Africa. Thecourt also made passing reference to the need to bringother aspects of South African law in line with the

constitutional emphasis on the sanctity of humanlife. With reference to section 49 of the CriminalProcedure Act, the court warned that if the statewas no longer permitted to take a life inpunishment of a convicted criminal, then howcould the law allow anyone to take the life of aperson they are trying to arrest.

The same reasoning would surely apply to someonewho takes the life of the thief who steals hisproperty. Evading lawful arrest is equally, if notmore, serious than theft.

Similarly, the more recent decision of Govender vthe Minister of Safety and Security,13 followed theMakwanyane reasoning in respect of using deadlyforce. The court held that the use of lethal force ineffecting an arrest may only be used if the fleeingsuspect poses an immediate threat of bodily harmto members of the public. This decision was laterfollowed in the constitutional court case of Ex ParteMinister of Safety and Security and other: In re S vWalters and Another.14 In short, these cases confirmthat use of deadly force can only be justified whenthe suspect poses a threat to the lives and safety ofothers.

If we apply this to the car theft scenario above, thenit is clear that you would not be able to use lethalforce to prevent the theft of your vehicle. You wouldhave to resort to other non-lethal methods of tryingto prevent the crime. If during your lawful attemptsto prevent the theft, the thief retaliates and poses athreat to your life or anyone else, only then wouldyou legally be entitled to use necessary force todefend yourself or others.

It is important to remember that before you can actin self-defence, the attack against you should havecommenced, or at least be imminent. For example,if the thief pulls out a firearm and aims in yourdirection, then you would be justified in usinglethal force to protect your life. However, youcannot shoot the unsuspecting thief on the premisethat if you confront him, he would place your life indanger. The pre-emptive strike principle is notapplicable in private defence cases.

Consider another set of circumstances. You wake upone night and discover that an intruder has broken

SA CRIME QUARTERLY No 8 JUNE 2004 3DU PLESSIS

into your living room. The thief is armed with afirearm and is sneaking through the house,gathering valuable items as he proceeds.

You know that if he is startled he might shoot youor your family. Can you lawfully shoot him? Do youhave to take your family and flee from your home?Do you have to wait for him to attack you or yourfamily?

Unlike the scenario with the car thief, this time theintruder is in your home. However, the same legalprinciples apply. You cannot use lethal force toprevent him from walking out with your TV. Instead,you or your family would have to be in immediatedanger. It could be argued that the mere fact thatthe intruder is in your home is sufficient threat tojustify your using lethal force against him. Again,each case could be judged separately, but thelegally safe option would be to avoid using lethalforce until you have no other option.15 Rather avoidconfronting intruders. It could save your life andkeep you out of jail.

In shortThe principle is simple: the life of the attacker canonly be taken in order to protect your or someoneelse’s life or to prevent serious bodily harm. It isunlawful to use lethal force in any othercircumstances. In other words, your property is notworth the life of the person that is stealing it fromyou!

Endnotes1 News24, 26th of May 2004.2 These include: the value of the property, nature and

extent of the danger, the time and place of theoccurrence, etc.

3 Self-defence is commonly referred to as private defence. Private defence captures the broader scope oflegally recognised interests.

4 Act 57 of 1977.5 Most legal systems have approached the question of

what interests may be protected by private defence in acasuistic way, with the result that not all potentiallyrecognisable interests have been recognised as thesubject of self-defence. Examples of legally recognisedinterests include: life, limb, property, dignity, personalfreedom, chastity and sexual integrity.

6 1967 (1) SA 488 (A).7 1982 (2) SA 587 (T).

SA CRIME QUARTERLY No 8 JUNE 2004 4 DU PLESSIS

8 Section 11 of the Constitution.9 Section 25 of the Constitution.10 This approach is supported by CR Snyman, Criminal

Law sixth edition 2003 at 108, who argues that lethalforce may be used when it is your last availablealternative to defend your property.

11 Lethal force would also be justified to prevent serious bodily harm and to prevent rape.

12 1995 (3) SA 391 (CC).13 2001 (2) SACR 197 (SCA).14 Case unreported.15 When judging whether or not the defence was

reasonable in the circumstances, the court will avoidassuming the role of an armchair critic and will takethe traumatic and emergency nature of the incidentinto account.


An important aspect of police accountabilitythat the South African Police Service(SAPS) needs to address relates to systems

for recording information about the conduct ofpolice officials. The police are the most publicmanifestation of government authority and havelegal power to use lethal force when necessary.They also represent the front-line in combatingcrime and enforcing the law, which makes holdingthem accountable even more important.

All government departments are required by law topresent their annual report to parliament. The SAPS’2000/01 annual report was criticised by the PublicService Commission for its lack of information onimportant aspects of what the police do, such asadministrative or human resources practices.1

Although the 2002/03 annual report showsimprovements in these areas, similar concerns havebeen raised about SAPS reporting systems on policemisconduct and the use of force.

In 2002, the Independent Complaints Directorate(ICD) indicated to parliament that “it had found thatcases of misconduct were grossly under-reported bythe South African Police Service and Municipal

Police Services (MPS)…probably because there isno obligation on them to do so”.2 The ICD doesnot, however, have a mandate to tackle structuralissues in the police or any authority to enforce itsrecommendations.

Data on misconduct is important for transparencyand public accountability, but also for policemanagers who need to exercise internal control andmonitor their staff. If internal systems on policeconduct are neglected, other efforts to produce anefficient and professional police force will beundermined.

Use of lethal forceIt has long been recognised that because policecarry lethal weapons – in the form of guns – theforce they may use could have lethalconsequences. Adams points out that “the capacityof the police to use coercive and deadly force is socentral to understanding their functions, one couldsay that it characterises a key element of theirrole”.3 The ICD has noted that for thetransformation of the SAPS to be successful, athorough understanding of the extent of police useof force is necessary.4

Themba Masuku Centre for the Study of Violence and [email protected]

NUMBERS THATCOUNT

National monitoringof police conduct

Publicly available data on police misconduct and the use of force has been found to be unreliable and

inconsistent. Such information is important for public accountability, but also for police managers to monitor

their staff and thereby improve performance and service delivery. Indicators of police use of lethal and non-

lethal force, torture, public complaints and corruption must be developed and monitored by the police, and

reported to the public via the SAPS annual report.

Given that the ability to effectively use firearms isan important component of police work, their useand abuse should be of key concern to policemanagement. Furthermore, abuse of firearms canundermine civil liberties, human rights andultimately, democracy.

This means that the police service should havesystems to record and analyse all usage of firearmsby their members. This would enable problematictrends to be identified and addressed throughfocused interventions such as training or increasedsupervision. The success of such interventions couldeasily be demonstrated by changes in key indicatorssuch as those relating to civil claims or policesafety. The benefit would be that over time, allpolice members would become competent andtherefore confident in the use of their firearms.

The only publicly available data on police use oflethal force is published by the ICD. However, theICD data on police shootings is largely limited tothose that result in deaths “caused, or reasonablybelieved to have been caused, by a member of theSAPS while [on duty] or in his or her capacity as amember of the Service…”. The ICD hasdiscretionary powers to investigate shootingincidents that do not end in death, but this has tobe as a result of a formal request, and only arelatively small number are investigated.

The ICD’s 2002/03 annual report shows that of the311 deaths as a result of police action, 294 werecaused by shootings. The report does not, however,indicate how many of those shootings were illegaland how many were legitimate.

Questions have been raised about the capacity ofSAPS systems to provide a coherent picture ofmembers’ use of their firearms and of lethal force.Research published in 2001 indicated a lack ofcredible systems or the non-existence of suchsystems in many provinces in South Africa.5

Incidents of shooting were not always entered onthe centralised database, and six of the nineprovinces did not have the relevant records. Thiswas despite the existence of SAPS Standing Order251, which requires a “full factual report” to berecorded immediately in the centralised system

following any incident in which a member “fires aweapon, allows a weapon to be fired or orders thefiring of a weapon”.6

The police still face challenges in the administrationof this data. While some improvements have beenmade, in general, problems remain around ensuringthat the data is collected and managed properly.Part of the problem is that there are no seriousconsequences for not recording the relevantinformation.

The responsibility for monitoring shooting incidentsand improving the use of firearms is that of theSAPS. Parliament, via the Minister of Safety andSecurity, should insist that effective systems are inplace and are properly managed. The results shouldbe published in the SAPS annual report.

Use of non-lethal forceThe use of a firearm constitutes ‘lethal force’ whilethe use of weapons such as batons, pepper spray,dogs and flashlights is regarded as non-lethal force,although in some cases there can be fatalconsequences. Non-lethal force is the mostcommon type of force used by police in the courseof their duties. And without adequate monitoringmechanisms, the opportunities are many for thistype of force to be used for purposes other thanfighting crime. When this happens, police brutalityis the result.

The SAPS is obliged, through the commitments ofthe South African government, to ensure that policebrutality does not occur. South Africaacknowledged the obligation to prevent and protectits people against police brutality or torture with thesigning in 1994 of the United Nations Conventionagainst Torture and Other Cruel, Inhuman orDegrading Treatment or Punishment. By signing theConvention, the government undertook to worktowards its ratification, thus binding the state to theConvention. The right not to be tortured is alsoentrenched in the constitution of the country.7

Despite these commitments there are no knowncredible data sources on the use of non-lethal forceby the police. As a result, the extent of the problemis not known.8 ICD records, for example, do not

SA CRIME QUARTERLY No 8 JUNE 2004 6 MASUKU


distinguish between police use of non-lethal forceduring an arrest, and police criminality. In 2002/03,the ICD recorded 1,002 allegations of criminaloffences against the police which included cases ofassault with intent to cause grievous bodily harm,common assault and attempted murder.9 It isunclear whether these acts took place whilemembers were on duty (presumably during thecourse of an arrest) or while they were off duty.

It is crucial that the types of force used whenarresting a suspect are documented, especiallyconsidering that most fatalities at the hands of thepolice (58%) happen during arrests.10 The ICDcannot, however, be expected to be the mainsource for this kind of information. Given that itdepends on the public to lodge complaints aboutpolice use of non-lethal force, ICD records can onlyever reflect part of the problem.

Sources other than the ICD on non-lethal force haveproved equally limited. Research has shown thatinquest reports – given the lack of detail in the J56form – do not provide much insight on the natureand type of force used by police either.11

TortureHow torture is defined will obviously have a directbearing on how levels of torture are determined.The SAPS Prevention of Torture Policy includes aneven more expansive definition of torture than thatcontained in the United Nations ConventionAgainst Torture (CAT).12 However, SAPS complaintsrecords do not distinguish complaints of torturefrom any other complaints, making it difficult toidentify and monitor this practice within police.

Once again, the ICD provides figures on only thosecases that are reported to them. Although it adoptedthe SAPS definition of torture, “[the ICD] does nothave, by its own admittance an accurate picture oftorture, and available statistics provide little insightinto rates of prevalence”.13 Indeed, the Directoratehas thus far utilised a very narrow definition oftorture in which certain methods (such as electricshocks, suffocation, and suspension) governwhether an act is regarded as torture or not.14 Assuch, ICD data does not necessarily correspondwith either the CAT or SAPS definitions of torture,

which means many cases falling within thesedefinitions would be excluded.

ICD statistics are further limited by the fact that, likecases of non-lethal violence, there is no legalobligation on the police to refer cases of torture orassault to the ICD. The Directorate thus relies on thepublic to bring such abuses to its attention. Manymore cases of assault are reported to the SAPS itselfevery year, some of which are likely to fall into thecategory of torture, as defined by the SAPS policy.However, because no distinction is made betweentorture and other kinds of assault, the extent of theproblem is not known.

Deaths in police custodyDeaths in police custody have been a major humanrights issue in South Africa since the days ofapartheid. It is not surprising then that the treatmentof people in police custody is regulated in terms ofthe constitution, the SAPS Act, and numerous SAPSregulations that govern the handling of suspectsfrom the time of their arrest to when they arehanded over to the Department of CorrectionalServices.15

The SAPS Act imposes a statutory obligation on thepolice to notify the ICD in all cases of deaths inpolice custody. This is done to ensure that deathsare accounted for and investigated by an externaloversight body. The ICD distinguishes betweendeaths in police custody and ‘police-action’ relateddeaths. Deaths in custody are limited to those thatoccur inside the police holding cells, while deathsas a result of police action refers to fatalities thatresult from the actions (or non-actions) of thepolice.16 This distinction is not, however, reflected inthe ICD’s published data, making it difficult toestablish the numbers of either type.

Nevertheless, a study on custody-related deathsfound that some deaths could have been avoided ifthe police had acted, by for example, providingimmediate medical attention.17 Similar issues areraised by the fact that 50% of deaths in policecustody during 2002/03 were as a result of naturalcauses, implying that “the deceased either becameill or was already ill when they were taken to policecustody”.18 The ICD needs to establish whether

MASUKU

SA CRIME QUARTERLY No 8 JUNE 2004 8 MASUKU

these deaths could have been prevented andwhether appropriate steps were taken to ensure thewell-being of inmates.

Although the ICD investigates all incidents broughtto its attention, it should be the SAPS that examinesthe incidents and trends with a view to preventingfurther deaths. Stringent action should be takenagainst police officials when deaths in custodyoccur, if it is found that procedures were notfollowed. Police management must send a clearmessage that contraventions of regulations andguidelines that lead to deaths in custody will beseverely dealt with.

Civilian complaintsThe SAPS policy on civilian complaints is containedin Standing Order 101, which obligates members toregister complaints immediately in the OccurrenceBook and to issue a reference number to thecomplainant. The Standing Order also obliges thestation commissioner or senior officers at the stationto immediately investigate any reported civiliancomplaint and report the results to the relevant Arealevel office.

Research has, however, shown that these proceduresare not always followed. A study conducted by theICD in 2001 highlighted that the SAPS system forhandling complaints was not being implementeduniformly.19 The study revealed that complaints –both verbal and telephonic – were handled in ahaphazard manner. Police stations had their ownunique ways of handling civilian complaints, andsome did not even have a complaints registry.

The ICD is mandated to investigate publiccomplaints of police misconduct or alleged criminaloffences by the police. It can receive complaintsdirectly from individuals or from the police,although the SAPS is not obliged to report suchcomplaints to the ICD. The Directorate hasrecommended that all complaints be registered, inline with the relevant Standing Order.

To this end, stations need to establish a user-friendlysystem that formally records all complaints whetherthey are written, verbal or anonymous. These shouldbe analysed, and the trends reported on. The

2002/03 SAPS annual report only covers thenumbers of complaints received through theNational Complaints line.20 No analysis of thesecomplaints is included in the report.

Complaints data from stations, as well as the areaand provincial offices, will enable police managersto analyse trends of civilian complaints and dealwith them. Apart from keeping managers informedabout public dissatisfaction with particularmembers, units or stations, it also offers a realpossibility of improving police–community relationsand service delivery to the public.

Police corruptionThe United Nations Office on Drugs and Crime’srecent Country Assessment Report concludes that interms of public perception, the SAPS is regarded asthe most corrupt public service in the country.21

While this is a perception, it is backed by the factthat the SAPS recognises the impact of corruptionwithin its ranks:

…corruption among police membersseverely compromises the functioning andcredibility of the SAPS. Internal corruption isdetrimental to the morale of police membersand causes the public to perceive the policeas being unable to provide an effectivepolicing service.22

Although the problem has been acknowledged bythe SAPS,23 the closure of the Anti-Corruption Unitin 2002 undermined public confidence in thepolice’s commitment to fighting corruption.Between 1996 and 2001 the SAPS’ Anti-CorruptionUnit (ACU) handled 20,779 allegations of policecorruption, 3,045 of which resulted in arrests,followed by 576 convictions.24

While the 2002/03 SAPS annual report providesdata related to corruption, the information isunclear and does not indicate whether efforts todeal with the problem are succeeding or not. Thereport states that for the years 2001 and 2002combined, there were 2,370 cases of corruptioninvestigated, of which 1,332 resulted in criminalprosecution and 641 in internal disciplinaryhearings.25 The report does not say what happened


to the remaining 397 cases. In the text, it states that872 police members were suspended as a result oftheir involvement in corruption, but theaccompanying graph shows that only 188 weresuspended under the category ‘corruption’.26

Furthermore, the SAPS annual report does notdefine what constitutes corruption. It separates arange of categories such as “assisting escapes”,“defeating the ends of justice”, “bribery” and“fraud” from the category “corruption”.27 Given theconfusing categorisation, the report is not entirelyclear on the number of members convicted ordismissed as a result of their involvement incorruption. What is worrying is that the availabledata reflecting the outcomes of disciplinary hearingsinto corruption shows that only 18 members weredismissed out of the 143 that were found guilty.28

While the SAPS annual report has improved withregard to the amount of information provided, theway it is presented is unclear. This makes it difficultto assess whether or not the initiatives to tacklecorruption in the SAPS are working.

Better oversight of police conductThe SAPS annual report is an important documentfor parliament in particular and the public ingeneral. It presents the key priorities and activities ofthe police, and during a period of organisationalreform, should demonstrate improvements onpreceeding years.

While the latest annual report is a significantimprovement on previous years, there are stillshortcomings. When it comes to key indicators andinformation on police conduct and abuse of poweras discussed in this article, substantial improvementsare still necessary.

It seems logical that departments will generally wantto report on their good performance rather than onissues that might cause embarrassment. However,the SAPS’ ability to report adequately on theseindicators will promote public confidence in thepolice. It will demonstrate that police managementtakes these issues seriously and is able to effectivelytackle them. Moreover, ensuring that policemembers are able to use appropriate force when

necessary, and that abuse of power can be quicklyidentified and effectively dealt with, will improvethe performance of, and public support for, thepolice.

Parliament, as the representative body of all SouthAfrican citizens, has the duty to ensure that theSAPS improve their internal systems of recording,monitoring, managing and reporting on aspects ofpolicing that are fundamental to the constitutionaldemocracy.

Endnotes1 Public Service Commission, Annual reports as an

accountability mechanism, in Review of Department’sAnnual Reports as an Accountability mechanism,2002.

2 SAPS misconduct ‘under-reported’, statement made by ICD to Parliament’s Safety and Security PortfolioCommittee, Dispatch Online, 25 April 2002.

3 K Adams, What we know about police use of force, in Use of force by police: Overview of national andlocal data, research report jointly published with theBureau of Justice Statistics, National Institute ofJustice, Washington DC, 1999.

4 See D Mistry, A Minnaar, J Redpath and Dlamini, The use of force by members of the SAPS: A case study ofseven policing areas in Gauteng, Institute for HumanRights and Criminal Justice Studies, Technikon SA,compiled for the ICD, 2001.

5 See D Bruce and O’Malley, In the line of duty? Shooting incidents reports and other indicators of thepolice use of force and abuse of force by members ofthe SAPS, compiled for the ICD, 2001.

6 Ibid.7 Chapter 2 of the Constitution of South Africa, 1996.8 Bruce and O’Malley, op cit.9 See 2002/2003 ICD Annual Report, pp 49–50.10 Ibid.11 See D Mistry et al, op cit.12 The SAPS anti-torture policy states that “No member

may torture any person, permit anyone else to do so,or tolerate the torture of another by anyone. The sameshall apply to an attempt to commit torture and to anact by any person which constitutes complicity orparticipation in torture. No exception, such as a stateof war or a threat of war, state of emergency, internalpolitical instability or any other public emergency willserve as justification for torture”.

13 P Pigou, Monitoring police violence and torture in South Africa, paper presented at the InternationalSeminar on Indicators and Diagnosis of HumanRights, April 2002 or visit<http://www.csvr.org.za/papers/papigou1.htm>

14 ICD annual report, 2002/2003, <http://www.icd.gov.za/reports/torture.htm>

MASUKU

15 South African Police Service Act 68 of 1995, Section 50; Section 10 of the Constitution of the Republic ofSouth Africa Act 108 of 1996; Standing Orders (G)341, 349, 350, 361, 362.

16 Visit <http://www.icd.gov.za/about/brochure.htm> for ICD categorisation of cases.

17 BD Bhana, Custody related deaths in Durban, South Africa (1998-2000), American Journal of ForensicMedicine and Pathology, 24(2), June 2003.

18 ICD 2002/2003 Annual Report, p 44.19 ICD, Recommendations pertaining to complaints

registry, 2001, <www.icd.gov.za/reports/registry.htm>20 See 2002/2003 SAPS Annual Report p 121.21 Country Corruption Assessment Report South Africa,

United Nations Office on Drugs and Crime RegionalOffice for Southern Africa, Department of PublicService and Administration, April 2003.

22 SAPS Annual Report 2002/2003, p 8.23 The SAPS has introduced several reporting systems for

its members and the public. Corruption can bereported via the SAPS Complaints Line (0860 11 1213), Community Service Line (0860 13 0860),Emergency Line (10111), through SAPS managementstructures (National, Provincial and AreaCommissioners), or in the case of SAPS members, viathe Crime Intelligence Corruption InformationManagement System.

24 SAPS Annual Report 2001/2002.25 SAPS Annual Report 2002/2003, p 10.26 Ibid.27 Ibid.28 Ibid, p 11, see graph 5 “Outcome of departmental

hearings”.

SA CRIME QUARTERLY No 8 JUNE 200410 MASUKU

The problems with this approach are manifold.There is little reason to believe that the socialservices departments have any better insight intowhy people commit crime, or how to change theirattitudes and behaviour, than do the police. Sincecrime prevention does not explicitly lie withinsocial service departmental agendas, performanceindicators are not geared to encourage theirparticipation in interdepartmental efforts. In theend, this approach blurs the line between crimeprevention interventions and the eternally recedinghorizon of ‘social regeneration’.

This has meant that crime prevention often fallsbetween the cracks. It is seen as a long term projectin an area where there is a blaring demand for shortterm results. When resources are allocated todealing with the problem of crime in society, theconcrete requirements of the police are generallymore compelling to lawmakers than ‘pie-in-the-sky’social development projects.

Chief among these police demands are funds forsalaries; salaries needed so they can do more ofwhat they know how to do. On the one hand, thisencompasses their vital reactive function –responding to calls and reports from the public, andinvestigating crime. On the other is what the police


Who is responsible for preventing crime?Most people would say “the police”.They would have the South African

Constitution to back them up, which lists “crimeprevention” as the first responsibility of the SouthAfrican Police Service (SAPS).

But there appears to be widespread consensus thatthe causes of crime are rooted deep in social andeconomic inequalities. This would indicate that thesolutions to crime problems must, in some way,address these inequalities. The police are ill-suitedto doing this. They are trained, rather, in theequally important task of maintaining social orderby reacting to crime incidents and other emergencysituations. They are not social engineers.

The South African approach to this dilemma hasgenerally been to make the SAPS the ‘lead agency’in coordinating inter-departmental crimeprevention efforts. The idea is that departmentssuch as Social Development, Education, and evenHealth and Housing may be able to contribute tochanging the social conditions that generate crime.While neither the police nor the other agencieshave the skills needed on their own, it is hopedthat as partners they will be able to come up withsolutions.

Ted Leggett, Institute for Security [email protected]

WHY WAIT?

By-laws and regulationsfor high impact crimeprevention

Reducing crime is not just about making arrests and convicting criminals. The social and economic inequalities

that cause crime require ‘crime prevention’ measures that can take years to show any results. But there is an

alternative. This article argues for locally based interventions that can change social behaviour in the short term

and have an immediate impact on safety and security. By-laws, for example, can be used to target those with

something to lose and to regulate the ‘free-for-all’ environment that grips many of our inner cities.

SA CRIME QUARTERLY No 8 JUNE 200412 LEGGETT

call ‘crime prevention’, which seems to be any‘proactive’ work that is not motivated by a specificcall for assistance from the public. This generallyincludes visible patrols, raids of suspect buildings,roadblocks on routes used for getaways bycriminals, and cordon and search operations inhigh crime areas.

The thinking behind this is that crooks can bescared into behaving by the sight of a blue uniform.Those who aren’t, can be incapacitated by lockingthem behind bars for long periods of time. This willscare their friends into behaving too, because theywill see that government is serious about crime. Nomatter how poorly raised or economically needy,everyone can be expected to respond to fear.

While the above portrayal verges on caricature, itdoes set out a live dilemma. On the one hand, wetrust the police to protect us from crime. On theother, we recognise that they are ill equipped todeal with its social causes. This article suggests away around this impasse – a way of changing socialconditions in the short term. Since most crimeproblems are local problems, the key is the kind oflocal law designed to regulate social conditions: theby-laws.

Crime prevention in South AfricaThe 1998 White Paper on Safety and Securitydefines “crime prevention” as:

All activities which reduce, deter, or preventthe occurrence of specific crimes, firstly, byaltering the environment in which theyoccur, secondly by changing the conditionsthat are thought to cause them, and thirdlyby providing a strong deterrent in the formof an effective Criminal Justice System.1

The document goes on to differentiate between thetypes of activities described in the first two pointsand that of the third, drawing the distinctionbetween “social crime prevention” on the one handand “crime prevention through effective criminaljustice” on the other. This dichotomy has persistedthroughout the discourse on crime prevention inSouth Africa, with social crime prevention beingdescribed as a long term process, and law

enforcement-based crime prevention as a short termoption.

For example, the National Crime CombatingStrategy (NCCS), the present operational strategy ofthe police, is broken into two phases. Phase one,which was to have been conducted between 2000and 2003, has been nicknamed “OperationCrackdown”.2 In an attempt to ‘stabilise’ escalatingcrime levels in the 145 station areas that produce50% of the crime in the country, joint police andmilitary operations were launched, involvingsaturation patrols, building searches, roadblocks,and cordon and search operations. In theseoperations, a ‘zero tolerance’ approach was taken,and a massive number of arrests were made for awide range of charges.

In contrast, phase two, which is supposed to runfrom 2004 to 2009, is designed to ‘normalise’ crimelevels through interventions aimed at addressing thecauses of crime. It would appear that the SAPS’primary mechanism for accomplishing this will besector policing, a geographically-focused form ofcommunity policing in which the police engagewith community members to identify and solvepersistent crime problems.

Thus, the first phase was intended to have a quickimpact on the crime figures through lawenforcement, while the second is intended toaddress the causes of crime over a longer period oftime. What is actually going to be done to stopcrime in phase two has not yet been determined: itis hoped that together, the police and thecommunity will be able to come up with solutionsappropriate for the particular localities in which thesectors are established.

The two-phase approach has allowed the police toapply their existing skills for three years and put offdealing with the issue they know is actually beyondtheir scope: addressing the social causes of crime.

An alternative approachIs it possible to change social circumstances in theshort term? One aspect of social reality can bechanged immediately: the law. Some would arguethat legal change is irrelevant without


commensurate enforcement capacity, but SouthAfrica provides many examples to the contrary.

Anti-smoking laws, which many believed would be‘unenforceable’, resulted in major renovationsamong many restaurant chains. While the policemay be too busy to arrest incorrigible tobaccoaddicts, the law has provided the basis for moreinformal types of social coercion to be employed.Non-smokers now have a legal basis on which tochallenge those violating the law, although this ismore likely to be done with a glare than with a tripto the courthouse. The net result is a world lessfriendly to cigarettes, which is likely to be ahealthier world.

Looking at another example, employment equityinspectors may be few in number, but theconsequences of being caught out are just tooterrible for most major concerns to contemplate.The law provides the moral authority for thepreviously disadvantaged to challenge unfair hiringdecisions and to demand corrective action. Withthe dash of a pen in Cape Town, workcircumstances countrywide were changed forever.

Similarly, the law requiring that domestic workersbe registered for unemployment insurance has seenwidespread compliance by thousands of privateindividuals motivated more by a genuine desire tocomply with the law than by fear of prosecution.Domestic workers, who are the consummateexample of workforce members with little coercivebargaining power, have been alerted to their rightsthrough the media and word of mouth, despite thefact that many are illiterate.

What these examples have in common is thechallenging of people who cannot afford to be, ordo not desire to be, on the wrong side of the law.There may be crime problems that can bedispatched just as expeditiously if similar playerscan be identified whose decisions resonate at streetlevel.

The biggest problem with deterrence theory is thatmany, if not most, people who engage in criminalacts are, in fact, undeterrable. Most acts of violentcrime are committed in the heat of the moment,

when the possibility of incarceration is utterlyirrelevant. This ‘heat’ is turned up when alcohol ordrugs are involved. And there are people in anysociety whose lives are lived ‘in the heat of themoment’. What would deter the legislators whocraft our criminal codes becomes just another partof the drama of lives lived under an entirelydifferent set of values.

The trick, then, is to find a way of changing thechoices faced by these undeterrables by targetingthe actions of those with something to lose.

Targeting those with something to loseSouth Africa is different from many developedcountries in that it is possible to live almost entirelyoutside the ambit of the law. People can and dobuild un-inspected homes on property they do notown, draw free water from untreated sources, eatun-regulated food, and dispose of their waste inunauthorised ways. It is very difficult to touch thesepeople by changing the law, since they owe theirway of life to ignoring it.

But there are large groups of people who defy thelaw while simultaneously enjoying its protection.This is most obvious in urban contexts, whichcomprise some of the most notorious crimehotspots in the country. These people owe theirlifestyle not to the absence of the law, but to thefact that it is not enforced. They rely on unregulatedenvironments.

It is not a coincidence that crime and grime gotogether, but neither causes the other. Crime canonly thrive when people don’t give a damnanymore, a sentiment that is most pungentlymanifest in neglect of basic hygiene. Squalordoesn’t generate crime. Rather, the two are bothsymptomatic of the same disregard for the value ofpublic order that permeates localities the stateneglects.

Inner-city areas pose many lifestyle advantages tothe urban criminal – advantages most would beloathe to give up. The desire to make a crookedbuck does not automatically imply an ignorance ofthe advantages of indoor plumbing. And especiallyfor criminal businesses, it can be difficult to attract

LEGGETT

a moneyed clientele to locations without access topaved roads and other amenities the buying publictends to take for granted.

The high-rise environments that characterise inner-city areas do not spring up on their own accord.They are comprised of buildings owned by people.These people have something to lose: theirbuildings. So while the criminals on the city streetsmay not have much law enforcers can threaten, thepeople who control their living environments do.

These people do control the little bit the criminalshave to lose: their relatively comfortable inner-cityliving arrangements, and the indoor component oftheir playing field. Building owners control theircrooked tenants’ access to indoor plumbing. Andwhile the threat of prison may loom too remotelyfor some to take notice, there are few things moreimmediate than needing a place to relieve oneself.

Accountable environmentsInner city crime can only ferment in dimly litplaces. Criminal enterprises require housing wherenobody asks too many questions. If theseenvironments were to suddenly become regulated,criminals could literally face eviction from theircosy, if somewhat dingy, cocoons.

Several basic tools are found in the by-laws.Municipalities are allowed to regulate in the areasof health and safety. They can set down basicbusiness and licensing requirements. They cancollect rates and taxes. They can zone.

Let’s apply these to your basic inner-city area.Imagine a by-law that requires all owners of rentalproperty (including hotels) to keep copies of theidentity documents of residents. This would ensurethat all foreign residents keep their visas up to date,that all runaway sex workers can be traced by theirfamilies, and that the next time there is a shoot-outin the hallway, the host can match the photocopiesto the bodies. They might even be able to point outthose who hastily relocate after the incident.

Failure to maintain these records should result in afrightening and escalating series of fines – fines thatcan pay the salaries of inspectors and may, in the


case of repeat offences, exceed the value of thebuilding. If this is the case, the building can beattached in settlement.

Municipalities may presently be loath to takeownership of crumbling tracts of residential realestate, but there are law-abiding South Africans whoneed roofs over their heads. These South Africanshave land-reallocation grants that can be used torefurbish the buildings. As proud owners of newhomes, they comprise a class of people who mostdefinitely do give a damn, and this is the strongestbulwark against backsliding into the anarchy thatprevailed before.

Suddenly, criminal fugitives may have a hard timefinding a place to hang their hat in the slums theyonce called their own. They may have to invest in atin roof, plastic sheeting, and cardboard. They mayeven have to commute. Whatever their response,there will be new points of vulnerability to exploit,new environments ripe to be regulated.

Guns and boozeWhile having a gun might not make you a killer, itdoes make you a lot more dangerous when you loseyour temper. And if you have had a few beforetaking aim, chances are a simple kneecappingcould turn into something far more serious. Gunsand booze may not cause crime, but theirproliferation does aggravate the situation.

Firearms and alcohol are two legal commodities,and are thus subject to regulation. In theory, this isdone at national or provincial level, but clearly itcould use some local tweaking. For example, justbecause it is legal to own a firearm does not meanthat the city needs to allow people to carry theirweapons on their persons. ‘Check your guns at thecity lines’ has a Wild West sound to it, but it couldresult in lives being saved in scuffles thataccidentally become slayings.

All this is legally contentious, of course. Nationallegislation and constitutional rights to property andmovement will be invoked. But there are creativeways around the problem. A municipal tax could beinstituted, with rebates or exceptions for those whodeclare their buildings firearm free zones, for


example. With dedication and perhaps a littlelitigation, municipalities should be able to takeextra measures to ensure the safety of their streets.Moral authority is on their side.

Similarly, the right to a liquor license seems to havebecome second only to the right to vote in the newSouth Africa. But just because you have the piece ofpaper does not give you the right to open shopwherever you choose. Zoning is a powerful tool formunicipalities to use in relegating undesirableactivities to a well-watched basket. Like with thesmoking regulations, this is unlikely to reform manyalcoholics, but it may make bar-hoppinginconvenient enough to interrupt a few binges.

In the short term, this is likely to result in the furtherproliferation of unlicensed premises, but a similarapproach must be taken. Most shebeens are housedinside buildings with owners. If nothing else, by-laws could require confiscation of liquor stocks,which can be sold to further fund enforcement,such as rewards for those who provide evidenceagainst unlicensed vendors.

In the United States, owners of bars have beensubject to various state ‘dram shop laws’ that holdthem accountable for any damage done by peoplethey have served while visibly intoxicated. Theprimary witness for the state in these cases is oftenthe drunk himself. This follows the same principle ofdeterring those who have something to lose.

Of course, drunk drivers have vehicles andmunicipalities have the right to enforce trafficregulations. As with inner city buildings, thisproperty could be made directly forfeit onconviction, or the fines could be stiff enough thatthey exceed the value of the car. If appliedaggressively, the city police should never suffer forlack of transport.

Follow the moneyPolice in Hillbrow think that one major factorbehind the remarkably high rate of robbery in thearea3 is that many of the residents are foreignnationals whose permits for being inside the countrymight not be entirely in order. Many of theseforeigners engage in street trading or other informal

enterprises, which means they often carry a lot ofcash, but they lack the identification documentswith which to open a bank account. Their homesare even more insecure than their pockets, and thelocal crooks know this.

As the police in the area have suggested, one rapidway of regulating the finances of thesequintessentially unregulated people would be toprovide keys for secure rental strongboxes at anominal fee, with no identification required. Untilsuch time as residential control makes the inner citya less attractive destination for illegal immigrants,these devices could allow a safe stash for one’s lifesavings. The demand for such boxes may quicklygive us an idea of the scale of foreign commercialactivity in our inner city areas.

Increasingly drawing marginalised people into moreregulated lifestyles – whether they be illegalimmigrants or local street sex workers – couldprovide a possible bridge into the mainstream,while simultaneously minimising the harms sufferedboth by the individuals and by the society at large.These people, who are not malicious criminals, arebest addressed with a carrot instead of a stick, asthe hardships they face daily are often heftier thanthe stoutest knobkerrie.

Not about zero toleranceProponents of the crime and grime link boast anarray of macho-sounding approaches to ‘takingback the streets’, many of which are ostensiblyrooted in New York’s zero tolerance experience.Most of these have to do with enforcement of lawsrooted in social norms, such as the prohibition ofpublic drunkenness or lewdness, drinking in public,and even jaywalking. The idea is to ‘send a signal’that people do give a damn and that deviance willnot be tolerated. It is rooted in the notion thatmoral decay can be rolled back by enforcing decentconduct and respect for the laws, however trivial.

While this approach may have some utility in areaswhere the majority of the people still subscribe to acommon set of norms, it is most emphatically notwhat is being argued here. The potential of by-lawenforcement lies in the realm of market disruptionfar more than that of moral regeneration. It is about

LEGGETT

Changing the rulesOur inner cities suffer for our neglect. The peoplewho live there are victims, even if some are alsoperpetrators. It is time these unregulatedenvironments be shown the light of day. Many moreinterventions could be added to the suggestionsabove, but they would all conform to the samecentral idea: social circumstances can be changed,locally, today.

It is possible that similar measures could be taken inareas other than the inner city, like peri-urbaninformal, and even rural, areas. Although theenvironment and types of crime problems maydiffer, the principles should be the same. Regulatethose areas that have been forgotten. Deter thosewho have something to lose. Lead those who havebecome marginalised back into the mainstream.

Endnotes1 <http://www.info.gov.za/whitepaper/1998/safety.htm>2 In fact, this term only applies to one component of the

overall strategy.3 My 2002 survey of 1,100 households in the Hillbrow

and Johannesburg Central station areas found that 30%of the respondents said they had been robbed in the lastyear alone. See ISS Monograph 78 Rainbow Tenement:Crime and policing in Inner Johannesburg.

4 The work of the Hatfield Magistrate’s Court may provide a good example of the potential this route contains.


functionally disabling the infrastructure on whichurban criminals rely – not scaring skid rowalcoholics into keeping their zippers up andcrossing at the robot.

The so-called ‘zero tolerance’ school of thoughtposits massive criminal justice resources to ensureblanket police coverage and plentiful jailhouseaccommodation to receive their work product.What is being suggested here requires neither.

The enforcersGiven the case loads endured by the police in thiscountry, along with the endless stream of seriouscrimes like murder, rape and robbery, is there thecapacity to enforce these minor local laws?Surprisingly, most people forget that this is actuallythe primary responsibility of the municipal police.

At present, the municipal police in many areasseem to be focusing either on acting as a forcemultiplier for the SAPS or on ‘business as usual’ –traffic enforcement. This should change, and wouldif city management got a sense of the potential ofby-law enforcement.

But while the municipal police are likely to do mostof the heavy lifting, who should be driving theprocess? There are many possible options. Since theprocess may involve the drafting and passage ofnew laws, it makes sense that someone in the locallegislature be involved. But it is also important thatsomeone with clout over a range of executivedepartments be included. The city manager is anobvious choice, but there are many others. Thelocal deputy director of public prosecutions is apossibility,4 as is the local municipal police chief.

Since no single individual is likely to have directcontrol over all the forces that must be marshalled,we are likely to find ourselves squarely back in the‘crime prevention by committee’ dilemma. But thiscan be avoided by a single, well-placed individualtaking the reigns and cultivating a series of bi-lateralrelationships with the pivotal people. In the end,the types of interventions available may be limitedby local personalities, but this simply means thatthe champion of the cause may need to be a littlemore creative.


Duxita Mistry, Institute for Security [email protected]

FALLING CRIME,RISING FEAR

2003 National Victimsof Crime Survey

For several years, the police have maintained that crime levels in South Africa are ‘stabilising’. Without

alternative sources of crime statistics, it is impossible to test these claims. The most reliable supplements to

police data are national victim surveys, which are now conducted regularly in several countries for precisely

this purpose. The 2003 National Victims of Crime Survey shows that crime levels, as measured by the surveys,

have indeed declined since 1998. Public sentiment does not reflect this good news however – feelings of safety

are much worse now than they were five years ago.

In 2003, the Institute for Security Studies (ISS)conducted a national victim survey with the aimof measuring crime trends in the country, public

perceptions about crime and safety, as well asconfidence in the criminal justice system. The studywas planned and carried out to allow directcomparisons with the national survey conducted in1998 by Statistics SA for the Department of Safetyand Security and the United Nations InterregionalCrime and Justice Research Institute (UNICRI).

The survey was conducted between September andOctober 2003. Households were randomly selectedacross the country based on the 2001 Census, and anational sample of 4,860 people, over the age of 16years, was realised. The sample was stratified byprovince and urban/rural areas, and the data wasweighted to reflect the actual composition of thepopulation.

In a nutshell, the findings revealed a drop in crimelevels since 1998, although accompanied by rising,and high, levels of public insecurity. In particular,people were most afraid of violent crimes such asmurder and sexual assault. This article provides anoverview of some of the results, and explores the

relationship between fear of crime and actualexperiences of victimisation.1

Why the need for a national victim survey?National victim surveys provide invaluableinformation on victimisation rates and vulnerablegroups, because they focus on the victims of crime(rather than the perpetrators as is the case withpolice and court data), and cover a representativesample of the population in a specific geographicarea. The surveys also provide an understanding ofpublic perceptions of crime and safety, and the fearof crime, as well as victims’ actual experiences ofspecific types of crime. They also offer insight intothe underreporting of crime, or the so-called ‘darkfigure’ that describes the incidents that do not makeit into police records. As such, victim surveyscomplement police statistics by adding to theinformation that is already on the official database.

Victim surveys have a distinct advantage in that theyshow the extent of multiple victimisation andwhether crimes are concentrated in a small numberof people who are frequent victims, or are spreadout among the general population. Disadvantages ofvictim surveys are that respondents do not always

Table 1: % of South Africans, over the age of 16 years, who were victims of crime in

1998 and 2003

1998 2003

Any crime 24.5 22.9

Housebreaking 7.2 7.5

Corruption* - 5.6

Theft of personal property 4.8 4.7

Stock theft 4.9 2.5

Theft out of vehicle 2.5 2.5

Assault 4.2 2.2

Robbery 2.4 2.0

Deliberate damage to vehicle 1.3 1.3

Bicycle theft* - 1.2

Car theft 1.2 1.0

Deliberate damage to buildings 1.1 0.9

Fraud 3.0 0.8

Crop theft* - 0.7

Car hijacking** 1.4 0.5

Other crime 1.6 0.2

Murder 0.5 0.2

Theft of motorbike 0.0 0.1

Sexual assault/rape 0.4 0.1

SA CRIME QUARTERLY No 8 JUNE 200418 MISTRY

some extent. A more dramatic decrease is evidentfor crimes like stock theft, assault, and fraud.

The levels of sexual offences/rape as measured inboth surveys should be treated with caution asthese crimes are underreported in victim surveys aswell as official police statistics. The apparentdecline in sexual offences from 0.4% in 1998 to0.1% in 2003 is therefore unlikely to reflect the realtrend, as data on these crimes is not consideredreliable in the survey context.

recall precise details of their experiences, the resultsare subject to sample error, and the surveys are apoor means of collecting data on crimes such asrape, domestic violence, fraud and corruption,because members of the public are often reluctantto discuss such matters with survey fieldworkers.General victim surveys also do not record crimesagainst businesses, crimes against children, anddrug and firearm related offences.

Nevertheless, the value of these studies has beenrecognised by governments in several developedcountries such as the United Kingdom and theUnited States, where victim surveys are nowconducted annually to supplement police statistics.Together with the official crime data, the surveyresults are key for crime prevention, policing andjustice policy formulation, for identifying gaps inresource allocation, as well as improving victimsupport services.

The good news: crime rates down since 1998 In order to gain the maximum benefit from nationalvictim surveys, similar studies must be conducted atregular intervals. A comparison between the 2003and 1998 surveys shows whether or notvictimisation rates have increased, and how levelsof fear of crime may have changed over time.

In the latest survey, just more than one fifth (22.9%)of all South Africans had been a victim of crime inthe 12 months preceding the survey. This is slightlyless than the overall victimisation rate recorded bythe 1998 survey, in which one quarter (24.5%) ofSouth Africans had experienced crime over thepreceding year. This means that overall, thevictimisation rate dropped by 1.6% over the pastfive years.

It follows that most of the crime types measured inthe surveys would also show a decrease between1998 and 2003. The one exception to this trend ishousebreaking, which is the only category of crimein the survey that increased, albeit fractionally,since 1998 (Table 1). Rates of theft out of motorvehicles and deliberate damage to motor vehiclesremained the same, while other offences such astheft of personal property, car theft, deliberatedamage to buildings and robbery have decreased to

* crime types not covered in the 1998 survey** In the 1998 survey the category ‘car

hijackings’ included attempted and‘successful’ hijackings, while in the 2003survey only successful hijackings wererecorded. This could account for the decreasein the hijacking rate between 1998 and 2003reflected here.

they did in 1998, if the “very safe” and “fairly safe”categories are added together (Figure 1). However,significantly more felt only fairly safe in 2003 asopposed to very safe in 1998. The tendencytowards feeling less safe becomes a clear trendwhen the night-time results are considered. SouthAfricans felt significantly less safe when walkingalone after dark in 2003 than they did five yearsago (Figure 2). In fact, more than double thenumber of people in 2003 than in 1998 felt veryunsafe walking in their area after dark (58% in2003 as opposed to 25% in 1998).

Another indicator of public concern about crimerelates to views about how the crime level haschanged. Despite the decline in the crime rate,more than half of South Africans (53%) felt thatcrime has increased over the past three years in theareas where they live. These views were particularlyprevalent among people in metropolitan and urbanareas, and among Indian and white South Africans.In general, more people believed that propertycrime had increased (55%) than those who thoughtviolent crime had gone up (47%).

In terms of the ordering of crime types by theirprevalence, little has changed since 1998. The mostprevalent crimes five years ago were housebreaking,theft of livestock, theft of personal property, assault,fraud, theft out of motor vehicle and robbery. In2003, the same crimes were among the top sevenmost prevalent offences, with the exception offraud. In both years, property crimes occurred morefrequently than violent crimes.

A notable finding in the 2003 survey was that asignificant proportion of South Africans (5.6%)reported being asked by a government official for abribe in the form of money, a favour or a present inreturn for a service that the official was legallyrequired to perform. This suggests that pettycorruption was the second most prevalent crimetype in the country.

The bad news: less people feel safe Despite the decline in crime rates indicated by thevictim surveys and the official crime statistics, SouthAfricans feel less safe in 2003 than they did in1998. Perceptions of safety can be measured invarious ways, one of which is used internationally,2

and asks survey respondents how safe they feelwhen walking alone in their area during the dayand after dark.

In 2003, 85% of South Africans said they feel safewalking alone in their area during the day, whileonly 23% felt safe walking alone at night. On thepositive side, the percentage feeling safe at night ishigher than that recorded in other comparable site-based victim surveys in the country, including CatoManor, Hillbrow/Inner Johannesburg, Cato Crest orMeadowlands.3 However, significantly less SouthAfricans felt safe walking in their area at night thanthose surveyed in other developing countries. TheICVS found that on average, 60% of those surveyedin African countries, 56% in Latin Americancountries and 55% in Asian countries said they feltsafe walking in their areas after dark. In South Africaonly 23% said the same.4

Of more concern than the internationalcomparisons, is that South Africans are much morefearful now than they were five years ago. Duringthe day, the public felt generally as safe in 2003 as

SA CRIME QUARTERLY No 8 JUNE 2004 19MISTRY

1998

Veryunsafe

Fairly safe

Figure 1: Respondents’ feelings of safety when walking alone in their area during the day,

1998 and 2003

Source: ISS National Victims of Crime Survey,2003

0

30

50

70

60

40

20

10

Very safe

%

2003

25

60 60

25

10 95 6

A bit unsafe

The survey results show that the risk of becoming avictim of property crime was greater than violentcrime, and that people were more inclined to thinkproperty crime had increased than violent crime.Despite this, five of the top six crimes that SouthAfricans were most afraid of, were violent, withmurder topping the list even though it was among theleast prevalent of the crimes investigated (Figure 3).

It is possible that more sensational crimes such asmurder and rape have a greater impact onperceptions, and are more intensely covered in themedia. Moreover, how the police are believed todeal with particular crimes could contribute to publicconcerns about them. In general, these viewsindicate the types of crime that respondents thoughtthey were most susceptible to, as well as theirconcerns about the impact of the offences. Althoughcrime has levelled off since 1998, the results indicatethat violence remains the key challenge as far as thepublic is concerned.

Although fears about certain crimes did not matchthe risk of actually becoming a victim, public views


about which crimes are most common were acloser match to reality. When asked “what one typeof crime occurs most in your area?”, respondentswere most likely to say housebreaking (38%),followed by robbery (14%), theft of property (10%),murder (7%), stock theft, bag snatching, and assault(all 6%), rape (4%), car theft (3%) and hijacking(2%). The order of crimes believed to be mostcommon was fairly similar to the actualvictimisation rates (Table 1), with the exception ofmurder and robbery, whose prevalence wasoverestimated by respondents.

Impressions of police performanceRespondents were asked about their physical accessto the police, whether they had actually been to thenearest police station, and how they rated theperformance of the police in their area.

Access to the police was generally good: almost allSouth Africans (97%) knew where their nearestpolice station is, and two thirds were able to reachthe police station within 30 minutes or less usingtheir usual mode of transport.5 This should improve

1998

Veryunsafe

Fairly safe

Figure 2: Respondents’ feelings of safety when walking alone in their area after dark,

1998 and 2003



0

30

50

70

60

40

20

10

Very safe

%

2003

10

32

13

24

19 19

58

25

A bit unsafe

0

Not afraid

5 10 15 20 25 30

Livestock theft

Pick pocketing/bag snatching

Theft of property

Car hijacking

Assault

Robbery

Sexualassault/rape

House breaking

Murder

%

Figure 3: The one type of crime thatrespondents were most afraid of in the area

where they live

25

23

19

13

5

4

4

3

2

1


doubt informs people’s opinions on the fear ofcrime and their reliance on the police forassistance.

An important issue related to police visibility andperformance is the level of reporting of crime byvictims. High reporting rates reflect, among otherthings, levels of public confidence in the police.The reporting rate also gives an indication of howmany crimes are never registered in the police’sofficial database. Generally, serious property andviolent crimes are reported, while offencesregarded as petty (such as pick-pocketing), that maycause embarrassment to victims when reporting(such as rape), or are believed to be a matter for theparties concerned and not the state (such asdomestic violence) are often not reported to thepolice.

This pattern of reporting was found to be true forthe most prevalent crimes recorded in the 2003survey, with the exception of robbery. The vastmajority of car theft victims reported the crime tothe police, no doubt for insurance purposes (Figure4). Similarly, a smaller majority reported theft out ofvehicles, and housebreaking. The reporting rate for

the chances that victims will report crime to thepolice. Unsurprisingly, those living in the highlyurbanised Gauteng and Western Cape provinceswere closest to police stations, while those in ruralLimpopo and Eastern Cape had to travel the furthestto reach their local station.

Just under half (46%) of the respondents indicatedthat they had visited their nearest police station inthe last three years. Given that those with first handexperience of dealing with the police are betterplaced to articulate their views on policeperformance, the opinions of these respondents areimportant. Of those who had been in contact withthe police, more than half (56%) said it hadchanged their opinion of the police, and of these,54% said their opinion had improved. A little morethan one tenth (12%) claimed their opinionremained unchanged, while just over one third(35%) said it had made their opinion worse.

Perceptions of police performance were also testedin a question to all respondents about how theythink the police are doing in their area of residence.Just over half (52%) of South Africans said thepolice were doing a good job in their area, whilemore than two out of five (45%) thought they weredoing a bad job. The main reasons cited for why thepolice are doing a good job were their commitment(25%), that they arrest criminals (24%), respond ontime (23%) and come to the scene of a crime(15%). The main reasons for saying the police aredoing a bad job included that they do not respondon time (35%), are corrupt (13%), don’t come intothe respondent’s area (12%) and are lazy (11%). Theimportance of police response times in both sets ofreasons indicates an area that is directly within thecontrol of the police that could be worked on toimprove public perceptions.

Police visibility is also an important factor regardingperceptions of safety. When asked how often theysee the police on duty and in uniform in their area,respondents were most likely (29%) to say they seea police officer at least once a day. Just more than aquarter (25%) said they see the police at least oncea week. A major cause for concern is that one fifth(21%) reported that they “never” saw a policeofficer on duty in their area of residence. This no

MISTRY


0

Robbery

20 40 60 80 100 120

Livestock theft

Damage tomotor vehicles

Theft

Bicycle theft

Assault

Housebreaking

Theft out ofmotor vehicle

Car theft

% of victims who reported

Figure 4: Reporting to the police for crimes with rates >1%

97

64

57

55

44

41

36

36

29

10

assault is fairly high, considering that this is a crimethat is often regarded as not important enough tobother reporting, or not a matter for the police toresolve. In the case of robbery however, the fact thatonly 29% of victims reported the offence isworrying, particularly considering that most of therobberies recorded in the survey were armedrobberies and thus of a serious nature.

Reporting rates have nevertheless improved since1998 for several of the more prevalent crimes, againwith the exception of robbery (Figure 5). The resultssuggest that confidence in the police as measured byreporting rates has grown in the past five years. It ishowever important to bear in mind that victims’decisions about whether or not to report are basedon a range of factors, some of which are not directlyrelated to policing, such as the view that it isunnecessary to report, or the fear that the perpetratorwill take revenge on the victim if he or she reports.

Views of the courtsRespondents were asked a similar set of questionsabout their physical access to the courts, as well astheir views of court performance. As in the case of


the police, access to the courts was generally good:more than two thirds (84%) of South Africans knewwhere the nearest magistrate’s court is located andjust over half (51%) said they can get to the courtwithin 30 minutes or less using their usual mode oftransport. Access was better in the more urbanisedprovinces. Respondents living in the Eastern Capeand Limpopo were most likely to have to travel longdistances, while those in Gauteng and WesternCape travelled for the shortest time.

On the whole, slightly more South Africans (59%)felt the courts were performing their dutiesadequately than the 52% who said the police weredoing a good job. Levels of satisfaction with thecourts were even higher among those who haddirect experience with the court system: of the onefifth (22%) who had been to court in the last threeyears, most (70%) were happy with the serviceprovided by the state prosecutor/state advocate. Asimilar majority (71%) was happy with themagistrate or judge that presided over the case.

All respondents, regardless of whether they hadbeen to court or not, were asked whether they weresatisfied with the way courts generally deal withperpetrators of crime. Just over half (51%) said theywere, with almost as many (45%) expressing theirdissatisfaction. The main reasons given for beingsatisfied were that the courts pass appropriatesentences (60%), have high conviction rates (22%)and are not corrupt (17%). Dissatisfaction centredon the courts being too lenient (34%), releasingperpetrators unconditionally (32%), not enoughconvictions (16%), and matters dragging on for toolong (14%).

These results indicate that sentencing was the mainissue about which the public formed their opinions,both positive and negative, of the way courts dealwith suspects. It is also revealing that the secondmost common reason for criticising courtperformance was that perpetrators are released“unconditionally”. This suggests that the public donot understand the bail and sentencing processes.

Views about perpetrators of crimeThe above results suggest that the public favour stiffsentences for perpetrators. But who do South

Source: ISS National Victims of Crime Survey,2003 and Stats SA, 1998

0 20 7030 40 50 60

Robbery

% of victims reporting to police

Figure 5: Comparative reporting rates, 2003 and 1998, selected crimes

41

19982003

29

Theft of personalproperty

2841

Assault 3855

Housebreaking 5757

Theft out ofvehicle

5764

Africans think is responsible for committing mostcrime, and what are the motivations of thesecriminals believed to be? Public opinion on theissue is likely to be informed, considering the highnumber of people who know someone in their areawho makes a living from crime: 29% of respondentsadmitted to this, which is not that surprising giventhat crime rates are relatively high. Respondentswere further asked about the residency and origin ofthe perpetrators. The responses clearly indicate thatcontrary to popular opinion, the vast majority ofSouth Africans believe that people born in SouthAfrica are responsible for most crime. Only 4%thought that most crime was committed byforeigners. Respondents were also of the view thatmost violent and property crime is carried out bypeople who live in their area, rather than by‘outsiders’.

When asked about perpetrators’ motivation forcommitting crime, the most frequent answers forboth property and violent crimes were “greed” and“non-financial motives” as opposed to “real need”.Real need was however almost as common anexplanation for property crime as the other reasons(Figure 6). Although a common perception is thatcrime is caused by poverty, these results suggest thatthe public think otherwise.

No matter what the motives for crime were believedto be, most South Africans said developmentalsolutions are most important for solving theproblem. When asked which one of three options(crime prevention and law enforcement includingmore police; the judiciary and courts includingharsher sentences, punishment and prisons; andsocial development including job creation)government should spend money on to reducecrime in their area, most South Africans opted forsocial development. A further one quarter saidmoney should be spent on crime prevention andlaw enforcement, with the remainder identifying thejudiciary and courts as important (Figure 7).

ConclusionThe results of South Africa’s second national victimsurvey, as well as the police statistics, show thatcrime rates have either decreased or levelled offover the last five years. However, according to the


Violent

Figure 6: Views on what motivates mostperpetrators of property and violent crime



0

30

40

20

10

25

45

35

15

5

Greed%

Property

3639

Non-financialmotives

34

43

Real need

30

18

Violent crime

Figure 7: ‘Which one of the following shouldgovernment spend money on to make your

area safe from crime?’

0 10 20 30 40 50 60 70

Socialdevelopment

(createemployment)

%

Property crime

63

50

Crimeprevention

and lawenforcement(more police

etc)

23

24

The courts and prisons

(punishment, harsher

penalties)

15

26

MISTRY

victim survey the public’s fear of crime hassimultaneously increased. This counter intuitivetrend may be explained by a number of factors suchas increasing public awareness of other people’svictimisation and the high level of violence thattypifies some criminality. However, more research isrequired in order to understand the complexdynamic between the increasing fear of crime anddecreasing crime rates.6

Endnotes1 The full report of the survey results will be published

shortly. See P Burton, T Leggett, A Louw, D Mistry andH van Vuuren, National Victims of Crime Survey: SouthAfrica 2003, ISS Monograph Series, July 2004, Institutefor Security Studies, Pretoria, forthcoming.

2 This question is asked in the International Crime Victim Surveys (ICVS) conducted by the United NationsInterregional Crime and Justice Research Institute(UNICRI) over the past 15 years in 24 industrialisedcountries and 46 ‘countries in transition’.

3 See T Leggett, Rainbow Tenement: Crime and policing in inner Johannesburg, ISS Monograph Series, No 71,ISS, Pretoria, April 2003.

4 The ICVS results are reported in A A del Frate and van Kesteren, The ICVS in the developing world,International Journal of Comparative Criminology, 2(1),de Sitter Publications, 2003, pp 57-76.

5 Since the 1997/1998 financial year 28 police stations, 13 satellite stations and 9 contact points have beenestablished (figures obtained from SAPS EfficiencyServices, March 2004).

6 A series of focus group discussions will be undertaken in due course by the ISS.



Prohibiting ‘abusers’ from entering the commonhomeAlthough section 7(1)(c) and 7(2)3 of the DVAprovides that magistrates may grant an order thatprohibits the respondent from entering his/herresidence that is shared with the complainant – or apart of that residence – magistrates are reluctant toenforce this provision. Many believe that thisdecision, when the respondent is legally entitled byownership or tenancy to occupy the home, is anextremely sensitive one. Some magistrates arguethat it is tantamount to eviction, which is thejurisdiction of the High Court.

Although the Act refers specifically to “prohibitingentry into a shared residence”, there is some debateas to whether this prohibition results in the actualremoval of the respondent or a temporary restrictionfrom entering the residence. Opinions aboutremoval vary greatly among magistrates, who raisethe following important questions about theenforcement of this provision:• Does section 7 of the DVA specifically intend to

give the right to remove or temporarily evict therespondent from the shared residence?

Lillian Artz, Gender, Health & Justice Research Unit University of Cape [email protected]

TOUGH CHOICES

Difficulties facingmagistrates inapplying ProtectionOrders1

The second in a series of articles on the Domestic Violence Act considers some of the most difficult issues that

magistrates must decide on. These include the temporary removal of the ‘abuser’ from the common home,

emergency monetary relief for ‘victims’, and orders specifying the terms of contact with children. Magistrates’

opinions on these controversial issues vary greatly, with the result that victims get uneven assistance from the

courts. Magistrates, however, argue that the variation of opinion reflects their independence and discretion, as

well as the various capacities of the lower courts to implement the Act.

In the first part of the series on magistrates andthe Domestic Violence Act (DVA), magistrates’opinions about the general substance and

workability of the Act were discussed (see SA CrimeQuarterly No. 7, March 2004). The second part ofthe series considers magistrates’ interpretation ofmore controversial issues such as the temporaryremoval of the abusers (‘respondents’) from thecommon home, emergency monetary relief, andorders specifying the terms of contact withchildren.2

These issues have been described as the mostdifficult ones for magistrates to decide upon. Thecomplexity of granting orders that prohibit anabuser’s access to his or her residence and/orchildren, and the obvious problems associated withforcing the respondent to pay for the victim’s(‘complainants’) expenses, is not to beunderestimated for lower court magistrates. Thevariation of opinion does not necessarily imply‘division within the ranks’, however. It may insteadreflect the variety of cases brought before individualcourts, and the varying capacities of the lowercourts to implement the Act.

period between the granting of the interim order andthe finalisation of the final order can be a highlydangerous one for complainants.

Respondents are generally only prohibited fromentering the shared residence for a temporary period,until the complainant and respondent agree on theliving conditions or until the complainant applies fora formal order to remove the respondent through theHigh Court.

An exception to this approach is when the courtestablishes that the violence was of a ‘serious nature’and thus warrants the immediate removal of therespondent. Factors that help the application forimmediate removal of the respondent includeevidence of acts of domestic violence againstchildren and other vulnerable family members, andthe existence of criminal charges against therespondent.

Magistrates, however, are cautious about grantingrestrictive conditions prohibiting respondents fromentering areas surrounding the home and other‘common’ sites frequented by both parties. Such sitesmay include places of employment, schools andother family homes. They argue that it is difficult toenforce, particularly in communities where livingconditions are crowded. Magistrates do neverthelessacknowledge that concerns about deprivingrespondents of their right to occupy the sharedresidence, except under “extreme circumstances”,also deny the right of complainants and theirchildren to live in their own homes without violence.

Emergency monetary reliefThe Act makes provision for the granting ofemergency monetary relief (EMR) which is definedas:

the compensation for monetary lossessuffered by the complainant at the time of theissue of a protection order as a result ofdomestic violence, including:a) loss of earnings;b) medical and dental expenses;c) relocation and accommodation

expenses; ord) household necessities.6

SA CRIME QUARTERLY No 8 JUNE 200426 ARTZ

• Does section 7 apply to situations in which the respondent has full legal rights to his/her propertyand therefore full legal rights to occupy thehome?

• Should section 7 be applied as an ‘exclusion’ or eviction order?

• If section 7 can be applied in the same manner as an eviction order, at what stage of theproceedings should it be applied? (Is it fair togrant this condition at the interim stage, when theproceedings are ex parte?)4

• If section 7 can be applied in the same manner as an eviction order, should this condition onlybe granted on the Return Date5 when therespondent can have an opportunity to respond tothe allegations of violence?

• Should the prohibition of entry into the shared residence be a temporary measure until formalproceedings are conducted in the High Court toestablish permanent removal or eviction from theresidence?

• If the DVA gives magistrates the power to prohibit entry into the shared residence for anextended period of time, does the order have thesame legal standing as a High Court evictionorder?

Some magistrates argue that they will only grant thisremedy when the respondent is present to argue hisor her side of the case. Others say this approach isincorrect as the Act does not specify at which stagethe prohibition may be granted.

The general approach to section 7, however, is toprohibit the respondent from entering the sharedresidence only at the final stage of the protectionorder. However, when magistrates believe thatcomplainants are in serious danger of further abuse,the condition will be granted at the interim stage.

But what constitutes ‘serious danger’ may not becompatible with the complainant’s real or perceivedvulnerability to further violence. Indeed, theconsequences of not thoroughly examining the extentof vulnerability to further violence can have harmfuleffects on the complainant and her dependents. IfReturn Dates are set months away, or are postponeddue to the fact that the notice to the respondent toappear on the Return Date is not properly served, the


The intention of the Act is to allow complainantsaccess to emergency funds to ensure that they canprovide for their immediate safety and well-being,and that of their dependents. Magistrates stated thatsome complainants have interpreted this relief as asubstitute for maintenance payments, and foundsome of the requests “unreasonable”. The Act,however, is careful not to use the term maintenanceto describe this remedy.

Magistrates suggested that some complainants usethe Domestic Violence Act when the maintenancesystem fails them – either when they have beenunsuccessful in securing a maintenance order orwhen they have waited a long time for themaintenance order to be served, granted or varied.As expected, there were a range of opinionssurrounding both the purpose and application ofEMR.

On the one end of the continuum are magistrateswho believe that respondents who are notmaintaining their families or the shared residence –as required by the law – are committing what theAct refers to as economic abuse. Referring tosections 1(ix)(a) and (b) of the Act,7 it would followthat defaulting on maintenance, not paying monthlyrent or mortgage payments, and not providing forbasic family necessities would all constitute‘economic abuse’ and therefore warrant EMR.

Some magistrates suggested that it was perfectlyacceptable for the court to provide the complainantwith EMR for a temporary period, while the lengthywaiting periods for the appearance of maintenancedefaulters in maintenance courts were pending. TheAct is clear that a protection order may be granted ifthere is evidence of any act of domestic violence,including economic abuse. One magistratedefended this position:

We are all well aware of the huge delaysexperienced by maintenance courts.Maintenance hearings are set down months inadvance and the courts sometimes have to sendthe sheriff out three or four times before hereceives his summons to appear. This is the casefor both new applications, defaultingrespondents and for variations of maintenance

orders. Some women wait for over a year to gettheir first maintenance payment. How long dotheir children have to go without proper food,without school because fees are not paid,without medical care? The maintenance systemis a mess and fathers know it. They use thedelays to avoid supporting their families. TheDVA is clear, it is an immediate and effectiveremedy. The one part of justice can help theother. If the defaulter ends up double-payingbecause of back payments, that’s his problem.He should have obeyed the maintenance orderin the first place. Really, the maintenance courtcan take the amount of EMR off the payments ofmaintenance. I would consider that a fairjudgment. It provides the applicant withimmediate funds and it means he doesn’t haveto pay twice.

This view, however, was not shared by othermagistrates who considered maintenance acompletely separate issue that should becontemplated in the maintenance courts. The ideathat EMR could potentially provide the complainantwith ‘bridging’ funds until the maintenance matterswere settled, was not viewed favourably.

Instead, the allocation of EMR, it was suggested,should only cover expenses that are a “direct resultof domestic violence”. This, however, implies thateconomic abuse is not a ‘real’ form of domesticviolence and that provision for EMR should only bemade when other more ‘serious’ forms of violenceare committed by the respondent.

It was further suggested that EMR should cover veryspecific costs associated with domestic violence,such as relocation expenses, payments for rental orbond, medical costs incurred due to the acts ofphysical domestic violence and other immediateliving costs incurred by the complainant as a resultof violence. The latter was not adequately definedin the discussions with magistrates.

There was very little consensus about what toprovide complainants in terms of EMR. However,magistrates were in agreement that the amountgranted should be fixed – in one lump sum or for alimited monthly period – and that the courts should

ARTZ


favourable approach is to refer these matters to theChildren’s Court or the High Court.

One argument was that granting a no-contact or asupervised-contact order at the interim protectionorder stage was unfair to the responding party. Theinterim order, being an ex parte order, takes away therespondent’s fundamental right to have access tohis/her children – an issue that should be the purviewof the Children’s Court or the High Court. It wassuggested that contact orders with children shouldonly be granted at the finalisation of the protectionorder, when the respondent could argue his or herposition.

It was further suggested that when a High Court orderis in place, the Magistrate’s Court must not grant anorder that contradicts the order already in place.When a High Court order has granted particularcustodial or access rights to the children, theMagistrate’s Court is not in a position to vary theconditions of the order, even though the DVA makesprovision for the protection of children either by theremoval of the respondent or a regulated ‘contact’agreement between the parties.

It was counter-argued, however, that the purpose ofthe interim protection order was to provideimmediate relief to the complainant until the returndate. When the magistrate is convinced that an “actof domestic violence” has taken place, his or herduty is to ensure the protection of the complainantand her dependants. Limiting the granting of contactorders to the finalisation stage therefore defeats theobjective of protecting complainants from imminentdanger.

It was widely accepted that approaching the policeor the courts for protection was one of the mostdangerous periods for the escalation of domesticviolence and, on this basis, every available remedyprovided for in the Act should be made available. Itwas also forcefully argued that the act of domesticviolence does not have to be committed solely on thecomplainant, but that the court should also protectchildren from the damaging effects of this violence.

When there are existing High Court orders, it wasrecommended that magistrates should have the

inform the complainant that they may apply formaintenance at the maintenance court.

Contact orders with childrenIn terms of section 7(6) of the Domestic ViolenceAct, if the court is satisfied that it is in the bestinterests of the child, it may:

a) refuse the respondent contact with such a child; or

b) order contact with such child on such conditions as it may consider appropriate.

Contact orders for children were raised as anotherserious issue facing magistrates. The ambiguity andthe variety of approaches used by magistrates inrelation to orders which specify extent of contactthat the respondent has with children, was notableduring all phases of this study. The First Reportfound that between 10-50% of requests forsupervised contact with children (in the researchsample) were refused by magistrates. Whenconditions of contact were granted (in s. 3.1.2.8 ofForm 4 of the application), the following conditionswere ordered: • that the respondent not remove the child

without the complainant’s consent;• that the respondent not come within a certain

distance of the child;• that the respondent not visit the child at

school/crèche/day mother; or• that the respondent not contact the child in any

way, including telephonically.

Granting both interim and final protection ordersthat limit the respondent’s contact with his/herchildren is a complex and contentious issue for themagistracy. The approach to non-contact orsupervised contact orders has been a cautious onebut some courts are more willing to entertaincontact orders than others.

Magistrates are acutely aware that the decision tolimit contact with children is serious, and that themanner in which application forms are currentlyfilled out does not provide the court with sufficientinformation to make such a weighty decision. As aresult, magistrates tend not to use the DomesticViolence Act to establish temporary ‘custody’ of, or‘access’ to children, and feel that the more

SA CRIME QUARTERLY No 8 JUNE 2004 29ARTZ

power to grant immediate relief to a complainantuntil such time that an alteration is made to the HighCourt order. Magistrates felt strongly that eventhough, in principle, High Court orders should bevaried at the High Court, the Domestic Violence Actshould provide a victim of domestic violence withsome temporary relief.

Some insisted that when the original High Courtorder was issued, domestic violence may not havefeatured in the decision to grant the order. TheDomestic Violence Act specifically provides forimmediate relief to victims of domestic violence, andon this basis, magistrates argued that they shouldhave the power to intervene in cases in whichviolence is present and to provide relief to applicantsuntil such orders can be varied in High Court.

The court records that were analysed for the FirstReport showed that violence against women hadconsiderable effects on children. Some of the effectsof witnessing domestic violence on childrenincluded:• insomnia/restlessness;• acute anxiety;• diarrhoea and vomiting;• abdominal pain;• eating problems (such as not eating or excessive

eating);• notable problems at school when violence

intensified (i.e. poor performance or troubles with teachers or peers);

• depression/sadness;• bed-wetting;• running away from home/staying with other

family members/refusing to come home;• poor general health (chronic cold or flu

symptoms; exhaustion); and• increasingly aggressive behaviour/discipline

problems.

The report also argued that the magistrates presidingover domestic violence cases should request a reportfrom a social worker on the child. One magistratesuggested that with respect to children as applicantsor dependents of the protection order:

We need permanent social workers at courtdealing with domestic violence cases. In fact,

we need a special domestic violence court,or at minimum we need to legislate a socialworkers report, which must be attached tothe application. High court won’t deal withaccess/custody issues regarding childrenwithout a social worker’s report, surely thesame should happen in the case of lowercourts who are also expected to addressthese issues through this legislation.

Recommendations regarding the child (asan applicant for a protection order) and in

terms of the placement of children:

• If a child applies to the court for a protection order, the court must considerthe application, and if it deems fit, grantan interim protection order. The courtmust then, if it finds the child to be inneed of care, refer the child to theChildren’s Court in terms of section11(1)(c) of the Child Care Act.

• When adult applicants request, as part of the protection order, an order for theplacement of children (i.e. structured orspecific visits), the magistrate shouldinform the applicant that thearrangement is a temporary one. Whenmaking provision for the placement ofchildren, magistrates must consider:- the safety, health and well-being of the

applicant, child/children or any otherperson affected by the domesticviolence;

- the applicant’s perceived risk of further harm or violence;

- the personal and material interests of the applicant; and

- the best interests of the child/children.

Like the concerns surrounding the prohibition ofrespondents from entering the shared residence,magistrates are cautious about making decisionsthat may appear to be the purview of the HighCourts. The development of guidelines for presidingover cases that involve contact orders with childrenmust be compatible with a number of other legal

not be finalised, in whole or in part, the order may beset aside (cancelled) or varied (changed).

6 Section 1(x)(a).7 Sections 1(ix)(a) and (b) define economic abuse as:

(a) the unreasonable deprivation of economic orfinancial resources to which a complainant is entitledunder law or which the complainant requires out ofnecessity, including household necessities for thecomplainant, and mortgage or bond repayments orpayment of rent in respect of the shared residence; or(b) the unreasonable disposal of household effects orother property in which the complainant has aninterest.


instruments, namely the Child Care Act, remediesavailable through the High Court for the custodyand access of children, and the forthcomingChildren’s Bill. Since the latter has not beenfinalised, magistrates felt that decisions relating tocontact with children can only be decided on thefacts presented before them, and what is currentlyset out in the Act itself.

Endnotes1 Sections of this article were originally published as L

Artz, Magistrates and the Domestic Violence Act:Issues of Interpretation, Institute of Criminology,Faculty of Law, University of Cape Town, 2003.

2 The results of this research are based on the opinions of magistrates themselves and not the author. Thisstudy was conducted to investigate the variousapproaches by magistrates in implementing the Act.Broadly, the study involved the re-examination of ourmonitoring database on the DVA (see P Parenzee, LArtz & K Moult, Monitoring the Domestic ViolenceAct: First Report, Institute of Criminology, Faculty ofLaw, University of Cape Town, 2001), in-depthinterviews with magistrates from each of the nineprovinces, the analysis of the outcomes of two majorconferences (including over 350 magistrates and HighCourt judges, facilitated by this author and herassociates) as well as the outcomes of monthlymeetings with the ‘Domestic Violence WorkingGroup’; a group consisting of magistrates representingeach province, the Justice Training College, theGender Directorate of the Department of Justice andthe author.

3 7(1) The court may, by means of a protection order referred to in section 5 or 6, prohibit the respondentfrom:(c) entering a residence shared by the complainantand the respondent: provided that the court mayimpose this prohibition only if it appears to be in thebest interests of the complainant.7(2) The court may impose any additional conditionswhich it deems reasonably necessary to protect andprovide for the safety, health or wellbeing of thecomplainant …

4 Ex parte, refers to court proceedings where the respondent is not in attendance and decisions aremade in absence of the respondent. Decisions madeat ex parte hearings are only enforceable until theReturn Date, when both the applicant and therespondent appear before the magistrate.

5 Return Date is the date set by court, once an interim protection order has been granted, when both theapplicant and respondent appear before themagistrate. The magistrate may finalise the protectionorder on the return date, or should the respondentprovide reasons for why the protection order should


The launch of the Scorpions was announcedin September 1999, and the organisationbecame formally known as the Directorate of

Special Operations (DSO) in January 2001 when itsfounding legislation was promulgated.1 The DSO isthe investigative arm of South Africa’s NationalProsecuting Authority, falling under the authority ofthe National Director of Public Prosecutions,Bulelani Ngcuka (referred to below as the ‘NationalDirector’).

High-profile since its inception, the organisation isgenerally viewed by the South African public asembodying the ultimate crime fighters. The DSOinvestigation into the arms deal concluded by theSouth African government in 1999, and itsinvestigation into the role of the deputy president inthis deal, upped this public profile considerably –and highlighted the unresolved issues that havedogged the DSO since inception.

The cherry-picking accusationAlmost as soon as successful DSO cases began tobe publicised, accusations of DSO ‘cherry-picking’arose. Specifically, the DSO is accused of choosingto investigate and prosecute only matters whichthey are sure to win. Sometimes this accusationwent further, to suggest the DSO had a tendency to

take over cases already substantially investigated,taking all the credit for the subsequent successfulconclusion of the matter. More generally, there isdiscomfort as to which cases become DSO‘matters’ – in law enforcement language,uncertainty as to the DSO’s mandate.

It is easy to dismiss the accusations as ‘sour grapes’,but the very real uncertainty remains as to how amatter comes to be pursued by the DSO. Whatthen, is the DSO’s mandate? Given the complexityof the answer to this question, it is unsurprising thatuncertainty exists.

The legislation creating the DSO describes alegislative mandate encompassing a broad conceptof organised crime – any crime committed in an“organised fashion”2 – which is so wide that justabout any matter could be argued to fall under theDSO mandate. The legislation furthermorespecifically retains all of the police’s powers ofinvestigation, so that this mandate is not exclusiveto the DSO. Clearly, such a broad mandate is notpractical for those on the ground who must put itinto practice. The legislature appeared to haveenvisaged that a negotiated operational mandatewould emerge, and that a Ministerial Committee,consisting solely of Cabinet members, created by

Jean Redpath, criminal justice research [email protected]

WEATHERING THE STORM

Tough questions forthe Scorpions

The ‘Scorpions’ are probably the most recognised law enforcement body in South Africa. Yet their existence

appeared to be under threat during 2003 when highly placed figures suggested that they ‘cherry-pick’ their

cases, and are open to political manipulation. There was even talk of disbandment or restructuring under the

police on the grounds of unconstitutionality. While the Scorpions appear to have weathered that particular

storm, the issues remain. Is there any substance to the accusations?

the legislation would confirm procedures for thetransferral of matters to the DSO.3

However, acrimony between the SAPS and DSOarose soon after the launch of the organisation. Withthe appointment in 2000 of Jackie Selebi as SAPSNational Commissioner, a man who was keen todefend the reputation of the police and who clashedwith the National Director, a negotiated mandatebecame unlikely. The Ministerial Committee neversat. As a result, the DSO was forced to carry out aninternal case review, in which it considered whetherthe matters it had already taken on were appropriateor not. In doing so, the DSO came up with its ownoperational mandate – more in the nature of internalterms of reference – which outlines the requirementsa case must meet before being taken on. This wasdubbed “Circular One”.4

In terms of Circular One, the first criterion is that thematter concerned must fall within the strategic focusareas of the DSO. These have been refined toinclude: drug trafficking, organised violence(including taxi violence, urban terror and streetgangs), precious metals smuggling, human trafficking,vehicle theft and hijacking syndicates, serious andcomplex financial crime, and organised publiccorruption.

There are a further 14 general criteria or factors thatmust be considered, covering such questions as ‘Isthe criminal activity involved complex, and does itcomprise at least five persons?’ There are alsofinancial thresholds that must be met. For example, acorruption matter has a threshold of R500,000, whileserious economic offences must involve actual lossof R5m to meet threshold requirements.

All of these requirements must be met andconsiderations canvassed before the head of the DSOwill authorise an investigation or ‘declare a matter interms of s28’. Prior to such an authorisation, no DSOmembers are designated to a case and DSOmembers do not, until designation, enjoy the DSOspecial powers of investigation, which includesomewhat expanded powers of search and seizure.

This, then is the mandate of the DSO. While it doesto some extent put to rest the cherry-picking

accusation – no case which meets all of theserequirements can by any stretch of the imaginationbe considered ‘easy’ to prosecute – it does providemany loopholes for the DSO to justifiably turn downthe investigation of a case. Indeed, the majorconcern of the DSO head office does appear to havebeen to avoid being overloaded with trivial matters.

However, one of the main flaws from which thismandate suffers – arguably due to politicalcircumstances beyond the control of the DSO – isthat it has been drawn up without any outside input.The question of whether a case is taken on or notremains one which is settled internally, by the DSOalone. Furthermore, there seems to be no oversightbody which on a regular basis reviews caseselection. Outside accountability of the DSO as awhole rests entirely with the Minister of Justice andConstitutional Development, and with the justiceportfolio committee of parliament.

The constitutional questionDuring 2003, some in government suggested thatthe DSO in its current form might be‘unconstitutional’. Again, it is tempting to dismisssuch accusers as having ulterior motives.Nevertheless, the basis for this claim appeared to bethe constitutional provision which states that SouthAfrica must have only one police force (thesuggested cure for the defect is the removal of theDSO from the NPA to fall under the control of theSAPS).5

The first thing to note is that if the DSO were to belocated anywhere other than under the NPA, itwould no longer be the DSO as conceived – that is,a unit comprising prosecutors and investigatorsworking together as a team on a daily basis, carryinginvestigations through to prosecutions seamlessly.Nevertheless, some kind of secondment arrangementwith prosecutors working with SAPS investigatorscould be set up to approximate DSO operation.

Yet the question remains a moot point – is the DSO,as conceived, constitutional? While an expertconstitutional opinion cannot be offered, somerelevant constitutional provisions can be discussed.The provision in question prohibits any other entityfrom operating as a police force – from having the

SA CRIME QUARTERLY No 8 JUNE 200432 REDPATH


“objective powers and functions” of the police.Only in the matter of investigation of crime doesthe DSO overlap with the SAPS.

Yet each of the objectives, powers and functions ofthe police on their own are manifestly not exclusiveto the police – for example, the object of “securingthe inhabitants of the Republic” would also beshared by the SANDF. It could therefore be arguedthat the ‘investigation of crime’ on its own issimilarly not the exclusive preserve of the SAPS.This would tend to suggest that the DSO within theNPA is not excluded from investigating crime. But isit empowered to investigate crime?

The constitution provides that the National Directorhas the power to carry out any “necessary functionsincidental to instituting criminal proceedings”. Itcould be argued that this includes the furtherinvestigation of certain crimes to ensure successfulprosecution. Many institutions other than the DSOcarry out investigations incidental to their functions– such as the Auditor-General, or the South AfricanRevenue Service.

Furthermore, Parliament itself considered thequestion and took the unusual step of making itsopinion clear: the preamble to the NPA amendmentact creating the DSO says, “the constitution doesnot provide that the prevention, combating orinvestigating of crime is the exclusive function ofany single institution”.

But the constitutional provision regarding a singlepolice force is a distraction from a more deep-seated question of constitutional theory, which isrelevant not only to the DSO alone but to the NPAas a whole – the question of the immense power ofthe National Director, and its impact on theprinciple of the separation of powers. That power –which resides largely in prosecutorial veto – wasgreatly increased once the DSO was created,effectively providing the National Director withauthority also over a powerful investigative tool.

Separation of powers‘Separation of powers’ refers to the principle ofdemocratic constitutional theory that the business ofgovernment should be divided along natural lines

into the power to make law (legislative), the powerto enforce law (executive), and the power to resolvedisputes arising under law, including deciding onwhether actions undertaken by the other twobranches fall within the law (judicial). The idea isthat each branch of government must have thepower and the incentive to guard its own sphereand to counter the abuses of the other two.

In 1998 legislation was passed that gave SouthAfrica a single, national prosecuting authority, interms of which the National Director could veto theprosecutorial decisions of provincial directors – andindeed of any prosecutor.6 These powers werealmost immediately subjected to constitutionalscrutiny, but in 1996 the Constitutional Court heldthat this provision did not unjustifiably infringe thedoctrine of separation of powers and that there weresufficient safeguards against the abuse of power bythe National Director.7

However, in the modern state, a prosecutor,particularly a national prosecutor, plays a role thatto some extent impinges on the principle ofseparation of powers. The DSO, as an entity of theNPA, falls directly under the National Director, andits decisions to prosecute are also subject to hisveto. (Even were the DSO to fall under the SAPS,the National Director would still have a veto powerover any prosecution).

The position of the prosecution service in anycountry is interesting in that close analysis revealsthat it runs the risk of straddling both the executiveand judicial spheres. The prosecution in reality has aquasi-judicial function (only those crimes it choosesare prosecuted, and only those it prosecutes run therisk of being convicted) yet it is firmly positionedunder the executive branch of government.

The decision to prosecute a matter is not undulyproblematic because whether a conviction isobtained depends in the final analysis on thejudiciary; in theory, even a malicious prosecutionwill not succeed if the judiciary finds there is notenough evidence to prove the charge. However, adecision not to prosecute is more problematic, asthere is no input into the outcome of such adecision from another branch of government.

REDPATH

In theory, the failure on the part of the prosecutionto carry out its obligations, in particular, bydeclining to pursue allegations of wrongdoing bymembers of the executive, leaves only recourse tothe legislature, to whom the prosecution isaccountable.8 Parliament can therefore call theprosecution to account for the decisions it takes,particularly decisions to prosecute or not toprosecute. Although this is theoretically possible,an academic paper has argued that in politicalsystems where the president is elected by thelegislature and therefore by the majority party inParliament, the probability of Parliament callingthe prosecution to account for its failure toprosecute is low.9 This thesis appears to hold truein South Africa, where the majority party inparliament effectively elects the president.

A person who feels aggrieved by the prosecutingauthority’s decision not to prosecute may also optto institute a private prosecution. However, thiscould become particularly complex and expensivefor an unsuccessful private prosecutor.

The political manipulation accusationClosely allied to the question of separation ofpowers are the accusations levelled against theDSO that it is open to political manipulation.Such accusers would say that given the reputationfor excellence of the DSO, the mere fact that aninvestigation is carried out against a particularperson suggests guilt and could be used to tarnishthat person in the public mind. On the otherhand, the DSO could also decline to investigate orprosecute a matter when it should indeed do so,in order to protect particular individuals.

In essence, these concerns can be reduced to thequestion of whether the DSO is at the same timeindependent (acts against the executive orlegislature when necessary) impartial (refrainsfrom acting maliciously against politicalopponents of the executive) and accountable(answerable for its actions). These traits are neithermutually exclusive nor mutually compatible. Forexample, consider the situation in which theMinister of Justice and ConstitutionalDevelopment, to whom the DSO is accountable,is implicated in an investigation.

This balancing of independence, impartiality andaccountability is an extremely difficult one. Inconsidering whether the DSO has succeeded inmaintaining the balance, there are two questions:first, whether the DSO has in fact thus far behavedindependently, impartially and accountably, andsecond, whether the DSO is structured such that atany stage in the future, it is unlikely to fail on any ofthese points.

Without access to detailed information on thematters with which the DSO has engaged, it isextremely difficult to judge the first question ofactual independence, impartiality, andaccountability. The mere fact of the existence of aDSO investigation into the arms deal – which islikely to implicate members of the executive – iscited as evidence of the DSO’s independence.However, similarly, the refusal to prosecute theDeputy President, combined with the publictarnishing of his name, is cited by others asevidence of partiality. Again, without more in-depthknowledge, these questions are almost impossibleto judge.

But in an unknown future with unforeseen facts andunknown players, how prone would the DSO be tofail to be independent, impartial, or accountable?After all, President Mbeki has only one more term,and Ngcuka’s tenure ends in 2008.

The DSO as it is structured is overshadowed by thefact that the National Director, its de facto head, isdependent on the President’s continued goodopinion that he is a fit and proper person, to avoidbeing removed from office. It could, however, beargued that this requirement protects the NationalDirector from outside interference. He is appointedfor a non-rewewable term of ten years at the salaryof at least a High Court judge, and the groundsunder which the president can remove or suspendhim are limited.10 Also, such a decision by thepresident is subject to ratification by parliament.

Furthermore, while the DSO is accountable to theMinister of Justice and Constitutional Development,and to parliament, this form of accountability isinsufficient and perhaps even counter-productive(on the independence front) when decisions on


SA CRIME QUARTERLY No 8 JUNE 2004 35REDPATH

whether to investigate (not to mention prosecute)members of the executive and legislative arms ofgovernment are at issue, if the executive andlegislature are closely intertwined as they are inSouth Africa. The Ministerial Committee created bythe DSO legislation, which inter alia may determine“where necessary the responsibility of the DSO inrespect of specific matters”12 is fundamentallyflawed in that it consists solely of Cabinet Ministers.By contrast, the United Kingdom’s National CrimeSquad (NCS) is directed by an NCS ServiceAuthority consisting of 11 members, none of whomare high-ranking members of the executive, five ofwhom are entirely independent, and four of whomare elected by associations of police officers.

Consider that Italian Prime Minister SilvioBerlusconi in 2003 succeeded in getting hisparliament to pass legislation immunising him fromprosecution while he remained in office. (Italy’sConstitutional Court subsequently declared the lawunconstitutional.) How much easier for him if hecould have got a committee consisting of hisCabinet to instruct investigators not to pursue thecorruption matter which implicated him?

Lastly, there is no provision for misconduct byindividual members of the DSO to be investigatedby an independent body. Alleged misconduct ofSAPS members as well as those of municipal policeforces fall under the mandate of the IndependentComplaints Directorate.

RecommendationsWhat is clear is that the current legislative structureof the DSO makes its independence, accountabilityand impartiality almost entirely dependent on theintegrity of its officers and of the National Director.While the incumbents may indeed thus far haveacted blamelessly, what could improve the situation(and safeguard the future) while not interfering withthe ability of the DSO to do its job? • An independent committee consisting partially

of persons outside of the executive andlegislature should exercise (post facto) oversightin respect of DSO case selection (the exercise ofits mandate), and review the general conduct ofinvestigations and prosecutions after theirconclusion.

• A decision not to prosecute should be reported to the committee after the conclusion of a DSOinvestigation, and should be reviewable by theSupreme Court of Appeal at the instance of thecommittee.

• Removal of the National Director by the President should only be possible afterconfirmation by the Supreme Court of Appeal.

• Alleged misconduct by individual DSO members should fall under the mandate of theIndependent Complaints Directorate.

The DSO is an innovation in South African lawenforcement which has had a profound impact onthe investigation of complicated cases. Whileallegations of cherry-picking, at least since theadoption of Circular One, do not appear to befounded, DSO case selection is a laborious andopaque process. The legal infrastructure withinwhich the DSO is situated is not without problems,especially in respect of ensuring a balance of theDSO’s independence, accountability and impartiality.

Endnotes1 National Prosecuting Authority Amendment Act 61 of

2000.2 ‘Organised fashion’ includes the planned, ongoing,

continuous or repeated participation, involvement orengagement in at least two incidents of criminal orunlawful conduct that has the same or similar intents,results, accomplices, victims or methods ofcommission, or otherwise are related by distinguishingcharacteristics.

3 See s31(1) National Prosecuting Authority Act 61 of 1998 as amended.

4 The Office of the Head of Operations: DSO. Circular:1. Effective Date 8 November 2001. Signed:Acting Investigating Director Adv. LF McCarthy 7-11-2001.

5 See for example, Skerpioene word dalk deel van polisie, sê Mbeki, Die Burger, 30 July 2003.

6 National Prosecuting Authority Act 32 of 1998.7 See s179, Constitution of the Republic of South Africa

Act 108 of 1996. The Constitutional Court confirmedthis provision of the Constitution at paragraphs 140 and141 in CCT 23/96, Certification of the Constitution ofthe Republic of South Africa, 1996.

8 See s35(1) of the National Prosecuting Authority Act 61 of 1998. However, the Minister of Justice andConstitutional Development exercises finalresponsibility over the prosecution, in terms of s 179(7)of the Constitution of the Republic of South Africa Act108 of 1996.


9 Van Aaken, Salzbuger & Voigt, The Prosecution of Public Figures and the Separation of Powers:Confusion within the Executive Branch, GermanWorking Papers in Law and Economics, Paper 11,Volume 2003.

10 The president may remove the National Director on the grounds of the National Director’s continued ill-health, misconduct, incapacity to carry out his dutiesefficiently, or on the grounds that the NationalDirector is no longer a fit and proper person to holdoffice. See s12(6)(a) National Prosecuting AuthorityAct 32 of 1998.

11 J Chafetz, The Independent Counsel after Nixon, Yale Political Quarterly 19(3), 1998.

12 See s31(1)(c)(i) National Prosecuting Authority Act 32 of 1998.

WHEN CAN I FIRE? Use of lethal force to defend property · of lethal force to protect property was...

Documents

Transcript of WHEN CAN I FIRE? Use of lethal force to defend property · of lethal force to protect property was...