Critical Analysis (ii) Literature Review, Crime profile

Articles to help you counter hoplophobic arguments.

Moderator: GenMod

Critical Analysis (ii) Literature Review, Crime profile

Postby Martin_tu » Wed, 2006-04-12 13:32



Desiree Hansson's dissertation is a useful resource that focuses on the fatal use of guns by civilians, police and private security officers in the Cape Town area prior to 1991. While her dissertation covers the period prior to democratisation and her focus is not specifically on the criminal use of firearms in the perpetration of violent crime, it provides a useful review of South African and international research.

In a manner somewhat unique to the South African literature, it specifically delineates all the major issues inherent in the gun control debate such as the lethality of firearms relative to other weapons, the possible substitution effect if access to firearms is curtailed, defensive gun use etc. In particular, her dissertation gives a unique insight into firearm related issues in the previous dispensation. In that way it provides a background to the current firearm control debate.

The most comprehensive work to date on firearm distribution in South Africa has been compiled by Robert Chetty 2 of the National Secretariat for Safety and Security. Chetty aims to provide policy makers with fundamental data relating to firearm proliferation in South Africa.

Chetty's compilation of material on firearm- related crime, firearm deaths and injuries as well as a detailed account of the distribution of licensed firearms in South Africa, is an invaluable source of statistical information. His book is effectively an audit of the available factual information pertaining to firearms and he deliberately presents the data without commentary or interpretation.

Despite Chetty's attempt at objectivity, one of the basic presuppositions of the research is that the proliferation of firearms contributes to high levels of violence in South Africa. In the foreword Dr. Fanaroff writes, "There is no doubt that the easy availability of firearms contributes to the high level of violence and violent crime." 4 It is this very pre-existing presumption that undermines the validity of most of the research in this area.

While the factual information presented in his book is undoubtedly valid, the way in which information is presented as well as certain omissions, belie the predisposition of the authors.

No attempt is made to detail those cases where individuals have effectively used their firearms to deter violent criminals, only those in which firearms have been wrongfully used. Rather than the focus of the book being on firearm use as the title would suggest, all the information presented relates to the abuse of firearms in South Africa.

Nonetheless, the information provided is an extremely useful compilation of official data. Whereas Chetty aims to provide policy makers with raw factual data pertaining to firearms in South Africa, Gamba 5 attempts to use available research to persuade policy makers on the need for more stringent gun control. Of all the South African publications considering firearms and related issues in the post-apartheid era. Society Under Siege: Managing Arms in South Africa is one of the most academically sound and fair treatments of selected aspects of the topic, despite its avowed commitment to strict gun control. The research conducted is reliable, well informed and the claims made are well substantiated.

This book is primarily concerned with the connection between legal weapons, illegal weapons and the culture of violence in South Africa. It highlights the regional flow of illicit weapons explaining how the flux of weapon's flow depends largely on supply and demand. It also discusses how the distinction between legal and illegal weapons becomes blurred through the loss and theft of licensed weapons. This publication specifically aims to be policy relevant6 and contains recommendations for the revision of firearm legislation in South Africa.

The findings contained in Society Under Siege: Managing Arms in South Africa are certainly important and cognisance must be taken of them. However, the authors make an ardent plea for stricter gun control on the basis of their research findings, while neglecting to investigate other relevant factors that impinge heavily on the relative efficacy of stringent gun control measures.

Anthony Altbeker's case study 7 on gun crime and self-defence, is one of the few South African studies that focuses specifically on the issue of defensive gun use. The scarcity of indigenous research in this area is unfortunate given the pivotal place it deserves in the gun control debate.
Altbeker reviewed 602 police dockets on violent crimes occurring in Alexandra and Bramley in early 1997. He found that only fifty (8 percent) of the victims had a licensed firearm with them at the time of the attack. In only one quarter of these cases was the victim able to use a firearm in self-defence. Altbeker uses these results to assert that victims were four times more likely to have their firearms stolen than to use them in self-defence.

This statement certainly holds true for the cases that Altbeker examined. However, these were by definition cases in which the perpetration of a crime had been successful. It ignores the entire set of cases in which crimes were prevented by the defensive use of a gun. For this reason, his findings are not generalisable and give little or no indication of the prevalence or effectiveness of defensive gun use in South Africa.

7 Anthony Altbeker, "Guns and Public Safety: Gun Crime and Self- Defence in Alexandra and Bramley" (Unpublished Research Report commissioned by Gun Free South Africa, 1999).
8 Altbeker. "Guns and Public Safety," 2.

9 By contrast, research conducted by John Mann in 1994, according to Hammond detailed 206 cases in which civilians had used licensed firearms to defend themselves. In 36 percent of the cases the attackers were killed or arrested by their intended victims and 64 percent of attackers fled. Eight of the victims died in the confrontation and no innocent bystanders were injured or killed.

While this study alone cannot paint an accurate picture of defensive gun use in South Africa, it certainly seems to indicate that the utilisation of a firearm for self-defence may be more effective and prevalent than gun control advocates would like to concede.

Katherine McKenzie's work attempts to substantiate the basic presupposition of the gun control movement. By examining evidence from countries in the Southern African Development Community (SADC) she endeavours to establish the alleged correlation between the prevalence of civilian gun ownership and levels of violent crime. From a comparison of ten case studies McKenzie concludes that "the convergence of poverty, unemployment, a gun culture and the availability of firearms is a lethal combination which results in high levels of gun crime."... "Countries in the region with effective gun control policies and fewer firearms in circulation, have less gun crime and are safer than countries with permissive gun control policies and more firearms in circulation."

McKenzie's work has been subject to much criticism, most notably from Richard Wesson. Wesson considers those countries for which McKenzie provides murder rates and calculates an estimated murder rate for Zimbabwe. Motivating for the exclusion of Swaziland and South Africa on the basis of their dissimilarity to the other cases in important respects, he analyses the remaining cases. Wesson finds that in fact there appears to be a negative correlation between "ease with which private civilians can obtain firearms" and homicide levels.

Peter Hammond, "The Proposed Firearms Control Bill, Government Gazette 21193," (United Christian Action Submission on Firearms Control Bill, 2000) ( (03/08/2000).
10 Katherine McKenzie, "Domestic Gun Control Policy in Ten SADC Countries" (Unpublished Research Report commissioned by Gun Free South Africa, 1999).
11 Ibid., 3.
12 Richard Wesson, "An Inspection into the Relationship Between Murder Rates and Legalised Private Ownership of Firearms in Southern Africa, with Special Regard to South Africa," 1999 (http:// (Updated. M_tu). "Id.

10 Both McKenzie and Wesson provide an interesting foray into aspects of the domestic gun control debate, but neither study in its current form holds up to scientific scrutiny. Even rudimentary quantitative analysis of the qualitative data gathered by McKenzie certainly does not support her conclusion as Wesson amply demonstrates.

Despite the potential of Wesson's work, it is limited by the fact that it is primarily a response to McKenzie and other gun control advocates, rather than a proactive study in its own right. In both cases the results and conclusions must be treated with caution. What both documents illustrate most of all, is the need for further, more comprehensive research in this area, South African research into firearm control and related issues is most notable for its scarcity and manifold limitations.

Of the studies that have been conducted, most have been funded and commissioned by those with an outspoken predisposition towards stringent gun control such as Gun Free South Africa. Consequently, the research tends to be rather one-sided which results in an impoverished policy debate in the halls of power that lacks rigour and precision.


Research from abroad, and the U.S.A. in particular, is much more diverse and comprehensive and has generated important data which reflects on the efficacy of gun control in curtailing violent crime in other countries. In America specifically, the debate on weapons and crime is much more robust than in South Africa, given the sophistication of the "gun lobby", the longevity of the debate and the existence of outspoken academics and activists on cither side.

In the late 1970s Wright, Rossi and Daley undertook a comprehensive review of all the available American literature pertaining to weapons, crime and violence.14 Despite being almost two decades old Under the Gun: Weapons, Crime and Violence in America provides a very useful review of the research conducted to that point. Some of the main issues examined include, whether or not a causal link between private civilian firearm ownership and violent crime can be demonstrated, the alleged deterrent effect of civilian gun ownership, the intrinsic lethality of firearms and the effect of weapons control legislation on violent crime. After surveying the available evidence, Wright, Rossi and Daley concluded that most of the research was marred by severe methodological shortcomings and on the whole, the evidence was inconclusive.15

In 1982 Wright and Rossi undertook ground breaking research of their own16 in which questionnaires were administered to almost two thousand convicted felons in correctional services facilities across the United States. Prisoners were questioned on the acquisition, carrying and use of guns and other weapons in criminal acts.

Of particular relevance is their finding that the prospect of meeting an armed victim seemed to have a fairly large deterrent value. About three fifths of their sample agreed that "a criminal is not going to mess around with a victim he knows is armed with a gun." Four fifths agreed that "a smart criminals always tries to find out if his potential victim is armed." 1?

One of their most oft quoted findings is that three fifths of their sample were more concerned about meeting an armed victim than running into the police.18 While the vast majority of felons were not concerned about being arrested or imprisoned, they were certainly concerned about the prospect of meeting an armed victim.19

It also appears that this concern actually affected criminal behaviour. Data reported later indicated that about two fifths of the sample had at some time chosen not to commit a crime because they knew or believed their potential victim to be armed.20 One third reported that they had actually been "scared off, shot at, wounded or captured by an armed victim."2t

15 Ibid., 12. 16 James D. Wright and Peter H. Rossi, Armed and Considered Dangerous: A Survey of Felons and their Firearms (New York: Aldine De Gruyter, 1986). 17 Ibid., 145. 18 Ibid., 147. 19 Ibid., 144-145.
20Id. 21 Ibid. 154. 12

Wright and Rossi's findings support the claim that civilians armed with firearms constitute a significant deterrent to the criminally inclined. Their work suggests that defensive gun use may be both effective and fairly frequent, judging from felon's encounters with armed victims.
Kleck and Gertz 22 set out to determine how frequent defensive gun use (DGU) actually is in the United States.

They conducted a comparatively large telephonic survey amongst the American population, making a special attempt to correct the errors and methodological shortcomings of previous surveys. Their results indicate that there are approximately 2.2 to 2.5 million incidents of DGU in the United States each year. If this is the case then incidents of DGU outnumber the use of guns to perpetrate a crime, four to one.

These results imply that restrictions imposed on civilian ownership of firearms in order to reduce incidents of violent crime, would be counter-productive.

While Kleck and Gertz's results do not shed light on the frequency of DGU in South Africa, they do suggest that DGU may be more frequent than often asserted. Given the weighty consequences of the relative frequency of DGU, it is vital that a more concerted effort be made to establish the regularity with which this phenomenon occurs in South Africa.

Lott and Mustard's research23 is consistent with the findings of Kleck and Gertz. Using cross-sectional time-series data for U.S. counties from 1977 to 1992, Lott and Mustard found that allowing citizens to carry concealed weapons deterred violent crime. When state concealed handgun laws went into effect in a county, murders fell by 7.65 percent, rapes by 5 percent, aggravated assaults by 7 percent. Lott and Mustard also found evidence that criminals were deterred from confrontational crimes, rather engaging in less confrontational property crimes.24

22 Gary Kleck and Marc Gertz, "Armed Resistance to Crime: The Prevalence and Nature of Self-Defence with a Gun," in Journal of Criminal Law and Criminology vol. 86 (Fall 1995). 23 John R. Lott, Jr. and David B. Mustard, "Crime, Deterrence, and Right-to-Carry Concealed Handguns," in Journal of Legal Studies vol. 26 (1997) ( (15/08/2000). 24 Ibid., 1.



As Wright, Rossi and Daley point out it is unwise to take the literature on weapons and crime as a lesson in scientific objectivity.25 Firearm violence and gun control are very emotive issues and research has often been conducted by stakeholders already firmly entrenched on one side or other of the great divide. Much of the research is genuinely impaired by methodological shortcomings. It appears that in some cases the research design itself has predetermined the outcome, thus undermining the validity of the research findings.

However, this criticism is also blatantly utilised by some merely to discredit any research that opposes their point of view. Despite claims to the contrary, it does appear that Kleck and Gertz 26 and Lott and Mustard 2? have managed to overcome the methodological flaws hampering much of the research in this area.

In South Africa it appears that much research has been done specifically with policy makers in mind. Although there is definitely a need for policy relevant research, this has resulted in a bias towards formulating the somewhat contradictory research findings into a coherent public policy. As this dissertation will highlight, effective firearm control may simultaneously promote and hamper violent crime in South Africa. The net effect of these opposing tendencies, remains to be seen.

25 Wright, Rossi and Daly, Under the Gun, 3. 26 Kleck and Gertz, Armed Resistance to Crime. 27 Lott and Mustard, Crime and Deterrence.



Policy responses to South Africa's burgeoning crime problem need to be grounded in an accurate understanding of the current realities. Statistical evidence, despite its shortcomings, provides important information on the crime situation and can be used to inform crime-fighting efforts.


There are three main sources of crime statistics: police or government department crime statistics, victim surveys and self-report surveys,1 each of which is subject to inherent shortcomings. In South Africa self-report surveys have been under-utilised and currently most statistical information is generated by (he police and victimisation surveys.2

1.1 Police Statistics

Until recently, the South African Police Service (SAPS) Crime Information Analysis Centre (CIAC) published semester and monthly reports on the twenty most serious crime tendencies in South Africa,3 Crimes pertinent to this discussion fall mainly under the classification of violent and social fabric crimes. The category "violent crime" comprises murder, attempted murder and robbery with aggravating circumstances.

Robbery with aggravating circumstances includes car hijacking, hijacking of trucks, robbery of Cash-in-transit and bank robbery as well as other acts of robbery involving a high level of violence or the threat thereof.

Caution should be exercised when using a compilation of data gathered from various sources as definitions and classifications are not standard across the board. Crimes are sometimes classified differently by the police and victim surveys, making data from these sources not directly comparable.

2 Some medical statistics, such as the number of patients with firearm related injuries treated in hospitals, can be used to supplement or verify official crime statistics. These should also be used with caution as only the injury sustained is recorded and not the circumstances surrounding the incident. It is these circumstances which would determine whether the act which inflicted the wound should be considered a criminal act or a non-criminal act such as in the case of self-defence or a lawful police shooting.

3 Steve Tshwete, the Minister for Safety and Security, made a public announcement on 20/7/2000 declaring a moratorium on all crime statistics for the year 2000 onwards. 15

"Social fabric crimes" cover rape, common assault and assault GBH (assault with intent to inflict grievous bodily harm).4
Whilst no doubt every effort is made to ensure the accuracy of this data, there are reasons that official police statistics should be used circumspectly.

The first reason is that of underreporting to police. This is not a peculiarly South African phenomenon, but given the historical mistrust of the police, underreporting may be especially high at least with regard to certain crimes.5 Secondly, in South Africa police statistics reflect only those cases in which a docket has been opened. Where the victim is related to the perpetrator, the victim often decides to drop the charges and these cases are then not reflected in police statistics.6

There are also crimes, such as illegal possession of firearms, which are heavily dependent on the police for detection as there is no complainant to report (he crime. An increase or decrease in these figures may reflect more on policing and detection than on an actual fluctuation in the incidence of this crime. In any event these statistics reflect only those crimes which the police uncovered which may constitute a fraction of the real number of criminal incidents.

For the reasons cited above, caution needs to be exercised when utilising police statistics to calculate the exact extent of any particular crime. However, assuming that, all other things being equal, underreporting and other errors are fairly consistent over time; police statistics are fairly reliable indicators of short-term crime trends. The CIAC reports also provide valuable information on some of the circumstances surrounding selected crimes. This proves beneficial when devising intervention strategies.

4 SAPS CIAC, "The Crime Situation at National, Provincial, Area and Station Level," in Semester Report 1/2000 fhttp^/ (13/08/2000).
5 Lala Camerer et al., Crime in Cape Town: Results of a City Victim Survey in ISS Monograph no. 23 ( ... rtmg.htnil) (23/04/2000).
6 Englander, Understanding Violence, 13.

Unfortunately police databases were not designed for analytical purposes. They do not take account of explanatory variables and do not link crime statistics or occurrences with socio-economic, demographic and policing data7. If this could be rectified it would aid in ascertaining the relationships between different types of crime and some of these variables. In the mean time victim surveys provide a valuable source of information on the relationship between the incidence of crime and some of these explanatory variables.

1.2 Victim Surveys

Victim surveys such as those conducted by the Institute for Security Studies (ISS) in 1998 are the second source of crime statistics.8 These surveys have the advantage that victims may be more forthcoming with a civilian interviewer than with the police. Victim surveys also provide valuable information about the relationship between victimisation and life circumstances such as socio-economic status, geographic variants, and demographic variables on which official statistics are often mute.9 Being able to identify who is most at risk (as these surveys interview both those who have been victimised and those who have not) is very valuable in identifying the groups most vulnerable to certain crime types.10 Victim surveys can also aid in establishing the magnitude of underreporting to the police. In this way they are able to help provide a more accurate picture of the full extent of crime.11

The main shortcoming of these surveys is that human memory is fallible. The respondent's reporting may be inaccurate especially if the crime was perpetrated some time before the survey was conducted.12 Also the results generated by various surveys may differ because of disparate sampling methods.
7 SAPS CIAC, "The Generators of Crime in South Africa," in Semester Report 1/2000 ( (13/08/2000).
8 Camerer et al, Crime in Cape Town and Antoinette Louw and Mark Shaw et al.. Crime in Johannesburg:
Results of a City Victim Survey, Monograph no. 18
( ... ctims.html) (15/08/2000).
9Ibid., 3.
10 Antoinette Louw, "Comparing Crime in South Africa's Major Cities: Results of Four City Victim Surveys," in
African Security Review, vol. 8 no.1 (
1 According to Camerer et al., Crime in Cape Town, the levels of reporting ranged from 93 % in the case of
murder to 36 % in the case of sexual assault and as low as 25 % in the case of sexual harassment.
12 Englander, Understanding Violence, 14-15.

1.3 Self-Report Surveys
Crime statistics can also be gathered from self-report surveys where perpetrators themselves divulge information about crimes that they have committed and other pertinent information.13

It is improbable that hardened criminals will voluntarily confess to their crimes. However, conducting these surveys amongst the general public is useful in terms of eliciting information about crimes like domestic violence where questions can be framed in such a way that people do not feel they that are implicating themselves or making themselves liable to prosecution.

Surveys can also be done amongst convicted felons,14 which may provide valuable insight into the mentality and motivation of the criminal. Unfortunately, this kind of survey has proved difficult in South Africa. A project was launched by the CIAC to establish the role that HIV/AIDS myths play in the perpetration of rape, but had to be aborted because of a lack of co-operation by the offenders.15

The International Arrestee Drug Abuse Monitoring Programme (I-ADAM) only recently commenced in South Africa but appears to be having more success. The pilot studies all indicate a strong double nexus between alcohol and /or drug abuse and crime. The contention is that often people commit crime under the influence of alcohol and/or drugs. Their habit may also become very expensive to maintain and in this way "impel" them to participate in criminal activities.16

It would be very valuable if self-report surveys could be utilised to shed some light on various blind spots in our knowledge about criminality in South Africa. For instance, the efficacy of deterrents such as sentence length, arrest and conviction rates, or an armed citizenry, the primary motivations for criminal involvement in certain crimes, incentives and disincentives of a life of crime and the ease with which weapons can be obtained, could be explored.

13 Ibid., 15.
14 Such as the survey conducted by in the U.S.A. by Wright and Rossi, Armed and Considered Dangerous.
15 SAPS CIAC, "Statistical Analysis of Reported Rape Cases." in Semester Report 1/2000 ( 2000/rapeanalysis.htm) (13/08/2000).
16 SAPS CIAC, "Generators of Crime."


In South Africa the level of violent crime is extraordinarily high and of utmost concern to the government and population alike17 Approximately one third of all reported crimes in South Africa are violent in nature, including murder, attempted murder, robbery, assault (common and GBH) and rape. The murder rate stands at about eight times the international average18 and South Africa is reported to have one of, if not the, highest rape rate in the world.19

The need to develop effective strategies to combat violence in South Africa requires that violent crime be dis-aggregated and that the specifics surrounding the incidence of each crime type be understood. It is contended that the dynamics surrounding a gang rape differ substantially from those involved in a bank robbery or cash-in-transit heist. If this is the case then no "catch all" policy will effectively deter all types of violent crime equally.

However, there are commonalties that all gang rapes or bank robberies share that are relevant to preventative initiatives. Compiling a profile of the typical incident of each type of violent crime aids in tailor making interventions to address that particular crime type.

For the purposes of evaluating the potential impact of strict firearm control on violent crime, it is important to establish the set of violent crimes routinely perpetrated with a firearm.

It is unreasonable to contend that a crime not perpetrated with a firearm would have been precluded by preventing the perpetrator having had access to a firearm. Whilst any crime can potentially involve the use of a firearm, police classification distinguishes between crimes on the basis of the level of violence involved. Weapon use is one of the factors taken into account.

Common assault, by definition, excludes the use of a weapon and is, therefore, omitted from further analysis in this regard. Assault GBH could conceivably involve use of a firearm, but in reality all non-fatal attacks with a firearm are recorded as attempted murder.20 Thus, using official police categories those crimes that could potentially involve the use of a firearm are murder, attempted murder, rape, robbery with aggravating circumstances and attempted robbery with aggravating circumstances.

17 Inter-departmental Strategy Team, NCPS, 4; Helen Taylor and Robert Mattes "The Public Agenda" in Opinion 1999, Press Release, 24 May 1999 ( ..sd5.1 ff319&NS-doc-offset==0) (15/05/00).
18 Nedcor, Nedcor Project on Crime, 6-7.
19 SAPS CIAC, "Annexure E" to Semester Report on the Incidence of Serious Crime in South Africa January to December 1998, Semester Report 1/99 (Pretoria, 1999).

For the purpose of developing intervention strategies, it is useful to classify violent crimes on the basis of the victim-perpetrator relationship and the degree of organised intent. Following the World Health Organisation 21 violence can be classified as:
• Intimate Violence (abuse of children, spouse or partner)
• Acquaintance Violence (such as between drinking partners, date rape)
• Stranger violence (most aggravated robbery)
• Organised Violence (such as factional conflicts and taxi-violence)
• Self-directed violence (suicide and self-mutilation)

These categories will be utilised in the following examination of violent crime in South Africa in an attempt to establish the relative frequency of each type of violence for each relevant crime type.22 Attempting to determine the proportion of murders, robberies etc that are a result of each type of violence, aids in formulating an appropriate intervention.

2.1 Murder and Attempted Murder

CIAC figures indicate that murder and attempted murder decreased during the period 1994 to 1999.23 Given the high degree of the reporting of murder,24 of all the crime statistics available, these figures are probably the most accurate.
20 Chetty, "Firearm Crime in South Africa," 26.
21 The World Health Organisation, Violence: A Public Health Priority (Unpublished Memo) cited in Alexander Butchart, Johan Kruger and Victor Nell, ''Neighbourhood Safety: A Township Violence and Injury Profile," in Crime and Conflict no. 9 (Winter 1997), 13.
22 As previously explained, self-directed violence will be excluded from analysis.
23 SAPS CIAC, "Table 1," in Semester Report 1/2000 ( .htm) (13/08/2000).
24 Hirschowitz et al, Victims of Crime Survey, (Pretoria: Statistics South Africa, 1998). 53 and Camerer et al. Crime in Cape Town.

Murder decreased in real figures and in terms of the ratio per 100 000 in the period 1994 to 1999, whilst for attempted murder the actual number of attempted murders was in fact higher in 1999 than 1994.25 However, given the population increase in those five years the ratio of attempted murders per 100 000 decreased. The decrease in murders may be partially related to a decrease in political violence that began to wane in the years subsequent to 1994.26

An interesting feature of the murder and attempted murder figures is that while murder decreased from 69.5 per 100 000 in 1994 to 55.3 per 100 000 in 1999, attempted murder decreased from 70.7 per 100 000 in 1994 to 66.6 in 1999.27 Thus the rate of murder has dropped to a much greater degree than the rate of attempted murder. This would indicate that there has been a much smaller decrease in potentially fatal attacks, than the murder figures alone would appear to indicate.

The national Victims of Crime Survey revealed that 73.4 percent of all murders were perpetrated with a weapon. Of those perpetrators carrying a weapon, 58 percent were armed with a gun and 42 percent with a knife or other sharp object. Where the perpetrator was unarmed, death was by beating, assault, throwing off train, strangulation etc.28

SAPS data indicate that whilst murder and attempted murder have been on the decrease since 1994, the number of murders committed with firearms has been steadily increasing from 41 percent in 1994 to 49 percent in 1998. Interestingly, the use of firearms utilised in attempted murders has been steadily decreasing from 86.7 percent of all attempted murders in 1994 to 74.7 percent in 1998.

** It should be noted that the statistics concerning the use of weapons in attempted murders are somewhat misleading as almost all non-fatal attacks with a firearm are recorded as attempted murder whereas attacks with a knife or other weapon tend to be recorded as assault GBH **.29
25 SAPS CIAC, "Table I."
26 Mark Shaw, "Violent Crime in South Africa," in Firearm Use and Distribution, ed. Robert Chetty, 13.
27 SAPS CIAC, Table 1."
28 Hirschowitz et al. Victims of Crime Survey, 50-51.
29 Chetty, "Firearm Crime in South Africa," 26.

2.1.1 The Provincial Breakdown

An analysis of murder dockets by the CIAC office in the Eastern Cape during 1996, indicates that only 4 percent of the murder cases could be attributed to factional conflicts, 3 percent to taxi related violence and 1 percent to gang related violence. The remaining 92 percent of murder cases were predominantly associated with drug and alcohol abuse. Only 34 percent of all murders under consideration were perpetrated with firearms.30

Data from the Northern Cape demonstrate similar findings. Alcohol and family disputes seemed to characterise these cases. Most murders occurred on Saturdays, especially Saturday evenings. Of all murder victims, 68 percent were stabbed, 52 percent with a knife and the remainder with another object. In the vast majority of cases, no premeditation was evident.31

Docket analysis done in the Free State indicates the same tendencies. Most murders were committed with knives or other objects and firearms were used in a minority of cases. The most common motive for the murders was personal or inter-personal problems such as marital problems, jealousy and self-defence. In the majority of cases either the perpetrator or the victim, or both, were under the influence of alcohol when the act occurred. Many murders also occurred in or in the vicinity of drinking places.32

The CIAC office in Mpumalanga confirms the patterns observed in other provinces: 59.1 percent, of the murder cases analysed at the beginning of 1997, occurred in shebeens and other drinking places. In an additional 20.1 percent of the cases the evidence proved inconclusive but it appeared that the victims were murdered on their way to or from drinking places. Evidence suggests that only 15.7 percent of cases may have involved premeditation.

30 SAPS CIAC, "A Selection of Research Results Received from Provincial Crime Information Management Centre Offices" in Monthly Report 2/97(http^/ (13/08/2000).
31ld 32Id.

Data from the Western Cape in 1999 indicate similar tendencies. Most murders took place on Saturday evenings. More than half the murders were stabbings with a knife or other sharp object. Only one quarter were firearm-related homicides.33

The information gathered from the provinces discussed above indicates that the dominant feature of murder in these provinces is alcohol abuse and intoxication coupled with interpersonal disputes. This is partly a consequence of the historic 'tot system" whereby farm workers were compensated (or at partly compensated) with wine. Despite the termination of the system, the alcoholic tendencies that it engendered remain. Geographic crime pattern analysis reveals that those areas in which the "tot system" once prevailed, have consistently experienced the highest ratios of assault, rape and murder.34

It appears that most cases of murder occur over the weekends and especially in the festive seasons, often in and around drinking holes. Most perpetrators and victims are known to each other and more often victims are stabbed rather than shot. It can be reasonably concluded that in the provinces discussed above, murder is predominantly the result of intimate or acquaintance violence, In the most typical scenario inferred from available evidence, an argument occurs and leads to a common assault.

This escalates as one or more party is under the influence of alcohol and, therefore, less rational, less inhibited etc. At least one party lays their hands on the first available weapon (most often a knife or broken bottle) and the end result is homicide.35

The breakdown of typical characteristics surrounding murder and attempted murder in Kwa-Zulu Natal and Gauteng may differ in significant aspects from the remainder of South Africa. It is possible that the proportion of stranger and organised violence is higher in these provinces. However, at this point the breakdown of murder statistics into intimate, acquaintance, organised and stranger violence etc. is not known.

Generally, the inhabitants ofthe North- Eastern part of the country seem to be less inclined to endemic excessive alcohol consumption and concurrent interpersonal fights.36 Of all the provinces in South Africa, it is only in Kwa-Zulu Natal and Gauteng that firearm murders exceed murder by means of other weapons. Firearm murders in these two provinces together account for about 65 percent of all firearm-related murders and 49 percent of the total murders in South Africa.37

33 Western Cape SAPS CIAC, Semester Report 2/99 (December 1999), 5,
34 SAPS CIAC, "Generators of Crime."
35 Id.

Attempted murder in general appears to be less strongly associated with alcohol abuse than murder. Attempted murder is predictably high in areas experiencing factional/political/ ideological conflicts such as the Natal Midlands, Umzimkulu and Durban as well as areas where gang warfare and gang- PAGAD conflicts occur, such as the Cape Town Metropolitan area.

Attempted murder is also high in Gauteng especially in Johannesburg, Soweto and the East Rand, where armed robbery (most commonly hijackings and robbery at residential premises) is at its highest level.38
Kwa-Zulu Natal and Gauteng are the main financial centres of South Africa, are more densely populated and have high unemployment and crime rates.39 As a result these two provinces also have a disproportionate share of aggravated robbery and the violence that accompanies it.40

However, contrary to popular perception and despite the fact that the vast majority of robbers are armed, police reports indicate hijacking of trucks and cars, bank robberies and cash-in-transit heists result in less than one percent of all homicides in South Africa,41 It is unknown how many other cases of serious robbery result in the death of the victim.
36 SAPS CIAC, "Generators of Crime."
37 Chetty, "Firearm Crime in South Africa," 22-23.
38 SAPS CIAC, "Crime Situation."
39 Ettienne Hennop, "Illegal Firearms in Circulation in South Africa," in Society Under Siege, ed. Virginia Gamba, 29.
40 Ibid., 30.
41 SAPS CIAC. "Generators of Crime."

On the basis of the available, yet somewhat limited evidence, it is asserted that the majority of murders in South Africa are the result of intimate and acquaintance violence,42 very often characterised by intoxication and interpersonal disputes. Organised violence and stranger violence may form a higher proportion of the murder and attempted murder rate in Kwa-Zulu Natal and Gauteng than in other provinces, but it is doubtful that for South Africa as a whole the proportion exceeds 15 percent.43

This can only be postulated, and not proven conclusively because official murder and attempted murder statistics are not routinely analysed on the basis of the victim-perpetrator relationship and the degree of organised intent. As a result, official figures for murder and attempted murder broken down into gang-related incidents, factional conflicts, homicidal robbery and interpersonal violence etc., are not easily available.

2.2 Rape

According to the CIAC, the incidence of rape stabilised during the period 1994 to 1999 at 119 reported rapes per 100 000 of the population in 1999. In 1998 the CIAC reported a national ratio of 234.6 rape cases per 100 000 females.44 The SAPS themselves estimate that only one in thirty-five rapes is reported. If this is the case, then the incidence of rape would stand at 8211 per 100 000 of the female population or 4165 incidents of rape per 100 000 of the total population.45 Various victim surveys indicate that underreporting for sexual offences could range from 55 percent to 66 percent.46 Based on this, a very conservative estimate would put the actual incidence of rape at about 250 cases per 100 000 of the total population or 450 to 500 per 100 000 of the female population.

It is highly likely that real number of rapes in South Africa, exceeds these estimates, which, in any event is inordinately and unacceptably high.
42 It should be noted that "acquaintance" does not denote someone with whom the victim was necessarily on friendly terms. This term could be used to describe an opposing gang member, a prostitute, shopkeeper, taxi-drivers client etc. as well as a drinking buddy or social acquaintance. Given the extremely broad definition of the term it could be used to describe almost anyone that the victim ever had any kind of contact with. Hammond, "UCA Submission on Bill."
43 Wesson, "Murder and Private Firearms."
44 SAPS CIAC, "Rape Analysis."
45 Bollen et al, Introduction to Violence Against Women in Metropolitan South Africa: A Study on Impact and Service Delivery, ISS Monograph no. 41 (1999). ( ... ction/html) (03/08/2000)
46 Hirschowitz et al. Victims of Crime 58 ; Camerer et al., Crime in Cape Town and Antoinette Louw et al., Crime in Johannesburg.

Victim surveys indicate that 46 percent of perpetrators of sexual offences were armed.47 Of those armed, almost 80 percent were armed with a knife and about 15 percent with a firearm.48 This would mean that overall about 8 percent of sexual offences are committed by a perpetrator armed with a gun.

Conversely, 92 percent of rapes in South Africa are committed without the perpetrator being armed with a gun. This confirms Kleck's finding 49 that generally perpetrators of rape are not armed with a firearm.

The fact that about half of all rapes are committed without the perpetrator's use of any weapon at all testifies to the fact that in many cases inherent superior physical strength alone is sufficient to ensure successful execution of this crime. Of those perpetrators who were armed, the vast majority were armed with a knife. It is submitted that the very nature of this crime makes utilisation of a firearm less than efficacious.

The use of a firearm may be more apt where there are multiple perpetrators such as in the case of gang rape. Whilst this cannot be proven, it is submitted that in those small number of cases where a firearm was utilised, another weapon would have been substituted should a firearm have been unavailable.

Research results based on rape cases in 1996 from the Western and Northern Cape and the Free State indicate that rape also exhibits the cyclical nature of other interpersonal violent crimes and peaks nationally over weekends, especially long weekends, and the holidays around Christmas and Easter. About a third of the suspects were under the influence of alcohol at the time of the rape. According to the police about 60 percent of the suspects were known to their victims, which implies a high degree of acquaintance rape.50
47 Hirschowitz et al., Victims of Crime, 50.
48 Ibid., 51.
49 Gary Kleck, "Policy Lessons from Recent Gun Control Research," in Law and Contemporary Problems, 49 (1986) in Hansson, "Guns and Control," 18. 50 SAPS CIAC. "Selection of Research Results."

According to the CIAC some of the most important aspects of South Africa's high incidence of rape are the excessive consumption of alcohol and drug abuse, societal tolerance of violence towards women, HIV/AIDS myths that intercourse with a virgin can cure the infected and the initiation rites of gangs.51

Given the complex societal factors contributing to the high incidence of rape in South Africa and the very low rate of firearm utilisation in the perpetration of rape, the available evidence suggests that firearm availability is not a decisive factor in determining the incidence of rape in South Africa.

Thus, it can reasonably be assumed that firearm control is unlikely to significantly decrease the incidence of rape in South Africa. However, given that a firearm in the hands of the would-be victim can potentially be used to deter a rape, it is contended by Lott and Mustard 52 as well as Kleck and Gertz53 that firearm control may in fact increase the incidence of violent crime including rape. The argument for defensive gun use will be examined in some detail at a later stage.

Consequently, in further analysis of the Firearms Control Bill's potential to decrease violent crime in South Africa, rape will be excluded.

2.3 Aggravated Robbery
Aggravated robbery encompasses a fairly diverse group of crimes including all acts of robbery accompanied by high levels of violence or the threat thereof. Cash-in-transit heists and bank robberies make up a very small proportion of all aggravated robberies.54 Hijacking of trucks and cars are much more common and comprise about one fifth of all cases of serious robbery.55 Robbery with aggravating circumstances is rapidly on the increase and rose more than four times its expected population growth linked increase, in the period 1994 to 1999.56
51 SAPS CIAC, "Rape Analysis."
52 Lott and Mustard, "Crime and Deterrence."
53 Kleck and Gertz, "Armed Resistance to Crime."
54 Shaw. "Violent crime in South Africa," 16-17.
55 Robert Chetty, "Summary," in Firearm Use and Distribution, 10.
56 SAPS CIAC, "Crime Situation."

Property related crimes are generally perpetrated by strangers to the victim and typically involve the use of a weapon. Whilst street robbery may frequently be perpetrated with a knife, the firearm is the weapon of choice in most other types of aggravated robbery. Firearms were used in 85 percent of all serious robberies in 1998.57 This is presumably because firearms tend to be very intimidating and are more likely to ensure the victim's compliance. They also provide maximum protection against armed victims.

Hijacking of trucks, carjacking, cash-in transit heists and bank robberies are most often perpetrated by crime syndicates operating mainly in the large metropoles. Contrary to popular perception, despite the fact that over 90 percent of hijackers are armed, less than 1 percent of hijackings result in a fatal shooting. It appears that hijackings result in about sixty murders of the total ± 25 000 murders committed per year (that is only 0.2 percent of the total number of murders). Bank robberies and cash-in-transit heists result in the death of ± 100 people (quite possibly some fatalities are those of the perpetrators), which represents 0.4 percent of all murders.58

Armed robbery has spread to all comers of the country but remains concentrated in Gauteng and Kwa-Zulu Natal. These two provinces account for 73.1 percent of all firearm robberies that occurred in 1998.59 As previously mentioned this could be because these provinces are the main financial centres, more densely populated and have high unemployment and crime rates.60 Their geographical proximity to the borders of South Africa would also facilitate the movement of stolen goods and contraband out of the country.
57 Chetty, "Firearm Crime in South Africa," 27.
58 SAPS CIAC, "Generators of Crime M
59 Hennop, "Illegal Firearms in Circulation," 30.
60 Ibid. 29.


The preceding analysis of violent crime in South Africa suggests that the vast majority of murders, assaults and rapes, as well as some attempted murders, can be classified as intimate and acquaintance violence.

Organised violence and stranger violence may form a higher proportion of the murder and attempted murder in Kwa-Zulu Natal and Gauteng than in other provinces, but for South Africa as a whole the proportion probably does not exceed 15 percent.61 The vast majority of armed robberies can be classified as 'stranger violence'.

The seasonal nature of most "violent crimes against the person" is very telling. There is a definite peak around the Christmas holidays in December and January and also on long weekends, especially the Easter weekend. These violent incidents tend to be characterised by excessive alcohol consumption and inter-personal disputes. This suggests that interventions aimed at preventing these crimes would need to address the excessive abuse of alcohol and the culture of violence in South Africa, if they are to be at all effective.

In South Africa, those violent crimes most commonly committed with firearms are armed robberies, murders and attempted murder. Thus, for the purposes of analysing firearm policy interventions, a focus these particular crimes can be justified.
61 Wesson, "Murder and Private Firearms." 32
"I'd rather have a gun and not need it than vice versa."
Posts: 84
Joined: Thu, 2006-02-02 13:56
Location: Kempton Park, Gauteng, South Africa

Return to Educational Articles

Who is online

Users browsing this forum: No registered users and 1 guest