The framers of South Africa's crime prevention policies are clearly of the persuasion that firearms in civilian hands do more damage than good. The NCPS asserts that "the number and easy accessibility of firearms is a major contributor to violent crime."... "The fact that a large proportion of the citizenry is armed,¹ serves to escalate the levels of violence associated crime." ²

However, it appears that many South African citizens are not convinced. Most civilians who own guns cite self- defence as at least one of, if not the, motivating factor/s.³ Indeed, the alleged frequency and efficacy of defensive gun use (DGU) is one of the main reasons for opposition to stringent restrictions on civilian ownership of self-defence weapons. It also constitutes a large part of the argument disputing the alleged correlation between the prevalence of privately owned firearms and violent crime.

Gun control advocates respond that the perception that a firearm provides increased security is misguided. They assert that 'the cases of people attacked and robbed of their guns, and often shot with their own guns, far outnumber the occasions when somebody successfully defends themselves with a gun." ⁴

For any meaningful discussion on gun control to take place, both sides must admit that firearms can be used, both to injure and take life, as well as to prevent injury and protect life. Whether firearms in the hands of civilians prove ultimately to be hazardous or helpful, depends on the relative frequency (and efficacy) of legitimate gun use and illegal abuse.

If the sum total of violent crimes perpetrated with firearms that were once legally licensed to civilians, exceeds the number of incidents in which privately owned firearms are used to deter violent crime, then effective gun control would, on the whole, save lives. However, if armed civilians deter more violent crimes than incidents perpetrated with previously licensed civilian weapons, then the net effect of gun control would be a greater loss of life and limb. Thus proving or discrediting the efficacy and frequency of defensive gun use becomes vital to the case of those on both sides of the gun control debate.

1. THE EFFECTIVENESS OF DEFENSIVE GUN USE

There are some spectacular examples of private citizens using their licensed firearms to ward off criminal attack. One such example is the assault on St. James Church on 25 July 1993 where APLA terrorists killed eleven people and wounded fifty-five. One of the congregants returned fire wounding one attacker and the heavily armed terrorists took flight. The police applauded the congregant's prompt action, contending that fatalities would have been much higher had Charl van Wyk not taken the decisive action that he did.⁵

From examples such as this one, it is clear that a firearm can be used very effectively to ward off a criminal attack. However, the gun control lobby would presumably argue that this is the exception rather than the rule. In responding to claims that law-abiding citizens require firearms to protect themselves from armed criminals, gun control advocates claim that having a gun does not in fact make one more secure. If attacked by an assailant, simple compliance is supposed to be a safer strategy than resistance. Defensive gun use supposedly escalates the conflict as armed attackers will open fire if fired upon, and this will increase the risk of injury to oneself and bystanders.⁶

This preference for compliance is most often supported by reference to the United States Department of Justice's National Crime Victimisation Survey. If passive behaviour is compared to all forms of active resistance lumped together, passive behaviour is indeed slightly safer than active resistance. However, the category "active resistance" includes all kinds of resistance like using one's fist, kicking, yelling and screaming, running away, using Mace, a baseball bat, a knife, or a gun. Many of these behaviours are very ineffective and certainly much more dangerous than passive behaviour. However, using a gun is not.

In the case of a woman who is attacked by a stronger man, fighting back with her fists is the most dangerous course of action as it is likely to elicit a physical response from the attacker and increase the chances of serious injury or death to the woman. For women, by far the safest form of defence is to have a gun.

By examining data from the National Crime Victimisation Survey, Southwick found that the probability of sustaining serious injury from an attack is 2.5 times greater for women offering no resistance compared to those resisting with a gun. For those who resist without a gun the probability is four times as high than for those resisting with a gun.⁷ For men, behaving passively is 1.4 times more likely to result in serious injury than resisting with a gun. Resistance without a gun is 1.5 times more likely to cause injury than resistance with a gun.³

For the prevention of rapes specifically, it is highly advantageous for a women to be armed. Another U.S. Justice Department study of more than thirty-two thousand rapes or attempted rapes, found that the best protection against rape is for a woman to be armed. When a potential victim of rape was armed with a firearm or knife, only 3 percent of the attempted rapes succeeded.⁹

Kleck's research reinforces that the safest response when confronted by an armed robber is for a victim to display his/her own firearm. In 98 percent of the cases examined, the mere brandishing of the firearm proved sufficient to deter the attacker. Potential victims who defended themselves with a gun suffered lower rates of injury than did those who resisted without a gun, or even those that complied fully with the criminal's demands.¹⁰

A study conducted by John Mann in 1994 researched 206 cases of DGU (excluding police or military personnel) reported in the Gauteng area. Despite victims usually being outnumbered by attackers (on average 2,75 to 1) victims armed with a gun managed to defend themselves very effectively against armed assailants. 36 percent of attackers were killed or arrested by the potential victims and the other 64 percent of attackers fled. Only eight victims died and no bystanders or innocent parties were injured by actions taken by defenders.¹¹

Gun Free South Africa relies on the research conducted by Altbeker to assert that an individual is four times more likely to have their weapon stolen than to use it in self-defence. Altbeker examined 602 police dockets of criminal incidents in Alexandra and Bramely, of which 506 involved use of a firearm. His findings included that fifty of the victims were known to have been carrying guns at the time of the incident. Of these fifty, thirty-nine had their weapons stolen. In only twenty-four cases was a firearm actually used by the victim or third party in defence.¹² Gun possession, it is proposed, actually increases a person's chance of attack as criminals may specifically target gun owners in order to obtain the firearm.¹³

However, by utilising police dockets of cases in which the perpetration of a crime was successful, Altbeker excludes all those cases in which a gun was actually used to prevent a crime from occurring. Altbeker himself acknowledges that his findings are not generalisable and, therefore, cannot be used to estimate the frequency or efficacy of defensive gun use in South Africa.

2. THE PREVALENCE OF DEFENSIVE GUN USE

On the basis of the evidence presented in the previous section, it appears that using a firearm in self-defence is one of the safest and most effective strategies that can be employed if attacked by an assailant. There appears to be no foundation for the claim that compliance is a safer strategy than defensive gun use. However, it is also vitally important to establish how frequently private citizens actually do deter violent criminals.

In the United States, fairly comprehensive research has been done in this area. One of the interesting features of this research is the large discrepancy in the incidence of defensive gun use that respective surveys report.

Using the National Crime Victimisation Survey, Phillip Cook contends that annually there are only between eighty and eighty-two thousand defensive uses of guns during assaults, robberies, and household burglaries.¹⁴ By contrast, other surveys imply that private firearms may be used in self-defence up to two and a half million times each year, with four hundred thousand of these defenders believing that using the gun "almost certainly" saved a life.¹⁵

One of the most famous studies pertaining to DGU was conducted by Kleck and Gertz in 1995. They conducted an extensive telephonic survey amongst the American population, making a concerted effort to correct the errors and methodological flaws that marred previous surveys. The results from their survey indicate that in America there are over two million protective uses of firearms each year. Their analysis of ten other nation-wide polls implies that there are 760 000 to 3. 6 million defensive uses of guns per year.¹⁶ Thus in all likelihood the real incidence of DGU is much higher than the National Crime Victimisation Survey estimates.

While the incidence of defensive gun use abroad does not shed any light on the regularity of DGU in South Africa, these figures serve to indicate that defensive gun use could possibly be more frequent than gun control advocates would like to admit. Importantly the research findings from America suggest that it may be difficult to establish an objective estimate of the frequency of DGU. Reasons for this could include disparities with regard to the definition of DGU, variance in survey design and the framing of questions, as well as potential bias on behalf of researchers who could interpret their results in the light of pre-existing convictions.

Given the significance of DGU, one would expect extensive research to have been conducted in South Africa to measure the frequency with which law abiding citizens utilise their licensed firearms to ward off criminal attack. However, information on this is extremely scarce.

According to the Human Sciences Research Council survey, 2 percent of respondents indicated that they had used a firearm to defend themselves or a family member. As a proportion of the total South African population, this would suggest that there are approximately 81160 incidents of DGU (possibly per year as period not specified.)¹⁷ It is not known how many of the respondents actually owned a licensed firearm and were thus in a position to use a firearm in defence of themselves or their family.

By contrast, a reader survey conducted in South Africa by Magnum Magazine reported that private citizens used their licensed weapons in more than 2.5 million incidents over a two-year period. As in the American surveys, it was reported that in most cases attackers retreated at the sight of the potential victim's gun without any shots being fired by the potential victim.¹⁸ Given that Magnum Magazine caters largely for the armed population it is likely that a very high percentage of respondents owned licensed firearms. This could explain why the reported defensive gun usage was much higher in this survey than the HSRC one.

Given that the number of defensive gun uses has yet to be decisively determined, it is difficult to see how in can be asserted, as Leher and Hansson do,¹⁹ that a domestic firearm is more likely to be used on a friend or family member than on an intruder. Nonetheless this claim is often used to counteract the argument for defensive gun use. This assertion is made frequently by gun control advocates and is often based on research conducted by Kellermann and Reay.²⁰

However, the likelihood of a licensed firearm owner shooting a friend or family member relative to the likelihood of them shooting an intruder, does not say anything about how often firearms are actually used to defend and protect. This information may be pertinent to the relative advantages and disadvantages of a firearm in the home, but not to the frequency or efficacy with which that firearm may be used in a defensive manner.

3. THE DETERRENT EFFECT

The previous section established that in South Africa an accurate estimate of the incidence of defensive gun use remains elusive. Despite difficulties that may be encountered, victimisation surveys could possibly establish an accurate estimate of DGU at some point in the future. What is even more difficult to measure is the number of crimes not even attempted because of the possibility of encountering an armed victim.²¹

Based on the assumption that criminals are by and large rational actors, it can be expected that increasing the costs of crime relative to the potential benefits would deter people from engaging in criminal activity. This is known as the deterrent effect.²²

Given current high levels of violent crime in South Africa it would appear that violent criminals are not deterred by the threat of legal sanction. Even in America the threat of prison does not seem to effectively deter crime as most criminals do not expect to get caught. Over seventy five percent of those criminals interviewed by Wright and Rossi said that they were unafraid of getting caught at the time of the offence.²³ Opponents of strict gun control assert that the only effective deterrent is "the fear of being caught, injured or killed during the pursuance of an illegal violent act. A firearm is the only really effective deterrence against violent criminals." ²⁴ If a large proportion of the population is armed, this may create a "positive externality" for others. The nature of concealed handgun laws would mean that most of the time criminals would not know who was armed and who was not. Thus the greater the number of armed individuals, the greater the chance of any particular victim being armed, the greater the potential costs of crime to the perpetrator.²⁵ The increased cost could offset potential benefits and thus make criminals more wary of confrontational crime. Criminals may not be deterred from crime altogether but would possibly consider criminal activities that minimised confrontation between victim and perpetrator.²⁶

Wright and Rossi's research among convicted felons strongly supports the idea that armed victims constitute a real deterrent to confrontational crime. 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" and three fifths agreed that "criminals are more worried about meeting an armed victim than they are about running into the police." It is interesting to note that those criminals in states with the highest levels of civilian gun ownership appeared to be the most worried about armed victims.²⁷

About three quarters said that burglars tend to avoid breaking into occupied residences because they fear being shot and about one third had actually personally been "scared off, shot at wounded or captured by an armed victim." ²⁸ 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.²⁹ Data reported later indicated that about two fifths of the sample had at some time actually chosen not to commit a crime because they knew or believed their potential victim to be armed.³⁰

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 may be both effective and fairly frequent, judging from felon's encounters with armed victims.

Lott and Mustard's seminal study also strongly supports the notion that criminals respond rationally to deterrence threats. Their work is by far the most comprehensive study of the relationship between gun control laws and crime with data spanning eighteen years drawn from approximately three thousand counties in the U.S.A. Lott and Mustard examined the incidence of crime before and after the implementation of "shall issue" concealed carry laws, that demand that firearm licenses be issued to all qualified adults without discretion.

Their research finds unequivocally that increased civilian gun ownership deters confrontational violent crime. When concealed handgun laws were implemented the rate of murders fell by 7.65 percent, rapes by 5 percent and aggravated assaults by 7 percent.³¹ Concurrently, the incidence of non-confrontational property crimes rose after the implementation of "shall issue" laws. An additional decrease in violent crimes continued for years after the laws went into effect. The annual decrease in violent crimes averaged about 2 percent, while the annual increase in property crimes averaged about 5 percent.³²

Lott and Mustard conclude that criminals responded substantially to the threat of being shot by rather engaging in less risky criminal activities that minimised contact between victim and perpetrator. They stipulate that concealed handguns are the most cost-effective method of reducing crime analysed so far by economists, proving more effective than increased policing or incarceration, other private security devices, or social programs.³³

4. CONCLUDING COMMENTS

Despite the importance of Wright and Rossi and Lott and Mustard's findings, there are no indigenous studies that indicate whether, on the whole, South African offenders are deterred by the potential of meeting an armed victim. The incidence of robbery of a firearm (the use of force to take possession of the firearm) indicates that at least a number of South African criminals are not at all deterred by the prospect of an armed victim.

It is quite possible that both South African criminals and armed law abiding citizens differ in significant respects to their American counterparts. South African criminals may be more brazen and fool hardy. Law abiding citizens may be abysmal shots, ill trained or ill prepared to defend themselves. It is also possible that the circumstances under which crime is commonly perpetrated in South Africa may differ to those in the United States and may either facilitate or hinder the potential for DGU.

Unfortunately, at the moment these possibilities remain untested. Until such time as comprehensive indigenous research is conducted into these matters, debates on the frequency and efficacy of defensive gun use and the deterrent effect in South Africa will remain based on unqualified assumptions.

----------------------------------------------------

1 Interestingly only 6 percent of South African civilians legally own arms. Meek, "Legal Firearms in South Africa." 37.
² Inter-departmental Strategy Team, NCPS, 22.
³ Meek, "Legal Firearms in South Africa," 36.
⁴ Hammond, "UCA Submission on Bill."
⁵ Ibid.
⁶ GFSA. "Gun Free South Africa " (http://www.wits.ac.za/csvr/gunfree.htm) (25/08/2000).
⁷ Lawrence Southwick, Jr., "Self- Defence with Guns: The Consequences," in Managerial and Decision Economics (forthcoming) and Gary Kleck, Point Blank: Guns and Violence in America (Hawthorne, N.Y.:Aldine de Gruyter Publishers, 1991) cited in John Lott, Jr., More Guns, Less Crime: Understanding Crime and Gun Control Laws (Chicago: University of Chicago Press, 1998), 4.
⁸ Lott, More Guns, Less Crime, 4.
⁹ Law Enforcement Assistance Administration, Rape Victimisation, 1979 cited in Hammond, "UCA Submission on Bill."
10 Gary Kleck, "Crime Control," in Social Problems, (1998) in Hammond, "UCA Submission on Bill."
¹¹ Hammond, "UCA Submission on Bill."
¹² Altbeker, "Guns and Public Safety," 2.
¹³ GFSA, "Gun Free South Africa."
¹⁴ Phillip Cook, "The Technology of Personal Violence," in Crime and Justice: Annual Review of Research (1991) cited in Lott and Mustard, "Crime and Deterrence," 2.
¹⁵ Kleck and Gertz, Armed Resistance to Crime, 177.
¹⁶ Kleck and Gertz, Armed Resistance to Crime, 158, 182.
¹⁷ Robert Chetty, "Perceptions and Experiences of Firearm Use," in Firearm Use and Distribution; 63,
¹⁸ Juan de Greef, "The Role of NGO's in the Control of Light Weapons: A Case Study of the South African Gun Owners Association." in Society Under Seige. ed Virginia Gamba, 235
¹⁹ L.B., Leher and Desiree Hansson. "Smoking Guns and Public Health." South African Medical Journal vol. 83 (Sept 1993).
²⁰ Arthur L. Kellermann and Donald T. Reay, "Protection or Peril: An Analysis of Firearm Related Deaths in the Home," in New England Journal of Medicine vol. 314 no. 24 (1986) 1559.
²¹ Wright, Rossi and Daley, Under the Gun, 124.
²² Wright and Rossi, Armed and Considered Dangerous, 6.
²³ Wright and Rossi, Armed and Considered Dangerous, 144-145.
²⁴ Wesson, "Murder and Private Firearms."
²⁵ Lott and Mustard, "Crime and Deterrence," 4.
²⁶ Ibid-, 24,64.
²⁷ Wright and Rossi, Ibid. 144-145.
28 Ibid., 154.
²⁹ Ibid., 144-145.
³⁰ Id.
³¹ Lott and Mustard, "Crime and Deterrence," 19.
³² Ibid.. 34.
³³ Ibid., 65.