6 August 2018 – The State has served us with a no&#...

Read More


6 August 2018 - GOSA has been invited down to Parliamen...

Read More
Legal Fund Donations
R1 000 000.00
Updated: 10/24/2018 10:54:11 AM

More Guns, Less Crime: Does the Lott Model Work in South Africa?

Way back in October 2000, a gentleman by the name of Dr. Richard Wesson wrote a research paper placing the conclusions of Prof. John Lott's More Guns, Less Crime in a South African context. This is an incredibly important piece of academic work, and I was saddened and disappointed about only recently learning of its existence, nearly 18 years later. None the less, this body of work remains completely relevant, and it is a highly important read for all South Africans who care about their constitutional Right to Life. It is presented here in full, with the document attached at the bottom in Word format.

By Dr. Richard Wesson

Executive Summary

Analysis of the latest data and opinion relating to South African murder rates was carried out with regard to Professor Lott's Model; “More Guns, Less Crime”.

Sociological aspects were examined to determine whether pressure was being inflicted on society to increase crime or to reduce it. It was determined that overwhelmingly, pressure, especially economic pressure should be driving people to crime.

Antony Altbeker's research was examined to determine whether carrying a legal firearm increased the likelihood of robbery, or other violence to the person. No evidence could be found since the research was not representative of persons defending themselves with licensed firearms, because most successful defences were not reported to the SAPS. Unsuccessful defences could not be compared against the unknown number of successful defences.

The most recent murder figures available from CIAC/SAPS were examined. It was found that total murder rates, firearm murder rates and handgun murder rates have all fallen since 1994. Indeed handgun murder rates have fallen by 22% in that period.

The figures and opinions were compared against Lott's Model. Statistics relating to domestic violence and rape are unreliable. Furthermore, South African Culture inhibits firearm ownership by women. These two factors mean that aspects of Lott's Model relating to South African Women cannot be examined, at present. However, this does not affect other parts of his model.

The other factors and statistics demonstrate that the only common factor which affects the drop in all types of murder rates, a reduction in the increase in armed robbery, a stabilisation of hijackings and other confrontational economic crimes is the increase in licensed firearm ownership.

The increasing ownership of licensed firearms in conjunction with the dropping of murder rates, especially handgun murder rates, further disproves the hypothesis that licensed firearms, and especially licensed handguns contribute to crime and violence.

A final discussion presented where it was shown that no evidence could be found to show that licensees are criminally minded because they own firearms, and that there was a real need for the private ownership for protection purposes.





Murder rates are perhaps the best indicator of the violent crime which affects a country. Murders have arguably the highest rate of being reported of all crimes. The reporting rate is probably in the order of 93-98%. It is doubtful that the rate of reporting any other crime in South Africa, except perhaps car theft (including hi-jacking), comes close to that.

Despite the anomaly of firearm murder rates rising and then falling for the six-month period prior to the general election of 1999, murder rates and firearm murder rates have fallen since 1994. Even more interestingly, actual handgun murders have dropped from 9 144 to 7 183 (CIAC/SAPS), a drop from 81.75% to 59.80% of firearm murders.

Reported murders rose from 8 662 in 1974 to 26 832 in 1994 and have been steadily dropping since then to 24 210 in 1999 (CIAC/SAPS). These figures do not take into account the rise in population, which will cause a dropping of rates, in themselves, and the possibility of poor or corrupted data gathering prior to 1994. However, since 1994, murder rates and firearm murder rates, over the long term, have been steadily falling – why?

It was therefore decided to analyse the most recent data and compare it with relevant work by Antony Altbeker, other sociological factors and to put these new data into perspective.

It was realised that if allowances were made for South African culture, it would be possible to compare the currently available information with Professor John Lott's Model – “More Guns, Less Crime”.

Each aspect is discussed, and conclusions drawn at the end.


There are considerable reasons, mainly socio-economical, why the murder rates and other crimes should be rising. These are listed below.

1) Lack of Respect for the Law – there is a general lack of respect of, and faith in the whole judicial process. The purpose of the Law is to allow society to arrest, try and punish (or incarcerate) a wrongdoer. Research worldwide, but especially in the United States shows that the Law only acts as a deterrent if the criminal believes that s/he may get caught, and more importantly, if caught, the rate of convictions is high.

The arrest rate and the rate of convictions in this country are too low for the judicial system to act as a deterrent. A criminal does not participate in a crime where s/he thinks that s/he will get caught! This lack of respect for the Law removes inhibitions, which would inhibit wrongdoers.

2) Reductions in the Number of Police Personnel – The number of policemen and policewomen has been steadily dropping since 1995. Bureaucratic requirements have been increasing. This means fewer personnel on the street, and less affective investigations. This is less effective policing, and therefore less deterrence to crime.

3) Detectives work alone – this allows greater opportunity for corruption. Furthermore working in teams increases efficiency and allows detectives to discuss cases and thereby increase success rates. Clever cops are involved in more cases.

4) Increasing urbanisation – urbanisation results in people living in more crowded conditions and settlements having larger populations. Both these factors increase stress. This results in increased violence. Increased populations result in increased availability of goods to rob and more places to hide, both of which causes an increase in crime.

Most urbanisation, in the underdeveloped world, is in the form of informal settlements (squatter camps). There is a dramatic differential between the rich and the poor.

The poor seek to put food on the table, but the image of the very rich may dissuade some of these people from attempting to find honest employment (which may not always be available) and therefore they resort to crime, usually violent (it requires less skill).

Initially the intent may be to rob only the rich but this consideration soon drops away until anybody becomes a suitable candidate for robbery, especially their own people. (The various reports by Antony Altbeker graphically demonstrate this, if one cares to read between the lines.)

5) Reduction in Employment – Actual numbers of employed positions have been dropping since 1995. With an increasing population, this results in ever increasing unemployment. (One only has to listen to COSATU) This in turn, results in increasing rates of crime to put food on the table.

6) Loss of Hope – The reduction in employment in conjunction with increasing informal settlements has resulted, in some people, in a loss of hope and expectation. There was considerable expectation after the 1994 election, which has not been met. This disappointment has undoubtedly resulted, in some, in a disregard for the Rule of Law, and a determination to ‘improve their lot' by any means available. Crime is the most obvious option.

7) Immigration of other African Criminals – Since 1994, RSA has opened her borders to other African nationals. This has resulted in criminals immigrating to his country, albeit mostly illegally. However, this has resulted in an increase in the number of criminals, usually organised and violent, in this country. This, in turn, increases the probability of crime and violence.

All these factors are considered as engines to drive an increase in violent crime, including murder rates and firearm murder rates. These reasons, in themselves, would be sufficient to explain any rises in violent crimes without resorting to blaming legal ownership of firearms, yet murder rates and firearm murder rates are falling.


Without doubt Antony Altbeker has attempted to put the highest integrity into his research. He has ensured randomness in his research by analysing finalised SAPS dockets. With the filing system that is currently available to the SAPS, there is absolutely no way dockets can be chosen for result.

Unfortunately, there does appear to be confusion between randomness and representativeness. His research is based on analysis of dockets. At best, his findings represent incidents, which were reported to the police.

Initial research (Anger 2000) in South Africa, and considerably more formal research in the US (e.g. Rand 1994, Kleck 1993)) indicates that if a firearm owner uses a firearm, but suffers no major loss, and is not witnessed, s/he will not report the incident to the SAPS. Indeed non-firearm owners are unlikely to report non-serious (and sometimes serious) incidents to the police due to the belief that they will get no help from the police, nor will the criminal be caught.

Indeed, many people, under such circumstances, believe that the criminal is probably a friend of the cop concerned. Such people do not want any more bother; they just want to get on with their lives.

3A. Are South Africans responsible Gun Owners? – Evidence from 1000 dockets

The following are aspects of a report, which required four research assistants and a considerable amount of hard work. This research looked at the incidents of firearm theft and losses in Cape Town, Durban, Johannesburg and Nelspruit. However, these are mostly criticisms relating to the unrepresentativeness of the data. These criticisms also contain different interpretations, perhaps of a more worldly nature?

Antony Altbeker fully admits that he has absolutely no idea how many people have resisted a confrontational crime with a firearm and then NOT reported it. The data does not support the conclusion that licensed firearm owners cannot protect their firearms since it is not known how many licensees successfully resisted a robbery, which could have involved the loss of their firearms, but did not report it. 13% of robberies of firearms were committed on policemen and soldiers (mostly cops, 98 out of 107) while 19% were carried out on security guards (a total of 32%, almost 1/3rd). It is submitted that this tells us that if a gangster wants a gun, all he has to do is to wait outside a shabeen until a uniformed cop or security guard staggers out, drunk, and then hit him over the back of the head with a brick and take his service pistol. There is no need to rob sober persons in the street not knowing if they were armed or not and how they will react. Not all violent criminals are idiots; otherwise they'd all be in prison. The criminal tells the victim to hand over his gun. This leads to the untenable assumption that the criminal is only after a ‘gun'. It is submitted that:

a) the first thing that a criminal needs to do, is to ensure that a victim cannot defend him/herself, then he goes about searching the victim for valuables. Hence the first words spoken,” Give me your gun (weapon?).” It used to be “hands up” but this would attract too much attention, besides this is not the wild west.

b) the dockets state that often only a firearm was stolen. It is likely that the firearm was the only thing of real value that a victim possessed when only a firearm was taken. It would be the only item worth listing which a poor person has been robbed.

Of the 298 694 firearms recovered in 1996-1998, no attempt had been made to determine how many of these firearms were used in a crime! Unfortunately, no attempt was made to use the results to disprove that the internationally researched conclusion that most criminals acquire a handgun to protect themselves from other criminals.

This research does not offer any evidence to show that this hypothesis is wrong. It would have been useful if data were made available to determine, or at least discuss, the applicability of this hypothesis. [Also many people cannot afford a legal firearm or are confused by the very long, difficult, expensive and complicated process (they have no faith in the cops and yet wish to protect their loved ones and themselves).]

The number of firearm robberies in townships was surprisingly high. This suggests that crime in townships is a lot worse than previously thought. It suggests that firearm theft is reported because of the nature of the item stolen, but that other crimes are seriously under reported due to the lack of faith in the SAPS. Perhaps, much more than previously thought. This supports the hypothesis discussed in item 1, above. [Could this be the reason why kangaroo courts are springing up everywhere? At least the victim gets attention and believes that s/he gets some justice and protection. (This is not to be taken as any support for vigilantism.)] A lot of thefts from safes are reported. Two aspects come to mind:

a) often only the firearm, or one firearm is stolen from the safe. No ammo, no valuables, no cash. This suggests that the firearm was lost or mislaid, and therefore the owner has made up the story to avoid prosecution, having otherwise lost the firearm.

b) media reports suggest that most thefts from safes occur under duress. The owner has to open the safe with a knife at his/her throat. As far as I am aware, it is extremely rare to hear about safes being broken onto. If they can't get the key, they may kill the victim(s) in disgust. This type of case shows flaws in the GFSA hypothesis that safes are of no effect. Also it shows that the current regulations relating to safes are sufficient.

3B. The Alexandra Report

This is a 1997 report relating to usage of firearms for self-protection purposes. Again, analysis of dockets is used as the means to research this topic.

Again, it is admitted that no attempt has been made to determine how many incidents of self-protection using a firearm occurred, which were not reported. This means that figures relating the probability of death or injury because of ownership or use of a firearm for defence are de facto academic inaccuracies.

These figures are not representative of the actual situation. These numbers only refer to reported crimes where a docket was opened. Given the nature of South Africa and South Africans, only unsuccessful defences are likely to result in the opening of a docket.

What does come through is that those persons who carry a firearm should be aware of their surroundings if they wish to benefit from owning a firearm. Possession of a firearm in itself is insufficient defence. Training by instructors from reputable civilian training organisations would solve that problem. The Alexandra report is extremely interesting with respect to the differences in sociology of two distinct cultural groups. It further confirms that violence is often associated with an excessive intake of alcohol.

Three aspects come through Antony Altbeker's research:

Despite real efforts to be unbiased, unrepresentative data can easily give rise to academically untrue results. In this case, the probability of injury of persons whose robbery became docketed has been transferred to all persons who attempts were made to rob and who were armed. This is the same as saying; “All persons of a particular racial group are mad because all the inmates of a particular mental institution are of that racial group.” This is obviously wrong, and so are the broad statements from Altbeker's work. Far too many firearms are being lost or stolen. It should be pointed out that firearms are also lost from governmental and para-governmental institutions. However, it is not only those who are fanatically opposed to non governmental persons legally owning firearms, but everybody else who believes that there should be less leakage. There is a considerable problem in encouraging police personnel to properly investigate and administer any firearm legislation. The boring bits are often ignored. This leads to an arbitrary enforcement of the law, and to possible (indeed probable) misunderstanding of the legislation and therefore enforcement of non-existing legislation. Even supporters of draconian legislation doubt whether it can be properly enforced, at present.


South African murder rates and firearm murder rates were calculated for the period 1994 -1999 using data from the CIAC/SAPS (Total No. of murders and No. of firearm murders – issued 03/10/99) and Statistics SA (Estimated mid-year population for South Africa).

Murder rates (per 100 000) versus licences (per 100 000), below, suggest an inverse relationship. This can be tested by plotting the two data sets against one another.

Our mandate is clear:

Ensure that the CFR sticks to it's mandate
Transparency in firearms legislation
Equal treatment before the law
Reasonable and Rational Licencing requirements

GOSA - We fight for YOU!


    • Contact Us
    • Girls On Fire
    • Events
    • Shop
    • GOSA court action


      • Silver Member
      • Silver (Monthly)
      • Ordinary Member
      • Donate Now! ​​​​​​​
      • GOSA success