No blueprint of gun control promises to eliminate gun violence from communities. But the wide range of murder rates – between the extremes of Colombia (167.6 male homicides per 100,000), where guns are readily available and Japan (0.7 per 100,000),1 which has the strictest gun control in the world – raises obvious questions about whether there might be some connection between gun availability and total and gun homicide rates. Similar questions can be asked about gun suicide rates.

The main reforms introduced after the Port Arthur massacre were designed to reduce the number of guns in the Australian community by making it much harder to qualify for gun ownership; tightening requirements on registration and storage; and removing certain weapons deemed unacceptable for civilian use. These policies were expected to reduce gun ownership, gun availability and therefore gun violence (murders, suicides, accidental deaths and injuries).

This chapter summarises the main arguments for reducing the number of guns in communities and the number of people permitted to own guns. The concern here is with the central question of whether reducing access to guns can be expected to reduce the number of people killed and injured from guns through murder, suicide and unintentional injury. I will summarise several key pieces of research on the relationship between the prevalence of guns in communities and their death/injury rates, and comment on the gun lobby’s counter-arguments.

74

Reduction not elimination

The tightest gun control laws imaginable would not eliminate all gun violence from a community. No nation has a zero death rate from guns. As with all preventive public health policies, gun control policies seek to reduce rather than eliminate the problems they address. Road safety policies are acclaimed when they reduce deaths and injuries. If someone condemned random breath testing or highway patrols because they only halved rather than eliminated drunk driving or speeding, they would hardly be taken seriously. Yet this was what the gun lobby sought to argue in regard to the new gun laws.

Mike Ascher, from the NSW Amateur Pistol Association, bellowed to the crowd at the 15 June 1996 pro-gun rally in Sydney, ‘Get rid of the guns and you stop homicides. Get rid of the guns and you stop robberies. Get rid of the guns and you stop home invasions. What a load of unadulterated bloody rubbish!’2 Similarly John Tingle argued: ‘Handguns have been registered in NSW since 1927. It has not and will never stop their use in crime.’3 Here Tingle, like Ascher, adroitly sets up the straw man argument that the only worthwhile test of gun registration is whether it eliminates all criminal use of guns. Anything less is presumably considered worthless.

International comparisons

The task of comparing countries head-to-head on two variables – for example, number of guns per population and number of incidents of violence (usually death from gun homicides, suicides and accidents) – is fraught with methodological problems. There are many cultural, historical and economic factors that can influence the scale of a nation’s gun violence. The Colombian gun death rate, for example, plainly has much to do with the presence of cocaine cartels and their private armies – something few other nations share. Civil wars and persistent sectarian

75violence in places like Bosnia, Sri Lanka and Northern Ireland also confuse the gun availability/violence picture.

The extent of a nation’s ‘gun culture’ depends on many factors including whether guns have been historically revered and glamorised, patterns of urbanisation, the extent of all violent crime, and whether guns are kept in homes as part of strictly regulated civil defence programs (as in Switzerland). These potentially complex and confounding variables are difficult to define and quantify, and make direct international comparisons problematic. All but the most intractable in both gun control and pro-gun factions admit there is clearly room for debate whenever precise claims are made about the relationship between availability and gun violence.

However, these groups differ radically over the question of whether there is an acceptable degree of research support for the general proposition that more guns means more deaths. The gun lobby argues this is nonsense and champions the case of nations with high rates of gun ownership and low rates of gun homicide (for example, Norway). Gun control advocates are less concerned about exceptional cases and more with the pattern across countries. There have been several recent studies comparing rates of gun ownership and gun violence between countries that are broadly similar culturally and economically.

A comparison of homicide rates between the US and Britain found that the non-gun US homicide rate (per 100,000 population) was 3.7 times higher than the British rate, while the rate of handgun homicides was 175 times higher.4 This suggests that, even if Americans were inherently somewhat more murderous than the British, the easy availability of handguns has produced a massively disproportionate number of homicides.

A study comparing the cities of Seattle (US) and nearby Vancouver (Canada) found the cities had almost identical rates of burglary, assault and robbery without a gun; but that the Seattle rate of assault with a firearm and homicide with a handgun were seven and 4.8 times higher

76respectively than the corresponding rates in Vancouver.5 Again, handguns are much more accessible in the US than in Canada.

A later study by Centerwall6 compared homicide rates from 1976–89 in Canadian provinces and US states on the US/Canada border. He concluded that there were no consistent differences between the homicide rates of the two countries, despite handgun ownership being much higher in the US. But other researchers have pointed out that Centerwall’s comparison gives the same weight to sparsely populated states and provinces as it does to densely populated urban areas. If the comparison is made between states and provinces of similar population densities, the homicide rate is between two and four times higher on the US side of the border.7

The prolific suicide and homicide researcher David Lester examined the relationship between gun availability and both the total and gun homicide rates in 16 European countries. He used the proportion of all murders and suicides where guns were used, as well as the annual accidental gun death rate, as proxy measures for gun availability, commenting that ‘no country of the world provides accurate measures of firearm ownership’. He found a strong correlation between these measures of availability and gun homicides, demonstrating that nations with more firearms have higher gun homicide rates. He also found that the rate of homicide by other means did not present any compensatory corollary; that is, in countries where guns were less available, there was

77no significant increase in alternative methods of killing.8

Swiss criminologist Martin Killias published in 1993 a review of international correlations between gun ownership and rates of homicide and suicide in 14 countries,9 later updating this to 18 nations.10 The data on percentages of households with guns was obtained from the 1989 and 1992 International Crime Victimisation Surveys (ICVS) conducted by telephone interviews with householders. This is the largest international attempt to use a standard method to measure gun ownership. Table 2.1 shows Killias’ results, listed by ascending levels of gun ownership.

78Killias found that across the nations, the rate of household gun ownership correlated with the rates of homicide and suicide committed with guns, the proportion of all homicides and suicides committed with guns, and the overall rates of homicide and suicide. In other words, countries with high gun ownership rates tended to have higher rates of fatal gun violence than those with lower gun ownership. Nor did the study find a ‘substitution’ effect; that is, in countries with low rates of gun ownership, the rates of homicide and suicide by other means did not compensate for the lack of access to guns.

Killias noted:

79

The gun lobby has criticised Killias’ study because of the high non-response rates (47% averaged across all 18 nations).12 It is also generally sceptical about surveys of gun ownership, arguing that many people will under-report ownership, due to fear of government persecution. But these critics cannot explain why the non-respondents would produce any systematic bias. In other words, they provide no compelling reasons why the correlations obtained would be any different had more people responded – if gun owners had consistently high rates of non-response to the surveys across the 18 countries, and the data were adjusted to account for this, this would not have altered the correlational results. Nor have they responded to the fact that the ICVS reported that rates of gun ownership generally corresponded well with estimates obtained by other surveys undertaken in some of the countries.13

The gun lobby also claims that the argument works the other way round: high homicide rates in a community may prompt people to purchase more guns for self-defence; that is, high gun murder rates lead to higher rates of gun ownership. This argument fails to explain why Killias and others have also found a link between high gun ownership levels and high suicide rates. It would be ludicrous to suggest that high gun suicide rates prompt communities to arm themselves more.

Gun violence in Australia

Table 2.2 shows the data on gun deaths in Australia for the five years 1990–95.

In summary:

• Between 1990 and 1995, an average of 560 people died from guns each year.
• Of all gun deaths, 81% were suicides; 14% resulted from violence; 4% were unintentional (accidents); 1% were of unknown intent. 80 In addition in 1994:
• 95.2% of gun suicides were males; 63.2% of gun homicides were males.
• Firearms were used in one in four homicides and one in five suicides.14
• One in four male suicides involved a firearm.
• One in 18 female suicides involved a firearm.
• 55% of female firearm deaths resulted from an assault.

In New South Wales, between 1984 and 1988, guns were the third leading cause of injury and death (8% of total), behind motor vehicle accidents (35%) and falls (12%).16

81Measuring the effect of gun laws in Australia has been difficult in the past because the eight states and territories all had different laws and the absence of interstate border checks meant guns were easily transported between jurisdictions, rendering the laws largely inoperable. This meant that guns bought freely in a state with a liberal law (for example, Queensland) could be carried into a stricter state where they would be illegal and not easily detected. The absence of gun registration in NSW, Tasmania and Queensland also meant there were no national records of number sand types of guns in circulation. Before the new laws, Australia’s strictest gun law was in Western Australia, where many of the 1996 APMC agreement terms were law already.

The most striking feature of the distribution of firearm mortality in Australia is that the two states with the most permissive laws, Tasmania and Queensland, have gun death rates significantly above the average.17 (The Northern Territory also has a relatively high rate, but it has such a small population (170,000) that a single gun death can dramatically increase the rate per population.) Gun death rates per 100,000 population in 1990–1994 were: Tasmania 7.97, Northern Territory 7.52, Queensland 5.04, South Australia 3.55, NSW 2.83, Victoria 2.66, Western Australia 2.55, Australian Capital Territory 1.56.18 The Tasmanian gun death rate in 1994 was 7.2 per 100,000, based on 34 gun deaths for that year. Port Arthur doubled this rate in one afternoon.

One of the few Australian studies on the relationship between firearm availability and violence was conducted by Chris Cantor and Penelope Slater. They examined the effect of the Queensland Weapons Act 1990, which introduced firearm licensing and a 28-day waiting period for all applications.19 The researchers said their study provided preliminary evidence that tightening gun control may reduce suicides

82rates, especially among young men. (They found reductions in both gun suicides and overall male suicides in provincial areas, and in gun suicides in the cities.)

Gun violence in the United States

As described in Chapter 1, the American gun problem has long imprinted itself in the Australian consciousness as the embodiment of all that should be avoided. For US gun control campaigners, the introduction of the most basic restrictions are seen as major victories; for example, the radical requirement that handgun buyers endure a six-day cooling-off period to allow background checks. The State of Virginia has introduced what the gun lobby bemoans as a tough handgun law, allowing residents to purchase a maximum of only one handgun per month.

The gun lobby likes to make the point that there is no shortage of gun laws in the US; rather, there are 20,000 separate gun laws which have not curbed the epidemic of violence.20 The large number of gun laws illustrates the same fundamental deficiency which undermined gun control in Australia until 1996: the lack of national uniformity. A ban on military-style semi-automatics in her home state of Western Australia did not prevent 21-year-old Kate Scott from being killed when she visited Port Arthur, because Tasmania’s law still allowed these weapons to be freely sold. The safety of an entire nation is undermined when gun laws are more feeble in some states than in others. But at least Australian gun laws are state and territory laws, so our legal patchwork was made up of only eight parts. US gun laws are often municipal ordinances – local bans which in reality have only symbolic value: the equivalent of banning handgun sales in downtown Perth while a brisk trade continues in neighbouring Fremantle.

In 1996, 31 of the 50 states in the US had laws permitting citizens to carry concealed loaded handguns21 – a situation the National Rifle

83Association (NRA) claims actually reduces the gun homicide rate. In 1996 a great deal of publicity was given to a paper by two American researchers, Lott and Mustard, focusing on ‘shall issue’ laws which require officials to issue permits for citizens to carry concealed guns.22 The researchers examined whether these laws deterred violent crime. They concluded that states with such laws had lower rates of violent crime than those without. The gun lobby rapidly championed the study on the internet and in countless editorials. But reviews by other researchers found the study was methodologically weak enough to render its conclusions suspect.23 The ‘guns make us safe’ perspective could probably only be advanced with a straight face in a country where gun violence is as rampant as it is in the US. For outside observers, the first and most obvious question to ask is: ‘If carrying guns and arming homes prevents gun violence, why does the most heavily citizen-armed nation in the Western world have about 38,000 annual gun deaths, including 18,000 gun homicides a year, giving it the highest per capita rate of gun violence?’ The debate on gun control in the US is profoundly parochial, with the horizons of many internal commentators seldom reaching beyond US shores to consider the perspective of other nations’ gun violence records.24 To argue from the US situation that guns make communities ‘safer’ is rather akin to arguing that war-torn Bosnia is safer than Rwanda during the genocide. Ever since Port Arthur, the NCGC’s web site has attracted virulent email from US shooters gasping in near apoplexy over our statements, apparently seldom realising that other nations’ gun laws differ from those of the US and seemingly unaware that the benchmarks of ‘safer’ communities in the US are frequently outrageously high compared with other nations with tougher gun laws.

84

Here are some recent statistics on gun violence in the US (population about 254 million):

• From 1968 to 1991, annual deaths due to motor vehicles declined by 21% (from 54,842 to 43,536), while during the same period deaths due to firearms increased by 60% (from 23,875 to 38,317).25 If this trend continues, by the year 2003 the number of US citizens killed annually by firearms will exceed the number killed in motor vehicle crashes.26 In some states this crossover has already occurred.
• In 1993 there were 39,595 firearm-related deaths, a rate of 15.36/100,000. Of these, 18,940 (47.8%) were suicides, 18,571 (46.9%) were homicides, 1,521 (3.8%) were unintentional, and the remainder were of undetermined intent.27 Firearm-related deaths were the fourth leading cause of ‘years of potential life lost’ before age 65.2885
• A total of 26,009 homicides were reported nationally in 1993, of which 71% (18,466) were firearm related. Overall rates of homicide increased from 1985 to 1991 and decreased from 1992 to 1994. During these two periods, rates for total firearm-related homicides and homicide among people aged 15–24 years increased and then stabilised, but remained at record high levels.29
• For every person killed by a gun, another 2.5 are injured seriously enough to be admitted to hospital. An estimated 99,025 people were treated for non-fatal firearm-related injuries in US hospital emergency departments in 1992.30
• In 1992 the US had more gun dealers (about 245,000) than petrol stations (about 210,000).31 Since recent law reform outlawing ‘kitchen table’ dealerships, the number of dealerships has fallen to a mere 142,094.32

In summary, the US has nearly 14 times Australia’s population. It has 69 times our total gun deaths, and 172 times Australia’s gun homicides (Table 2.3). By contrast, Japan has nearly seven times Australia’s population, yet has nearly six times fewer gun deaths and nearly three times fewer gun homicides.

86

Gun violence in England, Wales and Scotland

The gun control and gun violence situation in mainland Great Britain (excluding Northern Ireland, because of the continuing sectarian violence there) is particularly instructive. In a population of 57 million, Great Britain has very low rates of gun violence compared to those of Australia and the United States. Table 2.4 shows that the gun homicide rate in England and Wales is eight times less than in Australia, and nearly 56 times lower than the US rate. The gun suicide rate is nine times lower than Australia’s and 19 times lower than it is in the US.

After the 1987 Hungerford massacre, Britain banned virtually all semi-automatic rifles in 1988 from private ownership. In response to the Cullen report on the March 1996 Dunblane massacre, the British Government passed The Firearms (Amendment) Act 1997 which banned all private possession of handguns over .22 calibre and all multishot .22s. Tony Blair’s new Labour Government, elected in June 1997, lost no time in outlawing all private ownership of handguns, attributing the decision to the Dunblane shootings. Intriguingly, Lord Cullen’s report did not apply the reasoning behind his recommendations about ex-residential handgun storage, to all guns. He wrote:

One cannot hunt or target-shoot with rifles and shotguns inside homes. But long arms are often stolen from homes, entering the criminal subculture. Being more common than handguns, shotguns and rifles are also more commonly used in domestic homicides and suicides. Having to take the time and trouble to check out a shotgun from a community armoury can put a critical cooling-off period between a raging impulse to shoot oneself or one’s family and actually doing it. It is regrettable that Lord Cullen apparently chose to apply his terms of reference only to ‘public’ safety from guns, when gun murder and suicide behind the closed doors of gun owners’ houses claim more lives each year.34

The NCGC frequently exemplified Britain’s strict gun control post Hungerford and low death rates as the direction Australia should take. The gun lobby had no answers to this, except to claim that as Britain’s gun laws had become tougher, its gun crime rate had risen. In fact, between 1993 and 1994 both gun homicide and armed robbery fell in Britain (Table 2.4). But this did not stop the gun lobby claiming that ‘firearm deaths are increasing at a higher rate than [in] the so-called “gun crazy” United States’.35

88

The gun lobby has argued that high gun homicide rates in some nations result not from high gun ownership, but from high levels of ‘violent culture’, which manifests itself in gun homicides.37 On this basis, nations which do not have ‘violent cultures’ can safely have access to guns without this translating into high rates of gun homicide. The gun lobby put this argument in submissions to the Dunblane inquiry. Unfortunately for the gun lobby, the argument that Britain is a ‘less violent’ and less crime-ridden nation than the US does not stand up to scrutiny. Table 2.5 lists rates of various crimes in the US and in England and Wales. It shows that property crime and non-gun assault rates are actually higher in England and Wales than in the US. This contradicts the suggestion that the British ‘culture’ is inherently less ‘criminal’ than the American, as the gun lobby seeks to imply. Gun-related crime rates, though, are dramatically different – suggesting that the availability of guns increases both fatal and non-fatal gun crime considerably more than might be expected from differences in recorded levels for other offences.

89

Gun violence in Japan

The extremely low gun death rate in Japan is an object lesson for other nations. Japan has the world’s toughest gun laws. Citizens are not permitted to own handguns or swords. In 1971 the Japanese Government stopped the private transfer of all rifles then held by citizens, and required family heirs to surrender guns when the owners died. As a result, by 1986 only 27,000 people were licensed to keep a rifle or shotgun39 and the number of licensed guns fell from 652,000 in 1981 to 493,373 in 1989.40 In 1992 Japan had 133 million people (seven times Australia’s population of 17.5 million), yet in 1995 just 34 people were murdered with guns – about one-third of Australia’s gun homicide total. More people were shot dead at Port Arthur in one afternoon than in Japan in all of 1995.

90Despite this, the editor of Australian Gun Sports told his readers: ‘Japan has a high murder rate and its people are prohibited from owning firearms.’41 The NCGC repeatedly raised the example of Japan in media interviews and radio compere John Laws emphasised it many times. Yet a SSAA pamphlet distributed in December 1996 stated: ‘Even though guns are virtually prohibited in Japan their murder rate is the same as Australia’s – a fact gun prohibitionists never mention.’ The pamphlet cited the World Health Organisation as its source.42 The reason why ‘gun prohibitionists’ such as the NCGC never mention this ‘fact’ about Japan is because it is utter nonsense. At 0.7/100,000, Japan has the world’s lowest homicide rate.

Gun violence in Switzerland

For many years the gun lobby has played what it believes is a rhetorical trump card in promoting the case of Switzerland, a nation whose civilian militia is said to hold some 400,000 military guns at home. By European standards, Switzerland has a low rate of homicide overall and a middle-range rate of homicide by guns (Table 2.1). The gun lobby believes this proves that heavily armed populations can live in safety. Immediately after Port Arthur it seemed that every second opponent of gun law reforms felt he had had the final word if he cited the case of Switzerland.43

Martin Killias, a Swiss resident, has pointed out that the image of Switzerland as a society with unrestricted access to guns is a ‘criminological myth’.44 The firearms kept in many Swiss homes are military guns that many citizens are obliged to keep as a part of their nation’s civil militia. This kind of gun ownership differs fundamentally from that in countries like Australia and the US. The social meaning of gun

91ownership has much more to do with Swiss civic responsibility than any personal interest in guns.45 Further, as one writer to the Sydney Morning Herald pointed out, the gun lobby seldom acknowledges that in Switzerland both arms and ammunition must be kept under strict locked storage conditions that are checked every year.46 The military weapons are heavy and difficult to carry inconspicuously, and military ammunition is not sold. There are extremely harsh penalties for failure to comply with these storage requirements and for any misuse of the military weapons. In terms of non-military firearm ownership, the Swiss gun law is ‘definitely not among the most liberal in Europe’.47

As for gun violence, Killias noted that Switzerland has a high rate of gun homicide and one of the highest rates of gun suicide in Europe (see Table 2.1). This evidence, he says, contradicts the contention that the high gun ownership rate in Switzerland is not accompanied by frequent illegal use.48 In fact the Swiss Government and people themselves believe guns are too easily available: in 1996 the population voted overwhelmingly to replace their patchwork of cantonal gun laws with one national law. At the time of writing, the new national law, which is stricter than the old cantonal laws, was expected to pass through Parliament.

Killias’ work confirms what would seem to be intuitively the case: that societies with high rates of gun ownership pay the price with high rates of gun death, and those with strict gun controls tend to have low death rates. Easy access to guns places lethal and efficient killing instruments in the hands of people who might otherwise use less efficient means.

92The gun lobby constantly disputes the relationship between gun availability and gun violence. A SSAA official agreed that the US as a whole had a high gun death rate, but insisted, ‘there are many American States with murder rates as low as anywhere in the world that are safer to live in than parts of Australia. North Dakota and Nebraska are two examples’.49 The reminder that there are some peaceful places in America was little comfort to Australians when our own quiet little backwater, Tasmania, was so savagely rocked by gun violence. (Although Tasmania’s ‘quiet’ reputation was hardly deserved. Along with the most liberal gun laws in the country, it has the highest rate of firearm deaths.)

One senior journalist reviewing the gun control debate concluded: ‘The crux of the guns lobby’s argument – that there is no relationship between the murder rate and the level of gun ownership – simply is not tenable. It makes about as much sense as saying that there is no connection between the road toll and the number of cars, and moreover that Formula One cars should be allowed on the open road.’50

Suicide prevention

Since 81% of all Australian gun deaths are suicides, suicide prevention is clearly a major goal of gun control policy. Public gun massacres and other homicides capture more media attention, but are statistically overshadowed by gun suicides as the main type of gun death. Opponents of gun control believe restricting access to firearms will not prevent suicides because people who are unhappy will kill themselves anyway, by another method. But expert opinion suggests that only a proportion of people intent on suicide will substitute other methods. Of those who do, many will substitute less lethal methods and therefore have a greater chance of survival. Suicide prevention is the primary reason why the campaign for tighter gun laws has been strongly supported by the Royal Australian and New Zealand College of Psychiatrists, the Australian

93Medical Association, Suicide Prevention Australia, the Public Health Association, the Australian College for Emergency Medicine, and the Centre for Adolescent Health – among others.

Many suicide attempts are not planned well in advance. Many are relatively impulsive acts, precipitated by bouts of depression or incidents such as sexual conflict or financial crises.51 A Scottish study of 522 self-poisonings, for example, found that two-thirds were impulsive.52 Many suicide attempts occur when the person’s judgment is impaired by intoxication.

Choice of suicide method by individuals and across different populations may be influenced by many factors: availability, cultural acceptability, technical skills (for example, being able to tie a fail-safe noose), whether planning is necessary, certainty of death, time taken to die, scope for second thoughts, chances of intervention, courage needed, perceived consequences of failure such as pain, disfigurement or humiliation, consequences for others (for example, danger, contaminating the family ‘nest’), scope for concealing or publicising the suicide, symbolism, cultural ritual (for example, Japanese hara-kiri, seppuku, shinju and oyako-shinju) and dramatic impact.53

Whatever the relevance of the above to the choice of using a gun, three factors stand out as highly important:

Many more suicide attempts arise from temporary despair, rather than from a genuine decision to die. A Sydney hospital conducted an eight-year retrospective study of 33 survivors of self-inflicted firearm injury, finding that most were young men who did not suffer from major depression or psychosis. Most shootings occurred in the context

94of personal disputes with sexual partners or family members. They shot themselves impulsively in a crisis, were not psychotic, and had ready access to firearms.54 Significantly, after medical and counselling intervention, almost none of the survivors attempted suicide again. This contradicts the popular misconception that ‘they’ll do it again anyway’; that is, that suicidal people will kill themselves no matter what efforts are made to prevent them from doing so. For every suicide in Australia, it has been estimated that there are 30–40 attempts (parasuicides).55

Guns are a particularly efficient means of killing. Different methods of attempting suicide have different ‘completion’ or fatality rates. In NSW, for example, attempts using guns are second only to hanging (see Table 2.6). The contrast in completion rates is particularly stark between firearms (75%) and the most common methods of attempting suicide – slashing oneself (6%) and drug overdoses (4%). The data suggest that if more people attempting suicide used guns, the number of suicide deaths would almost certainly rise. Alternatively, if more ‘attempters’ used knives, razors or drugs instead of guns, the number of suicides would fall. If access to guns was more difficult, some attempting suicide – particularly in impulsive contexts – might chose these less fatal methods.

Suicidal intent can dissipate in the time taken to locate or construct a method of attempting suicide. If a gun is not immediately available, by the time it is obtained or a substitute method found, the suicidal crisis may be over. US research suggests the availability of a highly lethal method like a gun in a house can precipitate an impulsive suicide attempt.56 In other words, the availability of firearms influences not only

95the outcome of an attempt, but also the likelihood of a person attempting suicide at all.

Hanging has a suicide fatality rate comparable to that of firearms. So if a gun was not available, might not those seriously attempting suicide choose a rope, which of course is readily accessible? ‘Why don’t you try to ban ropes!’ the gun lobby may sneer. Obviously some attempters may indeed choose hanging if a gun is not available. But hanging requires skill with knots, a suitable location and a degree of organisation. Some may reject this method as too difficult and too likely to fail.

96There is an extensive body of research on whether people intent on suicide switch to alternative methods if the first choice of suicide is unavailable.58 There is compelling evidence from the UK on the importance of limiting access to easily accessible, culturally accepted means of suicide. In England and Wales between 1963 and 1975, the number of suicides suddenly and unexpectedly declined from 5,714 to 3,693 when suicide rates were rising in most European nations.59 The fall was almost wholly attributed to the detoxification of domestic gas when carbon monoxide was removed from the gas supply. Many people attempting suicide did not switch to other means when they were no longer able to kill themselves in their homes using a gas oven or heater.

But what about gun control? Does restricting access to firearms prevent suicides? Canadian criminologist Thomas Gabor identified 16 local, national or international studies conducted since 1980 on firearm availability and suicide rates.60 He observed that 15 of the 16 studies found a strong link between gun availability and gun suicides. (The only study not confirming this relationship was conducted by Gary Kleck, the researcher whose work is most heavily relied on by the gun lobby.) Gabor found eight of these 15 studies also concluded that high gun availability led to higher suicide rates overall. Six found no relationship, and one study did not consider overall suicide trends. But Gabor noted that if the two factors were not connected at all, one would expect to see a balance of positive and negative correlations in the research results; for example, some studies suggesting that high gun ownership reduces suicide rates. The absence of any such studies led Gabor to conclude that the research is ‘strongly suggestive of a link between suicide and firearm availability’. Even more important for clinicians, counsellors

97and families to keep in mind is the evidence from case-control studies of the relationship between domestic gun availability and suicide. Three important US studies have found that the presence of a gun in the home increases the risk of suicide.61

Domestic violence

Gun control has long been a concern of those working to reduce domestic violence in Australia. The largest category of homicides (around 40%) is domestic, and guns are more commonly used in family killings than among homicides in general. A NSW Bureau of Crime Statistics and Research study noted: ‘The difference between a fatal and a non-fatal episode of domestic violence may be entirely due to the presence of a dangerous weapon such as a gun.’62 US Research shows that domestic assaults involving guns are three times more likely to result in death than those with knives; 23 times more likely than unarmed assaults; and 12 times more likely than non-firearm assaults overall.63

Victim surveys confirm abundant anecdotal evidence from domestic violence groups that guns are frequently used to intimidate women and children in their homes. For example, 14% of victims who contacted the 1988 Queensland domestic violence phone-in said they had

98been threatened or injured with a gun.64 In 1991 Sydney’s Domestic Violence Advocacy Service (DVAS) reported that 15% of clients said their partners had a gun.65 The DVAS and other agencies have described some of the ways guns are used to terrorise family members, including direct threats, shooting pets as a warning, mock executions, sleeping with the gun nearby, and cleaning the gun during or after arguments, especially during custody disputes.

Danielle Mazza’s recent study of 1,500 women attending their doctor found that of those women who were in couple relationships, one in 50 had been threatened with a knife or a gun in the previous year, and one in 100 had had a knife or gun used against them.66 The 1996 Women’s Safety Survey revealed that 0.5% of all Australian women – that is, about 34,500 – had been either threatened or harmed with a knife or gun in the previous 12 months.67 However vicious and traumatic a knife threat or attack may be, the availability of a gun greatly increases the likelihood of conflict or depression leading to death.