Gay-hate related homicides: an overview of major findings in NSW


Each year some men are murdered in Australia because they are homosexual. Since 1990, the New South Wales Police Service, through its Gay and Lesbian Client Consultant has systematically recorded gay-hate related killings in New South Wales. The Australian Institute of Criminology has collected data on all homicides coming to police attention in Australia as part of the National Homicide Monitoring Program. Using these datasets, this paper compares the victims and offenders in gay-hate related homicides with other male homicides in New South Wales and finds that the victims are generally older than other male homicide victims and are more likely to have been beaten to death. The offenders are much more likely to be younger than other homicide offenders, and more likely to be unemployed (82%) and unmarried (77%). This paper summarises the differences and similarities between gay-hate related homicides against men and other male homicides in an attempt to achieve a fuller understanding of the most extreme form of hate crime—homicide. For this reason, this study is an important contribution to hate crime research.

The literature on hate crime in the last 15 years covers a wealth of surveys, research and theories on crimes of bias/prejudice which target another person because of their sexuality, race, ethnicity or religion. There are inherent difficulties associated with research into the most extreme form of hate crime—homicide, and more specifically gay-hate related homicides. Some of these difficulties include the accurate identification, classification and recording of these incidents once the primary witness has been killed. A gay prejudice related homicide or gay-hate related homicide is one where the victim may have been gay or perceived to be gay and the offender’s actions motivated to some significant degree by prejudice or homophobia.

The New South Wales Police Gay/Lesbian Client Consultant has compiled and collated details of these homicides for 10 years, monitored trends, and encouraged researchers to conduct valuable studies such as those by Tomsen (1994) and Tomsen and George (1997). For a case to be included in the dataset, there must be a primary, but not necessarily sole, causal link between the offender’s apparent prejudice towards gay men/lesbians and their lethal act of violence.

Similar to the monitoring role of the New South Wales Police Gay/Lesbian Client Consultant, the Australian Institute of Criminology (AIC) through its National Homicide Monitoring Program (NHMP) has collected details of all homicide incidents, victims and offenders that have occurred in each Australian state and territory since  1 July 1989. The existence of these two datasets has facilitated comparative research such as the present study.

Aims of the Present Study

This study sought to compare male gay-hate related homicides with other New South Wales male homicides. However, it should be noted that since the research was completed, police identified a possible lesbian-hate-related homicide in rural New South Wales in August 1999. This study examines differences and similarities between male gay-hate related and other male homicides, focusing on characteristics associated with the incident, victim and offender, and the victim-offender relationship. Another aim was to achieve a fuller understanding of the most extreme form of hate crime—homicide, and to discuss preventative strategies.


Data Sources

One of the data sources used in this study was the National Homicide Monitoring Program (NHMP) held at the AIC. The present study analysed NHMP data for the state of New South Wales only and this data was supplemented by data provided by the New South Wales Police Gay/Lesbian Client Consultant. All statistical analyses were then conducted using NHMP data, comparing male gay-hate related homicides with male non-gay hate related homicides (other homicides). Between 1 July 1989 to 30 June 1999—the period of this study—a total of 1072 homicide incidents occurred in New South Wales, perpetrated by 1163 identified offenders and resulting in the death of 1165 victims (65% male; 35% female).

The gay-hate related homicides compiled by the Police Gay/Lesbian Client Consultant were classified as such after consultation with detectives, general duties police, witnesses and researchers who have read coroners’ files and/or court transcripts.

During this 10-year period, there were 37 male victims of gay-hate related homicide (excludes seven cases where sufficient details were not available in the NHMP dataset in earlier years), and 454 male victims of other homicide (after exclusions discussed in Study Parameters below) identified in New South Wales.

Some of the possible indicators of a gay-hate related homicide are:

  • formal or informal admissions by perpetrators;
  • anti gay/lesbian abuse;
  • proximity to a known gay social club/venue;
  • proximity to a beat (A beat is a meeting place for men who have sex with men, usually a park, bushland or beachside area. A sexual encounter may occur in the location or be arranged);
  • information on motive from persons known to offender/victim;
  • nature of injuries (for example, sexual overtones to injuries, mutilation);
  • frenzied nature of attack;
  • time-related to major gay community event;
  • absence of other motive; and
  • alleged sexual proposition/suggestion/advance by victim to perpetrator.

Study Parameters

The main reason for limiting this study to New South Wales is that supplementary information identifying and classifying gay-hate related homicides was only available for that jurisdiction. The New South Wales Police Service is the only police service that systematically collects information on gay-hate related homicides. The opportunity to supplement the NHMP database with information provided by the Gay/Lesbian Client Consultant has allowed a fairly accurate account of gay-hate related homicides and the characteristics thereof. Other Australian states have recently established similar full/part-time specialist positions that should facilitate the identification of gay-hate related homicides in their jurisdictions.

There were a number of other exclusions. As previously mentioned, female homicide victims were excluded from the comparison group (other homicides), there were no lesbian-hate related murders recorded in New South Wales during the period under review (although Queensland, South Australia and Tasmania recorded such incidents [Howe 1997]). Victims under eighteen were excluded from both groups. Intimates were also excluded, because by definition it is not a gay-hate related homicide when committed by an intimate partner but rather an intimate partner homicide. A gay-hate related homicide differs from a gay person being killed by another gay person­—such an incident is not motivated by a hatred for the individual because they are part of a particular group, such as gays or lesbians (Golden, Jackson and Crum 1999). Not all homicide incidents that involve gay victims are considered gay-hate related homicides.


During this period, on average, approximately four men were killed each year in New South Wales in attacks related to prejudice or homophobia which may also be linked to notions of gender and masculinity. The incidence of male gay-hate related homicides in New South Wales has remained relatively stable over the 10-year period.

In approximately 78 per cent (n=29) of gay-hate related homicides an offender was charged. Comparisons will therefore focus predominantly on the 29 cases of gay-hate related homicide where offender information is available. However, given the size of the dataset for male gay-hate related homicides, it should be noted that percentage shifts and differences may only involve small numbers.

Incident Comparison

A comparison between the characteristics of a gay-hate related homicide and other homicides involving male victims is shown in Table 1. In summary, the main differences are:

  • A male gay-hate related homicide incident is more likely to occur on a Friday between midnight and 6 am, whereas other male homicide incidents are most likely to occur on Sunday between 6 pm and midnight.
  • The victim of a gay-hate related homicide incident is more likely to be killed at a residential premise (62% vs 51.4%), with the majority of these incidents occurring in the victim’s home (only one incident occurred in some other person’s home).
  • The majority of gay-hate homicide incidents involved multiple offenders in comparison to other homicide incidents (54.5% vs 44.2%).
Table 1 NEW SOUTH WALES, 1 July 1989 – 30 June 1999: Comparison of Incident Characteristics Between Male Gay-Hate Related Homicides and Other Male Homicides Involving Victims Aged Over 18
Characteristics Male Victims of Gay-Hate Related Homicide (n = 29) Other Male Victims of Homicide (n=454)
Day of the Incident
Sunday 14.3% 20.7%
Monday 3.6% 11.7%
Tuesday 17.9% 11.9%
Wednesday 10.7% 9.4%
Thursday 14.3% 13.9%
Friday 21.4% 13.9%
Saturday 17.9% 18.6%
Time of the Incident
Midnight to before 6am 34.5% 28.5%
6am to before Noon 10.3% 7.3%
Noon to before 6pm 20.7% 19.5%
6pm to before Midnight 27.6% 35.8%
Unknown/Not Stated 6.9% 8.8%
Location of the Incident
Residential Premise 62.0% 51.4%
Public Transport & Related 0.0% 4.2%
Street/Open Area 31.0% 19.5%
Places of Entertainment 0.0% 10.7%
Other Location* 3.4% 14.2%
Number of Victims & Offenders Involved
Single Victim – Single Offender 45.5% 46.0%
Single Victim – Multiple Offenders 54.5% 44.2%
Multiple Victims – Single Offender 0% 6.5%
Multiple Victims – Multiple Offenders 0% 3.3%

*Includes corrective/health institution, shops, shopping malls, banks/credit union, car parks/public garages/service stations, workplace/schools, and other commercial premises.

Victim Comparison

There are also a number of differences between gay-hate related homicide victims and other male homicide victims (see Table 2). For example:

  • The victims of gay-hate related homicides are more likely to fall in the older age groups (35 years and above) (median age of 43 years vs median age of 36 years).
  • The majority of gay-hate related victims were of Caucasian appearance (93.1% vs 76.1%).
  • The most likely cause of death for gay-hate related homicide victims was beating. A beating can be carried out in one of two ways: using hands/feet or using a blunt instrument. Other male homicide victims most likely died of a stab wound.
  • The most common weapon used to kill both categories of the victims was a knife or other sharp instrument. Only 3.4 per cent of gay-hate homicide victims were killed with a firearm, in comparison to 27.8 per cent for other male homicide victims. Many perpetrators of prejudice related violence employ a “hands-on” attack of a very physical nature.
  • Gay-hate related incidents are significantly more likely to involve a high level of brutality. For example, it is not uncommon to find victims of gay-hate related homicide incidents that have been repeatedly stabbed to death, with up to 75 stabs wounds being recorded. A study by the New South Wales Gay/Lesbian Client Consultant (Thompson 1999) showed that 70 per cent of the 42 gay-hate homicides recorded in 1990-99 involved savage beatings, repeated stabbings, mutilation and/or dismemberment.This expectation of a higher level of brutality has been presented in hate crime literature—see Levin and McDermitt (1993), Mason (1993), Martin (1996) and has been observed by hospital staff particularly in regard to gay-hate homicide—see Herek and Berrill (1992).
    The complete NHMP dataset has recorded three cases of mutilation and dismemberment involving male victims, all of these cases involved gay-hate homicides.
  • Gay-hate related homicide victims were more likely than other homicide victims to have consumed alcohol at the time of the incident (48.3% vs 32.3%). Many attacks occurred after some degree of socialising.
Table 2 NEW SOUTH WALES, 1 July 1989 – 30 June 1999: Comparison of Victim Characteristics Between Male Gay-Hate Related Homicides and Other Male Homicides Involving Victims Aged Over 18
Characteristics Male Victims of Gay-Hate Related Homicide (n=29) Other Male Victims of Homicide (n=454)
Age Group
18 to 24 years 10.3% 19.3%
25 to 34 years 20.7% 24.3%
35 to 49 years 37.9% 30.6%
50 to 64 years 20.7% 15.5%
65+ 10.3% 6.3%
Unknown/Not Stated 0.0% 4.0%
Racial Appearance
Caucasian 93.1% 76.1%
Indigenous 0.0% 3.4%
Asian 3.4% 10.1%
Other 3.4% 4.0%
Unknown/Not Stated 0.0% 6.5%
Cause of Death
Gun Shot Wound 3.4% 27.8%
Stab Wound 34.5% 37.4%
Beating 37.9% 20.8%
Drug Overdose 0.0% 0.2%
Drowning/Submersion 0.0% 1.1%
Strangulation/Suffocation 13.8% 3.5%
Smoke inhalation/Burns 3.4% 0.7%
Other Cause*/Unknown 6.8% 8.2%
Weapon Used to Kill Victim
Firearm 3.4% 27.9%
Knife & Other Sharp Instrument 41.3% 37.5%
Blunt Instrument 20.7% 7.5%
Hands/Feet 31.0% 22.6%
Poison/Drugs/Fire/Other 3.4% 4.7%
Alcohol Consumption
Yes 48.3% 32.3%
No 51.7% 67.7%

*Includes hanging, poisoning, injection, electrocution, employer negligence, shaking.

Source: Australian Institute of Criminology, National Homicide Monitoring Program & New South Wales Police Gay/Lesbian Client Consultant, NSW Police Service.

Offender Comparison

In terms of policy implications one of the most important differences between the offenders of male gay-hate related homicides and other male homicides is their youth (see Table 3). For example:

  • The offenders in male gay-hate related homicide were almost three times more likely to be aged between 15 to 17 compared to other offenders (29.5% vs 8.4%).
  • Approximately 39 per cent of gay-hate related offenders were aged between 18 to 24 years (median age of 20 years), compared with 34 per cent (18 to 24 years) of the offenders of other homicides (median age of 25 years) (see Tables 3 and 4). In total, 68.1 per cent of gay-hate homicide offenders were aged between 15 and 24 years, whereas 42.2 per cent of other offenders were in this age group.
  • The offenders of gay-hate related homicides are more likely to be of Caucasian appearance (93.2% vs 68.3%), unemployed (81.8% vs 47.9%) and unmarried (77.3% vs 63.7%).
Table 3 NEW SOUTH WALES, 1 July 1989 – 30 June 1999: Comparison of Offender Characteristics Between Male Gay-Hate Related Homicides and Other Male Homicides Involving Victims Aged Over 18
Characteristics Offenders of Male Gay-Hate Related Homicide (n=44) Offenders of Other Male Victims of Homicide (n=678)
Age Group
10 to 14 years 0.0% 1.8%
15 to 17 years 29.5% 8.4%
18 to 24 years 38.6% 33.8%
25 to 34 years 25.0% 29.4%
35 to 49 years 4.5% 18.4%
50 to 64 years 2.3% 2.5%
65+ 0.0% 0.9%
Unknown/Not Stated 0.0% 4.8%
Racial Appearance
Caucasian 93.2% 68.3%
Indigenous 4.5% 7.2%
Asian 2.3% 10.7%
Other 0.0% 3.8%
Unknown/Not Stated 0.0% 9.9%
Marital Status
Single 77.3% 63.7%
Married/Defacto 18.2% 13.1%
Separated Married/Defacto 0.0% 4.0%
Divorced 0.0% 1.0%
Widowed 0.0% 0.3%
Unknown/Not Stated 4.6% 17.8%

Source: Australian Institute of Criminology, National Homicide Monitoring Program & New South Wales Police Gay/Lesbian Liaison Unit, NSW Police Service.

Victim-Offender Relationship Comparison

The victim-offender relationship allows us to determine the dynamics of the interaction that led to the homicide. Contrary to popular belief, “stranger” homicides also involve an element of social interaction—from a brief nod in the street or some short association, such as meeting someone in a bar (Mouzos 1999).

A greater proportion of both gay-hate victims and other homicide victims were killed by a friend or acquaintance (see Table 4). However, a much higher proportion of gay-hate related victims were killed by a stranger in comparison to other homicide victims (45.5% vs 28.2%).

Table 4 NEW SOUTH WALES, 1 July 1989 – 30 June 1999: Comparison of Victim - Offender Relationship Characteristics Between Male Gay-Hate Related Homicides and Other Male Homicides Involving Victims Aged Over 18
Characteristics Male Gay-Hate Related Homicides (n=44)* Other Male Homicides (n=678)
Victim-Offender Relationship
Family 0.0% 8.4%
Friends / Acquaintances 52.3% 36.7%
Strangers 45.5% 28.2%
Other 2.3% 11.9%
Unknown / Not Stated 0.0% 14.7%
Median Age of Victim 43 years 36 years
Median Age of Offender 20 years 25 years
Victim Younger than Offender 9.1% 30.4%
Victim Same Age as Offender 0.0% 4.7%
Victim Older than Offender 90.9% 64.9%
Employment Status
Victim & Offender Working 4.6% 10.9%
Victim Working & Offender Not Working 43.2% 22.9%
Victim Not Working & Offender Working 2.3% 8.3%
Victim & Offender Not Working 50.0% 58.0%
Alcohol Consumption
Both Victim & Offender Drinking 25.0% 27.9%
Victim Drinking But Not Offender 11.4% 5.8%
Offender Drinking But Not Victim 13.6% 9.6%
Neither Victim Nor Offender Drinking 50.0% 56.8%

*Offender based analyses.

Source: Australian Institute of Criminology, National Homicide Monitoring Program & New South Wales Police Gay/Lesbian Liaison Unit, NSW Police Service.

Another difference between gay-hate related homicides and other homicides is that in just under half of these cases, the victims of gay-hate related homicide were employed, and their offenders not in the workforce at the time of the offence (43.2%). This is in comparison to less than a quarter of victims who were employed, and their offenders not in the workforce in other homicides (22.9%) (see Table 4). Although information as to the type of employment was not available for comparative purposes, previous research (Thompson unpub.) suggests that victims of gay-hate related homicide were more likely to be employed in professional/managerial positions. This may indicate that male gay-hate related homicides differ from homicide in general which most often occurs between persons who belong to what can be described as an under-class in Australian society (James and Carcach 1997). Thus, offenders of gay-hate related homicide who were not in the workforce were more likely to target middle-class gay men, rather than men with the same socioeconomic status as themselves.

Discussion and Policy Implications

This study has found that there are differences between gay-hate related and other New South Wales homicides relating to incident, victim and offender characteristics, and the victim-offender relationship. In brief, some of the main findings relating to gay-hate related homicides are that:

  • Incidents are more likely to involve multiple offenders, and less unlikely to involve multiple victims.
  • The victim is more likely to be killed in the privacy of their own home.
  • The victim is more likely to be older than the offender.
  • The victim is more likely to be brutally beaten to death (with hands or feet or some blunt instrument), or repeatedly stabbed to death with a knife or some other sharp instrument.
  • The victim is more likely to be killed by a stranger.
  • The gay-hate homicide offender is more likely to be aged between 15-17 years; and on average 5 years younger than the offender of a non-gay-hate homicide.
  • Victims and offenders are more likely to be Caucasian.
  • Victims are more likely to be in the workforce than the offender.

As a result of these underlying differences, intervention and prevention strategies and policies can be tailored to address the findings of this study.

Foremost, however, it should be recognised that intervention with individuals committing hate crimes is not easily accomplished (Golden et al. 1990). The main obstacle is that “hate crimes are as much a product of the society as they are of the individual” (Golden et al. 1999, p. 267). Homophobia and violent conceptions of masculinity are said to be reflected in cultural phenomena that are widespread, if not intractable (Tomsen and George 1997). Such deeply ingrained attitudes promote an atmosphere that condones violence against gay men and lesbians. Intervention therefore needs to focus on societal attitudes at large, not just at the perpetrator of the crime.

Mason (1993, p. 2) outlines immediate prevention measures which should be addressed across Australia, but her overriding premise is “that violence prevention measures must start by removing the anti-homosexual bias in Australian society . . . Violence will fail to serve a function for the perpetrators if the prejudicial attitudes under riding such violence are no longer supported by societal norms or by religious, legal and political doctrines”. Herek and Berril (1992) held the same view when pointing out that laws which stigmatise homosexual behaviour can contribute to anti-gay violence. In this connection, the New South Wales Anti-Discrimination Board has compiled a list of many New South Wales laws which deal differently with gay men and lesbians.

Since 1990, the New South Wales Police Service has implemented, trialed and nurtured a range of strategies to reduce homophobic hate violence (Sandrousi and Thompson 1995; Thompson 1998). These include the specialised Police Gay/Lesbian Liaison Officers (GLLOs), communication and education campaigns, working with schools to reduce youth involvement in anti-gay violence, targeted recruitment advertisements, an award winning web-site, improving liaison with the gay and lesbian community, and marching in the annual Sydney Gay and Lesbian Mardi Gras Parade under the banner “We’re Here Because We Care”.

Prevention can begin by parliaments and the judicial system placing a stronger emphasis on the incarceration of those who commit gay-hate related homicide, and so reaffirming the seriousness of these offences, and the importance of equality. Nonetheless, jury verdicts in the mid-1990s have been characterised as reflecting “tenacious community attitudes against homosexuals . . . that a homosexual advance is the greatest affront to red-blooded masculinity. And that a man who reacts violently to defend his honour is behaving as any ordinary man would have” (Horin 1996, p. 2A). It should be noted, however, that these verdicts were prior to New South Wales removing the unsworn statement from the dock and prior to government attention focussing on these trials via the New South Wales Attorney General’s Homosexual Advance Defence (HAD) Monitoring Group.

One of the significant findings in the present study in terms of possible policy directions was that the majority of perpetrators of gay-hate homicide were young men under the age of 25, and that more than half of the offenders committed their crimes in groups.

Given the relative young age of the offenders and their over-representation as offenders of gay-hate homicides, one possibility is to focus on early intervention through education and training. Education should focus specifically on imparting knowledge of the incidence of gay/lesbian hate violence and on lessening prejudices, on understanding and on creating an atmosphere conducive to tolerance, and respect for gay men and lesbians.

It has been emphasised that the school environment is a place of education as well as a place where socialisation with other children with different characteristics occurs. Therefore, schools are in a unique position whereby they are able to teach children the general values of accepting one another and of valuing and respecting each other despite differences. “Children and adults armed with strong values are likely to be better able to resist the misinformation and invitations to violence found in the community” (Golden et al. 1999, p. 270).

Concluding Comments

This comparative study reveals that there are distinct differences in the prejudice motivated killings of men who are gay or perceived to be gay, when compared to other types of homicide. As a society we inevitably react with revulsion when we realise the young age of many of the perpetrators of these homicides, and at the level of brutality involved when a person is killed because of hate or prejudice or indeed to elevate one’s own sense of social status at the expense of another group.

These crimes are considered “message crimes” or crimes that send a message of fear and terror based in bigotry. They impact on the target group and the wider community, and in turn affect social harmony and well-being. This study reiterates the need for the whole of society to remain vigilant in protecting all of its members.


  1. The complete report of this study was presented at the Hate Crime Conference, 9-10 December 1999, convened by the Institute of Criminology and the Department of Gender Studies, both of the University of Sydney.
  2. The complete report will also be published in a forthcoming issue of Current Issues in Criminal Justice.


  • Golden, C.J., Jackson, M.L. and Crum, T.A. 1999, “Hate Crimes: Etiology and Intervention”, in H.V. Hall and L.C. Whitaker (eds), Collective Violence— Effective Strategies for Assessing and Interviewing in Fatal Group and Institutional Aggression, CRC Press, Boca Raton.
  • Herek, G. and Berrill, K. 1992, Hate Crimes Confronting Violence Against Lesbians and Gay Men, Sage Publications, Newbury Park, California.
  • Horin, A. 1996, “Gay Deaths: Is Justice Being Murdered?”, Sydney Morning Herald, 27 May, p. 2A.
  • Howe, A. 1997, “More Folk Provoke Their Own Demise—Homophobic Violence and Sexed Excuses”, Sydney Law Review, vol. 19, pp. 336-65.
  • James, M. and Carcach, C. 1997, “Homicide in Australia 1989-1996, Research and Public Policy Series, no. 13, Australian Institute of Criminology, Canberra.
  • Levin, J. and McDevitt, J. 1993, Hate Crime: The Rising Tide of Bigotry and Bloodshed, Plenum, New York.
  • Martin, S. 1996, “Investigating Hate Crimes: Case Characteristics and Law Enforcement Response”, Justice Quarterly, vol. 13, no. 3, pp. 455-80.
  • Mason, G. 1993, “Violence Against Lesbians and Gay Men”, in Violence, no. 2, Australian Institute of Criminology, Canberra.
  • Mouzos, J. 1999, “Femicide: The Killing of Women in Australia”, Research and Public Policy Series, no. 18, Australian Institute of Criminology, Canberra.
  • Sandrousi, J. and Thompson, S. 1995, Out of the Blue, New South Wales Police Service, New South Wales.
  • Tomsen, S. 1994, “Hatred Murder and Male Honour”, Criminology Australia, November, pp. 2-6.
  • Tomsen S. and George, A. 1997, “The Criminal Justice Response to Gay Killings: Research Findings”, Current Issues in Criminal Justice, vol. 9, no. 1, pp. 56-70.
  • Thompson, S. 1992 (unpub.), “Violence and Harassment Against Lesbians”.
    • 1998, “Improving Police Gay Lesbian Relations and Targeting Hate Crimes Against Gays and Lesbians 1985-997”, New South Wales Police Service, Sydney.
    • 1999, “Fact Sheet—Youth Involvement in Gay Hate Homicides”, New South Wales Police Service, Sydney.
    • 1999, unpub. Monitoring of Gay Hate Related Homicides, New South Wales Police Gay/Lesbian Liaison Unit

About the authors

Jenny Mouzos is a Research Analyst with the Australian Institute of Criminology.

Sue Thompson is the New South Wales Police Gay/ Lesbian Client Consultant.