Australian Institute of Criminology

Skip to content

Indigenous and non-Indigenous homicide in Australia

Research in practice no. 37

Tracy Cussen & Willow Bryant
ISSN 1836-9111
Canberra: Australian Institute of Criminology, May 2015

Indigenous people (Aboriginal and/or Torres Strait Islander Australians) are disproportionately victims and offenders in homicide incidents both in relation to their relative proportion of the Australian population and in comparison with their non-Indigenous counterparts. In 2011–12, Indigenous people comprised three percent of the Australian population (ABS 2009; ABS 2012) yet constituted 13 percent of homicide victims (n=35) and 11 percent of homicide offenders (n=32; Bryant & Cussen 2015). The rate of both victimisation and offending by Indigenous people was approximately five times higher than that of non-Indigenous people (Bryant & Cussen 2015).

Available research suggests that victims and offenders may be exposed to, or experience, a number of vulnerabilities that increase the likelihood they will be involved in a violent offence and further, that these factors may be more pronounced for Indigenous people. Research undertaken by Wundersitz (2010), Bryant (2009) and Bryant and Willis (2008) has linked substance abuse, personal history (such as sexual abuse as a child), housing mobility, and social stressors (such as witnessing violence, gambling addiction, mental illness or serious accident) to an increase in offending and victimisation risk. A previous comparative analysis of Indigenous and non-Indigenous homicides in Australia (Mouzos 2001) also identified that the majority of Indigenous homicides occurred between family members in the context of domestic conflict.

This paper describes selected characteristics of Indigenous and non-Indigenous homicides as recorded within the AIC’s National Homicide Monitoring Program (NHMP) from 1 July 1989 to 30 June 2012. Over this time period, the NHMP has recorded:

  • 6,744 homicide incidents (1,096 involving at least one Indigenous person);
  • 7,217 victims (of whom 951 or 13% were Indigenous people); and
  • 7,599 identified offenders (of whom 1,234 or 16% were Indigenous).

Both the total number of victims and offenders is greater than the total number of homicide incidents over the 23 year period because some incidents involve multiple offenders and/or the death of multiple victims.

Homicides contained within the NHMP are reported to the AIC by police services and data are augmented with information from the National Coronial Information System, media reports and/or publicly available sentencing remarks from relevant court proceedings. Victim and offender Indigenous status is principally identified by the police and is likely derived from subjective assessments based on appearance and/or offender self-reported status. It is therefore likely that the number of victims and offenders identified as Indigenous within the NHMP is under-estimated and this limitation should be considered with reference to the data presented in this report. It should also be noted that there were 1,126 homicides (17%) where the Indigenous status of victims and/or offenders was not recorded.

Incident level analysis

Table 1 depicts the distribution of homicide incidents by the Indigenous status of offenders and victims. It shows that an Indigenous person was involved (as an offender or victim) in 16 percent (n=1,096) of homicides. Seventy percent (n=765) of these homicides involved both an Indigenous offender and an Indigenous victim. Of the 4,853 (72%) homicides involving a non-Indigenous person (as an offender or victim), 93 percent involved a non-Indigenous offender and a non-Indigenous victim.

Table 1: Distribution of homicide incidents according to the Indigenous status of offenders and victims, 1989–90 to 2011–12
n %
Indigenous offender on Indigenous victim 765 11
Indigenous offender on non-Indigenous victim 230 3
Non-Indigenous offender on Indigenous victim 101 2
Non-Indigenous offender on non-Indigenous victim 4,522 67
Unknown/not stated/missing 1,126 17
Total 6,744

Source: AIC NHMP 1989–90 to 2011–12 [computer file]

Over the 23 years of recorded data within the NHMP, 73 percent (n=4,946) of all incidents involved a single victim and offender. Homicides involving Indigenous people rarely involved multiple victims or offenders. Over 90 percent of Indigenous homicides (where both the victim and offender were Indigenous) (93%; n=708), and 86 percent (n=947) of all homicide incidents that involved at least one Indigenous person, were single victim/offender homicides (see Table 2).

Although rare, it is important to note that incidents involving more than one victim or offender are more difficult to classify by the relationship between the parties involved. Within the NHMP, the relationship between any victim/offender pairing is ranked and classified according to the ‘closeness’ of the relationship (see Bryant & Cussen 2015 for further detail). For example, if a person is murdered by their intimate partner and close friend, the homicide would be categorised as an intimate partner homicide. All homicides that involve intimate partners or other family members are broadly categorised as domestic homicides.

Comparatively, Indigenous homicides were more likely to involve intimate partners and other family members. Of the 765 incidents identified as Indigenous, 67 percent (n=511; see Table 2) were classified as domestic homicides. This compares to 26 percent of homicides that involved either an Indigenous victim or offender (but not both) and 44 percent (n=1,977) of non-Indigenous homicides.

Homicides occurred across numerous settings but primarily in a home setting whether that was the home of the victim, offender or another person. Overall, since 1 July 1989, almost half (48%) of all homicide incidents (n=3,230) have taken place within the victim’s home and a further 13 percent (n=888) occurred in the offender’s home (not shared with the victim) or other home. Indigenous homicides occurred less frequently in a home setting (41%; n=317) than non-Indigenous homicides (51%; n=2,325).

Homicides involving Indigenous people were more likely than non-Indigenous homicides to occur in public/open spaces. Almost one in five (19%; n=143) of Indigenous homicides occurred in open areas such as public parks/bush land or waterways, compared with seven percent of non-Indigenous homicides (n=329).

Table 2: Selected incident characteristics by victim/offender Indigenous status, 1989–90 to 2011–12a
Characteristic Indigenous victim and offender (n=765) Inter-racial (Indigenous victim OR offender) (n=331) Non-Indigenous victim and offender (n=4,522)
n % n % n %
Plurality of victims/offenders in incidents
Multiple victims 9 1 10 3 221 5
Multiple offenders 46 6 82 25 656 15
Multiple victims/ offenders 2 <1 0 0 39 1
Single victim/offender 708 93 239 72 3,606 80
Unspecified offenders or offender not identified n/a n/a n/a n/a n/a n/a
Incident classification
Domestic 511 67 85 26 1,977 44
Friend/Acquaintance 242 32 142 43 1,946 43
Stranger 12 2 104 31 599 13
Unknown/unspecified 0 0 0 0 0 0
Location
Victim home 317 41 117 35 2,325 51
Other homeb 124 16 39 12 647 14
Health/mental care facilityc 3 <1 2 1 38 1
Retail/recreation establishmentd 14 2 26 8 247 5
Public transporte 9 1 11 3 112 2
Workplace/School 1 <1 1 <1 56 1
Street/road/Highway 115 15 67 20 484 11
Sporting oval/facility 8 1 5 2 19 <1
Open area/waterway 143 19 37 11 329 7
Otherf 25 3 25 8 238 5
Unknown 6 1 1 <1 27 1
Presence of alcohol use
Alcohol use by both victim and offender indicated 532 70 142 43 1,015 22
Alcohol use by victim indicated 30 4 34 10 312 7
Alcohol use by offender indicated 76 10 43 13 432 10
Alcohol use by neither victim nor offender indicatedg 127 17 112 34 2,761 61

a: excludes 1,126 incidents where the Indigeneity of the victim, offender or both was unknown

b: includes offender home and other home

c: includes hospitals and psychiatric facilities

d: includes shopping malls and recreation/food venues

e: includes car parks and public transportation connected facilities

f: includes private motor vehicles and prisons

g: includes incidents where alcohol use was recorded as ‘no’ and ‘unknown’

Source: AIC NHMP 1989–90 to 2011–12 [computer file]

NHMP data on alcohol involvement within homicide incidents relies on two sources:

  1. a toxicology report identifying the presence of alcohol in the victim at the time of death (accessed by the AIC through the National Coronial Information System); and
  2. police assessment of whether the offender was ‘under the influence’ at the time of the incident (derived from data collection templates completed by police for the AIC)

The conclusions that can be drawn regarding an indication of the alcohol use by the victim, offender, or both, at the time of a homicide incident are limited. The effect of alcohol (including the level of intoxication) on the victim or offender is unknown, therefore the contribution of alcohol to the events that led to the homicide cannot be ascertained. Conclusions are also limited because the results presented may be an underestimate. First, offender alcohol use is based on the subjective assessment of police and is frequently recorded as ‘unknown’. Second, whether the police record alcohol use and/or the threshold for being ‘under the influence’ will be subject to recording practices and definitions which may vary between jurisdictions. The data presented in Table 2 and Figure 1 relate to alcohol use by the principal victim and principal offender in each incident only.

However, even with these limitations, the use of alcohol by both victims and offenders appears to be a risk factor for homicide. Since 1989, alcohol use by both victims and offenders prior to the incident has been identified in 26 percent (n=1,726) of homicide incidents. Seventy percent (n=532) of Indigenous homicides were recorded as involving alcohol use by both victims and offenders, as were 43 percent (n=142) of homicides involving at least one Indigenous person (see Table 2). This compares with non-Indigenous homicides where 22 percent (n=1015) were characterised by alcohol use by both victims and offenders. Alcohol use prior to the homicide incident was far more frequently indicated for Indigenous victims (69%) and offenders (72%) than for non-Indigenous victims (27%) and offenders (31%) (see Table 3 and Table 4). Data from the Australian Institute of Health and Welfare (see for example, AIHW 2011a; AIHW 2011b) has consistently shown that Indigenous Australians are no more likely than non-Indigenous Australians to consume alcohol, but when they do, they are more likely to do so in harmful amounts. Again, although the effect of alcohol on the victims and offenders is unknown, it is possible that both parties may have been intoxicated at the time of the incident.

Trend analyses were conducted to assess the change, over time, in the proportion of incidents that indicated alcohol use by both victims and offenders (see Figure 1). The results indicated that over the 23 years from 1989–90 the proportion of incidents involving alcohol use by Indigenous victims and offenders or by non-Indigenous victims and offenders has remained relatively stable, while the proportion of incidents that involved alcohol use by victims and offenders in inter-racial (involving both Indigenous and non-Indigenous individuals) homicides has declined.

Figure 1: Alcohol use by both victims and offenders in homicide incidents by victim/offender Indigenous status (% and trend line), 1989–90 to 2011–12

Source: AIC NHMP 1989–90 to 2011–12 [computer file]

Victim and offender characteristics

Within the NHMP Indigenous status was not recorded for 207 (3%) victims and 420 (6%) offenders. For the purpose of this analysis, victims and offenders whose Indigenous status was unknown were excluded from the final analysis population.

Both Indigenous and non-Indigenous victims of homicide were more likely to be male than female (see Table 3), however, the proportion that were female was slightly higher in the Indigenous victims population than among non-Indigenous victims (41% cf. 35%). Indigenous victims were on average five years younger (mean=31) than non-Indigenous victims (mean=36). Similarly, both Indigenous and non-Indigenous offenders (88% cf. 80%) were more likely to be male. Indigenous offenders were, on average, three years younger (mean=29) than non-Indigenous (mean=32) offenders (see Table 4); a finding that is consistent with Australian population data which shows that Indigenous people are typically younger than their non-Indigenous counterparts.

It has already been noted that a majority of incidents involving Indigenous people involved family members. Intimate partner homicides involving both an Indigenous victim and offender were almost double the proportion of non-Indigenous intimate partner homicides (38% cf. 20%) but similar for homicides involving other family members (20% cf. 17%) (see Figure 2). Further examination by victim sex identified that the proportion of Indigenous male victims of domestic/family homicide was double that of non-Indigenous men (44% cf. 22%). Over three-quarters (78%; n=304) of all female Indigenous victims compared with almost two-thirds (64%; n=1,374) of all female non-Indigenous victims were victims of domestic/family homicides.

The apparent cause of death of victims has, historically, been most frequently attributed to a wound resulting from the use of a knife or other sharp instrument by the offender. Analysis by victim Indigenous status revealed that stab wounds contributed to the death of almost half (49%; n=466) of Indigenous victims compared with almost one-third (31%; n=1,858) of non-Indigenous victims. Gun shot wounds were far less likely to contribute to the death of Indigenous homicide victims (n=42; 4%) than they were to the death of non-Indigenous homicide victims (n=1,313; 22%).

Table 3: Selected victim characteristics by victim indigenous status, 1989–90 to 2011–12a
Indigenous Non-Indigenous
n % n %
951 6,059
Sex
Unknown 2 <1 1 <1
Maleb 560 59 3,908 65
Intimate partner 108 19 286 7
Other family 139 25 586 15
Friend or acquaintance 201 36 1,371 35
Stranger 35 6 396 10
Other relationshipc 42 8 731 18
Unknown 35 6 538 14
Femaleb 389 41 2,150 35
Intimate partner 251 65 952 44
Other family 53 14 422 20
Friend or acquaintance 37 10 241 11
Stranger 13 3 70 3
Other relationshipc 12 3 218 10
Unknown 23 40 247 11
Age group
Under 1 21 2 173 3
1–9 39 4 343 6
10–14 14 1 94 2
15–17 33 3 173 3
18–24 195 21 850 14
25–34 278 29 1,320 22
35–49 281 30 1,673 28
50–64 67 7 854 14
65 and over 8 1 515 9
Unknown 15 2 64 1
Mean age 31 36
Apparent cause of death
Gunshot wound 42 4 1,313 22
Stab wound 466 49 1,858 31
Beating 338 36 1,516 25
Drug overdose (administered by offender) 2 <1 69 1
Drowning/submersion 7 1 88 1
Strangulation/suffocation 17 2 546 9
Poisoning 0 0 49 1
Smoke inhalation/Burns 10 1 143 2
Otherd 51 5 311 5
Unknown 18 2 166 3
Presence of alcohol in victim at time of death 655 69 1,635 27

a: excludes victims whose Indigenous status was unknown or not recorded

b: relationship breakdown excludes victims where relationship status or sex was unknown

c: includes employer/employee, relationship rivals etc

d: includes electrocution and hanging

Source: AIC NHMP 1989–90 to 2011–12 [computer file]

Figure 2: Victim offender relationship by Indigenous status (%), 1989–90 to 2011–12

Source: AIC NHMP 1989–90 to 2011–12 [computer file]

Table 4: Selected offender characteristics by offender Indigenous status, 1989–90 to 2011–12a
Indigenous Non-Indigenous
n % n %
1234 5945
Sex
Male 985 80 5,241 88
Female 249 20 703 12
Unspecified 0 0 1 <1
Age group
10–14 21 2 48 1
15–17 105 9 355 6
18–24 352 29 1,545 26
25–34 406 33 1,790 30
35–49 289 23 1,548 26
50–64 44 4 476 8
Over 65 3 <1 141 2
Unspecified 14 1 42 <1
Mean age 29 32
‘Under the influence of alcohol’ at time of incident 887 72 1,874 31

a: excludes offenders whose Indigenous status was unknown/not recorded.

Source: AIC NHMP 1989–90 to 2011–12 [computer file]

Conclusion

There is currently limited research evidence on how homicide offending and victimisation risks differ between Indigenous and non-Indigenous people and there remains a need for further research to explore the differences between these homicides. This could ensure tailored and culturally appropriate prevention responses, as well as enhancements to victim/survivor rehabilitation and the support services for affected families. Data from the NHMP indicated a greater proportion of family homicides occurring between Indigenous people and a greater use of alcohol at the time of incidents by both Indigenous victims and offenders. As noted by Bryant (2009: 4) however, ‘no single data source is able to provide a comprehensive overview of Indigenous violent victimisation (particularly homicide) and each data source (interviews, surveys, criminal justice data) has strengths and weaknesses’. Therefore, using both qualitative and quantitative data from a variety of sources (such as administrative police and court data, interviews with offenders and the families of victims) would allow for more detailed contextual information regarding the differences between Indigenous and non-Indigenous homicides.

References

  • Australian Bureau of Statistics (ABS) 2012. Australian Demographic Statistics. ABS cat. no. 3101.0 Canberra: ABS
  • Australian Bureau of Statistics (ABS) 2009. Experimental estimates and projections, Aboriginal and Torres Strait Islander Australians 1991 to 2021 series b. ABS cat. no. 3238.0 Canberra: ABS
  • Australian Institute of Health and Welfare (AIHW) 2011a. 2010 National Drug Strategy Household Survey. Canberra: Australian Institute of Health and Welfare.
  • AIHW 2011b. Substance use among Aboriginal and Torres Strait Islander people. Canberra: Australian Institute of Health and Welfare.
  • Bryant C & Willis M 2008. Risk factors in Indigenous violent victimisation. Technical & Background Paper no. 30. Canberra: Australian Institute of Criminology. http://aic.gov.au/publications/current%20series/tbp/21-40/tbp030.html
  • Bryant C 2009. Identifying the risks for Indigenous violent victimisation. Indigenous Justice Clearing House Research Brief no. 6. Canberra: Australian Institute of Criminology and Standing Council on Law and Justice. http://indigenousjustice.gov.au/briefs/brief006.pdf
  • Bryant W & Cussen T 2015. Homicide in Australia: 2010–11 to 2011–12: National Homicide Monitoring Program report. Monitoring report no 23. Canberra: Australian Institute of Criminology.
  • Mouzos J 2001. Indigenous and non-indigenous homicides in Australia: A comparative analysis. Trends and Issues in Crime and Criminal Justice, no. 210. Canberra: Australian Institute of Criminology. http://aic.gov.au/publications/current%20series/tandi/201-220/tandi210.html
  • Wundersitz J 2010. Indigenous perpetrators of violence: prevalence and risk factors for offending. Research and Public Policy series no 105. Canberra: Australian Institute of Criminology. http://aic.gov.au/publications/current%20series/rpp/100-120/rpp105.html