Skip to main content

Indigenous Deaths in Custody

Indigenous Deaths in Custody


Part A Methodology


Paul Blackmore’s photographs illustrate this report. They depict an inspired local initiative. Five years ago the Royal Commission into Aboriginal Deaths in Custody drew attention to the appalling conditions in police lock-ups in rural Australia. The cells in Murgon, three hours north-west of Brisbane, were typical - dirty, disgusting, depressing. Detective Sergeant Ryan and Senior Sergeant McReight got together in 1991, put chalkboards in the cells and started what has become the Murgon Watch-house Art Project. The results have been excellent. Vandalism has almost disappeared. Aboriginal people in the cells, who some police may have perceived merely as drunks or troublemakers, are now recognised for their talents. It makes policing more interesting. The officers take snapshots of the artists with their work before the chalkboards are cleaned, and try to get a copy to them afterwards.

While removing hanging points in cells has benefits, what remains can be a stark environment of  enforced inactivity. The number of ‘Injury in Custody’ forms submitted by police to the Ombudsman’s Offices attests to extent of the problem. But it is very difficult to take one’s life with a piece of chalk. The project is a brilliant and culturally adapted solution to the problem of the onset of suicidal depression in cells. Hopefully the concept will spread, and benefit Aboriginal and non-Aboriginal people alike.

Chapter 1



1.1 This Report uses findings of coronial inquests as a means of externally auditing the implementation of the recommendations of the Royal Commission into Aboriginal Deaths in Custody.

1.2 The circumstances of actual deaths can provide a strong indicator of the extent of implementation by governments.

1.3 A death in custody includes a death occurring whilst a person is watched or guarded by police. In other words, if that person is not at liberty to come and go as he or she wishes.

1.4 Profiles of the 96 deaths from 31 May 1989 to 31 May 1996 are constructed from information obtained from coronial inquests.

1.5 The profiles record the circumstances of death, relevant issues, coronial recommendations and breaches of Royal Commission recommendations.

1.6 Further comment is provided in the profiles where information from coronial inquests is poor, incomplete or other information is available.

1.7 The profiles and other data is statistically presented in Part B. The characteristics of those Aboriginal and Torres Strait Islander people who died in custody are compared to those who died during the Royal Commission period and to non-Indigenous people who died in custody. Arrest and imprisonment rates and types of offences are also presented.

1.8 The profiles are then analysed under Royal Commission recommendations in part C. Each chapter in Part C provides a schedule of recommendations breached and a discussion drawn from the profiles under each recommendation.


This Report uses findings of coronial inquests as a means of auditing the implementation of the Royal Commission’s recommendations. Circumstances of actual deaths provide a strong indicator of the extent of implementation by governments.

The short radius of coronial investigation from the point of death constrains the Report to those recommendations concerning the criminal justice system.

Many coronial findings are brief and have a tendency to focus only upon custodial health and safety. This necessitates the occasional use of supplementary material such as other reports, legislation and communications with relevant organisations.

The following chapter sets out the methodology, parameters and limitations of the Report. The definition of deaths in custody is addressed in section 1. Section 2 outlines the methodology employed in the development of the profiles. Section 3 sets out the scope and approach of the analysis undertaken.

1. Definition of Death in Custody

The definition of a death in custody has been both problematic and contested. Persons who are detained by custodial authorities clearly fall into the classic definition of persons in custody. Inclusion of persons who police are attempting to detain has encountered resistance from police authorities. This question was raised in Eatts v. Dawson. 1 Police officers submitted that the Royal Commission could not investigate a certain death since the person was shot before being arrested. The Full Bench of the Federal Court rejected this narrow definition and held a person to be in custody if they are ‘watched or guarded by police’2. In other words ‘if that person is not at liberty to come and go as he or she wishes to do’.3

Royal Commission recommendations 6 and 41 provide a non-exhaustive definition of a death in custody.4 The first paragraph of both recommendations provides a broad encompassing definition:

a. The death wherever occurring of a person who is in prison custody or police custody or detention as a juvenile. 

The following three paragraphs elaborate the definition: 

b. The death wherever occurring of a person whose death is caused or contributed to by traumatic injuries sustained, or by lack of proper care whilst in such custody or detention; 

c. The death wherever occurring of a person who dies or is fatally injured in the process of police or prison officers attempting to detain that person; and

d. The death wherever occurring of a person who dies or is fatally injured in the process of that person escaping or attempting to escape from prison custody or police custody or juvenile detention. 

Therefore deaths in custody include deaths in institutional settings such as a police van, police or prison cell or in a hospital after transfer from a custodial authority. It also includes deaths from police pursuit such as police car chases or gunshot (by police or deceased) when the police are attempting arrest. 

Police pursuits were generally not investigated by the Royal Commission. This has two consequences. First, comparison of Royal Commission deaths and post-Royal Commission deaths can be misleading. Statistical analysis in this Report will therefore exclude police pursuits when conducting comparisons of the two periods. Second, there are no specific recommendations of the Royal Commission directed towards police pursuit, particularly car chases. Nevertheless, this Report will comment upon the appropriateness of certain police practices. 

The profiles examined in this Report coincide with the database maintained by the Australian Institute of Criminology. However, Appendix II includes a number of other deaths which raise similar issues but technically are not classified as Aboriginal deaths in custody.5 

Postscript. Just prior to publication the Australian Institute of Criminology has removed two deaths from their list of Aboriginal and Torres Strait Islander deaths in custody. Recent information indicates that the woman who died at Brisbane Women’s Prison (11QLD) was not Indigenous. The circumstances of the death of the juvenile who died on Papunya Road (12NT) were considered too remote for it to be classed as a death in custody. However, as discussed in the case profile of 12NT, this death may fall under sub-paragraph (b) of Royal Commission recommendation 6. Furthermore, two of the deaths set out in the appendix (2ASA and A4WA) should be considered as deaths in custody for future analysis. 

2. Profiles 

The methodology adopted draws upon, and develops, the approach utilised by Cuneen and Behrendt in their earlier study6. The profiles are presented in Part E.

The profiles provide a summary of the coronial inquiry. Each profile includes the coronial finding, circumstances of death, issues raised by the Coroner, recommendations of the Coroner and identified breaches of Royal Commission recommendations. Further comment is provided on other areas of concern and additional breaches of Royal Commission recommendations are listed. These two levels are further explained below. 

In deaths where coronial inquests have not been completed, personal details of the death and the known circumstances of the death are recorded. In cases where there is serious concern, comment has been made. 

2.a Coronial Inquests

Coroner’s findings have been chosen as the primary source material for a number of reasons. First, they provide a detailed account of each case. Second, submissions are made by interested parties before the summing-up and a finding is made. Third, coroners are usually empowered by legislation, and were directed by the Royal Commission7, to examine the broader circumstances of a death. Coroners have often the failure of to implement Royal Commission recommendations as well as using the recommendations to reinforce or underline their conclusions. Fourth, coroners generally have the ability to make recommendations to prevent further deaths. 

Each profile provides the personal details of the deceased, the Coroner’s name, date of finding and the formal finding as to direct cause of death. Information is then provided under the following three categories.


The Coroner’s summing up, and transcripts where available, are used to provide the circumstances of death and issues arising in the inquest. In cases where the summing-up is brief and the transcript is not available recourse is made to submissions of legal representatives, statements, exhibits and media reports.8 It should be noted that this section does not attempt to cover all the issues and incidents raised by the Coroner. The emphasis is upon revealing systematic breaches of Royal Commission recommendations. 


Coroners are empowered to make recommendations to prevent further deaths under coronial legislation.9 Recommendations made by coroners are reproduced or summarised. Unofficial recommendations made during a summing-up are also reproduced.

The recommendations are both an indicator of governmental implementation and a specific indicator of implementation by coroners of Royal Commission recommendation 13.10  

Royal Commission Recommendations Breached

This Report views the Royal Commission recommendations as identifying substantial outcomes to be achieved. A recommendation can therefore be breached in two ways: (i) where there is a failure to enact certain legislation or establish and communicate a policy under a recommendation (breach of the express terms of the recommendation); and (ii) where that legislation or policy is not followed in a specific instance, usually through the act or omission of custodial, judicial or medical persons (breach of the substance of the recommendation). For example, prison medical staff should have training in Aboriginal health issues under recommendation 154a. This recommendation may be breached through the absence of training or the failure of staff to utilise the training in a specific situation. The approach of the Report to determining implementation of recommendations therefore differs from government implementation reports which only consider whether policies and legislation have been promulgated. Implementation, if it is to have any real meaning, must be about operational effectiveness, not superficial compliance.  

Where breaches of Royal Commission recommendations are identified explicitly or implicitly by the Coroner these are noted. Positive compliance with the recommendations is not recorded. However, where a systematic breach is found, best practice from other jurisdictions is considered in the Part C Analysis.  

Recording of a breach, however, does not necessarily indicate that the breach contributed to the death. The breach is noted because it has the potential to cause future deaths. 

Many deaths occurred in the interim period between the Royal Commission’s cut-off date for investigations into deaths and the release of the National Report in April 1991. For these deaths, breaches of the 56 recommendations in the Interim Report (Muirhead Report) are listed. The Interim Report, released in December 1988, was intended to provoke immediate action to prevent custodial deaths. Breaches of the recommendations of the National Report have also been recorded in these cases. This allows a comparison over time to determine the extent of progress.

2.b Social Justice Commissioner

Exclusive use of coronial findings is problematic. First, some coroners do not examine the broader circumstances of the death. Second, the coronial radius from the time of death is restrictive in many of the cases concerning prison authorities, which comprise two-thirds of the deaths. The focus is usually upon custodial conditions to the exclusion of issues concerning arrest and imprisonment. Third, coroners’ conclusions cannot always be justified in light of the evidence presented. Fourth, information has not always been available from the coroner: one transcript was destroyed and 25 deaths are yet to be investigated.  

Therefore, further comment is provided where necessary. Additional recommendations breached in light of these comments will also be listed. This section of the profiles does not attempt to fully rectify the conservative nature of coronial processes. It seeks to provide a different perspective in cases which have been insufficiently investigated. In many cases the comment is on breaches of recommendations which appear on the face of the coroners findings.  

3. Analysis 

a. Statistical Analysis

Chapters 2-4 (Part B) of this Report provide statistical analysis. Chapter 2 profiles the characteristics of Aboriginal and Torres Strait Islander people who have died in custody since the Royal Commission and compares them to the deaths investigated by the Royal Commission. The characteristics examined are year of death, gender, age, custodial authority (police custody, prison or juvenile detention centre), the jurisdiction and cause of death. 

Chapter 3 compares Aboriginal and non-Aboriginal deaths in custody between 1990 and 1995. The rate at which Indigenous people and non-Indigenous people die in custody is contrasted by comparing the number of deaths in custody to the different general populations and custodial populations. The characteristics of Aboriginal and non-Aboriginal deaths in custody, as examined in chapter 2, are also compared. 

Chapter 4 examines the principal cause behind the high rate of deaths in custody, the high arrest and imprisonment rates of Aboriginal and Torres Strait Islander peoples. The rates of arrest and imprisonment for Indigenous and non-Indigenous peoples is contrasted. The types of offences for which Aboriginal people are arrested and convicted are examined in comparison to offences for non-Aboriginal people. 

In each of the chapters a number of statistical methods are employed. First, two statistical tests, the t-test and chi-square (?2) test are used to examine whether the difference between numbers is statistically significant.

Second, a two stage process is employed in calculating the different rates of death, arrest and imprisonment for Indigenous and non-Indigenous people. The first stage is to separately calculate a rate which takes into account the different populations of Aboriginal and non-Aboriginal people. This rate is expressed in terms of 100,000 people when the general population is used as the denominator. It is expressed in terms of 1,000 people when custodial populations are used as the denominator. For example, as chapter 3 shows, 7.4 Aboriginal people for every 100,000 Aboriginal people in Australia died in custody in 1994. For non-Aboriginal people, 0.5 out of 100,000 people died in custody in that year. The ratio of these rates is then calculated to give the over-representation rate. Using the same example, Aboriginal people were 15.8 times more likely to die in custody in 1994.  

b. Profile Analysis

Chapters 5-10 (Part C) of the Report draw together the issues raised by the profiles. Royal Commission recommendations concerning the criminal justice system have been broken down into five areas: (i) policing practices; (ii) courts; (iii) custodial conditions; (iv) juveniles; and (v) coroners.  

Each chapter contains a schedule, listing the frequency of breaches of recommendations. The schedule breaks down the recommendations by reference to the authority which recorded the breach (Coroner or the Office of the Aboriginal and Torres Strait Islander Social Justice Commissioner). It also breaks down the recommendations by jurisdiction and into two time periods to determine whether there has been any improvement over the past six years.  

The implementation of each recommendation is then examined. The analysis notes government claims with respect to implementation. Relevant profiles and related material are then discussed.

c. Recommendations

Recommendations by the Office of the Aboriginal and Torres Strait Islander Social Justice Commissioner are contained in most of the chapters. These recommendations have been made where: (i) there is significant evidence of non-compliance with Royal Commission recommendations; (ii) useful recommendations were made by coroners or other organisations; and (iii) there are new circumstances not considered by the Royal Commission.

back to top


  1. G208 of 1990, 23 May 1990, per Morling, Beaumont, Gummow JJ

  2. Ibid. at p.23

  3. Wootten, J, Report of the Inquiry into the Death of David John Gundy, Royal Commission into Aboriginal Deaths in Custody (1991).

  4. recommendation 6 relates to post-death investigations while recommendation 41 os addressed to the Australian Institute of Criminology.

  5. These are:(i) an uncounscious person who died after being taken to a sobering up shelter by police in the belief that the person was intoxicated (A1SA); (ii) a death which occurred after prison authorities had given the deceased an early discharrge (A2SA); (iii) a death where the Coroner did not accept the deceased's Aboriginality (A3NSW); (iv) two motor vehicle crashes after police pursuit in western australia (A4WA, A5WA); (v) a death which occurred after contact with police and court system (A6NSW);and (vi) a death of a person in a mental in under the Mental Health Act (A7QLD)

  6. Cuneen, C and J. Behrendt, Aboriginal and Torres Strait Islander Custodial Deaths Between May 1989 and January 1994; a report to the National Committee to Defend Black Rights, 1994

  7. Recommendation 12

  8. Each summing-up begins with the relevant offence and length of imprisonment if relevant. This information was generally not provided by coroners

  9. This power is limited in Queensland (see chapter 10).

  10. Recommendation 13 states: That a Coroner inquiring into death in custody be required to make findingsas to the matters which the Coroner is required to investigate and to make such recommendations as are deemed appropriate with a view to preventing further custodial deths. The Coroner should be empowered, further, to make such recommendations on other matters as he or she deems appropriate.

back to top




cover of report

A Report prepared by the Office of the Aboriginal and Torres Strait Islander Social Justice Commissioner for the Aboriginal and Torres Strait Islander Commission