Indigenous Deaths in Custody:

Indigenous Deaths in Custody

Part D - Implementing the Recommendations

Chapter 11. Accountability and the Reporting Process
Chapter 12. Alternative Mechanisms to Ensure Implementation

One is irresistibly reminded of the likening of bureaucratic activity to  the sex life of elephants: much trumpeting, a lot of activity at high level  and no outcome for three years.

Aboriginal and Torres Strait Islander Social Justice Commissioner,
Second Report, 1994

Chapter 11
Accountability for Implementation - The Current Process


11.1 The reporting process was flawed from the outset, and has not resulted in accurate evaluations of progress in implementing recommendations at either Commonwealth or State and Territory level. This is a fundamental question of public accountability.

11.2 The recommendations are still largely current, although there are gaps (for example, no reference to the 'trifecta' in the Royal Commission reports) and structural changes have occurred in custodial arrangements (for example, speedier transfer to remand centres, which may, in part, account for the proportional shift in the location of death from police to prison custody).

11.3 Monitoring is not useful unless there is a considered plan for the implementation of Royal Commission recommendations by Commonwealth or State and Territory Governments. Responsible departments are encouraged to develop plans in a six stage process, by:

1. reviewing current activities;

2. developing policies and programs; 

3. setting goals or targets;

4. allocating responsibility for implementation;

5. ensuring adequate communication and training supports the plans; and

6. establishing evaluation mechanisms.

The implementation of Royal Commission recommendations currently starts and stops at step 1.


How many of the recommendations have been implemented? What progress has been made since the Royal Commission into Aboriginal Deaths in Custody? Indigenous people, the general public and journalists who ask this question are often frustrated by the lack of simple answer.

The Royal Commission produced 110 reports, 10,000 pages of script: 339 recommendations were made. They all reflect the diverse causes of high levels of incarceration, and corresponding disproportionate numbers of Aboriginal deaths in custody, in this country.

There are nine jurisdictions in Australia with responsibility for implementing the recommendations - the Commonwealth, the six states and the two territories. A very large number of 'implementations' were required, closer to 3000 than 339. Accordingly, the implementation and monitoring process had to be co-ordinated and well planned if it were to yield transparency and effectiveness.

The Commonwealth Government promised $400 million towards the implementation of the recommendations in 1992. The States and Territories made commitments to most of the recommendations, and promised to fund their areas of responsibility.

Recommendation 1 established the monitoring and reporting structure. All Australian governments committed themselves to it. It states in part:

  1. That the Commonwealth Government and State and Territory Governments, in consultation with the Aboriginal and Torres Strait Islander Commission (ATSIC), agree upon a process which ensures the adoption or otherwise or recommendations and the implementation of the adopted recommendations will be reported upon on a regular basis with respect to progress on a Commonwealth, State and Territory basis;

  2. That such reports should be made not less than annually and that, subject to the agreement of its Commissioners to do so, ATSIC be given special responsibility and funding to enable it to monitor the progress of the implementation of the adopted recommendations and to report thereon to the Aboriginal and Torres Strait Islander community...

The recommendation has not been meaningfully implemented. Unfortunately, accountability for the implementation of recommendations of the Royal Commission is still unacceptably poor. First, the mechanics of the federal system work against accountability. The Royal Commission was a Commonwealth undertaking. The Commonwealth has a funding role, a leadership role in pressing the states for implementation, and an operational role in limited areas. But the recommendations were largely directed at state and territory governments and agencies, such as police, prisons, corrections health services, Attorneys-General and the courts. The limitations placed on the Government Response Monitoring Unit within ATSIC were overwhelming - the Unit was only to report on the activities of Commonwealth agencies, and was given no powers to require agencies to provide them with information and documents relevant to the implementation of the recommendations.

1. The Reporting Process

At a meeting of the Ministerial Council on Aboriginal and Torres Strait Islander Affairs in 1992 all governments committed themselves to regular reporting on the implementation of the recommendations of the Royal Commission.

Since then, bookshelves of implementation reports have been published by the nine governments. Unfortunately, the reports tend to take a 'public relations approach' to monitoring. Characteristically, the material is very poorly organised.

The responses often miss the point of a recommendation. The approaches taken to the concept of what constitutes implementation are arbitrary. General assertions of compliance are made without any evidence to back them up.

The main reason the reports are such poor indicators of performance by State Governments is the procedure for their production. Towards the end of a reporting year, an employee of a department which has the responsibility for the implementation of certain Royal Commission recommendations is told to come up with evidence. That employee must frantically unearth any existing policies which sound as if they might resonate with a recommendation and list them as evidence that the department is doing something. The process is fatally flawed - departments have usually not drawn up a coherent plan which the implementation report can address. The result is an enormous waste of resources in publishing and distributing the reports.

The reports give a very misleading impression of the daily experience of Aboriginal people. A 1994 publication from South Australia, Voices Behind the Razor Wire, was written from a prisoner's perspective. It strongly criticised the lack of reality content in the Government's implementation report. 1 A companion report noted that the South Australian Government had claimed implementation of 107 recommendations, most of which the Aboriginal Legal Rights Movement believed were clearly not implemented. 2 State and territory reports brazenly claim implementation of recommendations at the same time that coroners, in inquests into Aboriginal deaths in custody, are criticising the same agencies for failure to implement them.

The lack of independence in reporting is a consistent feature. It is completely unacceptable. The departments responsible for implementing the recommendations prepare or send their responses to the agency responsible for publishing the report. By way of illustration, the New South Wales Police Service itself prepares the response to Recommendation 87, that arrest should be a measure of last resort, and sends it to the Department of Aboriginal Affairs.

There is very little incentive for the Head of a Department to submit a critical response. There have reportedly been instances in which departmental employees have prepared frank responses, only to have a more senior employee censor unpalatable facts. The employee who prepared the response is prevented by secrecy legislation and career recriminations from revealing what has happened. The responses are often submitted months after they are prepared and close to the publication date, giving the Aboriginal Affairs Department or equivalent agency little time to ascertain the quality of the material being published. In some jurisdictions, the State or Territory Cabinet Office has then been responsible for long delays in publication.

Often the recommendations are treated as a strict code, and not placed in the context of the Royal Commission reports. Responses by governments must shift towards a deeper understanding of the intended purpose of any single recommendation and recognise the inter-relatedness of the recommendations.

A former employee of an Aboriginal Affairs department described not being permitted to use their experience as a solicitor in an Aboriginal Legal Service to critically comment on the input of other departments into an implementation report. They were told that the minister responsible for the Aboriginal Affairs portfolio would not want to offend the minister responsible for the Police or Corrective Services portfolios. The employee also described unsuccessful attempts to seek further information when another department claimed implementation by referring to 'policy no. XYZ,' details of which were not disclosed to the Aboriginal Affairs department.

Continuing the example of Recommendation 87 in New South Wales, the 1993/94 report claims the recommendation is implemented without giving any evidence at all. 3 Surely it is reasonable to expect that figures indicating the increased use of summons or court attendance notices as an alternative to arrest would be reported. The New South Wales Bureau of Crime Statistics could have had input in verifying the claim. Unfortunately, the New South Wales Police Service is only now beginning to include Aboriginality in the information routinely gathered when a person is arrested. By not collecting this information for so long, the Police Service effectively avoided accountability.

SJC Recommendation

59. State Police Departments publish regular statistics on, and analysis of, the incidence use of arrest, summons and court attendance notices to back up claims of implementation. The research should be conducted to reveal variations between districts, and should convey ethnicity and charge information.

60. All government departments and agencies should record and publish statistical or other objective data relevant to the implementation of all Royal Commission recommendations for which that department or agency has responsibility. Annual targets should be set by reference to such statistics or data to provide a measure of effective implementation or otherwise.

Agencies which could potentially ensure that the reports better reflect reality, are being left out of the reporting process. Apart from coroners, probably the best example would be the Office of the Ombudsman, or the equivalent body dealing with police complaints. The Ombudsman receives complaints which relate directly to the law and justice recommendations, has ready access to the prisons and can gauge compliance with the recommendations from a more critical perspective. Other agencies which could provide valuable input include the agency responsible for crime statistics and the Auditor-General's Office.

In the world of accounting, auditors have to certify that the financial reports of a company represent a true and fair view of its financial state. The auditor can be sued if the reports give a misleading picture. Compare this model to reporting on the implementation of Royal Commission recommendations. Reports are published whether or not they represent reality. They leave out vital details. There is no independent body allocated to ensure self-interest does not influence the quality of reports. Given the level of expenditure of public funds for so little return, a stricter method of accountability is well overdue. A recommendation proposing a better model appears at the end of this chapter. A survey of most of the jurisdictions follows.

2. The Commonwealth

The first Commonwealth implementation report was scrutinised by this office in a submission to the House of Representatives Standing Committee on Aboriginal and Islander Affairs in 1994. The criticism is summarised in the Social Justice Commissioner's 1994 Second Report. 4

There has been some progress since, with the inclusion of some indicators and details of programs, but the criticisms remain largely valid.

The leadership role of the Commonwealth in ensuring that recommendations are implemented has not been taken up. The Commonwealth can influence the states in a number of ways in areas in which it does not have primary responsibility, such as policing, imprisonment and the criminal law. One method is to put performance conditions on grants to State and Territory Governments. Another is to take an active leadership role. One important forum for criminal law reform is the Standing Committee of Attorneys General, which consists of Attorneys General from all the States and Territories as well as the Commonwealth. Another important forum is the Council of Australian Governments (COAG), which co-ordinates policy between the two levels of government. 5

Commonwealth reporting does not adequately respond to vitally important recommendations, such as recommendation 87 about arrest practices, even where it has direct responsibility. The Commonwealth Implementation Report 1994 stated that in Jervis Bay, the only place the Commonwealth actually has operational responsibility for custody, a handful of Aboriginal people were summonsed rather than arrested. 6 In 1994, the Australian Bureau of Statistics revealed the extent of the problem: 45-50 per cent of Aboriginal men in their twenties had been arrested at least once in the past five years. 7 In any case, the Commonwealth was not fully addressing its responsibility, as Commonwealth legislation - such as the social security laws - results in the imposition of a significant number of penalties of imprisonment each year.

The Commonwealth Government Response Monitoring Unit only holds a mandate to report on the progress of Commonwealth agencies. An allocation of $7 million of Commonwealth funds over five years for the purpose of monitoring recommendations was inefficiently used because police, prisons and the courts are state and territory responsibilities.

To sum up, a good proportion of the $400 million of Royal Commission funding has been given to the States and Territories where the real action is. They hold responsibility for police, prisons and the courts.

The following questions are unanswered. Where are the details of State and Territory expenditure? What has been achieved? Where are the bench marks of past and future achievement? This is a matter of basic accountability for the expenditure of tax-payers' money. ATSIC is closely audited. Where is the audit of State and Territory expenditure? Where are the results? The statistics of imprisonment rates and deaths in custody are damning evidence of ineffective expenditure and waste.

The Monitoring Unit within ATSIC must devote its energies towards a more thorough evaluation of policies and practices designed to achieve the objects of the recommendations. To this end, it requires power to examine the activity of Commonwealth departments and agencies, not merely to act as their publishing house. It must act in this area.

There has been no Commonwealth response to Recommendation 92, that imprisonment be a last resort for Aboriginal people, despite the Commonwealth's primary responsibility for imprisonment in social security matters.

3. The States and Territories

The States and Territories have the everyday responsibility for police, prisons, prison health services and criminal law. Overall their reporting has not been satisfactory. For example, relevant criticisms by coroners have been conspicuously absent from the reports. The exception is the South Australian report for 1994/95, which is to be commended for addressing more difficult issues and for taking a more critical and credible approach to reporting progress in that State. 8

The Western Australian Government indicated in its latest implementation report that most of the recommendations it had adopted have been implemented. The Western Australian Aboriginal Legal Service also reports on the progress of the Western Australian Government in implementing the recommendations. 9 It analysed 216 recommendations, and found that 16 (8%) have been fully implemented, 107 (49%) have been partly implemented and 43 per cent have not been implemented at all, or implemented at a level which is entirely unsatisfactory. The difference in the substance of the two reports is extreme. Even if the true position were only mid-way between these extremes, the position would be extremely poor.

The Northern Territory Government's report indicated that it had implemented 216 of the recommendations for which it has primary responsibility, with 45 in the process of implementation and only six as yet not implemented. 10 Scrutiny by this office indicated a large number of recommendations which were either not implemented or there was no evidence given of implementation.

Implementation reports have not provided frank reporting or objective evidence of progress on Royal Commission recommendations. Brief examples from state and territory reports further illustrate the point.

a. Western Australia

Recommendation 164 called for prisoners and detainees not to be charged for inflicting self-harm. The Western Australian Government made a one word response - 'implemented'. 11 The Chief Justice of the District Court imposed a three month sentence for criminal damage on a twenty-one year old Aboriginal prisoner who set fire to cell while locked inside it. He was refused medication for a chronic migraine, and was fortunate not to die from smoke inhalation. 12 He had already been punished by two months segregation. 13

b. New South Wales

Recommendation 174 calls for the employment of Aboriginal Welfare Officers in prisons. There are 22 prisons in New South Wales. The response to the recommendation indicates that there are seven designated positions for Aboriginal Welfare Officers in New South Wales prisons. What the response does not mention is that these positions have a very poor record of being filled. Even when they have been filled, on many occasions they have been filled by non-Aboriginal people. Coroners have indicated in two recent inquests (67NSW and 72NSW) that there was no Aboriginal welfare worker at the prison.

c. The Northern Territory

Recommendation 95 calls for programs to reduce imprisonment for driving offences. Unspecified community policing initiatives, general road safety campaigns, school visits and television advertisements are put forward to support the claim that the recommendation has been implemented. However figures from 1993/94 reveal that 92.4 per cent (231/250) of persons imprisoned for driving offences were Aboriginal.

By contrast only 45 per cent of people given Community Service Orders for driving offences were Aboriginal. 14 Preliminary research by the Northern Australian Aboriginal Legal Aid Service indicates that the ratio of Aboriginal to non-Aboriginal people charged for more serious driving offences is 1:1. The imprisonment rate of more than 9:1 indicates a failure to implement recommendation 95.

d. South Australia

Recommendation 226 calls for the independent investigation of complaints against police, and the employment of Aboriginal people in an independent police complaints body. It is one of the 107 recommendations that the South Australian Government has claimed is fully implemented, and accordingly no longer reports on. The latest implementation report does not even mention this recommendation. Ironically, the Independent Royal Commission Into Aboriginal Deaths in Custody Monitoring Report 15 in South Australia notes major problems with the police complaints system, including non-compliance with recommendation 226, and devotes more space to this recommendation than to any other.

e. Victoria

Recommendation 62 called for programs to reduce the involvement of young Aboriginal people in the criminal justice system. The Victorian Government's implementation report outlines a great deal of activity and a number of programs. 16

The response by the Victorian Aboriginal Justice Advisory Committee indicates that a number of programs listed had either been defunded or were unfunded. 17

f. Tasmania

While frank in recognising some recommendations had not been implemented, the implementation report does not substantiate its responses, especially in regard to conditions in juvenile detention centres (recommendation 167). A review is reported to have resulted in further training for staff and a closer liaison with the Aboriginal community. There are no details of conditions in the centres.

The Tasmanian report is the worst example of using inconsistent and imprecise terminology. Instead of standard terms 'implemented', 'part implemented or 'not implemented', the report uses terminology such as 'covered,' 'supported' and 'ongoing'.

g. Queensland

Recommendation 80 called for sobering-up shelters to which intoxicated persons could be taken as an alternative to prison cells. The Queensland implementation report indicates that three such centres have been established, with difficulty experienced in obtaining sites in Townsville and Rockhampton. 18 The centres were the main initiative of the $5.5 million diversion from custody program, which formed the bulk of government's $10.5 million commitment to implementing the recommendations. Of the five promised sobering-up shelters, the Mount Isa facility has been a success, but the Brisbane and Cairns centres are operating in temporary facilities (the Cairns shelter doubles as a place to stay for people visiting the town for medical treatment and a student accommodation hostel). There is therefore only one facility operating in appropriate, permanent facilities.

4. The Extent to which the Recommendations are Outdated

In some cases, the recommendations need updating. An example is recommendation 86, which calls upon police departments to take various measures to ensure they refrain from arresting Aboriginal people for offensive language. The recommendation contemplated situations in which Aboriginal people were provoked by police into offending. Instructions not to use this charge have often led to the laying of more serious charges, such as malicious damage to property (often a police uniform), resisting arrest, or assaulting police. Responses indicating reductions in the use of offensive language charges may merely indicate that the consequences of 'answering back' to police have become more grave.

Governments in their implementation reports certainly show no sign of taking a dynamic approach to recommendations such as the above. True monitoring would respond and adapt to situations not falling within the precise terms of a recommendation, but within its spirit.

Some structural changes have occurred since the Royal Commission. One change is that people are spending less time in police custody before they are placed in a remand centre. The recommendations aimed at police were intended to cover the dangerous initial period of incarceration when detainees may be withdrawing from alcohol or drugs, or depressed because of the incident which led to their detention. If detainees are moved straight into remand centres, the recommendations aimed at police should also apply to prisons that hold such fresh remand prisoners. The case of the 19 year old man who had consumed drugs heavily before he died in Parklea Prison (67NSW) illustrates the need to extend the reach of recommendations, such as recommendation 137, in order to meet their purpose.

There are many other recommendations which could have been drafted more broadly. For example, recommendation 144 provides that Aboriginal prisoners should not be placed alone in police cells unless there is a good reason for doing so. This recommendation should apply equally to remand facilities, as well as to prisons and other types of custody, such as involuntary detention in mental institutions. The recommendation about visits to police facilities is another example - it should have covered remand centres as well, as the case of the man who died in the Arthur Gorrie Remand Centre (see 45QLD) demonstrates.

The growing problem of drugs in prison was one which the Royal Commission did not address in adequate detail. Part D documents a number deaths in custody which resulted either from overdoses or as a result of drug seeking behaviour. This study does not have the scope to do justice to the issues. However, problems which may not immediately be apparent to an outsider include: a) the control by prisoners of people on the outside; b) the effect on the children and family of prisoners; c) the lack of involvement of groups such as narcotics anonymous in prisons; d) deaths immediately upon release from prison as a result of 'hot shots' (overdoses which occur because drug users are unaccustomed to the 'purity' of street heroin); e) poorly run methadone programmes with inadequate counselling; f) offenders (including young minor offenders) who first become addicted to heroin or methadone inside prison; g) the storage and trading of prescribed drugs; h) HIV issues, and the fact that needles are valuable commodities and must be shared; i) violence following the incurring of drug debts; j) the dangers to prison employees; and k) the brutality which results from the state of virtual civil war existing inside many prisons as opposing groups vie for control of the prison drug trade. Options such as drug-free zones or drug-free prisons, monitored by random drug testing, should be explored.

With these qualifications, the evidence from the coroners' reports into deaths in custody indicates that the recommendations are still all too current. Coroners inquiring into Aboriginal deaths in custody often made recommendations identical to those made by the Royal Commission without referring to the recommendations. Often when they did not, they should have. If government agencies responded to their implementation in a committed and co-ordinated manner, many of the problems facing Aboriginal people which lead to deaths in custody could be addressed.

There are no indications that there are any grounds to discontinue the implementation process. Reporting must be radically improved and maintained until the problems addressed by the recommendations are resolved.

Currently, state agencies responsible for the implementation of recommendations reach the end of their reporting cycle without any coherent plan for the implementation, and without any real ability to assess progress. Reform is urgently needed.

A better system for reporting would result if a precedent drawn from the field of disability discrimination were followed. Part 3 of the Disability Discrimination Act 1992 (Cth) sets down criteria for action plans for the implementation of the objects of that legislation, which state departments can lodge with the Human Rights and Equal Opportunity Commission. 19 State departments are encouraged to draw up plans in a six stage process, by:

1. reviewing current activities;

2. developing policies and programs;

3. setting goals or targets;

4. allocating responsibility for implementation;

5. ensuring adequate communication and training supports the plans; and

6. establishing evaluation mechanisms.

By comparison, the implementation of Royal Commission recommendations currently starts and stops at step 1. 20


SJC Recommendation

61. State and Territory Governments use standard terminology ('implemented,' 'part implemented' or 'not implemented') in their reports, and support these claims with appropriate evidence drawn from their own records and data from other agencies with direct involvement. Three examples of interested agencies who should provide supporting evidence in the criminal justice area are the State and Territory Ombudsmen or equivalent, Anti-Discrimination Commission or equivalent, the Bureau of Crime Statistics and the Office of the Auditor-General.

62. State and Territory Governments acknowledge the shortcomings in their reporting on implementation of Royal Commission recommendations, and commit to the model of comprehensive Action Plans by responsible departments (similar to those provided for in Part 3 of the Disability Discrimination Act (Cth) 1992). Interested Aboriginal organisations should be invited to comment as part of the process in the drafting of these action plans.


1. The Sansbury Association Inc and Patricia Tresize, Voices Behind the Razor Wire, Sansbury Association and the Aboriginal Legal Rights Movement, November 1994.

2. Aboriginal Legal Rights Movement and the Aboriginal Justice Advocacy Committee, 107  Recommendations - Have They Been Implemented? A Joint Report Responding to the South Australian Government's 1993 Implementation Report, September 1994.

3. Government of New South Wales, Implementation of Government Responses to the Recommendations  of the Royal Commission into Aboriginal Deaths in Custody: New South Wales Government Report, Vol.1, 1993/94, p.141-144.

4. Aboriginal and Torres Strait Islander Social Justice Commissioner, Second Report, 1994 p 75-97.

5. The National Commitment to the Provision of Services to Aborigines and Torres Strait Islanders, produced by COAG, specifically referred to the co-ordination of responses to problems identified by the Royal Commission into Aboriginal Deaths in Custody.

6. ATSIC Government Response Monitoring Unit Implementation of the Commonwealth Government Responses to the Recommendations of the Royal Commission into Aboriginal Deaths in Custody, Annual Report, 1994, p130-131.

7. Australian Bureau of Statistics, National Aboriginal and Torres Strait Islander Survey 1994 Detailed Findings, Catalogue No 4190.0, 1995, p57.

8. Department of State Aboriginal Affairs, Royal Commission into Aboriginal Deaths in Custody: 1994/95 Implementation Report, July 1996, p.35-43.

9. Aboriginal Legal Service of Western Australia (Inc) Striving for Justice Volume 3: Report to the Western Australian Government on the Implementation of the Recommendations of the Royal Commission into Aboriginal Deaths in Custody 1996 p1.

10. Northern Territory Office of Aboriginal Development, Royal Commission into Aboriginal Deaths in  Custody: Implementation of the Recommendations 1993/94 p.2.

11.. Aboriginal Affairs Department, Government of Western Australia 1995 Implementation Report: Royal Commission into Aboriginal Deaths in Custody Vol 1 p.158.

12. R v Button (Unreported, District Court of WA, Hammond CJDC, 3 May 1996)

13. In a similar case, an Aboriginal man in Townsville prison in Queensland (43QLD) did the same in 1994, three days before he hanged himself.

14. Northern Territory Government, Department of Corrective Services Annual Report 1994/95, p72

15. Aboriginal Legal Rights Movement Inc (SA) & Aboriginal Justice Advocacy Committee (Inc), 1996, p85-89.

16. Victorian Government, Royal Commission into Aboriginal Deaths in Custody: Victorian Government 1994 Implementation Report , p.86-89 ('Victorian Implementation Report 1994').

17.. Aboriginal Justice Advisory Committee (Victoria), Response to the Victorian Government 1994 Implementation Report, p.11.

18. Queensland Government, Royal Commission into Aboriginal Deaths in Custody: Queensland Government Progress Report on Implementation, December 1994, p.106.

19. Sections 59-65.

20. For a detailed guide to these implementation plans see eg Australian Local Government Association Disability Discrimination Act - Action Plans: A Guide for Local Government 1995.

 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