A fierce exchange between Warren Mundine and former Indigenous Affairs Minister Nigel Scullion exposed a difficult question that governments still have to confront: when a serious problem has been studied repeatedly, is the better response another inquiry or the implementation of recommendations already sitting on the shelf?
Mundine, who chaired the prime minister’s Indigenous Advisory Council during the dispute, criticised a proposed investigation into the overrepresentation of Aboriginal and Torres Strait Islander people in prisons. He described the decision as a waste of taxpayers’ money and argued that governments already had sufficient evidence to act. Scullion rejected the criticism in unusually direct language, saying Mundine’s assessment was wrong and defending the need to examine the causes and responses in a coordinated way.
The disagreement was not about whether Indigenous incarceration mattered. Both were speaking about a problem with profound consequences for families, communities and the justice system. They differed over what another formal review could contribute. Mundine wanted practical action on existing evidence; Scullion supported a fresh examination across the complicated legal and policy arrangements involving the Commonwealth, states and territories. That distinction makes the confrontation more useful than treating it as a personal insult alone.
Mundine’s comments were striking partly because of his advisory position. Rather than privately endorsing the government’s process, he publicly attacked it. His rhetoric put pressure on the ministers responsible to explain why another study would achieve more than the reports that had come before it. This was political criticism, not a finding that particular officials had misused public funds. The conclusion that an inquiry would be wasteful was Mundine’s judgment about competing priorities.
Scullion responded through ABC radio from Alice Springs. He suggested that Mundine was having an unusually bad day and made clear that he strongly disagreed. He also described their relationship as friendly despite their willingness to argue. The exchange should therefore not be retold as a rupture or a physical confrontation. It was a sharp disagreement expressed publicly by two men familiar with each other’s work and immersed in the same broad policy area.
The government announced the fresh inquiry in the context of longstanding concerns about imprisonment rates and the treatment of Aboriginal and Torres Strait Islander people in custody. Then attorney-general George Brandis argued that the situation amounted to a national tragedy and that the Commonwealth would work with states and territories, which have primary responsibility for policing, criminal law and prisons. The division of powers is central to understanding why both better evidence and better action can be difficult to achieve.
Criminal justice is not controlled by one minister in Canberra. Decisions affecting arrest, bail, sentencing, rehabilitation, parole and prison operations are made across state and territory institutions, while the Commonwealth influences programs, funding, national policy and cooperation. A federal inquiry can investigate systemic drivers and propose changes, but many practical reforms still depend on others accepting, financing and implementing those proposals. Simply publishing a report does not reduce incarceration by itself.
That is the strongest part of Mundine’s criticism. If earlier investigations have already identified problems, governments should be able to explain what has been done with the recommendations and why similar failures persist. A new report without accountable implementation risks consuming professional time and public money while providing little additional benefit. The expenditure is not automatically improper, but taxpayers are entitled to ask what outcomes are expected from the work.
The strongest response is that data and circumstances change. A formal inquiry can identify differences between jurisdictions, revisit ineffective assumptions and gather evidence from affected communities whose experiences are not properly captured in administrative numbers. It can also place a problem on the public record in a way that is harder to dismiss. Whether that advantage is worth the cost depends on how the inquiry is designed, whom it hears from and whether its proposals lead to a practical response.
It is essential not to freeze the story at the moment of the argument. The Australian Law Reform Commission subsequently completed its Pathways to Justice inquiry into the incarceration rate of Aboriginal and Torres Strait Islander peoples. Its final report addressed factors contributing to imprisonment and proposed reforms across the criminal justice system. The later report did not retroactively prove that Mundine’s scepticism was frivolous, or that Scullion’s confidence was guaranteed. It demonstrated that the inquiry proceeded and produced a concrete piece of work whose impact can be assessed separately.
The justice questions have real household and community consequences. Imprisonment can remove an income earner or caregiver from a family, interrupt education and employment, create housing insecurity and contribute to intergenerational disadvantage. Those are reasons to evaluate prevention and rehabilitation alongside prison administration. But assigning a precise financial cost to any community without proper evidence would risk reducing a human-rights and public-safety issue to an invented budget calculation.
Safety matters too. Communities need protection from violence and repeat offending. A sound justice policy cannot be assessed solely by whether prison numbers rise or fall; it also needs to examine harm to victims, fairness in bail and sentencing, access to treatment, and whether people leaving custody can reintegrate without reoffending. The fact that Indigenous Australians are overrepresented in custody makes those questions more urgent, not less nuanced.
An evidence-based debate should distinguish the number of people held in prisons from the rates of imprisonment relative to population. Historical figures cited around the announcement describe the scale of the problem at that point, not a current snapshot. Using them in a new article as if they measured present conditions would be misleading. The more enduring point is the disproportionality that prompted the inquiry and the need to track comparable outcomes over time.
Consultation also matters. Aboriginal and Torres Strait Islander communities are not a uniform group with one view about policing, sentencing or reform. Local priorities, histories and service availability differ markedly. A national review can provide a framework, but durable implementation usually depends on governments working with the people affected and the organisations delivering legal assistance, family support, housing, health and rehabilitation services.
The political incentive to announce a review can be strong because it signals concern. The incentive to implement difficult changes may be weaker because benefits take time to show and responsibility is divided across budgets and agencies. That gap is precisely why governments should identify measurable commitments, publish progress and explain departures from expert recommendations. A review becomes more valuable when its findings inform an accountable program rather than simply joining an archive of reports.
For readers weighing the Mundine-Scullion clash, neither man’s strongest language should decide the policy question. The important test is whether the justice system’s outcomes improve and whether public money is linked to effective interventions. Mundine demanded action rather than another diagnosis. Scullion believed there was a role for further investigation. Both positions were responses to a real and serious problem, not evidence that one of the participants opposed the wellbeing or safety of Indigenous Australians.
One measure of implementation is whether services can reach a person before contact with the criminal justice system intensifies. Access to suitable housing, mental health care, alcohol and drug treatment, legal assistance and family support may affect the options available to courts and communities. The scale and effectiveness of any intervention cannot be assumed without evaluation. But the range of services shows why incarceration policy is not a problem that police and prisons can resolve alone. Any government announcing a review should say how it will coordinate these responsibilities.
Better data can also help reveal where different decisions contribute to unequal outcomes. Police contact, remand, sentencing, access to legal representation and conditions imposed on release are separate stages. A reform that changes only one stage may have limited effect if problems elsewhere remain unaddressed. Independent researchers and community-controlled organisations can help identify these differences. Government should not treat an increase in the volume of data as a substitute for explaining the action it will take on that evidence.
The ALRC inquiry gives the disagreement a tangible point of reference. Its work considered the rate of Indigenous incarceration and options across legal and policy systems. The fact that a report exists allows later governments to check which recommendations were accepted and what followed. That exercise needs to distinguish recommendations requiring legislation from measures that can be delivered administratively, and reforms dependent on states from those the Commonwealth can undertake. Otherwise responsibility can be shifted endlessly between jurisdictions without clear accountability for results.
The financial argument should be made in the same disciplined way. An inquiry costs money, and its value should be assessed against the public benefit of improved policy decisions. Prison operations, remand and repeated contact with justice services also have costs, as do failures of prevention. Comparisons should use reliable and compatible data rather than a dramatic national figure created for a headline. The strongest case for action is both moral and practical: reducing avoidable harm while making government responsibilities measurable and transparent.
The argument remains an instructive case study because it forces a sequence of questions. What do governments already know? Which recommendations have been adopted? What gaps in evidence remain? Who has the power and funding to act? And how will affected communities be able to judge the result? The sharp rebuttal attracted attention, but the answers matter far more than the insult. A credible public policy process must both learn and deliver, and be prepared to demonstrate that it has done each.
