PAINTING THE PICTURE OF INDIGENOUS WOMEN IN CUSTODY IN AUSTRALIA

This article seeks to paint the contemporary picture of Indigenous women in custody in Australia. In particular, the article presents and analyses the most recent data available on prisoner numbers, imprisonment rates, age, sentence length, offence type and recidivism. The article then considers some of the characteristics of Indigenous female prisoners, including their physical and mental health, their role as mothers, and their exposure to family violence. The implications of Indigenous women’s representation and circumstances in Australian prisons are also examined.


INTRODUCTION
This article seeks to paint the contemporary picture of Indigenous women 1 in custody in Australia.In particular, the article presents and analyses the most recent data available on prisoner numbers, imprisonment rates, age, sentence length, offence type and recidivism.The article then considers some of the characteristics of Indigenous female prisoners, including their physical and mental health, their role as mothers, and their exposure to family violence.The implications of Indigenous women's representation and circumstances in Australian prisons are also examined.
As has been noted previously, 2 a silence often pervades the consideration of Indigenous women in the criminal justice system.Where information is available, it tends to focus on Indigenous women as victims, 3 not as offenders, although, as the discussion below demonstrates, this is often inextricably linked with their offending.
In addition, much of the existing literature is qualitative and/or anecdotal in nature.Where quantitative information is presented, it tends to consider Indigenous status or gender, but not the intersection of the two.For example, in its Prisoners in Australia dataset, the Australian Bureau of Statistics (ABS) includes information on aggregate sentence lengths on the basis of offence types and Indigenous status, but this information is not disaggregated on the basis of gender.Information of this nature would assist however in determining whether the gender-and culturally-specific needs of prisoners are being met, taking into account differences in offence types and the length of sentence received.
The issue of intersectionality was a key focus in the 2002 annual report of the then Aboriginal and Torres Strait Islander Social Justice Commissioner (ATSISJC), where it was noted: In a general sense, intersectionality refers to the connection between aspects of identity, such as race, gender, sexuality, religion, culture, disability and age.An intersectional approach asserts that aspects of identity are indivisible and discussing them in isolation from each other results in concrete disadvantage.'Intersectional discrimination' refers to the types of discrimination or disadvantage that compound on each other and are inseparable. 4e ATSISJC also devoted a significant portion of the report to the issue of Indigenous women in the criminal justice system, noting that: Aboriginal women remain largely invisible to policy makers and program designers with very little attention devoted to their specific situation and needs.This is of critical importance, particularly because of the impact that imprisonment has on Indigenous families and communities (especially through separation from children). 5her observations in the 2002 ATSTSJC report which regrettably remain just as relevant today include the difficulties in tracking national trends in crime and sentencing, due in part to the way data are collected and the issues associated with small numbers in some jurisdictions, especially Tasmania and the ACT, which 'may result in outcomes which appear disproportionate to the true conditions … [and] reduces the extent to which meaningful analysis can be undertaken'. 6In particular, although the quarterly ABS data give some understanding of the flow of prisoners through the system through changes in the daily averages, especially in relation to the number of prison receptions, these are not disaggregated by Indigenous status.The prison census data, on which analysis of the types of crimes committed by prisoners is based, only collect information on prisoners who are in custody on 30 June of each year.Prisoners who only serve short sentences are therefore not recorded in this dataset, and 'underestimate Indigenous women coming through the prison system on shorter sentences for more minor offences'. 7Baldry has commented on the limitations of relying on prison census data, suggesting that such information is 'misleading for those working with people being released, especially when considering women prisoners.Such census figures do not take into account the flow-through numbers -the numbers of prisoners who flow through the system over the period of a month, or 6 months or a year'. 8Notwithstanding these limitations, it is suggested that the following data are valuable for contextualising our understanding of Indigenous women in the Australian correctional system, and contribute to the emerging knowledge base on this doubly marginalised population.This section presents key data on Indigenous women in the correctional system, drawing on the most recent publications from the ABS. 9 prison in the Northern Territory and 43 percent in Western Australia. 13In Victoria, by contrast, Indigenous women accounted for only a small proportion of female prisoners (7%).Although most jurisdictions' correctional agencies make specific mention of the needs of Indigenous women in their policies,14 the numbers will not always be sufficient to justify the development or implementation of programs are adapted for address both culture-and genderspecific needs.A clearer understanding of the distribution of Indigenous women across Australian prisons is therefore required to ensure the rollout of services that meet these needs, whether an Indigenous woman represents a lone figure in her jurisdiction's correctional system or, as in the Northern Territory, is in the majority.

C Imprisonment rate
As set out in Indigenous women were therefore more than 15 times more likely to be imprisoned than the general adult female population, compared with 13 times for Indigenous men.Although the overall imprisonment rate for Indigenous women is clearly much lower, this demonstrates a higher level of over-representation for Indigenous women than men, a reality rarely reflected upon in the criminal justice discourse.

D Age
Behrendt, Cunneen and Liebesman15 have noted that Indigenous female prisoners tend to be younger than their non-Indigenous counterparts and this is borne out by the ABS data.As at 30 June 2011, the median age for Indigenous women in prison was 31.6 years, compared with 36.3 years for non-Indigenous women; the mean ages were 32.4 years and 37.9 years.By way of comparison, the median ages for Indigenous and non-Indigenous men were 30.4 years and 34.8 years respectively, with a mean of 32 years and 36.8 years. 16Again, this suggests the distribution of age for Indigenous women in prison mirrors that of their male counterparts, which may be obscured if the needs of Indigenous women are simply considered as against their non-Indigenous counterparts.
The median ages for female prisoners by Indigenous status and jurisdiction are set out in Table 1, indicating that Indigenous women were younger in all jurisdictions, with the contrast particularly marked in the smaller jurisdictions of Tasmania (29.5 years vs 37.8 years) and the ACT (27.5 years vs 37.5 years).The gap between the two groups was smallest in the Northern Territory (33.5 years vs 35 years).
The foregoing section presented Indigenous women's imprisonment rate for the June 2011 quarter.The figures from June 2011 17 enable a breakdown by age; the overall figures at this time were 358 per 100,000 for Indigenous women and 16 for non-Indigenous women.As can be seen in Table 2, the imprisonment rate for Indigenous women peaked at 30-34, at 673 per 100,000, 26 times the comparable non-Indigenous population.The disparity between the two groups was greatest, however, for women aged 18 years, where Indigenous women faced an imprisonment rate 40 times that of their non-Indigenous counterparts, although it should be recognised that this is in the context of very small numbers.The fact that non-Indigenous women were more likely to be imprisoned in the older age groups (>39) may be a reflection of Indigenous women's significantly reduced life expectancy. 18There does not appear to be any research which has specifically explored the different age distribution of Indigenous women in prison, but it is possible that other factors may include the younger age profile of the Indigenous population generally, the role of alcohol and possibly younger ages for childbearing, all of which may tend to shift the offending profile downwards.Table 2 also sets out information on the age distribution of the prison population by Indigenous status, indicating that 41 percent of Indigenous female prisoners were aged between 25 years and 34 years, whereas this group accounted for only 35 percent of non-Indigenous women.By contrast, only one percent of the Indigenous female prisoner population was aged 55 years and over (n=6), compared with seven percent of the non-Indigenous female population.At the other end of the spectrum, very young women (18-19 years) accounted for four percent of the Indigenous population, compared with 1.3 percent for their non-Indigenous counterparts.These findings may have implications for correctional planning, for example, as to women's health, training and parenting needs. 17Ibid. 18The life expectancy for Australian Indigenous women, at 72.9 years, is almost 10 years lower than for non-Indigenous women, at 82.

E Sentence length
Recent research from Western Australia suggests that, compared with non-Indigenous women, Indigenous women are in fact less likely to be sentenced to a term of imprisonment when appearing before the court for comparable offending behaviour and histories, 19 but this does not suggest there are not differences in the sentences Indigenous and non-Indigenous women receive.As can be seen in Table 1, Indigenous women generally serve shorter sentences than their non-Indigenous counterparts.Nationally, the median sentence length was 18 months for Indigenous women, compared with 36 months for non-Indigenous women, which may suggest that Indigenous women are being imprisoned for more trivial offences. 20The median sentences for Indigenous women ranged from six months in the Northern Territory to 36 months in Victoria, while the ranges for non-Indigenous women were from 30 to 46 months.Interestingly, median sentences for Indigenous men were about halfway between the two groups of female offenders, at 25 months, with a range of 12-35 months.
The data on expected time to serve set out in Table 1 indicate that Indigenous women have a median time of 10 months, compared with 19 and 16 months respectively for non-Indigenous women and Indigenous men.This information is again of significance in the context of correctional planning and ensuring appropriate programs are available to address prisoners' criminogenic needs.However, the relatively shorter time periods to be served by Indigenous women may be outweighed by the fact that they are more likely to serve multiple sentences, as will be discussed below in the context of the rate of prior imprisonment, which again raises issues in terms of program availability, duration and completion.The fact that Indigenous women's sentences are so much shorter than those of non-Indigenous women (which are already much shorter than men's sentences) requires further analysis in the context of correctional planning.It may also raise issues in relation to the diversion of such women from custody.

F Offences committed
The ABS provides information on the types of offences committed by prisoners, as assessed by the most serious offence for which they were convicted. 21As has been noted previously, however, this sort of information is limited because it fails to contextualise the criminal behaviour, for example, the drug addiction which might fuel someone's property offending. 22This point obviously applies to all prisoners, regardless of gender or Indigenous status, but may mask differences on this basis.Notwithstanding this caveat, the data indicate a very different offending profile for Indigenous and non-Indigenous women.
The most common offence type for Indigenous women was acts intended to cause injury (AICI), which includes all assaults (33%, compared with 11% for non-Indigenous women).This confirms Behrendt, Cunneen and Liebesman's assertion that 'Indigenous women are more likely to be imprisoned for violence- 19  related offences than non-Indigenous women.' 23 They suggest this might be in response to domestic violence and other forms of abuse.The nexus between Indigenous women's exposure to violence and offending conduct is discussed further below, however it should also be noted that AICI also accounted for a much higher proportion of offences committed by Indigenous men, compared with non-Indigenous men (32% vs 15%).
The second most common offence for which Indigenous women were imprisoned was burglary offences (12% vs 6%), followed by offences against justice procedures, government security and operations, which include attempts to pervert the course of justice (10% vs 8%).Earlier literature indicated Indigenous people were particularly likely to be imprisoned for public order offences, 24 and this appears to be borne out by the most recent data, which shows such offences account for a higher proportion of offending by Indigenous women than non-Indigenous women (1.4% vs 0.5%), although offences of this nature remain at low levels overall.Parker, Kilroy and Hirst noted recently that '[a] significant factor in the incarceration of indigenous women is fine defaulting … These women are incarcerated because the burden of paying fines is very difficult for them.' 25 As will be seen in the following section, one of the key characteristics of Indigenous women in prison is their uncertain financial circumstances.
Non-Indigenous women are most likely to be convicted of drug offences (24% vs 2%), followed by fraud offences (13% vs 3%).These numbers suggest a very different offending profile for the two cohorts, which may not be accounted for in the correctional programs currently available.As discussed further below, there are few correctional programs specifically designed for Indigenous women in Australian prison systems, and the programs that are available may therefore fail to recognise the differences in offending patterns.Further research is therefore required to promote the development of correctional programs that reflect the true picture of Indigenous and non-Indigenous offenders and their respective offending patterns.In doing so, consideration should also be given to Baldry's analysis of desistance theory in this context, specifically the assumptions that Indigenous women are 'offending in the conventional understanding of that term [and] can choose to stop it.' 26Indigenous women are also less likely than non-Indigenous women to be imprisoned for homicide (9% vs 13%), although other research 27 has found that Indigenous women are 14 times more likely than non-Indigenous women to 23  commit homicide (5.3 vs 0.4 per 100,000).This latter figure is somewhat anomalous; one explanation may be because the number of women committing homicide is very small.Accordingly, while Indigenous women may be more likely to commit homicide, this still results in only a very small number of women being held in custody for homicide, with little proportional representation in the prison population.Even though non-Indigenous women are less likely as a proportion of the overall population to commit homicide, presumably there are still more individual non-Indigenous female homicide offenders than Indigenous female homicide offenders, in terms of actual numbers of individuals.As homicide offenders tend to get long sentences, they tend to remain in the population from one census to the next, so become over-represented in prisoner counts.

G Recidivism rates
In its 2002 report, the ATSISJC noted that a 'significant factor among the Indigenous female prisoner population is the high rate of recidivism.' 28 The report highlighted the need for investigation of the circumstances that place Indigenous women at risk of repeated imprisonment and noted that: Statistics in relation to previous offending are a useful indication of a prison's achievements in rehabilitating offenders and these figures suggest a need to focus on the women prisoners' offending and background with a view to effective interventions.Development and support of effective programs for Indigenous women is clearly a priority to reduce rates of re-offending.

29
Lawrie's survey of female Indigenous prisoners in NSW revealed that 98 percent had at least one previous adult conviction, while 26 percent had 15 or more prior convictions. 30n research on people released from prison and homelessness, Baldry et al 31 found that 68 percent of Indigenous women were back in prison nine months after release, compared with 36 percent of Indigenous men and 40 percent of all interviewees.
The most recent ABS data on the proportion of female prisoners who have had prior adult imprisonment 32 are set out in Table 1 and indicate that two-thirds of Indigenous female prisoners have had prior experience of imprisonment.Although it is acknowledged the figures in Tasmania (44%) and the ACT (100%) are too small to be of particular significance, the proportion of women who had previously been imprisoned ranged from 50 to 72 percent.By way of comparison, the figures for non-Indigenous women ranged from 30 percent in the Northern Territory to 67 percnet in the ACT, with a national average of 36 percent.Notably, although the figures for all female prisoners have decreased since 1999, these fell from 44 percent to 36 percent for non-Indigenous women but, to a much more limited extent, from 69 percent to 67 percent, for Indigenous 28 ATSISJC 2002, above n 4, 140. 29 women. 33ccordingly, research should examine the factors that may have contributed to the reduced rate of non-Indigenous women returning to prison, and seek to adapt any promising initiatives (for example, correctional programs) for the Indigenous population.
By way of comparison, three-quarters of Indigenous men had had prior experience of prison; while the figures for Indigenous women are not as poor as this, they much more closely mirror these patterns than the patterns of non-Indigenous women.This factor should also be taken into account in developing appropriate correctional programs and measures which seek to reduce recidivism.

III CHARACTERISTICS OF FEMALE INDIGENOUS PRISONERS
The preceding discussion presented a quantitative representation of Indigenous women in Australian prisons.This information, while important, fails to recognise the lived experiences of these women, and the extent to which they differ from both Indigenous men and non-Indigenous women in prison.In the 2002 ATSISJC report, it was noted that '[s]tudies of indigenous women in prison reveal experiences of life in a society fraught with danger and violence.' 34Some of the key characteristics and experiences of Indigenous female prisoners are discussed in this section, in an attempt to provide a fuller picture of the Indigenous women who populate our prisons and to better understand the reasons why they find themselves there and how to engage sensitively and effectively with them while they are incarcerated.
The Australian Medical Association noted in its 2006 report card on Indigenous people in prison that it 'is a normal part of life for many of these women to rotate through prisons and to have illicit drug problems, little material or social capital and to be at risk of dying from a drug overdose once released from jail.' 35 This claim is supported by a study of Indigenous police detainees which found that they were 'younger, had less education, were more likely to be caring for children and to be living in public housing, and were less likely to be employed than non-Indigenous women'. 36It is not suggested that all of these issues are equally relevant to all Indigenous women, nor that they do not also apply to the non-Indigenous prison population, but Indigenous women appear to demonstrate more extreme markers of disadvantage and have specific needs relevant to the effective service delivery in the correctional system. 37

A Mental health issues
It is beyond dispute that Indigenous women in custody are over-represented among prisoners at risk, although there are limited comprehensive data from which to draw conclusions about the scope, prevalence and burden of mental health problems among Indigenous people, especially for vulnerable groups, including prisoners and juveniles in detention. 38The available data suggest that the majority of Indigenous women in prison have serious psychiatric issues, 39 and are more than 11 times more likely to experience severe psychosocial distress than the general population.By way of comparison, the figures for non-Indigenous women were eight times the general population, while male prisoners were four times and three times respectively. 40 its 2009 Overcoming Indigenous Disadvantage report, 41 the Productivity Commission referred to research indicating that rates of hospital admissions for mental disorders were three times as high for Indigenous female prisoners as in the Indigenous population of Western Australia generally.One might infer that hospitalisation rates for the general Indigenous population are already likely to be higher than the non-Indigenous population.Earlier data cited in the 2002 ATSISJC report 42 indicated that 22 percent of Indigenous women had self-harmed in custody, compared with 13 percent of non-Indigenous women, while Blagg et al's interviews with female prisoners in Victoria revealed that the women: saw mental health as being the single biggest issue they faced in prison, and saw mental health as inextricably linked with other issues such as family violence, sexual abuse and addiction. Thy raised concerns about assessment processes, drug treatment and continuity in services.43 What emerges clearly from this information is that there is an urgent need for appropriate mental health treatment in correctional centres which is not only targeted to the specific needs of female prisoners, but takes into account the cultural needs of Indigenous women.This is particularly vital, given the data presented above that around two-thirds of Indigenous women in prison have already served time.Addressing their mental health needs is a critical step in breaking the cycle of recidivism.However, the evidence above about the short duration of Indigenous women's sentences also needs to be recognised in terms of planning for and delivery of appropriate mental health treatment.38

B Physical health issues
Indigenous women in custody also experience physical health issues to a greater extent than their non-Indigenous or male counterparts, with the Australian Medical Association reporting that 95 percent of Indigenous women in prison report at least one chronic condition, such as hypertension, diabetes and renal disease, compared with 78 percent of Indigenous men. 44The 2007 National Prison Entrants' Bloodborne Virus and Risk Behaviour Survey data found that 72 percent of Indigenous female prison entrants screened positive to Hepatitis C, compared with 33 percent of non-Indigenous entrants.For Hepatitis B, 39 percent of Indigenous female entrants tested positive, compared with 22 percent for non-Indigenous women. 45Finally, in a retrospective cohort study of adults imprisoned in NSW between 1988 and 2002, Indigenous women were 12.6 times more likely than the general NSW population to die after release from custody. 46y way of comparison, Indigenous men were 4.8 times more likely to do so. 47ain, these figures raise concerns in terms of the experience of Indigenous women in prison and their access to health treatment which responds appropriately to the prisoners' specific health needs.There is some evidence of this occurring, for example, all women's prisons in NSW reportedly run Indigenous vascular health clinics, 48 however such initiatives would appear to be unusual.

C Substance abuse issues
Drug and alcohol abuse issues have long been identified as a significant factor in Indigenous offending. 49Indigenous women in prison are much more likely than non-Indigenous women to have alcohol dependency issues (54% vs 17%), and also have a higher rate of drug dependency (57% vs 48%); overall, only 20 percent of the Indigenous women in the Drug Use Careers of Offenders study did not report any drug or alcohol dependency, compared with 36 percent of non-Indigenous women. 50ndigenous women were also more likely to causally attribute their offending to alcohol and drugs equally (24% and 21% respectively), while non-Indigenous women attributed their offending primarily to drug use (33%) rather than alcohol (2%).51 The findings from the Drug Use Monitoring in Australia (DUMA) program, which seeks to measure drug use among those people who have been recently apprehended by police, are also instructive in this context.In particular, Loxley and Adams found that while Indigenous female (police) detainees were more likely than non-Indigenous women to have used alcohol and/or cannabis in the previous 12 months, they were less likely to have used other illicit drugs.52 Indigenous female detainees appeared more likely to consume alcohol and displayed higher levels of violent offending than non-Indigenous women.Significantly, alcohol dependence did not appear to be associated with property offending in Indigenous women, although illicit drug dependence was.These findings should be considered in light of the data above about the extent to which violent offences and burglary dominate Indigenous women's offending patterns.

D Parenting responsibilities
According to Behrendt, Cunneen and Liebesman, 53 80 percent of Indigenous female prisoners are mothers, but it has been suggested that they do not appear readily able to access Mothers and Children's Units, which are generally available to female prisoners in most Australian jurisdictions. 54A needs analysis with Indigenous women with dependent children leaving prison in NSW has indicated that the women were generally not consulted in relation to their future plans or required changes did not eventuate. 55ogram planning in this context should also acknowledge the extended kinship ties of many Indigenous communities (for example, allocating adequate time when it comes prison visits), especially in light of the finding of Lawrie's survey of NSW Indigenous female prisoners that 29 percent of respondents had primary care responsibilities for children other than their own; the same proportion were normally responsible for the care of other people, principally their parents and other family members. 56 noted previously, 57 more research is required to better understand the needs of Indigenous women with infants and young children in prison and the appropriateness and ease of access to programs which enable such prisoners to keep their children with them, as well as their needs upon release.The Aboriginal Women with Dependent Children Leaving Prison Program in NSW 58 may provide a model for further consideration.This program is currently being evaluated by Eileen Baldry, who is also conducting research in relation to the social and cultural resilience and emotional wellbeing of Aboriginal mothers in prison. 59

E Financial circumstances and homelessness
In her study of Indigenous women in prison in NSW, Lawrie found that 43 percent of respondents with dependent children had not been in receipt of any income from paid employment or Centrelink (eg, parenting payment) at the time of their last offence.Lawrie suggested that this left 'a huge gap for Aboriginal women, especially those trying to support a family or provide care for extended family members, and places additional pressure on an already difficult situation.' 60 This finding clearly has implications for correctional agencies, and programs which promote financial literacy may be of particular relevance; however, such programs should be adapted to be culturally sensitive and appropriate, for example, by recognising the prevalence of humbugging in some Indigenous communities and the impact this may have on financial planning.
In addition, Baldry et al's 61 study of prison releases in NSW and Victoria indicated that most of the Indigenous women in the study were unable to secure public housing on their release from custody to their financial circumstances and prior poor relations with the respective housing authorities.Almost all of the women had moved more than once in each of the three month periods between follow-up interviews, with most moving a number of times.Baldry has also suggested that Indigenous women are the group of prisoners least likely to find appropriate housing and support services after release, especially if they have dependent children, and that they return to prison faster and at a higher rate than other prisoners. 62These findings clearly point to the need for post-release support and better interaction with housing departments to ensure a smooth transition to reliable housing for Indigenous women released from custody.

F Exposure to family violence
The data above indicate that Indigenous women were three times more likely than non-Indigenous women to have assault recorded as the most serious offence for which they were imprisoned.Marchetti has noted that there was 'little, if any, discussion of the prevalence of family violence within Indigenous communities in the official RCIADIC reports.' 63 Arguably, the extent of Indigenous women's exposure to family violence now appears to have been acknowledged (see for example the emphasis in the National Plan to Reduce Violence Against Women and their Children 64 ).Even if we are, lamentably, no closer to reducing its incidence, recognising this is crucial, given the links between Indigenous women's exposure to such violence and their offending patterns and incarceration. 65In this context, a report by the NSW Aboriginal Justice Advisory Council found that: at least 80 percent of the women surveyed said that their experience of abuse was 60 Ibid 27. 61  an indirect cause of their offending.Some women revealed that the underlying cause of their drug and criminal habits was to avoid dealing with, or because they had not been able to address, the abuse that they had suffered as a child, in particular child sexual assault. 66ere is also evidence of a very high level of victimisation among Indigenous female prisoners, with the majority having been subjected to physical or sexual abuse. 67Lawrie's survey of Indigenous women prisoners in NSW found that 70 percent of respondents had been subjected to physical and sexual abuse as children; 78 percent reported being physically assaulted and 44 percent sexually assaulted as adults. 68Shockingly, the NSWLRC referred to reports indicating abuse rates of between 90 percent and 100 percent among female Indigenous prisoners. 69ere appears to be a perception that Indigenous women were becoming more likely to retaliate against on-going family violence than previously, which has been considered in the literature. 70The ATSISJC has referred to '[a]necdotal evidence suggest[ing] increased arrest for violence is the result of Indigenous women who behave violently to protect or defend themselves, because they know that they would not receive police protection.' 71 The findings of a study by Stubbs and Tolmie are instructive in this context. 72They examined cases between 1991 and 2007 where Indigenous women killed their abusive partner and found that the battering the women had experienced and their disadvantaged circumstances were generally read as indicators of personal deficits and any evidence of structural disadvantage was muted.Stubbs and Tolmie argued that the large number of Aboriginal women serving sentences in Australia for killing violent men in part may reflect a disjunction between their stories and dominant representations of battered women. 73They also suggested that Aboriginal women in some Australian communities may have fewer reservations than other women about responding to physical force with force.Whether or not this perception is accurate, it is clear that Indigenous women's exposure to and trauma from violence is a critical aspect in their offending and requires closer consideration by correctional agencies.For example, the Walking Together program developed by NSW Corrective Services provides Indigenous women perpetrators of family violence with strategies for reporting violence against children, as well as sessions on alcohol consumption and its relation to offending behaviour.Over 50 women graduated from the program between October 2005 and February 2008. 74rograms of this nature should be independently evaluated and, if effective, made more widely available.

IV CONCLUSION
In September 2011, Luke Grant, the Assistant Commission of NSW Corrective Services described the rapidly increasing overrepresentation of Indigenous women in prison as 'one of the worst statistics you could possibly have.' 75 Norrington has observed that 'Intersectional groups, or "minorities within minorities", struggle to have their voices heard.' 77 While this article does not purport to represent the voice of this doubly marginalised population, 78 it goes some way to filling the research gap in this area.In particular, this article has presented and analysed the most recent publicly available Australian data on Indigenous women in custody, including information on imprisonment rates and numbers of Indigenous women in prison across Australia, the offences for which they are imprisoned, the length of sentences imposed and recidivism rates.Some specific characteristics of Indigenous female prisoners, especially in the context of their physical and mental health, parenting responsibilities, financial circumstances and exposure to violence, were also reviewed.In addition, the practical implications of these data, especially for correctional agencies were considered.Clearly, the extent to which Indigenous women are overrepresented in our prisons, and the level of disadvantage they have commonly experienced, indicates a critically high level of need.However, it is equally apparent that most prisons are designed around the needs of non-Indigenous male prisoners; 79 Indigenous women, representing a minority within a minority, appear to be something of an afterthought in correctional planning and management.
In their recent review of correctional programs in Australia, Heseltine, Sarre and Day commented that there had been 'limited progress made in the development of programs for special needs groups, including female and Indigenous offenders.'Interestingly, although the authors went on to say that there was still 'much work to be done in developing effective rehabilitation programs for both Indigenous prisoners and female prisoners,' 81 the authors did not consider the specific needs of prisoners who were both Indigenous and female, with none of the programs they reported on including both measures.As a result, the report continued a trend previously remarked upon that Indigenous women tend to 'disappear' from criminal justice data, being represented either in the statistics on women or in the statistics on Indigenous people, with little attention being paid to the intersection of these two groups. 82is is not to say that there are not examples of programs and policies which address the specific needs of this subgroup across Australia, some of which were detailed in this article, 83 but it remains the case that much more is required.As Stubbs noted recently, 'an intersectional analysis that recognises the specific circumstances that contribute to Aboriginal women's criminalisation and incarceration, coupled with an approach to the provision of services and support that focuses on substantive equality is crucial.' 84 Ten years on from the ATSISJC report, we urgently need to move to a position where Indigenous women in prison no longer have to suffer from a 'rights and policy structure which identifies groups of needs and rights holders such as women and Indigenous people, but fails to provide for the needs of people who dwell at the intersection of these groups.' 8581 Ibid 41. 82 For discussion, see Bartels, n 74; Marchetti (2008), n 63; Norrington, above n 77. 83See also Bartels

Table 2 : Adult female imprisonment rate (per 100,000) and distribution of female prisoners (%), by age and Indigenous status
Behrendt, Cunneen and Liebesman, above n 15, 16.See also Holly Johnson, 'Drugs and Crime: A Study of Incarcerated Female Offenders' (Research and Public Policy Series No 63, AIC, 2004); Wendy Loxley and Kerryn Adams, 'Women, Drug Use and Crime: Findings from the Drug Use Monitoring in Australia Program', (Research and Public Policy Series No 99, AIC, 2009). 24See Lucy Snowball and Don Weatherburn, 'Indigenous Over-representation in Prison: The Role of Offender Characteristics' (Crime and Justice Bulletin No 99, NSW Bureau of Crime Statistics and Research, 2006) for discussion.
25Judy Parker, Debbie Kilroy and Jonathan Hirst, 'Women, Health and Prisons in Australia' in Diane Hatton and Anastasia Fisher (eds), Women Prisoners and Health Justice: Perspectives, Issues and Advocacy For An International Hidden Population (Radcliffe Publishing, 2009) 45.26Baldry, above n 8, 260. 27Jack Dearden and Warwick Jones, 'Homicide in Australia: 2006-07 National Homicide Monitoring Program Annual Report' (Monitoring Report No 1, AIC, 2008).
Steering Committee for the Review of Government Service Provision, Overcoming Indigenous Disadvantage: Key indicators 2009 (Productivity Commission, 2009). 39See Behrendt, Cunneen and Liebesman, above n 15.See also Tony Butler et al, 'Mental Health Status of Aboriginal and Non-Aboriginal Australian Prisoners' (2007) 41 Australian and New Zealand Journal of Psychiatry 429. 40Sharan Kraemer, Natalie Gately and Jenny Kessell, HoPE (Health of Prisoner Evaluation) Pilot Study of Prisoner Physical Health and Psychological Wellbeing (Edith Cowan University, 2009). 41Steering Committee for the Review of Government Service Provision, above n 38.
42ATSISJC 2002, above n 4. 43 Harry Blagg et al, Systemic Racism as a Factor in the Overrepresentation of Aboriginal People in the Victorian Criminal Justice System, Report to the Equal Opportunity Commission and Aboriginal Justice Forum (2005) 48.(ed), Women and the Law (LexisNexis Butterworths, 2011) 389 for discussion.
Social and Legal Studies 155.64SeeDepartment of Families, Housing, Community Services and Indigenous Affairs, Indigenous Women and the National Plan to Reduce Violence Against Women and Their Children <http://www.fahcsia.gov.au/our-responsibilities/women/programs-services/reducingviolence/the-national-plan-to-reduce-violence-against-women-and-their-children> at 23 July 2012. 65See Bartels, above n 1, for discussion.
As the data in this article demonstrate, with the exception of recent trends in prisoner numbers in NSW, things have generally worsened since the ATSI Social Justice Commissioner observed in 2002 that: the intersection between race, gender and class is of particular relevance to Indigenous women … [but] the discrimination faced by Indigenous women is more than a combination of race, gender and class.It includes dispossession, cultural oppression, disrespect of spiritual beliefs, economic disempowerment … and more.76 80 75 Luke Grant, cited in Kate Lumley and Stephen Norrish, 'Exchanging Ideas about Aboriginal Contact with the Criminal Justice System' (2011) 23(10) Judicial Officers' Bulletin 83, 85. 76 ATSISJC 2002, n 4, 155. 77Caroline Norrington, 'Intersectionality: Indigenous Women' in Patricia Easteal (ed), Women and the Law (LexisNexis Butterworths, 2011) 228, 228. 78Indeed, the limitations of non-Indigenous researchers writing in relation to Indigenous issues are acknowledged: for a recent discussion, see Allan Ardill, 'Non-Indigenous Lawyers Writing About Indigenous People: Colonisation in Practice' (2012) 37(2) Alternative Law Journal 107. 79Bartels and Gaffney, above n 14. 80 Karen Heseltine, Andrew Day and Rick Sarre, 'Prison-based Correctional Rehabilitation Programs: The 2009 National Picture in Australia' (Research and Public Policy Series No 112, AIC, 2011) 40.
, above n 74; Bartels and Gaffney, above n 14.See also Dot Goulding, 'Breaking the Cycle: Addressing Cultural Difference in Rehabilitation Programmes' in Rosemary Sheehan, Gill McIvor and Chris Trotter (eds), Working with Women: Offenders in the Community (Willan, 2011) 173 for a recent example of a best practice program run in Bandyup Women's Prison in Western Australia designed not only for but also by Indigenous women. 84Stubbs, above n 2, 59. 85 ATSISJC 2002, above n 4.