Contents

• Abstract
• Introduction
• Canada
• United States of America
• New Zealand
• Australia
• Legislative frameworks
• Cultural competence
• Implementing culturally competent policy
• Client/agency relationships
• Indigenous community control
• Family preservation versus child protection
• Family group conferencing and other similar models
• Early intervention
• Service delivery
• Conclusion
• References

ABSTRACT

Many of the problems associated with child abuse and neglect in communities are directly related to experiences of colonisation. Indigenous peoples in different countries and communities may have experienced colonial processes differently, but have experienced some common impacts. This paper provides a review of legislation and services delivered to Indigenous communities in Canada, the United States and New Zealand. While some of the issues and ideas may be useful and relevant in the Australian context, a key finding in the research is that a 'one size fits all approach' does not work. Particular attention is paid to new approaches to family and community wellbeing that focus work on community strengths and healing, and provide a less problem-focused picture of Indigenous communities.

There has been much interest in overseas legislative reform that transfers real authority for children and families' wellbeing to Indigenous communities. This is widely considered to be important to the long-term empowerment of Indigenous peoples and a basis for the development of more effective support systems. The United States Indian Child Welfare Act currently represents the highest level of transfer of decision-making authority to Indigenous peoples.

The paper describes a range of service models that focus on Indigenous collaboration, community development, community participation and community control. Although local solutions will need to be found for different Indigenous communities, there is a preference across Indigenous communities for holistic, multifaceted approaches that heal all sections of the community and address the underlying causes of health and social problems. Some key policy and practice recommendations for the development of better child protection and child welfare/family support systems are described.

• It is contended that statutory child protection services be based on an understanding of communal identity and a 'whole-of-community' rather than individually-focused response. Further, they require a more collaborative, community-based 'grass-roots' approach. This should be based in part on the development of comprehensive neighbourhood-based supports and services, which draw on family networks and other informal resources.
• Good partnerships and meaningful collaboration between government and Indigenous organisations are vital to the development of effective child protection and broader child welfare strategies. Such partnerships must be founded upon inclusiveness andempowerment strategies. They will usually require government agencies to relinquish some power and to recognise the authority of the Indigenous community or organization. Ideally, this collaboration will be supported via appropriate legislation.
• Culturally competent service delivery requires not only the employment of Indigenous staff in support services, but the incorporation of cultural knowledge into the service delivery framework via community engagement and active participation.
• Community satisfaction with service delivery will be increased by a holistic approach to healing, autonomy and flexibility in service provision, and the capacity to respond or tailor services to meet local needs. Family preservation is a particularly important part of Indigenous family support systems. The flexible family and home-based nature of family preservation services renders it suitable for use with a variety of cultural traditions.

INTRODUCTION

In Clearinghouse Issues Paper 19, Stanley, Tomison and Pocock (2003) provided an overview of the issue of child abuse as it affects Australian Indigenous communities. Stanley and colleagues identified the failure of current responses to child abuse in Indigenous communities and the urgent need to identify more effective ways of addressing this problem. It noted the relationship between traumatic colonial experiences and ongoing intergenerational trauma experienced by many Indigenous communities. It was also noted that if the situation is to improve, Indigenous people need to be resourced to provide future directions.

While Indigenous peoples in different countries and communities have experienced colonial processes differently, many communities have experienced some common impacts. Frequently occurring problems that have been documented in Indigenous communities in Australia and other countries include: intergenerational traumas, such as the effects of child removal; social dislocation; community dislocation; consequent or related mental health problems; marginalisation from social services for health, housing, education and policing as well as from family and child welfare services; and more generally, the loss of power and community cohesion associated with colonial experiences. Common manifestations of these problems within communities include: alcohol and substance abuse, high levels of family violence and violence generally, economic deprivation, and related impacts on children's wellbeing. These issues are discussed in the Australian context in Issues Paper 19 (Stanley et al. 2003).

This paper provides an overview of legislation and service delivery programs in Canada, the United States and New Zealand, which attempt to meet Indigenous children's needs. Much of the legislation and many of the programs and processes are responses to these commonly experienced problems. New approaches to family and community wellbeing, which focus work on community strengths and healing, provide a less problem-focused picture of Indigenous communities.

However, work in the area of Indigenous children's wellbeing has been saddled with the legacy of this problem-focused characterisation of Indigenous communities. It is embedded in much of the work in this field, and is often reproduced in Indigenous-run programmes, through funding, reporting and other requirements, which may lend themselves to program-focused rather than holistic approaches to children's wellbeing.

While local solutions will need to be found for different Indigenous communities, it was noted by Stanley and colleagues that there is a preference across Indigenous communities for a holistic approach, which heals all sections of the community and addresses the underlying causes of health and social problems. Australian and international reviews and reports into the delivery of child welfare services to Indigenous communities have found that conventional individualistic responses to children's wellbeing do not substantially improve conditions for Indigenous communities and families and that a more holistic and community based response is required.

The research underpinning this paper was designed to respond specifically to the growing interest in overseas models. The intention is to provide information and ideas for consideration in the Australian context. It does not consider Australian models which are reviewed elsewhere (for example, Urbis Keys Young 2001). The research was undertaken in response to calls from many Indigenous organisations for consideration of international models for the delivery of children's services to their communities. For example, many of the submissions from Indigenous organisations made to the National Inquiry into the Separation of Aboriginal and Torres Strait Islander Children from their Families (Human Rights and Equal Opportunity Commission [HREOC] 1997).

Consideration is given to models that focus on Indigenous collaboration, community development, community participation and community control. There has been particular interest in legislative models which provide greater control to Indigenous communities such as those implemented in parts of Canada and in the United States. For example, the Secretariat of National Aboriginal and Torres Strait Islander Child Care Organisations (SNAICC) and Link-Up NSW called for legislative models such as those found in the United States and Canada.

The research was carried out through extensive database searches and by writing to the providers of Indigenous children's services in New Zealand, Canada and the United States between 1999 and 2000. As noted by other researchers, there is extremely limited information available on Indigenous children's services, particularly information prepared by and for Indigenous organizations (Carr and Peters 1997; McKenzie 1997). Wherever possible such information has been reviewed.

LEGISLATIVE FRAMEWORKS

Arange of legislative models for the delivery of child welfare services to Indigenous communities have developed in a number of countries within different historical and political contexts. However, in each country there has been a resurgence of Indigenous political demands for greater self-determination and control over family life, particularly from the 1970s onwards. Legislative reform which transfers real authority over children and families to Indigenous communities is widely considered to be important to the long-term empowerment of Indigenous peoples in the area of children's wellbeing. The legislative models reviewed range from the transfer of legislative, judicial and administrative functions to Indigenous communities to those which retain all of these functions within the mainstream child welfare system.

In this section of the paper a range of legislative models for the delivery of child welfare services to Indigenous communities are considered. The legislative models considered range from:

• complete autonomy with the recognition of Indigenous jurisdiction over legislative, judicial and administrative matters pertaining to Indigenous children;
• shared jurisdiction with the transfer of some functions to Indigenous communities;
• delegated authority with jurisdiction over child protection matters retained by the state but delegation of some child protection functions to Indigenous communities; and
• mainstream legislation which integrates Indigenous input into existing structures.

Before considering current legislative models in each country a brief historical overview of state responses to Indigenous families is provided.

Canada

Canada, like Australia, is a federation with responsibility for child welfare residing with the provinces. Some First Nations organisations have called for national umbrella legislation that would provide a framework for the delivery of all child welfare services to First Nations communities. To date no such legislation has been passed. Canadian Indigenous people refer to themselves as First Nations people. This definition arose in recognition of the sovereign nature of Indigenous groups: each nation occupies a specific geographical area and has distinct traditions and decision-making structures (Community Panel 1992: 5-6).

History of Indigenous Child and Family Services

As with Aboriginal people in Australia, there is a long history of the forcible removal of Canadian Aboriginal children from their families and communities. In Canada, children were placed in 'residential schools', modelled after British industrial schools. The Canadian government established the first of these in 1883; schools were also operated by 25 different religious orders (Morrisette 1994). Conditions at the schools were appalling. By the late 1960s, the resulting poor academic achievement of resident children led to decisions to close most schools. The residential school system is now widely regarded as reflecting an assimilationist policy, which was intended to break down Canadian Aboriginal culture, and integrate Aboriginal people into white society. This policy dominated First Nations child welfare in Canada until the closure of the residential schools in the 1960s.

On the closure of the schools, a policy of integrating Aboriginal and non-Aboriginal services was introduced. This led to a greatly increased incidence of removal of Aboriginal children. This pattern was repeated in other parts of Canada (Johnston 1983). Aboriginal leaders have characterised this phase as a period of cultural genocide. Integration of services persisted. By the 1970s, placement and adoption had replaced residential schools as the primary alternative care system for First Nation children. Children were removed without consideration of cultural difference, according to the ethnocentric assumptions of social workers regarding matters of perceived health, housing, diet etc. These children were more culturally isolated than those earlier sent to residential schools, due to the absence of their peers in the placement. This isolation engendered a greater degree of assimilation than under the previous overtly assimilationist policy (Armitage 1993).

Restoring First Nation control

After two provincial tripartite reviews of child welfare in 1978 and 1980, agreements with First Nation bands were increasingly made. By the 1980s, broader policy began to favour some degree of community control under tripartite agreements between federal government, provincial governments, and Indian bands or tribal councils (Armitage 1993).

However, the implementation of this policy has been hampered by funding requirements which have resulted in constraints in the number and type of agreements made. While provincial government child welfare systems remain in control of legislation, a number of First Nation communities have developed proposals and negotiated agreements transferring service delivery. First Nation communities are continuing to develop culturally appropriate service models of their own. In most jurisdictions, a number of First Nations have established child and family services that provide a range of services mandated under the provincial legislation. In some instances this encompasses the full range of services including protective responsibility, while in other instances services are provided in conjunction with departmental authorities.

There have been calls for a national Act similar to the United States Indian Child Welfare Act (1978), in order to provide an umbrella under which individual First Nation agencies could operate while allowing for regional cultural differences (Durst 1998). On the one hand, Canadian Indigenous people are faced with the likely impracticality of 600 Bands each having their own act, but on the other, they are faced with the difficulty in finding consensus on a single federal act.

In the early 1980s, Manitoba First Nation people developed the first Canadian Indian child welfare agreements with federal and provincial governments, providing for province-wide First Nation control under the one model. Tripartite agreements were negotiated between the band, provincial government, and federal government. All rural and reserve areas were covered by these agreements. However the Aboriginal Justice Inquiry Report (1991), which examined the relationship between the Aboriginal peoples of Manitoba and the justice system, found that Aboriginal peoples living off reserves were not being served well by the non Aboriginal system. Its recommendations included the establishment of a province-wide Metis agency and the expansion of existing First Nation agencies authority to enable them to serve band members who are living off reserves.

In 2000 the Province, First Nations and Metis leaders signed an agreement to work together to restructure the child and family services in Manitoba, so that it recognises Metis and First Nations people's authority and responsibility to care for all their children. This agreement is in the process of being implemented. In November 2003 The Child and Family Services Authority Act was proclaimed. This legislation transfers the responsibility for delegating the delivery of chid and family services, including child protection services, from the Province to four new child and family service authorities. Three of these are Aboriginal. Under the new Act all Aboriginal children in Manitoba can be served by an Aboriginal agency. A detailed plan for the transfer of cases to the most appropriate authority, and the broad implementation of the new system, has been developed and is being implemented.

The most appropriate authority for clients will be determined by applying the following values: children, families and communities belong together; decisions will be in the best interests of children; and service arrangements should be culturally appropriate, stable and timely.

It is anticipated that the transfer of cases will be complete by the end of 2004. The new authorities are operating under the Manitoba Child and Family Services Act while the next stage of the process, the drafting of new legislation, takes place. Guiding principles under the Manitoba Child and Family Services Act include: acknowledgement of the special needs of Aboriginal people in respect of culture, geography and past experience; the importance of the preservation of Aboriginal cultural identity; and the provision of services must involve Aboriginal people and recognise their priorities (Armitage 1993).

The Manitoba Child and Family Services Act 1984 (Principle 11) recognises that 'Indian bands are entitled to the provision of child and family services in a manner which respects their unique status as aboriginal peoples'. This includes provision of child protection services by Indian bands. Where child protection matters are brought to court by a non-Aboriginal agency, the Aboriginal agency which serves the child's community must be given notice of the proceedings (Manitoba Child and Family Services Act 1984, s 30(1) (e)). This is in addition to the parents and other relevant parties being notified of such proceedings. Arrangements under the Child and Family Services Authorities Act provide for First Nation control over service delivery, with Provincial responsibility for funding, legislation and standards. The Province will provide funding to the four authorities, which will design and manage the delivery of services and fund agencies to provide the services. The authorities will work in partnership with the Province to design laws, policies and standards for service delivery.

In British Columbia, First Nation involvement in child welfare has evolved in an ad hoc manner. In the absence of treaties or compensation for dispossession (which have existed in other provinces), First Nation people have taken a more militant approach to asserting their rights. The 1980 Spallumcheen band child welfare by-law, authorising the band to conduct its own child welfare program, was recognised by the provincial Social Services Department only after a live-in protest at the home of the then Minister (Armitage 1993; Wharf 1992). While negotiation of most agreements continues at the local and informal level, most children on reserves are covered by agreements, which grant effective control to Indian people.

The Child, Family and Community Services Act provides that an Aboriginal child's cultural identity must be considered when determining the best interests of an Aboriginal child. If a child is to be removed from a family their community must be notified of the hearing and an Aboriginal child placement principle applies. The Aboriginal Child and Family Services Branch, which is part of the Ministry of Children and Family Development, helps to build capacity in communities and encourages formal agreements with communities to provide services. The agreements vary with some communities' delegated full responsibility for all child protection services and others with authority to provide some services such as voluntary care services, voluntary care agreements, recruitment and development of foster homes.

The Child and Family Services Act (Ontario; s1 (2)-(5)) also recognises that Aboriginal people should be entitled to provide, wherever possible, their own child and family services, in a culturally appropriate manner. However, this and all other purposes of the Act are subject to the paramount purpose of promoting the best interests, protections and wellbeing of children (s1 1(1)). An Aboriginal child placement principle applies where a child is placed in out of home care. The Minister may enter into agreements with communities, which can then appoint agencies with full or partial delegated powers under the Act. There are five designated First Nations agencies in Ontario; Tikinagan, Payukotayano, Weechi-it-te-win, Abinoojii Family Services and Dilico. These agencies are required to consult regularly with their band or Aboriginal communities with respect to matters affecting children, including maters relating to placement of children, provision of family support services, preparation of care plans, temporary care and special needs agreements, amongst other matters.

Under the Children and Family Services Act (CFSA) the Mi'knaw Family and Children's Service is mandated to provide the full range of child protection services to all First Nations people living in First Nations communities in Nova Scotia. The Mi'knaw agency was also established through a tripartite agreement between the Government of Canada, the Department of Community Services (Nova Scotia) and the First Nations communities through the 13 chiefs of the Nova Scotia bands.

The Mi'knaw agency also provides support and consultative advice with respect to First Nations people not living in First Nations communities. The CFSA provides that the Mi'knaw agency will be notified like other parties to proceedings where the child who is the subject of the proceedings is Indian. The Mi'knaw agency with its consent may be substituted for the agency which commenced the proceedings (Children and Family Services Act (Nova Scotia) s36 (3)). The Act also provides that the Mi'knaw agency must be involved with the voluntary placement or adoption of Aboriginal children (s68 (11-12).

The Yukon Territory, where Indian people make up approximately 20 per cent of the population, provides an example of the difficulties experienced by small and remote groups in gaining control over child welfare. Although the Yukon Children's Act was modified in 1984 to allow delegation of child welfare authority to Aboriginal groups, only one agreement has been made, with the Champagne/Aishihik band. More than half of the Yukon's 15 First nations have selfgovernment agreements which enable them to pass their own child welfare laws. However, as at 2000 no such laws had been passed.

The Child and Family Service Act Saskatchewan enables the Minister to enter into agreements for the provision of children's services including child protection to First Nations families living on reserves (s 61(6)). As at 2000 there were 17 agencies established. The agencies are also involved in First Nation children's cases where the child resides off a reserve. The Act provides for First Nation bands to be notified with respect to placement hearings, for the appearance of bands in court hearings, and for their involvement in cases from the point of initial contact with the Department (s25, (37)). The Federation of Saskatchewan Indian Nations and the Provincial and Federal Government are engaged in self-government discussions, with child and family services one of the program areas under consideration.

While the Child Welfare Act (1999) Alberta does not include specific provisions for agreements with First Nations children's agencies, the Minister does have the power to delegate duties and powers (s87(1)). The Minister has exercised this power to establish 13 First Nations agencies. The Act also includes provisions with respect to consultation with a child's band and provision of culturally appropriate services (ss 62 and 73).

The Northwest Territories and Nunavat both have legislation which allows for extensive delegation of authority and responsibility for child welfare to Aboriginal corporations under community agreements. As at 2000 no community agreements had been reached.

The legislation in Newfoundland and Labrador, Prince Edward Island and New Brunswick does not contain any statutory provisions specific to First Nations, however protocols and agreements require culturally appropriate delivery of services.

While Provincial legislation with respect to First Nations children differs across Canada, there is a trend towards tripartite negotiated agreements with First Nations and Aboriginal peoples. These recognise the specificity of Indigenous peoples children's needs and the benefits of local control over children's services and decision making. In many instances in legislation, but otherwise in practice, the importance of including Aboriginal agencies in all aspects of decision making with respect to Aboriginal children is recognised. The implementation of Aboriginal control over children's services and decision making is however hampered by financial and other resource restraints and in some instances by the ad hoc implementation of reforms. However, the trend is to rectify these problems, with more comprehensive agreements which include Aboriginal peoples at all stages of development and implementation, as seen in Manitoba.

United States of America

While the United States is a federation, national legislation, the Indian Child Welfare Act 1978 (ICWA), regulates welfare with respect to Native American children. The ICWA is often referred to as a model for consideration by Indigenous peoples in other countries. The legislation transfers legislative, administrative and judicial decision making to Indian bands where children domicile on a reserve.

History of child and family services

Between 1850 and 1960, Native American children were forcibly removed from their families and communities and sent to boarding schools in what is now regarded as a policy of assimilation and cultural genocide. All types of abuse and neglect were widely inflicted on children resident at the schools.

History of tribal jurisdiction

However, as early as the 1820s, limited tribal jurisdiction was recognised by the United States Supreme Court in a trilogy of 18th Century United States cases (Johnson v McIntosh Supreme Court of United States, 1823, 21 US (8 Wheat) 543, 1823; Cherokee Nation 30 U.S. (5 PET) 1 (1831); Worcester 31 US 16 PET 5115 (1832) -hereafter Johnson v McIntosh, Cherokee Nation and Worcester).

In the United States, the recognition of aboriginal title was closely associated with the recognition of Indian Nations and Indian treaty rights. Cherokee Nation and Worcester were both cases contesting the right of the State of Georgia to make laws which undermined the Cherokee Nation's laws and which contravened the Hopewell and Holston treaties. The limited jurisdiction recognised in these early cases has been affirmed in legislation.

In the case of Fisher v District Court of Rosebud County [424 U.S. 383 (1976)], the Supreme Court of the United States affirmed tribal authority in child placement cases where all parties were members of the tribe and resided on the tribe's reservation.

The origins of the Indian Child Welfare Act (ICWA) are found in cases such as Fisher. In this case core principles found within the ICWA, including recognition of exclusive tribal jurisdiction in child welfare matters, where all parties are members of the tribe and reside on a reserve, are recognised. The dual interests of individual parties and the tribal group embodied in the ICWA were also recognised in Fisher. However, competing and sometimes contradictory principles developed as cases were brought before state courts testing tribal court's jurisdiction, for example, in cases where one or both parties lived off the reserve. This conflicting case law, together with hugely disproportionate removals of Indigenous children from their families and their placement in non Indigenous environments, lead to the passing of the Indian Child Welfare Act 1978 (US).

Indian Child Welfare Act

A United States Congress Commission into American Indian policy was established in the mid 1970s. This Commission's terms of reference were broad ranging and included child welfare. It was estimated that 25-35 per cent of all Indian children were raised at some point by non-Indian families or institutions. The Commission found that Indian children suffer severely, particularly at adolescence, in terms of identity and life crises, when removed from their culture and brought up in a non Indian environment. The hearing recognised that cultural survival of Indian tribes depended on retention and teaching of culture to Indian children. The ICWA was passed with the dual objective of protecting the best interests of Indian children and to promote the stability and security of Indian tribes, communities and families (Indian Child Welfare Act of 1978, 25 U.S.C s 1902).

The ICWA is premised on recognition of limited tribal sovereignty and the collective interest of tribes in children. It is essentially a jurisdictional statute directing that Tribal courts have authority over Indian child welfare where the child is residing or domiciled on a reservation (Id.at 1911 (a)). State and Tribal courts have shared jurisdiction over Indian children who are not residing on a reserve (about half of all Indian children). Proceedings in a State Court must be transferred to a Tribal Court, unless there is good cause not to transfer the proceedings, if a transfer is requested by a parent or the tribe (id. at 1911 (b)). Either parent can veto the transfer of proceedings (id. at 1911 (a) ). The tribe, Indian custodian and parents, all have full standing in matters involving Indian children in State courts. The ICWA also provides directions to State Courts where they hear Indian child welfare matters.

A party seeking foster care placement or termination of parental rights of an Indian child in a State Court has to demonstrate to the Court that they have made positive efforts to provide assistance to prevent the breakdown of the relationship which led to the action being taken. The parent or Indian custodian of a child against whom involuntary proceedings are being brought has a right to a court appointed lawyer. The Court also has the discretion to appoint a separate representative for the child.

When adopting or fostering Indian children, State Courts must follow a preferred order of placement which is similar to the Aboriginal and Torres Strait Islander Child Placement Principle. The descending order of preference to be followed is: with a member of the child's extended family; with other members of the child's tribe; with another Indian family; and if the above three options are not possible, with a non-Indian family.

An Indian child may be removed, under state law, for a limited period of time for emergency placement to prevent imminent physical harm.

United States 'intergovernmental child welfare agreements' (in the context of the ICWA) are usually based on the 'Model Tribal-State Indian Child Welfare Agreement' drafted by the American Indian Law Centre in 1986. Agreements are made between a tribe/tribes and a state department. Tribes sometimes also develop their own Tribal Resolutions. Intergovernmental agreements are usually modelled on the ICWA itself, and include sections following the ICWA structure in areas including confidentiality, authority, jurisdiction, child protection service practice, and fiscal structure. Native American child welfare is delivered through a number of agencies including nongovernment organisations, Tribal agencies, and State and Federal Agencies.

New Zealand

New Zealand has national legislation which governs both child welfare and juvenile justice. The central feature of the New Zealand legislation, for all children, is family group conferencing. The New Zealand family group conferencing model has been adapted and used in many parts of the world.

History of Indigenous child and family services

While New Zealand had a policy of assimilation between 1847 and 1960, this policy did not include a program of forced removals of Maori children from their families. Up until this period, Maori children's welfare needs were usually left to the whanua (Maori extended family). Extensive contact with the child welfare system occurred from the 1960s onwards, when mainstream child welfare legislation was applied without regard to Maori culture and community values.

Post 1960s, there was a large movement of Maori to urban centres. The Maori extended family, which had played a significant role in Maori children's wellbeing, was not recognised and supported by the official child welfare system. Rather, non-Indigenous child protection services began what was to be an unsuccessful, and ever increasing, involvement with Maori children. In 1981, 49.2 per cent of all children in need of care were Maori children. In 1991 Maori constituted 13 per cent of the population and they would have constituted a smaller percentage of the population in 1981.

In the 1980s and 1990s, Maori staff of the Department of Social Services, and other activist groups, accused the Children and Young Persons Act 1974 (NZ) and the Department of being racist. Criticisms included making assessments of children and their needs without due consideration of the context of the whanau; the imbalance in staffing of social services by non-Indigenous staff rather than in a manner proportionate to the representation of client groups who have contact with the Department; and the control of power and resources by Non-indigenous decision makers. The Department of Social Welfare commissioned a Committee to inquire and report on planning and delivery of services to Maori communities. In light of this report draft legislation to amend the Children and Young Person Act 1974, which had not taken on board Maori child welfare needs, was rejected and the Children, Young Persons and Their Families Act 1989 was developed.

Children, Young Persons and Their Families Act 1989

The Children, Young Persons, and Their Families Act 1989 (New Zealand) addresses child protection and juvenile justice in a single piece of legislation with the objective of focusing on the wellbeing of children and young persons in the context of their families, whanau (kin group), hapu (extended kin group with many whanau), iwi (descent group with many hapu) and family groups.

Section 5 of the Act outlines principles to be applied in the exercise of powers under the Act. These principles include participation of family, whanau, hapu, and iwi in decisions affecting the child wherever possible. Wherever possible family relations should be maintained and strengthened. Consideration must always be given to how decisions will effect both the child and the stability of the child's family, whanau, hapu, iwi and family group.

Section 13 outlines principles applying to children in need of care and protection. These principles affirm that intervention in a child or young person's family life should be the minimum necessary to ensure their safety and protection. Wherever possible assistance should be provided to the family to facilitate a child remaining within their family. A child or young person should only be removed from their family if there is a serious risk of harm to the child or young person.

Where a child must be placed in out of home care, wherever practicable, the child should be placed with a member of the child's or young person's hapu or iwi (preferably with hapu members) or, if this is not possible, with a person who has the same tribal, racial, ethnic, or cultural background as the child or young person and who lives in the same locality as the child or young person.

This Act attempts to give effect to principles of Maori self-determination by prioritising and emphasising the involvement of Maori family with decisions which affect Maori children. However, this approach to child protection requires sufficient resourcing to address underlying problems. This has been a key criticism of the implementation and effectiveness of the Act.

Further, the Act privatises child protection issues by strongly advocating their resolution within the confines of a family conference rather than a forum subject to public scrutiny. The principles of family maintenance and the best interest of the child are subject to each other without legislative guidance as to how to resolve the potential conflict of interest between these principles. This aspect of the Act has been criticised by commentators and within cases brought before the New Zealand High Court and Court of Appeal. Judge Walsh held that the principles in sections 5 and 13 of the Act are subject to the 'over-riding arch' of the 'welfare and interests of the child' as defined in s 6. (The Matter of an Application about the L Children FC Wanganui, CYPF 1/95, 7 April 1998 at 10).

The potential for family group conferences to exacerbate the vulnerability of weaker family members has been noted in the Mason Review (Mason Report 1992) and in the Social Policy Agency Study (Gilling, Patterson and Walker 1995). It was noted in the Mason Report (1992: 12), which reviewed the Children, Young Persons and Family Act 1989, that: 'The idea of bringing the wider Whanau and other players under the umbrella of the Act has increased the number of competing interests, and in our view has rendered the child or young person increasingly vulnerable.' The Review recommended that: 'The Act be amended to make it clear that any Court or person who exercises any power conferred by the Act shall at all times treat the interest of the child or young person as the first and paramount consideration' (Mason Report 1992:12).

This raises similar issues to those which arise under the Indian Child Welfare Act 1978 (United States) in terms of the dual interest of the community or group and its maintenance and the individual rights of parents or the child. However, the additional factor of privatising the decision making under the New Zealand legislation undermines the protection provided by judicial decision making under the ICWA. Where legislation implements processes of self-determination, it is important that the structural framework, and preferably legislative framework, within which new forms of decision making for vulnerable people are implemented, are clear, well resourced and well defined.

It has been noted that the reality of family empowerment depends on resources and support. The following problems with family groups conferencing were identified by the Social Policy Agency Study (Gilling, Patterson and Walker 1995):

• inadequate information for families about the type of situations which give rise to care or protection concerns and the family group conferencing process;
• the need to wait for the family group conferencing process to commence before receiving help;
• difficulties regarding the process for inviting participants to the conference;
• inadequate management of relationships between participants at family group conferences;
• undue influence of officials and some family members in the decision making process;
• failure to ensure decisions meet the needs of the child and address the underlying issues;
• resourcing of family group decisions;
• unequal participation of attendees; and
• lack of effective monitoring of implementation and failure to address non-implementation.

These criticisms give rise to concern about the efficacy of the Act in terms of its dual goals of strengthening families and protecting children. However, it is not possible to evaluate the success of this method compared with traditional methods of child protection, without comparative information with respect to these issues under the previous legislation. As noted above, the previous child protection system was criticised as a form of institutionalised racism (Rine 1998).

Australia

In Australia, each state or territory has statutory responsibility for the provision of child welfare services. Peak Indigenous (and non-Indigenous) groups have called for national legislation to provide a framework for the provision of child welfare services to Indigenous communities for many years. Indigenous people's experience of child welfare services is closely associated with experiences of forced separations of children from their families and colonial policies of dispossession. In the following sections, the state of New South Wales is used to highlight child welfare issues for Indigenous peoples.

Background

The HREOC National Inquiry into the Separation of Aboriginal and Torres Strait Islander Children from their Families (1997) brought legislation and policies of forced and unjustified separations of Aboriginal and Torres Strait Islander children into the public arena. The report of the Inquiry, 'Bringing them home', drew attention to the importance of understanding contemporary separations in the context of colonial policy towards Indigenous Australians, with particular reference to the impact of the removal of children from their families.

National Inquiry findings with respect to current welfare issues

The National Inquiry found that Indigenous children continue to be over-represented in their contact with child welfare agencies across the country. They found that there are many reasons for ongoing separations and these include the intergenerational effects of past removals, poor socioeconomic conditions in communities, and systemic racism.

These factors combine to produce cultural difference between welfare departments and communities and the myriad of social conditions which underlie removals.

Indigenous communities across Australia gave evidence to the HREOC Inquiry of their need for resources to implement programs to ensure the wellbeing of their children. Not a single submission to the Inquiry from Indigenous organisations saw interventions from welfare departments as an effective way of dealing with Indigenous child protection needs. Fear and distrust of welfare agencies continue. Evidence confirmed that Indigenous families perceive contact with welfare departments as threatening the removal of their children. In evidence to the Inquiry all state and territory governments stressed the need for Indigenous communities to exercise control over their children. They all claimed to support policies of self management or self-determination with respect to Indigenous child welfare needs. The Inquiry concluded:

Departmental attempts to provide culturally appropriate welfare services to Indigenous communities have not overcome the weight of Indigenous peoples historical experience of 'The Welfare' or the attitudes and structures entrenched in welfare departments.... For many Indigenous communities the welfare of children is inextricably tied to the wellbeing of the community and its control of its destiny.

If welfare services are to address Indigenous children's needs they need to be completely overhauled.... Ultimately child welfare appropriate to each community and region should be negotiated with those whose children, families and communities are the subjects of the system. Negotiation clearly implies empowerment of Indigenous parties and recognition of their true partnership in the reform process (HREOC 1997: 458-459).

Recommendations with respect to child welfare legislation

The HREOC National Inquiry made recommendations which recognised the need to address the underlying social and economic causes of child abuse and neglect, and the need to address the underlying colonial practices which continue to impact on Indigenous families. The Inquiry addressed specific child welfare and juvenile justice legislative reform in the context of principles of self-determination.

It recommended a two-tiered approach in order to address longer term aspirations for self determination, and the immediate need to establish minimum standards for services to Indigenous children whether delivered by Indigenous or non-Indigenous providers.

• Recommendation 43a proposes a negotiation process between governments and Indigenous organisations to establish a new legislative framework. This recommendation recognises the need for the involvement of Indigenous peoples in the process of creating a new framework for Indigenous children's wellbeing as well as in the outcome.
• Recommendation 43b (2) proposes negotiation at the community level of agreements for children and families wellbeing. This recommendation recognises that a one size fits all approach is not appropriate ie that different communities will have different needs, aspirations and requirements.
• Recommendation 44 proposes that legislation set minimum standards for child welfare and juvenile justice legislation.
• Recommendation 50 proposes that a child's Indigenous status must be recognised and considered in any children's court matter and that the child must be separately represented.
• Recommendation 51d proposes that where placement is with a non-Indigenous carer, principles of family reunion, continuing contact with the child's family and community, and the proximity of the carer to the child's family and community, should guide choice of a carer.

These recommendations have not been adopted in any jurisdiction. The changes to New South Wales legislation after the National Inquiry are discussed below.

Review of two NSW Acts

In 1994, a review of the Children (Care and Protection) Act 1987 was initiated by the New South Wales Premier and Minister for Community Services. The Legislative Review identified ways in which a new legislative structure could improve the provision of child protection services to the community generally. A number of recommendations were made specifically with respect to Aboriginal and Torres Strait Islander children.

Recommendation 6.3 from the Legislative Review Committee proposed that the Act should give the Minister for Community Services the power to delegate certain functions to Aboriginal and Torres Strait Islander people to enable a greater degree of self determination in the provision of services to Aboriginal and Torres Strait Islander children.

The Review report noted Aboriginal and Torres Strait Islander people called for greater emphasis on prevention and support programs and that there was community support for including the concept of self-determination in legislation. They also expressed the view that greater Aboriginal and Torres Strait Islander control over child protection would ensure more culturally appropriate and effective child protection. However their recommendations with respect to these matters were ambivalent.

While the National Inquiry (HREOC 1997) recommended a process of negotiation of potentially all child support and protection functions, including legislative, judicial and administrative responsibilities, with Aboriginal communities, the New South Wales Review recommendations fall short of transferring any decision making powers where statutory intervention may take place. Its recommendations also fall short of the transfer of child protection functions that have occurred under the United States, New Zealand and most mainstream Canadian child welfare statutes.

The functions which the Legislative Review Committee envisaged the Minister should delegate to accredited Aboriginal and Torres groups, included preventative and support work, organisation and provision of out of home care, and where negotiated, joint departmental and community policy and program development.

The NSW Children and Young Person's Care and Protection Act 1998 implemented a 'self determination' provision which falls short of the National Inquiry and Legislative Review Committee's recommendations. Section 11 provides:

Aboriginal and Torres Strait Islander Self-determination

• It is a principle to be applied in the administration of this Act that Aboriginal and Torres Strait Islander people are to participate in the care and protection of their children and young persons with as much self-determination as is possible.
• To assist in the implementation of the principle in subsection (1), the Minister may negotiate and agree with Aboriginal and Torres Strait Islander people to the implementation of programs and strategies that promote selfdetermination.

This open-ended section fails to provide recognition of Aboriginal and Torres Strait Islander peoples' rights as a group to control their children, or the specificity and safeguards as to how, and by whom, resources and programs should be implemented. No definition or criteria are provided for understanding self-determination. The meaning of this section is left to the discretion of the Minister. The legislation only enables the Minister to outsource programs and discuss strategies with Aboriginal and Torres Strait Islander communities. It does, however, provide a positive obligation to consult and facilitates participation.

The Legislative Review Committee made a similar recommendation to the National Inquiry with respect to consultation with relevant Aboriginal and/or Torres Strait Islander people and organisations about all significant decisions affecting Indigenous children. They recommended that a requirement to consult about all relevant decisions, not just decisions about placement, be included in the legislation. Section 12 of the Children and Young Persons (Care and Protection) Act 1998 provides that:

Aboriginal and Torres Strait Islander families, kinship groups, representative organisations and communities, be given the opportunity, by means approved by the Minister, to participate in decisions made concerning the placement of their children and young persons and in other significant decisions made under this Act that concern their children and young persons.

The Act does not provide a method for identifying or accrediting 'representative' or 'appropriate' Aboriginal and/or Torres Strait Islander organisations. This provision, like the self-determination provision, places Aboriginal families' capacity to have input into decisions or control over their children at the discretion of mainstream political and bureaucratic processes. This is problematic because it is these very processes which have been incapable of addressing Indigenous child welfare needs to date.

The Review Committee recommended, and the recommendation was adopted in the Children and Young Person (Care and Protection) Act 1998, that a child be placed with:

• A member of the child's or young person's extended family or kinship group, as recognised by the Aboriginal and Torres Strait Islander community to which the child or young person belongs; or
• If it is not practicable or would not be in the best interests of the child or young person to be so placed, then they may be placed with a member of the Aboriginal and Torres Strait Islander community to which the child or young person belongs.

This order of placement can be displaced by a court if it is in the best interests of the child to do so, where the child is to remain with an Indigenous carer. However, the test from the 1987 legislation applies if the placement is to be with a non-Aboriginal family. That is, the order of placement can only be displaced if it is not practicable, or the placement would be detrimental to the child.

Where a child is placed with a non-Aboriginal carer, the Director-General must consult with the child's extended family or kinship group and Aboriginal and Torres Strait Islander welfare organisations as are appropriate to the child (s 13 (d) ). In accordance with section 13 (6), if a child or young person is to be placed with a non-Aboriginal carer, the principle guiding their placement - subject to the best interests of the child, and if the child is old enough his or her own wishes - is that a fundamental objective must be reunion with the child's family or community and continuing contact with the child's culture and community.

In November 2001, the Children and Young Persons (Care and Protection) Amendment (Permanency Planning) Bill 2001(2) was passed by both houses of Parliament. This shifts the focus of the Children and Young Persons (Care and Protection) Act from ongoing assistance to families who are vulnerable and assessed as not being able to provide adequate care, to placement of children permanently where the court assesses no 'realistic possibility' of reunion. While many children in the care and protection system experience an enormous degree of instability, the permanent removal of children from their families is not necessarily the most appropriate solution for Aboriginal and Torres Strait Islander families. A consistent message from reports and consultations with Aboriginal and Torres Strait Islander communities has been the need for a holistic community and family response to the needs of children, whether in terms of child protection or in a broader context.

An individualistic approach, which looks at Indigenous children's needs without proper consideration of the context of their parent/s' or communities' circumstances, has been criticised by Indigenous groups in Canada, New Zealand and Australia. This does not imply that children's and parents' interests will always coincide, but acknowledges the complexity of factors that impact on a child's longer-term needs. Section 78A of the amended Act provides for permanency planning. Section 78A (3) provides that a permanency plan for an Aboriginal or Torres Strait Islander child must comply with the Aboriginal and Torres Strait Islander Child and Young Person Placement Principle in section 13 (discussed above). Section 78A (4) provides limitations on when an Aboriginal or Torres Strait Islander child can be permanently placed with a non-Aboriginal or non-Torres Strait Islander person or persons.

The importance of Indigenous communities control over their children and families has been acknowledged in the United States, Canada, New Zealand and Australia. However governments have found it difficult to relinquish the power and dedicate the resources which are necessary to translate this recognition into legislative terms. The United States Indian Child Welfare Act represents the highest level of transfer of decision making authority to Indigenous peoples. Thousands of child welfare matters are adjudicated each year in the United States in the various Tribal Courts. In New Zealand the inclusion of Maori people in the primary decision making process, the family group conference, is mandated in legislation.

In Canada decision making with respect to First Nations children is frequently transferred to First Nations organisations through the negotiation of tripartite agreements. However, unlike Tribal groups in the United States, these organisations operate within the parameters of Provincial legislation and decisions are adjudicated in Provincial Courts. In New South Wales, as in other parts of Australia, legislation recognises the importance of Indigenous children remaining in their communities and families through an Indigenous child placement principle. While Australian Indigenous people may have input into some decisions, all powers are retained by mainstream child welfare departments. Regardless of the legislative structure regulating child protection, effective provision for children's wellbeing also requires adequate resourcing and appropriate delivery of services.

SERVICE DELIVERY

For effective collaboration between government departments and Indigenous communities it is necessary for departments, and individuals who work within them, to have a meaningful understanding of the history and experiences which impact on the communities to be serviced. This requires personal and institutional reflection on values inherent in attitudes and presumptions within individual behaviour and within service delivery models. A key issue identified for consideration when working with Indigenous communities, is an understanding of communal identity and a related whole-of-community rather than individually-focused responses to child protection.

Cultural competence

Cross-cultural communication problems and cultural difference militate against collaborative planning, responsibility and accountability. Cultural competence has been defined as 'a set of congruent behaviours, attitudes and policies that come together in a system, agency, or among professionals that enable them to work effectively in cross-cultural situations' (Tong and Cross 1991:12).

This is of particular relevance when working with Indigenous communities. Striving for cultural competence in social services in the United States is now widespread, and occurs partly in recognition of the ethnocentric history and values of social welfare services. A culturally competent program 'appreciates and values diversity; understands the cultural forces which impact the program; understands the dynamics which result from cultural differences; institutionalises cultural knowledge; and adapts its services to fit the cultural context of the clients it serves' (1991:13).

However, empirical models for cultural competence are few, and those tailored for Indigenous people are fewer still. Social work policy and practice can discourage flexibility and innovation in approaches to cultural difference (Tong and Cross 1991). There is a great and largely unfulfilled need for practitioners, policy-makers and other professionals to be aware of the cultural specificity of policy and practice (Kalyanpur 1998). Many authors who discuss cultural competence emphasise the importance of practioners' ability to reflect on their own personal cultural backgrounds and possible biases. Effects of the cultural incompatibility of social service models, particularly those relating to child and family services, have been overwhelmingly negative.

Problems with conventional social work and child welfare methods

In the United States, a number of authors and reviews suggest that social work methods impose alien cultural values of individualism, materialism and empiricism on Native American people (Voss et al. 1999; Weaver 1998; Kalyanpur 1998; Awasis Agency 1997; Ricks et al. 1990). Voss et al. comment on the exclusion of traditional Native American ideas from social work literature and the common characterisation of Indian people as a 'problem' group. They suggest that social work practice 'rigidly reinforce a kind of clinical colonialism' (1999:233).

Weaver (1998) notes that the high value placed on independence in the dominant culture has led to conditions such as 'enmeshment' and 'co-dependency' being regarded as dysfunctional. However, such judgements are culturally relative, and can lead to misunderstanding and misdiagnosis. 'It is not unusual for non-Indian members of the formal child welfare system to misinterpret a parent's reliance upon extended family members for child care as a sign of neglect . . . [yet this behaviour represents] normal and healthy interdependence among Native Americans' (Ronnau et al. 1990: 91)

It is important not to make generalisations about Indigenous identity and selfhood. The great diversity within Indigenous groups always calls for practitioners to obtain specific knowledge about the community, nation or client group. This information is best obtained from the client. Indigenous social services and cultural agencies are further sources of information (Weaver 1998, 1999).

The conventional individually focused models applied by child and family service agencies and treatment services are often culturally inappropriate for use with Indigenous client groups due to differences in the nature of personal and communal identity. Individually focused treatment models often disregard the complexities of extended family networks in First Nations communities (Connors 1993).

Many authors and community consultations find that a 'whole-of-community' approach to child protection and other social service and treatment interventions is more appropriate and likely to lead to success (Manitoba Justice Inquiry 1991; Durst 1998, First Nations Child and Family Task Force 1993). For example, the Awasis Agency, a regionalised peak body for the Indigenouscontrolled child and family services of 18 northern Manitoba Aboriginal communities, integrates child protection with other services, observing that this inclusive approach mirrors the Aboriginal concept of self in that region (Awasis Agency 1997).

Professionals dealing with Indigenous families may be unaware of the potential effects of their 'cultural blindness'. Indigenous parents tend to be disempowered in relations with professionals, who must develop strategies to increase 'levels of participation' (Kalyanpur 1998). In a qualitative study of social work professionals working with on-reservation Native American mothers, Kalyanpur (1998) found that although the workers were acting according to best practice, their assumptions of the universal applicability of 'objective' theories was false. Kalyanpur found that although the parents in the study had perceived parenting deficits according to professional criteria, they were raising their children 'to become competent adults within their culture' and therefore possessed appropriate parenting skills.

Partnerships and collaboration

Good partnerships and meaningful collaboration between government and Indigenous organisations are vital to the development of effective child welfare strategies which empower Indigenous communities. However, to date, power has resided almost solely with the state, as outlined by the Awasis Agency (1997: 24-25): 'The power structures that underlie traditional approaches to social work practice often work against collaborative decision-making with families. Even when social workers try to share decision-making power with families the power and authority attached to the role of social worker erode this attempt.'

Collaboration is vital for 'both understanding the specific limitations and ineffectiveness of existing services and programs, and for identifying the changes necessary to create culturally appropriate solutions' (Aboriginal Family Healing Joint Steering Committee 1993: ii).

In describing a number of Native American child and family services entities considered exemplary, one report identifies collaboration as the key feature to their success. Several of these organisations had complex partnerships between various combinations of state agencies, tribal organisations, and non-governments organisations (American Humane Association 1997).

A project conducted by the American Humane Association (1997) examined sources of conflict and collaboration in areas of child welfare in which both tribes and government agencies have an interest. Project sites were located on five reservations covering seven tribes in three states - Arizona, North Dakota and Washington. The research found North Dakota was an exemplary case for positive tribal-state relations. Some of the qualities which contributed to this status were: the long history of tribes and government working together; mutual understanding of history and cultural context; recognition of the 'sovereignty nationhood' of tribes by government; provision of training on the means to obtain federal funds; and a collaborative approach.

The report identified that the qualities and capacity of individual people involved were a key factor in successful tribal-state relations. People consulted for the project (individually and as representatives) discussed perceived personal skills and qualities as important to good working relations between tribes and states. These were grouped and summarised, including as follows: communication skills; sensitivity to different values; cultural broker skills ('cultural brokers' have the ability to 'walk in two cultures' with comfort in the different roles required); teamwork skills; and comfort in ambiguity.

In 1991, the United States National Indian Child Welfare Association produced a paper providing strategies for the development of effective cross-cultural partnerships for child abuse prevention. They found two vital factors in successful strategies to be inclusiveness and empowerment. Involvement of and consultation with community members should take place throughout the project cycle, from design through to evaluation. Natural community support networks should be used and developed, while natural helpers and natural prevention networks should be engaged. This can be achieved through attending formal and informal community gatherings, or by sponsoring joint training or public awareness events with Indigenous organisations. The history of disempowerment and attendant feelings of helplessness must be overcome by harnessing community strengths and resources.

Strategies which have been used include using the influence of Indigenous leaders to disseminate information, and seeking information and advice from Indigenous organisations. Programs should be designed so that they are sustainably incorporated into the local Indigenous culture.

Factors contributing to culturally competent work

There are a number of key issues which have been identified as relevant to culturally competent work with Indigenous people. Weaver (1998) discusses a number of topics important for practitioners to be aware of when working with Native Americans. These are issues which appear to also have relevance in the Australian context:

• History - Weaver (1998) states that interventions addressing trauma are often best approached through a group method, as a) much trauma has been perpetrated on people as a group, and b) Native American identity is focussed on groups. Community healing projects are becoming more common. Validation of historical grief is important in assessment and healing.
• Citizenship - The lack of recognition of (a) Indian nations by the state, and (b) individual Native Americans by nations, leads to problems with identity and self-esteem.
• Cultural identity - A thorough cultural assessment is essential. For example, how much does a client identify with Native American culture, or with a blend of Indian culture, or a blend of Indian and non-Indian culture?
• Sovereignty - Practitioners need to be aware of ICWA and rights of Native American agencies and communities to provide care and intervention.

It is important that child and family service providers are able to integrate knowledge and reflection with practical skills (Weaver 1998).

Implementing culturally competent policy

Some features of culturally competent policy, and some of the related practices that have been implemented in Canada, are outlined in Table 1. This table is an edited version of a table included in Hart (1997: 14-15), which was derived from a review of 15 Family Violence Prevention projects planned and implemented by Indigenous people and funded by Health Canada.

Client/agency relationships

The legacy of historical removals

An understanding of the impacts of trauma resulting from a history of forced and unjustified removals of children and culturally inappropriate service provision is necessary to develop effective social services policy analysis and child welfare programs within Indigenous communities (McKenzie 1997).

However, the impacts of this history are seldom considered by non-Indigenous agencies. The Awasis Agency of Manitoba (1997: 24) points out that: 'Social work cases are not looked at within the larger context of social, economic, historical, political, and cultural realities. Blame rests with the individual . . . Child and Family Services within a First Nations context must adopt a contextual perspective for service delivery to be effective.'

An atmosphere of taboo and shame still exists around the history of maltreatment of Indigenous children in a number of countries. Yet by 'better understanding client cognitions and behaviours that stem from this experience, treatment plans can be designed to overcome problematic parenting patterns' (Morrisette 1994: 381-382).

Strategies such as culturally appropriate placement may not resolve underlying problems. Much evidence suggests that parents who themselves spent lengthy periods in adoptive placement or residential schools as children often have parenting or substance abuse problems which lead to the removal of their children, establishing an intergenerational pattern of trauma and removal (Morrisette 1994; Mannes 1993). Factors which contribute to a lack of parenting skills include: the absence of positive parental role models; destroyed transmission of parenting knowledge and behaviours; absence of experience of family life; and sexual abuse (Morrisette 1994; Horejsi et al. 1992).

Some program models aim to raise awareness and educate Indigenous people about how the effects of historical factors have contributed to their contemporary realities, experiences and circumstances. In so doing, these innovative models attempt to address root causes of child abuse and neglect in Indigenous communities.

Healing possibilities also exist in 'productive encounters with representatives of the responsible religious orders' who were involved with removals and residential schools.

Collaborative evaluation of programs

Conventional evaluation criteria and frameworks are 'severely tested' in the context of Aboriginal child welfare. Beliefs and values underlying conventional approaches are those of the mainstream, not Aboriginal culture. Different belief systems can mean differences in objectives, indicators, who does the evaluation and how the information is used. Evaluation should contribute to the goals of a project, not just measure outcomes. There is a need to render values underlying evaluation processes explicit as part of the process.

Standards

Culturally inappropriate standards used for determining a child's need for substitute care have been a major contributor to disproportionate rates of removal in Indigenous populations (First Nations Task Force 1993; Mannes 1993; Community Panel 1992). In many places, culturally inappropriate alternate care standards lead to the placement of Indigenous children with non-Indigenous carers (First Nations Task Force 1993; Community Panel 1992).

Expanding on this point, the report of a Manitoba community consultation notes: 'The standard and procedures followed by First Nations agencies for apprehensions, placements and adoptions are provincially defined. The standards relating to foster homes on reserves are viewed from the mainstream society perspective. Most First Nations homes are unable to meet these standards.... It is not always possible to find foster or adoption homes that will pass the provincial test in the communities' (First Nations Task Force 1993: 51).

In the United States, Native American child welfare programs have successfully developed culturally sensitive placement standards, but have had to battle with states for acceptance. Tribally-controlled kinship care placements with aunts and uncles or grandparents are often seen by the non-Native child welfare system as foster care settings, with tribal agencies struggling to assert the legitimacy of these placements (Mannes 1993).

Staffing and training issues

A factor inhibiting increased Indigenous control of child and family services, which appears likely to apply in most countries and areas including Australia, is an inadequate supply of Indigenous workers (Durst 1998; Armitage 1993).

Consultations with the British Columbia Aboriginal population found that culturally inappropriate standards for health care and social worker education have contributed to the 'gross underrepresentation' of Aboriginal people in these fields (Community Panel 1992). Contributors to another community consultation, in Manitoba, argued that academic qualifications were not the most important criterion for workers, as they believed that mainstream social work curricula don't meet the needs of First Nations people (First Nations Task Force 1993). As well as the lack of Indigenous workers, a lack of supervision and administrative support is another impediment to the development of First Nation agencies (Durst 1998).

Other reasons for the short supply of Indigenous child and family services workers include difficulties in educating Indigenous social workers, especially those from isolated areas, and problems with retention of qualified First Nations staff, with few ongoing career development opportunities existing for staff at First Nations agencies (First Nations Task Force 1993; Armitage 1993). A review of Native American child protection teams found that permanency should be a critical factor in the choice of team members - high membership turnover brings problems with training, confidentiality and cohesion (Carr and Peters 1997).

The under-representation of Indigenous staff in Indigenous child and family services needs to be addressed as a priority. The British Columbia consultation found that it has led to culturally inappropriate service delivery, and the devaluing of traditional Aboriginal healing practices (Community Panel 1992).

Where non-Indigenous workers must be employed, the importance of cross-cultural training is very important. The Manitoba consultation stated that 'cultural differences created chasms between non-First Nations workers and their clients' (First Nations Task Force 1993: 56). Stereotypical views might lead to the belief that these issues and differences might not be so relevant to Indigenous people living apparently enculturated lives in cities. However, the consultation also found that the 'same concerns were expressed in urban areas as well as in First Nations communities' (First Nations Task Force 1993: 56).

Indigenous community control

Around the world, child welfare systems and agencies are struggling to protect their reputations and carry out their responsibilities in an environment of ever-increasing reports of abuse and neglect. There is a growing consensus among professionals and the public that there is a need for fundamental change in how child protection services should be conceptualised and delivered, for mainstream as well as Indigenous populations.

In the United States the 'Executive Session on Child Protection' concluded that a more collaborative, community-based approach to child protection was required (Farrow et al. 1998). The Session proposed that rather than child protection service agencies bearing sole responsibility for protecting children, other agencies, parents and the public should jointly share responsibility in 'community partnerships for child protection'. States including Missouri, Michigan and Florida, are developing new laws, policy and practice in response to these ideas.

The Session envisaged the development of comprehensive neighbourhood-based supports and services, which draw on family networks and other informal resources. These networks are closer to and more trusted by families in need than traditional services. The Session saw the development of formal community boards responsible for child protection as a viable alternative.

Given the parallel histories of dispossession and wholesale removal of children from Indigenous peoples in a number of colonised countries, the issue of community control is particularly important for Indigenous people.

The process of change

A report based on a review of 15 Health Canada-funded Family Violence Prevention projects planned and implemented by Aboriginal people had this to say about Indigenous control of child welfare services: 'As ownership of family-related services has increasingly passed to Aboriginal control, it has become evident that simply staffing those services with Aboriginal people is only part of the answer. The services themselves need to be designed by Aboriginal people to make them work as a reflection of the host community and the belief system found there' (Hart 1997: 12).

A First Nations task force set up in order to investigate Manitoba First Nations child and family services agencies envisaged the integration of child protection and other family services. The task force's consultations with 15 Manitoba communities also led to the development of the idea of 'Local Child Care Committees', which would play a major role in developing and/or approving case plans, as well as being involved in placement and child and family services (First Nations Task Force 1993).

In 1993, an Ontario Aboriginal committee produced an Aboriginal family healing strategy, developed through a community consultation process involving 7000 Aboriginal people throughout the Province. The Strategy saw the empowerment of Aboriginal people as being a central component in the healing of individuals, families, communities, and Aboriginal Nations (Aboriginal Family Healing Joint Steering Committee 1993: iii). The strategy required Aboriginal community control and funding for design and implementation. This process depended on a provincial government commitment to devolving authority to Aboriginal communities.

This phased handover of authority proposed in the Ontario Strategy involved the establishment of a joint management committee, with provincial government and Aboriginal community members. In the first phase, programming continued under provincial Ministry mandates while beginning to share control over family healing programs. In the medium to long term, full control will be devolved to the Aboriginal community. In 1994, the Ontario government created the Aboriginal Healing and Wellness Strategy with the signing of 13 implementation agreements with Aboriginal organizations and chiefs of First Nations. These agreements were renewed for a further five years in 1999. The Aboriginal Health and Wellness Strategy operates a number of programmes including healing lodges and treatment centres which offer traditional approaches to the treatment of sexual assault, physical abuse, addictions and family dysfunction; shelters from violence for women and children; and a recent program which focuses on child development from the prenatal stage to six years of age.

The phasing aspect of the scheme was designed to accommodate differing levels of community readiness. This aspect of the scheme may be particularly relevant to Australian Indigenous child welfare, as the levels of social, physical, economic and political resources and infrastructure are likely to vary considerably between communities. A relative resource deficit is not necessarily a good reason to postpone a phased handover of responsibility for children's wellbeing to Aboriginal communities. An advantage offered by the phased handover concept is that it allows for some real change and development in delivery of Indigenous child and family services without, or prior to, legislative change.

Two case-studies

Many existing Indigenous-controlled child and family services appear to have a good record for improving child welfare outcomes in their communities. Below are two examples of successful Canadian services.

Decentralisation and community-based services

The view that Indigenous child and family service provision must get in touch with grass roots issues and circumstances by operating at the local level has been expressed repeatedly (Awasis Agency 1997; Weechi-it-te-win Family Services 1995; First Nations Task Force 1993; Ronnau et al. 1990).

The community requirement today is to design services from the bottom-up or from the community's perspective, which is grounded in a more complete understanding of its social reality. The challenge is to move from mandates which emphasise efficient delivery of services to mandates that focus on effective service outcomes. (Awasis Agency 1997: 106)

This service, the Awasis Agency of Manitoba, reports that devolution of child and family services authority to the local level may improve responsiveness, management, flexibility and integration of services, and increase local community support and voluntarism (Awasis Agency 1997). The agency criticises centralised services on grounds that they are: unresponsive to the needs of locals; alienating due to their inaccessibility and over-specialization; and undemocratic, in the absence of community control (Awasis Agency 1997).

First Nations community representatives responding to a 1993 Manitoba consultation were alienated, frustrated, and angry with the centralised nature of First Nations agencies. They said that there was little local involvement, and agencies demonstrated a heavy-handed approach. (First Nations Task Force 1993).

A separate issue related to service decentralisation concerns the setting in which service delivery occurs. A number of studies suggest that services should not only be locally-based, but, where possible, offered at the client's home. Several researchers found that the institutional or office environment is alienating to Native Americans (Norton and Manson 1997; Tafoya 1990; Ronnau et al. 1990); it is likely that this also applies to many other Indigenous people, particularly given the common legacy of traumatic past child welfare interventions.

Besides their alienating atmosphere, there are several other reasons why institutional settings may not be ideal for delivery of Indigenous social services. Norton and Manson, consultants to a domestic violence program at an urban Indian health centre, found that early attempts at providing counselling services to battered women at the clinic were unsuccessful due to problems with: inadequate child care; lack of transport; and poor rapport (Norton and Manson 1997).

Norton started experimenting with home-based counselling, which proved successful: 'Home visits involve additional time and effort, but relative to the alternative of under utilisation of office-based interventions, home visits significantly enhance care and the effectiveness of counselling' (Norton and Manson 1997:336). Norton noted that the home-based approach allowed for greater flexibility, and facilitated a trusting relationship.

An assessment of six placement prevention and reunification projects in Native American communities found that the two most successful projects were home-based. By making the first contact in the client's home, the client's value in the relationship is established. The home visits also help to overcome the perceived reluctance of Native Americans and Alaskan natives to seek help outside the extended family (Smollar and French 1990).

Accountability

A number of accountability-related issues arise in the international literature on Indigenous child welfare. Political or personal interference with, and influence over, Indigenous-controlled child and family services is a very serious issue, which compromises the probity and effectiveness of some Indigenous agencies, and leaves Indigenous women and children the greatest losers. Other issues associated with devolved authority include: the problem of determining specific responsibilities where divided authority creates multiple accountability; the capacity of local services to provide assured child protection; and confidentiality.

The most critical reference to political interference found in this review related to the report of a Manitoba Indigenous community consultation. The Manitoba report cited political interference by powerful community members as an impediment to the development of First Nation child and family services agencies (Durst 1998). Gray-Withers (1997) states that First Nation women's groups accused Chiefs of 'complicity and political self-serving interference'. The apparent prevalence of political interference in Canadian Indigenous child welfare matters is closely linked to the small size of many First Nations communities. Health or social workers and police are likely to know or be related to the victim or the perpetrator. The close proximity of these various parties involved in child protection matters is likely to engender bias (Dumont-Smith 1995).

In Canada, serious family violence problems occur in many First Nations communities (First Nations Task Force 1993; Giesbrecht 1992). Although women had a powerful place in traditional First Nations culture, men dominate today. The colonising culture is a factor which has impacted on the adoption by First Nations men of negative attitudes and behaviour, including chauvinism, abuse and control of power (Gray-Withers 1997; First Nations Task Force 1993). This abuse is often seen as a private family matter in Aboriginal communities. As a result, little intervention from relatives or others occurs. Support services are often unavailable, and Chiefs or council members are unlikely to be charged over domestic violence offences (Dumont-Smith 1995).

These problems can be exacerbated by the process of instituting self-government and First Nations' control of child and family services. Native women's groups have been vocal in their criticism of self-government where it entails the further domination of First Nations men over the lives of women and children (Gray-Withers 1997: 86). Gray-Withers also contends that this genderbased power imbalance undermines child protection: 'In many communities, the male-dominated Native leadership has hidden and perpetuated problems of child abuse.... A process of empowerment for women and their communities will need to occur to allow for true community development and the acceptance of responsibility for current problems' (1997: 89). Women tend to favour regional control of child welfare, in the hope that Chiefs would have less influence over child welfare outcomes in the absence of local control (Gray-Withers 1997; Durst et al. 1995).

When authority for child and family services is handed over to Indigenous agencies, accountability becomes more complicated. Armitage (1993:169) highlights the coordination problems which often ensue between organisations and jurisdictions:

The establishment of independent First Nation family and child welfare organisations has the effect of dividing authority between mainstream provincial agencies and independent First Nation organisations. The result is diminished accountability in the child welfare system as a whole. At a practical level single accountability for the welfare of children and advocacy for them as individuals is lost because of the fragmentation of authority.

It is important to stress that this 'diminished accountability' is not a specific result of the involvement of Indigenous organisations, but simply a result of adding to the number of stakeholder organisations. First Nations agencies may be accountable to provincial and federal governments, as well as to their people.

The last accountability issue to be dealt with is confidentiality. One Native American evaluation report points out that attempts by child welfare workers to keep information classified were not always successful. Confidentiality is difficult to maintain in small communities (Carr and Peters 1997: 98).

Suggested resolutions

It is important to recognise that current mainstream child welfare systems are also likely to have unresolved accountability gaps or problems. The complex accountability maze which Indigenous agencies are presented with under partial or interim authority arrangements would be far simpler under full Indigenous control. Facilitating Indigenous communities to design programs and policies in line with their own needs may prove a more effective way of ensuring accountability (Gray-Withers 1997).

The establishment of regional peak agencies may result in disputes between these bodies and their constituent community groups, however a regionalised model does appear to offer better accountability than a fully localised one (Gray-Withers 1997; McKenzie et al. 1995). WRCFS is a good example of a regionalised service. WRCFS has a regional abuse unit which initially investigates notifications, and assists local workers who then take responsibility for follow-up services and case management. McKenzie (1997: 106) states: 'This model is quite effective in assuring required expertise in investigations, while protecting local community staff from some of the conflicts that can occur around initial abuse referrals in small communities.'

The establishment of regional agencies is one possible response to some of the accountability issues facing Indigenous child and family services. Other suggested strategies and initiatives include:

• a system of accountability to an authority outside the community political leadership (Gray- Withers 1997; McKenzie 1997);
• agency adoption of a political interference/conflict of interest protocol which involves sanctions for non-compliance (First Nations Task Force 1993);
• the creation of suitable fora for disputes and grievances to ensure fair and just process (First Nations Task Force 1993);
• the establishment of inter-community child protection teams where members from each community in a given area 'could help protect abused children caught in a political battle within a tribe' (Carr and Peters 1997:9 9);
• the creation of a national Indigenous child welfare commission for investigation of complaints (Durst 1998);
• the substitution of state or provincial legislation with comprehensive federal legislation, in order to simplify the accountability maze (Durst 1998); and
• formal confidentiality agreements signed by child protection team members (Carr and Peters 1997).

Traditional healing and cultural revival

Much of the literature on Indigenous child welfare from Canada and the United States describes or advocates the use of traditional healing methods in child welfare cases. A number of authors and reports emphasise that for many Indigenous peoples, mental, emotional, spiritual and physical health are integrated, interdependent and inseparable (Voss et al. 1999; Barlow and Walkup 1998; Awasis Agency 1997; Connors 1993). However, the 'spirit dimension' is badly neglected in conventional social work practice. A report by the Awasis Agency of Manitoba states that 'innovative approaches to dealing with families are seldom examined . . . First Nations practice requires the adoption of an integrative approach, addressing cognitive, emotional, physical and spiritual development' (1997: 25).

McKenzie (1997: 108) notes that holistic healing is important: 'because it transcends the notion of helping in the narrow therapeutic sense. Instead, it emphasises the resilience of First Nation people, and their ability to utilize self-help and cultural traditions as a framework both for addressing problems and supporting future social development at the community level.'

Traditional Native Americans often use Western medicine for physical conditions, and prefer treatment by traditional healers for emotional and spiritual healing (Barlow and Walkup 1998). Horejsi et al. (1992: 335) contend that: 'The most effective parent training programs are those that blend principles derived from modern child development with the spirituality, customs, traditions and other cultural ways of their tribe.'

A successful First Nations psychotherapist has developed a model for treating First Nations sexual abuse victims, where clients are assessed prior to treatment to determine their degree of acculturation. After assessment, treatment is based on either Western psychotherapeutic practice, traditional First Nations practices, or a combination of the two. First Nations elders and psychotherapists cooperate in designing healing strategies (Connors 1993).

Strengths versus deficits

Conventional social work practice generally operates using a 'deficit reduction' model of intervention, which attempts to respond to perceived weaknesses in the individual (Voss et al. 1999; Awasis Agency 1997). The Awasis Agency considers this model demeaning as it is based on the client's admission of inadequacy. Research tends to approach Native American families with a deficit model, rather than looking for strength, yet the diagnosis of Indian behaviours using Western clinical notions may in itself be iatrogenic (Red-Horse 1997: 247).

The 'strengths perspective' (Saleebey 1992) in social work embraces concepts of empowerment, collaboration, healing from within and suspension of disbelief (Voss et al. 1999). Native American and Canadian professionals report that the strengths perspective is more compatible with their communities than prevailing social work pedagogy and practice, which is generally Eurocentric (Voss et al. 1999, Awasis Agency 1997). Indigenous child and family services will be enhanced by harnessing cultural strengths (Tong and Cross 1991; Ronnau et al. 1990). However, as discussed with respect to accountability issues, whatever the approach there needs to be checks and balances to ensure that children's safety is prioritised.

<5>Healing through education and decolonisation

Indigenous groups involved with child welfare agree that child abuse and neglect in their communities result to a large extent from the effects of colonisation. A Canadian service puts it this way:

We understand the child welfare system as a system which has evolved in the dominant culture, to deal with the problems of industrial society. Within the Native community, the child welfare system is a system that deals with the symptoms of larger social problems - racism, poverty, underdevelopment, unemployment, etc. [We regard] child welfare problems as the result of the colonial nature of relations between the aboriginal people and the Euro-Canadian majority (Ma Mawi Wi Chi Itata 1985, cited in Armitage 1993: 159).

Few child welfare service models developed for or by Indigenous people respond directly to the colonial causes of problems. Helping the client to get in touch with Indigenous identities is an important part of the process. The following models consider the above issue:

Ma Mawi Wi Chi Itata is an urban Indian social agency established in Winnipeg in 1984: Ma Mawi integrates mainstream social work practices with Indigenous traditions in its work. The emphasis is on positive relations with other agencies, advocacy, exchange and collaboration. Ma Mawi is the largest urban Native agency in Canada. They describe their practice as a process of decolonisation.... We see this as a conscious process through which we regain control over our lives and resources (Ma Mawi Wi Chi Itata 1985 quoted in Armitage 1993: 159).

Research by Brave Heart-Jordan (1995, cited in Duran et al. 1998) found that Lakota clients in the United States, who engaged with traditional healing found workshops 'made their lives more meaningful and helped to liberate them from the symptoms of ongoing neo-colonialism that may have been imposed on them by other health systems that were not aware of [the history, dynamics, and politics of the American Indian soul wound]' (Duran et al. 1998: 351-352).

Other findings included that: (a) educating people about historical trauma leads to increased awareness of its impact, and symptoms; (b) the process of sharing experiences with others of similar background leads to a cathartic sense of relief; and (c) the healing and mourning process initiated, resulted in an increased commitment to ongoing healing work at an individual and community level.

Very high proportions of respondents were favourable about the traditional healing workshops, in terms of grief resolution, and feeling better about themselves. Parenting was improved.

Rokx (1998) and colleagues in New Zealand have developed a parenting model which takes the effects of colonisation on Maori child-rearing practices into consideration. The Atawhaingia Te P? Harakeke model is currently being delivered to male Maori clients in two New Zealand prisons. The model's 'decolonisation' process is intended to educate participants about contemporary Maori socio-political contexts, and the role of colonial history and ongoing neocolonial factors on them.

Participants are taught about the initial and ongoing breakdown of traditional systems, values, beliefs and practices around caring for children, and traditional family structures, which occurred as a result of white settlement. Participants gain a detailed understanding of various specific factors of influence, including: the effects of mixed marriages on family structures; introduction of Christianity; the decline of the Maori language; and the government's policy of assimilation, particularly through European schooling.

The training then moves on to focus on issues of power and control, in general, then personal terms. Participants are encouraged to position their own family backgrounds into this social history of New Zealand, and to focus on the specific circumstances of their upbringing. Participants are encouraged to set parenting goals, with a view to connecting with traditional values through improved parenting.

Community awareness raising/education

Some child abuse and neglect intervention projects attempt to bring about change through strategies involving community-wide awareness raising, as distinct from individual interventions with abuse cases. Cross and LaPlante (1995) argue that the greatest constraint to child abuse and neglect interventions in Native American communities is denial, and that grassroots community involvement is the best antidote. They point out that prevention can be grounded in traditional values and principles. Although acknowledging the substantial breakdown of tradition in some communities, what remains can be drawn upon. Cross and LaPlante contend that grassroots efforts work well because 'no one knows the community better than the community itself' (1995: 27).

The Hollow Water community of Ottawa began a program of community awareness and education in 1987. The environment of greater trust which followed saw a dramatic increase in the number of sexual abuse disclosures (Lajeunesse 1993).

Cross and LaPlante (1995) contend that conventional services, including tribal, have not succeeded in prevention. They say that, approaches which draw on local experience are valuable. The authors cite the example of a grassroots child abuse and neglect prevention campaign developed by the Siletz Tribe in Siletz, Oregon, where there was a high rate of abuse and neglect.

They planned an awareness activity: community members were sent a blue ribbon with instructions to affix it to their cars at a designated time, to show support for child abuse and neglect prevention. Another activity was a 'Family Fun Fair', with a focus on children's activities, but also an outside speaker who related her own experience with child abuse and neglect.

The authors make a range of suggestions for other communities considering similar interventions. These include:

• when forming a local committee, invite 'key participants' such as teachers, spiritual leaders, elders, and community health workers;
• organise the meeting in an appropriate informal environment - for example, meet over a meal; set a simple and achievable agenda;
• structure the group in a way which helps to bond committee members;
• boundaries of the community need to be defined, to help focus the group; and
• a needs assessment should be done to find out community strengths, weaknesses and requirements; and to determine the extent of child abuse and neglect - assessment models used by agencies and programs in the community may be available for adaptation.

Sexual abuse: Traditional healing and offender treatment

Rates of child abuse and neglect are almost universally higher in Indigenous compared with general populations. The unique histories of trauma and injustice suffered by Indigenous people under colonial regimes are clearly associated with the disproportionately high rates of sexual abuse in communities today. These specific circumstances demand consideration in health and welfare responses. Conventional approaches are, in any case, unlikely to be culturally appropriate.

Many Canadian First Nations communities are looking to more holistic methods for dealing with sexual abuse. A British Columbia community consultation recommended that courts should be empowered to offer first-time sexual abuse offenders 'extensive treatment' as an alternative to incarceration, and that culturally appropriate treatment should be available to sex offenders (Community Panel 1992). Sex offenders were regarded as needing healing rather than punishment, and contributors emphasised that the healing should take place within the community.

Many Canadian First Nations communities have recently adopted alternative strategies for dealing with sexual abuse. A number of these strategies have evolved from a 1992 sexual abuse treatment program developed by Oates for use in northern British Columbia communities. Oates' model is based on an 18 step consensual 'Traditional Process' involving extended family gatherings (effectively a version of family group conferencing). (Connors and Oates 1997).

Connors and Oates (1997) conclude that community-based responses to sexual abuse should involve the following basic elements: some esteem-enhancing form of punishment; victim protection; and treatment for all members of the family. Solutions may also include: community service; restricted access to children; native-oriented treatment program; and attendance at community support groups.

The Hollow Water program, the Community Holistic Circle Healing Project (CHCH), is used with sexual abuse cases in Manitoba Indigenous communities. CHCH heals by providing support, guidance and counselling to all those affected by sexual abuse, including the victim, the perpetrator, and respective families (see Family Group Conferencing).

By dealing with the needs of all involved, CHCH is seen as healing the community, not addressing an individual problem. The method is seen as a long-term solution, as the whole process is estimated to take five years. CHCH empowers communities by allowing members to generate their own response to individual situations, in a manner which gives consideration to the specifics of each case. The Hollow Water program transfers power from the mainstream legal system to the community (Awasis Agency 1997).

The Hollow Water program is widely viewed as a successful example of an Indigenous-controlled sexual abuse treatment program (Connors and Oates 1997; Awasis Agency 1997; Lajeunesse 1993). A cost-benefit analysis of the Hollow Water program found that for every $2 which the Provincial and Federal Government spent on the program the community receives well over $6.21 to $15.90 worth of services. Further, the program has a very low recidivism rate with only two clients re-offending over a ten-year period. (The cost benefit analysis did not take into account the costs saved from perpetrators not re-offending.) Other benefits from the program included improved holistic health for children, more people completing their education, better parenting skill, an increase in sense of safety, a return to traditional ceremony and a decrease in overall violence (Native Counselling Services of Alberta 2001).

Family preservation versus child protection

Family preservation has been broadly described as: 'an expansive concept that incorporates: (1) core philosophic tenets enjoining society to recognise that every child should grow up in a permanent family and proposing that the best way to accomplish permanency is by preserving families and preventing the placement of children outside the home; (2) a set of action theories prescribing how agencies should respond to children and families in need; and (3) a portfolio of practice technologies promoting the usage of specific service delivery techniques and interventions' (Mannes 1990: 1).

There has been an enduring international trend towards recognition of family preservation in Indigenous child welfare policy. This contrasts with the previous placement-oriented child protection emphasis. However, in the United States, Canada and New Zealand, as well as Australia, Indigenous rates of placement in substitute care are much higher than for other groups. This imbalance indicates that family preservation is of particularly acute relevance to Indigenous child welfare. The suitability of home-based family preservation initiatives is underlined by the high value placed on family and extended family in traditional Native American cultures (Tafoya 1990). The family and home-based nature of family preservation services has an adaptability which renders it suitable for use with the variety of cultural traditions represented in Native American groups (Ronnau et al. 1990).

However, contemporary Native American child welfare has emphasised protecting children above family preservation. Native culture has been protected through a policy of Native placement rather than family preservation (Mannes 1993). However, research claims that family preservation is likely to be a better means to cultural preservation than culturally appropriate placement (Mannes 1993).

The report of a Manitoba community consultation outlines a simple, clear argument for family support services:

It is difficult to understand why children are taken out of homes, then, perhaps some time later, placed back in the home where the problems began. The problems do not go away. Why not fix the home, [First Nations people] wonder, but there is little or no funding allocated to services for families ... [the community] perception is that government will pay astronomical costs for someone else to give custodial care to their children while they stand by in helpless poverty because someone else controls the money and has the power to make decisions about their children(First Nations Task Force 1993: 49).

Anderson's (1998) research provides another example of Indigenous community support for family preservation. Interviews were conducted with urban Canadian Aboriginal clients of Native Child and Family Services of Toronto (NCFST). Interviewees viewed existing non-Aboriginal child protection agencies negatively. The feeling was that parents' needs were ignored by non-Native agencies, who lied to them and didn't care about them. In contrast, clients liked how NCFST treated the whole family as a unit in programs and therapy. Through its focus on families, NCFST was seen as actually strengthening Native culture. Clients mentioned NCFST program outcomes as including: better parenting, less violent child, increased openness and sharing for both parents and children. To genuinely implement family preservation policy and programming requires a paradigm shift in child welfare, from a model based on child rescue and placement to one of family support.

Family preservation models create a 'service continuum', by delivering services that support and strengthen families in normalised environments such as the home, and focussing on basic life skills and environmental problems (Mannes 1993). Research, evaluations and community consultations (derived from Mannes 1993, 1990; Community Panel 1992; Smollar and French 1990) highlight a number of important features and orientations for family preservation programs:

• building on existing family strengths;
• intensive home-based support services;
• community education to engender support for family preservation;
• recruitment and training of Indigenous staff;
• fostering cooperation among multiple service providers;
• effective coordination between various agencies at a given site;
• secure long-term funding;
• longer program timeframes (greater than the usual four to six weeks; Community Panel 1992);
• reunification work; and
• attempts to minimise the impact of placements, where placements are unavoidable.

In order to enhance child abuse and neglect prevention, and therefore family preservation, services, such as the following, must be provided to Aboriginal communities: respite and homemaker services, particularly for single-parent families; day-care; family support; counselling; drug and alcohol treatment programs; suicide prevention services; and educational and recreational facilities and resources (Community Panel 1992).

However, devolution of authority for provision of child protection to communities does not necessarily result in lower placement rates, at least in the short term. For example, after the introduction of the ICWA in the United States, Indigenous placements increased. The number of Native American children in substitute care grew by 25 per cent over the 1980s (Mannes 1990:11). Surprisingly, tribally-run programs were the primary contributors to this post-ICWA increase. However, it is important to recognise that trends in placement numbers are not always a reliable indicator of service effectiveness as a wide range of factors influence these trends (McKenzie 1997: 110).

Placement rates may also increase after handover of authority for child protection because Indigenous agencies elicit more trust. In the case of Weechi-it-te-win Family Services, an Ontario tribal agency responsible for child and family services to ten First Nations reserves, disclosures of abuse rapidly increased in number after they took over control of child protection. This did not reflect higher rates of child abuse and neglect, but that children and families in need of help were more readily identified and supported. The agency has established high levels of trust in the communities, contrasting with the mistrust of the previous children's aid society (Weechi-it-tewin 1995).

A number of obstacles to the implementation of family preservation initiatives have been recorded in the literature. For example, 'organisational and administrative structures and state and local financing practices appear to be barriers in shifting service provision from child placement focused to family-centred services' (cited in Mannes 1990: 21). There are also legislative impediments.

Canadian provincial child welfare agencies generally only provide services to families in crisis, so perpetuate high removal rates by not providing services to families at risk: children are removed from situations where early intervention could have preserved the family. Financial support for a child is usually only available once the child is removed (Community Panel 1992). In the United States, federal discretionary funding announcements have also favoured placement over preservation (Mannes 1993).

The Community Panel of British Columbia First Nations (1992) found that legislative definitions for children needing protection should differentiate between 'children who are in immediate danger in their present environment, and children who would suffer in the long term if an intervention did not occur' (1992: 67). Such a delineation would allow for limitations to be placed on the power of authorities to remove children in less critical circumstances, which would in turn reduce placement rates. The panel also recommended that 'judicial bodies must be able to order the removal, from the home, of the person presenting a threat to the child, rather than removing the child' (1992: 67).

Ultimately, family preservation also depends on factors outside the scope of most social welfare programs. Most family preservation initiatives focus on psychological therapeutic interventions, while the factors contributing to family crisis include wider social, political and economic issues (which include poverty, unemployment, isolation, disempowerment, the legacy of separations and other colonial interventions).

In their study of six Native American placement prevention projects, Smollar and French (1990) found that the two most successful projects both attempted to meet both the substance abuse and social service needs of the client families. Their effort to provide a coordinated response with this dual mandate appears to have been the most important factor in the success of these projects. It has been estimated that substance abuse is involved in about 90 per cent of all child abuse and neglect cases in American Indian communities (Cross and LaPlante 1995), and anecdotal evidence suggests that a similar correlation exists in many other Indigenous groups, including many Australian Aboriginal communities. Smollar and French's finding is acutely relevant, and should be considered in the development of child abuse and neglect intervention models.

Although there has been substantial support for family preservation programs operating in Indigenous agencies, it is not unanimous. The 1995 Gove Report into child welfare in British Columbia was commissioned following the death of a five-year-old boy resulting from injuries inflicted by his mother (Schmidt 1997). In the Report it was concluded that child welfare services were focused on the benefits to parents rather than children, and that the child died as a result of a policy that put parents' needs before children's.

Some researchers have suggested that models such as the 'strengths' approach and FGC neglect the fundamental rights of children (Schmidt 1997). However, Schmidt (1997) points out that this response represents a Eurocentric view of the rights of children. Schmidt has no objection to Gove's conclusions in themselves, but asserts that they reflect a position incompatible with child protection for First Nations communities, who 'regard the larger community as parent'. The importance of community and extended family in First Nations (and other Indigenous) parenting is not given consideration in Gove's conclusions. A number of emerging child abuse and neglect intervention models used by First Nations are variations on the family group conferencing theme.

Family group conferencing and other similar models

There are many models and programs in place which could appear under the broad heading of Family Group Conferencing.

Community Holistic Circle Healing - Canada

The Hollow Water Community's Holistic Circle Healing (CHCH) in Ottawa 'aims to restore balance by empowering individuals, families and the community to deal productively, and in a healing way, with the problem of sexual abuse' (Lajeunesse 1993: 1).

CHCH has an Assessment Team and a Management Team. The Assessment Team provides prevention and intervention, develops support systems, and provides assessment and liaison with lawyers and institutions. The team consists of various professionals such as family violence workers and nurses, volunteers and Tribal Council representatives. The Management Team is responsible for administration. The CHCH process follows a 13-step process. The assessment team assigns an individual (trained) worker to each of the victim(s), the perpetrator and their families.

Children are removed to a safe home 'if necessary', and the laying of charges is encouraged. Where the perpetrator pleads guilty in court, CHCH prepares and presents a report outlining the tailored CHCH alternative plan for that individual. A formal protocol has been developed with Crown prosecutors in the Manitoba Department of Justice; the judiciary supports CHCH fully. Perpetrators making a guilty plea are put on three years' probation; very few plead 'not guilty'.

Another aspect of the integrated CHCH program is the 'Self-Awareness For Everyone' (SAFE) program, a personal growth training program implemented in the communities. At one community, 60 per cent of adults and youth participated in the program, which 'had a profound and positive effect on general levels of self-awareness, alcohol consumption, family communication, and the healing process of many people, including the victims of sexual abuse' (Lajeunesse 1993: Appendix C).

Treatment combines contemporary and traditional methods in 'healing contracts'. A psychologist provides assessment and counselling services for all referrals. A male psychologist facilitates an 'Adult Male Sex Offenders Group', while various other groups for survivors and families are facilitated by workers. Traditional 'sharing circles' are used in group sessions. 'CHCH offers balance rather than the punishment that is offered by the court system' (Lajeunesse 1993: 9; see also the discussion of CHCH under Sexual abuse: Traditional healing and offender treatment above).

Family Group Decision Making Project - Canada

The Family Group Decision Making (FGDM) project was modelled on the New Zealand Family Group Conferencing model. There are three main stages to the process. First, preparing for the conference: family members are contacted but not compelled to take part; consultations with the family take place; abused young people usually select a support person. Second, holding the conference: the meeting is opened in a culturally appropriate manner; the coordinator presents necessary information to the group, followed by private deliberations amongst the family group (and chosen support people) to develop a plan; the plan is finalised with the coordinator; the plan must contain contingency plans, and monitoring and review processes. Third, the plan is approved by a field worker and supervisor.

Negative critique of family group conferencing

Critics note that Family Group Conferencing can lack the safeguards of due process and legal representation. Further, it is difficult to find a balance between child protection and family preservation. Where the child is not removed, or the couple stays together, child abuse may continue (Burford and Pennell 1995). For example, 'Pre-Sentence Reports' attached to a report on the wellknown Community Holistic Circle Healing (CHCH) model implemented by the Hollow Water community of Ottawa include a case where a father continued to sexually abuse his adopted daughter after treatment using the CHCH system.

Burford and Pennell (1995) suggest that some family members may privately see the Family Group Conferencing process as unfair, secretly preferring formal action, and that a family 'conspiracy of silence' may continue under FGC. Burford and Pennell also note that resources must be made available for the model to work, and that there is a risk that authorities may try to apportion too much responsibility to the family, where there is still a vital role for protective services.

Finally, power imbalances often occur between child and family services agencies, legal staff, and families; and gender inequities are often not addressed by Family Group Conferencing approaches. The use of Family Group Conferencing, particularly when used in small communities, may inhibit disclosure by women and children or result in coercion by the offender not too disclose. Burford and Pennell (1995: 46) suggest that some family members may be too intimidated to make disclosures at conferences due to the presence of the perpetrator or others. Confidentiality problems are more prevalent in small communities and these may have an impact on the family group conferencing process (see also Tafoya 1990).

Positive critique of family group conferencing

Schmidt (1997: 81) claims that FGC approaches 'emphasise community responsibility in protecting the child through the offender's accountability to the group. Discounting [the FGC] approach will serve to perpetuate the present system which mitigates against the possibility of reconciliation and holistic healing.'

Burford and Pennell (1995) argue that family and extended family are marginalised by justice, health, education and social service systems, and the assumption is that families are not part of the solution. In contrast to the views above, they see the FGDM process as a means to overcome the 'conspiracy of silence' surrounding abuse of both children and adults. The method does not attempt to keep families together 'at all costs', and the authors acknowledge that the system will not always work, and that involvement of abusers is not always appropriate. They suggest that the majority of family conference participants participate responsibly and devise useful plans. They claim that the model is highly adaptable across cultures, so long as local people are very involved in the adaptation process. They say that the family is empowered, and family ties are renewed, re-established and strengthened. Feedback from Inuit participants has mainly been positive.

Burford and Pennell (1995: 41) quote the District Director, Correctional Services of Canada: 'We can say with a high level of confidence that the conferences in which our staff and clients participated were extremely cost effective . . . [and] . . . the benefits derived for both staff as well as our clients were numerous.'

Early intervention

Numerous studies show that in the United States the Head Start program - an early intervention initiative targeting at-risk children of pre-school age - has achieved considerable success over a 30-year period (for example, Zigler and Styfco 1996). The Canadian 'Aboriginal Head Start Initiative' is an early intervention program for Canadian Aboriginal children, in both urban and remote communities.

The program involves parents and community in design and implementation of projects, which include promotion of culture and language, education, health and improved social supports. Head Start does not specifically target child abuse and neglect. However, research into the effects of early intervention programs indicates many benefits including some linked to child abuse and neglect issues, including: support for families; better relationships between parents and children; improved social and emotional stability in participating children; and enhanced community capacities (Health Canada 1998).

Evaluation of another early intervention program found that it was also successful, enhancing life for children and families. The project was piloted in seven urban Ontario communities, targeting high risk Aboriginal three to five-year-olds (Becker and Galley 1996). Children involved demonstrated improved confidence, better behaviour, improved language skills, and better communication and expressiveness.

In September 2002, the Canadian Federal Government reaffirmed its commitment to early intervention programs for First Nations children. The Canadian Government has committed a further $320 million over five years to enhance Aboriginal Head Start, First Nation and Inuit Child Care Programs, and to address Fetal Alcohol Syndrome/Fetal Alcohol effects in First Nation Communities.

CONCLUSION

Legislative schemes provide the framework within which programs for the delivery of child welfare services to Indigenous communities are developed and delivered. Legislation reviewed for this paper ranges from that where all aspects of decision making including legislative, judicial and administrative decision making are transferred to Indigenous communities such as under the Indian Child Welfare Act 1978 (USA), to that where all jurisdiction is retained by the state such as under the Children Young Persons Care and Protection Act 1998 (NSW). Within each model considered measures have been taken to recognise and facilitate Indigenous input into decisions affecting Indigenous children.

Good partnerships between agencies and Indigenous organisations require some equality in the relationship. This usually requires the agency to relinquish power and to recognise the authority of the Indigenous community or organisation. Effective recognition often requires legislative recognition. To work in a culturally competent manner with communities it is necessary to understand the historical influences on those communities, including trauma generated by previous colonial policies, and related personal, family and community issues with respect to identity (Weaver 1998). For service delivery to be culturally competent it needs to move beyond incorporation of Indigenous staff in standard delivery programs, to the incorporation of cultural knowledge into the service delivery framework (Tong and Cross 1991). Services need to develop and incorporate locally identified knowledge, skills and values in order to achieve cultural competence. This includes knowledge of the peoples in the area, their communication systems and culture, and their contemporary realities including local inter and intra community politics, and socio economic situations (Weaver 1999).

Where communities are not designing and delivering programs themselves they need to be involved at all stages of program development. From initiation of programs and identification of community needs, to the evaluation of programs. Conventional criteria, whether they are to be used to determine children's needs, or for measuring 'success' (program evaluation), will often need to be developed (Ricks et al. 1990; Awasis Agency 1997). This can be a complex and difficult process which requires considerable resourcing and support. Likewise, facilitating the training and retention of Indigenous staff working on projects requires adequate resourcing, lateral consideration of qualifications, and consideration of shortfalls in current training programs (First Nations Task Force 1993; Durst 1998).

Many mainstream children's services which are grappling with the vexed question of how to deliver child protection services more effectively are looking to more collaborative, community based responses to service delivery (Wharf and Mckenzie 1998). The histories of Indigenous communities' dispossession, and their experiences of forced removals of children from their families, make the issue of community control over children's services particularly important for Indigenous people. A key component to the devolution of authority to local communities is empowerment of the community. In some instances the phased devolution of authority to Indigenous communities can facilitate empowerment of communities while accommodating different levels of resourcing and readiness for taking control of children's services (Aboriginal Family Healing Joint Steering Committee 1993).

Factors which influence community satisfaction with service delivery include a holistic approach to healing, autonomy and flexibility in service provision, and a collaborative approach to service provision (McKenzie 1997). An important and often referred to aspect of effective service provision is the capacity to respond to local needs (Awasis Agency 1997). This includes provision of the service in an environment which clients feel comfortable and secure in and which they can realistically get to and use.

While the larger part of literature examining Indigenous control of children's services emphasises devolution of authority to a local level, a significant countervailing issue is that of compromised accountability through political and family interference in child protection services (Giesbrecht 1992; Gray-Withers 1997). This is a serious issue which can compromise the probity and effectiveness of some agencies and can leave Indigenous women and children vulnerable. Accountability gaps also exist in mainstream service provision. A number of suggestions have been made to respond to accountability concerns. These include establishment of an independent body to whom the service provider is accountable, adoption of agency protocols, and establishment of independent dispute resolution mechanisms (First Nations Task Force 1993; Gray-Withers 1997).

Much of the literature reviewed for this paper associated a holistic response to children's needs with incorporation of cultural revival/acknowledgement and traditional healing. This is viewed as empowering because use of cultural traditions to respond to problems models communities' capacity for self-help and provides a framework for future social development at the community level. For similar reasons an approach to social work which focuses on community strengths is viewed as empowering and enhancing of the holistic healing approach to children's wellbeing which is supported by many communities.

Many of the problems associated with child abuse and neglect in communities are directly related to experiences of colonisation. Much of this has been experienced as a group. Group responses which educate people about the historical trauma which they as a group and individually have faced can lead to increased self-awareness, be cathartic, and provide people with a sense of relief. This has been used in healing programs to provide communities with a framework within which to understand their problems. Other projects have attempted to address abuse and neglect through community strategies which raise awareness within the community of child abuse and neglect and related issues such as substance abuse. Specific programs considered include traditional healing and offender treatment for sexual abuse and various forms of family group conferencing.

The two most frequently cited criticisms of existing child welfare services is the high level of out of home placement for Indigenous children and the limited resources and programs available to help families in need of assistance (Community Panel 1992). For policies of early intervention to be effective a broad range of services are needed. These may include parent support programs, respite care and day care, counselling and substance abuse treatment programs, suicide prevention programs, and educational and recreational facilities. Incorporation of such services in child protection work is consistent with the whole community approach to child protection which is advocated by many communities (Mannes 1993). The factors which contribute to family crises are often deep set and include intergenerational experiences of trauma, poverty, isolation and disempowerment. Coordination of services provided by a range of government and nongovernment organisations and financing of services needed have been identified as barriers to effective preventative programmes.

This paper provides a review of legislation and services delivered to Indigenous communities in Canada, the United States of America and New Zealand. While some of the issues and ideas may be useful and relevant in the Australian context, a key finding in the research is that a 'one size fits all approach' does not work. Research and programs for children's wellbeing need to be developed, implemented and evaluated locally.

REFERENCES

• Aboriginal Family Healing Joint Steering Committee (1993), 'For generations to come: The time is now: A strategy for Aboriginal Family Healing', Unpublished Final Report, Ontario.
• American Humane Association - National Resource Centre on Child Abuse and Neglect (1997), Tribal-State Relationships in Child Welfare: Ensuring High Quality Services to American Indian Children and Their Families, American Humane Association, Englewood.
• Anderson, Kim (1998), 'A Canadian child welfare agency for urban natives: The clients speak', Child Welfare, vol. 77, no. 4, pp. 441-460.
• Armitage, Andrew (1993), 'Family and child welfare in First Nation communities', in B. Wharf (ed.) Rethinking Child Welfare in Canada, McClelland and Stewart, Toronto.
• Atkinson, J. (2002), Trauma Trails: Recreating Song Lines, Spinifex Press, Melbourne.
• Australian Institute of Health and Welfare (1998), Juvenile Justice and Youth Welfare: A Scoping Study, AIHW, Canberra.
• Awasis Agency of Northern Manitoba (1997), First Nations Family Justice: Mee-noo-stah-tan Mi-nisi- win, The Agency, Thompson, Manitoba.
• Barlow, A. & Walkup, J. T. (1998), 'Developing mental health services for Native American children', Child and Adolescent Psychiatric Clinics of North America, vol. 7, no. 3, pp. 555-577.
• Becker, J. & Galley, V. (1996), 'Aboriginal Head Start Summer Pilot Program: Evaluation and Final Report', Unpublished report.
• Brady M. (1999), 'Dealing with alcohol: Indigenous usage in Australia, New Zealand and Canada', Oceania, no. 70, pp. 201-203.
• Brave Heart, MYH (1999), 'Oyate Ptayela: Rebuilding the Lakota Nation through Addressing Historical Trauma among Lakota Parents', Journal of Human Behaviour and the Social Environment, vol. 2, no. 1/2pp. 109-126.
• Burford, G., & Pennell, J. (1995), 'Family group decision making project: Implementation report summary', Family Group Decision Making Project, School of Social Work, Memorial University of Newfoundland, Newfoundland, Canada.
• Carr, J. & Peters, M. (1997), 'Assessment of multi-disciplinary child protective teams in five western Washington communities: Chehalis, Nisqually, Shoalwater Bay, Skokomish, Squaxin Island', Unpublished MPA thesis, The Evergreen State College, Washington.
• Cheroke Nation 30 U.S. (5 PET) 1 (1831).
• Child and Family Services Act 1990 (Ontario).
• Children and Family Services Act (Nova Scotia).
• Community Panel (1992), Liberating our Children, Liberating our Nations: Report of the Aboriginal Committee, Community Panel, Family and Children's Services Legislation Review in British Columbia, The Community Panel, Victoria, BC, Canada.
• Connors, E. (1993), 'Healing in First Nations: The spirit of family', in M.R. Rodway & B. Trute (eds) The Ecological Perspective in Family-Centered Therapy, Studies in Health and Human Services No.19, Edwin Mellen Press, New York.
• Connors, E. & Oates, M. (1997), 'The Emergence of Sexual Abuse Treatment Models Within First Nations Communities', in D. Wolfe, R. McMahon & R. Peters (eds) Child Abuse: New Directions in Treatment and Prevention Across the Lifespan, Sage, Thousand Oaks, London.
• Cross, T. L. & McGregor, K. (1995), Linking Child Abuse and Neglect with Substance Abuse in Native American Communities (Evaluation report), National Indian Child Welfare Association, Portland, Oregon.
• Cross, T. & LaPlante, J. (1995), Grassroots Prevention of Child Abuse and Neglect in Indian Communities: A Guide for the Community Organiser, National Indian Child Welfare Association, Portland, Oregon.
• Cunneen, C. (1997), 'Community conferencing and the fiction of Indigenous control' Australian and New Zealand Journal of Criminology, vol. 30, no. 3, pp. 292-311.
• Cunneen, C. & Libesman, T. (2002), 'Removed and discarded: The legacy of the Stolen Generations', Australian Indigenous Law Reporter, vol. 7, no. 4, pp. 1-20.
• Department of Indian and Northern Affairs (1989), Indian Children and Family Services Management Regime: Discussion Paper, Department of Indian and Northern Affairs, Ottawa.
• Dumont-Smith, C. (1995), 'Aboriginal Canadian children who witness and live with violence', in E. Peled, P. Jaffe & J.L. Edleson (eds) Ending the Cycle of Violence: Community Responses to Children of Battered Women (pp. 275-283), Sage Publications, Thousand Oaks, CA, USA.
• Duran, E., Duran, B., Brave-Heart, M.Y.H. & Yellow Horse-Davis, S. (1998), Healing the American Indian Soul Wound, in Y. Danieli, (Ed), International Handbook of Multigenerational Legacies of Trauma, Plenum Press, New York.
• Durst, D. (1998), 'The wellness of Canadian First Nations children: Seeking solutions through selfgovernment', Unpublished draft research paper.
• Durst, D., McDonald, J. & and Rich, C. (1995), 'Aboriginal Government of Child Welfare Services: Hobson's Choice?', in J. Hudson & B. Galaway (eds) Child Welfare in Canada. Research and Policy Implications, Thompson Educational Publishing, Toronto.
• Farrow, F. & the Executive Session on Child Protection (1998), Child Protection: Building Community Partnerships, John F. Kennedy School of Government, Harvard University, Cambridge, Massachusetts.
• First Nation's Child and Family Task Force (1993), Children First, Our Responsibility: Report of the First Nation's Child and Family Task Force, Manitoba, November 1993, The Task Force, Winnipeg.
• Frenza, L. (1993), 'An early intervention approach to ending child abuse and neglect: Hana Like home visitor program', Journal of Psychohistory, vol. 21, no. 2.
• Gant, F. & Grabosky, P. (2000), The Promise of Crime Prevention, Second Edition, Research and Public Policy Series, No 31, Australian Institute of Criminology, Canberra.
• Gilling, M. Patterson, L. & Walker, B. (1995), 'Family members experience of care and protection family group conference process', in Social Policy Agency Study, Ministry of Social Development, Wellington, New Zealand.
• Gray-Withers, D. (1997),'Decentralised social services and self-government: Challenges for First Nations', in J. Pulkingham & G. Ternowetsky (eds) Child and Family Policies, Struggles, Strategies and Options, Fernwood, Halifax.
• Gwa'sala-'Nakwaxda'xw (1993), 'Gwa'sala-'Nakwaxda'xw family violence and sexual abuse intervention program', Unpublished document.
• Hart, R. (1997), Beginning a Long Journey: A review of Projects Funded by the Family Violence Prevention Division, Health Canada, Regarding Violence in Aboriginal Families, Health Canada, Ottawa.
• Health Canada (1998), Aboriginal Head Start Initiative: Principles and Guidelines, Health Canada, Ottawa.
• Health Canada (1999), 'Aboriginal Head Start on reserve: National evaluation framework', Unpublished report of Health Canada
• Health Research Council of New Zealand website: http://www.hrc.govt.nz/
• Herbert, A. & Te Kanawa, D. (1998), 'The Mana Kainga Program: Practice and research perspectives: A Manaipoto model of family empowerment', Paper presented at the 12th ISPCAN International Congress on Child Abuse and Neglect, September 1998, Auckland, New Zealand.
• Herron, J. (1999), 'Principles for the provision of community-based Indigenous family violence programs, Unpublished paper.
• HREOC (Human Rights and Equal Opportunity Commission) (1997), Bringing them Home: Report of the National Inquiry into the Separation of Aboriginal and Torres Strait Islander Children from their Families, Human Rights and Equal Opportunity Commission, Commonwealth of Australia, Sydney.
• Horejsi, C., Bonnie Heavy Runner, C. & Pablo, J. (1992), 'Reactions of Native American parents to child protection agencies: Cultural and community factors', Child Welfare, vol. 71 no. 4, pp. 329-343.
• Hudson, P. & McKenzie, B. (1987), Evaluation of the Dakota Ojibway Child and Family Services, Department of Northern and Indian Affairs, Winnipeg.
• Hudson, P. & Taylor-Henley, S. (1987), Agreement and Disagreement: An Evaluation of the Canada- Manitoba Northern Indian Welfare Agreement, University of Manitoba, Winnipeg.
• Indian Child Welfare Act of 1978.
• Johnson v McIntosh Supreme Court of United States, 1823, 21 US (8 Wheat) 543, 1823.
• Kalyanpur, M. (1998), 'The challenge of cultural blindness: Implications for family-focused service delivery', Journal of Child and Family Studies, vol. 7, no. 3, pp. 317-332.
• Lajeunesse, T. (1993), Community Holistic Circle Healing: Hollow Water First Nation, Minister of Supply and Services, Ottawa.
• Mannes, M. (1990), Family Preservation and Indian Child Welfare, American Indian Law Center, Albuquerque, New Mexico.
• Mannes, M. (1993), 'Seeking the balance between child protection and family preservation in Indian child welfare', Child Welfare, vol. 72, no. 2.
• Mansfield, H. (1997), 'Hawaii's Hana Like Home Visitor Program: A Healthy Start program', Journal of Psychohistory, vol. 24, no. 4, pp. 332-338.
• Manitoba Child and Family Services Act 1984.
• Ma Mawi Wi Chi Itata (1985), First Annual Meeting Report, Winnipeg.
• Mason Report (1992), The Report of the Ministerial Review Team to the Minister of Social Welfare the Hon Jenny Shipley (The Mason Report), New Zealand Government, Wellington, NZ.
• McKenzie, B. (1997), 'Connecting policy and practice in First Nations Child and Family Services: A Manitoba case study', in J. Pulkingham & G. Ternowetsky (eds) Child and Family Policies, Struggles, Strategies and Options, Fernwood, Halifax.
• McKenzie, B., Seidl, E. & Bone, N. (1995), 'Child welfare standards in First Nations: A communitybased study, in J. Hudson & B. Galaway (eds) Child Welfare in Canada, Research and Policy Implications, Thompson Educational Publishing, Toronto.
• Morrisette, P.J. (1994), 'The Holocaust of First Nation People: Residual effects on parenting and treatment implications', Contemporary Family Therapy, vol. 16, no. 5, pp. 381-392.
• National Crime Prevention (1998), Pathways to Prevention: Developmental and Early Intervention Approaches to Crime in Australia, National Crime Prevention Report No 6, Attorney-General's Department, Canberra.
• Native Counselling Services of Alberta (2001), A Cost-benefit Analysis of Hollow Water's Community Holistic Circle Healing Process, Solicitor General of Canada, Ottawa, Ontario.
• Northwest Indian Child Welfare Institute (1986), Positive Indian Parenting: A Model Indian Parent Training Manual, Northwest Indian Child Welfare Institute, Portland, Oregon.
• Norton, I. M. & Manson, S.M. (1997), 'Domestic violence intervention in an urban Indian Health Centre', Community Mental Health Journal, vol. 33, no. 4, pp. 331-337.
• Oates, M. (1992), 'Native treatment programs: Circle Of Harmony Healing Society', Unpublished document.
• Prince George (1992), 'Prince George Native Friendship Centre Sexual Abuse Treatment Services', Unpublished document.
• Public Health Commission (New Zealand) (1995), He Matariki: A Strategic Plan for Maori Public Health, Public Health Commission, Wellington.
• Red-Horse, J. (1997), 'Traditional American Indian family systems', Families, Systems and Health, vol. 15, no. 3, pp. 243-250.
• Ricks, F., Wharf, B. & Armitage, A. (1990), 'Evaluation of Indian child welfare: A different reality', Canadian Review of Social Policy, vol. 24, pp. 41-46.
• Rine, M.H. (1998), Te Mana Te Kawanatanga, The Politics of Maori Self Determination, Oxford University Press, New Zealand.
• Rokx, H. (1998), 'Atawhaingia Te Pa Harakeke: Nurture the family', Paper presented at the 12th ISPCAN International Congress on Child Abuse and Neglect, September 1998, Auckland, New Zealand.
• Ronnau, J., Lloyd, J., Sallee, A. & Shannon, P. (1990), 'Family Preservation Skills with Native Americans', in M. Mannes, Family Preservation and Indian Child Welfare, American Indian Law Center, Albuquerque, New Mexico.
• Schmidt, G. (1997), 'The Gove Report and First Nations Child Welfare', in J. Pulkingham & G. Ternowetsky (eds) Child and Family Policies, Struggles, Strategies and Options, Fernwood, Halifax.
• Shields, C. (1995), 'Improving the life prospects of children: A community systems approach', Child Welfare, vol. 124, no. 3.
• Smollar, J. & French, R. (1990), 'A study of six Native American placement prevention and reunification projects', Unpublished report, Prepared by CSR Incorporated and Three Feathers Associates.
• Stanley, J., Tomison, A.M. & Pocock, J. (2003), Child Abuse and Neglect in Indigenous Australian Communities, National Child Protection Clearinghouse Issues Paper no. 19, Australian Institute of Family Studies, Melbourne.
• Tafoya, N. (1990), 'Home-based family therapy: A model for Native American communities', in M. Mannes, Family Preservation and Indian Child Welfare, American Indian Law Center, Albuquerque, New Mexico.
• The Matter of an Application about the L Children FC Wanganui, CYPF 1/95, 7 April 1998 at 10.
• Tong, C. & Cross, T. (1991), Cross Cultural Partnerships for Child Abuse Prevention with Native American Communities, Northwest Indian Child Welfare Institute, Portland, Oregon.
• Urbis Keys Young (2001), Aboriginal Child Welfare and Juvenile Justice Good Practice Project, Urbis Keys Young, Milsons Point.
• Van Stelle, K.R., Allen, G.A. & Moberg, D. P. (1998), 'Alcohol and drug prevention among American Indian families: The Family Circles program', Drugs and Society, vol. 12, no. 1-2, pp. 53-60.
• Voss, R., Douville, V., Little Soldier, A. & Twiss, G. (1999), 'Tribal and Shamanic-Based social work practice: A Lakota perspective', Social Work, vol. 44, no.3.
• Weatherburn, D. & Lind, B. (1998), Social and Economic Stress, Child Neglect and Juvenile Delinquency, New South Wales Bureau of Crime Statistics and Research, Sydney.
• Weaver, H. N. (1998), 'Indigenous people in a multicultural society: Unique issues for human services', Social Work, vol. 43, no. 3, pp. 203-211.
• Weaver, H. N. (1999), 'Indigenous people and the social work profession: Defining culturally competent services', Social Work, vol. 44, no. 3, pp. 217-225.
• Weechi-it-te-win Family Services Inc. (1995), 'Building healthier communities: A report of the operational review of Weechi-it-te-win Family Services', Unpublished report by Minsitry of Community and Social Services, Ontario.
• Wharf, B. (1992),'First Nation Control of Child Welfare Services', in B. Wharf (ed.) Communities and Social Policy in Canada, McLelland and Stewart, Toronto.
• Wharf, B. & McKenzie, B.(1998), Connecting Policy to Practice in the Human Services, Oxford University Press Canada, Don Mills, Ontario.
• Worcester 31 US 16 PET 5115 (1832).
• Zigler, E.F. & Styfco, S. (1996), 'Head Start and Early Childhood Intervention: The changing course of social science and social policy', in E. F. Zigler, S. L. Kagan & N.W. Hall (eds) Children Families and Government: Preparing for the Twenty-first Century, Cambridge University Press, New York.

• This discussion paper is a summarised version of a larger report by C. Cunneen and T. Libesman, A Review of International Models for Indigenous Child Protection, available at <www.austlii.edu.au/au/other/IndigLRes/2002/1/>