Diversity is the focus of the first 'key issue' designated by the National Council for International Year of the Family in Australia. The full statement is: 'To recognise the diversity of families in Australia in terms of their composition, culture and race and to celebrate their central contribution to Australia's social and economic welfare and cultural heritage.'

On the face of it, this is a straightforward statement, with which many would agree. Nevertheless, the statement raises some questions. Families vary enormously under these three broad headings and there are numerous aspects related to composition, culture and race which could be 'recognised'. What does it actually mean to recognise diversity? Why is it important? Are there limits to the celebration of diversity?

WHY RECOGNISE THE DIVERSITY OF FAMILIES?

Recognising diversity can be a two-edged sword. It depends very much on how the 'unlikeness' of families is interpreted and the implications of those interpretations. In reviewing the first ten years of the Australian Institute of Family Studies, Don Edgar (1990) argued the importance of challenging the taken-for-granted notions about families upon which public policies are often based. He maintained that sound research inevitably exposed family diversity, and the benefits in doing so were considerable. There was potential for better policies for families across the board and for greater understanding of factors which could affect any family. Overall, research which re-interpreted the 'taken- for-granted' reality could be both challenging and liberating.

As accepted wisdom is challenged and members of families who were previously not seen as fitting the norm have their experiences recognised and validated, they are less likely to feel alienated, guilty, or out of step, and they are likely to have an increased sense of wellbeing. There is also a likelihood that the wellbeing of families in general will be enhanced and parents and children will have broader and more flexible opportunities in life.

However, recognition of diversity can have its roots in, and be sustained by, discriminatory social attitudes and public policies. For example, social attitudes towards separation, divorce and single- parenthood have become much more accepting over the past couple of decades, as have attitudes towards pre-marital sex and cohabitation. But it is not so very long ago that separation and divorce were synonymous with coming from a 'broken' home. Children were automatically assumed to suffer long-lasting psychological damage, or worse still were ostracised, pitied, treated differently in schools, removed from their family networks or dealt with in various ways which frequently were far more damaging than the experience of their parents' separation or divorce. Many women who were sole-parents had to contend with punitive social attitudes as well as poverty.

Periodic verbal attacks on single mothers and discriminatory statements about children from sole-parent families indicate that such attitudes are still just below the surface. And there are still quite widespread differences in the way many people view widows (those women whose partners have died) as opposed to divorcees, separated women and women who have chosen single-parenthood.

DESCRIBING FAMILY COMPOSITION

A demographic description of Australian families is the starting point for recognising diversity of family composition, but how differences are interpreted and what is done with the information are crucial issues. We need first to mention two factors which make it difficult to give a clear picture of family composition. They are the distinction between household and family, and the dynamic nature of family over the course of an individual's life.

Official statistics from the Australian Bureau of Statistics, which are most frequently used when talking about the numbers of families in Australia, refer to household families - that is, two or more related people who usually live together in the same household. They do not refer to families as people usually define them - that is, a group of people related by blood and/or strong emotional ties irrespective of where they live. For example, most people see their parents or their children (or their brothers and sisters) as part of their family, even though they may not at the time live with them. In some cultures, cousins are part of the immediate family but do not necessarily live together. An older person may appear in official statistics as someone in a 'sole- person household', but may live close to adult children, mind grandchildren frequently and often stay with a married sister and brother-in-law. So they are very much part of a family. Similarly, a young person in a sole-person or group household may have strong connections with parents or other family members and be partially financially dependent on parents.

Co-residence is likely to be even less satisfactory in describing many migrant families. As well as strong family obligations extending beyond households, there are ties to other countries. Remittances sent overseas, family reunion sponsorship and arrangements for marriages across countries testify to extensive family obligations and connections beyond Australia. In addition, household composition may be determined by the family members who are in Australia. Some types of families may not be possible because relatives are still in the country of origin, or the composition of families may be limited by government immigration policies (McDonald 1991). In the reverse of an earlier pattern for Chinese settlers, for example, there is now a significant minority of Chinese families from Hong Kong where the wife and children live in Australia and the husband spends much of his time in Hong Kong.

Furthermore, it is important to remember that families frequently change in form and composition over the course of a person's life. Couples become parents; sole-parents remarry or live in a de facto relationship; children leave home; children are born; old people die. All of these normal life transitions have an effect on the composition of household families and on the composition of a family as perceived by an individual.

Current definitions of family, as reflected in the collection of national statistics, exclude groups of people, some of whom would regard themselves as families. Gay and lesbian couples, with or without children, are an example. The next national Census will allow people to designate a relationship between themselves and someone else in the household of the same sex as a partner. They will then be coded as a couple, rather than members of a group household, as at present.

So, while official statistics give us a useful snapshot picture of household families at a particular time they cannot convey the complexity and diversity of family functioning and relationships, how people identify themselves and their families for different purposes, and the shifting composition of individual families.

THE DEMOGRAPHIC COMPOSITION OF FAMILIES

With the above provisos, official statistics tell us much about the diversity of family composition (and household composition) along many dimensions - number of family members, number of hildren, whether young children are present or not, whether there are older dependent children or not, whether couples are married or not, age distribution of children, whether the household contains families of more than one generation, the presence or not of step- children, the number of persons employed in the household, the number of full-time students in the household. In reality, some of these factors tend to get much more attention than others.

At the 1991 Census, 75.4 per cent of Australian households were family households. Sharing of households by two or more families is very uncommon in Australia: less than one per cent of households consisted of two or more families. Among the family households, 85.3 per cent were couple families, 12.8 per cent were one-parent families and 1.9 per cent were 'other' families such as two siblings living together. In turn, couple families divided into those with dependent children (52 per cent), couples with non- dependent children (11 per cent) and couples with no children present (37 per cent).

From the perspective of families with dependent children, in 1991, almost 17 per cent of these families were sole- parent families compared with about 9 per cent in 1971. On the other hand, the proportion of sole-parent families in 1991 was the same as it was in 1891, but last century the death of a parent was more often the cause than marriage breakdown (McDonald, 1993). Thus, experience of sole- parent families is by no means only a modern- day phenomenon.

From the perspective of children, in 1986, 80 per cent lived with both their own parents (natural or adoptive), 12 per cent with one parent only and 5 per cent with one parent and a step-parent.

Thus, the family consisting of two parents and their dependent offspring is still by far the most common form of household family followed by the family consisting of a couple without children. The third most common family form is the sole-parent family. A more dynamic approach to the measurement of family patterns, however, would not show these family types as mutually exclusive, but rather as transitions. That is, most sole-parent families were once two- parent families and many will return to a two-parent situation through re-partnering. Most couple families without children have had children in the past or will have in the future.

When considering diversity, therefore, we need to remember that the family type in which a person lives at a particular point in time does not represent a choice which is immutable across time, but instead is dependent upon the life cycle stage in which the measurement is taken. The incidence of each of these stages in the population is in turn affected by demographic patterns such as the timing of marriage and childbearing and the extension of the life span.

An example of change in family life cycle stages is the increase in the past ten years or so in the proportion of families with older offspring (15-24-year-olds) living with them. Later marriage patterns, extended periods in education, affordability of housing and unemployment have contributed to the trend. Unemployed under- 21-year-olds receive less than the adult rate of unemployment benefit, so their resources to establish independence are limited. Their continued financial dependence on parents puts an enormous strain on low income families and those where one or both parents are unemployed.

In recent years a considerable amount of research, public debate and government policy activity has focused on the incidence of two- parent, sole-parent and stepfamilies, and issues surrounding their functioning. This is certainly not a new phenomenon. Single mothers in particular have over the years been the target for social comment, public condemnation, and government regulation (or neglect).

Explanations for the recent focus are not difficult to find. More realistic divorce laws and greater social acceptance of divorce, an increase in the number of sole-parent families arising because of separation and divorce and (to a lesser extent) as a result of personal choice, have led in the past 20 years to the greater incidence of sole-parent families described above.

Widespread interest in this aspect of family composition stems from many different ideological stances and includes both conservative and radical positions, moral and economic arguments. Some people are concerned that any trend away from the notion of a life-long nuclear family structure will lead to the 'death of the family'. Others look realistically at social and economic factors which affect families and seek to understand exactly why the shifts in marriage and the nature of intimate relationships are occurring. Concern about the economic disadvantage and continuing poverty of many sole-parent families, most particularly those headed by women, has highlighted this aspect of diversity in family composition.

The language of conservative attacks on sole parents mirrors the paternalism which saw government agencies in the past remove Aboriginal children from their parents: 'we recognise the struggles of these poor people, but it is no way to bring up a child'. Indeed, some recent commentators have gone so far as to suggest that children in sole-parent families should be adopted out into good, two-parent families. A much more enlightened approach is found in the statement of the Australian Catholic Bishops' Conference (1993) which states that 'sole parents are as important to our community as two-parent families' and that the efforts of sole parents under difficult circumstances 'need to be supported adequately through allowances and concessions that acknowledge the special challenges that confront sole parents'.

ECONOMIC DIVERSITY

Compared with the situation 25 years ago, there is now much greater diversity in the employment patterns of Australian families. Two trends are of particular note: the increase in the labour force participation rate of women in couple families and the increase in unemployment of family members.

Between 1971 and 1991, the labour force participation rate for married women aged 35-44 years increased from 41 per cent to 71 per cent. The increases at other ages were from 33 per cent to 61 per cent for wives aged 25-34 years and from 36 per cent to 63 per cent among those aged 45-54 years. Much of the increase, however, was in part-time work.

Considering women with dependent children, the labour force participation rate varies according to the age of the youngest child, with 47 per cent in couple families being in the labour force when the youngest child is aged less than five years, rising to 66 per cent when the child is aged 5- 9 years and 73 per cent when the child is aged 10-14 years. These statistics, however, mask a greater level of movement in and out of the labour force by women with children. Thus, not only do children have parents with widely diverging employment situations, but also the individual child experiences a diversity of patterns of employment of his or her parents while growing up.

The frequent movement between couple families being one- earner or two-earner is an argument against structuring of family policy upon the basis of whether a family has one or two incomes. Income-splitting for taxation purposes, for example, provides large benefits to one-income, high income families, but little to one- income, low income families and nothing to the rising number of two-income, low income families. Income-splitting also provides nothing to sole parents and is a disincentive to work for second earners in a family. While we would support the proposition that families with children require a greater level of support from the tax- transfer system, any new support should be upon the basis of income level compared to needs, not on the basis of the number of incomes in the family.

In the debate about one and two earner families, the more serious and immediate problem of no-earner families is ignored. In June 1993, there were 215,000 married-couple families where neither parent was employed and 230,000 sole-parent families where the parent was not employed. These families contained a total of over 800,000 children. Among children aged less than five years, 21 per cent had no parent employed and 30 per cent had no parent with a full-time job. For all dependent children aged less than 25 years, 23 per cent had no parent employed full-time.

Particularly at risk because of unemployment are some families from non-English-speaking backgrounds (unemployment rates for Vietnamese and Lebanese born settlers are consistently higher than other groups), members of small recently arrived communities, and those arriving as refugees (Moss 1993). Overall, refugee families are especially disadvantaged in the labour market as they face special problems stemming from their status as refugees (unpreparedness for departure, experiences of trauma and torture and disruptions to education and working life) as well as 'normal' problems of adjustment (Iredale and D'Arcy 1992).

In the context of high levels of unemployment among families with children, it is disappointing that the Committee on Employment Opportunities (1993), while containing some positive suggestions about the current unemployment crisis, pays little attention to the specific needs of unemployed families with children. These families are much less likely to be able to sell up and move to a new area than single people or couples without children. This is because families with children will be more likely to have invested heavily in their own housing. The Australian Living Standards Study has shown that the investment in housing is the leading factor which would prevent families moving to a new area in search of work. Furthermore, families with young children often require the assistance of extended family members living nearby. In recommending major extension of training programs for the long- term unemployed, the Committee also does not consider the child care needs of couples with children.

The only suggestion made by the Committee which specifically relates to couples with children is the introduction of a new parenting allowance which would apply to the unemployed and to low wage-earner families. We are concerned that this allowance would provide additional disincentives for women in these families to seek part-time work. The Committee itself notes that where a woman with an unemployed spouse earns less than $160 per week the couple will be worse off under this proposal. A common pattern nationally is for mothers to return to work for less than 15 hours a week when their children are young, increasing their hours as their children grow older. Withdrawal of a parenting allowance plus the withdrawal of benefits such as the Home Child Care Allowance and the Additional Family Payment would not make it worthwhile for a mother in a low-income family to work, unless she were to work full-time or substantial hours of part-time work. The same women, having been discouraged to work while their children are young would then, under the proposal, be expected to seek full- time work when their youngest child reached age 16 years. This approach seems to be totally counter to the preferred work patterns of most mothers.

CULTURAL DIVERSITY

The Aboriginal population in Australia, though relatively small in numbers, is an essential element of the Australian national identity. The proportion of the population identified in the 1991 Census as Aboriginal was 1.4 per cent. The age structure of the Aboriginal population is quite different from the total Australian population, being much younger. This has implications for family formation and structure in the future.

The general limitations of Census data for describing family diversity have been outlined above. Additional problems in relation to Aboriginal families are lack of relevance of some questions, language difficulties and sensitivity of some topics in Aboriginal society (ABS 1991). The ABS National Aboriginal and Torres Strait Islander Survey being carried out in 1994 should provide a better picture of diversity amongst Aboriginal families than we have had to the present.

A dominant characteristic of Aboriginal families is a sense of kinship which extends beyond the non-Aboriginal concept of a nuclear family (Walker 1993). Traditional kinship structures have been difficult and sometimes impossible to maintain in the face of white colonisation, years of oppression and the removal of children from their parents. However, the strength of family affiliation has had much to do with the preservation of a distinct Aboriginal culture in the face of extreme disadvantage (Walker 1993). The 1991 Census indicated that Aboriginal people tend to live in larger households than the population as a whole and that only a very small percentage live alone. (The Census definition of an Aboriginal household includes households where there are non- Aboriginal people present also.)

Post-second-world-war migration from Europe began a major shift in the composition of Australia's population, which still continues. Since 1947, the initially high proportion of migrants from English- speaking countries has fallen, and the proportion from continental Europe has risen, with recent increases from New Zealand and the countries of Asia. Some have come as part of a family, others as individuals who later formed families. In most cases, they left behind significant members of their families; in some cases they have had to leave everyone behind.

It is not appropriate in this article to explore in any detail the place family has played in government migration policy. However, it is worth noting that, particularly in the immediate post-war years, the rationale for migrant family intakes had much more to do with increasing the pool of workers and consumers than in promoting family life.

Although there was considerable rhetoric in the 1950s and 1960s concerning the desirability of family migration to Australia, racial preferences frequently determined immigration policies and the result was a favouring of British and northern European migrants over southern European (Kunek 1993). For example, the Dependants' Nomination Scheme was much more restrictive for women from Greece and Italy than from northern European countries. It was only after large groups of single men began to be seen as a 'problem' that single female migration from southern Europe was promoted. The fact that the situation had been created by a specific policy of gender and race selection was not often acknowledged.

Later on, regulations concerning family reunion became more relaxed generally, then were tightened as economic conditions deteriorated in the 1980s. Families play a very important part in the immediate and longer-term settlement process and attachments and obligations towards absent family members are often keenly felt. One issue in debates concerning family reunion has been different cultural expectations and norms about what constitutes close family ties and family obligations.

The 1991 Census indicated that of Australia's population of 16.8 million people, over 3.8 million (23 per cent) were born in 224 countries, while a further 3.3 million had one or both parents born overseas. Approximately 13 per cent of people living in Australia were born in a non-English- speaking country. Despite the relative increase of migrants from non-English-speaking areas, the United Kingdom remains the most significant source country for migrants to Australia. Its importance overall, however, is declining (from 20.9 per cent of migrants in the period 1981-85 to 14.8 per cent in 1986-1991). (Unless otherwise indicated, information in this section is taken from Australia in Profile, an ABS summary of the 1991 Census.)

Census questions relating to cultural diversity are country of birth, parents' birthplace, language spoken at home, proficiency in English, and religion. They can be very useful in determining overall trends and for general planning, but they are only a starting point for the total picture of family cultural diversity in Australia. They do not identify later than second generation migrant families, a growing proportion of the Australian population. Nor can they portray the culturally diverse backgrounds of many overseas- born settlers; for example an adult woman born in Eastern Europe who came to Australia via Central America, having migrated there with her family as a child; a Sunni Muslim whose family had lived in Beirut for 100 years, but were originally from Damascus.

In summarising the outcome of intermixing in Australia, Price (1993) predicts that by the turn of the century, well over 40 per cent of Australia's population will be ethnically 'mixed'. He estimated that by 1988 the Anglo-Celtic element made up 48 per cent of the population, the non-Anglo-Celtic element about 22 per cent and the mixture between the two, the remaining 30 per cent. If mixing within the major non- Anglo-Celtic groups were added to this 30 per cent, the result was an overall 'ethnic mix' of at least 37 per cent. While it must be said that these measures of ethnic mixing tell us little about 'Australianness', Price's analysis challenges notions of homogeneity and needs to be acknowledged.

Cultural diversity brings diversity of values and attitudes, which is perhaps the greatest challenge for a heterogeneous community. However, cultural influences are complex and dynamic. They constantly interact with class and gender and may change with time and circumstance. Policies based on simple conceptions of migrant family adjustment and adaptation, based on the response of 'traditional' family structures to a society with 'modern' family structures, are no longer appropriate, if they ever were (Morrissey, Mitchell and Rutherford 1991). People migrate from societies which are changing at a more or less rapid rate; they come from the same geographical space but with very different experiences based on class and gender; their experiences of migration inevitably bring further change; they interact with others in Australia who themselves are participating in, adjusting to, or resisting in varying degree, family and social change.

LANGUAGE DIVERSITY

Language is an important element of cultural and personal identity and first-generation parents are frequently concerned to have their children maintain the language of their heritage. Public policies favouring multilingualism did not become accepted until the 1970s, following a long period of positive discouragement for children to learn or speak anything but English.

The form of the main language question asked in the Census (which refers to language spoken at home) probably gives a slightly inaccurate picture of multilingualism as it does not pick up people who may regularly speak the language of their cultural heritage in some other context, for example, at parents' or grandparents' home or in a business context. Nevertheless 1991 Census figures show that around 15 per cent of the population speak a language other than English at home. Italian, Greek, Cantonese and Arabic are the most widely spoken languages in the home after English.

Whether the children of parents born overseas maintain the language of their parents depends on factors such as time spent in Australia, and cultural similarities to the Australian-born population and social structures within different cultures (ABS 1993). Clyne (1991) showed how other factors, such as settlement patterns, government policies, shifts in nationalist attitudes and 'ethnic revival', are also important in language maintenance.

There appears to be a fairly wide acceptance of second language teaching in schools amongst English-speaking background, as well as non-English-speaking background parents (Clyne 1991). More generally, Clyne argues that bilingualism is now recognised as a social good and a personal asset. He suggests that we have reached a point where multiculturalism 'affords legitimation and rights to all languages used in Australia. Whether a language is maintained as part of ethnic awareness or for its own sake in a family, one's 'Australianness' is not called into question.'

The other side of the coin is providing opportunities for new settlers (and long established Australians of migrant background) to learn and become proficient in English. Lack of proficiency in English may well be the single most important factor contributing to educational, economic and social disadvantage, for young people and for their parents. High levels of unemployment in some migrant communities are partly the result of lack of government foresight in ensuring that opportunities for becoming proficient in English were provided, in and out of the workplace.

RELIGIOUS DIVERSITY

Religion may be a strong determinant of the values which are transmitted through families. The 1991 Census provides some indication of religious affiliation in Australia, but the information has limitations for determining the impact of religious diversity. The question on religion is voluntary and in 1991, 10 per cent of persons did not state their religion. Nor does the Census measure the degree of commitment or involvement in a religion.

Australians overwhelmingly identify as Christian (74 per cent of the total population). The three largest Christian denominations were Catholic, Anglican and Uniting, accounting for over 80 per cent of all Christians. The number of Buddhists and Muslims has grown considerably in the past ten years, reflecting recent migration from Indochina and the Middle East particularly. Even so, in 1991, the two religions represented only 0.8 per cent and 0.9 per cent respectively of the Australian population. Around 13 per cent of the total population indicated that they had no religion. (The figure in 1981 was just under 11 per cent, the increase being greater for younger age groups.)

RESPONSES TO CULTURAL DIVERSITY

The two statements below (quoted in Department of School Education, Victoria 1992) illustrate changes in responses to cultural diversity at the Commonwealth Government policy level. In 1969, the then Federal Minister for Immigration, The Hon. Bill Snedden, said:

We have a single culture. If migration implies multicultural activities within Australian society, then it was not the type Australia wanted. I am quite determined we should have a monoculture with everyone living in the same way, understanding each other and sharing the same aspirations. We do not want pluralism.

Twenty years later, the Federal Government's National Agenda for a Multicultural Australia contained the following statement.

The National Agenda . . . is an attempt to redress the historic failings and, just as importantly, to facilitate the processes of continuing adjustment in the future. Australia's population will continue to change and we need to create an attitudinal and institutional environment that can accommodate those changes - so that the rights of the individual are recognised and the interests of the community advanced. In this sense, multiculturalism affects and serves the interests of all Australians.

The change of emphasis has inevitably meant differences in the settlement experience and subsequent life in Australia for individuals and families arriving over the past 20 years. It would be gratifying to say that the extreme ignorance and lack of sensitivity to cultural difference has gone, but we know that this is not so. Legislative protections and legal sanctions against discrimination have been put in place and there is much wider understanding and acceptance of difference in the community.

However, people from different cultural backgrounds still suffer greatly through lack of understanding and ignorance as a result of government and individual action or inaction. Nor are they and their experiences adequately represented in a range of social structures, from participation in the different levels of government to realistic portrayal in the mass media.

Diversity is challenging. On the one hand, we need to ensure that policies and support services are appropriate for all families, not just those who fit narrow social and cultural criteria. In many ways, we have only gone a short way towards such a goal. There are some ethno-specific services, and increased sensitivity in some 'mainstream' services to differences in needs. However, the scope for recognising needs and providing a variety of support services is great.

For example, lack of sensitivity to different cultural practices concerning pregnancy and childbirth make the experience unsatisfactory for some women and more frightening and confusing than it needs to be. The powerlessness experienced by women suffering domestic violence is frequently exacerbated for women from non- English-speaking backgrounds. They may have little or no English, be isolated by cultural attitudes to women's roles in families, have poor opportunities for employment, be confronted by racism and prejudice in their daily lives and have limited access to information and support targeted at those from non-English-speaking backgrounds (Moss 1993).

Equally, we need to be aware of how child care, family relationships, schooling, workplace training, domestic violence, care of the aged and a range of other issues, can be made much more difficult for families by the failure of services and agencies to recognise cultural diversity.

On the other hand, there are undoubtedly limits to community tolerance of cultural diversity which need to be clearly stated. An example is female genital mutilation. In general, where limits are set, legal sanctions need to be accompanied by community education, a point emphasised by some commentators in recent public debates of the issue. The Law Reform Commission has tackled some difficult questions in its report on multiculturalism and the law (Law Reform Commission 1992). Australia's signing of various international conventions outlining individual rights guided some of their recommendations in areas where there is clear cultural division.

The challenge is to recognise the commonalities between families and individuals across cultures, as well as the uniqueness of different cultural backgrounds.

DYNAMISM OR DIVISION?

This paper outlines the many faces of Australian families but argues that diversity is much more than simply a matter for recognition and celebration. Recognition of diversity is an essential first step in gaining an understanding of Australian family life, but unless that understanding is translated through policy into real improvements in the lives of all Australians, there is no cause for celebration. We may recognise the face of the unemployed or socially isolated family, but few people wish to celebrate its existence.

Social diversity can be a source of dynamism or of division. With the obvious exception of the disenfranchising of the Aboriginal people, the Australian experience overall has been more one of dynamism than division. However, in periods of economic insecurity such as we have been experiencing in recent years, blame is all too easily attributed to some of the faces, particularly those of more recent origin. Rather than a dynamic society ever moving forward, all too often the catch-cries are about a return to some mythical golden age of the past, to one set of 'traditional or basic' values and to the one face that goes with it. At the same time, the everyday realities and diversity of people's lives make it obvious that a return to the past is about as likely as a return to our former belief that our earth is flat. But we must be very vigilant as it would be relatively easy to lose some of the gains we have already made in affirming diversity.

Responding to diversity can expand and extend definitions of family and how they function. There is a substantial body of research focusing on sole-parent families and stepfamilies, but the standard Western nuclear one- generation family with its relatively narrow definition of close kin remains the underlying family model for research and for development of family policy.

Applying this narrow model to families from quite different cultural backgrounds can have very negative consequences. At best, it ignores what is important for those families; at worst, it can place at risk fundamental aspects of a culture and destroy individual lives, as in the case of Aboriginal families.

Diversity renders policy making more difficult because a single formula with a single set of parameters will not be fair and just for all. We cannot base policy upon some single past or future, idealised notion of family life which is at variance with present reality. For example, in the policy area of children's services, we must base policy on a recognition that there may be two, or one, or no parent in paid employment, because substantial proportions of children live in each of these three situations.

To be fair and just to all children and all parents, policy must take account of all the main social and cultural settings in which children live. That is, we should attempt to do the best by children within the particular parental context in which they live. On the other hand, we cannot have mainstream policies which are so diverse as to be tailored to the individual case. The challenge then is to design policies for families which have the best possible balance between recognition of diversity, fairness, and administrative efficiency.

REFERENCES

• ABS (1991), Census 86 - Australia's Aboriginal and Torres Strait Islander population, Catalogue No. 2503.0.
• ABS (1993), Australian in Profile. 1991 Census, Catalogue No. 2821.0
• Australian Catholic Bishops' Conference (1993), Families: Our Hidden Treasure, A statement on family life in Australia, Aurora Books, Melbourne.
• Clyne, M. (1991), 'Monolingualism, multilingualism and the Australian nation', in C.A. Price (ed.) Australian National Identity, The Academy of the Social Science in Australia, Canberra.
• Committee on Employment Opportunities (1993), Restoring Full Employment, A discussion paper, AGPS, Canberra.
• Department of School Education, Victoria (1992), Combating Prejudice in Schools: The Middle East in Focus.
• Edgar, D. (1990). 'A decade of family research - and the next', Family Matters, No.26, April, pp.2-11.
• Iredale, R. and D'Arcy, P. (1992), 'The continuing struggle: refugees in the Australian labour market', BIR Bulletin, 7, pp.7-8.
• Kunek, S. (1993), The Brides: A Multimedia Installation, Kunexion, Melbourne.
• Law Reform Commission (1992), Multiculturalism and the Law, Report No.57, Commonwealth of Australia.
• McDonald, P. (1991), 'Migrant family structure', in K. Funder (ed), Images of Australian Families: Approaches and Perceptions, Longman Cheshire, Melbourne.
• McDonald, P. (1993), Family Trends and Structure in Australia, Australian Family Briefings No.3, Australian Institute of Family Studies, Melbourne.
• Morrissey, M., Mitchell, C. and Rutherford, A. (1991), The Family in the Settlement Process, Bureau of Immigration Research, AGPS, Canberra.
• Moss, I. (1993), State of the Nation: A Report on People of Non- English-Speaking Backgrounds, Federal Race Discrimination Commissioner, Commonwealth of Australia, Canberra.
• Price, C. (1993), 'Ethnic intermixture in Australia', People and Place, Vol.1, No.1, pp.6-8.
• Walker, Y. (1993), 'Aboriginal family issues', Family Matters, No 35, August, pp.51-53.