Abstract
This paper will discuss the place of concepts and practices of assimilation within liberal strategies of government generally, as the foundation for understanding the ways in which they were mobilised in relation to Indigenous Australians. The overall argument is a dual one: to see particular logics of assimilation as integral to liberal government, and also to remain sensitive to what was specific about the assimilation of Indigenous Australians as a particular example of the exercise of political power under settler-colonialism.

It appears to me that we are paying altogether too much....it is a lamentable state of affairs that we should practically have a camp - which, to all intents and purposes, appears to be white - of children who are brought up in laziness and vicious surroundings, when they ought practically to be merged in the general population of the state. It has been the policy of the board not to allow children, many of whom are almost white, who have been removed from camp life to return thereto, but to eventually merge themselves in the white population, and thus cease to be a charge upon the state.2


It is not a question of stealing the children, but of saving them.3

Introduction
A strong linking thread running through almost every discussion of assimilation, supportive and critical, throughout the history of the idea in all its forms, consists of a set of presumptions about what we can call `the already-assimilated', that which Aboriginal people (and later, migrants) were to be assimilated into. If we look at public statements supporting assimilation, for example, we see phrases like: the Australian community, a European way of life, the white community, the wide[sic] community, the balance of the community, the general community, the Australian way of life, other Australians. Diversity and difference seems only to exist within the `to-be-assimilated' Other: the `already-assimilated' Self was characterized, in relation to `culture, religious beliefs, hygiene, standards of living and conduct', by uniformity, unity, sameness, consistency and coherence.
Our understanding of assimilation, I would like to suggest, is deeply flawed to the extent that we neglect to problematise this aspect of its perception, for the compelling reason that it is simply wrong. We all know it to be wrong: any analysis of the social fabric, structure and dynamics of non-Indigenous societies will have at least something to say about the lines of conflict which divide their `community', as well as the strategies and practices of power mobilised precisely to constitute and construct whatever gets understood as `the white community'. To the extent that monarchs, churches, political elites, legal systems and governments at every level have exercised authority over their populations at all, the idea of `a European way of life' or `the white community' is highly problematic, probably best understood as a rhetorical device employed to create the very thing it is meant to describe.4

The cultural logic of child removal
The example of the deeper roots of practices normally associated with assimilation which I would like to draw on here is the NSW Government's introduction of the Aborigines Protection Amending Bill in November 1914 and January 1915 in order to expand the power of its Aborigines Protection Board, in particular to grant it legal guardianship of all Aboriginal and part-Aboriginal children. The primary purpose of the legislation seems to have been, over the longer term, to empty the camps of part-Aborigines by removing `the rising generation' of part-Aboriginal children and `merging' them into the white population, thus saving the State the costs incurred by such `camps'. Along with the companion legislation in the other States and the Northern Territory, it is one of the central examples of policies and practices of assimilating Indigenous people into the non-Indigenous population, one which continued, as we all know, in spite of shifts in the accompanying rhetoric, until around the 1970s.

The problem which seemed to face the Aborigines Protection Board and the NSW Government was that although they wanted to present their interventions into the lives of part-Aboriginal children as more or less equivalent to interventions founded on neglect or abuse into the non-Indigenous population, in fact the same categories, concepts and legislative powers did not appear to enable them to remove as many part-Aboriginal children as they wanted. Despite all the best attempts to explain to mothers the benefits of the `opportunity' of being removed, `the almost invariable experience has been that the parents or relatives have raised some frivolous objection and withheld their consent'. Proceedings under the Neglected Children & Juvenile Offenders Act, complained Mr Flowers, `have proved unnecessarily cumbersome and ineffective'.5 This was because mothers would take their children across the border into Victoria, and also because `the difficulty of proving neglect where children are fairly clothed and fed is insurmountable'.6 What the Board wanted was the capacity to remove children because it assessed their `environs' as being `as bad as possible', subjecting these poor unfortunates to `the dangers of indifferent and immoral surroundings', dangers which went over and beyond the understandings of neglect or abuse which sufficed to bring non-Indigenous children into care under the existing legislation. JH Cann, the Colonial Secretary, put it this way: `if the aboriginal child happens to be decently clad or apparently looked after it is very difficult indeed to show that the half-caste or aboriginal child is actually in a neglected condition, and therefore it is impossible to succeed in the court'.7

Mr McGarry, the Member for Murrumbidgee, protested that it was cruel to separate any child from their parents; in the name of `civilisation' we had already `over-run their country and taken away their domain,' and taking Aboriginal children away was simply to compound the injury. McGarry then warmed to his theme by declaring that he had `some knowledge and experience of the operations of boards in boarding out children of the white race,' and that his knowledge and experience led him to the conclusion that he never wanted `to see the cruelties perpetrated upon the aborigines which years ago were perpetrated upon the children of white parents'.8 His own experience of the State's interventions into the lives of white children had revealed `many shocking instances of the treatment meted out by mean people in the interior to children boarded out with them'. He was in favour of `instilling the ideals of citizenship' and `teaching the methods of production', but, said McGarry:

....that can only be done by allowing the child to remain with the parent. Improve the parent if you can, but you will never improve the child by taking it away from the parent. Would hon. members advocate that with the white race?9

Indeed, as G Black added, many part-Aboriginal children borne of young Aboriginal mothers were a result of precisely this sort of removal of Aboriginal girls from their families to situations under the power and at the mercy of a variety of white station hands, sons of station owners, and so on. He also argued for interventions which kept children with their families.10 But J Storey was uncompromising:

We have truant inspectors always carefully watching to see whether white parents do the right thing by their children, and if in their opinion the parents are guilty of immorality or neglect, the children are forcibly taken away from them, and possibly they do not see them again for some years. Surely the same thing should apply to the black child; if she is unable to look after herself and her parents are unwilling to do so, it is obviously the duty of the board to take action.11

Mr Scobie countered by suggesting that the Bill essentially meant the reintroduction of slavery, despite his agreement that part-Aborigines needed to be `absorbed' into the white population. Cann then fell back onto the rhetoric of profligacy, immorality and degeneracy, having fore-armed himself with a number of supposedly illustrative `cases': an 11-year old boy with no mother and a `profligate' father who let him wander the streets `like a stray dog'; a family of three children, 7, 9, and 11, parents always quarrelling, exercising no control over them and allowing them to `meander about the camp like mangy dogs'; a deserted mother leading `an immoral and profligate life'; two children, 1 and 4, with the mother `undoubtedly leading an immoral life'.12 Despite protests from Scobie that the existing child welfare legislation must surely encompass any relevant case of abuse or neglect, the House, Cann declared, `ought readily to accede to the proposal, and give these poor little ones a chance',13 as indeed it did.

What is striking about this exchange is that although there is clearly a very particular piece of legal machinery being constructed which concentrates on part-Aboriginal children in a way which never applied to white children, one actually has to work fairly hard to find those aspects of its underlying cultural logic which are specific or unique to Aboriginal families and children. Accusing mothers of immorality, promiscuity and poor hygiene, fathers of low intelligence, profligacy, alcoholism and violence, children of `wandering', laziness and irregular habits,14 was all stock in trade for the moral discourse surrounding child welfare in the 19th century, and both supporters and opponents of the Bill framed their remarks in terms of the correspondences between this piece of legislative machinery and its equivalent for non-Indigenous children. The comments of observers like McGarry and Storey indicate that what was being invented here was not something entirely new: there was already a particular social technology in place to deal with problems of social discipline among the degenerate convicts and working classes, and this was what was being turned to achieving the goal of `emptying the camps'.15

Liberalism, civilization and governance
The idea of `rescuing the rising generation' had been central to European church and state agencies' policies in relation to the children of the poor and the working class since the 16th century, and it was a central element of the modern State's conception of the intersection of family life and liberal citizenship. The idea of `assimilation' underlying the removal of part-Aboriginal children was thus based on pre-existing philosophies, policies and institutional practices concerning unacceptable, `problem' populations in all the Western European countries and their colonies. It is possible, then, to chart the parallels and affinities between the racism of removing part-Aboriginal children for their Aboriginality, and the class ideology underlying the removal of non-Indigenous children for the `immorality' and `viciousness' of their impoverished surroundings. Indeed, the foundations of class ideology can be traced to ideas of race and racial difference, particularly in the idea of `blood' and its `inheritance'. In many respects, the anxieties concerning the rapid reproduction of part-Aboriginals on the border between white and black cultures followed a similar logic of assimilatory governance to that underlying the fears of the equally sexually dangerous and prolific non-respectable working-class, especially women, on the fringes of the metropolis. This is not a claim that there was some sort of equitable distribution of state violence between Indigenous and non-Indigenous families, and that all forms of assimilation are the same. Its meaning is more that assimilation can usefully be understood as referring to a broad diversity of policies and practices, applying to a range of target populations, which signal a certain degree of isomorphism between `race,'`class' and `gender'.16 The interesting question then becomes one of understanding the relationships between these diverse policies and practices and their particular effects on different target groups.

More broadly, the development of European societies have always involved what Talal Asad has called `the reformation of subjectivities and the reorganization of social fields in which subjects act and are acted upon'.17 This reformation of subjectivity involved an `implantation' of `the desire for progress' that was based on the employment of various political powers - if necessary, violence18 - which were in turn organised around particular discursive strategies. Zygmunt Bauman has also identified the more general features of what he describes as the `assimilatory project' within European state formation, and the centrality of that project to the very nature of the modern state.19 The assimilatory project was part and parcel of the process of dismantling older, deeply rooted forms of communal life which provided alternative, sometimes oppositional frameworks of social power. Assimilation, he suggests, `was an exercise in discrediting and disempowering the potentially competitive, communal or corporative sources of social authority'.20 As part of the liberal political and legal program to secure the modern state's `monopoly of law-making and coercion',21 assimilation was organized around a tolerance of individuals based on a profound intolerance of differing collective cultural identities, so that `tolerant treatment of individuals was inextricably linked to intolerance aimed at collectivities, their ways of life, their values and, above all, their value-legitimating powers'.22 The price to be paid by individuals for entry to liberal citizenship in the modern state, at least in its juridical form, has always been to leave all their previous communal cultural identities behind, apart perhaps from some remnant in the form of quaint customs wheeled out at ceremonial occasions.
One aside comes to the surface here: the fact that Bauman's discussion of assimilation in the European context engages with the experience of Jewish peoples prior to as well as including the Holocaust also suggests an observation on the way in which the Australian experience of colonialism has tended to be compared with the European experience. The concentration on the concept of `genocide' has generally led this comparison to revolve around the establishing the linkages between Australian practices of colonization with the genocidal annihilation of Jewish, Gypsy and otherwise `marginal' peoples in the Holocaust. However, it may be more meaningful in a broader sense to compare the assimilation of the Aboriginal people in Australia, not with the Holocaust, but with the processes and structures of assimilation of `outsiders' and `strangers' in the centuries prior to the Holocaust itself. It has become a truism in Holocaust studies that the event can only be understood in its context, as the result of a long history and the outcome of long-term processes - this idea can also be useful applied to the question of how we should compare settler-colonial (Australian) and metropolitan (European) versions of assimilation and the flattening-out of cultural diversity.

In fact, of course, liberal models of individual rights can never really detach themselves from an accompanying conception of `society as a whole' to which individuals are to be `assimilated'. What happens is that the rhetoric of liberal democracy pulls our attention away from the models of society and community which are in fact being drawn upon, making their problematic effects that much harder to perceive, let alone respond to. The pursuit of egalitarianism is in itself no guarantee of cultural diversity; indeed, it is arguably precisely the reverse, as Christopher Lasch suggested:

The rise of egalitarianism in western Europe and the United States seems to have been associated with a heightened awareness of deviancy and of social differences of all kinds, and with a growing uneasiness in the face of those differences - a certain intolerance even, which expressed itself in a determination to compel or persuade all members of society to conform to a single standard of citizenship. On the one hand, egalitarian theory and practice insisted on the right of all men (and logically of all women as well) to citizenship and to full membership in the community; on the other hand, they insisted that all citizens live by the same rules of character and conduct.23

When the model of `community' with which liberalism is combined is an organic, mono-cultural, and unitary one, the forms of governance which results have a strongly normalising edge to it which can, in situations where the boundaries between the `normal' and the `pathological' communities are drawn strongly enough (as with racial divisions), have effects very similar to more authoritarian regimes based on quite different political philosophies. This is why the concept of `genocide' has had such appeal in capturing the underlying logic of the assimilatory project.24

There is in fact a powerful tension at the heart of liberal understandings of individuals and their place in society, between, say, `the best interests of the child' and `the best interests of society'. If we simply assume that the two work in harmony, the former will almost always be defined in terms of the latter. For example, Uday Mehta25 argues there are intimate internal bonds between liberalism's theoretical claims to universalism and its actual practices of political exclusion that are based on a particular `anthropology' concerning what it means to be `human'.26 Mehta speaks of `a thicker set of social credentials that constitute the real bases of political exclusion'27 and qualify, heavily, liberalism's claimed universalism. What is then generated is a practical particularism which allows the two to exist side by side without the contradiction causing too much embarrassment. In other words, the liberal commitment to individual freedom, privacy and autonomy depends for its successful realisation on the parallel formation of behaviour and conduct according to a particular understanding of what constitutes human `welfare',28 and this formation of conduct in turn rests on a certain `anthropological minimum'.29 An important part of the story of liberal political rationality then becomes the securing of that anthropological minimum, both among the population generally and more specifically in relation to Indigenous populations under colonial conditions.

Now, the argument I have been working towards in my own research into the social and political rationalities surround part-Aboriginal child removal30 has been that it is important to understand those collective self-images, and the boundaries drawn around colonizers and colonized, as having generally being organized around the concept of `civilization' and the pre-exising processes of `assimilation'/`civilization' of the non-Indigenous population. Liberal universalism is articulated through the filter of `civilization', so that the freedoms, rights and liberties of liberal citizenship are made conditional on effective assimilation, on acceptance into the forms of civility recognized by various, sometimes competing, authorities: Church, state, local communities, schools and welfare agencies. Civilization constitutes the criteria for exclusion from liberalism's freedoms, rights and liberties, but also the reference point for those on-going projects aimed at the assimilatory (re)shaping of conduct permitting eventual re-entry into `the wider community' and its attendant identity as a free liberal agent, a modern citizen. The history of Australian liberal governance then needs to be understood as running alongside a parallel history of Australian civilization.

Conclusion
I would like to conclude with an observation that Bob Connell has made31 in relation to the topic of the removal of part-Aboriginal children: given the `power over bodies' that liberal states regularly exercise over their populations, moving especially the lives and limbs of children around as it suits differing strategies of governance, the important analytical question is not `why did it happen?', but `why did it stop?' - which makes possible another question: `has it stopped?'. Instead of taking as `normal' or as a `starting point' what we think to be the current political situation of a critical attitude towards assimilationist ideals, it may be far more useful and politically productive to see liberal government as inherently concerned to `assimilate' all of its diverse populations to particular (changing) modes of conduct, habits and being. By seeing what projects of assimilating Aboriginal people to a supposed `white community' have in common with the contested construction of that white community itself, it will be possible to gain a much more effective understanding of how the two types of project differ from each other,32 as well as the directions we might like to nudge them towards in the future. It might be better to ask, not `what do want, assimilation or something else?', but `what form do we think the current assimilatory project should take?'. Instead of alternating between being subjugated by and resisting the central rhetoric of assimilation, perhaps it is a matter of seizing it.

Notes

1 Thanks to Robyn Arrowsmith for undertaking the bulk of the research on which this paper is based, as well as the University of Sydney's Research Institute for Humanities & Social Sciences, through its Extended Research Secondment scheme, and the Australian Research Council, for their financial support of that research and its writing-up. I am also grateful to Tim Rowse and Virginia Watson for their comments and observations on various aspects of the larger project of which this paper is a component. Parts of the paper are also drawn from two earlier publications: `The barbarism of civilization: cultural genocide and the `stolen generations'', British Journal of Sociology 50(2) 1999: 295-313 and `The "stolen generations": on the removal of Australian indigenous children from their families and its implications for the sociology of childhood' Childhood 6(3) 1999: 297-311.
2 H Flowers MLA, 24 Nov 1914, NSWPD 56 1914-15, p. 1354.
3 JH Cann (Colonial Secretary), 27 Jan 1915, NSWPD 57 1914-15, p. 1951
4 My discussion here resonates, then, with Tim Rowse's observation that discussions of Indigenous citizenship need to be accompanied by examinations of `the transformations in non-Indigenous citizenship....the constitution of `us', not only as citizens' but as colonists': `Indigenous citizenship' in Wayne Hudson & John Kane (eds) Rethinking Australian Citizenship Melbourne: Cambridge University Press, 2000, p. 87. On the question of the historical formation of subjectivity broadly, see Robert van Krieken,`The organisation of the soul: Elias and Foucault on discipline and the self' Archives Europeénes de Sociologie 31(2) 1990: 353-71, and Robert van Krieken,`Social discipline and state formation: Weber and Oestreich on the historical sociology of subjectivity' Amsterdams Sociologisch Tijdschrift 17(1) 1990: 3-28.
5 24 Nov 1914, NSWPD 56 1914-15, p. 1354.
6 Ibid.
7 27 Jan 1915, NSWPD 57 1914-15, p. 1951.
8 Id, p. 1953.
9 Id, p. 1954.
10 Id, p. 1958.
11 Id, p. 1960.
12 Id, p. 1966.
13 Id, p. 1967.
14 For example, the 1915 State Children Relief Board's Annual Report contained the following: `As a class such children are preternaturally lazy and unreliable and require almost constant driving to obtain any good from them'; `Coloured children from camps...are usually grievously neglected; have been allowed to become exceptionally precocious in immorality and vice (or are at least quite non-moral); and these failings or propensities they bring with them into the Homes, and there the white children are not slow to detect them or to emulate them in spite of the closest supervision'; `Paternity is casual and conjectural, and promiscuous association is the rule; sanitation is ignored. Dirt is the dominating element. In this mire of moral and physical abasement, tended by semi- imbecile mothers, children are allowed to wallow through the imitative stages of childhood'; `Many of such children are so white that, were it not for their presence in camps or in association with blacks, the average individual would characterize them as practically normal. Beneath the skin, however, the taint is more marked, and it is in the correction of degenerate traits and the eradication of demoralised habits that the work of the expert psychologist and educationalist lies, in order that these children may be fitted by suitable training under approved conditions, to become assimilated with the white community without prejudice to their mutual interests'.
15 See Robert van Krieken, Children and the State: Social Control and the Formation of Australian Child Welfare Sydney: Allen & Unwin, 1992.
16 Ann Laura Stoler, Race and the Education of Desire: Foucault's History of Sexuality and the Colonial Order of Things. Durham: Duke University Press, 1995.
17 Talal Asad, `Conscripts of Western civilization' in Gailey, Christine Ward (ed.) Dialectical Anthropology: Essays on Honor of Stanley Diamond, Vol. 1: Civilization in Crisis: Anthropological Perspectives Gainesville: University Press of Florida, 1992, p. 337.
18 Id, p. 339.
19 Zygmunt Bauman, Modernity and Ambivalence Cambridge: Polity, 1991.
20 Id, p. 106.
21 Id, p. 111.
22 Id, p. 107.
23 Christopher Lasch, `Origins of the asylum' in The World of Nations: Reflections on American History, Politics & Culture New York: Alfred Knopf 1973, p. 17.
24 Human Rights & Equal Opportunity Commission, Bringing them Home: Report of the National Inquiry into the Separation of Aboriginal and Torres Strait Islander Children from their Families. Sydney: Sterling Press, 1997.<http://www.austlii.edu.au/rsjlibrary/hreoc/stolen/index.html>

25 Uday S Mehta, `Liberal strategies of exclusion' Politics & Society 18(4) 1990: 427-54.
26 See also Mariana Valverde's discussion of the `despotism' at the heart of liberal thought, particularly JS Mill: `"Despotism" and ethical governance', Economy and Society 25(3) 1996, pp. 357-72, and Barry Hindess's (1998) discussion of the politics of citizenship across national boundaries: `Divide and Rule: the international character of modern citizenship', European Journal of Social Theory 1(1) 1998, pp. 57-70, as well as more generally Philip Pettit, Republicanism: A Theory of Freedom and Government, Oxford: Clarendon Press, 1997, pp. 241-70 and Nikolas Rose Powers of Freedom: Reframing Political Thought, Cambridge: Cambridge University Press, 1999, pp. 71-8.
27 Above n25, p. 429
28 Above n26, pp. 48-9.
29 Above n25, p. 431.
30 Above n1.
31 In personal correspondence.
32 How did the two projects differ? Some tentative suggestions:
1. the cultural differences being engaged with were greater and more deeply anchored in individual habitus, the outcome of completely distinct `civilizing processes'; difference of degree;
2. the political presence of Aboriginal parents was weaker and there were fewer advantages to effective governance associated with recognizing the autonomy of Aboriginal parents;
3. making it possible for the legal powers of intervention to be more extensive, and for it to take longer to wind back those powers, to `civilize' the state's and the church's interventions into their children's lives.