And what is law, after all, but a partially realized promise to overcome disorder and aggression, tame and domesticate force, and subject action and instinct to reason and will? (Sarat & Kearns 1992: 211-2)

We have to give attention to hygiene. So long as natives are not living in a way that makes them physically acceptable - to put it crudely, so long as natives live in a way that makes them smell - then there is no hope for them. We have to improve their hygiene in order to make them acceptable (Paul Hasluck, Commonwealth Parliamentary Debates 8 (New Series) 21st(1st) 6 October 1955: 1333)

Introduction: why do we need a theory of civilization?
It hasn't been a century for Occidentals to be proud of, when you think of the aspirations of Western liberals at its outset, the efforts directed at all manners of progress and improvement, and how much so many millions of people have ended up suffering, and continue to suffer, at each other's hands. The management of violence in all its diverse forms is arguably a problem of similar significance in the year 2000 as it was in 1900 - or 1800, 1700. Indeed, it could be said that it has simply become more complex and differentiated.

How should we understand the continuing problem of violence? Often we do so by turning to the idea of 'barbarism', and this is the first reason why we need a theory of civilization: the normative and rhetorical power of the opposition between barbarism and civilization continues to be enormous. Civilization remains an important conceptual and rhetorical reference point, the focus of whatever we understand by 'progress' or, if that word offends, the 'point' of any human action on the social and political world. Since the post-1989 enthusiasm for 'civil society', there has also been a movement in social and political thought towards an analysis of the processes by which societies become more or less 'civil', and this has much to do with the current appeal of the idea of civilization.

It is, however, precisely this normative usage of civilization which should alert us and make us uneasy. This approach to the idea of 'the civilizing of the world' is in fact firmly rooted in the history of colonization and imperialism; as Norbert Elias wrote in 1939 'It is not a little characteristic of the structure of Western society that the watchword of its colonizing movement is 'civilization'' (1994: 509). If we approach the period since 1989 as being related to the increasing 'success' of a particular model of society, politics and economics, in addition to analysing this development in terms of capitalism, liberalism, democracy or civil society, it is equally accurate to see in connection with a particular conception of civilization being promoted throughout the world along more or less the same lines as it has been pursued since the fourteenth century when Europeans began colonizing the world beyond Europe (Becker 1988). What we currently understand as globalization or the reinvigoration of civil society can thus be seen as entirely continuous with the Enlightenment mission of civilizing the world, the way in which we today perceive the 'barbarism' of the genocide in Rwanda, Kosovo and East Timor as intimately related to the way that Europeans understood the 'savagery' of American Indians in the eighteenth century.

The second reason behind the need for a theory of civilization is that it constitutes an often hidden dimension to citizenship and governance, which are generally qualified in terms of the differing degrees of civility displayed by those members of a liberal society hoping to enjoy the full range of its rights and entitlements, their differing capacities to meet the accompanying obligations (Pettit 1997: 241-70; Rose 1999: 71-8). The freedom which liberal political rationality promotes is a conditional one, accompanied by an equally strong concern with the production and maintenance of a configuration of shared norms, values, codes of conduct and rules of behaviour, pursued in the realms of social welfare, education, public health, psychology, psychiatry, counselling, religion, social work, the mass media, town planning and public architecture, all the activities characterizing what have been called 'civilizing offensives' (van Krieken 1990).

Just one obvious example here is the way in which the exclusion of Australian Indigenous peoples from citizenship was until only quite recently based on the extent to which particular norms of civility had or had not been achieved by particular individuals or groups - citizenship was not something to which anyone born in Australia was entitled, it was conditional on being able to demonstrate the requisite degree of civilization (Chesterman & Galligan 1997; Petersen & Sanders 1998), and the whole post-1945 assimilatory project of Australian liberalism can be seen as precisely a 'civilizing offensive' designed to bring both Indigenous and migrant Australians within a particular frame of civility, in order to render them proper and governable citizens.

Third, if civil society is seen as the source or foundation of the project of civilising the state, it is important to recall civil society's own sources and foundations, and this endeavour can often lead one full circle back to the state. Civil society is not something "natural", but produced and developed, the subjects of civil society being the targets, since the sixteenth century, of a range of disciplinary and governmental strategies designed to produce them as appropriate citizens, rather than simply taking them, and whatever social forms they generate, as they are. The whole history of civil society and the state since Adam Ferguson has been that of a partnership, albeit one with up and downs, rather than a war of opposites. In addition, the field of law lies at the heart of this production of civil society as a realm of relative individual and communal freedom separate from the state. The distinction made in jurisprudence between civilized and barbaric societies has always been centred on the 'rule of law', on the existence of a system of abstract, rule-governed restraints on the exercise of violence and force, the rules of which have an existence of their own independent of whichever individuals or groups happen to dominate at any particular time - a 'government of laws, not of men'. To the extent that civil society concerns the restraint of violence and power, law has played a central role, especially in civilizing the state (originally, civilizing the King). Indeed, the construction and defence of the boundaries between civil society and the state is a large part of law's concern, and a society without a rule of law is one with no distinction between state and civil society.

The conceptual task which I would like to engage with in this paper, then, is a "knitting together" of the three ideas in the title: law, governance and civilization. Each of them on their own has been explored in enormous depth, and sometimes in relation to one of the other two, but my concern here is to work towards the 'triangulation' between all three.

Government, rule of law and civility
There is a relationship of mutual interdependence between the concepts and practices of government and citizenship on the one hand, and civilization on the other, which is in turn mediated by law:

Government/Citizenship

<---------> Law

Civilization

Law's role in mediated the connection between government and civilization arises, first, from its centrality to whatever we understand as civilization. As Dennis Lloyd has suggested,

[E]xperience has shown that law is one of the great civilizing forces in human society, and that the growth of civilization has generally been linked with the gradual development of a system of legal rules, together with machinery for their regular and effective enforcement. (Lloyd 1964: 7)

The object of law is to a large extent the absence of a sufficient degree of civilization, and legal reasoning is thus concerned both with identifying what constitutes civilization, and with going about its business in a suitably civilized way, consistent with its own standards of civility. Indeed, the 'rule of law' can be defined precisely as constituting, in a definitive way, civilization itself - to be civilized is to know and operate within the law (probably all cultures defined the relationship between law and civilization in this way). James Tully, for example, following Kant, describes representative government and the rule of law as the 'threshold institutions of civilisation' (1995: 80).

For example, in the Mabo judgement(Mabo and others v Queensland (No. 2) [1991-1992] 175 CLR 1), Justice Brennan argued that a law which

took from indigenous inhabitants any right to occupy their traditional land, exposed them to deprivation of the religious, cultural and economic sustenance which the land provides, vested the land effectively in the control of the Imperial authorities without any right to compensation and made the indigenous inhabitants intruders in their own homes and mendicants for a place to live

should be regarded as 'unjust', with a questionable claim to be part Australian common law, 'judged by any civilized standards' (pp. 28-9). A little further in the judgement, however, we are also reminded that 'civilization' was precisely the concept underpinning the legal reasoning refusing an indigenous interest in land. It was, Brennan J remarks, 'the supposedly barbarian nature of indigenous people [which] provided the common law of England with the justification for denying them their traditional rights and interests in land'; Lord Sumner is cited in his Privy Council judgement as proposing that:

The estimation of the rights of aboriginal tribes is always inherently difficult. Some tribes are so low in the scale of social organization that their usages and conceptions of rights and duties are not to be reconciled with the institutions or the legal ideas of civilized society. Such a gulf cannot be bridged. It would be idle to impute to such people some shadow of the rights known to our law and then to transmute it into the substance of transferable rights of property as we know them.

There was seen to be a sort of isomorphism between the approach to land and to family life; just as Aborigines were seen as bereft to rights to land because they did not cultivate it, and thus 'uncivilized', so too were they seen as bereft of rights to their own children, because they did not 'cultivate' them into a form of civilization recognizable to Europeans (Dorsett 1995). Just as it was the duty of Europeans to cultivate the land to its maximum capacity, so too it was their duty to 'cultivate', educate Aboriginal children to their 'maximum' capacity, that is, as Europeanised. Civilized society is in this usage, exactly what Aborigines are not part of, and it was this exclusion which supported the denial of their access to full citizenship, apparently leaving unchallenged the broader conceptions of egalitarianism and equity on which Australian national identity was supposed to rest.

The concept of civilization is also used to legitimate the idea of the rule of law in a more general sense, particularly in constitutionalism and the separation of powers idea, as operatively central to civilization, particular the restraint of violence and power, as it is to liberalism. What characterizes both civilization and the rule of law is the embedding of the exercise of power in impersonal rules and structures, rather than in the unregulated will of dominant individuals and groups, a 'government of laws, not of men'.

However, the form taken by individual subjectivity also plays a central role in determining the operation of different modes of government and law. For David Hume, it was the application of human effort to practical, productive labour which would achieve this effect:

Laws, order, police, discipline; these can never be carried to any degree of perfection, before human reason has refined itself by exercise, and by an application to more vulgar arts, at least, of commerce and manufacture. Can we expect, that a government will be well modelled by a people, who know not how to make a spinning wheel, or to employ a loom to advantage? (Hume 1987: 273)

A central focus of Western political theory has been to argue that human beings are not born reasonable, and that some process of cultivation, refinement, education or formation of 'public reason' - let us call it civilization - is a crucial dimension of a peaceful and productive civil society. Bauman thus argues that the project of civilization was centrally about the production of governable subjects:

.....the concept of civilisation entered learned discourse in the West as the name of a conscious proselytizing crusade waged by men of knowledge and aimed at extirpating the vestiges of wild cultures - local, tradition-bound ways of life and patterns of cohabitation. It denoted above all else a novel, active stance taken towards social processes previously left to their own resources, and a presence of concentrated social powers sufficient to translate such a stance into effective practical measures. In its specific form, the concept of civilisation also conveyed a choice of strategy for the centralized management of social processes: it was to be a knowledge-led management, and management aimed above all at the administration of individual minds and bodies. (Bauman 1987: 93)

Until the early 1700s, suggests Febvre, the concept of police had been regarded as most central to superior government, and cites Furetière referring in 1690 to 'police, laws, systems of conduct to be observed for the subsistence and government of states in general, in opposition to barbarity' (1998: 166), the defining feature of 'savage' societies was their lack of laws and a concept of police. However, in the course of the eighteenth century, there was additional concern for a concept which grasped 'the triumph and spread of reason not only in the constitutional, political and administrative field but also in the moral, religious and intellectual field' (Febvre 1998: 167), with the second effectively absorbing the first, and this concept became civiliser. By the end of the eighteenth century, it was 'civilization' which represented the form of human perfectability to which European societies drove both their own populations and as much of the rest of the world as they could lay their hands on.

What is this thing called civilization? Norbert Elias, civilizing and de-civilizing processes
Lucien Febvre suggested that it is possible to identify two main ways in which concept 'civilization' is used. It can have an 'ethnographic' meaning, referring simply to a way of life, a particular assembly of cultural, moral, political and economic forms. In this sense it is interchangeable with 'culture' or 'society', and is of little analytic utility. In the second usage, wrote Febvre,

when we are talking about the progress, failures, greatness and weakness of civilization we do have a value judgment in mind. We have the idea that the civilization we are talking about - ours - is in itself something great and beautiful; something too which is nobler, more comfortable and better, both morally and materially speaking, than anything outside it - savagery, barbarity or semi-civilization. Finally, we are confident that such civilization, in which we participate, which we propagate, benefit and popularize, bestows on us all a certain value, prestige and dignity. For it is a collective asset enjoyed by all civilized societies. It is also an individual privilege which each of us proudly boasts that he possesses. (1998: 161)

However, we can add a third consideration, namely those social and political conditions, practices, strategies and figurations which produce whatever ends up being called civilization, founded on a reflexively critical awareness of the way in which particular conceptions and experiences of 'being civil' get constructed and produced in one way or another. This was the concern of the German sociologist Norbert Elias.

Processes of Civilization
By the end of the eighteenth century, wrote Elias, civilization had come to be defined by Europeans 'simply as an expression of their own high gifts' (1994: 41). It became a crucial part of Europeans' sense of superiority over all other peoples in the world: 'the consciousness of their own superiority, the consciousness of this "civilization," from now on serves at least those nations which have become colonial conquerors, and therefore a kind of upper class to large sections of the non-European world, as a justification of their rule' (1994: 41). Elias organized his observations about the development of modern social life around this perception of themselves as particularly 'civilized', at the very hour of Europeans' engagement with in horrific barbarism, because he felt it went to heart of the constitution of the psychic structure characteristic of contemporary Western societies.

Elias believed that what we experience as 'civilization' is founded on a particular habitus, a particular psychic structure which has changed over time, and which can only be understood in connection with changes in the forms taken by broader social relationships. Elias insisted that 'the molding of instinctual life, including its compulsive features, is a function of social interdependencies that persist throughout life', and these interdependencies change as the structure of society changes. 'To the variation in this structure correspond, 'wrote Elias, 'the differences in personality structure that can be observed in history' (1994: 249). The first point was explored by Elias in relation to the successive editions of a variety of etiquette manuals, and the second in relation to the history of state formation in Britain, France and Germany.

The first volume of The Civilizing Process traces gradual changes in expectations of people's interpersonal conduct in European societies, as well as the way they approached their own bodily functions and emotions. In outlining 'correct' behaviour, Erasmus pointed to 'attitudes that we have lost, that some among us would perhaps call "barbaric" or "uncivilized",' and it spoke 'of many things that have in the meantime become unspeakable, and of many others that are now taken for granted' (1994: 44). Elias suggested that typical medieval conduct was characterized by 'its simplicity, its naïvete,' emotions were 'expressed more violently and directly' and there were 'fewer psychological nuances and complexities in the general stock of ideas' (1994: 50).

As time went by, he found that the standards applied to violence, sexual behaviour, bodily functions, eating habits, table manners and forms of speech became gradually more sophisticated, with an increasing threshold of shame, embarrassment and repugnance. Gradually more and more aspects of human behaviour become regarded as 'distasteful', and 'the distasteful is removed behind the scenes of social life'.

Again and again, wrote Elias, we see 'how characteristic is this movement of segregation, this hiding 'behind the scenes' of what has become distasteful' (1994: 99). For example, a French etiquette manual from 1672 suggested that in the past, 'one was allowed to take from one's mouth what one could not eat and drop it on the floor, providing it was done skilfully. Now that would be disgusting' (1994: 76).

Elias described medieval society as being characterised generally by 'a lesser degree of social control and constraint of instinctual life' (1994: 159), in particular by a violence which dominated everyday life and was rarely subject to much social or self-control. His interpretation of his evidence was that it suggested 'unimaginable emotional outbursts in which - with rare exceptions - everyone who is able abandons himself to the extreme pleasures of ferocity, murder, torture, destruction, and sadism' (1978: 248). Elias observed that there seemed to be great pleasure in killing and torturing, describing it as 'a socially permitted pleasure'; indeed, to some degree 'the social structure even pushed its members in this direction, making it seem necessary and practically advantageous to behave in this way' (1994: 159).

It was the social process of 'courtization' which subjected first knights and warriors, and then ever-expanding circles of the population (1994: 88), to an increasing demand that such expressions of violence be regulated, that emotions and impulses be placed more firmly in the service of the long-term requirements of complex networks of social interaction. Slowly and gradually, argued Elias, 'the code of behaviour becomes stricter and the degree of consideration expected of others becomes greater,' and 'the social imperative not to offend others becomes more binding' (1994: 64). In court society we see the beginnings of a form of mutual and self-observation which Elias referred to as a 'psychological' form of perception (1994: 63). Elias did not see courts as the 'cause' or driving force of this process, but as its nucleus, and he drew a parallel with the form taken by a chemical process like crystallization, 'in which a liquid...[being] subjected to conditions of chemical change.....first takes on crystalline form at a small nucleus, while the rest then gradually crystallized around this core' (1994: 95).

The result was a particular kind of habitus or 'second nature', an 'automatic self-restraint, a habit that, within certain limits, also functions when a person is alone' (1994: 113). Elias argued that the restraint imposed by increasingly differentiated and complex networks of social relations became increasingly internalized, and less dependent on its maintenance by external social institutions, developing what Freud was to recognize as a super-ego (1994: 154).

He did say that these developments in habitus were not unilinear, that 'the civilizing process does not follow a straight line' and that 'on a smaller scale there are the most diverse crisscross movements, shifts and spurts in this or that direction' (1994: 153). Nonetheless, at this point he felt that there was a more significant overall tendency with a particular direction, towards increasing 'regulation of affects in the form of self-control' (1994: 153). 'Regardless,' then, 'of how much the tendencies may criss-cross, advance and recede, relax or tighten on a small scale, the direction of the main movement - as far as is visible up to now - is the same for all kinds of behaviour' (1994: 154).

The second volume of The Civilizing Process dealt with the explanation of the transformation of psychic structure revealed by the etiquette books and other historical evidence. 'When enquiring into social processes,' he wrote' 'one must look at the web of human relationships, at society itself, to find the compulsions that keep them in motion, and give them their particular form and their particular direction' (1994: 288).

Among those changes in the 'web of human relationships' there was, first, 'the process of state-formation, and within it the advancing centralization of society' (1994: 269), especially as it was expressed in the absolutist states of seventeenth and eighteenth century Europe. Second, he stressed the gradual differentiation of society, the increasing range, diversity and interdependence of competing social positions and functions composing European societies. There were other, related changes which he also mentioned, such as the development of a money economy and urbanization, but it was these two processes of social development which he emphasised most (1994: 457).

There was, Elias believed, a powerful 'logic' built into any configuration of competing social units, such as states, towns or communities, towards an increasing monopolization of power and, correspondingly, of the means of violence. He saw this 'logic' as emerging from the dynamics of social, political and economic competition, and saw it as being organised around two 'mechanisms': the 'monopoly mechanism', which 'once set in motion, proceeds like clockwork' (1994: 343), and the 'royal mechanism'. The operation of the 'monopoly mechanism was that as social units competed with each other, 'the probability is high that some will be victorious and others vanquished, and that gradually, as a result, fewer and fewer will control more and more opportunities, and more and more units will be eliminated from the competition, becoming directly or indirectly dependent on an ever-decreasing number' (1994: 347). Unless some countervailing process is set in motion, argued Elias, competition would generally drive any human figuration towards 'a state in which all opportunities are controlled by a single authority: a system with open opportunities has become a system with closed opportunities' (1994: 347).

Accompanying the monopoly mechanism was another tendency, that of what Elias called the 'royal mechanism', which was a feature of the evenness or indecisiveness of any pattern of competition. If social conditions are not bad enough for any one group to risk the loss of their current position, and power is distributed so evenly that every group is fearful of any other group gaining the slightest advantage, 'they tie each other's hands' and ' this gives the central authority better chances than any other constellation within society' (1994: 397). The position of a central authority is not based simply on some greater power that they might have over any other social unit, but on their function as a mediator or nodal point for the conflicts between the other groups in society, which can neither individually overcome any of the others, nor stop competing to the degree required to form an effective alliance with each other.

The consequence of these mechanisms in terms of power relations was not, however, simply to increase the power-chances of those individuals and groups in more central positions of authority and influence, which is how we usually think of any process of monopolization. Elias emphasised that 'the more people are made dependent by the monopoly mechanism, the greater becomes the power of the dependent, not only individually but also collectively, in relation to the one or more monopolists'. This was because those in the more central, monopoly positions were also made increasingly dependent on 'ever more dependents in preserving and exploiting the power potential they have monopolized' (1994: 348). The greater monopolization of power-chances is thus accompanied by a greater collective democratization, at least, because a monopoly position is itself dependent on a larger and more complex network of social groups and units. Examples here would include the position of the head of government in any of the advanced industrial countries, or the managing director of a large corporation.

The state-formation process in Europe was accompanied, necessarily, by an increasing monopolization of the means of violence, and a pressure towards other means of exercising power in social relations. Rather than the use of violence, social 'success' became more and more dependent on 'continuous reflection, foresight, and calculation, self-control, precise and articulate regulation of one's own affects, knowledge of the whole terrain, human and non-human, in which one acts' (1994: 476). Elias argued that this 'rationalization' of human conduct, its placement at the service of long-term goals and the increasing internalization of social constraint was closely tied to the process of state formation and development of monopolies of physical force (1994: 447). The 'requirement' placed on each individual is not a direct one, but one mediated by one's own reflection on the consequences of differing patterns of behaviour (1994: 450).

Underlying the processes of state-formation and nation-building were also others of increasing social differentiation, increasing density, complexity, and what Elias called 'lengthening chains of social interdependence'(1994: 448). A central developmental process in European societies was their increasing density, produced by a combination of population growth and urbanization, and the ever-larger circles of people that any single individual would be interdependent with, no matter how fleetingly.

He spoke of the 'conveyor belts' running through individuals' lives growing 'longer and more complex' (1994: 452), requiring us to 'attune' our conduct to the actions of others (1994: 445), and becoming the dominant influence on our existence, so that we are less 'prisoners of our passions' and more captive to the requirements of an increasingly complex 'web of actions' (1994: 445), particularly a demand for 'constant hindsight and foresight in interpreting the actions and intentions of others' (1994: 456). Just as important as the 'length' of chains of interdependence was the increasing ambivalence of overlapping and multiple networks: as social relations become more complex and contradictory, the same people or groups could be 'friends, allies or partners' in one context and 'opponents, competitors or enemies' in another. 'This fundamental ambivalence of interests,' wrote Elias, is 'one of the most important structural characteristics of more highly developed societies, and a chief factor moulding civilized conduct' (1994: 395).

All of these processes of civilization 'tend to produce a transformation of the whole drive and affect economy in the direction of a more continuous, stable and even regulation of drives and affects in all areas of conduct, in all sectors of his life' (1994: 452). We are all compelled more and more to regulate our conduct 'in an increasingly differentiated, more even and more stable manner'. Elias referred to this increasing self-regulation as a process of 'psychologization' and 'rationalization', because it revolved around the growing reflexive understanding of our own actions, those of others, their interrelationships and their consequences.'The web of actions grows so complex and extensive,' wrote Elias, and 'the effort required to behave "correctly" within it becomes so great, that beside the individual's conscious self-control an automatic, blindly functioning apparatus of self-control is firmly established' (1994: 445-6).

It is useful, too, to recall the qualifications which Elias added in response to his critics. First, his concept of a civilizing process in European social history did not imply the existence any sort of original 'state of nature' in some early historical period. There is no 'zero point in the historicity of human development' (1994: 131), no example of human existence without social constraints. Second, there was no particular beginning to civilizing processes, so that in any given period people will regard themselves as more civilized than the peoples in the preceding periods. 'Wherever we start,' he wrote, 'there is movement, something that went before' (1994: 48). Third, civilizing processes were never-ending, and that we can never regard ourselves as having attained a state of 'true' civilization. Although he was confident that considerable social development had taken place since antiquity, he was equally sure that we had by no means stopped 'civilizing' ourselves and each other, which was why the final line in The Civilizing Process included these words from Holbach: 'la civilisation....n'est pas encore terminèe' (1994: 524); see also 1996: 173).

...and decivilization
The Civilizing Process was completed in 1939, and both Elias himself and his interpreters, supportive as well as critical, have tended towards the view that his understanding of the development and dynamics of Western societies did not change substantially afterwards. The development of Elias's ideas between the 1960s and 1980s reveals, however, a more nuanced picture, and his writings can be regarded as ranging from a reiteration of his arguments in The Court Society and The Civilizing Process, through a development or refinement of his ideas, to a distinct change of direction and emphasis.

One of the themes running through the way that Elias changed and developed his approach after The Civilizing Process was an examination of the contradictory and ambivalent character of processes of civilization, their 'dark' sides and the question of 'civilized barbarism'. In The Civilizing Process, the relationship between barbarism and civilization had been presented largely as mutually exclusive, one turning into the other, with possible 'reversals' of direction. To a large extent Elias's later work remains consistent with this line of argument, raising the possibility that specific processes of state-formation produce either a 'deficient' process of civilization, or result in a clear process of decivilization encouraging the more widespread manifestation of brutal and violent conduct.

However, Elias also suggested that civilization and decivilization can occur simultaneously. For example, he made the point that the monopolization of physical force by the state, through the military and the police, cuts in two directions and has a Janus-faced character (1996: 175), because such monopolies of force can then be all the more effectively wielded by powerful groups within any given nation-state, as indeed they did under the Nazi regime. Pursuing a line of thought he had been developing since the 1970s (Wouters 1977: 448), in 1986 he argued for the reversibility of social processes, and suggested that 'shifts in one direction can make room for shifts in the opposite direction,' so that 'a dominant process directed at greater integration could go hand in hand with a partial disintegration' (1986: 235). Similarly, in The Germans he remarked that the example of the Hitler regime showed 'not only that processes of growth and decay can go hand in hand but that the latter can also predominate relative to the former' (1996: 308). In a critique of Kingsley Davis' understanding of social norms (1996: 159-60), he argued that Davis emphasised the integrative effect of norms at the expense of their 'dividing and excluding character'. Social norms had an 'inherently double-edged character', since in the very process of binding some people together, they turn those people against others.

Elias had said that a large part of his motivation in writing The Civilizing Process was to come to a better understanding of the brutality of the Nazi regime, since 'one cannot understand the breakdown of civilized behaviour and feeling as long as one cannot understand and explain how civilized behaviour and feeling came to be constructed and developed in European societies in the first place' (1994: 444-5). In other words, Elias was proposing that barbarism and civilization are part of the same analytical problem, namely how and under what conditions human beings satisfy their individual or group needs 'without reciprocally destroying, frustrating, demeaning or in other ways harming each other time and time again in their search for this satisfaction' (1994: 31). The problem for Elias was both to make events such as the Holocaust - and one could add any number of other examples of 'modern barbarism' - understandable as the outcome of particular social figurations and processes of socio-historical development, and also to explain what it was about the development of modern state-societies which generated organized critical responses to such large-scale genocide (1994: 445).

An example: the stolen generations
Although law and jurisprudence are frequently included in social and political analyses of the state, government and liberalism, there remains an odd abstractness to the discussion, a focus on the history of legal thought and a construction of 'law' or 'justice' as rather rarified entities floating about in the philosophical ether (Rawls 1973; Dworkin 1977, 1986). Few judgments are examined in any detail - Mabo being, in Australia, the exception which proves the rule - actual law-making processes (in legislation and in courts) and the evolution of legal principles not at all. Our understanding of the actual institutionalization of civilization would, therefore, be much enhanced with a more finely grained grasp of the precise ways in which the management of the tensions between civil society and the state are actually managed in our courts, by our judges, in the field of law.

The example I would like to turn to in illustrating the institutional form actually taken by civilization is that of the stolen generations, the forcible removal of Indigenous children from the families from around the turn of the century to the 1970s. A profound challenge was posed to the European conception of civilization was raised in the Australian context in May 1997, when the Australian Human Rights and Equal Opportunity Commission (HREOC) issued its Bringing them Home, stressing that the treatment of indigenous Australian children by both State and Church agencies throughout this century falls clearly within the terms of the UN definition of genocide, which includes 'forcibly transferring children of the group to another group' committed 'with intent to destroy, in whole or in part, a national, ethnical, racial or religious group, as such'.

The particular significance of the relationship between European and Indigenous Australian children is that the removal of Aboriginal children was itself centrally a 'civilizing project', despite the fact that the subsequent critique of that practice is also undertaken in the name of 'civilization'. 'Civilization' was the foundation for citizenship in the modern nation-state, with its achievement the key condition for the attainment of citizenship rights. The story of the 'stolen generations' as part of a particularly Australian set of civilizing processes is thus an important example of the multiple meanings of the concepts 'civilization' and 'citizenship'.

The primary and overarching concern of the assertion of legal guardianship by the state over all indigenous children was to 'solve' the 'half-caste problem' by breeding out the colour of both body and mind through a program of social engineering; in this sense the removal of Aboriginal children meshed with the other, accompanying, strategy of controlling sexual relations and reproduction among adult Aborigines. This was certainly the most strongly articulated argument in the writings of the politicians, administrators and anthropologists central to the development of the various forms of legislative and administrative action. 'Merging', 'absorption' and 'assimilation' into the ways of 'civilization' were the key concepts around which this discourse was organized. In 1936 a conference of the leading authorities in Aboriginal affairs declared its belief 'that the destiny of the natives of aboriginal origin, but not of full blood, lies in their ultimate absorption by the people of the Commonwealth' (Commonwealth of Australia 1937: 3). By the 1950s this kind of conception had been replaced by one more organized around a liberal conception of citizenship, and in 1950 Paul Hasluck told the House of Representatives that 'Their future lies in association with us, and they must either associate with us on standards that will give them full opportunity to live worthily and happily or be reduced to the social status of pariahs and outcasts living without a firm place in the community' (1953: 6).

Within this second conception of 'citizenship as assimilation', it was also possible to regard the state's and church's intervention into Aboriginal family life as advancing the 'welfare' of the Aboriginal population as a whole, by posing a stark and uncompromising contrast between membership of the European community, on its terms, and exclusion from civilization itself. Aboriginal culture and its way of life, especially once it had encountered European civilization, was presented by Hasluck and almost every other administrator in Aboriginal affairs as inherently flawed, fragile and basically worthless, producing only illness, disease, drunkenness, filth and degeneracy in the 'thousands of degraded and depressed people who crouch on rubbish heaps throughout the whole of this continent' (1953: 9; see also Read 1983: 20). Aboriginality was constructed simply as a 'primitive social order' composed of 'ritual murders, infanticide, ceremonial wife exchange, polygamy'(1956: 2), so that for Hasluck and most white Australians, the permanent elimination of Aboriginality from the fabric of Australian social life was self-evidently synonymous with civilization and progress itself, a crucial element of the truth that 'the blessings of civilization are worth having'. 'We recognise now,' said Hasluck, 'that the noble savage can benefit from measures taken to improve his health and his nutrition, to teach him better cultivation, and to lead him in civilised ways of life.... We know that the idea of progress, once so easily derided, has the germ of truth in it' (1953: 17).

The destruction of Aboriginal culture and society, both inexorable and planned, via the assimilation of Aborigines as individuals and as children, was thus posed in terms of a humanitarian concern for the welfare of indigenous Australians, and this interpretation is still an important element of the 'common sense' understanding of the practice of forced child removal.

How was such a linkage between welfare and attempted cultural genocide, mediated through policies concerning children and the 'rising generation' possible, and how should we understand it? Whatever it was about European models of the relationship between state and society which produced Aboriginal child removal is not simply a "mistake" for which apologies might be issued, but something much more deeply rooted in European social, political and legal thought, with profound on-going implications for social relationship in Australia, between as well as among indigenous and non-indigenous people. The model of citizenship and the evolution of individual rights - especially children's rights - underlying much of the debate displays considerable amnesia about exactly how 'colonizing' European social history has been in relation to its own subject populations (Weber 1976), and especially in relation to children and family life (e.g., Hearst 1997). This raises the question of the relative success of European 'internal' colonization in comparison to the overall failure to eradicate non-European identities. The 'community' destroyed by European child welfare policies was defined in class rather than ethnic or racial terms, whereas an ethnically defined community seems to be much more difficult to eliminate, short of physical genocide. Ethnocultural groupings appear to be more resistant to attacks on their citizenship status than class-defined groupings - paradoxically, because they are also the most likely to have such restrictions on their citizenship status imposed on them.

Liberal social and political thought also rests on a delicate balance between individual rights and some conception of 'the social', or the particular and the universal, making it possible for civilisation and modernity to have barbaric effects to the extent that this balance takes particular forms. This is what is striking about the sheer disgust which European Australians have tended to feel for Aborigines (Read 1983: 20). Hasluck, for example, gave the House of Representatives this folksy advice, that '[w]e have to give attention to hygiene. So long as natives are not living in a way that makes them physically acceptable - to put it crudely, so long as natives live in a way that makes them smell - then there is no hope for them. We have to improve their hygiene in order to make them acceptable' (Commonwealth Parliamentary Debates 8(New Series) 21st(1st) 6 October 1955: 1333). (2)

The specific significance of the Australian 'stolen generations' history is that it is not simply one of a dominance of communal identities over individuals, with all the negations of individual freedoms and rights which that entails. In fact, it shows how the two are interlinked and that we are by no means not out of the woods with the introduction of the factor 'liberalism'. It is useful to distinguish here, as Tim Rowse does, between different versions of liberalism, between what Rowse call 'juridical' liberalism, focused on 'a sometimes militant concern to unfetter individuals from pernicious social bonds and to imagine individuals and to act towards them in terms of their abstract universal equivalence from the point of view of the state' (1999: 127), and a 'sociological' liberalism which instead conceives individuals as integral parts of collectivities as well, with their communal identity an essential rather than expendable element of their relationship to the state. Aimed as juridical liberalism was towards assimilation into a mono-cultural form of citizenship, the removal of indigenous children was structured around an individualised conception of well-being and welfare, but it was this assimilationist focus on individuals which helped undermine communal identity, in turn inflicting significant long-term psychological and social violence on the individuals who make up communities.

Zygmunt Bauman has identified the more generalizable features of what he calls the 'assimilatory project' within European state formation, and the centrality of that project to the very nature of the modern state. It was part and parcel of the process of dismantling older, deeply rooted forms of communal life which provided alternative, sometimes oppositions frameworks of social power. Assimilation, he suggests, 'was an exercise in discrediting and disempowering the potentially competitive, communal or corporative sources of social authority' (1991: 106). As part of the liberal political and legal program to secure the modern state's 'monopoly of law-making and coercion' (p. 111), 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' (p. 107). 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. (3)

Liberal models of individual rights can never, on the one hand, really detach themselves from an accompanying conception of 'society as a whole' to which individuals are to be 'assimilated'. Indeed, the rhetoric of liberal democracy tends to draw 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. On the other hand, when combined with an organic, mono-cultural, and unitary conception of citizenship and community, individualistic liberalism has 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.

There is, then, a powerful tension at the heart of liberal understandings of children and their place in society, between 'the best interests of the child' and 'the best interests of society'. The assumption that the two simply work in harmony means that the former are almost always defined in terms of the latter. Rather than simply being an error in judgement, a mistake for which Australians today should or should not apologise, the policies and practices surrounding the 'stolen generations' reveal two much more fundamental flaws buried deep within 'civilized', 'juridical' liberalism: the difficulty which its 'juridical' or 'contractual' individualism produces for a comprehension of individuals as socially located, inter-generational, inter-subjective beings, their essentially communal identities 'stretched' over time both backwards and forwards (van Krieken 1997), and its mono-cultural and organicist conception of 'society', which allows only for assimilation to a single, individualised and de-communalised 'way of life'. It is only to extent that both these features of 'juridical' liberalism are addressed that we can hope for any degree of certainty that similar histories will not re-emerge in relation to any group without the protection afforded by a visible and publicly respected cultural and political presence.

The continuing project of civilizing the state
A major problem for any conception of civilization, then, is that the rule of law's 'restraint' of violence and force is itself founded on violence and force (Cover 1986; Sarat & Kearns 1992; Derrida 1990). Civilizing processes which are organized around a self-perception as 'civilized' without a corresponding identification with the different humanity of others, adhering to different civilizational standards, tend to be accompanied by aggression and violence towards those who remain uncivilized, largely because of the threat they pose to the fragility of the achievements of civilisation, It is this aggression which then underlies the associated civilizing offensives. The state monopolization of violence in fact involved the exercise precisely of violence on groups seen to lie outside the prevailing standards of civilization, so that civilizing processes involve not simply the reduction of violence and aggression, but their rearrangement.

In addition, apart from a few passages in The Civilizing Process (pp. 41, 461-3, 465, 509), we do not get a very clear sense that at the very time that civilization was developing in Western Europe, it was busily spreading itself over the whole globe in the most violent of ways, so that it is not unfair to say that the ritualised civility of European court society was built on the blood of murdered 'primitives' and bought with the land, labour and raw materials which marauding Europeans plundered from 'their' empires. In all the discussions of the state's monopolisation of violence, there is little clear sense of what states actually did with that monopoly, in relation to both their own populations and those of the parts of the world they set about colonizing. Elias himself, for example, spoke of the 'spread' of Western civilization, the 'transformation of Oriental or African peoples in the direction of Western standards' (1994: 464), and the 'integration' of the rest of world within European standards of behaviour as an essential element of the 'civilization of the colonized' (1994: 509) in a way which glossed over exactly how violent a process that really was.

Elias did recognize the ideological centrality of the concept of civilization to the whole project of colonialism, pointing out that this was its 'watchword', and that in addition to seizing land 'it also became necessary...to rule people in part through themselves, through the moulding of their super-egos' (1994: 509). But he always seemed simply to assume that this mode of rule was destined to succeed; he was able both to see 'habituation to foresight' and 'the stricter control of behaviour and the affects' as 'instruments of dominance' (1994: 462) and to observe that 'this civilization is the characteristic conferring distinction and superiority on Occidentals' (1994: 463, emphasis added). This suggests that he was persuaded of the actual as well as the self-perceived superiority of Western civilization, based on a supposed greater ability to manage their emotions and impulses, with far less attention paid to their instrumental technological and military superiority. A central problem, then, is to unpack the various ways in which processes of 'civilization' are simultaneously processes of colonization, and to work towards an understanding of colonialism and decolonisation which might be usefully and productively organised around the concept of 'civilization'.

Second, the concept of "civilizing the state" is itself an element of the contemporary rationality of "advanced liberal" governance, promoting the assumption of responsibility within individuals and groups themselves for the ethical mode and substantive civility of contemporary practices of government. The demands made by civil society of the state - that it become, for example, more civilized, or that it subject itself to the critical scrutiny of economic, social and political actors in society - have thus always been central to a particularly liberal political rationality. The question, 'how can non-government organisations play socially positive roles in building civil society and civilising their states?' should be seen, then, as exactly the kind of question which liberal government inherently poses for itself, since it is what makes political rationality specifically 'liberal'.

The contribution which these arguments can make is to help broaden the issues we engage with in thinking about the connections between governance, law and civilization. As well as asking whether the infrastructure of civil society can support democracy and popular participation, or how non-government organisations might contribute towards building civil society and civilizing their states, we can also address questions concerning the nature of law and civilization themselves, including: in the process of civilizing the state, can we identify a 'dark underbelly' of barbarism inherent in precisely that process? What are the colonizing dimensions of what we have chosen to see as civilization? How can the constraint of violence which civilization and law is meant to encourage be undertaken without reproducing merely different forms of violence? in what ways are the interrelationships between governance and civilization configured by particular sets of legal conditions, by underlying legislative and judicial frameworks? What kind of debate is it possible to have about the definition of 'civility' within highly differentiated societies, particularly its degree of homogeneity or heterogeneity? The practical realization of answers to questions such as these, I would suggest in conclusion, will play an important role in developing forms of government which are more likely to produce genuinely peaceful and civil societies.

References

Bauman, Zygmunt (1987) Legislators and Interpreters: On Modernity, Post-Modernity and Intellectuals Cambridge: Polity.

-- (1991) Modernity and Ambivalence. Cambridge: Polity.

Becker, Marvin B. (1988) Civility and Society in Western Europe, 1300-1600 Bloomington: Indiana University Press.

Burchell, Graham (1991) 'Peculiar interests: civil society and governing 'the system of natural liberty'' in Graham Burchell, Colin Gordon & Peter Miller (eds) The Foucault Effect: Studies in Governmentality London: Harvester, 119-50

Chesterman, John and Galligan, Brian (1997) Citizens without Rights: Aborigines and Australian Citizenship Cambridge: Cambridge University Press.

Commonwealth of Australia (1937) Aboriginal Welfare: Initial Conference of Commonealth and State Aboriginal Authorities. Canberra: Government Printer.

Cover, Robert M. (1986) 'Violence and the word' Yale Law Journal 95: 1601-29.

Derrida, Jacques (1990) 'Force of law: the "mystical foundation of authority"' Cardozo Law Review 11 : 921-1045

Dorsett, Shaunnagh (1995) 'Civilisation and cultivation: colonial policy and indigenous peoples in Canada and Australia' Griffith Law Review 4(2): 214-38

Dworkin, Ronald M. (1977) Taking Rights Seriously London: Duckworth.

-- (1986) Law's Empire London: Fontana.

Elias, N. (1978) 'The Civilizing Process revisited: interview with Stanislas Fontaine', Theory & Society, 5: 243-53.

-- (1983) The Court Society, Oxford: Basil Blakwell.

-- (1986) 'Soziale Prozesse', in B. Schäfers (ed.), Grundbegriffe der Soziologie, Opladen: Leske en Budrich, 234-41.

-- (1994) The Civilizing Process, Oxford: Blackwell[1939].

-- (1996) The Germans, Cambridge: Polity.

Febvre, Lucian (1998) 'Civilisation: Evolution of a word and a group of ideas' in John Rundell & Stephen Mennell (eds) Classical Readings in Culture and Civilization London: Routledge, 160-90 [1930]

Hasluck, P. (1953) Native Welfare in Australia: Speeches and Addresses. Perth: Paterson Brokenshaw.

-- (1956) 'Policy of Assimilation', National Archives of Australia, NTAC 1956/137.

-- (1959) 'The future of the Australian Aborigines' Dawn 8(1) 1959: 11-15.

Hearst, A. (1997) 'Domesticating reason: children, families and good citizenship' in A. McGillivray (ed.) Governing Childhood, pp.200-24. Aldershot: Dartmouth.

Human Rights & 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. Sydney: Sterling Press.

(http://www.austlii.edu.au/rsjlibrary/hreoc/stolen/index.html)

Hume, David (1987) 'Of refinement in the arts' in Essays Moral, Political and Literary Indianapolis: LibertyClassics

Keane, John (1998) Civil Society: Old Images, New Visions Cambridge: Polity Press.

Lloyd, Dennis (1964) The Idea of Law Ringwood: Penguin.

Pagden, Anthony (1988) 'The "defence of civilization" in eighteenth-century social theory' History of the Human Sciences 1(1): 33-45.

Peterson, Nicolas and Sanders, Will (eds) (1998) Citizenship and Indigenous Australians: Changing Conceptions and Possibilities Melbourne: Cambridge University Press.

Pettit, Philip (1997) Republicanism: A Theory of Freedom and Government Oxford: Clarendon Press.

Rawls, John (1973) A Theory of Justice London: Oxford University Press.

Read, P. (1983) The Stolen Generations: The Removal of Aboriginal Children in New South Wales, 1883 to 1969. Sydney: Ministry for Aboriginal Affairs.

Rose, N, (1999) Powers of Freedom: Reframing Political Thought Cambridge: Cambridge University Press.

Rowse, T. (1999) 'The modesty of the state: Hasluck and the anthropological critics of assimilation' in T. Stannage, K. Saunders and R. Nile (eds) Paul Hasluck in Australian History: Civic Personality and Public Life St. Lucia: University of Queensland Press: 119-32.

Rundell, John & Mennell, Stephen (1998) 'Introduction: civilization, culture and the human self-image' in John Rundell & Stephen Mennell (eds) Classical Readings in Culture and Civilization London: Routledge, 1-35

Sarat, Austin and Kearns, Thomas R. (1992) 'Making peace with violence: Robert Cover on law and legal theory' in Sarat, Austin and Kearns, Thomas R. (eds) Law's Violence. Ann Arbor: University of Michigan Press: 211-50.

Seligman, Adam B. (1992) The Idea of Civil Society New York: Free Press.

Tester, Keith (1992) Civil Society London: Routledge.

Tully, James (1995) Strange Multiplicity: Constitutionalism in an Age of Diversity Cambridge: Cambridge University Press.

van Krieken, R. (1990) 'The organisation of the soul: Elias and Foucault on discipline and the self' Archives Europeénes de Sociologie 31(2): 353-71.

-- (1997) 'Sociology and the reproductive self: demographic transitions and modernity' Sociology 31(3): 445-71.

-- (1998) Norbert Elias London: Routledge.

-- (1999) 'The barbarism of civilization: cultural genocide and the 'stolen generations'', British Journal of Sociology 50(2): 295-313.

Weber, E. (1976) Peasants into Frenchmen: The Modernization of Rural France, 1870-1914. Stanford: Stanford University Press.

Notes

1. Much of the research on which this paper is based has been supported by a Large Australian Research Council Grant, No. A1267, as well as by an Extended Research Secondment provided by the Research Institute for the Humanities and Social Sciences at the University of Sydney in 1998. It also draws on two previous papers published earlier this year, '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, as well as my book, Norbert Elias London: Routledge, 1998.

2. In the Hasluck papers in the National Library of Australia, there is a draft of a report by an Inter-Departmental committee on 'Matters Affecting Native Welfare', where the following recommendation 'That any aboriginal who has reached a standard of general education which makes his attendance at a secondary school advisable should be admitted to a "State" secondary school' has added to it, in Hasluck's writing, the following emendation: 'provided that his standard of personal hygiene and mode of life make him acceptable' (NLA MS 5274 Hasluck Papers, Box 32).

3. As Hasluck put it: 'the loss of any valid and distinctive aboriginal culture is certain in the course of time. The ancient pride can remain - and in fact may grow. Those people of Scottish ancestry who delight in strange capers at Hallowe'en, and those people of Irish origin who whenever they do something fine exclaim "It must be my Irish blood" are examples of the sort of cultural pride I had in mind. But how real are the bagpipes and the kilts and the poetry of Burns as a cultural force in Australia? The Scot and the Irish and the English are "assimilated", not "integrated" into Australian life. I look to a future when a person whose great grandfather was an Australian aboriginal will be as proud of the fact as a Scot is rightly proud of his barbaric ancestry' (Hasluck 1959: 15). The argument for a distinctive cultural identity he dismissed as so much romanticism, such as is expressed in 'the Moomba Festival in Melbourne, in the vogue of the Central Australian paintings and in the sale of factory-made koala bears and boomerangs' (p. 15).