The above quote is from Cotgrove's Catastrophe and Cornucopia (1982) which is one of the earlier studies of the clash of positions (`discourses') over `nature' which has been the hallmark of modern environmental conflict. Like many other writers reviewed in the previous chapter, Cotgrove regards moderm environmentalism as located in new middle class politics. Importantly, he also regards the advent of radical-oppositional environmental conflict as representing a struggle for power. However, the quote is used here to illustrate what is considered in this dissertation to be a very common but mistaken view of power.

The above quotation from Cotgrove seems to present power as hegemony: the struggle of a group to have its specific definition of norms accepted as general. `Power as hegemony' locates power and struggle in conscious agents (such as a class). Alternatively, the approach followed in this dissertation derives from the work of Michel Foucault who sees classes as effects, rather than agents, of contestation. Accordingly, strategy and struggle precede the identity of a group. A class is recognised as a class after its actions and does not have some prior or ideal identity. Further, power cannot be located within a single category (of classes) but cuts across a number of competing domains. Foucault does not deny the existence of class and class struggle. He considers these in a different way, as he does the operation of power. Foucault does not see power as a property/capacity of a class; rather it is the overall effect of its strategic positioning. Thus one cannot start an analysis with pre-existing classes holding various amounts of power or none at all. Power is rather the strategy of classes in action (Deleuze 1988:25). In this view of power and class, the key area to study is the nature of `struggle' itself and the practical strategies deployed (Foucault 1988:225). Foucault suggests that `struggle' is at the heart of power relations (Foucault 1988:240).

The methodological problem is one of explaining how actions and decisions are not coincident with interests. This problem has not been overcome in post-Marxist positions, despite their claimed influence from post-structuralist theory (eg, Laclau and Mouffe 1985), which continue the `power as hegemony/ideology' thesis and still continue to posit real or authentic interests. By dispersing power in discourse Foucault overcomes the identity of power with interests, and sees interests as the effect of strategies. An empirical application does not have to identify `real' interests but describes strategies, effects, and outcomes.

The recent struggle over Australia's native forests is about which contending discourse on `nature' will dominate politically, about how `nature' is to be governed and for whom. As discussed in the previous chapter, the political struggle over `nature' has required the creation and political organisation of a popular new social movement able to engage in the practical world of political power. I contend that an account of political power is essential to a theoretical and methodological framework in which to analyse recent environmental conflict.

The above are prefacing comments on the subject of this chapter which deals with power, discourse and environmental conflict. The purpose of this chapter is to outline a particular approach to power, recognising that conceptually `power' is highly contested in political theory. It will be argued that what can broadly be termed a materialist-discourse approach based on the work of Foucault is the most adequate way to account for the operation of power, the form of struggle and the nature of change, constituted in the actual politics of environmental conflicts. Further, such an approach enables interpretation and assessment of the political significance of recent environmental conflicts. It also provides an understanding of how `nature' is governed and makes possible conclusions as to the liberatory or interventionist-policing effects of contemporary environmental conflict.

The chapter is organised into four parts. First, the conventional theoretical and methodological approach to power is critically outlined using as the focus the influential work by Steven Lukes, Power: A Radical View (1974). Second, an overview is given of Michel Foucault's alternative theory of power which is centred on a materialist understanding of the nature, operation and function of political discourse. Third, Foucault's concept of `governmentality' and its development by Rose and Miller (1992) into an analytic of `problematics of government' are introduced. This approach to the operation of modern power offers an important conceptual framework in which to analyse the political functions of public discourse, discursive designs and offical discourses. This framework is briefly applied to the situation of recent environmental conflict, showing how `nature' has increasingly become a subject of government producing its own policing apparatus. Finally, the methodological considerations which can be drawn from the theoretical discussion in this and the previous chapter are discussed and the research procedure employed in the case study is outlined.

Recent reviews of the theories of power point to two distinctive theoretical and methodological positions (Clegg 1989, Hindess 1996). There is, firstly, a long tradition of theories of power that are connected by the same basic foundational concepts. Stretching from Hobbes to Lukes these theories have represented the normal or conventional approach to power in the social sciences (Clegg 1989:211). Although various theories making up this conventional approach have presented differing models, methods and political inferences, they are all underpinned by the view that power is a capacity. Thus power is seen as a capacity for action enabling A's interests to prevail over B's. It is a capacity which can be possessed and transferred. Here power is regarded as `power over' someone or something. This is the liberal-juridico concept of power in which the "...central figure is the sovereign" (Philip 1983:37) with legitimate rights and command of the right to be obeyed. Such sovereign power is given up by individuals to the modern state in a social contract or other cooperative arrangement. In this sense, power is possessed or held by an ultimate authority of which power is a property and can be legally transferred (Clegg 1989:159).

An alternative view of power has been developed by Foucault, who argues that power is exercised rather than possessed. Rather than a property, a quantitative capacity, power is considered by Foucault as a positive strategy which has no privileged centre, "...the sovereign, the ruling class, or the state" (Hennessy 1993:19). Unlike the conventional view of power as `power over', Foucault takes the view that power is `power to', it is an "...enabling phenomenon" (Law 1986:167). For Foucault, power is productive rather than prohibitive (Fraser 1989:18):

Foucault's radical departure from conventional theories of power centres on his linking of power with knowledge and the critical role of discourse and discursive practice. I will consider his contribution in the rest of this chapter, but for now I return to conventional theories and examine Lukes' three-dimensions of power and the major methodological requirements in empirical applications of his model of power.

Lukes (1974) argued that there were three dimensions to power. The first and second dimensions are based on his critical review of major debates in the USA between elitists and pluralists about the distribution of power. Lukes added the radical proposal that there was a critical and more far reaching third dimension to power dealing with the shaping of concsiousness.

The one dimensional view of power, the decisional approach associated with Dahl's (1961) study of power in New Haven, focuses on concrete action or behaviour in the formulation of actual decisions, "...on issues over which there is an observable conflict of interests, seen as express policy preferences, revealed by political participation" (Lukes 1974:15). In this view interests are seen as subjective and are "... equivalent to the expressed wants, preferences, choices of the actors concerned" (Benton 1981:163). This first dimension of power accords with "...the pluralist view of power as influence" (Alford and Friedland 1985:7).

The second dimension, non-decision making, is associated with the critique of Dahl by Bachrach and Baratz (1962). In their view power can be found to occur in the "...ways in which decisions are prevented from being taken on potential issues over which there is an observable conflict of interests" (Lukes 1974:20). In this view, by institutional practice or rituals, dominant individuals or groups can prevent, suppress or divert potential issues from reaching the policy agenda and thus effect non-decisions (Benton:163). This is the managerial view of power as domination, or an elitist view of power (Alford and Friedland 1985:7, Cox 1985:224). In this second dimension of power, which attempts to look beyond actual decisions, "...there is still a requirement that such issues and conflict must be present for an exercise of power to be identified" (Benton 1981:163). Interests are still conceptualised in subjective terms as associated with wants and preferences.

Lukes' radical development is a third dimension of power. This proposal incorporates the first and second dimensions, suggesting that they refer to elements of power, but both fail to account for power as the shaping of people's preferences or self-perceived interests, so that neither overt nor covert conflicts of interests occur (Ham and Hill 1984:67). Lukes incorporates but goes beyond Barhrach and Baratz's notion of a mobilisation of bias situated in the way societies are politically constituted. As Benton puts it, the "...mechanisms by which potential issues, grievances and conflicts are excluded may not be reducable to individual decisions either to act or to not act (non decisions)" (Benton 1981:163). Lukes' third dimension of power is associated with the shaping of consciousness and ideology. In this third dimension interests are conceptualised as real interests, as objective. Lukes uses Gramsci's notion of cultural hegemony to show the means of preventing the realising and articulation of what people's actual interests are. This has been associated with class rule (Alford and Friedland 1985:8, Cox 1985:224).

The shaping of consciousness and ideology is central to Lukes' radical view of power, as it is in the Marxist approach to power (Isaac 1987, McQuaire and Spaulding 1989) where it has been the major preoccuption of the Frankfurt School of critical theory (Hindess 1996:ch 4, Hoy 1981). Control over consciousness is very much influenced by a negative view that power is `power over'; this form of power is regarded as repression. This `repressive hypothesis' rests on the claim that there is a realm of freedom outside of power relationships. The claim is that in the absence of power, false consciousness gives way to true consciousness (Hoy 1981:56).

What, then, are the major methodological considerations in empirical applications of conventional models of power? In Lukes' first dimension, the most limited view of power, Dahl argues that the study of power simply requires "...the careful examination of a series of concrete decisions" (Dahl 1958:140). This is based on a static positional view of power where the analysis is finished after "...it has demonstrated which positions are filled by which people" (Saunders 1983:23).

Lukes' second and third dimensions of power deal with forces that are conceptually unobservable. Empirically, this problem has been tackled in the case of non-decison making (Lukes' second dimension) by Crenson's (1971) use of comparative method. Crenson conducted a detailed historical case study of two adjacant steel making cities facing similar air pollution problems. The first city, which contained a large number of different steel companies, passed air pollution control laws in 1949. However, in the second city, dominated by a single company, no such laws were introducted until 1962. This, Crenson found, was due to the immense corporate influence over the city by US Steel who, it was believed, might close in the face of tough laws. This, for Crenson, was a case of non-decision making.

In the case of the shaping of consciousness, Lukes' third dimension, Gaventa's (1980) study of a depressed mining community, is often cited as a successful focus on structural and ideological conditions which demonstrate how actors are prevented from acting on or even conceiving certain interests. Miners, living below the poverty line with high local unemployment, held the belief over successive generations that there was no point in political action and participation, as this could not produce any change in their community. The dominance of the mining interests and the poverty of the miners led to a history of quiescence with only limited cases of rebellion.

In exploring this situation, which he regards as the operation of Lukes' third dimension of power, Gaventa's method undertakes a detailed historical research into the power relations and decision making in the mining community. Gaventa is able to avoid the reduction of his account to conscious interests (Clegg 1989:109). He is not concerned to "...impute real interests to actors, it is sufficient to show that subordinate actors are prevented from acting on or conceiving certain interests" (Shapiro and Wendt 1992:215). It is the rich descriptive historical narrative which in this case is used to show the causal mechanisms of power so conceived.

Conventional theories of power, of which the three dimensional model is the most refined, are ultimately based on the position that power is a quantitative capacity to secure outcomes over interests, whether they be objective or subjective. This position concentrates analytical attention and associated methods on the `score-card' of conflict, to the vector sum of several powers in any conflict situation (Hindess 1982:500). Hindess' citation from Benton's summary formulation neatly encapsulates conventional theories of power:

The central conclusion, that outcomes of conflict are essentially predictable and unvarying, is challenged by Hindess who argues that the "..securing of outcomes should always be seen as problematic..." (Hindess 1982:500). In studying power and outcomes, Hindess advocates a move from the strengths and weaknesses of the attendant players involved, to what he labels the `arena of struggle and conflict'. By this he refers us to an examination of the actual conditions of a struggle or set of struggles, to the limitations on outcomes contained within the conditions of conflict. Hindess rejects the idea that outcomes can be determined by adding up the relative capacities of conflicting groups. Such a calculative model precludes consideration of the dynamics of struggle, in which various actors:

This brings us to questions of political strategy and tactics in power relations and Foucault's contribution on the operation of power, which he regards as a facilitative technique in which discourse is crucial. For Foucault, "Power relations are both intentional and nonsubjective...they are imbued, through and through, with calculation: there is no power that is exercised without aims and objectives. But this does not mean that it results from the choice or decision of an individual subject" (Foucault 1978:95). The next two sections will consider Foucault's approach and some of the debates about power, discourse and `governmentality' he helped to generate.

The previous section illustrated how the concept of the `sovereign' and the related notions that power is a negative capacity and is repressive have dominated theories of power. Foucault argues that such a concept of power, while possibly applicable to the feudal era can no longer account for how power operates. His concern is that this approach to power still dominates current theories and empirical research into the nature of modern power (Clegg 1989:156, Constable 1991: 271-282, Minson 1980:4-7). Foucault clearly rejected the repressive hypothesis (Foucault 1981). Hoy sums up Foucault's case against power understood as repression:

For Foucault, truth is the product of power and all truth is contingent and based on various economies of discourse. Truth cannot be constituted without discourse and power operates to both produce and establish which discourses are regarded as true. According to Foucault:

Foucault's alternative account is based on his historical research into a series of micro-practices and techniques associated with the disciplinary control of expanding populations. These practices and techniques led to the emergence of what Foucault claims are distinctly new forms of modern power. This modern power is associated with the advent of industrial capitalism in the sevententh and eighteenth centuries. It is here that Foucault finds the emergence of the coupling of power with institutionalised knowledge which progressively formed regimes of truth connected to discourses. Unlike early forms of power which were imposed from above (as in the `sovereignty, positional and capacity' framework), in Foucault's concept, modern power should be seen as "...local, continuous, productive, capillary, and exhaustive" (Fraser 1989:20).

Such characteristics of the operation of modern power originated, according to Foucault, in the micro-techniques and practices of early industrial workplaces and schools, the early hospitals, asylums and prisons which were concerned with the "...various social techniques for the administration of corporeal, attitudinal and behavioural discipline" (O'Neill 1987:42). Foucault does not regard power as only contained in the state or other political structures; in his view "Power is everywhere" (Foucault 1978:93). Power is firstly found and constituted at the micro level in various local sites. From here Foucault argues it was incorporated and developed into larger more institutionalised structures of instrumental rationality, control and discipline. From a plurality of local points, power has developed into global or macro strategies of domination:

From his historical research into power/knowledge and the central role of discourse in modernity, Foucault is able to advance a number of hypotheses on the way modern power operates which are fundamentally different from the `sovereignty, positional and capacity' framework. Foucault's concept of power concerns the "...processes, procedures, and apparatuses whereby truth, knowledge, belief are produced" (Fraser 1989:19). And the functioning of discursive practices and regimes involves:

Foucault's position is that power circulates in and through the production of discourses in society where the nexus between power/knowledge and discourse produces and establishes, through normalisation, particular regimes of truth. This is a rather different approach to those found in agency/structure accounts. It has been remarked that Foucault dissolves the agency-structure question, the subject of the clash between structuralism and hermeneutics:

Whereas `agency' assumes that interests are transparent to actors, `structure' locates the formation of interests in an external and imposed system (such as ideology). Foucault, however, locates power neither `in' consciousness nor `above' minds (in ideology). Power is rather dispersed through actions, practices and discourse. Discursive practices are not reflections of distorted realities, as would be the case in Lukes' third dimension of power or in Marxist treatments of ideology and false consciousness. Foucault regards power as inscribed within discourses and not outside of them (Purvis and Hunt 1993:488). This view requires analytical attention on the power struggle over the social and political determination of discourses and discursive practices (Fairclough 1993:51). Resistance to dominant social discourses is constituted by the production of counter discourses:

When studying power, the empirical question for Lukes is `who exercises power over whom' (Hoy 1981:53). For Foucault, the question is `how is government possible?', where `to govern' means to "...structure the possible field of action of others" (Foucault 1982:790). Foucault's concept of power leads to analysis of the functioning of discourse, of discursive struggle and discursive change. This has sparked a very complex debate about the facilitative role of discourse and its connection to power/knowledge/truth and to the material constitution and defence/enhancement of collective positions.

In much of post-structuralist writings the words `language' and `discourse' are often used interchangeably. They are used to refer to social and political practices. Discourse is "...a form of social practice, rather than a purely individual activity or a reflex of situational variables" (Fairclough 1992:63). The rejection of traditional notions of truth and reality is central in post-structuralists' positions. It is via relations of power and discourse that particular regimes of truth are established: "...power enables some to define what is or is not considered knowledge" (Riger 1992:734):

The importance of language and discourse centres on their function in constituting social realities, individual, as well as group, identities and positions:

The central claim here is that language and discourse are not dependent on, nor simply reflective of, deeper social structures (eg, the economy). This claim has been central to post-structuralist positions, especially deconstruction and post-Marxism, which have struggled against the weight of structuralist dogma that in the end everything, including discourse, is determined by a structured material base. The counter argument rests on a thesis of the non-essentialist character of structures (eg; institutions, technologies, modes of production), which in the end are constituted and given meaning via discourse and discursive strategies.

The dispute over the determination of the discursive by material non-discursive forces, or the contrary position, can best be resolved by an understanding of Foucault's position on the materiality of discourse. Foucault was primarly concerned with the materiality of language and discourse (Schottler 1989:45). Discourse is a thing in itself and not just a representation in terms of the thing referred to. Discourses have a material existence:

Foucault clearly refers to the materiality of discourse, and he defines discourse as a material practice. Central to Foucault's proposal that discourse links power and knowledge (Foucault 1978:100) is the view that `thought', statements, and groups of statements making up a discourse, do not emerge as pure events. Rather discursive practice takes place in an already given field of possibilities which it can realign or reconfigure. Foucault conceptualised discourse as operating in an existing material world: against and in relation to prior and contemporary discourses; other discursive formations; and social, political and economic material forces (the non-discursive). Discursive practices are both contingent and strategic and operate in what Foucault called a "field of strategic possibilities" (Foucault 1972:37):

In the `field of possibilities' discourses are deployed where they can be "...both an instrument and an effect of power" (Foucault 1978:101). Their deployment is all about the practical struggle for the control of discourses and the effects of discursive practice. Foucault was particularly concerned with the historical deployment of discourses in political struggles and processes of social change. Foucault posed the question, "How is struggle for the control of discourses conducted between classes, nations, linguistic, cultural or ethnic collectivities" (Foucault 1991a:60). His theoretical and empirical focus was the material formation of discourse and the transformation or mutation of discourse which produce social change.

A discursive formation is constituted when there is a correlation and regularity among a number of discursive statements dealing with an object of discourse. For Foucault, a `discursive formation' operates by exclusion. He suggests that at any given moment particular statements are capable of having a truth function while others are excluded. This conception of Foucault has been put in the following way:

In examining political struggles, one should attempt to look at the contest between discourses for the discursive dominance of the object of struggle; for the control of a discursive formation. In order to do this we need to examine both the form, content and permeability of discourses making up the discursive field of conflict and struggle. What is important is how the political strategies of conflicting forces are deployed in discourse and how these discourses are articulated across a chain of successive events (Foucault 1972:167).

While Foucault pointed out that discourses are formed in the context of prior and contemporary discourses and the non-discursive conditions in which they are located, he also showed the vital role of transformations or mutations of discourse in producing important events of discursive and social change. Foucault sees transformations or mutations of discourse as operating in relational dependencies at three levels: intradiscursive dependencies (between the objects, operations and concepts of a single formation); interdiscursive dependencies (between different discursive formations); and extradiscursive dependencies (between discursive transformations and transformations outside of discourse) (Foucault 1991a:58). Foucault's Madness and Civilisation (1972) and Birth of the Clinic (1973) are major historical studies of the correlations between critical transformations in existing medical discourses and "...a whole play of economic, political and social changes" (Foucault 1991a:58). He said he was concerned to discover and describe "...the moment at which the mutation in discourse took place..." (Foucault 1976:xi). He went on to say:

Critical discursive events or moments of discursive tranformation or mutation are important because of the immense power effects they have, operating as points of change in a discursive field that provides a strategic hindrance or blockage, a point of resistance, or a starting point of an oppositional strategy (Foucault 1978:101). A critical discursive event involves an utterance or statement which re-aligns or alters the configuration of a discursive formation. It is from such mutations of discourse that different types of statements are granted a validity; new divisions between discourses are made and `truth' is attributed to a new type of statement. Moments of transformation can realign a discursive struggle, or if powerful enough, they can effect the reconfiguration of an entire discursive field of conflict. These critical events of discursive transformation produce new distributions and relations of power.

Foucault's concepts dealing with: (1) the formation/development of discourses; (2) their deployment in struggle; (3) their relative articulation over a series of historical events; and (4) their transformation or mutation; will be used as key theoretical concepts in the historical narrative case study of contested native forests in the next part of this dissertation. Here, I make a number of further explanations and concluding points regarding Foucault's theory of discourse and power.

In Foucault's view, discourses are not representations of distorted reality, which is the traditional conception of ideologies, but rather they should be seen as socially constructed "...economies with their own instrinic technologies, tactics, effects of power, which in turn they transmit" (Purvis and Hunt 1993:488):

Foucault is consistently materialist in asking how discourse functions (Rabinow 1991:10). Discourses, as material practices "...make statements possible, and discursive formations are made up of groups of statements" (Cain 1993:76). These statements are assets, "...an asset that consequently...poses the question of power; an asset that is, by nature, the object of a struggle, a political struggle" (Foucault 1972:120). If we look at discourses this way, as assets which constitute things, we must also recognise that the ability to make statements, the ability to produce dominant discourses, and the right to draw on pre-existing discursive statements (to transform discourses), are unequally distributed between individuals and social groups. Foucault recognised this point saying, "Power is not something that is acquired, seized, or shared, something that one holds on to or allows to slip away; power is exercised from innumerable points, in the play of nonequalitarian and mobile relations" (Foucault 1978:94). Foucault had earlier commented on the inequitable distribution of power in the following way:

In investigating discourse as a specific mode of production we need to ask questions concerning who has access to the machinery of discourse, who is consequently authorised to speak, to make statements. On a related point, Foucault has been accused in his empirical work of analysing discourses as predominantly professional products generated in institutionalised sites of production (Fraser 1989, Hennessy 1993:40, Purvis and Hunt 1993:489). This emphasis, it is suggested, can and has led to a focus on the importance of institutional and professional construction of knowledge and policy discourses. Fraser comments:

Whether these critics of Foucault are right is debatable. However, the important point is that in investigating discourse we need to account for the active role of struggle and contestation over knowledges and the impacts of alternative non-institutional discourses in the formation of public policy (Fraser 1989:185, note 21). On this point it is important to remember that Foucault said "Where there is power, there is resistance" (Foucault 1978:95), and further "Discourse transmits and produces power; it reinforces it, but also undermines and exposes it, renders it fragile and makes it possible to thwart it" (Foucault 1978:101).

Foucault's position on power/knowledge/discourse provides a powerful theoretical framework for an analysis of environmental conflicts and the struggle over the knowledges and the always provisional and contingent `truth' governing `nature'. In order to expand our framework concerning the constitution of environmental struggles it is important to consider the discursive functions and operations of government. Again I start with Foucault's concept of `governmentality' and then move to the associated analytics of `problematics of government' recently developed by Rose and Miller (1992). These concepts are used to understand the recent growth of the government of `nature'. This is the subject of the next section.

Foucault's approach to power was established through a large number of studies of the microphysics of power, in particular those micro-practices and techniques which emerged during seventeenth and eighteenth centuries which were evidence of the strategic linkage of power/knowledge/discourse. However, in the latter part of his life Foucault's attention turned to reflections on the operation of modern power in its global or macro settings. From his historical research Foucault detected that the origins of modern power, in the macro sense, could be found in late sixteenth century Germany with the advent of Polizeiwissenschaft, or `science of police' ("...the English word `policy' is arguably a better equivalent to this meaning of Polizei." [Gordon 1991:10]).

In this early `science of police' Foucault saw the first manifestation of a new mentality of government. He termed this particular historical development `governmentality'. By this term Foucault "...sought to draw attention to a certain way of thinking and acting embodied in all those attempts to know and govern the wealth, health and happiness of populations (Rose and Miller 1992:174). This term `governmentality':

`Governmentality' involved the rapid development of new and historically novel apparatuses and techniques of government concerned with the production of rational knowledges calculated to inform public policy and administration. By these developments Foucault was not referring to the advent of the modern state, he points out that what is historically important "...is not so much the State-domination of society, but the `governmentalisation' of the State" (Foucault 1991c:102-103). This `governmentalisation' was based on the "...proliferation of a whole range of knowledges and `know-how' about government, the means of its exercise and the nature of those over whom it was to be exercised" (Rose and Miller 1992:174). For Foucault, `governmentalisation' was achieved by an "...ensemble formed by the institutions, procedures, and reflections, the calculations and tactics, that allow the exercise of this albeit complex form of power" (Foucault 1991c:102).

The concept of `governmentality' characterised the macro dimensions of the operation of modern power for Foucault; it also continued his concern with the `how' questions concerning power, "...its mechanisms, its techniques, its objectives and effects, can be asked of the `global' forms of power just as they can be asked of the micropowers" (Dean 1994:179). It is from Foucault's `governmentality' concept and his emphasis on the `how' questions concerning power that Rose and Miller have formulated the notion of `problematics of government'.

A `problematic of government' is something which has been "...rendered practicable and amenable to interventiion" (Miller and Rose 1990:5). `Problematics of government' concern "...the associations formed between entities constituted as `political' and the projects, plans and practices of those authorities...who endeavour to administer the lives of others..." (Rose and Miller 1992:175). These `problematics' are not pre-given, rather they need to be problematised as such "...and rendered amenable to administration" (Miller and Rose 1990:4). A consequence of this view is that objects of policy, that is `problematics of government', such as crime, poverty, employment and unemployment, health, education and, as argued in this dissertation, `nature' are things to be explained.

`Problematics of government' are constituted in the knowing and representation of a problem and the intervening in administration of the known problem. Knowledge and knowledge production play a fundamental role in "...rendering aspects of existence thinkable and calculable, and amenable to deliberated and planful initiatives" (Miller and Rose 1990:3). The view that `governmentality' is discursively created is central to Foucault's concept of `governmentality' and in the notion of `problematics of government':

`Problematics of government' may be analysed in terms of their political rationalities and their governmental technologies and the inter-dependencies between these two elements (Rose and Miller 1992:175-176).

Firstly, political rationalities refer to the objects which have been made the subject or domain of government. They represent "...idealised schemata for representing reality, analysing it and rectifying it" (Rose and Miller 1992:178). They result from political discourse and, most often from discursive conflicts, over both the justification and formulation of governmental intervention. Political rationalities display a moral form in the reasoning for governmental actions and in the formulation of objectives or goals directing such action. Rationalities have an epistemological character in that they contain particular bodies of knowledge concerning the nature of objects to be governed. And rationalities are articulated via languages which aim to make objects of policy thinkable in ways conducive to political deliberation. In summary, political rationalities are "...morally coloured, grounded upon knowledge, and made thinkable through language" (Rose and Miller 1992:178-179).

Political rationalities result from the problematising activities of government (Rose and Miller 1992:179). Involved are the critical activities in which the assembled knowledges of problems or policy areas are translated into programmes of government aimed at intervention and management. Programmes of government then are based on the goal of reformulating reality which in turn rests on the view that reality is programmable (Miller and Rose 1990:5, Rose and Miller 1992:181-183). This perspective adds to what has been already contended in this dissertation (see Chapter Two). Clearly `nature' is not a pre-given object but is an example of an `empiricity' constituted through a problematics of government which produces `nature' as a thing to be known and governed.

Secondly, it is via technologies of government that "...political rationalities and programmes of government that articulate them become capable of deployment" (Rose and Miller 1992:183). The representation of objects to be governed is a technical process which involves government in a massive activity of inquiry to transform events and phenomena into information. Expertise and experts are critical in the information process. Information is not a neutral recording process "...it is itself a way of acting upon the real, a way of devising techniques for inscribing it in such a way as to make the domain in question susceptible to evaluation, calculation and intervention" (Rose and Miller 1992:185). Technologies of government involve an extremely large range of procedures, mechanisms and instruments of representation and intervention:

The concepts of `governmentality' and the `problematics of government' extend the approach that power operates as discourses and discursive struggle. This approach provides a framework to analyse environmental conflict. It also offers the means to illustrate the thesis advanced in Chapter One, that the generation of a contested public discourse over the question of `nature' has produced outcomes far from the liberatory aims of modern environmentalism. To date, the tendency has been for `nature' to become policed, further controlled and governed in system management strategies, and locked-up in new regimes of truth and policed through instrumental and technocratic administrative regimes. What is not questioned is the discursive production of `nature' and the embedded power effects of this discursive formation.

The rapid politicisation of `nature' since the 1960s has constituted a new problematic of government. Governments have responded to this politicisation by developing discursive and administrative interventions over issues raised in the new environmental agenda. Such interventions include investigative, administrative and regulatory activities where both the production of knowledge and definitions of environmental problems are controlled and managed, and the responses to environmental problems are directed via legislation, the influence of specialist agencies and technologies, and the growth of professional specialisations. Governments at all levels have set up technologies and modes of inquiry aimed at knowing, quantifying, assessing and managing the impacts of human actions on the environment. In many cases, entirely new mechanisms have been developed to facilitate governmental interventions, such as environmental/social effects statements, environmental monitoring indexes and environmental risk assessment. The ways in which these processes of `governmentality' are created and deployed is contingent on context, time and place. In Part 111 of the dissertation we will analyse how these processes were developed and deployed in the case of political conflict over the East Gippsland forests.

An historical account of the development of the constitution and technologies of this new problematic of `nature' is an immense task still to be written. The remainder of this section will briefly discuss some important indicators of the extension of the administrative state to the government of `nature'. The focus here will be the central discursive technologies and mechanisms characterising the approach of the state in Australia. These will be a major focus in the case study of contested native forests.

The advent of environmental conflict in Australia during the 1960s lead to a rapid response from government centred around extensions of public participation to environmental inquiry and decision making. Such exercises usually took the form of open public inquiries with formal procedures for public input into environmental problems. Inquiries where either one-off efforts or, as has increasingly become the case, involved the establishment of permanent bodies of inquiry (notably, in land- use and natural resource decision making). With the appointment of non-government persons to the role of chairperson, these inquiries were promoted as neutral or `hands off' from government.

These exercises linking participation with open inquiry have been called `discursive designs' (Dryzek 1987, 1992). They represent organised attempts at the construction of public spheres based on the view "...that legitimate policy decisions now require not just expertise and the backing of constitutional authority, but also informed participation by affected parties" (Dryzek 1992:34). Discursive designs produce "...a rational and scientific administrative discourse out of the raw materials of political struggle and debate" (Ashforth 1990:3). They represent mechanisms by which the state attempts to reign in contested public discourse by functioning as centralised points for the conduct of discourse and for the production and dissemination of public knowledge. They effect the translation of problematised issues and demands into the subject of policy and render them into items that can be administered.

Public participation has been a central demand of the environment movement. However, the opening of possibilities for participation in new discursive designs focussed on environmental inquiry and decision-making can be distinguished by the relative amount and quality of the participation offered to oppositional groups (Arnstein 1969). Further, the actual practice of these discursive designs has been called incipient "...on the grounds that they represent very imperfect approximations to ideals of free discourse (and occasionally gross violations of these ideals)" (Dryzek 1992:33). This assessment is based on evaluations of actual discursive designs to determine to what extent they approximate free and open public spheres of discourse (Dryzek 1990, Kemp 1985, 1987). In their Habermasian sense, discursive designs are not regarded as constituting autonomous public spheres because they are created, sponsored and resourced by the state (Dryzek 1992:32-33). The evaluations undertaken by Dryzek and Kemp have been based on Habermas' ideal of free and open public spheres of discourse, a clearly different meaning of discourse from that developed by Foucault. In this dissertation the concept `discursive design' is used in Foucault's sense as a technology of governmentality concerned with domains of inquiry and official discourse. In a more Foucauldian influenced view, discursive designs are `reckoning schemes of legitimation' and function as discursive mechanisms of state legitimation, incorporation and control; they are "...ways of speaking about social life which make possible the work of organising political subjection" (Ashforth 1990:17).

The critique of discursive designs, outlined above, has been extended to cover two additional discursive mechanisms of government prominent in the Australian case. First, the mandatory conduct of publicly-open environmental impact statements on public and private developments which may effect the environment (directly imported from the USA) are regarded as legitimation exercises, window-dressing decisions already taken or taken elsewhere, and co-opting opposition (Amy 1990, Dryzek 1992, Formby 1986, 1987a, 1987b, Kemp 1985, Mercer 1987). Second, incorporation of oppositional groups is regarded as the aim of attempts by the state to establish corporatist organisations formally bringing together government, industry and the peak environment groups, with the aim of producing consensus based environmental policy (Economou 1993).

As well as functioning as government technologies aimed at legitimation, incorporation and control; discursive designs, environmental impact statements and corporatist policy processes all generate discursive products in the form of `official state discourses'. These `official discourses' function to provide legitimate statements of `truth' and contain the programmes of government which Rose and Miller (1992) have identified as essential to `governmentality'. All this discursive activity of government is made possible by the production and assembly of knowledge. The government of `nature' is thus underpinned by experts and expertise engaged in knowledge production. Knowledge and the command of professional specialisations is important in order to participate in institutional mechanisms of government. The conduct of these mechanisms of government can privilege the knowledge, competence and qualifications of particular social groups and institutions (Gismondi and Richardson 1991:48). So much of the business of governing `nature' has become centred on professional discourses implicated in how issues are defined, and are crucial in determining the legitimacy of who may be heard and what is left as unspoken (Cohn cited in Gismondi and Richardson 1991:44).

In the above argument concerning the governmentality of `nature', with its reliance on professional expertise and technical discourses, there is the suggestion of overall dominance and control of `nature' by the state and its professional operatives. That there are no possibilities for resistance should be rejected. As a recent study of public hearings on a Canadian pulp mill development has shown, less powerful social groups, can, with well-organised action, "...undermine the discourses empowered by the dominant groups and in the process construct(ed) counter-discourses of their own" (Gismondi and Richardson 1991:49). Despite discursive governmental technologies being mechanisms of control, organisation and normalisation, oppositional participation in such technologies can in some cases "...constitute worms in the brain which further undermine the logic of administration and the skewed distribution of power in liberal democracy" (Dryzek 1992:34). This again refers to Foucault's contention that where there is power there is resistance (Foucault 1978:95).

The outline of problematics of government in this section complements Foucault's materialist-discourse approach to political power. The next task is to move from the broad theoretical discussions to the methodological considerations involved in a discourse-oriented empirical study of environmental conflict.

In this and the previous chapter a broad framework for theorising contemporary environmental conflicts, the socio-historical questions it poses and its political significance has been proposed. The task now is to move to a methodological position enabling empirical studies of environmental conflict and the outlining of the approach to a case study of environmental struggle and contest. The objective is to produce a theoretically informed descriptive and interpretive case study of conflict over the native forests of East Gippsland, Victoria since 1968, where the focus is discursive conflicts and their relationship to the material, structural and institutional contexts of the case.

Earlier in the chapter the major methods associated with studies of Lukes' three dimensions of power were briefly outlined. Here a brief outline of Foucault's methodologies is provided, with some additional points concerning historical discourse methods. The method adopted in the case study of this dissertation is then outlined.

In exploring Foucault's method (or methods) it needs to be noted how little he actually said on methodology (Dean 1994:14). Foucault characterised the method of his early historical studies of particular knowledges and their discursive regimes as `archaeology' which concerns the "...pure description of discursive events" (Foucault 1972:27). This archaeological method and analysis attempts to show "...the positivity of discourses, their conditions of existence, the systems which regulate their emergence, functioning and transformation..." (Foucault 1991a:69).

Foucault distinguished his archaeological method from those of structuralism and hermeneutics (often known as the `interpretive' method) which are based on an epistemology of the knowing subject and represent the major philosophies and methodologies in the social sciences. Foucault's archaeological method is "...not yet another mode of interpretation rendering into discourse the unsaid. It concentrates all its focus on what is said" (Dean 1994:16).

The methodological focus of archaeology is the positive and material reality of discourses. As a method it is highly descriptive and can be immensely detailed. The archaeological method hinges on Foucault's concept of the `archive'. He said the archaeological method is "the description of the archive" (Foucault 1991a:59) and his definition of the archive is "the general system of the formation and transformation of statements" (Foucault 1972:130). This method determines Foucault's view of documents and sources and their methodological use. His purpose and method is not to interpret documents or other data sources seeking to unearth hidden meanings, but rather his archaeological method "...organises the document, divides it up, distributes it, orders, arranges it in levels, establishes series, distinguishes between what is relevant and what is not, discovers elements, defines unities, describes relations" (Foucault 1972:6-7).

In his later political studies Foucault developed a complementary method which he called `genealogy'. Genealogy is a more politically engaging method which continues his prior concern with discourses and discursive formations. What Foucault adds with this method is a shift to the centrality of power and a focus on strategies, techniques, and the institutional mechanisms where oppositions between true and false statements are historically created.

The genealogical method continues Foucault's rich historical and social descriptive techniques. "Genealogy is gray, meticulous, and patiently documentary...and it depends on a vast accumulation of source material (Foucault 1984:76). In his attention to power Foucault's genealogical method does not attempt to legislate on legitimate or illegimate uses of power; rather he focuses on the actual ways and means of the operation of power. Its focus then, is on the deployment of discourses, and:

The genealogical method, based on description, attempts to reveal terrains of forces or power relations where discursive relations are constituted and changed. It is about the authorisation of discourse, the mechanisms of who can speak and the location of people, groups and institutions within a prevailing power/knowledge network. "Genealogy will uncover a positive and productive form of power underlying every movement of institutional or discursive delimitation of statements" (Dean 1994:33). It is also a method where one can examine discourses that are discounted or repressed within political conflicts.

Foucault's methods are based on detailed historical and social documentation and a reading of it, in order to show the internal and external constraints (or non-discursive structures), and the "...institutional systems that envelop discourse and subjects it to forms of exclusion, rarefaction, and appropriation" (Dean 1994:33).

The link between Foucault's `pure description of events' and causality can be seen in what he was to say, in a round-table debate in 1977, about `eventualisation' as a procedure of analysis. Methodologically, this is a highly important statement:

In many respects Foucault's method of eventualisation, which produces descriptions of causal processes and mechanisms, is similar to what is achieved in Gaventa's study. Like Foucault's studies of power, Gaventa uses intensive and qualitative case study techniques where a rich descriptive narrative is designed to reveal the multiple causal mechanisms of power so conceived. His epistemological framework accords with the realism advocated by Bhaskar (1979); it is based on the central research task of describing causal mechanisms. This realist approach amounts to "...formulating tentative explanations for the phenomena in question and then attempting to supply descriptions of the causual mechanisms posited by the proferred explanation" (Shapiro and Wendt 1992:213).

The history of discourse analysis has produced a variety of research strategies. Notable examples include Robin's use of Harris' lexicometric procedure to quantitatively analyse the use of specific words within a discourse (Schottler 1989:42-43) and, Pecheux's `automatic discourse analysis' which employs computer programmes to identify discourse formations in a corpus of texts (Fairclough 1992:30-35). However, the research strategy employed in this case study of environmental conflict remains close to the historical narrative method used by Foucault and Gaventa with an attention to `multiple processess' and no single site of causality (neither intention nor ideology).

The empirical research objective of this dissertation is the production of a descriptive and interpretative case study of discursive conflict surrounding native forests. This requires a reconstruction of historically specific `discursive economies' of the case, focussing on `key words' and `key statements' which have exceptional social or political significance within the events of the case and which become part of a historically new discursive formation on the forests. This involves tracing the names given to background concepts and knowledge positions -reading the key thematic issues to emerge - within the production of the discourses of various social groups and institutions. This method can be used to highlight how particular discourses involved in conflict are historically located and socially classed (McHoul and Grace 1993:28).

The historical trajectory of any discourse cannot simply be read off from texts and other sign systems. Discourses must be situated and analysed within particular and complex sociological and historical processes. This requires social-historical analysis. In this case study I will be concerned with the relationships between discursive conflict and the material, structural and institutional forces outside of discourse (the non-discursive) which are involved in the case. Regarding the latter, particular focus is placed on the socio-historical background and contemporary structure of the political economy of wood production in native forests, including the technologies embedded within this economy.

There are then, two parts of the method; (1) description of discourses; their formation/development, their deployment in struggle, their relative articulation over a series of historical events, and their transformation or mutation, and (2) social-historical analysis of the non-discursive context of a discursive conflict. Combined, these two parts of the research method represent the archive of the case. This archive provides the researcher with the base sources for undertaking what Foucault has described as `eventualisation': the procedure of establishing and describing a multiplication of forces, events, causes, and effects which constitute the case.

The final stage of the research strategy involves the critical reading and analysis of both the archive and its description in terms of `eventualisation'. This is always influenced by theory construction: in this case, it involves both the tracing and mapping of, and judgement about, the power effects of material contexts and discursive strategies. However, it is the power effects of discursive strategies which is the dominant focus. This involves the critical task of attempting to weigh up the relative `value' of the differing discursive statements making up the archive material. And following Foucault, the `value' of discursive statements refers not to their truth or the hidden content they might contain, but rather to the very place particular statements occupy and to "...their capacity for circulation and exchange, their possibility for transformation, not only in the economy of discourse, but, more generally, in the administration of scarce resources" (Foucault 1972:120).

This reading and analysis provides the basis for detecting which tactics and strategies of discourse have been successful and via what techniques. It should be possible then to assess outcomes of conflict and to specify certain discourses and positions that have been denied or ignored within the conduct of the conflict. This assessment would not be the calculative exercise which as noted is common in conventional models of power, but rather an open, positional and strategic assessment/judgement. It is on this basis then that proposals for change can be advanced. I will be particularly concerned to identify any possibilities which may be able to link social and ecological justice within the context of the case.

It is necessary to conclude this chapter with a methodological comment on the research procedure and source material used in this case study.

The first research task was to locate, collect, and assemble both the primary and secondary source material covering; (1) the historical background to the case study period, and (2) the period 1968 to the around mid-1996. The highly discursive nature of the East Gippsland forest conflict has produced a rich and voluminous primary source of documents. A consequence of this discursive density is that the research archive can at best only be a selected sample of the material. The sampling method used involves an initial reading of documents, selecting out key central discursive statements and isolating objects of discourse. It is these `statements' and `objects of discourse' which make up a conflictual public discursive formation on the forests, one that forms an ongoing `problematic' for government. In this sampling I attempt to trace and map out the elements or `eventualisation' of the case. This sampling is based on the theoretical concepts (outlied in this chapter) concerning central and powerful discourses; their formation/development, their deployment in struggle, their relative articulation over a series of historical events and their transformation or mutation.

The second research task is a detailed reading of the sample archive of the case. Here I attempt the judgement/analysis of conflictual discursive events. I also locate this discursive conflict in relation to and in correlation with the key non-discursive elements of the case. This reading identifies the power effects and outcomes of this event in forest politics. I record my descriptive and analytic distillation of the events of the case study in historical narrative form. This is developed in the next four chapters.

Additional to the archival material available in the public domain, I was able to draw on source material collected as part of my involvement in aspects of the case study events. Such involvement enabled access to some non-public material and insights into the corporate `resource politics' involved in the case. It also provided access (with permission) to private research studies conducted by the CSIRO for T J Andrews Sawmills P/L.

In addition a small number of semi-structured/conversational type interviews were conducted with selected participants in the case. These were carried out in order to clarify certain events of the case. They are cited in the text as a personal communication (`Name: per comm'). A brief note on each person interviewed is contained in the bibliography. Where it has been required to protect the identity of the source, the reference in the text is provided as `Anon: per comm'.

Power involves struggle over discourse which is always subject to the play of alternatives, or in Foucault's terms, the struggle over discourse occurs in a `field of strategic possibilities'. This chapter has been concerned to develop a theoretical approach to the questions of power, struggle and change that can be used to study modern environmental conflicts. While conventional theories of power, especially Lukes, have produced some significant empirical studies, the approach adopted is based on Foucault's theory of the material existence of discourse and his linking of power/knowledge/discourse. The chapter has placed emphasis on the processes of modern governmentality which give form to both power and discourse. Knowledge, its production and the various governmental techniques of organising and handling discourse and discursive conflict, have been highlighted as important areas to theorise.

Further, the political conflict of `nature' has formed a modern problematic of government producing a variety of discursive designs which seek to organise, define, contain and control environmental conflicts. These designs have been productive of official discourses of the state, where governments and others have sought to govern `nature'. It has also been suggested that these official discourses represent certain positions, containing only certain definitions or orders of `truth' which enable certain actions and deny others.

Also, in the previous chapter it was argued that the recent politicisation of `nature' in the post-industrial west has been driven by intellectually trained people normally engaged in state cultural and social functions. This political action has constituted a political class which has been conceptually called the `new class'. Consequently, the conflicting discourses on `nature' can be said to be socially classed. This is why it has been argued in this dissertation that class and class relationships are important in understanding contemporary environmental conflicts.

The purpose of this and the previous chapter has been to propose a theoretical framework in which to analyse environmental conflict. This discussion then provides a broad and provisional set of theoretical concepts to analyse and interpret actual cases of environmental conflict. In the case study to follow, an historical narrative method is used to describe, analyse and interpret a multiplication of forces, events and causes which constitute the history of environmental conflict over the forests of East Gippsland in Victoria.

The next four chapters contain the East Gippsland case study. In a further chapter aspects of the case study and the theory proposed are brought together in a summary discussion, and a proposal is outlined for change in wood production and forest use.