In search of arenas for democratic deliberation: a Habermasian review of environmental assessment

Transcription

1 Impact Assessment and Project Appraisal ISSN: (Print) (Online) Journal homepage: In search of arenas for democratic deliberation: a Habermasian review of environmental assessment Hans Wiklund To cite this article: Hans Wiklund (2005) In search of arenas for democratic deliberation: a Habermasian review of environmental assessment, Impact Assessment and Project Appraisal, 23:4, , DOI: / To link to this article: Published online: 20 Feb Submit your article to this journal Article views: 372 View related articles Citing articles: 31 View citing articles Full Terms & Conditions of access and use can be found at Download by: [ ] Date: 04 December 2017, At: 00:24

2 Impact Assessment and Project Appraisal, volume 23, number 4, December 2005, pages , Beech Tree Publishing, 10 Watford Close, Guildford, Surrey GU1 2EP, UK Democratic deliberation In search of arenas for democratic deliberation: a Habermasian review of environmental assessment Downloaded by [ ] at 00:24 04 December 2017 Hans Wiklund In this paper the deliberative potential of environmental assessment (EA) is explored. The analysis is structured around four principles derived from Jürgen Habermas s conception of discourse as an ideal procedure for rational and democratic decision-making. The results show that there are many barriers to an implementation of the Habermasian principles. Nevertheless, it is concluded that EA has a hidden deliberative potential, which follows from the institutional flexibility of EA, that is, because EA legislation specifies minimal requirement for public participation, requirements that developers and authorities can exceed by using more inclusive and more dialogue-based participatory tools. Finally, the need for additional investigations of deliberation in EA is discussed. Keywords: environmental assessment, deliberation, deliberative democracy, communicative planning, public participation Hans Wiklund is in the Department of Political Science, Jönköping International Business School, PO Box 1026, SE Jönköping, Sweden; Tel: ; Fax: ; Hans.Wiklund@ihh.hj.se. The author wishes to thank Aleg Cherp and Lars Emmelin for their helpful comments on a previous version of this manuscript, and two anonymous reviewers for their valuable comments on an earlier draft of this article. AMORE DELIBERATIVE style of decisionmaking has been put forward as a way to revitalise democracy at a time when citizens trust in established political institutions is dwindling. Deliberation can be defined as dialogue that induces reflection upon preferences in a non-coercive fashion (Dryzek, 2000, page 2), and it is assumed to lead to better decisions. Deliberative democratic theorists claim that deliberation increases the legitimacy of decisions by giving people a fair chance to have their views heard and considered. Deliberative planning theorists claim that deliberation enhances the rationality of decisions by integrating the local and situated knowledge of ordinary citizens (Fischer, 2003; Hajer and Wagenaar, 2003). A key challenge for deliberative democratic and deliberative planning theorists is bridging the gap between theory and practice (compare with Kymlicka, 2002, page 292; Uhr, 1996). At what levels should arenas for deliberation exist local, national or international? How should these arenas be integrated into the public policy-making process? Is the goal to make established democratic mechanisms more deliberative or to create novel arenas for deliberation? These questions, regarding the implementation of deliberative ideals, are currently being addressed in the literature. A number of deliberative tools have been developed, such as planning cells (Daniel and Renn, 1995), citizens juries (Crosby, 1995), deliberative opinion polls (Fishkin et al, 2000), consensus conferences (Joss, 1998), authentic dialogue (Innes and Booher, 2003) and deliberative mapping (Eames et al, 2004). There is also an Impact Assessment and Project Appraisal December /05/ US$08.00 IAIA

3 increasing interest in exploring the deliberative potential of established political and administrative institutions (Eriksen and Fossum, 2000; Premfors and Roth, 2004; Uhr, 1996). Recently, environmental assessment (EA) has been characterised as a deliberative tool with the potential to improve environmental decision-making (Barber and Bartlett, 2001; Bond et al, 2004; Elling, 2004; Palerm, 2000; Petts, 2000; 2003; Sager, 2001; Wilkins, 2003). As Petts puts it: EIA (including SEA) has the potential to be a decision process which includes deliberation, inherent learning and decision influence through stakeholder and public input (Petts, 2003, page 373). Richardson (2005, pages ) expresses a similar idea: What has been described as the communicative turn in planning seems to be repeating itself in EA. There is also a limited but growing number of empirical studies of deliberation in EA (for instance, Diduck and Mitchell, 2003; Fitzpatrick and Sinclair, 2003; Petts, 2000; 2001; 2003; Saarikoski, 2000; Sinclair and Diduck, 2001; Soneryd, 2002; Webler et al, 1995). The main aim of this article is to contribute to our understanding regarding the extent to which EA as a decision support tool enables citizens to participate in inclusive and deliberative processes. In other words, the purpose is to analyse the potential of EA as an arena for democratic deliberation. Another aim is to provide a democratic theoretical foundation for deliberative EA. This is necessary for understanding what kind of decision-making process and democracy deliberative EA is supposed to enhance. Certainly, there are theoretically informed models for studying deliberation in EA (Petts, 1999; 2001; Webler et al, 1995). But these models primarily connect the idea of deliberation as form of rational communication with notions of rational planning. By providing a thorough democratic theoretical grounding the paper complements the existing models. Method and material The deliberative democratic potential of EA is explored through an analysis of barriers to an implementation of four principles derived from Jürgen Habermas s notion of discourse as an ideal deliberative procedure. The main reason for focusing on Habermas s theory is that it is coherent and structured. Another reason is that most arguments for a more deliberative style of decision-making have been strongly influenced by his writings (Bohman and Rehg, 1997; Elster, 1998). In this connection, it should be noted that deliberative democratic and deliberative planning ideals in general, and Habermas s conception of deliberative democracy specifically, have been criticised. Political theorists have claimed that deliberative democratic and planning theories tend to disregard the importance of power and strategic action in political administrative decision-making and planning, and to overemphasise the importance and possibility of reaching consensus (for instance, Fishkin and Laslett, 2003; Macedo, 1999). Parallel criticisms have been put forward of EA as a deliberative tool (Richardson, 2005). The four principles derived from Habermas s notion of discourse are used as an ideal type. This is a conceptual construct that stylises a phenomenon, and an analytical tool that can be used for studying practice (Eliaeson, 1982, pages ). The difference between an ideal type and a hypothesis becomes clear when the two are confronted with empirical findings. While the hypothesis claims to describe and explain reality, the ideal type provides a perspective from which practice can be viewed. Accordingly, an ideal type, by definition, cannot be falsified; it can only be found to be less useful in some situations than others (Petersson, 1989, page 30). The Habermasian principles are used as a lens through which EA is viewed with regard to deliberative potentials and qualities. Given the normative nature of Habermas s theory, the principles could also be used as a deliberative democratic standard of best practice of public participation in EA. There are several such standards (for instance, André et al, 2004; Bond et al, 2004; Palerm, 2000; Webler et al, 2001). A weakness with most of these standards is that their democratic theoretical foundation is somewhat unclear. By describing Habermas s model of deliberative democracy and deriving the four principles from this model, the ideal type (and the potential standard of best practice of public participation) is given a solid democratic theoretical foundation. It should be noted that the focus is exclusively on deliberation in EA; deliberation in other contexts is beyond the scope of this article. The analysis is based on a selection of recent investigations of public participation in EA and the results are consequently not claimed to be fully representative of the public participation situation, but rather to be illustrative. Accordingly, the assessment of the deliberative potential of EA is based primarily on secondary empirical materials. In the next section of this paper Habermas s notion of democratic deliberation is described. Then four principles are derived from Habermas s notion of discourse as an ideal procedure. In the following section, the potential of EA as an arena for democratic deliberation is explored. Finally, conclusions are drawn regarding the potential of EA as an arena for deliberation, and the need for additional research on deliberation in EA is discussed. Deliberation and democratic decision-making There are many models of deliberative democracy, such as discursive democracy (Dryzek, 1990), contestatory democracy (Pettit, 2000), reasonable democracy (Chambers, 1996), communicative 282 Impact Assessment and Project Appraisal December 2005

4 In his model of deliberative democracy, Habermas formulates a notion of democratic politics by extracting the most attractive elements of the liberal and the republican traditions in political theory and incorporating them into a communicative framework democracy (Young, 1993), and deliberative politics (Habermas, 1996a; 1996b). In addition, there are several models of deliberative planning, for instance, communicative planning (Forester, 1993; Sager, 1994), collaborative planning (Healy, 1997), planning through consensus building (Innes, 1996) and deliberative planning (Forester, 1999). What these models have in common is the emphasis on the significance of voice, even though they disagree about what forms of voice should be enabled. In his model of deliberative democracy, Habermas formulates an alternative notion of democratic politics by extracting the most attractive elements of the liberal and the republican traditions in political theory and incorporating these elements into a communicative framework (Habermas, 1990; 1996a; 1996b; 1999; 2001). For pedagogical reasons, these two traditions are presented in a highly stylised manner in this discussion. It should be noted that the focus here is limited to the process level of Habermas s two-track model of deliberative democracy, that is, to the notions of deliberative politics and discourse as an ideal democratic procedure. The system level of his theory how the public sphere can steer the formal political system through procurement and withdrawal of legitimation is not described (see Wiklund, 2002, chapters 2 and 3 for details regarding the two-track model). Liberal tradition The liberal tradition typically characterises politics as private in nature and instrumental in purpose. 1 Political actors are assumed to enter the political arena with prepolitically formed, private interests. Also, they are supposed to be self-interested and strategically oriented, and are seen as ultimately sovereign in the political arena. This conception of political actors leads to an understanding of political practice as a competitive enterprise, as a market. Politics is pictured as a power struggle among self-interested actors who act strategically to achieve their private goals. As a result, political participation is viewed as purely instrumental, as a means of satisfying non-political ends. The market approach to politics even implies that an absence of participation is something positive, since less participation indicates less interference with the private interests of political actors (Elster, 1997, pages 3 34). Because political practice is depicted as a power struggle between opposed private interests, democracy simply becomes an instrument for aggregation of self-interested political actors private interests into social choices. Democratic institutions are justified on the grounds that they enable a fair and efficient aggregation of private interests into collective choices. In this context, fairness and efficiency refer to minimal interference with actors private-life projects. Accordingly, democratic legitimacy stems from the preservation of the individual freedom of the political actor and the non-interference with his/her private interests (Manin, 1987; Miller, 1992; Knight and Johnson, 1994). Republican tradition The republican tradition, in contrast, characterises politics as public in nature and non-instrumental in purpose. 2 It pictures politics, not as a market, but as a forum. Political actors are viewed as participants in public discussion, and politics is understood to be an open and public activity, distinct from the isolated and private choices of self-interested actors. Nor are actors supposed to enter the political arena with prepolitically formed private interests and strategic orientations. Instead, they are supposed to go beyond the private interests of the market and orientate themselves towards the common interest of the forum. Civic virtue, not self-interest, is considered to be the core organising principle of politics (Finley, 1973). This leads to a view of political practice as an open and public discussion in which the members of a community become aware of their dependence on one another and discover their common interests. Moreover, participation is not viewed simply as a means for political actors to reach non-political private ends. On the contrary, as emphasised by the theorists of participatory democracy (Pateman, 1970), participation is understood to educate political actors and transform their interests. Politics consequently tends to be viewed as an end in itself (Elster, 1997, pages 3 34). Hence, democracy is not solely or even primarily conceived of as an instrument for aggregation of private interests, but as a process of public discussion through which the members of a community discover their common interests. In the republican tradition there is, however, a tendency to define common interest in substantive terms, as shared traditions, values, virtues, and so on. Common interests are dependent on the existence of a homogeneous community, be it an ancient city-state or a modern Impact Assessment and Project Appraisal December

5 nation (Elster, 1997, pages 3 34; Held, 1996, pages 36 62). Politics consequently tends to become a succession of ethics, and substantive common interests the source of legitimation (Manin, 1987; Knight and Johnson, 1994). Deliberative politics So, the notion of deliberative politics represents an attempt to elicit the most attractive features of the liberal and the republican traditions and incorporate these into a communicative framework. The result is a conception of democratic politics as public in nature and, at least partly, instrumental in purpose. In agreement with the republican tradition, the alternative view of deliberative politics rejects the liberal conception of politics as private in nature and the related conception of political practice as a struggle among self-interested and strategically oriented actors. Rather than understanding political practice solely as a struggle, this view revives the republican conception of democratic politics as a public search for common interests. Political actors are viewed as participants in public discussions regarding common interests, and political practice is characterised as an open and public activity in which deliberation and participation form the core elements (Habermas, 1996a, pages ; 1996b; 2001). In agreement with the liberal tradition, the alternative view of deliberative politics rejects the idea that politics is an end in itself. Although politics is depicted as public in nature, this view does not reject the idea that private interests can and will conflict and that it is the function of political institutions to resolve such conflicts. However, conflicts and disputes are not resolved primarily through an aggregation of private interests into social choices, but through deliberation on the matter at hand with the objective of reaching a consensus regarding how it should be resolved (Miller, 1992, page 54). Accordingly, the republican view of politics as non-instrumental and mainly concerned with ethical questions of self-understanding, is too limited. Deliberative politics is understood to involve collective searches for common interests, and negotiation and bargaining between conflicting private interests (Habermas, 1996a, pages ; 1996b). Democracy is thus neither conceived of solely or primarily as a mechanism for fair and efficient aggregation of prepolitical, private interests, nor seen as a process of public discussion in which the members of a community discover substantive common interests. In contrast, democracy is conceived of as a process of social learning through rational argumentation (Bohman, 1990). It is described as a process in which participants exchange arguments and counterarguments and thereby become informed of the views of one another. Politics then is a search for generalisable interests, that is, a process in which the need to reach agreement in matters of common concern forces actors to put forward proposals under the rubric of general principles or policy considerations that others could accept (Miller, 1992, page 55). It is viewed as a process in which common interests are constructed, rather than discovered (McCarthy, 1992, pages 57 62). The deliberative process of crafting solutions to matters of common interest is a process in which actors initial private interests are transformed to take into account the views of other actors (Habermas, 1990, pages ; 1996a, pages ; 1996b). Deliberative politics refers to the kind of political practice whose source of legitimation is derived neither from actors prepolitical private interests nor from communities substantive common interests, but rather from the formation process of those interests (Carleheden, 1996, page 113). The regulative idea of deliberative politics is reaching consensus. This concept is foreign to both the liberal and the republican tradition. It is foreign to the liberal tradition and its emphasis on fair and efficient aggregation, in the sense of minimal interference with individual private interests. It is foreign to the republican tradition and its idea of discovering a community s substantive common interests through public discussion (Miller, 1992, page 57). In other words, the alternative view of deliberative politics is founded neither on instrumental rationality restricted by prepolitical principles of justice, nor on substantive common interests of a community, but on argumentation grounded in communicative rationality. If at all, political outcomes are legitimate because they survive a deliberative process among free and equal individuals (Habermas, 1996a, pages ; 1996b; Knight and Johnson, 1994). Discourse Habermas has identified discourse as an ideal procedure for rational and democratic decision-making. To fulfil the requirements of discourse, deliberation must be structured in a very special way. The notion of discourse specifies a number of conditions aiming at ensuring that the outcomes of public deliberation are nothing but the result of the forceless force of Habermas has identified discourse as an ideal procedure for rational and democratic decision-making: to fulfil the requirements of discourse, deliberation must be structured in a very special way 284 Impact Assessment and Project Appraisal December 2005

6 the better argument (Habermas, 1975, page 108). At the core of the notion of discourse is the view that the democratic legitimacy of an outcome is dependent on the soundness of the reasons provided for its support (Habermas, 1990, pages ; 1996a, pages , ; 1996b). In the essay Discourse ethics: notes on a program of philosophical justification, Habermas outlines the structure of discourse (1990, pages ). More precisely, he identifies three sets of rules applying to three levels of rational argumentation (Habermas, 1990, pages 86 89). The first set is based on the premise that argumentation is designed to produce intrinsically cogent arguments with which we can redeem and repudiate claims to validity (italics in original) and stipulates that participants in discourse must make use of the same logical semantic rules, for instance, they may not contradict themselves and they must use expressions in a consistent way over time as individuals and across individuals (Habermas, 1990, page 87). The second set is based on the principle that arguments are processes of reaching understanding that are ordered in such a way that proponents and opponents can test the validity claims that have become problematic. It states that participants must follow certain procedural rules, for instance, they must state and defend only what they believe, and they must provide reasons to justify their opinions (Habermas, 1990, page 87). The third set is based on the idea that argumentative speech is a process of communication that, in the light of its goal of reaching a rationally motivated agreement, must satisfy improbable conditions. The set of process rules insulates the communicative process from coercion and inequality and specifies that nobody with the competence to speak and act should be excluded from discourse, that everyone is allowed to question or introduce any assertion and to express his/her needs, beliefs and wants, and that nobody should be prevented by external or internal coercion from exercising these rights (Habermas, 1990, pages 88 89). From the notion of discourse, four principles can be derived against which institutional arrangements and practice can be assessed with regard to deliberative potential (compare with Chambers, 1996, pages ; Kettner, 1993): 3 generality: a principle derived from the first rule of the third set, stipulating that discourses shall be open to all competent speakers whose interests are, or will be, affected by a matter of common concern or the norms adopted to regulate a matter. The principle stipulates that all actors affected, or at least their interests, shall be included in the deliberative process. autonomy: a principle derived from the second rule of the third set, specifying that participants in discourse shall be granted the right to take sides with or against raised validity claims. They shall be granted the right to effective participation, that is, equal opportunities to express and challenge arguments and counterarguments in the deliberative process. power neutrality: a principle derived from the third rule of the third set, stating that in discourse only the forceless force of the better argument (or communicative power) shall be allowed to sway participants. 4 To produce legitimate and rational outcomes, asymmetries of the three kinds of power with distorting effects on deliberation, which can be derived from Habermas s model of modern society as lifeworld and system, must be neutralised. Administrative power finds expression in formal organisation in general and the political system in particular, economic power follows the logic of market exchange and is represented by financial resources, and cultural power finds expression in values and norms generated in the lifeworld. ideal role-taking: a principle derived from the first set of logical semantic rules and the second set of procedural rules, stipulating that participants in discourse shall adopt attitudes of reciprocity and impartiality. If participants do not do this, the deliberative process, no matter how structurally equal, will not be productive. Reciprocity implies that participants talk and listen sincerely and that they do not act strategically. Impartiality means that participants engage in sincere attempts to view matters of common concern from the perspectives of others and, against the background of this multitude of views, try to find an independent stance. Assessing the deliberative potential of EA In the following, the deliberative potential of EA is assessed through an analysis of barriers to an implementation of the four principles derived from Habermas s notion of discourse as an ideal deliberative procedure. In Table 1, the principles are defined and operationalised. Generality The principle of generality stipulates that all those affected, or at least their interests, shall be included in the process. The generality potential of EA is estimated in terms of access to, and scope of, the EA process. Are all legitimate stakeholders included? Is there a systematic procedure for identifying the public concerned? Are the interests of the public concerned reflected in the definition of the environmental issue and the description of its adverse impacts? Who has access to the EA process? EA regulations normally provide guidance by identifying a set of stakeholders who must be consulted (Petts, 1999, page 161). UNECE s (United Nations Economic Impact Assessment and Project Appraisal December

7 Table 1. A Habermasian ideal type Principles and conceptual definitions Generality All those affected, or at least their interests, shall be included Autonomy Everyone included shall be granted the right of effective participation Power neutrality Distortions related to administrative, economic and cultural power must be neutralised to ensure that only the forceless force of the better argument affects the outcome Ideal role-taking Participants must adopt attitudes of reciprocity and impartiality Operational definitions Are the relevant stakeholders included, is there a procedure for identifying the public concerned, and are the interests of the public reflected in the definition of the environmental issue and the description of its adverse impacts? Are the participants (developers, authorities, citizens, etc) provided equal opportunities to put forward and challenge arguments and counterarguments in the various stages of the EA process, ie in screening, scoping, prediction and evaluation, drafting of environmental statement, review, decision and monitoring? How do EA legislation, participants financial resources as well as institutional and expert culture and citizens level of knowledge affect the EA process with regard to generality and autonomy? Do the participants listen and talk sincerely, are they striving to find a collective solution, and do they try to find independent stands based on their own and others views? Commission for Europe) Convention on Access to Information, Public Participation and Access to Justice in Environmental Matters (UNECE, 1998) and Convention on Environmental Impact Assessment in a Transboundary Context (UNECE, 1991) are probably the two most recognised benchmarks of public participation in environmental decisionmaking. In article 2, paragraph 4 of the former, it is stated that the public means one or more natural or legal persons, and, in accordance with national legislation or practice, their associations, organisations or groups, and in paragraph 5 it is stipulated that the public concerned means the public affected or likely to be affected by, or having an interest in, the environmental decision-making. In article 2, paragraph 6, article 3, paragraph 8 and article 4, paragraph 2 of the latter convention, it is stated that states shall notify and consult the public in the areas likely to be affected by activities that will probably have a significant adverse environmental impact across boundaries. Naturally, the regulations provide inclusive definitions of the public. The problem is, however, that the public the persons affected and their associations and organisations is specific to time, site and issue (Glicken, 2000, page 307), while the regulations are static. This highlights the importance of how EA regulations are interpreted and implemented. A common critique is that public participation (in practice) is too limited and takes place too late in the EA process (Adomokai and Sheate, 2004; Almer and Koontz, 2004; del Furia and Wallace-Jones, 2000; Diduck and Sinclair, 2002; Palerm, 1999; Palerm and Aceves 2004; Petts, 2003). However, it should be noted that it is rarely possible to include all relevant stakeholders, and in some situations it is impossible, for instance, in matters that affect future generations. A more realistic goal to strive for might therefore be a participation map that is representative of the interests affected (Petts, 2001), but achieving this is also difficult. One complicating factor is, for example, that the organised interests of civil society are not necessarily representative of the general public (Petts, 1999, page 151). It should also be noted that different selection processes lead to different definitions of the public. A selection process in which it is up to the stakeholders themselves to announce their interest in participating will result in a different set of stakeholders from a process in which certain individuals, associations and organisations are invited to participate (Glicken, 2000). That the two processes result in different definitions of the public concerned does not question their relevance, but it highlights the problem of inclusion. The definition of the environmental issue and the impacts is critical because it has implications for whose interests are taken into account in an EA. Issues and impacts are often formulated in ways that give limited attention to socio-economic aspects (Palerm, 1999). Further, experts tend to stress physical aspects of environmental problems, while citizens tend to stress social aspects (Alton and Underwood, 2003). Defining the scope of an EA is further complicated by information gaps caused by insufficient collection by the assessor, inaccuracy in measurement and sampling, or by the fact that scientific models are based on simplifying assumptions (Wilkins, 2003). There is a generality dilemma. At the same time as there are barriers to generality, it should be noted that the effectiveness of EA is dependent on a successful delimitation of its scope. Too narrow a scope will exclude relevant stakeholders and important environmental issues and impacts, while too broad a scope may make an assessment either superficial or difficult to manage (Modak and Biswas, 1999, page 42 cited in Wilkins, 2003, page 405). Nonetheless, it is important that those affected are provided real opportunities to access key information (Bond et al, 2004). Autonomy The autonomy principle states that it is not enough that those affected are included; they shall also be granted the right of effective participation. This raises the operational question whether the participants are provided the opportunity to put forward and challenge arguments and counterarguments in the various stages 286 Impact Assessment and Project Appraisal December 2005

8 of the EA process screening, scoping, prediction and evaluation, drafting of environmental statement, review, decision and monitoring. There is an institutional flexibility in EA with regard to public participation. EA regulations normally set minimum requirements for public participation, but do not specify the specific institutional arrangements or tools. There is a variety of public participation tools that can be applied in EA (Petts, 1999, pages ; Sinclair and Diduck, 1995), and various tools have different autonomy potential. The differences in potential can be illustrated through the ladder of participation developed by Sherry R Arnstein (1969). This provides a typology including eight levels of participation, and a higher level on the ladder corresponds to a greater autonomy potential. Levels one and two (manipulation and therapy) are denoted as non-participation, since they assume passive citizens who are given information; levels three to five (informing, consultation and placation) are referred to as symbolic participation since individual citizens are provided the opportunity to hear of a matter or submit comments but not given influence over decision-making; steps six to eight (partnership, delegated power and citizen control) are referred to as real participation since citizens are given the opportunity also to discuss and debate matters or even have decision-making power. The public participation tools that are normally used in EA, such as public hearings and information meetings and comment periods, have a limited autonomy potential. These tools are primarily designed for information provision from the developer to the public (Almer and Koontz, 2004; Petts, 1999, pages ; 2003). However, there is increasing experimentation with tools that have greater autonomy potential, such as citizens juries and citizen advisory committees (Petts, 2001; Woltjer, 2002). In common for these novel tools is that they aim to create arenas for deliberation between experts and citizens. Whether the potential of these deliberative tools is realised in practice is a different question. In addition to requiring too weak forms of public participation, the legislation is criticised for requiring participation in too few stages of the EA process (Adomokai and Sheate, 2004, Almer and Koontz, 2004, del Furia and Wallace-Jones, 2000, Diduck and Sinclair, 2002, Palerm, 1999, Palerm and Aceves 2004, Petts, 2003). Public participation is normally required primarily at the stages of scooping and review, while the participation requirements are typically very modest at screening, prediction and evaluation, drafting of environmental statement, decision and monitoring (Petts, 1999, pages ). Power neutrality Discourse prescribes that nothing but the forceless force of the better argument shall affect the outcome of the deliberative process. In the EA literature, the importance of resources for participation is emphasised (Bond et al, 2004). Power neutrality is measured in terms of how EA as a decision process affects the generation and distribution of administrative, economic and cultural power. Administrative power finds expression in formal regulation of various kinds, such as legislation and organisational hierarchies. The EA legislation has been identified as a major barrier to public participation and deliberation. Further, the circumstance that the developer is responsible for drafting the EA is claimed to create incentives for narrowing down the assessment to reduce costs and limit participation to avoid opposition (Diduck and Sinclair, 2002; Petts, 1999, page 171). Also, the division of labour between the parliamentary and the administrative complex, and between different administrative units within the administrative complex is understood to reduce the influence of public participation in EA on final decisions (Petts, 2003). Economic power is based on financial resources and is expressed in many different ways, for example, access to legal and scientific expertise. Financial restrictions force assessors to prioritise considerations and delimit EAs (Wilkins, 2003). Moreover, different participants tend to have different financial resources at their disposal. The developers, who are responsible for conducting the assessments, normally have greater financial resources at their disposal than the citizens affected (Adomokai and Sheate, 2004, Diduck and Sinclair, 2002). The financial resources of developers are translated into access to expertise in the administrative complex, lobbying in the parliamentary complex, and marketing in the public sphere (Wiklund, 2002, pages ). In response to these inequalities of economic power, the need for participation funding programmes through which citizen groups are provided financial resources for collecting and compiling independent information has been emphasised (Fitzpatrick and Sinclair, 2003). Cultural power is informal and finds expression in values and norms. Institutional and expert culture has been identified as a significant obstacle to participation (Emmelin, 1998a; 1998b; 2000; Petts, 2003). Paternalistic institutional culture is reported to have hindered successful implementation of public participation in EA in developing countries (Adomokai and Sheate, 2004; Boyle, 1998). If developers and authorities are not convinced of the benefits of public participation, opportunities for it to take place tend to be minimised (Bond et al, 2004). EAs may also be delimited as a result of the technical approach to decision-making that experts tend to adopt (Petts, 2003) or the inability of assessors to address certain issues (Wilkins, 2003). Technical language makes it difficult for citizens to understand impact statements (Diduck and Sinclair, 2002) and the complicated structure of statements result in few citizens and decision-makers reading the statements (Alton and Underwood, 2003). Impact Assessment and Project Appraisal December

9 Related to this barrier is that citizens are deterred from participating because of a lack of knowledge of the EA process (Adomokai and Sheate, 2004; Diduck and Sinclair, 2002). Further, administrators tend to give most assistance to developers, while citizens, in contrast, are given very little or no assistance at all (Sinclair and Diduck, 2001). To reduce the exclusionary effects of the expert culture, it has been suggested that the focus of impact statements is shifted from other experts to ordinary citizens and decision-makers (Alton and Underwood, 2003) and that the general public is provided thorough and accurate information and support to understand this information (Almer and Koontz, 2004; Sinclair and Diduck, 2001). Ideal role-taking The principle of ideal role-taking stipulates that participants must adopt attitudes of reciprocity and impartiality. That ideal role-taking focuses on qualities in practice does not challenge the fact that the design of EA systems is of great importance. Systems can facilitate reciprocity and impartiality, and different participation tools have different roletaking potential. An institutional precondition for reciprocity is that tools allow for two-way communication. The need for two-way communication is also stressed in the literature on public participation in EA (Bond et al, 2004). However, as noted above, the public participation tools most frequently used in EA assume passive citizens who are given information by the developer. A precondition for impartiality is that tools allow for public communication. This indicates the need of the EA process to be transparent (Bond et al, 2004). However, the structure of the traditional participation tools tends to further conflict and strategic action rather than dialogue and consensus (Fitzpatrick and Sinclair, 2003). The principle of ideal role-taking also points to the importance of motivational factors. People are reluctant to participate as a result of lack of interest or because they prefer to spend their limited time on other things (Adomokai and Sheate, 2004; Diduck and Sinclair, 2002). Yet even when citizens choose An institutional precondition for reciprocity is that tools allow for twoway communication, but the public participation tools most frequently used in EA assume passive citizens who are given information by the developer to participate, they rarely engage in dialogue; they are primarily concerned with expressing their own private opinions (Diduck and Mitchell, 2003). Also, public participation tools are often used strategically. Developers tend to view EA as an instrument for generating popular acceptance for projects to which they have already committed themselves, rather than a process for gathering information on open issues (Petts, 1999, pages ). Moreover, it has been argued that participants rarely try to adopt independent positions. Even though there are some indications that participants gain insight into the interests of others [there is] little evidence that participants change their initial positions on fundamental issues (Diduck and Mitchell, 2003, page 358). The NIMBY (not in my backyard) syndrome is often considered by developers and politicians to represent an example of a situation in which citizens look to their narrow private interests instead of public interests (Petts, 1999, page 151). Does deliberation lead to better decisions? The impact of deliberation in EA on decisions can be measured either in terms of impact on the background material or on the final decision, and it can be interpreted in terms of rationality or legitimacy. EA has its origins in rational planning. It was designed as a tool intended to contribute to informed planning decisions by building science into political administrative processes (Leknes, 2001; Wilkins; Barker and Wood, 1999). According to this traditional view, the involvement of the public is assumed to delay, and thereby increase the cost of, the EA process and to lead to less rational decisions by obstructing neutral experts in their production of objective knowledge. The problem with the traditional view is that it disregards the fact that it is impossible to insulate the EA process from selfinterest and that knowledge and expertise are expressions of power. In addition, the traditional view does not consider the importance of values in EA (Richardson, 2005; Wilkins, 2003). Other researchers of public participation in EA appear more or less implicitly to adopt a radical democratic view, that is, they seem to assume that there should be a direct link between the opinions expressed by citizens and the content of the impact statement and the final decision. The problem with this view is that it ignores that the EA represents only a part of the background material, and that legislative regulations typically prescribe a balancing of environmental and economic concerns (Glasson, 1999, page 141). Also, the radical view disregards the question as to whether more participation is always better than less. This circumstance is troublesome since the right level and form of participation is most likely context-specific, and applying the wrong level and form may do more harm than good (Lawrence and Deagen, 2001). 288 Impact Assessment and Project Appraisal December 2005

10 Empirical studies show that the product of public participation in EA is normally reflected in the impact statement but only to a limited extent, and in most cases the impact on the final decisions is limited. Public participation rarely reverses decisions but it can contribute to project modification and the application of conditions (for instance, Glasson, 1999, pages ; Grandell, 1996; Leknes, 2001; Cooper and Elliott, 2000). In this connection, it should be noted that in order to reach an assessment of participation techniques it is necessary to have a clear conception of the role and purposes of participation itself (Thornley, 1977, page 3 cited in Campbell and Marshall, 2000, page 324). Different roles and purposes lead to different success criteria of evaluation. For example, if the goal is to legitimise, effectiveness is measured in terms of popular acceptance of decisions, and, if the objective is to rationalise, effectiveness is measured in terms of the knowledge reflected in decisions. The empirical findings of the effects of deliberation are limited. In some studies of deliberation in EA, it is concluded that deliberation contributes to more rational decisions by supporting social, experimental or collaborative learning. Deliberation is claimed to add local or situated knowledge and increase the public s and the developer s knowledge (Diduck and Mitchell, 2003; Fitzpatrick and Sinclair, 2003; Petts, 2001; Saarikoski, 2000; Webler et al, 1995). There seem to be few published empirical studies of the legitimating effects of deliberation, but there are indications of that deliberation in EA can increase the legitimacy of background materials and decisions (Diduck and Mitchell, 2003). Summary and conclusions This article has analysed the potential of EA as an arena for democratic deliberation. Viewed from the perspective of deliberative democratic and deliberative planning theory, the results indicate the importance of continuing the search of arenas for deliberation within established political administrative institutions as a complement to the experiments with novel deliberative tools. First, regarding generality, it is clear that it is difficult and, in many cases, impossible to include all those affected and that the environmental issue tends to be defined in ways that have exclusionary effects. However, there is a generality dilemma. At the same time as there are barriers to generality, the effectiveness of EA is dependent on successful delimitation of its scope. Second, regarding autonomy, it is clear that the participation tools that are normally used in EA, such as public hearings and comment periods, have a limited autonomy potential. They are primarily designed for information provision from the developer to the public. Interestingly, there is an increasing experimentation with more inclusive, dialogue-based tools. Third, regarding power neutrality, it is clear that administrative power in the form of EA legislation generally requires weak forms of public participation and that, in late stages of the EA process, economic power in the form of financial resources tends to be unequally distributed between the developer and those affected by a development, and that cultural power in the form of institutional and expert culture tends to hinder implementation of public participation in EA and to deter citizens from participating. Fourth, regarding ideal role-taking, it is clear that the participation tools that are used rarely allow for public two-way communication, which is an institutional precondition for ideal role-taking. In addition, citizens are often reluctant to participate, and, when they do participate, they rarely engage in dialogue. Although there are many barriers, I would like to claim that EA has a hidden deliberative potential. Much of this potential follows from the institutional flexibility that tends to characterise EA systems (for a detailed analysis of the Swedish EA system, see Wiklund (2004)). Typically, the EA legislation specifies minimal requirements, which means that the parties are free to use more demanding participatory mechanisms, for example, with regard to procedures for identifying those affected and the specific public participation tools to be used for disseminating, collecting and processing information. The increasing experimentation with deliberative tools in EA can be seen as an attempt to release this hidden potential. In this connection, it should also be noted that EA is of special interest as an arena for deliberation, since it is a decision support process that involves encounters between experts and ordinary citizens. To release the hidden potential of EA, the barriers to an operationalisation of the Habermasian principles have to be erased. Further, even though it is clear that the analysis above is limited in the sense that it is based on previous studies of public participation in EA with different methodological and theoretical approaches, it has made clear that the four Habermasian principles are a useful ideal type for investigations of deliberation in EA. The principles provide a lens through which EA institutions and practices can be systematically assessed with regard to deliberative potential and qualities. In addition, they make it possible to assess the claim that deliberation leads to better decisions, by linking an assessment of the deliberative quality of the participatory process with an evaluation of the impact of deliberation on decisions in terms of rationality and legitimacy. Irrespective of agreeing or not with the conclusion that EA has a hidden potential, it is clear that there is a need for additional research on deliberation in EA and its effects. What areas are then in need of additional study to make clear whether deliberation in EA leads to a more rational and more legitimate decision-making? First, there is a need to clarify to what extent public participation in EA fulfils the deliberative procedural ideal, for instance, Habermas s Impact Assessment and Project Appraisal December

11 notion of discourse. Second, it is urgent to explore how different institutional arrangements (participation tools) and popular constellations affect the opportunity to deliberate. Third, it is urgent to study, not only whether the views of participants change during the participatory process, but whether changes of views are a result of deliberation. This requires an analytical framework that links an assessment of the deliberative quality of the process with an evaluation of the effects on the outcome. Finally, it is important to clarify what it is in deliberation that causes the (possible) effects. Notes 1. This conception of politics is also found in most 20th century liberal models of democracy, such as the élitist model developed by Joseph Schumpeter (1976), the economic model developed by Anthony Downs (1957) and the pluralist model developed by Robert A Dahl (1956). 2. The republican tradition holds on to a long-standing tradition in political theory from ancient Greece, from the works of Aristotle, and onwards (Finley, 1973). A more recent expression of the republican tradition is the model of strong democracy developed by Benjamin Barber (1984). 3. For a more elaborated account of the four principles, I refer to my doctoral dissertation, see Wiklund (2002, pages 35 71). For an alternative application of the criteria, see Wiklund (2005). 4. In later works Habermas (for instance, 1996a) uses the term communicative power. However, this notion does not refer to power in the same sense as administrative, economic and cultural power; it refers to the view that the better argument is a force to take into account in organising and regulating collective action. References Adomokai, R, and W R Sheate (2004), Community participation in environmental decision-making in the Niger Delta, Environmental Impact Assessment Review, 24, pages Almer, H L, and T M Koontz (2004), Public hearings for EIAs in post-communist Bulgaria: do they work?, Environmental Impact Assessment Review, 24, pages Alton, C C, and P B Underwood (2003), Let us make impact assessment more accessible, Environmental Impact Assessment Review, 23, pages André, P, B Enserink, D Connor and P Croal (2004), IAIA public participation best practice principles, Proceedings of the 24th Annual Conference of the International Association for Impact Assessment, IAIA 2004, Vancouver, Canada, available on CD- Rom (IAIA, Fargo ND). Arnstein, S R (1969), A ladder of participation, Journal of the American Institute of Planners, 35(4), pages Barker, A, and C Wood (1999), An evaluation of EIA system performance in eight EU countries, Environmental Impact Assessment Review, 19, pages Barber, B (1984), Strong Democracy: Participatory Politics for a New Age (University of California, Berkeley CA). Barber, F W, and R V Bartlett (2001), Toward environmental democracy: rationality, reason and deliberation, Kansas Journal of Law and Public Policy, 11, pages Bohman, J (1990), Communication, ideology and democracy, American Political Science Review, 84, pages Bohman, J, and W Rehg (editors) (1997), Deliberative Democracy: Essays on Reason and Politics (MIT Press, Cambridge MA). Bond, A, J Palerm and P Haigh (2004), Public participation in EIA of nuclear power plant decommissioning projects: a case study analysis, Environmental Impact Assessment Review, 24, pages Boyle, J (1998), Cultural influences on implementing environmental impact assessment: insights from Thailand, Indonesia and Malaysia, Environmental Impact Assessment Review, 18, pages Campbell, H, and R Marshall (2000), Public involvement and planning: looking beyond the one to the many, International Planning Studies, 5(3), pages Carleheden, M (1996), Det andra moderna: Om Jürgen Habermas och den samhällsvetenskapliga diskursen om det moderna (Daidalos, Göteborg). Chambers, S, Reasonable Democracy: Jürgen Habermas and the Politics of Discourse (Cornell University Press, Ithaca and London). Cooper, L M, and J A Elliott (2000), Public participation and social acceptability in the Philippine EIA process, Journal of Environmental Assessment Policy and Management, 2(3), pages Crosby, N (1995), Citizen juries: one solution for difficult environmental questions, in O Renn, T Webler and P Wiedemann, Fairness and Competence in Citizen Participation: Evaluating Models for Environmental Discourse (Kluwer Academic Publishers, Dordrecht). Dahl, R A (1956), A Preface to Democratic Theory (University of Chicago Press, Chicago). Daniel, P C, and O Renn (1995), Planning cells: a gate to fractal mediation, in O Renn, T Webler and P Wiedemann, Fairness and Competence in Citizen Participation: Evaluating Models for Environmental Discourse (Kluwer Academic Publishers, Dordrecht). del Furia, L, and J Wallace-Jones (2000). The effectiveness of provision and quality of practices concerning public participation in Italy, Environmental Impact Assessment Review, 20, pages Diduck, A, and B Mitchell (2003), Learning, public involvement and environmental assessment, Journal of Environmental Assessment Policy and Management, 5(3), pages Diduck, A, and A J Sinclair (2002), Public involvement in environmental assessment: the case of the nonparticipant, Environmental Management, 29(4), pages Downs, A (1957), An Economic Theory of Democracy (Harper, New York). Dryzek, J (1990), Discursive Democracy: Politics, Policy and Political Science (Cambridge University Press, Cambridge). Dryzek, J (2000), Deliberative Democracy and Beyond: Liberals, Critics, Contestations (Oxford University Press, Oxford). Eames, M, A Stirling, J Burgess, G Davies, S Williamson, S Mayer and K Stanley (2004), Deliberative Mapping: Integrating Citizens and Specialists Appraisals in a Transparent and Inclusive Participatory Process (Policy Studies Institute, London). Eliaeson, S (1982), Bilden av Max Weber. En studie av samhällsvetenskapens sekularisering (Norstedts, Stockholm). Elling, B (2004), Modernity and communicative reflection in environmental assessment, in T Hilding-Rydevik and A Hlökk Theodórsdóttir (editors), Planning for sustainable development the practice and potential of environmental assessment, Proceedings from the 5th Nordic Environmental Assessment Conference, Reykjavik, Iceland, August 2003 (Nordregio, Stockholm). Elster, J (1997), The market and the forum: three varieties of political theory, in J Bohman and W Rehg (editors), Deliberative Democracy. Essays on Reason and Politics (MIT Press, Cambridge MA) pages Elster, J (editor) (1998), Deliberative Democracy (Cambridge University Press, Cambridge). Emmelin, L (1998a), Evaluating environmental impact assessment part I: theoretical and methodological considerations, Scandinavian Housing and Planning Research, 15, pages Emmelin, L (1998b), Evaluating environmental impact assessment part II: professional cultures as an aid in understanding implementation, Scandinavian Housing and Planning Research, 15, pages Emmelin, L (2000), Nordisk miljöförvaltnings professionskultur och några aktuella frågeställningar i miljöpolitiken, Tidsskrift for samfunnsforskning, 41(3), pages Eriksen, O and E Fossum E (editors), Democracy in the European Union. Integration through Deliberation (Routledge, London). Finley, M I (1973), Democracy: Ancient and Modern (Chatto and Windus, London). 290 Impact Assessment and Project Appraisal December 2005