Reflexive Governance Coping with Complexity, Uncertainty

Transcription

1 Reflexive Governance Coping with Complexity, Uncertainty and Ambiguity Paper for the XXI IPSA World Congress of Political Science Session New Approaches in Political Research II Please do not quote without permission of the author Dr. Andreas Klinke Eawag, Swiss Federal Institute of Aquatic Science and Technology, Domain of the ETH, Zurich/Duebendorf 1

2 1. Introduction Policymaking and governance might become a driving force in the transformation of modern statehood, if it is capable to address the particular challenges associated with the handling of conflict potentials resulting from complex and uncertain problems under the conditions of bounded knowledge, the emergence of new governance forms and the increasing role and inclusion of non-state actors. Attributes characterizing intrinsic conflict potentials are complexity, scientific uncertainty and socio-political ambiguity, which might be triggered and amplified by functional and structural interdependencies between natural systems and human societies including technological, economic, and socio-political realms. Issues featuring high levels of these attributes can exceed physical, chemical and biological boundaries, disciplinary bonds, territorially differentiated political units, and political spheres of influence. Corresponding effects are often pervasive and impacting areas beyond the primary issue area. Governance tackling the conflict potential might generate conflicts of objective with other public policies such as securing economic prosperity or the supply of infrastructures for economic activities. Therefore governance especially relies on integrative efforts so that policies are not thwarted by opposite measures in other areas. Since complex and uncertain issue do not stop at territorial borders, there is an apparent need for supranational and international coordination and cooperation. In particular, the multilevel system of the EU offers platforms and mechanisms to integrate policies and harmonize respective national policies. The traditional problem solving abilities and mechanisms of governments and the prevalence of governmental actors and institutions have to be questioned with regard to the handling of the attributes, because there are at least some plausible rationales to assume that a functional inclusion of non-state resources and capacities can contribute to dissolve the associated conflict potentials. The article will suggest a reflexive governance approach, which conceptualizes a functional division and coordination of labour, capacity and authority between public and private actors in order to cope with complexity, uncertainty and ambiguity. Policymaking by government has not become obsolete, but it needs to face functional alterations in order to cope adequately with the challenges. Reflexive governance implies a tendency towards the transformation of statehood in at least two respects: At first, the scope of state contribution to governance is focused on those functions and capabilities which can only or most efficiently be performed by the state. Second, reflexive governance does not arbitrarily integrate public and private capacities and resources, but functionally adequate with respect to the different requirements emerging with different levels and constellations of complexity, uncertainty and/or ambiguity. And third, the inclusion and participation of non-state actors contributes to a shift from state-centric, interventionist governing to non-hierarchical and participatory forms of governance and its corresponding institutional change. 2

3 On a conceptual level, the following rationales of reflexive governance affirmatively argue that the functional inclusion and participation of experts, stakeholders and the public might enhance the problem solving capability handling complexity, uncertainty and ambiguity. - Reflexive governance addresses the tension between input, i.e. the participation of stakeholders and the public, and output, i.e. the problem solving capability with respect to uncertainty, complexity and ambiguity. Input-oriented quality should not be achieved to the disadvantage of output-quality or vice versa. - Reflexive governance does not rely on a territorially defined polity of a state, it rather refers to a functionally defined publicity in a transnational space that can exceed national borders. - Reflexive governance disengages from the perspective of state-centrism and pays particular attention to the integration of non-state actors. Public-private networks will be established, which are significantly influenced by non-state actors. - The conditions of governance across national boundaries do not necessarily feature insurmountable structural restraints for the implementation of reflexive governance. There are institutional frameworks in which reflexive governance can evolve. The aim of this paper is to discuss the prospects of reflexive governance and its implications for the legitimacy and effectiveness of governance. Why does it make sense to assume that the inclusion and participation of non-state actors in reflexive governance increases the legitimacy and the expectations on sustainable and effective problem solving capabilities? How can reflexive governance cope with complexity, uncertainty and ambiguity? The article is organised as follows. First, I outline the process of transformation of public authority from state-centric policymaking to horizontal modes of governance including non-state actors. Subsequently, I address those phenomenological characteristics of public policy subjects, namely complexity, uncertainty and ambiguity, which trigger conflict potential in public policymaking. In the next section, I relate the identified conflict potentials to different discourse approaches and explain how the specific rationales of the different discourses foster the resolution of the conflict potentials. In the following section, I develop a normative-analytical model of public deliberation and argue that this approach is most adequate to incorporate the different discourses and to realise a legitimate and effective way of public policymaking. Afterwards, I examine the Great Lakes-regime by means of the categories and criteria of the model of public deliberation. I reveal a multilevel participation and network system of deliberative governance from which lessons can be learnt. The article concludes on prerequisites and means how to design a deliberative governance approach of consecutive public participation forms for public policymaking. 3

4 2. Contemporary trends in governance The first trend in governance concerns the move from state-centric approaches with hierarchically organized governments as the only locus of power to systems in which political authority is dispersed to separately constituted bodies with overlapping jurisdictions that do not stand in hierarchical relationship to each other (Skelcher 2005:89; see also Hajer 2003; Hooghe & Marks 2003). Sometimes referred to as governance without government (Rhodes 1996), the hollowing out of the state (Rhodes 1994), or polycentric governance (McGinnis 1999), this notion captures an increasingly complex and diversified political landscape in which a multitude of actors draw on various forms of material and symbolic power in order to influence decision-making processes and outcomes. Advocates of polycentric governance argue that institutional diversity offers considerable advantages when complex and uncertain problems need to be addressed because, first, issues with different geographical scopes can be managed at different scales, second, an inherent degree of overlap and redundancy makes polycentric governance systems more resilient and therefore less vulnerable, and third, the larger number of actors facilitates experimentation and learning. On the other hand, disadvantages include the possible loss of economies of scale; slow and costly collective decision-making; and the potential loss of democratic accountability (Skelcher 2005; Young 2002). Empirical work specifically linking polycentric governance to performance is relatively rare, with examples of the U.S.- Canadian Great Lakes region (Klinke 2006; Sproule-Jones 2002) and Australia (Bellamy et al. 2002). A second trend associated with the rise of governance relates to public participation, which can take different forms, including information supply, consultation, discussions with the public, co-decision making, or the transfer of management authority to non-governmental bodies such as users associations (Keohane & Ostrom 1995; Mostert 2003; Reed 2008; Wondolleck & Yaffee 2000). Although public participation is often promoted as a normative goal, the literature also advances several empirical claims. Public participation is thus said to improve the quality of decision making by making better use of information, creativity, and resources available in society; improve public understanding of the management issues at stake; enhance transparency; stimulate coordination among different government bodies; make management less controversial; and strengthen democracy (Coenen et al. 1998; Ridder et al. 2005). 4

5 A fair amount of empirical work has examined the real and potential advantages of public participation in policy making. Soncini-Sessa et al (2007) looked at the nature and feasibility of a participatory and integrated planning (PIP) model for integrated and participatory water resources management in the transboundary Verbano (Lake Maggiore) system; Junker and Buchecker (2007) analyzed two participatory approaches along the River Thur and the River Flaz, making a case for the careful use of stakeholder identification tools; Mostert et al. (2007) examined participatory processes in 10 European river basins and found improvements in public awareness, stakeholder relations, and, in some cases, policy outputs; and Sabatier et al (2005) synthesize insights from a systematic study of collaborative watershed management processes in the U.S. These empirical results support many of the claims made by advocates of public participation, but they also point to inherent challenges, including the frequent lack of clarity about stakeholder roles; project managers inadequate experience with participatory processes; insufficient time and resources; and the risk of co-optation (Leach & Pelkey 2002; Videira et al. 2006). 3. Conflict Potentials through Complexity, Uncertainty and Ambiguity The continuous increase of interrelations between actions and impacts in the realm of policymaking entails conflict potentials caused by a high degree of seriousness, complexity, scientific uncertainty and/or socio-political ambiguity. These characteristics are sometimes immediate and obvious, but more often subtle, latent and lagged. When these characteristics manifest a conflict potential the effects that result can cause a sequence of secondary and tertiary impacts with regard to time and space, i.e. the functional and territorial dimensions of political, social and economic spheres. The cross cutting consequences of these conflict potentials overextend the scope of purely state-driven policymaking. To rationalize divergent perceptions and interpretations, to harmonise controversial problem understandings, and to find rationally adequate and sustainable solutions in policymaking the maximum possible degree of cognitive reliability is essential. Therefore, first of all, a distinction and further explanation of the terms complexity, scientific uncertainty and socio-political ambiguity is needed (Klinke and Renn 2002; Renn 1999a). The three attributes are interlocked to a certain extent and feature some interactions. The following explication thus contains diffuse grey areas of fuzziness, which cannot be dissolved theoretically, analytically or empirically. 3.1 Complexity Complexity refers to the difficulty of identifying and measuring links between a multitude of causes and sources, for example pollution emitters, and specific adverse impacts or con- 5

6 sequences caused by the interplay of different factors such as the accumulation of persistent chemical substances. In many cases of environmental damages, for example, the relationship between polluters and aggrieved parties are characterized by significantly spatial and temporal distances. The nature of this difficulty may be traced back to interactive effects such as synergism and antagonisms, long delay periods between cause and effect, variations regarding different actors and groups, intervening variables, and others. It is precisely these complexities that make sophisticated scientific investigations necessary since the cause-effect relationship is neither obvious nor directly observable. Nonlinear response functions may also result from feedback loops that constitute a complex web of intervening variables. Complexity requires therefore sensitivity to non-linear transitions as well as to scale on different levels. It also needs to take into account a multitude of pathways and the composite effects of other factors and mechanisms that are present in specific situations. Under these conditions, models of scientific assessment (including e.g. hazard identification, exposure characterisation, risk characterisation) are the most appropriate instruments in policymaking to gain a better (but never) complete picture of the relative problems and risks associated with these complex causal chains. For example, in the last two decades scientific investigations and modelling have been increasingly demonstrated that the global emissions of CO 2 (in particular through traffic, heating, industrial production, and production of energy) and methane (through agricultural production) cause a human induced climate change. Some indications are an increasing temperature worldwide, the reduction of the Artic ice thickness (Johannessen et al. 1999), the decline of the Alpine glaciers (Lozan et al. 1998), and an augmentation of extreme weather events. In the case of highly non-linear or even chaotic relationships (such as climate change), which cannot be adequately addressed by probabilistic techniques, complexity turns into indeterminacy, which is a component of uncertainty. In this case, the limits of analytical modelling are reached. The same is true, of course, for the component of ignorance, which also describes an element of uncertainty. Furthermore, complexity refers to the interdependencies between society and environment. The interrelations are multilayer and utterly difficult and the knowledge about it is markedly unreliable. Bias of perceptions, social amplifications, cultural contexts, organizational influencing factors, and time lags regarding the recognition of ecological damages are challenges which need to be adequately addressed. 3.2 Scientific Uncertainty Scientific Uncertainty often results from the missing ability to completely or adequately reduce complexity in modeling cause-effect relations and refers to a lack of intelligibleness or quality of scientific or technical data. It is essential to acknowledge in the context of scientific assessment that human knowledge is always incomplete and selective and thus contingent on uncertain assumptions, assertions and predictions (Bruijn/ten Heuvelhof 1999; Functowicz/Ravetz 1992; Laudan 1996). It is obvious that the modeled probability 6

7 distributions within a numerical relational system can only represent an approximation of the empirical relational system with which to understand and predict uncertain events (Cooke 1991). It therefore seems prudent to include other, additional, aspects of uncertainty (Morgan/Henrion 1990; van Asselt 2000; van den Sluijs et al. 2003). Since there is no prevailing school of thought in the literature how to best itemize uncertainties, the following categories seem to be at least an appropriate approach of distinguishing the key components of uncertainty (Klinke/Renn 2002, 2009 forthcoming): Variability refers to different vulnerability of targets such as the divergence of individual responses to identical stimuli among individual targets within a relevant population such as humans, animals, plants, landscapes etc; Inference effects relate to systematic and random errors in modeling inducing problems of drawing extrapolations or logic deductions from small statistical samples, from animal data or experimental data onto humans or from large doses to small doses etc. All of these are usually expressed through statistical confidence intervals; Indeterminacy results from genuine stochastic relationship between cause and effects, apparently non-causal or non-cyclical random events, or badly understood non-linear, chaotic relationships; System boundaries allude to uncertainties stemming from restricted models and the need for focusing on a limited amount of variables and parameters; Ignorance means the lack of knowledge about the probability of occurrence of a damaging event and about its possible consequences. For example, in the 1950ies politics and societies were unaware of the impacts of chlorofluorocarbons (CFC) upon the stratospheric ozone, which caused the hole in the ozone layer. The first two categories of uncertainty address the epistemic dimension and therefore can be reduced by improving the existing knowledge and by advancing the present modeling tools. The last three categories are genuine components with an aleatory nature of uncertainty and thus can be characterized to some extent using scientific approaches but cannot be further resolved. If uncertainty, in particular the aleatory components, plays a large role then the estimation becomes fuzzy. The validity of the end results is questionable and, for management purposes, additional information is needed such as a subjective confidence level in the estimates, potential alternative pathways of cause-effect relationships, ranges of reasonable estimates, loss scenarios and others. Examples for high uncertainty, particularly aleatory uncertainty, include many natural disasters, possible health effects of mass pollutants below the threshold of statistical significance, and long-term effects of introducing genetically modified species into the natural environment. 3.3 Socio-political ambiguity 7

8 Interpretative and normative ambiguity relates to divergent or contested perspectives on the justification, severity or wider meanings associated with a given threat (Stirling 2003). Interpretative ambiguity denotes the variability of (legitimate) interpretations based on identical observations or data assessments results, e.g. an adverse or non-adverse effect. Variability of interpretation, however, is not restricted to expert dissent. Laypeople s perception often differs from expert judgments because it is also a response to qualitative characteristics such as familiarity, personal or institutional control, assignment of blame, and others. Moreover, in contemporary pluralist societies diversity of perspectives on the environment within and between social groups is generally fostered by divergent value preferences, variations in interests and very few, if any universally applicable moral principles; all the more, if problems are complex and uncertain. That leads us to the aspect of normative ambiguity. It alludes to different concepts of what can be regarded as tolerable referring e.g. to ethics, quality of life parameters, distribution of costs / risks and benefits, etc. A condition of ambiguity emerges where the problem lies in agreeing on the appropriate values, priorities, assumptions, or boundaries to be applied to the definition of possible outcomes. What does it mean, for example, if neuronal activities in the human brain are intensified when subjects are exposed to electromagnetic radiation? Can this be interpreted as an adverse effect or is it just a bodily response without any health implication? Many of the scientific disputes in the fields of analysis and management do not refer to differences in methodology, measurements or dose-response functions, but to the question of what all this means for human health, protection, security and management requirements (cf. Harrison/Hoberg 1994). High complexity and uncertainty favor the emergence of ambiguity, but there are also quite a few simple and highly probable problems that can cause controversy and thus ambiguity. Examples for high interpretative ambiguity include low dose radiation (ionizing and non-ionizing), low concentrations of genotoxic substances, food supplements and hormone treatment of cattle. Normative ambiguities can be associated, for example, with nuclear power and genetically modified food. Ambiguity may lead to social mobilisation and conflict. This may be based primarily on conflicting moral perspectives or in the first place on conflicting views of what constitutes a fair distribution of risks and benefits. It yet becomes more problematic when an issue does not only polarize a national public discourse, but triggers conflicts on the supra- and international level among states. For example, the BSE crises in the European Union and disputes about growth hormones in the WTO included the conflict potential of ambiguity. Again, high complexity and uncertainty favour the emergence of ambiguity, but there are also quite a few simple and almost certain issues that can cause controversy and thus ambiguity. It is important to note, that the challenges concerning complexity, lack of scientific certainty and socio-political ambiguity can hardly be considered as clearly separated. Instead, 8

9 the answers to resolve the conflict potentials in public policymaking must be conceived of as being interrelated and supplementary. Accordingly, the following proposed sequential approach of discourses is an attempt to cope with these interrelating challenges through reflexive governance. 4. A Model of Reflexive Governance Within representative democracies the societal and political system is coupled through general elections and the parliament. The relation between the society and the politicaladministrative system is ideally interpreted as principal-agent relationship, in which the members of the society commission the policy makers to address their concerns and interests and they again commission the administration to perform the tasks. This normative ideal has to be left behind, when governance has to cope with the complexity, uncertainty and/or ambiguity of problems and to resolve its corresponding conflict potentials. Reflexive governance has moved further away from traditional top-down policy making and its established processes because they are not capable to early absorb the conflict potentials and systematically address the attributes of complexity, uncertainty and ambiguity. This is particularly apparent, if quite different interests are affected, which cannot be reduced to a common denominator, and different value systems are at stake. Accordingly, legitimacy is lacking, since an acceptable political consensus or at least the absence of dissent cannot be achieved. The success of reflexive governance depends on creating procedures and institutional arrangements that foster effective processing based on division of labour. The proposed governance approaches primarily focus on the communication and the functional canalization of discourses as means of politics to systematically address complexity, uncertainty and ambiguity. Thereby it is take into account that the involvement of all affected is unrealistic, since the approximation of the ideal that all should be entitled to participate in a collective binding decision becomes implausible as soon as the appropriate definition of what we mean by all is called into question (Offe/Preuss 1991: 166). A pragmatic political implication might be that feasible governance approaches need not depend on making every decision the product of deliberation and participation of every affected citizen (cp. Bohman 1998). 4.1 State-centred regulation Handling issues, which are not associated with a significantly high degree of complexity, scientific uncertainty and/or socio-political ambiguity, requires hardly any alterations to traditional regulatory politics and respective administrative structures. Since the data is provided by statistical analysis and law or statutory requirements determine the general and specific political objectives, a top down oriented regulatory governance approach is ade- 9

10 quate to ensure that all necessary remedial actions and measures are implemented and enforced. The problem solving capability emanates from legal restrictions executed by powerful governmental authorities with interventionist capacities. Such a state-centric regulation is based on a hierarchical command-and-control model, which provides mechanisms to reduce transaction costs (cp. Williamson 1975) and enhances accountability. Enforcement authorities and administrative agencies impose implementing specifications of conduct through rulemaking by means of proscriptions, instructions and threats of sanction. They are also responsible for monitoring the issue area and assessing compliance in order to provide a reinsurance that no unexpected consequences may occur. An internal throughout line of steering and control commits the administration to subordinated governmental authorities even in federal systems, whereas courts, boards of audit or accounting offices and the public ensure external control. In many cases the supranational and national level is responsible for the legislation and the sub-national level is accountable for the legal execution and effective implementation. Inquiries and valuations about performances should be free from hierarchical control. The vertical integration of regulatory governance is based on legal requirements, in which the subordinated political levels possess relatively little autonomy. Accordingly, an instrumental discourse among departmental and agency staff and enforcement personnel within governmental institutions might be sufficient to handle relatively simple problems. The aim is to find the most cost-effective method for a desired regulation level. For this purpose traditional cost-benefit analyses including effectiveness and efficiency criteria are the political devices for finding the right balance between under- and overregulation. If necessary, stakeholders may be included in the deliberations as they have information and know-how that may provide useful hints for being more efficient. For example, the protection of drinking water by the EU water framework directive is a traditional regulatory governance approach of EU environmental policy making. However, regulatory politics is set boundaries, if effective and common good oriented problem solving capability depends on new, horizontal and dialogue-oriented governance modes including increased proactive research, stakeholder involvement and the engagement of citizens. For example, integrated water resource management might require the horizontal integration of public and private actors as outlined below because it induces complexity, uncertainty and even ambiguity. Even if these attributes are not highly significant, there are constraints for the hierarchical functionality of regulatory governance systems. Regulatory governance may not be confined to blind adherence of instructions and mechanical implementation, but include the situation-specific use of administrative freedom (cp. Döhler 2007: 50). The abandonment of hierarchy enables cooperative structures of labour and capacities for teamwork as well as an increase in efficiency. 10

11 4.2 Expert deliberation Resolving the conflict potential caused by complexity requires an epistemological discourse and deliberation among knowledge-based experts of the epistemic community aiming at the clarification and definition of problems for political decision making and how politics should tackle them. Epistemic communities are scientifically legitimized networks of professionals with recognised expertise and competence in a particular domain and an authoritative claim to policy-relevant knowledge within that domain or issue-area (Haas 1992: 3). Epistemic communities dealing with complexity are collaborative networks, in which the actors establish close relationships with each other, since no actor has the ability to meet the epistemic challenge of complexity without the cooperation with the others. The collaboration is based on mutual dependences of knowledge resources. Sharing and exchanging information and knowledge plays a major role in the interaction. The discussion and collaboration aim for the exertion of influence on political governance processes in order to participate how complexity in politics will be handled. However, they do not have decision making power and the authority to issue directives. But the epistemological discourse can get more political authority and influence, if it is canalized and institutionalized as professional consultancy in form of an expert advisory body, which is recognized as deliberative forum in form and content. The Intergovernmental Panel on Climate Change (IPCC) might be instructive as an example. The successful international collaboration within the IPCC seems to have a function of a role model as similar efforts towards chemical pollution (IPCP, International Panel on Chemical Pollution) 1 and an international expert advisory body on biological diversity 2 illustrate. The participants of an epistemological discourse scientifically ascertain and assess data and information of the phenomena in question and evaluate action consequences. An institutionalized advisory board representing the epistemic community could compile the research results from the different disciplines and frames a coherent delineation which forms the cognitive basis for political decision making. The more complex, the more multi-disciplinary and the more uncertain a phenomenon appears to be, the more necessary is a communicative exchange of arguments among the experts. The overarching goal is to achieve the most adequate description and explanation of the causal phenomena, a homogenous and consistent problem definition and understanding, a collectively accepted assessment of anthropogenic influences, and answering the question, which, political and/or societal impacts are to be expected by specific actions or policies. Furthermore, dissenting views should be clarified 1 2 For the IPCP see the website The German Advisory Board on Global Change has firstly suggested to establish an international advsiory board on biological diversity similar to the IPCC in its annual report on the conservation of the global biosphere (WBGU 2001). Meanwhile an international consultation process has been initiated called International Mechanism on Scientific Expertise on Biodiversity (IMoSEB), which shall develop proposals for an international panel. 11

12 by means of arguing (rather than bargaining) in the process. However, the epistemological discourse may reveal that there is more uncertainty and ambiguity hidden in the case than the initial appraisers had anticipated. If the discourse and deliberation include natural and technological as well as social scientists, future controversial debates may be anticipated. Controversies would occur less as a surprise than now. 4.3 Associational policy making If issues are associated with high uncertainty, scientific input is only the first step of a more complex evaluation procedure. It is still essential to compile the relevant data and the various arguments for the positions of the different science camps. Procedures such as using the Pedigree Scheme by Funtowicz and Ravetz (1990) might be helpful to organise and classify the existing knowledge. In a second step, information about the different types of uncertainties has to be collected and brought into a deliberative arena. Players in this arena are not only scientists, but also directly affected stakeholders and public interest groups (Yosie and Herbst 1998). Associational policy making encourages stakeholderbased politics that underpins the complex evaluation procedure. The objective here is to find the right balance between too little and too much precaution. There is no scientific answer to this question and even economic balancing procedures are of limited value, since the stakes are uncertain. Coping with uncertainty includes the clarification of knowledge (similar to the epistemological) and the assessment of trade-offs between the competing extremes of over- and under-protection. It is about decisions on risk-averse or risk-prone action in environment-related policy making such as energy politics, biotechnology, nanotechnology, climate change politics etc. For this purpose it might be instructive to bring together different actors from the public sector (governments and administration) as well as representatives of societal interests such as experts and stakeholders (civil society and economy), which are directly addressed by policies, and interest groups representing the public interests of those directly affected. Accordingly, associational policy making integrates territorially oriented governmental policy makers and public administrators 3, functionally oriented representatives of corporations and associations and scientifically oriented experts in institutional arrangements in order to search problem solving options in which the following questions are discussed and collectively answered: How much uncertainty is one willing to accept for some future opportunity? Are the risks taken worth while the potential economic and/or technological benefits? How much safety is adequate (classic question: how safe is safe enough )? The challenge for the associational governance approach is to integrate different actor groups with different logics of action. Policy makers 3 Actors of the public sector represent different territorial levels such as the international, European, national, regional and/or local level. 12

13 are bound to their political institutions and constituents and their action is influenced by political decision making structures. Economic actors are steered by market mechanisms and their action is focused to create benefiting basic conditions for production or services. Civil society stakeholders are based on ideational commonalities and the willingness of their members to cooperation. The action of their representatives is guided by the solidarity of the members and majority decisions. And scientists are driven by the rational search for the cognitive truth. They refer to existing facts, factual knowledge and logics, which can scientifically be verified. Except for the scientists, the participating actors in associational governance are representatives of organizations, i.e. they are not absolutely free in their collaborative behaviour, but need the affirmation of their own organizations. Hence, the actors need to produce consensus on different levels, i.e. on the associational governance level and within their organizations. For all practical purposes, associational policymaking is primarily adequate to be functionally organized for specific related issues. It is uncoupled from traditional political institutions, constitutional requirements and classic subsystems of the society and the political system (legislative authorities, bureaucratic agencies, economy, social organizations, associations etc). It is self-organized as network, in which horizontal and dialogue-oriented forms of interaction prevail. The interaction and communication might be formally canalized through special procedures such as negotiated rulemaking, mediation and round tables. The decision making process in these procedures is characterized either by arguing aiming for a deliberative process of consensus seeking with the orientation on common interests of the participants or by bargaining as search for a compromise, which allows all participants to realize their individual interests to a large extent, or by a mixture of both (cp. Elster 1992; Holzinger 2001). If arguing prevails, the actors tend more to mutual understanding and learning processes, in which they even redefine their interests to reach a consensus. If bargaining dominates, they tend towards mutual adaptation of positions to make a compromise. As structuring instruments value-tree analysis and other decision-aiding tools might be used in these procedures (Renn et al. 1993). If associational decision making is coupled with existing regulating systems, which are self chosen or at least negotiated, the exertion of political influence and the reliability of expectations increase, whereas transaction costs decrease. Political or economic advising committees, who propose or evaluate political options, could also be established to advise this core-group. 4.4 Inclusive participation Coping with disputed knowledge and/or socio-politically ambiguous interpretations of impacts necessitates an inclusive and participatory governance approach since causal and principal beliefs as well as differences about values are associated. There are no sovereign authorities in political-administrative systems, which would be able to hierarchically 13

14 achieve legitimate and effective policies resolving socio-political ambiguity. Instead different state and societal actors need to cooperate to search for sustainable solutions that are compatible with the interests and values of the people affected and to resolve conflicts among them. Rather than being centralized, an inclusive-participatory governance approach occurs within horizontally organized networks of discourse and dialogue involving representatives of governments and agencies, experts, civil-society and economic stakeholders, directly affected groups and representatives of the public. To focus the discourse and to establish stable relations of communication among public and private actors, traditional procedures of legal policy making, but also novel procedures might be instructive. On regional and local levels citizen panels or juries (randomly selected), and voluntary public advisory committees might be appropriate participative procedures. On the national level the Danish model of consensus conferences and other participatory techniques in order to resolve ambiguities and norm and value conflicts might be appropriate procedures (cf. Schneider at al. 1998). On the transnational level a three-step procedure consisting of mediations and citizen panels might be adequate for public participation across borders (cf. Renn and Klinke 2001). Since every single form of public participation has its merits as well as disadvantages the combination of different participation forms makes possible to compensate the deficits of single procedures and to aggregate their strengths. Anyway it is decisive that deliberative processes are institutionalized, in which public and private actors commonly search for consensus by means of arguing and argumentative action focussing on mutual learning and understanding. This involves the subjective weighting of evaluation criteria and an interpretation of the results. Issues of fairness and justice, visions on future technological developments and societal change and preferences about desirable lifestyles and community life play a major role in these discussions (Webler 1999). On regional and local levels citizen panels or juries (randomly selected), and voluntary public advisory committees might be appropriate participative procedures. On the national level the Danish model of consensus conferences and other participatory techniques in order to resolve ambiguities and norm and value conflicts might be appropriate procedures (cf. Schneider at al. 1998). On the transnational level a three-step procedure consisting of mediations and citizen panels might be adequate for public participation across borders (cf. Renn and Klinke 2001). The deliberative character of the process prompt the participating actors to impose norms of reciprocity upon their behaviour, to commonly search for the common good and sustainability, to consider future consequences of decisions, and to develop collectively acceptable policies and practices, which can be publicly justified. The arenas and procedures of inclusive participation are often decoupled from the institutions of representative democracy, in particular from the parliamentary arenas. The networks evolve ad hoc in many cases and the arenas are hardly formalized. In environmental politics, for example, participatory procedures have been rather rarely and tentatively used in Europe (Weidner 1998). However, procedures of public participation in environmental politics have been at 14

15 least partially institutionalized in the USA and Canada (Elliot 1998; Sigurdson 1998). The primary goal of this governance approach is the integration of affected people and policy addressees in the political decision making process because it can help to overcome problems of socio-political ambiguity by mobilizing the knowledge of those affected, by considering normative and value-oriented preferences and by fostering the communication and exchange across the expert-lay and principal-agent divide. 4.5 Escalator of discourses for the science-policy-stakeholder-interface The different approaches as set above lead to an escalating and consecutive sequence including different discourse types when on moves from simple to complex, from complex to uncertain and further to ambiguous conflict potentials. The political communication and decision making of reflexive environmental governance should be organised as a system of modules constituted on each other, in order to increase legitimacy and effectiveness. The more complex, uncertain and ambiguous are the issues, the more a collective responsibility how to handle the issues should be shared by public and private actors. But the increase of legitimacy and effectiveness in such a way is not unproblematic, which will be addressed in the final section. Fig. 2: An escalating sequence of four types of discourse Inclusive Participation State-Centred Regulation Routine operation Actors: Policy Makers Expert Deliberation Scientific Assessment Necessary Types of Conflict: cognitive Actors: Policy Makers External Experts Associational Policy Making Balancing Necessary Scientific Assessment Necessary Types of Conflict: cognitive evaluative Actors: Policy Makers External Experts Economic Stakeholders Directly Affected Groups Tradeoff Analysis and Deliberation Necessary Balancing Necessary Scientific Assessment Necessary Types of Conflict: cognitive evaluative normative Actors: Policy Makers External Experts Economic Stakeholders Directly Affected Groups Representatives of Public Discourse: Instrumental Discourse: Epistemological Discourse: Reflective Discourse: Participatory Simple Complex Uncertain Ambiguous 15

16 5. Conclusion: Equilibrium between Input and Output The legitimacy and effectiveness of reflexive governance depends on the balance between input, i.e. the participation of stakeholders and the public, and output, i.e. the problem solving capability with respect to uncertainty, complexity and ambiguity. from the perspective of deliberative and participatory democracy theory, the input-oriented legitimacy of reflexive governance rely on an equal opportunity of access to political influence for civil-societal and economic actors in processes of governance (cf.: Bohman 1999; Knight and Johnson 1997). This criterion has two dimensions: The procedural dimension raises issues and questions of inclusion and exclusion, gatekeeper-function, representativeness of interests, and forms of direct participation for affected or interested people such as in alternative dispute resolution (cf. Holzinger 2001). Which stakeholders or representatives of the public will be selected for participation, and how? Who does play the role of the gatekeeper? Who concedes whom, when and on which political-institutional level access to political influence? When have non-state actors the opportunity to participate directly, when will their interests be represented? Which participatory procedure or combination of different procedures is best adequate? Different forms of stakeholder and public participation have their advantages, but all also have shortcomings (cf. Renn et al. 1995). There is no single procedure which fulfils all normative requirements of public deliberation with regard to inclusion, fair representation, dialogue-oriented exchange of arguments, rational argumentation, and problem solving capability. The combination of different participation forms makes possible to compensate the deficits of single procedures and to aggregate their strengths (Klinke 2006; Renn/Webler 1998). The substantial dimension of the input-oriented legitimacy refers to the effective influence in policymaking, i.e. the influence goes beyond mere voting. Influence means something like the effective use of free public reason (Bohman 1999). Governance is not only based upon the structural character of argumentative-oriented procedures that virtually imposes the participating actors to bring forward publicly justifiable reasons for their interests and concerns, but it also relates to the individual capability, willingness and reliance of the participants to the use of public reason. The use of public reason demands from the participating actors to translate their claims into mutually justifiable language and to distinguish between what they find good and what can reciprocally and generally are argued for (Forst 2001: 362). Moreover, it is the epistemic capability to distinguish good and convincing from bad and unpersuasive reasons (Bohman 1997: 337). To cope with issues associated with high complexity, scientific uncertainty and sociopolitical ambiguity, a maximum of cognitive reliability is needed. While governmental policy makers as well as citizens are overstrained, scientific expertise has the potential to provide the necessary cognitive reliability. The systematic integration of independent scientists and representatives of the epistemic community facilitates the development of a 16

17 common knowledge and information base, which allows rationalising divergent problem perceptions, to harmonise contradictory problem understandings and to find rationally acceptable problem solving. It is the input of valid, causal knowledge for the formation of a collectively acceptable, cognitive basis, which makes more likely to agree on a common problem definition, a reasonable problem evaluation and a rationally adequate and common good oriented public policy. The scientific expert knowledge becomes a public resource that may be democratically shared and publicly verified through reflection (Bohman 1999). Structurally and functionally, policymaking is based on a cognitive division of labour between lay people and experts. From an institutional perspective the involvement of independent scientific expertise should be organised preliminary, parallel and downstream to the policymaking process. The output-effectiveness relates to the question how to achieve a public policy that is acknowledged by the affected people as effective. Effectiveness, which emanates from the argumentative procedures, means the discursive capability making good and right decisions. Since this capability is more than the lowest common denominator of preferences, there are two aspects, which the affected people and the addressees of the policies can approve as epistemic quality of the problem solving. First, the developed policy should be the substantially adequate reaction to improve the situation and to cope with the conflict potentials. Second, the selected strategies, regulative actions and measures within the policy have to take into account the common good of the affected people in the issue area and must not promote particular interests. Therewith an policy will be accepted as common good oriented, it needs a reflexive formation of a public conception of the common good (cf. Cohen 1989). Here, the question of inclusion or exclusion once again arises, whose public good ought to be taken into account and what, substantively spoken, is the public good (Dahl 1989: 306). The reference system of the public good gains validity, accountability and credibility, if the relevant framework of action will be politically, socially and/or territorially specified. Since the substantial dimension of the common good can cover prosperity, full employment, education, health, internal and international peace, social, military and civil security, sustainable use of natural resources, and sustainable growth, it is neither a hierarchy nor a charging for these units possible (Offe 2001: 482). But it is possible to resort to the category of justice, even if it is a relational term and has to be defined as, for example, compensatory, distributional, intergenerational, or procedural justice. Accordingly, output-effectiveness does not rely on an independent standard that is objectively measurable; it rather embodies the result of cognitive reflection and interpretation of the participants in argumentative procedures. The implementation of reflexive governance is faced with the challenge to ensure high quality of input-legitimacy and output-effectiveness. But there is a tension between the input-dimension, i.e. the direct participation of actors, and the output-dimension, i.e. the 17

18 result or product of policymaking. The more actors participate, the more the epistemic quality is at stake, i.e. the capability to achieve good and right public policies. The quantitative and qualitative enhancement of inclusion raises questions such as: Who is an affected actor? Who is a relevant stakeholder? Which interests should be represented? Who does play the role of the gatekeeper? In most cases governmental actors determine whom they concede access to political influence. The empirical research on procedures of alternative dispute resolution indicates that argumentative-oriented procedures are effective when the number of participants is limited and when they take place behind closed doors, which would violate democratic principles such as publicity and transparency. Furthermore, there is a lack of mechanisms and procedures to aggregate and integrate the results of participatory procedures taking place on different political-institutional levels. Fig. 3: Equilibrium between input and output Source: Klinke (2006) Costs or risks Input Output y 1 x 1 Participation and inclusion The simultaneous maximisation of input-oriented legitimacy and output-oriented effectiveness is impossible. From normative and empirical models we know that the increase on the input-side will be achieved to the disadvantage of the output-side. By the same token the enhancement of the output-quality will mostly be achieved at the expense of the inputquality. Therefore, a trade-off between the input and output-dimension is needed. Theoretically, decision-making moves between the decision making through one single actor, which is undemocratic, and the inclusion of all affected people, which is unrealistic. The increasing number of participating actors raises the costs and risks of effective decisionmaking and aggregation. Vice versa: The costs and risks of decision making decline, if the 18