Social Multi-Criteria Evaluation (SMCE)

15/2003 – UHE/UAB – 04.06.2003 “Social Multi-Criteria Evaluation” (SMCE)” Giuseppe Munda Universitat Autonoma de Barcelona Dept. of Economics and Economic History, building B and Institute for Environmental Sciences and Technologies 08193 Bellaterra (Barcelona), Spain e_mail: [email protected] Abstract: The main argument developed here is the proposal of the concept of “Social Multi-Criteria Evaluation” (SMCE) as a possible useful framework for the application of social choice to the difficult policy problems of our Millennium, where, as stated by Funtowicz and Ravetz, “facts are uncertain, values in dispute, stakes high and decisions urgent”. This paper starts from the following main questions: 1. Why “Social” Multi-criteria Evaluation? 2. How such an approach should be developed? The foundations of SMCE are set up by referring to concepts coming from complex system theory and philosophy, such as reflexive complexity, post-normal science and incommensurability. To give some operational guidelines on the application of SMCE basic questions to be answered are: 1. How is it possible to deal with technical incommensurability? 2. How can we deal with the issue of social incommensurability? To answer these questions, by using theoretical considerations and lessons learned from realworld case studies, is the main objective of the present article. Key Words: Multi-Criteria Analysis, Economics, Complexity Theory, Environment, Social Choice, Post-Normal Science, Incommensurability, Ethics 15/2003 – UHE/UAB – 04.06.2003 2 1. Methodological Foundations of Social Multi-Criteria Evaluation (SMCE): Complexity, Post-Normal Science and Incommensurability The world is characterised by deep complexity. This obvious observation has important implications on the manner policy problems are represented and decision-making is framed. As a consequence, one may decide to adopt a reductionistic approach trying to tackle one of the many possible dimensions or simply to deal with real-world complexity. This second approach is the one adopted in the present article. My firm conviction is that any representation of a complex system is reflecting only a sub-set of the possible representations of it. A system is then complex when the relevant aspects of a particular problem cannot be captured when using a single perspective (Funtowicz et al., 1999; O'Connor et al., 1996; Rosen, 1977). To make things more difficult, systems including humans are reflexively complex. Reflexive systems present two peculiar aspects: “awareness” and “purpose”, both requiring an additional “jump” in describing complexity. The presence of selfconsciousness and purposes (reflexivity) means that these systems can continuously add new relevant qualities/attributes that should be considered when explaining, describing or forecasting their behaviour (i.e. human systems are learning systems). Moreover, the existence of different levels and scales at which a hierarchical system can be analyzed implies the unavoidable existence of non-equivalent descriptions of it (Giampietro, 1994). Even a simple “objective” description of a geographical orientation is impossible without taking an arbitrary subjective decision on the system scale considered relevant. In fact the same geographical place, e.g., in the USA, may be considered to be in the north, south, east or west according to the scale chosen as a reference point (the whole USA, a single state and so on)1(Giampietro and Mayumi, 2000). Therefore, the problem of multiple-identities in complex systems cannot only be interpreted in terms of epistemological plurality (non-equivalent observers), but also in terms of ontological characteristics of the observed system (non-equivalent observations). The implications for multi-criteria evaluation of the scale issue are very important. For example, in generating evaluation criteria (e.g., in evaluating the impacts building a ski infrastructure in a mountain region, who are the relevant social actors to interact with? The inhabitants of the mountain region, the potential users in urban areas or even the ecological preservationists all around the world might sound reasonable answers) or in computing the impact scores (e.g., a contamination indicator has to be computed locally, or should it be computed at a larger scale? The use of hydrogen cars inside cities is clearly good at a local level, but it is not that clear at a global level, where the emissions depend on the technology by which hydrogen is produced - being hydrogen an energy carrier and not an energy source -) or in choosing the weight factors. A consequence of these deep subjectivities is that in any normative exercise connected to a social decision problem, one has to choose an operational definition of “value” in spite of the fact that social actors with different interests, cultural identities and goals have different definitions of “value” (O’Neill, 1993). That is, to reach a ranking of policy options, there is a previous need for deciding about what is important 1 These multiple-identity/multiple-scale systems can be defined as “Learning Holarchies”. A “holon” is a whole made of smaller parts (e.g. a human being made of organs, tissues, cells, atoms) and at the same time it forms a part of a larger whole (an individual human being is a part of a household, a community, a country, the global economy) (Koestler, 1969). 15/2003 – UHE/UAB – 04.06.2003 3 for different social actors as well as what is relevant for the representation of the realworld entity described in the model. In general, these concerns have not been considered very relevant by scientific research in the past. On the other hand, the new nature of the problems faced in this third millennium (e.g., the mad cow, genetic modified organisms, … ), implies that very often when deciding on problems that may have long term consequences we are confronting issues “where facts are uncertain, values in dispute, stakes high and decisions urgent” (Funtowicz and Ravetz, 1991, 1994). In this case, scientists cannot provide any useful input without interacting with the rest of society and the rest of the society cannot perform any sound decision making without interacting with the scientists. That is, the question on “how to improve the quality of a policy process” must be put, quite quickly, on the agenda of “scientists”, “decision makers” and indeed the whole society. This extension of the “peer community” is essential for maintaining the quality of the process of decision making when dealing with reflexive complex systems. In relation to this objective Funtowicz and Ravetz have developed a new epistemological framework called "Post-Normal Science", where it is possible to better deal with two crucial aspects of science in the policy domain: uncertainty and value conflict. The name "post-normal" indicates a difference from the puzzle-solving exercises of normal science, in the Kuhnian sense (Kuhn, 1962). Post-Normal Science can be characterized in relation to other, complementary, scientific strategies according to the diagram in Figure 1, which is based on two axes: "systems uncertainties" and "decision stakes". When both uncertainty and stakes are small, we are in the realm of "normal" academic science, where it is safe to rely on “codified expertise”. When either uncertainty or stakes are in the medium range, then the application of routine techniques and standardized and generalized knowledge is no longer enough. In these cases, skill, judgement, sometimes even courage are required to adjust the “general knowledge” available to the “special situation”. Funtowicz and Ravetz call this "professional consultancy", with the examples of the surgeon or the senior engineer facing a critical situation. Finally we arrive to cases, in which conclusions are not completely determined by scientific facts; inferences will (naturally and legitimately) be conditioned by the values held by the agents. When the stakes are very high (as when an institution is seriously threatened by a policy) then a defensive tactic will involve challenging every step of a scientific argument (this applies even to those cases in which systems uncertainties are actually small). Such a tactic should be considered wrong only when is conducted covertly, as by scientists who present themselves as impartial judges when, in reality, they are actually committed advocates of one view. When legitimate contrasting views are openly used to challenge scientific arguments, we are in the realm of Post-Normal Science. The previous discussion can be synthesised by using the philosophical concept of weak comparability (Martinez-Alier et al., 1998; O'Neill, 1993). Weak comparability implies incommensurability i.e. there is an irreducible value conflict when deciding what common comparative term should be used to rank alternative actions. Remembering that the presence of multiple-identities in complex systems can be explained in terms of epistemological plurality and in terms of ontological characteristics of the observed system, I argue that it is possible to further distinguish the concepts of social incommensurability and technical incommensurability. Social incommensurability can be derived from the concepts of reflexive complexity and Post Normal Science and refers to the existence of a multiplicity of legitimate values in society, that is, just in one word to democracy. 15/2003 – UHE/UAB – 04.06.2003 4 Technical incommensurability comes from the multidimensional nature of complexity and refers to the issue of representation of multiple identities in descriptive models. At this point, if we accept that real-world systems are multidimensional in nature, we have also to accept that the evaluation of public plans or projects has to be based on procedures that explicitly require the integration of a broad set of various and conflicting points of view. Consequently, multi-criteria evaluation is in principle an appropriate policy framework. Academic Science Professional Consultancy Post Normal Science Post Normal Science Uncertainty Decision Stakes facts uncertain values in dispute stakes high decisions urgent facts uncertain values in dispute stakes high decisions urgent S. Funtowicz, J. Ravetz Figure 1. Graphical Representation of Post-Normal Science (Source: Funtowicz and Ravetz, 1994) For example, the concept of sustainable development has a wide appeal mainly because it does not set economic growth and environmental preservation in sharp opposition. Rather sustainable development carries the ideal of a harmonisation or simultaneous realisation of economic growth and environmental concerns. Unfortunately, multi-criteria decision theory teaches us that a consequence of taking into account various dimensions simultaneously is that it is impossible to optimise all the objectives at the same time. So that we should learn how to look for “compromise solutions” i.e. the balance between conflicting incommensurable values and dimensions (Bromley, 1998; Faucheux and O’Connor, 1998; Munda, 1997). The arguments developed in this section imply that at least there could be 2 different compromise solutions: a social compromise solution coming from value conflicts and a technical compromise solution coming from conflicting non-equivalent representations of the same policy options. At this stage, basic questions to be answered are: 1. How is it possible to deal with technical incommensurability? 2. How can we deal with the issue of social incommensurability? 3. Which are the main consequences of technical and social incommensurability in a SMCE framework? To answer these questions is the aim of the rest of the present article. 15/2003 – UHE/UAB – 04.06.2003 5 2. Technical Incommensurability and Multi/Inter-Disciplinarity An effective policy exercise should consider not merely the measurable and contrastable dimensions of the simple parts of the system, that even if complicated may be technically simulated (technical incommensurability). To be realistic it should also deal with the higher dimensions of the system. Those dimensions in which power relations, hidden interests, social participation, cultural constraints, and other "soft" values, become relevant, and unavoidable variables that heavily, but not deterministically, affect the possible outcomes of the strategies to be adopted (social incommensurability). In this section the discussion will focus on the technical dimensions of a complex system; in the next section the issue of social incommensurability will be tackled. One should note that the construction of a descriptive model of a real-world system depends on very strong assumptions about (1) the purpose of this construction, e.g. to evaluate the sustainability of a given city, (2) the scale of analysis, e.g. a block inside a city, the administrative unit constituting a Commune or the whole metropolitan area and (3) the set of dimensions, objectives and criteria used for the evaluation process. A reductionist approach for building a descriptive model can be defined as the use of just one measurable indicator (e.g. the monetary city product per person), one dimension (e.g. economic), one scale of analysis (e.g. the Commune), one objective (e.g. the maximisation of economic efficiency) and one time horizon. If one wants to avoid reductionism, there is a clear need to take into account incommensurable dimensions using different scientific languages coming from different legitimate representations of the same system. This is what Neurath (1973) called the need for an “orchestration of sciences”. It is clear that a multi-criteria approach, being multidimensional in nature, seems an interesting framework to make Neurath’s idea operational. To clarify this point I will refer to one real world case study involving the water supply system of the city of Palermo in western Sicily (South Italy). This problem was part of a project which was commissioned by the Sicily region and executed in the frame of the European Commission DGXVI structural funds. This case study was developed in two years of interaction mainly between a multidisciplinary team and the management body of the water supply system of the city of Palermo (plus some social actors involved in the final step of the study) (for more information on this case study see POP Sicily, full final report European Commission contract No.10122-94-03 TIPC ISP I or for a shorter version Munda et. al., 1998). Water resource management is characterised by the presence of a strong competition among different categories of consumptive water uses and, as a consequence, among various interest groups. Such a competition also exists between consumptive uses as a whole and “ecological uses” which aim at limiting water diversion for off-stream uses in order to preserve the ecological equilibrium of ecosystems. This permanent condition of competition may become a real conflict under drought conditions, i.e. when there is a temporary reduction of available water resources due to a long and severe decrease of rainfall (compared to mean or median natural values). The problem of water shortages due to drought is particularly relevant in Southern Europe. In Sicily, the water distribution issue has deep historically routs. The mafia started from the fighting over water control. Water shortages not only depend on hydrological drought which in turn follows from meteorological drought, but also depend on water supply system characteristics and demand levels, which are both affected by different drought mitigation measures. As a consequence, the pure technical hydrological solutions cannot be separated from 15/2003 – UHE/UAB – 04.06.2003 6 their consequences on the socio-economic system. Although this was not evident in the beginning of the project, after a few meetings, hydrologists accepted that an economist could be of some help for this kind of problems. However, it was still very difficult to find a common language and to understand which contribution each could give to progress towards a possible solution (or at least a better understanding) of a such as complex problem. The water system of Palermo has to provide water supply to municipal, agriculture and industrial users by using surface water and groundwater; a reservoir is also used for energy production. It was agreed that alternative management options under drought conditions can be divided into two main groups: • alternatives that try to satisfy 100% of the water demands, • alternatives that do not satisfy completely the water demands. At the moment to specify the alternatives, it was necessary to work on the structure of the Palermo water supply system shown in Figure 2 and… it was immediately clear that this was the job of hydrologists. However, these alternatives had to be evaluated for the longest historic drought experienced in the water supply system (4 years) according to a set of criteria including the economic dimension (e.g. connected financial costs and benefits of the company managing the water supply system, the energy production company, and so on), the social dimension (e.g. hygienic risk and social discomfort) and the environmental dimension (e.g., the in-stream flow requirement defined as the discharge which maintains a stream ecosystem or aquatic habitat). At this point the advantage of the multi-criteria structuring of the problem was evident. Each expert suddenly knew which her /his comparative advantage was. From the experience of this case study, a first lesson can be learned. The use of a multi-criterion framework is a very efficient tool to implement a multi/inter-disciplinary approach. The experts involved had various backgrounds (mainly in engineering, economics and mathematics). While in the beginning, the communication process was very difficult, when it was decided to structure the problem in a multi-criterion fashion, it was astonishing to realize that immediately a common language was created. In terms of inter-disciplinarity, the issue is to find an agreement on the set of criteria to be used; in terms of multi-disciplinarity, the issue is to propose and compute an appropriate criterion score. The efficiency of the interaction process may greatly increase and its effectiveness too2. In the Palermo case study, it was also experienced that the possibility of taking explicitly into account distribution issues increases the transparency of the study and makes possible a process of interaction with various social actors in an effective way. This second lesson leads to the issue of social incommensurability and public participation. 2 Here I refer to the idea of orchestration of sciences as a combination of multi/inter-disciplinarity. Multi-disciplinarity: each expert takes her/his part. Inter-disciplinarity: methodological choices are discussed across the disciplines (this definition has been discussed with R. Strand). 15/2003 – UHE/UAB – 04.06.2003 7 Figure 2. Scheme of the Palermo Water Supply System (Source: Elaborations made by G. Rossi and his team at Catania University) 3. Social Incommensurability: Public Participation, Ethics and Transparency At this point of the discussion, one question arises, who is making the decisions? Some critics of multi-criteria evaluation say that in principle, in cost-benefit analysis, votes expressed on the market by the whole population can be taken into account (of course with the condition that the distribution of income is accepted as a means to allocate votes)3. On the contrary, multi-criteria evaluation can be based on the priorities and preferences of some decision-makers only (we could say that the way these decision-makers have reached their position is accepted as a way to allocate the right to express these priorities. This criticism may be correct if a “technocratic approach” is taken, where the analyst constructs the problem relying only upon experts’ inputs (by experts meaning those who know the “technicalities” of a given problem). For the formation of contemporary public policies, it is hard to imagine any viable alternative to extended peer communities (Funtowicz and Ravetz, 1991, 1994; Funtowicz et al., 1999; Gowdy and O’Hara, 1996). They are already being created, in increasing numbers, either when the authorities cannot see a way forward, or know that without a broad base of consensus, no policies can succeed. They are called "citizens' 3 One should note that indeed cost-benefit analysis can be easily criticised both from the distributive and environmental points of view (see e.g., Munda, 1996; Spash and Hanley, 1995). However I prefer not to deal with this issue here. 15/2003 – UHE/UAB – 04.06.2003 8 juries", "focus groups", or "consensus conferences", or any one of a great variety of names; and their forms and powers are correspondingly varied. But they all have one important element in common: they assess the quality of policy proposals, including the scientific and technical component. And their verdicts all have some degree of moral force and hence political influence. Here the quality is not merely in the verification, but also in the creation; as local people can imagine solutions and reformulate problems in ways that the accredited experts, with the best will in the world, do not find natural. However, one should not forget that even a participatory policy process can always be conditioned by heavy value judgements such as, have all the social actors the same importance (i.e. weight)? Should a socially desirable ranking be obtained on the grounds of the majority principle? Should some veto power be conceded to the minorities? Are income distribution effects important? And so on. The management of a policy process involves many layers and kinds of decisions, and requires the construction of a dialogue process among many stakeholders, individual and collective, formal and informal, local and not (see e.g., De Marchi and Ravetz, 2001; Corral-Quintana et al., 2001, Kasemir et al., 2003). This need has been more and more recognized in a multi-criteria decision-aid (MCDA) framework too. Banville et al., 1998 offers a very well structured and convincing argumentation in this direction. I agree with them on the need of extending MCDA by incorporating the notion of stakeholder; this is the reason why a social multi-criteria process must be as participative and as transparent as possible; although I argue that participation is a necessary condition but not a sufficient one. This is the main reason I propose the concept of “Social Multi-criteria Evaluation” (SMCE) in substitution of “Participative Multi-criteria Evaluation” (PMCE) or “Stakeholder Multi-criteria Decision Aid” (SMCDA) (Banville et al., 1998). To clarify this very important point, I will again use recent empirical examples. I will start with the experience of the so-called VALSE project (see VALSE full final report, Chapter 9, European Commission ENV4-CT960226, or for a synthesis De Marchi et al., 2000). Troina is a small town (10,000 inhabitants) in the North-eastern Sicily, Italy. On the one hand, it seems there is a common assumption that there is an actual water shortage, which could be remedied by more effective use of existing resources. (Paradoxically, although real water shortage is common in Sicily, Troina is an exception). On the other hand, there is a complex and heterogeneous collection of interests in the Troina water issue, who have hitherto had no effective dialogue. Hence an effective structuring of the water problem at this early stage is an important task, so that eventual negotiations among social actors can have a better chance of a positive outcome. The steps of the overall evaluation process followed are schematised in Figure 3. One has to note that policy evaluation is not a one-shot activity; on the contrary, it takes place as a learning process. It has to be realised that the evaluation process is usually highly dynamic, so that judgements regarding the political relevance of items, alternatives or impacts may present sudden changes, hence requiring a policy analysis to be flexible and adaptive in nature. This is the reason why evaluation processes have a cyclic nature. By this is meant the possible adaptation of elements of the evaluation process due to continuous feedback loops among the various steps and consultations among the actors involved (Nijkamp et al., 1990). The first question to be answered is the following: is "business as usual" a possible option in the long run? Business as usual is a situation where power and water management are fragmented among the main actors and where infrastructure actions are the only ones not requiring agreements. This can be considered the classic case of noncooperative resource exploitation. 15/2003 – UHE/UAB – 04.06.2003 9 For example, the Municipality of Troina is trying to become self-sufficient for its drinking water needs using its own spring water sources, even if this could be perceived as inefficient. To evaluate the business as usual option properly, it has to be compared to a set of different possible options on the basis of some evaluation criteria. At this point, an issue immediately arises: alternatives and criteria for whom? This leads to a need to take into account the preferences of some of the actors playing an important role in the problem at hand. Initially, only the actors playing an important role in the community of Troina (as a result of the institutional analysis) were taken into account. Later on, as a surprising feedback of the process of generation of alternative options, it was clear to everybody that additional interest groups outside Troina also had to be taken into account. This learning process was very interesting particularly for the local administrators of Troina, who fully realised the importance of Troina water resources outside their own territory. As the Mayor acknowledged, such a process of structuring the problem at hand was extremely useful for understanding the hierarchy of interests that is behind the exploitation of local natural resources. During the study, the top position of a course of action proposing an information campaign was an unexpected surprise. The response to this surprise was the idea of implementing, within a very short time horizon, an exposition on water management issues in the town of Troina. The Mayor and the municipal administration thought that the implementation cost of such a policy measure was quite low and the positive impacts on the community could be very high. Of course, the political risks for the administration can also be very high, since it was clear that a lot of powerful actors work hard to keep the status quo4. This point leads us to the initial and principal question, is business as usual a defensible option? Figure 3. Scheme of the Evaluation Process in the Troina Case Study 4 Actually, I must say that the Mayor and his administration lost the next elections …. Application of a Multicriteria Aggregation Procedure Application of an specific Conflict Analysis Procedure Construction of an Actors’ Impact Matrix Construction of a Criterion Impact Matrix Choice of set of Evaluation Criteria Generation of Alternative Options Identification of the Main Actors Formulation of the Explanation Hypothesis Interpretation of the Results Institutional Anal y sis MCDA 15/2003 – UHE/UAB – 04.06.2003 16 steps, to the optimum a*?). The final solution is more like a "creation" than a discovery. In Multiple-Criteria Decision Aid (MCDA) (Roy, 1985), the principal aim is not to discover a solution, but to construct or create something which is viewed as liable to help "an actor taking part in a decision process either to shape, and/or to argue, and/or to transform his preferences, or to make a decision in conformity with his goals" (constructive or creative approach) (Roy, 1990). This classical schematised relationship decision-maker/analyst and the related concept of “decision aid” as a learning process for the actors involved seems to me more adequate in situations such as the ones defined as applied science and professional consultancy by Funtowicz and Ravetz (see Figure 6). Since this process seems more adequate for the search of a technical compromise solution, I call it a “technocratic approach”. MCDM MCDA MCDM MCDA PMCE SMCE SMCE Figure 6. Multi-Criteria Approaches in Relationship to Funtowicz-Ravetz Classification of Science for Policy Expansions of MCDA to the social domain have recently been attempted by various scientists (e.g., Banville et al., 1998). For the reasons I discussed earlier in this article, I think that a MCDA participatory approach is still under the conditions of medium uncertainty and medium decision stake ranges. All the arguments and convictions discussed in this article have led me to the development of the concept of Social Multi-Criteria Evaluation (SMCE) whose very essence is the recognition that (see Figure 7): • The use of a multi-criteria framework is a very efficient tool to implement a multi/inter-disciplinary approach. • Science for policy implies a responsibility of the scientists towards the whole society and not just towards a mythical decision-maker. • Public participation is a necessary component but not a sufficient one. Participation techniques are a tool for improving the knowledge of the problem at hand and not for receiving inputs to be used uncritically in the evaluation process. Social participation does not imply lack of responsibility. 15/2003 – UHE/UAB – 04.06.2003 17 • Ethical judgements are unavoidable components of the evaluation exercise. These judgements always influence heavily the results. As a consequence, transparency on the assumptions used is essential. • In this framework, of course mathematical aggregation conventions play an important role, i.e. to assure that the rankings obtained are consistent with the information and the assumptions used13. This discussion leads to the need of defining the concept of evaluation as the combination of representation, assessment and quality check connected to a given policy problem in relation to a given objective14. This is the reason why I use the term “multi-criteria evaluation” and not “multi-criteria decision” when a social context is implied. QUALITY OF PRODUCT PROCEDURAL RATIONALITY LEARNING HOLARCHIES MCDA QUALITY OF “SOCIAL” PROCESS PARTICIPATION TRANSPARENCY MULTI/INTER-DISCIPLINARITY ETHICS RESPONSIBILITY CONSISTENCY Figure 7. Synthesis of a Social Multi-Criteria Evaluation Process Let’s finally have a look at the consequences of social and technical incommensurabilities on the so-called axiomatization issue. When different conflicting evaluation criteria are taken into consideration, a multi-criteria problem is mathematically ill-defined. The consequence is that a complete axiomatization of a multi-criterion aggregation convention is quite difficult. To deal with this problem, two main approaches can be distinguished. 13 I insist on the importance of the algorithmic component in SMCE. Indeed I used the term “nonalgorithmic” multi-criteria evaluation as an implementation tool for the incommensurability principle (Martinez-Alier et al., 1998). This term was intended to emphasize the importance of the decision process however I think was an unfortunate choice since it gives the impression that the algorithmic component is not useful at all. 14 This definition has been developed thanks to discussions with M. Giampietro. 15/2003 – UHE/UAB – 04.06.2003 18 1. The attempt to check under which specific circumstances each method could be more useful than others, i.e. the search of the right method for the right problem (e.g., see Guitouni and Martel, 1998). 2. The attempt of looking for a complete set of formal axioms that can be attributed to a specific method (e.g., Arrow and Raynaud, 1986; Vincke, 1994). Here, I will try to isolate some properties that may be considered desirable for a discrete multi-criteria method in the framework of SMCE. Of course in another framework, e.g. stock exchange investments, these properties can easily be irrelevant or even undesirable. A deeper discussion on this topic can be found in Munda (2003). The idea of technical incommensurability makes the following properties desirable: • Indifference and preference thresholds should be explicitly taken into account. • Mixed information of the widest type should be addressed in a consistent way. • Simplicity, meaning the use of as less parameters as possible, is a very desirable property to guarantee transparency. • The hierarchical dimension of a policy problem should be explicitly considered. The idea of social incommensurability makes the following properties desirable: • Weights in this framework are meaningful only as importance coefficients and not as trade-off. As a consequence, complete compensability cannot be implemented. • Sensitivity and robustness analysis have to check the consequences on the final ranking of only some clear ethical positions and not of all the possible combinations of weights. • Conflict analysis procedures explicitly looking for social compromises should integrate a SMCE exercise. • In a policy framework, to have a complete ranking of all the alternatives is more useful than just to select one alternative only; this implies that dominated alternatives cannot be excluded a priori. ACKNOWLEDGEMENTS Comments by the anonymous referees are gratefully acknowledged. This research has been partly financed by the European Commission, research project: Development and Application of a MultiCriteria Decision Analysis Software Tool for Renewable Energy Sources (MCDA_RES), Contract NNE52001-273. References Arrow K.J., Raynaud H. (1986) - Social choice and multicriterion decision making, M.I.T. Press, Cambridge. Bana e Costa C.A. (1990) - An additive value function technique with a fuzzy outranking relation for dealing with poor intercriteria preference information, in Bana e Costa C.A. (ed.)-Readings in multiple criteria decision aid, Springer-Verlag, Berlin, pp. 351-382. Banville C., Landry M., Martel J.M., Boulaire C. (1998) – A stakeholder approach to MCDA, Systems Research and Behavioral Science, Vol. 15, pp. 15-32. Bouyssou D., Vansnick J.C. (1986) – Noncompensatory and generalized noncompensatory preference structures, Theory and Decision, 21, pp. 251-266. Bouyssou D. (1990) - Building criteria: a prerequisite for MCDA, in Bana e Costa C.A. (ed.)- Readings in multiple criteria decision aid, Springer-Verlag, Berlin, pp.58-80. Bromley D. (1998) – Searching for sustainability: the poverty of spontaneous order, Ecological Economics, 24, pp. 231-240. 15/2003 – UHE/UAB – 04.06.2003 19 Climaco J.C.N. (2000) – Uma reflexao critica da decisao óptima, Abertura Solene das Aulas na Universidade de Coimbra, Coimbra. Corral Quintana S. A. (2001) - Una metodología integrada de exploración y comprensión de los procesos de elaboración de políticas públicas, unpublished Ph.D. Thesis, Univ. La Laguna, Dep.. of Economics, Spain. Corral-Quintana, S., B. De Marchi, S. Funtowicz, G, Gallopín, Â. Guimarães-Pereira, B. Maltoni (2001) - The Visions Project at the JRC, European Commission, Joint Research Centre, Institute for the Protection and Safety of the Citizens, EUR 19926 EN [http://alba.jrc.it/visions]. De Marchi B., Funtowicz S.O., Lo Cascio S., Munda G. (2000) – Combining participative and institutional approaches with multi-criteria evaluation. An empirical study for water issue in Troina, Sicily, Ecological Economics, vol. 34 No. 2, pp. 267-282. De Marchi B., Ravetz J. (2001) – Participatory approaches to environmental policy, Concerted Action EVE, Policy Research Brief, No. 10. Faucheux S. and O'Connor M. (eds.) (1998) - Valuation for sustainable development: Methods and policy indicators, Edward Elgar. Funtowicz S.O., Ravetz J. R. (1991) - A new scientific methodology for global environmental issues, in R. Costanza (ed.)- Ecological Economics, New York, Columbia, pp. 137-152. Funtowicz S.O., Ravetz J. R. (1994) - The worth of a songbird: ecological economics as a post-normal science, Ecological Economics, 10, pp.197-207. Funtowicz S., Martinez-Alier J., Munda G. and Ravetz J. – (1999) Information tools for environmental policy under conditions of complexity, European Environmental Agency, Experts’ Corner, Environmental Issues Series, No. 9. Geach, P. (1967) - Good and Evil, in P. Foot (ed.) - Theories of Ethics, Oxford University Press, Oxford. Giampietro, M. (1994). Using hierarchy theory to explore the concept of sustainable development. Futures 26 (6): 616-625. Giampietro M., Mayumi K. (2000) – Multiple-scale integrated assessment of societal metabolism: introducing the approach, Population and Environment, Vol. 22, No. 2, pp. 109-154. Giampietro M., Mayumi K. (2000) – Multiple-scale integrated assessment of societal metabolism: integrating biophysical and economic representations across scales, Population and Environment, Vol. 22, No. 2, pp. 155-210. Gowdy, J.M., and O'Hara S. (1996) - Economic Theory for Environmentalists. Saint Lucie Press. Guitouni A., Martel J.M. (1998) – Tentative guidelines to help choosing an appropriate MCDA method, European Journal of Operational Research, 109, pp. 501-521. Kasemir B., Gardner M., Jäger J. and Jaeger C. (eds.) (2003) – Public participation in sustainability science, Cambridge University Press, Cambridge. Kleijnen J.P.C. (2001) – Ethical issues in modelling: some reflections, European Journal of Operational Research, 130, pp. 223-230. Koestler A. (1969) – Beyond atomism and holism: the concept of the holon, in A. Koestler and J.R. Smythies (eds.) – Beyond reductionism, Hutchinson, London, pp. 192-232. Kuhn T.S. (1962) - The structure of scientific revolutions, University of Chicago Press, Chicago. Martí N. (2001) – Processos de decisió i instrumentalització de l’ avaluació d’ actuacions en el territori. Una proposta metodòlogica d’ avaluació integrada a l’ entorn del Parc Nacional d’ Aigüestortes i estany de Sant Maurici: el cas de l’ estudi DIAFANIS, Master Dissertation, Doctoral Programme in Environmental Sciences, Univ. Autonoma de Barcelona, Spain. Martinez-Alier J., Munda G., O’Neill J. (1998) – Weak comparability of values as a foundation for ecological economics, Ecological Economics, 26, pp. 277-286. Moreno-Jiménez J. M., Aguaron J., Escobar T., Turon A. (1999) – Multicriteria procedural rationality on SISDEMA, European Journal of Operational Research, 119(2), pp. 388-403. Munda G. (1993) – Multiple-criteria decision aid: some epistemological considerations, Journal of Multi-Criteria Decision Analysis, vol. 2, pp. 41-55. Munda G. (1995) - Multicriteria evaluation in a fuzzy environment, Physica-Verlag, Contributions to Economics Series, Heidelberg. Munda G. (1996) - Cost-benefit analysis in integrated environmental assessment: some methodological issues, Ecological Economics, Vol. 19, N. 2, pp. 157168. Munda G. (1997) - Environmental economics, ecological economics and the concept of sustainable development, Environmental Values, vol. 6, No. 2, pp. 213-233. Munda G., Paruccini M. and Rossi G. (1998) - Multicriteria evaluation methods in renewable resource management: the case of integrated water management under drought conditions, in Beinat E. Nijkamp P. (eds.) - Multicriteria evaluation in land-use management: methodologies and case studies, Kluwer, Dordrecht, pp. 79-94. Munda G. (2003) – An axiomatic solution for the discrete multi-criterion problem in a public choice framework, Unpublished Manuscript. 15/2003 – UHE/UAB – 04.06.2003 20 Neurath O. (1973) – Empiricism and sociology, Reidel, Dordrecht. Nijkamp, P., Rietveld P. and Voogd H. (1990) - Multicriteria Evaluation in Physical Planning, Amsterdam: North-Holland. O'Connor M., Faucheux S., Froger G., Funtowicz S.O., Munda G. (1996) -Emergent complexity and procedural rationality: post-normal science for sustainability, in R. Costanza, O. Segura and Martinez-Alier J. (eds.) -Getting down to earth: practical applications of ecological economics, Island Press/ISEE, Washington D.C., pp. 223-248. O'Neill, J. (1993) - Ecology, Policy and Politics, Routledge, London. Rauschmayer F. (2000) – Ethics of multicriteria analysis, International Journal of Sustainable Development, Vol. 3, No. 1, pp. 16-25. Roberts F. S. (1979) - Measurement theory with applications to decision making, utility and the social sciences, Addison-Wesley, London. Rosen, R. (1977) - Complexity as a System Property, Int. J. General Systems Vol. 3: 227 -232 Roy B. (1985) - Méthodologie multicritere d' aide à la decision, Economica, Paris. Roy B. (1990)- Decision aid and decision making, in Bana e Costa C.A. (ed.)- Readings in multiple criteria decision aid, Springer-Verlag, Berlin, pp. 17-35. Roy B. (1996) - Multicriteria methodology for decision analysis, Kluwer, Dordrecht. Roy B., Damart S. (2002) – L’analyse Coûts-Avantages, outil de concertation et de légitimation? Metropolis, N. 108/109, pp. 7-16. Saaty T.L. (1980) - The analytic hierarchy process, McGraw Hill, New York. Simon H.A. (1976) - From Substantive to procedural rationality. In: J.S. Latsis (Editor) Methods and appraisal in economics, Cambridge: Cambridge University Press Simon H. A. (1983) - Reason in Human Affairs. Stanford: Stanford University Press. Spash C., Hanley N. (1995) – Preferences, information, and biodiversity preservation, Ecological Economics, 12, pp. 191-208. Vansnick J.C. (1986) – On the problem of weights in multiple criteria decision making (the noncompensatory approach), European Journal of Operational Research. 24 pp.288-294. Vincke Ph. (1994) – Recent progresses in multicriteria decision-aid, Rivista di Matematica per le scienze Economiche e Sociali, Vol. 2, pp. 21-32.