Meso-Matrical Synthesis of the Incommensurable

Meso-Matrical Synthesis of the Incommensurable # Bojan Radej* Abstract: Many of those currently involved in the assessment of government interventions’ contribution to the social welfare have had significant difficulties in summarizing known but sometimes contradictory facts into summary conclusions There are incommensurable viewpoints with regard to many social realities, and they provide us with very different ‘numeraires’ and macro-views of the world which are not reducible to common denominator In particular, there is a disagreement over assumptions about the aggregation of the assessed policy impacts (micro) into summary conclusions (macro) that inform decision-makers operating at meso level A new method is proposed of meso-matrical impact assessment (MIA) of policy interventions to cope with social incommensurability in scale (micro-mesomacro) and scope (economic-social-human-natural) It is based on Leontief's square inputoutput matrix that evaluate overlaps between incommensurable sets on their margin In MIA secondary issues play a central role Practical example illustrates the achievement, schematic presentation generalises the principle Paper explains that social or political common ground is not the condition of unity and cohesion with the meso-matrical perspective in mind Keywords: Incommensurability, scope, scale, meso, matrix, evaluation JEL Code: A13, A10, F12, A19, C50, H40 # Revised 15 December 2008; submited in October 2008 * Bojan Radej, MSc in economics; independent social researcher; Slovenian Evaluation Society; Ljubljana, Slovenia, bojan.radej@siol.net 1 Introduction Sustainable development is a norm that obliges governments to put forward policies that ‘meet the needs of the present generation without compromising the ability of future generations to meet their own’ (WCED, 1987) This calls for multi-criteria assessment of policy interventions which significantly complicates summative evaluation There are incommensurable viewpoints and values with regard to different social realities, such as economic, social and environmental, or local and global, which provide us with very different ‘numeraires’ and views of the world, and they are not reducible to one common denominator (Funtowicz and Ravetz, 1994, in Martinez-Alier et al., 1998) For the fact that variegated forms of welfare need to be taken into account implies that ‘different principles of social primacy and legitimacy must be reckoned with and reconciled’ (Wacquant, 1997) in a multicriteria impact assessment (IA) Incommensurability is not accompanied only by value comparisons but also by scientific ones since its formal systems decisively differ in the propositions they take as their axioms When confronted with multidisciplinary issues involving scientific controversy, even competent, honest and disinterested scientists may arrive at different conclusions because of systematic differences in the way they summarise available information of one and the same world (Kuhn, 1970; Mumpower et al., 1996) Kuhn (1993 in Sankey, 1998) concludes that theories and scientists are unable to truly reflect reality When values and facts are based on incommensurable oppositions, there is no objective basis for rational choice between theories and than no neutral observation of social reality is possible No individual person or theory is than able to fully comprehend social complexity – such as sustainable development - and its multiple meanings are persistently disagreed collectively This alleges that multiplicity is one of the determining conditions not only of values and facts but also of social systems (Munda, 2004) The current welfarist model is on the other hand unable to conceptualize and operationally apply substantively different aspirations in a public sphere Bar-Yam (2003) thinks that governments have experienced systemic failures in addressing complex welfare issues such as large-scale and multi-scope matters They rely on reductionist approaches, which simplify the welfare concern by dividing it into independently ministered sub-problems The very act of simplifying by sub-divisions loses the interconnections and therefore cannot tackle overlapping concerns (Chapman, 2001) The analytical concept of reality, is for example based upon the idea that the total is equal to the sum of its parts and that the parts can be quantitatively described, is insufficient for studying complex issues A complex phenomenon is a structure of sets, connected by ordering relations of ‘super-, sub- and co-‘, where each sub-system (itself a system) has a plurality of relations of all three sorts with other subsystems (Ravetz, 2006) Reasoning on different levels is discontinuous and transgression between them (Weaver, Rotmans, 2006) breaks linearity For complex social considerations, where ‘sum differs from the aggregate of their parts’ (Veen, Otter, 2002), the summation procedure is far from trivial (Veen, Otter, 2002), so there is an aggregation problem (Foster, Potts, 2007) in the summative IA For sustainable development to flourish, an approach to welfare considerations is needed that is summative in a way that is equally sensitive for all incommensurable aspects Existence of an aggregation problem in policy studies is probably less obvious than the vast consequences it manufactures In policy-making even good individual policies, based on strong values and common sense, often lead to disappointing results Inability of a The Webster Dictionary (1971) extends the meaning of ‘incommensurable’ beyond the concept of measure: it also means lacking a common basis of comparison with respect to a quality normally subject to comparison government to coordinate and aggregate opposite claims can be hold responsible for generating growth antagonisms since 1960’s such as income inequality (Giraud, 1996; Milanović, 2006), and happiness paradox (Easterlin, 1974) where increased material wealth is not accompanied by a subjective satisfaction; an ecological footprint (Wackernagel, Rees, 1994) shows that the wealthiest nations actually consume resources at a volume that could be sustained only if the planet Earth were two to almost six times larger It was not that growth antagonisms took place in the absence of policy evaluation Evaluation has increasingly obtained its role in a policy cycle since the sixties, but it failed to address deep conflicts that accompany large-scale and multi-scope policy interventions So ‘ill development’ took place with our informed knowledge Thus the question how to think intelligently about the conflict nature of trade-offs between different aspects of welfare that are characterised by the incommensurable oppositions (Williams, 1972, in Martinez-Alier et al., 1998) arises again Arrow explained using his impossibility theorem (1951, Nobel Prize for Economics in 1972), that it is not possible to scale up from all individual preference functions to produce a social welfare or “public interest” function that satisfies desirable properties of an aggregation process (Evans et al, 2002) such as non-dictatorship The theorem has challenged the presumption that an aggregation of individuals’ wants such as in Condorcet’s majority voting rule or Bentham’s ‘the largest utility for all’ can simultaneously meet individual and collective expectations It further demonstrated that there is a discontinuity between rationality at the individual and group level (Evans et al., 2002) that mirrors the tension between private and public in the governance of the social Coleman (1986 in Åberg, 2000) maintains that this micro-to-macro link, also referred to as social causation (Sawyer, 2003) is a controversial and most poorly developed part of sociological theory Associated with these meta-theoretical concerns, there is an apparent paradigm crisis in IA (Virtanen, Uusikylä, 2004; Hertin et al, 2007) IA aims to suggest more consistent policy interventions (Hertin et al, 2007) However, evidence shows that policy response to evaluation recommendations is very limited (Picciotto, 1999) Existing IA methods are designed for the appraisal of homogeneous interventions or ‘projects’ with only limited diversity of impacts (Elbers et al., 2007; Rotmans, 2002), while governments work with heterogeneous issues IA conventionally forwards two-valued or binary logic which is focused on only two scopes of the assessed phenomena - cause and effect -, such as economy’s impact on environment But government operates in multi-value context, where causes and effects are not straightforward The problem is also that majority of IA models assess processes only at one scale level, micro or macro This all results in a ‘evaluation deficit’ – the situation in which most assessments provide the kind of information that does not inform policy-makers whether a global objectives can be met – consequently, evaluations remain under-utilized (Stame, 2004) or even get mis-used Summation is the Achilles heel of the evaluation effort (Scriven, 1994) It arises from a disagreement over assumptions about the aggregation of numerous policy impacts when assessment phenomenon is multiple in scope (economic, social…) and scale (micro, macro) As elaborated by Carlsson (2000), different evaluators assessing the same policy interventions may come to opposite conclusions solely because of differences in their assumptions about the summation of the assessed impacts across incommensurable scopes and scales We often think of them as coupled because of the most common ways in which we encounter them (Bar-Yam, 2004): consider observing a system through a camera that has a zoom lens For a fixed aperture camera, the use of a zoom couples scope and resolution (scale) in the image it provides “As we zoom in on the image we see a smaller part of the world at a progressively greater resolution This leads to a particular relationship of observations of parts and wholes, suggesting that when observing details of the system, the whole is not being observed We must allow a decoupling of scope and resolution, so that the system as a whole can be considered at differing resolutions as well as part by part For this purpose scale can be considered as related to the focus of a camera—a blurry image is a larger scale image— whereas scope is related to the aperture size and choice of direction of observation” (BarYam, 2004) For evaluator this means that what one sees is always predefined in scope and scale of his or her observation Recognition of this is central for the methodology of meso-matrical impact assessment (MIA) that is proposed here It builds on previous observations that to summarise the trade-offs between intrinsic sustainable values, evaluator needs to derive conclusions across different scales and scopes at which appraisal takes place (Weaver, Rotmans, 2006; Dopfer et al, 2004; Dopfer, 2006; Easterling, Kok, 2002) As Chapman (2001) elaborated: the theory of complex systems offers an alternative to conventional IA strategy for studying largescale issues: instead of going down to the elementary level as in the standard reductionist model, the system theory justifies the opposite, namely going up a level of abstraction which establishes a multi-scale view The aggregation problem in IA that apply the impact matrix method has been first addressed by Luna Leopold et al (1971) They proposed a detailed expert-based impact matrix at the micro-level from which the macro aspect remains absent It is concerned with only two scopes – economic and ecological – and assesses the possible impact of the former on the latter (second chapter) Recently Ekins and Medhurst (2003) proposed a method for IA in the complete multi-scope perspective They developed a reduced version of Leopold matrix, but with the scope dimension expanded on the four scopes of sustainability (Leopold-EkinsMedhurst impact matrix – LEM); they further allowed for aggregation of all impacts, represented in rows of their matrix on each particular scope – by column of the LEM An analogous approach is conventionally applied in various standard IA procedures, such as the strategic IA (2001/42/EC), territorial IA (ESPON – 3.2, 2006), and ex-ante IA of the contribution of the EU structural funds to sustainability of regional development (GHK et al, 2002) Nevertheless, LEM’s summation approach is inappropriate, as the effects of individual policy measures on each appraised scope are not homogenous (Rotmans, 2006), and therefore not commensurable So impacts could be aggregated by column in LEM only by source and area of impact, i.e partially This reorganises LEM into square matrix of Leontief (1951) with equal number of rows and columns (third chapter) We learn that social complexity does not prohibit aggregation in IA when one takes apart weak from strong incommensurability This permits two step summation procedure in IA - the second step is synthesis with the correlation of intersections between different scopes that are obtained from the Leontief matrix Both summative steps are illustrated with a practical example The regional development programme for the Slovenian region of Pomurje for 2007-2013 (RDPP) is first assessed exante with LEM and then with the MIA Their summary results are found diverging The second part of the paper studies reasons for this incompatibility To accomplish this task a complete multiple-scale perspective is introduced into the IA (fourth chapter) When scale and scope aspects of IA are combined it becomes obvious that inconsistency in standard IA arises from inappropriate interpretation of assessment results across scope and scale MIA offers new understanding on how to integrate policy concerns without imposing uniform norms The implications for the governance of the social close up the paper Work continues Summation in standard approaches The first impact matrix generation methodologies for the ex-ante assessment of large scale policy interventions such as the one elaborated by geologist Luna Leopold et al (1971; for review see Munn, 1979) explicitly reject the summation of multifarious policy impact into aggregate indicator They were concerned with the impact of economic policy measures on the components of the environment Their impact matrix lists the 100 most important economic policy actions horizontally and 88 environmental fields of impact vertically The intersection between these two scopes creates a detailed matrix with 8800 fields – each further divided into four subfields that characterise each impact by its size (strong, medium, small), direction (positive, negative), probability and risk factor (critical or not) In this way, all possible impacts are presented in sufficient detail to enable informed decision In the finest analytical manner, Leopold aims to provide the large picture of complex policy issue with a detailed description of all its elementary parts The first generation methodologies claim that impacts should be presented disaggregated, leaving policy-makers with full responsibility for the synthesis conclusions in the light of a policy decision There must be a clear demarcation in IA between the evaluator and policy-makers to ensure that value judgement rests with policy-makers and is not inappropriately shifted towards the evaluator Refusal of aggregation is then important because it protects the evaluator from the political interference (Kunseler, 2007) that accompanies decision-making The rejection of summation in evaluation of large-scale policy interventions is problematic Let us recall the impossibility theorem, which implies that the social optimum can not emerge from a reduction procedure It is precisely the inability of policy-makers as social aggregators that calls for a policy evaluation in the first place If evaluators fully accept their role in the policy cycle, they should not refuse summation of evaluation results! Another difficulty is that Leopold and even many contemporary authors observe incommensurable values as antagonistic, so they can not explain how opposing views work together, without which it is impossible to say anything about integrative capacity of a policy proposal Failing to summarize impacts in the assessments is “letting the client down at exactly the moment they need you most” (Scriven, 1994) Without aggregation, an appraisal will usually not generate clear-cut conclusions but ‘information overload’ fostering a piecemeal rationality Results that are too fragmentary and unrelated offer little value, fail to satisfy information needs at the strategic level and cannot provide an informed basis for a decision All too often disaggregated results produce banal answers to complex and multidimensional societal problems (Virtanen, Uusikylä, 2004) Evaluation methodology hewing simply to the production of non-overlapping information tends to underplay the need for learning and substantiation of findings, legitimating the disregard of stakeholder issues (Stake, 2001), which alone is sufficient to leave an evaluation utterly exposed to a political interference To synthesize or not to synthesize – this is how the question is again posed for Scriven (1994) In somewhat less tragic storyline he explains that we need to distinguish between cases in which it is improper to push for a synthesis and cases in which it is improper not to This fuzzy instruction may not be exciting for writers of tragedy, but it is very helpful for an evaluator To see how this advice helps us, we first need to broaden the conceptual framework of social incommensurability to identify the conditions under which summation is feasible even though commensurability of welfare indicators is abandoned Incommensurability is of two kinds: weak for individual considerations and strong for collective ones Two arguments aim to justify the distinction below: uniqueness of individual elements of the system must to be delimited from incommensurability of social issues; under certain conditions even different aspects of social incommensurability can be traded between each other At the elementary or micro level the social system does not entail incommensurability but uniqueness (Li Xiaorong, 2007): for example, even though individuals are unique, they can be compared because their unique formation of characteristics is obtained from the shared base of human characteristics The result is that individuals, their statements and acts are in some characteristics similar and comparable to some other individuals, but not to all others So similar sorts of individual uniqueness are partly commensurable and can be locally averaged The aggregation of diverse impacts takes not only that the positive (or negative) effects of one particular policy measures can be summed up More crucially it also assumes that a positive effect of one policy intervention outweighs its own or another policy’s negative effect Is it, for example, sustainable to trade tons of greenhouse emissions for euros in EU’s scheme of tradable pollution permits market, when we know that greenhouse emissions can cause irreversible changes in the atmospheric conditions? Such a trade-off is not adequate as a general or macro principle, because economic and environmental aspects of welfare are equally important However, one needs to reason in a multi-scale perspective, where micro is different from macro Trade-offs between economy and environment are not accompanied by incommensurability in every single case, or at least people and their communities are not willing to treat them as such To incorporate this peculiarity in policy evaluation, system thresholds of sustainability – such as ecological, human and social ones – have emerged as a new assessment determinant (for a survey of literature see Muradian, 2001) System thresholds exist that should not be crossed for this would endanger the basic integrity of the system Policy interventions that could jeopardize the thresholds should be avoided in order to circumvent (even more) antagonistic confrontation between, say, climate conditions and economic growth So the concept of thresholds is closely linked to the concept of incommensurability As Wiggins (1997) explains, two values are incommensurable if “there is no general way in which A and B tradeoff in the whole range of situations of choice and comparison in which they figure” Measurable social phenomena are incommensurable in IA ‘only’ beyond (or sometimes below) their threshold values However, within safety limits, an agent either does not sense the difference between two conditions, or refuses to declare a preference for one or the other (Luce, 1956 in Munda, 2006) such as in the cases of minor damages that stay within ‘safe’ ecological standards The practical consequence for policy evaluation is that positive and negative impacts can be aggregated (averaged, regressed, correlated, benchmarked etc.) only if they occur in the safe interval of system normality Reliance on thresholds also simplifies evaluation because risk factors would be already incorporated in the definition of thresholds Where the appropriate threshold level is in every particular case, that is certainly a hotly debated issue, but this does not obviate the need for their imposition in policy evaluation They simply account for the fact that there are discontinuities in the measurement of a valuebased phenomenon and value addition (Mason, 2006) that comply with discontinuity in individual and social values Thresholds set absolute limits to the possibility of trade-offs between different aspects of incommensurability (sustainability) on the micro level This further necessitates a distinction between weak and strong incommensurability of the appraised impacts Griffin (1986 in Morgan, 2007) argues against strong incommensurability because it repeatedly fails at every level of discourse except for debates about the nature of reality and truth (Morgan, 2007) Martinez-Alier et al (1998, in Stagl, 2004) pointed out that in complex social systems, when there is an irreducible value conflict, we can only search for weak comparability in marginal, non-core issues as a facilitator of collective discourse A consequence of this distinction for IA is that the summation algorithm should follow separate procedures for the aggregation of weakly incommensurable micro elements of the assessment and for the macro synthesis of conclusions that are based on components that are strongly incommensurable The first step is addressed in the remainder of this chapter, and the second one in the forthcoming chapter Acknowledgment of weak incommensurability on the micro level enabled a methodological progress in the IA aggregation procedures Among the first proposals was the methodology of strategic environmental assessment (SEA Directive, 2001/42/EC), that broadened the assessment focus to strategic issues that go beyond particular programmes and policy jurisdiction Further progress came from Ekins and Medhurst (2003) in their proposal for the IA of the contribution of the EU structural funds to regional sustainable development in the light of their four capitals – economic, social, environmental and human This four-valued logic of sustainability can be traced back to Brundtland report (WCED, 1987) and others (Munashinghe, 1992; Ekins, 1992) Proposal for the territorial IA in EU (ESPON Project 3.2, 2006; Camagni, 2007a, b) goes even further, proposing a composite indicator of territorial impact which is assuming commensurability of assessed impacts, and is therefore altogether contrary to Leopold For the reasons given earlier (refusal of composite as well as disaggregated measures), the most important recent proposal in the summative methodology may be attributed to Ekins and Medhurst They acknowledged the existence of social incommensurability (their four capital model) and also properly allowed for a partial aggregation of diverse policy induced impacts of a similar sort They have proposed highly compacted form of the Leopold matrix The columns are reduced from 88 fields of possible environmental impacts to four scopes of regional sustainability The first step in their procedure is standard for an assessment: composition of LEM starts when stakeholders of the appraisal select a smaller number of representative evaluation criteria for the assessment of policy impacts on each of the four scopes In the second step participating experts judge how the implementation of each of the proposed policy instruments impacts each evaluation criterion Impacts are expressed as relative differences on a scale between neutral impact (0; either no impact or positive and negative impacts mutually neutralised), weakly, medium and strongly positive (+, ++, +++) or negative (-, , -) impacts In the last step, all appraised impacts are aggregated within each scope for all criteria (horizontally) and than these aggregates are further aggregated (vertically) for each of the four scopes (vertical) The result is not one composite indicator but four composite measures of impact, in four column sums Table presents aggregate results of LEM, from practical example that will be also used in the next chapter to elaborate summation problems in IA Regional development programme for Pomurje region (2007-2013; RDPP) consists of 47 policy measures Their impacts were assessed against eight evaluation criteria (two for each scope: economic – E, human – H, social – S, and nature – N) All appraised impacts (47*8 fields; see Radej, 2007) were aggregated in two respects: the first impacts for both indicators of each scope (obtaining four columns in Table 1) and than also impacts for all 47 measures into six main policies (six rows in Table 1) Finally, the impacts of the RDPP are aggregated by columns to obtain four aggregate impact indicators Table 1: The LEM matrix of the RDPP’s impact on regional sustainability Incommensurable scope of regional SD* Sectoral policy Value added growth Tourism Health Rural development Infrastructure Environment Summary impact of RDPP E H S N +++ + +++ +++ + ++ + + + + + + + + ++ + + + 0 ++ ++ + + Source: Radej, 2007 Note: * SD - sustainable development Table suggests the following overall conclusions Impacts of RDPP on the four scopes of sustainability as identified in the summary row not appear unacceptable, they are all positive, and the differences are relatively small The most problematic thing is that growths of income will negatively impact social development RDPP will most improve economic situation (++), while it will have only a weak positive impact on the improvement of social, human and nature sustainability Individual policies support each others’ goals too weakly; nevertheless, for evaluator RDPP is satisfactory proposal because it promises a rather balanced and therefore sustainable improvement of all scopes of regional sustainability By deriving these conclusions, the broad purpose of the LEM is fulfilled But assessment results can be analysed further if properly rearranged Ekins and Medhurst incorporated social incommensurability into the evaluation, but only for policy goals and not for policy interventions This is important because sectoral policies not produce homogenous impacts; they are not neutral in scope, so they should not be aggregated by columns Partial aggregation If detailed expert assessments of policy intervention’s impacts are not summarized, as is the case in Leopold, the assessment produces findings that are too fragmented In contrast, full aggregation, such as in a LEM, causes that findings are too much amassed Both approaches waste information on the complex constitution of the appraised phenomenon In consequence both wade in inconsistent interpretation of the conclusions Column aggregation in a LEM assumes the homogeneity of the impacts of different policies on each evaluation scope However, many studies demonstrate that the policy impacts are differentiated (Schnellenbach, 2005) even for those policies that had previously been taken as the most neutral, such as monetary (Lucas, 1972) and tax policy (Leith, Thadden, 2006) In principle all institutional interventions should be due to their specialization addressed in terms of their inadequacy, in relation to the general interest they are supposed to serve (Donzelot, in Burchell et al., 1991) In such case secondary impacts should play a decisive role in policy evaluation Policy impacts are not only direct or primary when predictably affecting the target impact area, but also secondary when, by and large unpredictably, impact not targeted areas (Rotmans, 2006) For example, economic trends are only partly influenced by economic policies when other sectoral policies play a role Thus it is usually impossible to unambiguously link a sectoral policy to the development of the corresponding sector because too many other variables are changing (Chapman, 2004) Yet secondary impacts are routinely diminished in evaluation Primary and secondary impacts of different policies are therefore not comparable This is because part of the difference between impacts is structural or deep so it must be preserved during evaluation For example, industrial policy produces impacts on poor, that should not be mixed with their impacts on the environment because social and environmental concerns themselves are not commensurable In the assessment of issues that involve social incommensurability only partial aggregation for the same source and area of impact is allowed Partiality is crucial because it enables summary without recourse to a single type of value (Martinez-Alier et al., 1998) This sets a relative limit or relative condition of aggregation in this type of IA Together with the ‘absolute’ or thresholds limit discussed earlier, these two are necessary conditions to fulfil for preserving deep differences (Ostmann, 2006) in the assessment of complex social issues The vertical summation in LEM is thus only permitted for the same kinds of impacts of the same sort of sectoral policy This transforms LEM into a square Leontief's input-output matrix, the rows and columns of which are organised in the same way (Table below) Square matrix has been applied to assessing various aspects of sustainable development previously Daly (1993) has applied it to present flows between society and nature In their effort to develop an expanded measure of wealth, Dixon et al (1988) also followed the input-output principle when they observed social capital both as an input into the development process together with other forms of capital, and as an output of this process Input-output view is also known in IA (Rotmans, van Asselt, 2001a), even though not in a square form Russian economist Leontief proposed square matrix in the 1950s (Nobel Prize for Economics in 1973) initially to facilitate inter-sectoral studies because it is suitable for presenting the tension between sectors, direct and indirect, in an explicit way The technique is a descendent of the first macroeconomic modelling tool, a Tableau Economique (1758) of the French physiocrat Francois Quesnay, who visualized the economic activity of the whole country as one huge accounting system It follows from the obvious nature of economic transactions that each revenue item of an enterprise or household must reappear as an outlay item in the account of some other enterprise or household For instance, if we have agriculture, industry and services as three sectors in an economy, we can show in a matrix, by rows and columns, how each sector is linked with the other two The agricultural sector may require some capital goods and chemicals from the industrial sector and at the same time it supplies the industrial and service sectors with its product output as their intermediate input The transformation of LEM into Leontief’s matrix has important consequences for the evaluation methodology which will be explicated in steps by the end of the paper But let us first return to a practical example and see what has been miscalculated by LEM Table presents Leontief’s matrix of the RDPP’s estimated impacts organized by source and area of impact; its 47 measures are aggregated in four rows of the matrix by source of impact Squared matrix explains how RDPP is seen from the viewpoint of four regional scopes of sustainability The matrix indicates how RDPP as input (in rows) will impact the areas of sustainability (in columns) For example, the economic policy intersection with the human scope of regional sustainability is denoted as E∩H, etc – the intersection sign ‘∩’ from the set theory This rearrangement of LEM focuses the evaluator not on the relationship between policies and goals but on the intersection between different scopes and their causal, onedimensional trade-offs (where E∩H is of course differentiated from H∩E, etc) Table 2: Leontief’s impact matrix of the RDPP Outputs (area of impact) E H S N E∩E = +++ E∩H = + E∩S = - E∩N = + H H∩E = H∩H = + H∩S = + H∩N = S S∩E = +++ S∩H = + S∩S = ++ S∩N = ++ N N∩E = + N∩H = + N∩S = + N∩N = + Inputs (source of impact) E Source: Radej, 2007 The result of partial aggregation of the appraised impacts by source and area into a square matrix is obtaining 16 sub-aggregates This is still too fragmented to grasp IA findings One approach to synthesizing the square matrix is a correlation matrix, well known to econometricians who apply it to study connectedness between pairs of variables in their models Correlation compacts square matrix, so that it links one- directional trade-offs between two scopes (Leopold’s view, such as E∩N), with its symmetrical opposite, N∩E etc., to acquire bi-directional or reciprocal trade-offs, which explain how different scopes overlap and work together when they impact each other in the secondary way The square matrix is so further divided into intra-sectoral impacts (primary impacts - E∩E or shorter EE, S∩S as SS, etc), which are placed on the diagonal, and inter-sectoral (or secondary impacts) impacts, which are located below or above the diagonal Diagonal elements as incommensurable can not be aggregated any more from the Table 2; so they are interpreted as they are and may be removed from further assessment Non-diagonal elements of the square matrix are then correlated The correlation is only a matrical equivalent of four sets of intersecting scopes, which could be portrayed with four set Venn diagram Technical problems avert us from presenting it here, but it can be seen in a graphically simpler version with only three sets in Radej (2008) Three-valued logic is the simplest form of multi-valued logic that is entirely sufficient for our narrow purpose – to demonstrate unsuitability of two-valued logic for IA of complex social issues In any case, two different scopes can not correlate in core issues but only on their margins where their secondary concerns intersect – later in the paper is this recalled as a main route for the integration of systems that are socially incommensurable Co-relation of these marginal and secondary impacts is exactly the function of correlation matrix Next summative step in IA so correlates remaining six non-diagonal elements: HE (overlaps E∩H and H∩E; analogously for others), SE, NE, HS, HN, and SN are again paired bidirectionally to their opposites: (i) HE∩SN forms the second order intersection or component of the sustainability; (ii) SE∩HN forms component 2, and (ii) NE∩HS forms component In the last summative step these third order overlaps would be arranged in a correlation matrix of the second order which only reproduces Table In the case that sustainability were defined as a complex of five or six scopes, for example adding cultural and spatial capital to four capitals of Ekins and Medhurst, the whole aggregation procedure would be correspondingly longer, adding new layers of correlation For reasons of simplicity, the last step of summation in the four-valued model of Ekins-Medhurst will not be provided here Practical reasons of IA as a governance tool are therefore in favour of a simple three part definition of scope of the assessment, which would give a square matrix with only three rows and columns and only one layer of correlation Appraisal of complex social phenomena obviously demands that its definition is given on all multiple levels of their evaluative consideration This is not the case in Ekins and Medhurst, 10 neither in WCED but it is accurately followed by Camagni (2007a, b) He sets a three-level definition of the territorial cohesion as an overlap structure: spatial system consists on the micro level of physical (P), socio-cultural (S) and economic (E) subsystem They intersect on the margin and establish three relational overlaps at the meso level: territorial identity (I, as overlaps S∩E and E∩S), efficiency (C; E∩F, F∩E), and quality (Q; P∩S, S∩P) Territorial cohesion is then a macro correlate of bi-directional overlaps between I, C and Q, not of onedirectional intersections between P, S and E (input-output view) Unfortunately, Camagni has not capitalised deep understanding of the complex issues in multi-scale perspective, neither translate it (jet) in the proposed methodology for territorial IA After this methodological diversion we therefore rush to the conclusions in our practical example Table summarises impacts of RDPP in the view of reciprocal relations: (i) E is very successful in achieving its primary sectoral goals (E∩E), S is successful in medium extent, while E and H are only weakly successful; this urges doubt about equal possibilities for H, and N to contribute to RDPP; (ii) H and to a certain extent N are only weakly involved in the creation of regional welfare, while S is involved in interactions that are more beneficial for tangible (E and partly N) aspects of regional wealth As a proposal, the RDPP does not promise regional convergence of scopes, because they are not treated as equally important, which seriously calls into question the sustainability of RDPP Table 3: Correlation matrix of scope accounts for the RDPP Output Input E H E H S N EE = +++; Strong positive impact HE = (0, +); Very weak correlation (intersection) , impacts unbalanced in favour of H HH = +; Weak positive impact SE = (+++ , -); E & S are not complementary; RDPP unbalanced in favour of E HS = (+, +); Weak correlation, balanced impacts between S & H SS = ++; Positive impact NE = (+, +); Weak correlation, balanced impacts between E & N S N HN = (0, +); Very weak correlation, balanced in favour of N SN = (++, +); Medium correlation, unbalanced in favour N NN = +; Weak positive impact Source: Based on Radej, 2007 Compared to the LEM (Table 1), the correlation matrix offers a more comprehensive view of RDPP’s impacts and provides dissimilar policy advice LEM fails to present systematically biased impact of this programme on non-economic scopes of regional sustainability In particular it does not address overlaps between scopes, so it is not able to bring conclusion about integrity of RDPP So LEM can not identify deep conflicts between regional four capitals Conclusion from Table has been that impact of RDPP is balanced on the four scopes and thus sustainable Conclusion from Table is that impact of RDPP is systematically biased in favour of E and materialised aspects of development and therefore its sustainability could be disputed The different conclusions of Tables and are not due to altered detailed expert multi-criteria appraisal at the micro-scale which remain the same in both cases Discrepancy arises solely from the work of evaluator in selecting the method of summation of conclusions This conclusion complies with Weaver and Rotmans (2006), and Carlsson (2000) Recognising that two-valued logic of Leopold and LEM, compared to many-valued logic in Leontief produces narrower and even misleading results in IA, the first task of the paper is accomplished 11 Social incommensurability has two dimensions and multi-scale has yet to be addressed Square matrix exists hierarchically above the micro-level because it is derived from it As a partial aggregate it exists at a lower level than the macro The square matrix therefore presents policy impacts in an intermediate, or meso view In the next chapter a distinction between micro, meso and macro evaluation is added to the summative procedure Then multi-scale and multi-scope aspects are linked into a proposal for meso-matrical IA Even though Tables 1-3 are all an evaluator technically needs to accomplish his or her task practically, a broader concept should be explained to comprehend the aggregation problem into a whole This will improve understanding of differences between results in Table and Practical illustration will be replaced with the graphical presentation of the summation procedure that brings together achievements of this paper Meso-matrical impact assessment Previous chapter demonstrated that in the face of the complex configuration of society totally different issues are equally important in policy evaluation However, this chapter will explain that also one and the same issue – say climate change – can be perceived remarkably differently, either in the local specifics or in its global entirety So far, social issues have been addressed in IA either from the micro or from the macro perspective Dopfer et al (2004) took different way of reasoning when they adopted a multi-scale view Multi-scale decomposition of social complexity is present already in Schumpeter’s Capitalism, Socialism and Democracy (1942), where he explained economic processes in a multi-scale perspective in the case of innovation (micro) which causes a creative destruction of the system (macro) with the transformation of its determinants (meso) Multi-scale approach has been also put forward by Simon and Ando in 1956 The core of their idea is that large-scale problems should not be reduced to simpler ones (as by Leopold), but studied at different system levels in order to delimit intra-sectoral (micro) from inter-sectoral processes (meso; in Rogers, 1976) In the evolutionary perspective a proper analytical structure of complex social system is made in terms of micro – meso – macro (Dopfer et al., 2004) Micro is local, specific, diversified and short sighted, linked to experimentation, innovation and change Micro refers to the detailed view Macro, on the other side, pertains to the general evolution of a system The authors explain that at the meso-level, micro-events are organized into domains which then construct a base on which later the macro-view is founded It follows that macro is then not an aggregation of micro, but rather offers a system view in terms of its domains (meso) Similarly, for Dopfer et al micro is not the reduced essence of a system; it is a ‘bottom up’ view of social realities in terms of their parts Despite depending on each other, the micro and macro are not linked directly, but are intermediated through the meso lenses The result is that when micro and macro are intermediated through meso-perspective they can not be antagonised any more Meso possesses unique hybrid (micro-macro) characteristics (Schenk, 2006) that are needed for indiscriminate evaluation of scale-specific social issues In political discourse, meso as the social field is ‘a space of positions’ (Bourdieu, 1985), which intermediates between, “difference and sameness” (Allmendinger, 2002), or paraphrasing Gell (1992), ‘non-definitive but specific’ (or micro) judgements‘ and non-specific but definitive’ (or macro) judgements Because it equally tries to tread the path between deep system oppositions (as in Table 2), meso enables mid-level articulations that effectively cross divides (Knauft, 2006) and synthesize them (as in Table 3) 12 This suggests that the evaluator needs to assess complex social phenomena from the meso perspective which facilitates understanding of the basis for the oppositions (Mertens, 1999) The neutrality of evaluator standpoint is rooted in the ‘un-excluded middle’ (term is borrowed from French historian Braudel, 1969; in Wallerstein, 2004) of the meso-matrix It is not middle in the sense of being suspended in a tension or pendulum swing between antagonised alternatives because the force field of the previous poles has itself diminished (Knauft, 2006) at meso level Just the contrary, it is meant as an exclusion of exclusion – a reversal of the Aristotelian logic of excluded middle, of the two-valued system, adopted in modern science When someone sets up the law of excluded middle, it actually means that two options are put before us to choose from, and one and only one of them is right and corresponds to the fact (McGill, 1939) Social reality is not that simple because different truths exist and each of them is multitude of meanings Polish mathematician Jan Łukasiewicz (1917) was the first to work with the three-valued logic (true, false, possible) as an alternative to the bi-valued logic But this is not to say that many-valued systems dictate the rejection of the principle of twovaluedness; on the contrary, as explained by Łukasiewicz, the acceptance or rejection of this principle is one of the determinants of the form of abstract system which will be acceptable as a logic (in Baylis, 1936) In simpler diction, that is closer to the narrow subject: impacts may be observed in an antagonist relation (Leopold) or in an incommensurable (Leontief) - the evaluator needs to choose between opposing logics or philosophies of the IA either from the standpoint of excluded middle (micro, macro) or from the un-excluded middle (meso) The evaluator’s capacity for achieving neutrality in IA critically depends on this A necessary condition is that bi-valued antagonisms are rigorously classified apart from a multi-valued claims As the assessed social issues are complex, evaluator needs to identify antagonistic issues – as blind for secondary or marginal issues that are a key to synthesis – and decline to mix and aggregate them with plural issues of many-valued logic Evaluator will strive to translate binary proposals into plural form – for example intervene into discussions to help stakeholders of the assessed public programme, who have different belief systems, understand where their disagreements have epistemological roots (Van Eeten, 1999 in Bovens et al, 2005) so they should be not antagonised but treated as plural Therefore, the evaluator attains his neutrality against political interference not by his refusal of summation as such (Luna Leopold), but by his refusal of summation of irreconcilable oppositions only between their core meanings This does not prohibit aggregation of elementary similarities nor synthesis of secondary meanings As more susceptible to deep differences and more inclusive, the summative evaluation that transgresses multiple scopes and scales has a greater potential for objectivity (Harding, 1993 in Mertens, 1999) Thus, appropriate summation in IA is an effective shield against political influence (Chelimsky, 1995) This not only justifies the need for the summative IA of large scale government interventions, but also places aggregation concerns, via neutrality considerations, into the centre of the inquiries pertaining to paradigmatic crises in evaluation So it is (high) time to recapitulate achievements of this paper (Figure 1) As laid down in the introduction, social incommensurability is fundamental concept for new approach to summative and multi-criteria evaluation of policy interventions The axis of multi-scope is located at the meso level of the multi-scale axis (as seen at the end of third chapter); this means they can be presented orthogonally as in the Figure 1a The intersection between two axes locates the meso-matrix (square matrix) as an area of the un-excluded middle (Figure 1b) from which in two summative steps synthesis of impacts is obtained (arrows I and II in Figure 1c) Results of synthesis dictate policy recommendations (arrows α and β in Figure 1c) 13 Figure 1: Meso-matrical synthesis of social incommensurability Figure 1b MESO MATRIX Multi-scope Multi-scale Figure 1a COMPLEXITY Macro (unity) MesoMatrix (Un-excluded Figure 1c SYNTHESIS Strong incommensurability - Correlation II α Synthesis Middle) I β Weak incommens – Relativisation of minor differences Micro (diversity) Source: Author There are two analogous approaches to schematising complex systems – the hierarchical theory of O’Neil (1992) and the multi-scale level concept of Geels and Kemp As Easterling and Kok (2002) explain, hierarchy theory evolved to elucidate the multi-tiered structure of certain types of production systems The theory posits that useful way to deal with problems of global change in a multi-scaled system is to understand them on three levels – level (in our case this is meso), itself a component of a higher level (Level +1; macro); level +1 dynamics are generally slower moving and greater in extent than Level 0; they form boundary conditions that serve to constrain the behaviour of Level Level may then be divided into constituent components at the next lower level (Level -1; micro in our case) Processes operating at Level -1 are generally faster moving and lesser in spatial extent than those at Level 0; they provide mechanisms that regulate Level behaviour (Easterling, Kok, 2002) On the other hand, for explaining emergence of innovation of technologies Geels and Kemp formulated the multi-scale level concept (in Rotmans, 2002; Geels, 2002, 2007) Micro (niche) level is linked to individual agents; meso level is linked to the regime such as institution, and macro level to the socio landscape The niche-level accounts for the emergence of innovations; the regime level accounts for the stability of existing systems; and the landscape level accounts for exogenous macro-developments In fundamental logic Figure 1c closely relates to both preceding concepts, even though it is derived independently from Figures 1a and 1b Detailed impact assessment is thus summarised in MIA in two phases (Figure 1c): first with partial aggregation (arrow I; Table 2), next with correlative integration from the podium of square matrix (arrow II; Table 3) Partial aggregation shifts IA from micro to meso level: this involves a loss of assessment information and causes that the overall impact on the meso level is smaller than the sum of its micro impacts It is less because its parts are constrained by being organized and subject to 'objectification' (Rittel, 1972) of these minor differences Transition to meso takes place with local aggregation of similarities that are specific, not core, so all information that are not contributing to this specifics are lost within their groups when observed on a larger scale (Bar-Yam, 2003) But this is not really a terrible loss The reward is a formation of Leontief’s view that is systemic, which means that its claims can not be diminished any further Nevertheless, input-output view is in itself insufficient to assess complex phenomena To obtain this, it is necessary to impose a correlative view of intersections as synergetic fields of the social reality As social complexes can not be integrated between core values (strong incommensurability), the only possibility to integrate them is to overlap these core values on their margin (weak incommensurability) because their 14 boundaries are permeable (Morgan, 2007) This step is synergetic, so its result will be larger that the sum of its meso parts Yet, the main purpose of correlative integration is not to maximise the sum of impacts but to evaluate impacts against the system constitutive rules which define it, at least in our case, as socially incommensurable or complex (Figure 1a) Observed difficulties in the integration of different scopes (arrow II) will call for further deantagonisation of relations between multiple scopes (arrow α in Figure 1c) In practical example (Table 3), recognition of systematically discriminate emphasis of RDPP on the economic growth led the evaluator to the conclusion that RDPP, in the proposed form, might not be sustainable Observed difficulties of partial aggregation of micro similarities (arrow I) are associated with a ‘translation loss’ that need to be reduced in future – let us recall absolute and relative limits to aggregation Arrow β then involves measures that realign thresholds for the benefit of the discriminated parts of the system – in our practical example in favour of social conditions and restrictive to economic ones; arrow β also involves measures that diminish the sacrifice of individual’s uniqueness for accessing a collective discourse The arrow β addresses also institutional conditions for democratic openness of the assessment for different opinions Therefore, measures implied by arrow β are important for the production of micro diversity while arrow α relate to measures for reconstruction of macro-uniformity (Figure 1c) They both are simply two sides of the same evolution of a complex social system Social incommensurability in policy evaluation certainly prohibits calculating impacts of the policy intervention into a compound indicator But this is not a heartbreaking event for social researchers since it enables something much more important – to evaluate policy interventions in terms of their system integrity Integration is an increasingly important policy evaluation criterion (WCED, 1987; UNCED, 1992) that is still poorly understood This can be illustrated by an example The influential Guide for ex-ante evaluation of EU structural funds impact on the sustainable development of regions (TI et al, 2003), claims that “at a microlevel, the priority is integration of different skills and competences” Similarly, Ekins and Medhurst (2003) assert that integration at the level of individual measures and projects is required for sustainable development Some other authors and documents (SEC(2005)79) likewise comprehend that sustainability implies that each individual policy measure and strategy should be integral (for a discussion on this prevailing view see Holtz, 1998) This latter point on policy integration has already been rejected, when it was demonstrated that due to specialization policies should be treated as sector-biased towards only one of many public scopes Integration should not be confused for co-ordination While it is clear that a government must govern rationally and coordinate its own one-sided interventions, it is also clear that it should not have the ambition of integrating every aspect of its impact already at the decision level, as if it could manage it through central planning These authors assume that if every person and company, and each government measure already individually integrated in their micro behaviour or design every aspect of reality, then overall regional development would be ex-ante sustainable in macro The conclusion is that for the system to function well, individuals must be well trained to serve system goals The price for realizing a micro strategy of system integration is therefore the elimination of diversity and abolishment of freedoms and unique ways of operating at the micro-level This is exactly contrary to the conclusion obtained with MIA, where integration would be obtained in correlative transgression of IA conclusions from meso to macro system level Standard IA use the terms ‘integration’ and ‘balance’ inconsistently in scale perspective, just like it misapplies the concept of incommensurability Small political relevance of evaluation recommendations may be linked to the evidence that evaluators sometimes poorly understand the concepts they apply and problems which they wish to assess, and in particular their own position in the policy evaluation 15 Conclusions The opening problem has been that standard IA has difficulties in summarising its findings in particular when underlying social values are incommensurable such as in the case of sustainable development or territorial cohesion This paper aimed to demonstrate that the meso-matrical IA, where secondary relations on the system margins play a central role, widely opens a possibility for the summary evaluation of complex social issues If evaluation of complex social issues is both, multi-criteria and summative, then it needs to be rooted where marginal areas of social worlds in strong oppositions overlap Another lesson learned is that the standard multi-criteria evaluation is insufficient if it evaluates policy impacts only on the assessment criteria This is only a preparatory phase for the evaluation which constructs the micro base that informs evaluation with quantification of causal relations between individual policy measures and individual assessment criteria (Leopold view) At this level obtained assessment results not yet enable evaluative learning, nor can they propose substantive conclusions Bare facts not speak for themselves in complexity Information obtained in the first step of the assessment needs to be first interrelated such as in correlation matrix, where it obtains meaning through a bi-directional relationship Hence, it is not information abut impact but its meaning that can be learned, agreed or disagreed and decided one or the other way Standard IA, explicitly LEM and even more Leopold deal only with facts (scientific predictions) and they finish their otherwise respectful efforts – that are a precondition for every policy evaluation – exactly where the summative evaluation in meso-matrical context really sets off Evaluators try in vain to earn their political neutrality and relevance insisting on presents standards of ‘shallow evaluation’ As given at the beginning of the paper, the meso-matrical viewpoint reverses the social incommensurability problem in policy evaluation Impossibility to create consent between deep oppositions should not be seen as a threat to the indiscriminate assessment of welfare concerns from individual and collective viewpoints The post-modern loss of a social or political common ground as the condition of unity and cohesion is not an irresolvable challenge for evaluators, with the meso perspective in mind When social incommensurability is seen in meso frame, it can be understood in the coexistence without contradiction (FloresCamacho et al, 2007), because it is grounded in multi-valued logic MIA that aims to preserve deep differences at the macro and simultaneously to enrich diversity at the micro level Growth antagonisms are therefore not a consequence of the incompatibility of incommensurable aspects of reality, but only of the exclusive or two-valued treatment of a complex social world All things considered, the evaluator can be seen as a kind of social conflict accountant and the meso-matrix is his or her tool for diagnosing the ‘dark matter’ of the universe of social oppositions and a tool for re-processing them into a potential energy for progress Mesomatrical manipulation of weak and strong oppositions with approaches that go beyond simple minded reductionism can “increase capacity for thought and increase pool of possible reasons” (Habermas in Bohman, 1995) In this way it can invoke the Foucault effect (in Burchell et al, 1991), depriving binary and antagonistic social practices of their exclusive necessity, extending the bounds of the thinkable and progressing of reason to permit the invention of others, and therefore become a real factor of change In this way, an evaluator can make use of conflicts instead of trying to ignore or eliminate them (Donzelot in Burchell et al, 1991) 16 Bibliography Åberg Y Individual Social Action and Macro Level Dynamics: A Formal Theoretical Model Sage, Acta Sociologica, 43/3(2000):193-205 Allmendinger P Towards a Post-Positivist Typology of Planning Theory Sage, Planning Theory, 1/1(2002):77– 99 Arrow K 1951 Social Choice and Individual Values, (2nd ed., 1963), New Haven: Yale University Press 124 pp Bar-Yam Y 2003a Complexity Rising: From Human Beings to Human Civilization, A Complexity Profile Cambridge, Mass.: NECSI - New England Complex Systems Institute, 19 pp., http://necsi.edu/projects/yaneer/EOLSSComplexityRising.pdf , [VII/08] Bar-Yam Y 2003b Multiscale Variety in Complex Systems Cambridge, Mass.: NECSI - New England Complex Systems Institute, Technical Report, 17 pp., http://necsi.org/projects/yaneer/multiscalevariety.pdf, [VII/08] Bar-Yam Y 2005 Complexity Rising: From Human Beings to Human Civilization, A Complexity Profile, http://necsi.edu/projects/yaneer/EOLSSComplexityRising.pdf, 19 pp, [VII/08] Bar-Yam Y A Mathematical Theory of Strong Emergence Using Multiscale Variety Wiley, Complexity, 9/6 (2004):15-24 Baylis C.A Are Some Propositions Neither True nor False? The University of Chicago Press, Philosophy of Science, 3/2(April 1936):156-166, http://www.jstor.org/stable/184343, [VII/08] Bohman J Public Reason and Cultural Pluralism: Political Liberalism and the Problem of Moral Conflict Sage, Political Theory, 23/2(1995):253-279 Bourdieu P The social space and the genesis of groups Sage, Social Science Information, 24/2(1985):195-220 Bovens M., P ’t Hart, S Kuipers 2005 The politics of policy evaluation The Oxford Handbook of Public Policy, Chapter 15, p.317-333 Burchell G., C Gordon, P Miller (eds) 1991 The Foucault Effect: Studies in governmentality Hertfordshire: Harvester Wheatsheaf, 303 pp Camagni R 2007a The rationale for territorial cohesion and the place of territorial development policies in the European Model of Society, (in A Faludi (ed.), Territorial Cohesion and the European Model of Society), Lincoln Institute of Land Policy, pp 129-144 Camagni R 2007b TEQUILA - Un Modello Interattivo di Valutazione di Impatto Territoriale delle politiche europe Torino, Seminario AISRe-IRES Piemonte, Progettare e valutare la sostenibilità ambientale del territorio, 15-16 novembre, http://www.ires.piemonte.it/aisre/CamagniTEQUILA.pdf, [VII/08] Carlsson L Non-Hierarchical Evaluation of Policy Sage, Evaluation, 6/2(2000):201–16 Chapman J An Introduction to ‘Systems’ Systemist, 23/SE(Nov 2001):81-4, http://www.firstpages.com/ukss/publications/systemist/Vol23-Nov2001SE.pdf, [VII/08] Chelimsky E Politics, Policy and Research Synthesis Sage, Evaluation, 1/1(1995):97-104 Daly H.E 1993 On Economics as a Life Science, in Valuing the Earth: Economics, Ecology, Ethics (in Daly H.E, K.N Townsend, eds.), Cambridge Mass., London: The MIT press, p 249-266 Daly H.E The Economic Growth Debate: What Some Economists Have Learned but Many Have Not? Elsevier, Journal of Environmental Economics and Management, 14/4(1987):323-336 Dixon J., J Bakkes, K Hamilton, A Kunte, E Lutz, S Pagiola, J Xie, 1998 Expanding the Measure of Wealth Indicators of Environmentally Sustainable Development Washington: World Bank - Environment Department, Environmentally Sustainable Development Studies and Monographs Series No 17, 122 pp Donzelot J 1991 The Mobilisation of Society, in Burchell et al, p 169-79 Dopfer K 2006 The Origins of Meso Economics Schumpeter's Legacy Jena: Max Planck Institute of Economics, Evolutionary Economics Group, The Papers on Economics and Evolution No 610, 44 pp https://papers.econ.mpg.de/evo/discussionpapers/2006-10.pdf [IV/07] Dopfer K., J Foster, J Potts Micro–meso–macro New York: Springer, Journal of Evolutionary Economy 14/3 (2004):263–279, [VI/07], http://papers.ssrn.com/sol3/Delivery.cfm/SSRN_ID721599_code114131.pdf? abstractid=721599&mirid=1 Easterlin, R.A 1974 Does Economic Growth Improve the Human Lot? Some Empirical Evidence, (in P.A David and M.W Reder, eds.), Nations and Households in Economic growth, New York/London: Academic Press, pp 89-125 Easterling W.E., K Kok Emergent Properties of Scale in Global Environmental Modelling – Are There Any? Integrated Assessment 3/2–3(2002):233–246, http://www.ingentaconnect.com/content/routledg/nias/2002/00000003/F0020002/art00012;jsessionid=18gjekrjrn 3oh.alice [XII/06] Ekins P 1992 A Four-Capital Model of Wealth Creation; in Real-Life Economics: Understanding Wealth Creation (in Ekins P., M Max-Neef, eds.) London, New York: Routledge, p 147-155 17 Ekins P., Medhurst J., 2003 Evaluating the Contribution of the European Structural Funds to Sustainable Development Presented at the 5th European Conference on Evaluation of Structural Funds, Budapest, June 26-27, 48 pp, http://europa.eu.int/comm/regional_policy/sources/docgener/evaluation/rado_en.htm, [XI/04] Elbers C., J.W Gunning, K de Hoop Assessing Budget Support with Statistical Impact Evaluation: a Methodological Proposal Amsterdam: Tinbergen Institute, Discussion Paper 2007-075/2, 29 pp., http://ssrn.com/abstract=1016945, [VII/08] ESPON - 3.2, Vol 2006 Spatial Scenarios and Orientations in Relation to the ESDP and Cohesion Policy Territorial Impact Assessment, Final Report, p 2-97, [IX/07] http://www.espon.eu/mmp/online/website/content/projects/260/716/file_2786/fr-3.2_April2007-full.pdf, [VII/08] European Commission 2001 Directive 2001/42/EC on the assessment of the effects of certain plans and programmes on the environment European Commission 2006 Impact Assessment Guidelines SEC(2005)79 Evans T.P., E Ostrom, C Gibson Scaling issues with social data in integrated assessment modelling Swets & Zeitlinger, Integrated Assessment, 3/2–3(2002):135–150 Flores-Camacho F., L.Gallegos-Cazares, A.Garritz, A.Garciá-Franco Incommensurability and Multiple Models: Representations of the Structure of Matter in Undergraduate Chemistry Students Springer, Science & Education 16 (2007):775–800 Foster J., Potts J 2007 A micro-meso-macro perspective on the methodology of evolutionary economics: integrating history, simulation and econometrics The University of Queensland, School of Economics, Discussion Paper No 343, 24 pp Geels F.W Feelings of Discontent and the Promise of Middle Range Theory for STS Examples from Technology Dynamics Sage, Science, Technology, & Human Values, 32/6(November 2007):627-651 Geels F.W Technological transitions as evolutionary reconfiguration processes: A multi-level perspective and a case-study Elsevier, Research Policy 31/8-9(2002):1257-1274, http://linkinghub.elsevier.com/retrieve/pii/S0048733302000628, [X/06] Gell A 1992 The anthropology of time: Cultural Constructions of Temporal Maps and Images Oxford/Providence: Berg Publishers Press, 341 pp GHK, PSI, IEEP, CE, National Evaluators 2002 The Contribution of the Structural Funds to Sustainable Development: A Synthesis Report to DG Regio, EC 2002 Volume 1-2 London, Brussels, http://ec.europa.eu/regional_policy/sources/docgener/evaluation/doc/sustainable_annexes.pdf, [III/05] Giraud P.-N 1996 L’inégalité du monde Économie du monde contemporain Paris: Folio (Gallimard) Hertin J., A Jordan, M Nilsson, B Nykvist, D Russel, J Turnpenny 2007 The practice of policy assessment in Europe: An institutional and political analysis EU/FP6 Project MATISSE Working Paper 6, 52 pp., http://www.matisse-project.net/, [VII/08] Holtz S 1998 Integrating Environmental, Social, and Economic Policies, in The Cornerstone Of Development Integrating Environmental, Social, and Economic Policies (in Schnurr J., S Holtz, eds.) Ottawa: International Development Research Centre, Chapter 11, http://www.idrc.ca/en/ev-9406-201-1DO_TOPIC.html, [V/07] Knauft B.M Anthropology in the middle Sage, Anthropological Theory, 6/4(2006):407–430 Kuhn, T.S 1970 The Structure of Scientific Revolutions Chicago: The University of Chicago Press, 210 pp., http://sciencepolicy.colorado.edu/about_us/meet_us/carl_mitcham/courses_taught/5110/classic_sts/structure _of_scientific_revolutions.pdf, [VII/08] Kunseler E 2007 Towards a new paradigm of science in scientific policy advising Environment and Health Department NUSAP.net, pp., http://www.nusap.net/downloads/KunselerEssay2007.pdf, [VII/08] Leith C., L von Thadden 2006 Monetary and fiscal policy interactions in a new Keynesian model with capital accumulation and non-Ricardian consumers Working Paper Series No 649, 42 pp, http://ssrn.com/abstract_id=908620, [VII/07] Leontief W Environmental Repercussion and the Economic Structure: An Input-Output Approach Cambridge Mass.: The MIT Press, Review of Economics and Statistics, 52/3(1970):262-271 Leopold L.B., F.E Clarke, B.B Hanshaw, and J.R Balsley 1971 A procedure for evaluating environmental impact Washington: Geological Survey Circular 645, 13 pp., http://eps.berkeley.edu/people/lunaleopold/ (118)%20A%20Procedure%20for%20Evaluating%20Environmental%20Impact.pdf, [VII/08] Li Xiaorong A cultural critique of cultural relativism Part III: Rethinking Culture: Globalization and the Challenges of Interculturality Blackwell, American Journal of Economics and Sociology, 66/1(2007):151– 171, www.encyclopedia.com/doc/1G1-163939377.html, [I/08] Lucas R.E Jr 1972 Expectations and the neutrality of money Blackwell, Journal of Economic Theory, 4/2(1972):103-24 Martinez-Alier J., G Munda, J O’Neill Weak comparability of values as a foundation for ecological economics 18 Elsevier, Ecological Economics 26(1998):277-286 Mason E 2006 Value Pluralism; entry in Stanford Encyclopedia of Philosophy (in Zalta N.E, ed.), Stanford: Centre for the Study of Language and Information, Stanford University, http://plato.stanford.edu/entries/value-pluralism/, [II/07] McGill V.J Concerning the Laws of Contradiction and Excluded Middle The University of Chicago Press: Philosophy of Science, 6/2(April 1939):196-211, http://www.jstor.org/stable/184568, [VII/08] Mertens D.M Inclusive Evaluation: Implications of Transformative Theory for Evaluation Sage, American Journal of Evaluation, 20/1(1999):1-14 Milanović B 2006 Global Income Inequality: What It Is And Why It Matters Washington: World Bank Policy Research Working Paper 3865, 36 pp., http://www.un.org/esa/desa/papers/2006/wp26_2006.pdf, [VII/08] Morgan D.L Paradigms Lost and Pragmatism Regained: Methodological Implications of Combining Qualitative and Quantitative Methods Sage, Journal of Mixed Methods Research, 1/1(2007):48-76 Mumpower J.L., T.R Stewart Expert Judgement and Expert Disagreement Taylor and Francis, Thinking and Reasoning, 2/2-3(1996):191-211 Munasinghe M 1992 Environmental Economics and Sustainable Development Paper presented at the UN Earth Summit, Rio de Janeiro, Washington: World Bank, Environment Paper No Munda G 2006 A NAIADE based approach for sustainability benchmarking Inderscience, International Journal of Environmental Technology and Management, 6/1-2(2006):65–78, http://www.inderscience.com/storage/f113210112864597.pdf, [VI/07] Munn R (ed) 1979 Environmental Impact Assessment Paris: Scientific Committee on Problems of the Environment (SCOPE), http://www.icsu-scope.org/downloadpubs/scope5/chapter04.html, [II/05] Muradian R Ecological thresholds: a survey Elsevier, Ecological Economics, 38/1(2001):7–24 O’Neil R.V 1988 Hierarchy Theory and Global Change, (in T Roswall, R.G Woodmansee and P.G Risser, eds., Scales and Global Change), SCOPE 35, Chichester: John Wiley, pp 29–45 Ostmann A 2006 The aggregate and the representation of its parts Bonn: Max Planck Institute for Research on Collective Goods, Preprints of the Max Planck Institute for Research on Collective Goods 2007/11b, 38 pp., http://ssrn.com/abstract=1024681, [IX/07] Picciotto R Towards an Economics of Evaluation Sage, Evaluation, 5/1(1999):7–22 Radej B 2007 The Four Capital Model, Matrix and Accounts Ljubljana: Economic Faculty, Research Centre, Working Papers 193, 25 pp, http://www.ef.uni-lj.si/enote/rcef/delovniZvezki.asp, [VII/07] Radej B 2008 Sinteza vplivov nacionalnega energetskega programa na prostorsko kohezijo Slovenije (Synthesis of Territorial Impact Assessment for Slovene Energy Programme) Ljubljana: Slovensko društvo evalvatorjev, Delovni zvezek 2/2008, 47 pp., http://ssrn.com/abstract=1156757, [VII/08] Radej B.2006 Neenakost v svetu Ekonomija sodobnega sveta; Forword in Slovenian translation of Pierre-Noël Giraud, ’L’inégalité du monde, Éditions Gallimard, 1996 Ljubljana: Založba /cf*, transl by K Rotar Ravetz J Towards a non-violent discourse in science, (in B Klein Goldewijk and G Frerks, eds.), New Challenges to Human Security: Empowering Alternative Discourses Wageningen Academic Publishers, James Martin Institute for Science & Civilization, University of Oxford, September 2006 Rittel H On the Planning Crisis: Systems Analysis of the 'First and Second Generations' Bedriftsokonomen Nr 8(1972):390-6 Rogers A 1976 Aggregation And Decomposition In Population Projection Laxenburg: International Institute for Applied Systems Analysis, Research Memoranda 76/11, 37 pp Rotmans J Scaling in Integrated Assessment: Problem or Challenge? Swets & Zeitlinger, Integrated Assessment, 3/2–3(2002):266–279 Rotmans J Tools for Integrated Sustainability Assessment: A two-track approach Vancouver: University of British Columbia, The Integrated Assessment Journal, 6/4(2006):35–57 Rotmans J., M.B.A van Asselt Uncertainty in integrated assessment modelling: A labyrinthic path Kluwer Academic Publishers, Integrated Assessment, 2(2001):43–55 Sankey H Taxonomic Incommensurability University Melbourne: Department of History and Philosophy of Science, International Studies in the Philosophy of Science, 12/1 (1998):7-16, http://www.philosophy.unimelb.edu.au/staff/Sankey/howard/howardpaper6.PDF, [VII/08] Sawyer R.K Artificial Societies: Multiagent Systems and the Micro-Macro Link in Sociological Theory Sage Sociological Methods Research, 31/3(2003):325-363 Schenk N.J 2006 Modelling energy systems: a methodological exploration of integrated resource management Groningen: University of Groningen, PhD Dissertation, Chapter 6, p 97-115, http://dissertations.ub.rug.nl/FILES/faculties/science/2006/n.j.schenk/06_c6.pdf, [VIII/07] Schnellenbach J The Dahrendorf hypothesis and its implications for (the theory of) economic policy-making Cambridge Journal of Economics, 29/6(2005):997-1009 19 Scriven M The Final Synthesis Sage, American Journal of Evaluation, 15/3(1994):367-382 Stagl S 2007 Emerging Methods for Sustainability Valuation and Appraisal - SDRN Rapid Research and Evidence Review A report to the Sustainable Development Research Network Final Report (January), 66 pp, http://admin.sd-research.org.uk/wp-content/uploads/2007/04/sdrnemsvareviewfinal.pdf, [VI/07] Stake R.E A Problematic Heading American Journal of Evaluation; 22/3(2001):349–354 TI, GHK, IRS 2003 The Evaluation of Socio-Economic Development - The Guide London: Tavistock Institute - TI, GHK, IRS 157 pp., www.evalsed.info/, [IX/06] UNCED 1992 Agenda 21 United Nation Conference on Environment and Development, http://www.un.org/esa/sustdev/documents/agenda21/index.htm, [IX/06] Veen A Van Der, H S Otter Scales in Space Swets & Zeitlinger, Integrated Assessment, 3/2–3(2002):160–6 Virtanen P., P Uusikylä Exploring the Missing Links between Cause and Effect A Conceptual Framework for Understanding Micro–Macro Conversions in Programme Sage, Evaluation, 10/1(2004):77–91 Wackernagel M., W Rees 1996 Our Ecological Footprint: Reducing Human Impact on the Earth Gabriola Island, The University of British Columbia: New Society Publishers Wacquant L 1997 Reading Bourdieu's «Capital» Foreword to the English-language translation of The State Nobility Cambridge: Polity Press, http://www.homme-moderne.org/societe/socio/wacquant/capital.html, [I/07] Wallerstein I 2004 The Uncertainties of Knowledge Philadelphia: Temple University Press, 224 pp WCED 1987 Our Common Future Oxford and New York: Oxford University Press, World Commission on Environment and Development – WCED, 400 pp, http://www.un-documents.net/wced-ocf.htm, [IX/06] Weaver P.M., J Rotmans 2006 Integrated Sustainability Assessment: What? Why? How? EU/FP6 Project MATISSE, Working Paper (October), 22 pp., http://www.matisseproject.net/projectcomm/uploads/tx_article/Working_Paper_1_03.pdf, [VII/08] Webster’s Third New International Dictionary 1971 Chicago: Encyclopaedia Britannica Inc Wiggins, D 1997 Incommensurability: Four Proposals, (in Chang R (ed.), Incommensurability, Incomparability, and Practical Reason), Cambridge, Mass.: Harvard University Press, http://plato.stanford.edu/entries/value-incommensurable, [VII/08] 20 ... for IA is that the summation algorithm should follow separate procedures for the aggregation of weakly incommensurable micro elements of the assessment and for the macro synthesis of conclusions... jurisdiction Further progress came from Ekins and Medhurst (2003) in their proposal for the IA of the contribution of the EU structural funds to regional sustainable development in the light of their four... interventions The axis of multi-scope is located at the meso level of the multi-scale axis (as seen at the end of third chapter); this means they can be presented orthogonally as in the Figure 1a The intersection