Modelling dostal2
- 1. 1. Introduction In this contribution we are concerned with (i) complex character of geographical systems, (ii) methodological plurality in geographical research and (iii) different con- cepts of uncertainty and risk. We are considering these three themes of geographical conceptualisation of real systems, because they are closely interrelated. Complex character of the study object of geography is reflected in primary methodological problems or even limits of geographical knowledge and possibilities to assess risk processes in geographical systems. The complex character of geographical systems results from combination of two fundamental forms of complexity (Hampl, 2000): (i) structural complexity (complexes of heterogeneous phenomena) and (ii) develop- mental complexity (including societal phenomena and processes). Each of the two forms of complexity can further be considered in two specific forms. Structural complexity is characterised by (a) qualitative hybrid nature of geographi- cal systems and by their (b) relatively weak integrity that is resulting from important conditioning through exogenous factors, through wider external environment in respect to interactions between nature and society and also regarding relationships between macro-systems and micro-systems. In consequence, partial structures and processes in geographical systems are necessarily “determined” by different types of regularities that are also different in terms of variable degree of their probabilistic nature. In a simplified way, we can say that the variability in regularities ranges from functional relationships to “fully” random ones and thus to irregular situations. At the same time, our search for understanding of geographical reality makes it necessary to consider internal order of the systems, but also external conditions that are outside the systems. These complex aspects of geographical systems result in methodological plurality that obviously leads also to disintegration of geographical research. First of all, there is the relationship between research efforts in physical geography and social geography, respectively the problem of geographical dualism. In social geography, there is the current emphasis put upon the research strategy to follow contempo- raneous social sciences and accept underlying philosophical pluralism (Johnston et al. 2000, 356). On the other hand, attempts to find an integrating conception for the science of geography have led to some open proclamations or more or less hidden conceptions of geography as an idiographic science. Chapter 3: Systemic geographical approach, methodological plurality, uncertainties and risks petr dostál, martin hampl
- 2. 30 petr dostál — martin hampl The high level of developmental complexity of geographical systems results from the incorporation of the most developed (i.e. societal) phenomena and processes into the structures of geographical systems. Therefore, in analogical way as in social sci- ences, we can consider two key study problems. First, there is the activity and also the subjectivity of actors carrying societal development and there are corresponding processes of cognition and assessment. Further, there is the stochastic character and multi-level structuration of the societal system and its development. All these aspects of societal phenomena and processes provide room for speculative assessments, ideo- logical bias of knowledge and associated normative conclusions about societal real- ity. Second, there also is extraordinary dynamic and variable orientation of societal development. This circumstance complicates collection of information and data on societal development. In particular, the dynamics and variety in societal development result in changes in conditioning factors and imply shorter periods in which estab- lished regularities can be used and indicate possible causal factors in development of geographical systems. Accordingly, it is necessary to pay attention to these complicat- ing circumstances in the cognition process and timely recognise possible risks of our “objective” knowledge concerning the geographical systems. However, there is often also an excessive emphasis put upon (a) ideological approaches and (b) absence of law-like regularities in societal development. This excessive emphasis in social sci- ence on ideology and irregularity characterizes in particular current postmodern approaches and also resigns from any possibility of systematic and cumulative process of knowledge acquirement. Already this short and simplified specification of “dimensions” of the complexity of geographical systems shows the difficult character of their study. Obviously, there arises the question whether it is at all possible to indicate some methodological rec- ommendations regarding a workable conception of geographical studies. In spite of the above-mentioned problems it is justifiable to make such an attempt. In the first place, we can “revitalize” systemic approaches, i.e. general theory of systems in its original version with its various aspects of systemic ontology (Bertalanfy, 1969) that has been of an empirico-intuitive nature. We believe that systemic approaches can help us to put back together components of reality that standard research is being dismembering for specific purposes of narrow analytic approaches. Accordingly, we make a strong case for a systemic approach in which complex real systems are organised in terms of more dimensions in hierarchical orders (see also Hampl, 2000; Dostál and Hampl, 1995). The following dimensions can be considered as the most important ones: (i) dimension of rank or scale (macro-region–micro-region) (ii) dimension development (lower–higher development level in terms of qualita- tive development) (iii) dimension of structural complexity (for example distinction between societal and geo-societal systems). In a systemic approach, there appear multi-dimensional structuration of organisa- tional levels and development mechanisms as primary themes for the orientation of the cognition process. But, a systemic approach is also needed for elimination of con-
- 3. 31systemic geographical approach, methodological plurality ceptualisations concerned with seemingly contradictory principles or tendencies in development of real systems. In the field of regional development research, for exam- ple, there is continuing debate on “natural” dominance of tendencies to divergence or convergence, in spite of the obvious fact that dominance of a certain type of tendency can belong to a transient stage of development and also be dependent on particular development levels and scales of geographical systems. We can point out to examples of the current regional development in the European Union, or to the development in the global system (see also Dostál and Hampl, 2000; 2004). Another example provide studies in which economic, cultural and ecological themes are placed “side by side” without any hierarchical order or attempts to “incorporate” economy into hierarchical framework of socio-cultural system and further to “incorporate” such a system into an environmental system (see also Bennett and Chorley, 1978). However, in general, it is possible to identify a hierarchy of corresponding mechanisms: for example, market mechanisms function in a wider values system – an integrated societal system with political mechanisms and with deeply embedded socio-cultural mechanisms. The emphasis given to the ontological character of systemic or holistic approaches is in accordance with the specific subject matter of geography, but systemic approaches cannot only be considering different hierarchical levels. Of a key importance in sys- temic studies are interactions between various qualitative components and between different scale structurations. In the first place, the multi-level interactions among heterogeneous components characterise geo-societal systems in which such complex mechanisms are developed at a higher level. The studies of geo-societal systems can specially reveal mechanisms that lead to “identifications of compromises”; for exam- ple, between short-term and long-term effects (see distinctions between operative and strategic planning of major investors; Shephard, 1979, 274–276). Such an approach obviously necessitates solutions of many research questions. First, there is the ques- tion of how to distinguish in studies of geo-societal systems spontaneous mechanism from regulatory mechanisms when we know that formations of such mechanisms, respectively rules of the game, are an integral outcome of considerations of societal needs and possibilities, and also resulting from enforcements of particular inter- est articulations (see also Downs, 1957). In consequence, in studies of geo-societal systems there is insufficient knowledge of mutual and usually mediated qualitatively different mechanisms at different scales (ranks). Given these intricacies of geographi- cal research, it seems advisable to direct the research towards studies on long-term development tendencies of systems and their components that can bring insights into qualitative changes of mechanisms studied. In view of current levels of cognition of real systems in environmental and social sciences, we must suggest research efforts that are considering the above-mentioned types of mechanisms as important themes of current enquiries concerned with geographical systems.
- 4. 32 petr dostál — martin hampl 2. Methodological plurality Extraordinary complexity of societal and environmental reality can paradoxically be recognised in terms of heterogeneity and also according to simplicity of used meth- ods, or “useful” methods. This simplicity of used methods is obviously reflecting the mentioned inverse proportional relationship between the complexity of studied real- ity and the perfection of exact character of tools employed in studies of real systems. Consequently, there is considerable methodological pluralism in social sciences, but also in environmental sciences and the methodological pluralism is tending to be multiplied in geographical research. It is however not necessary to view the methodological pluralism as resulting in a disorientation of the cognition process. Undoubtedly, the difficult character of studies concerned with complex systems and also reflects the complex nature of real systems and stimulates the research searching for applications of various methods in “deciphering” the complexity. Moreover, it is possible to claim that heterogeneity of methods and approaches is useful. Yet, it is also necessary to note that there is a need for combination of various methods in the framework of the same strategic research programme. The complexity of the subject matter (i.e. studies of geographical systems) needs to be tackled by various methods, but it also necessitates establishing linkages between themes (i.e. attempting synthetic approaches) that enable more adequate interpretations of research results. Because, only corresponding research outcomes based on “differently sophisticated” meth- ods can provide convincing results. Therefore, we can suggest a parallel approach: (i) research efforts of a qualitative character (i.e. verbal and logical identification of relationships and interactions, conditions or “integral” organisation of systems), (ii) a “simple” quantitative research (interpretation of data with the help of statistical maps, simple correlations, etc.) and (iii) sophisticated quantitative research (multi- dimensional modelling, multivariate statistical analyses, GIS methods, etc.). Possible differences in research outcomes of various methods can suggest formulations of new research questions and necessitate new analyses and provide sensible themes for further research. Research concerning long-term developments of complex systems and particu- larly societal systems, has resulted in alternative approaches, different methodologi- cal conceptions and various ways of studies. In spite of the fact that debates on the research plurality are still going on, it seems that separate “alternatives” are having more or less similar scientific standing. In other words, it means that we are not concerned with “different truths”, but with considerations of different aspects of the same reality. In other words, the reality is integral, but its cognition is not unified. Unfortunately, different conceptualisations and approaches usually result in mutual exclusion than in complementary research strategies. In consequence, there seem to be more ideological disputes than synthesis-orientated research efforts. One can ask why reflections of difficulties in postulating of universal theories have had as an outcome, for example, the formulation of so-called middle range theories (Merton, 1957) and have not be resulting in conceptions of a hierarchical order of theories of different reach or scope.
- 5. 33systemic geographical approach, methodological plurality For example, there is made a principal distinction between extensive and intensive research designs (see Harré, 1979; Sayer, 1992) and that is reflecting in a sensible way the realities of current research strategies. However, we can ask why is this inter- pretation of research design not resulting in conceptualisations of methodological approaches that can enable research strategies connecting the two types of research design? Or why structuralist conceptions societal development are understood more or less as contradictory conceptualisations disproving voluntaristic conceptions, or why holistic methodology is “only” seen in contradiction with methodological individualism? Similar questions are also considered by Giddens (1984). The exiting methodological plurality can be illustrated by the above-mentioned distinction be- tween the intensive and extensive types of research design (see Table 1). It would be going too far to debate in detail implications of these two types of research design in our short paper. However, it is necessary to make a few remarks relating to research efforts orientated on cognition of geographical systems. First, the summary in Table 1 indicates that a vast majority of socio-geographical and demographical research is traditionally made according to the extensive research design with its emphasis upon abstract cognition and general applicability (see also Cloke, Philo and Sadler, 1991). Most of researchers see the two research types as a dilemma. Yet, it is not necessary. Second, there are analogous problems of extensive Table 1 Intensive and extensive research: a summary. APPROACH INTENSIVE EXTENSIVE Research questions How does a process work in a particular case or small number of cases? What produces a certain change? What did the agents actually do? What are the regularities, common patterns, distinguishing features of a population? How widely are certain characteristics or processes distributed or represented? Relations Substantial relations of connection Formal relations of similarity Types of groups studies Causal groups Taxonomic groups Type of account produced Causal explanation of the production of certain objects or events, though not necessarily representative ones descriptive “representative” generalisations, lacking in explanatory penetration Typical methods Study of individual agents in their causal contexts, interactive interviews, ethnography. Qualitative analysis. Large-scale surveys of population or representative sample, formal questionnaires, standardised interviews. Statistical analysis. Limitations Actual concrete patterns and contingent relations are unlikely to be “representative”, “average” or generalizable. Necessary relations discovered will exist where-ever their relata are present, e.g. causal powers of objects are generalisable to other contexts as they are necessary features of these objects. Although representative of a whole population, they are unlikely to be generalizable to other populations at different times and places. Problem of ecological fallacy in making references about individuals. Limited explanatory power. Appropriate tests Corroboration Replication Sources: adapted from Sayer, 1992, page 243.
- 6. 34 petr dostál — martin hampl and intensive research in the research considering complex themes in physical geog- raphy. Generally, this means that the complex subject matter of geography has been necessitating comprehensive (extensive) research orientations. Further, it seems that necessary combinations of extensive and intensive designs are realised seldom. This suggests that current level of cognition of societal and environmental themes does not allow deeper solutions of these fundamental research issues. It seems that pos- sible solutions would have to consider in studies concerning societal development a necessary transformation of the “principal” alternatives into “partial” alternatives and focus on integral co-influencing of partial structures, processes and articulations of interests. It is currently popular to stress subjectivity and individual, corporate or mass articulations of interests (Scharpf, 1997; Dostál, 2002) and choose particular ideological positions in scientific research claiming objectivity. However, we do not believe in an objective ideology and, in consequence, we believe that such a manner of “doing” social science has to be rejected. Third, we support the view that the process of cognition has to be a search for an order in the societal and environmental reality, a research effort bringing not only insight into the reality, but also estimating probable and potentially progressive (possibly desirable) changes in its development. Because if the complexity of studied reality (i.e. society in environment) makes such research efforts difficult, than ideological approaches make such a search “almost” impossible. The constitutions of ideological positions are conditioned through changing the real- ity. These considerations point out to the importance of an emphasis given in the framework of basic research avoiding deficiencies of ideologically biased research and orientations on applied research without sound theoretical conceptualisation. In brief, there is a need for a research that is avoiding replacements of scientific theory and methodology by ideologies. 3. Qualitative and quantitative modelling The preceding considerations indicate to a certain extent possibilities and necessity of a useful modelling of geographical structures and process and its importance in the process of cognition of real systems. It is obviously possible to reject the im- portance of modelling approaches. However, in general, quantitative approaches and particularly quantitative modelling provide important tools for more exact and deeper research efforts. It we are concerned with complex and less-understood systems than quantitative approaches can contribute to formulations of research questions. According to the well-known definition proposed by Chorley and Haggett (1967, 22) a model “is a simplified structuring of reality which presents supposedly significant features or relationships in a generalized form”. Due to the much demand- ing character of selecting and interconnecting fundamental features of the studied reality, it is necessary to define in clear terms phenomena and relationships under study and procedures of the cognition process. Therefore, modelling approaches can contribute to orientate studies by qualitative methods as well as by quantita- tive ones. Quantitative modelling approaches have to add to qualitative modelling
- 7. 35systemic geographical approach, methodological plurality procedures certain controlling features. In consequence, combinations of qualitative and quantitative modelling methods can contribute to the same integral process of cognition. 3.1 An example of qualitative modelling An example of qualitative modelling can be given considering structuration and development mechanisms of integral societal system (see Figure 1). The point of departure is a primary systematisation and generalisation of cognition of reality (see Hampl, 1994; 2000) providing a foundation for specification of the purpose of our investigation: types of regularities to be sought and what constitutes order in given parts of reality. This indicates basic direction of studies and subsequently also research efforts searching for regularities. It is therefore possible to derive methodo- logical approaches from the subject matter of the discipline. It appears that a primary distinction in modelling approaches would be the difference between the qualitative and quantitative methods. It is clear that primary schematisation and generalisation of qualitative models give necessary logical and theoretical cohesion to subsequently postulated quantitative models. The qualitative model of structuration and develop- ment of integral societal system, including specified interactions and development mechanisms (see also Dostál and Hampl, 1995; Hampl, 2000, 63–76) is obviously an attempt to incorporate geographical/environmental organisation into integral societal system. In other words, it is an attempt to elaborate further on internal structuration of the system. The model includes assessments of basic interactions, differentiation processes and regulatory mechanisms. It is necessary to note a typical feature of mod- elling is a simplifying approach to reality that can only emphasise substantial forms of structuration, major types of differentiations and subsequent reaction on evolving differentiations. There are three forms of structuration (organisation) in the integral societal sys- tem: (i) elementary form, corresponding with the relative homogeneity of humankind (human population) and also to the potential equality of human individuals as regards their role in society, basic interests, etc., (ii) semi-complex form corresponding to the hierarchisation (though often limited) in the distribution of wealth and power in society from the viewpoint of human individuals and their social group (internal structure of society), and (iii) complex form corresponding to the environmental/geo- graphical (external) organisation of society that is differentiated in an exceptionally asymmetric (uneven) way and hierarchically organised in respect to both the size of units and development conditions. These three forms of structures are in mutual interaction, while the central mediating function is having the semi-complex struc- turation. Depending on the degree of structural complexity (Hampl, 2000, 29–32) of these structurations the general nature of mutual interaction has a heterogenising or homogenising orientation. This conceptualisation claims that what basically matters is the creation of differences in the development of elements and partial systems and, simultaneously, an effort to suppress these differences (in other words a tendency towards a dynamic equilibrium).
- 8. 36 petr dostál — martin hampl These interactions are realised by various mechanisms that can be grouped in two categories. In the case of differentiation processes at the level of parts within the whole, competitive mechanism dominate. Market mechanism is an example, but also any form of competition, whether political, cultural or other. A similar interpreta- tion applies to the level of “spheres of competition” – among people just as among regions or entire countries. In this sense, Figure 1 conveys the notions of social and geographical competition. The second type is formed by co-operative mechanisms having the role in differentiating processes that is only secondary and indirect. Co-op- erative mechanisms are characteristically exemplified by a social as well as geographi- cal division of labour, exert in some respects suppressing influences on competitive Fig. 1 Structuration of integrated societal system and development mechanisms.
- 9. 37systemic geographical approach, methodological plurality mechanism, because competition mechanisms have positive influence in raising the efficiency and integrity of the whole system, and also in its capabilities to compete with other systems. The creation of social and territorial differences is the most fundamental result of the influences of the (dominantly competitive) mechanisms. The uneven outcomes are primarily caused through unevenness of external conditions for relatively even elements (people). It is obviously taken for granted that also the type homogeneity of elements has to be understood as relative and that there exist certain differences in individual efficiency. The “minor differences” are intensified by the impact of com- petitive mechanisms such as differences in education that are externally influencing potential of human individuals. Understandably, all of this provokes social tensions and efforts to redistribute the results of “competition” (see also Schumpeter, 1954, 284–302; Downs, 1957, 22–35). The differences that emerge are of a dual type, both as regards the extent of unevenness and the possibilities and especially needs to suppress them. The sources of social dissatisfaction are immediately linked up to the differentiation of relative levels of income, the share of the wealth and power of people, their groups or even entire nations (differences according to GDP per capita is a typical indicator). In contrast, people may not much care about differences in the spatial intensity of results or in their absolute volume in towns, regions or states (the GDP per square kilometre is a typical indicator in this case). The differences of the former type are relatively limited because their conditioning factors are primarily connected with semi-complex differentiation. Since the differences of the second type are conditioned by environmental differentiation, they attain extreme values and their suppression is very difficult; from the viewpoint of economic efficiency of the whole, this is basically always disadvantageous. However, both types of differences cannot be approached in an isolated way because forms of differentiation in intensity of settle- ment or economic productivity, on the one hand, and differentiation in the relative wealth of population, on the other, are usually similar in terms of their geographical distribution and obviously not in terms of levels of differences as we explained above. As a result, advantages of geographical position or the geographical combination of natural and social conditions often support the creation of social differences. At the same time, geographical differentiation in the relative wealth of people and general prosperity is an necessary impetus for migration of population and capital, and sub- sequently also the concentration tendencies which further intensify existing consider- able differences in the spatial intensity of settlement and economic production. These largely geographical concentration processes simultaneously tend to support increas- ing efficiency of an integrated societal system because they enhance the development of efficient forms of spatial organisation of labour (see agglomeration advantages). They also sustain the efficient exploitation of territorial potential by suitable forms of economic specialisation. The distinction between the two types of inequalities also indicates two sorts of spontaneous reactions of people and their groups on these inequalities and associ- ated processes. The view of spontaneity is here narrow and focused on “immediate“ auto-regulative processes and is not considering organised (i.e. institutionalised)
- 10. 38 petr dostál — martin hampl influences based on political interest articulations that are orientated on suppressing of inequalities. Such a wider view of spontaneity will be considered below. According the narrow view, the spontaneous processes are primarily conceptualised as accumu- lation or concentration processes, thus as relationships feeding back on increasing inequalities (see also the well-know conceptualisation of so-called backwash proc- esses in Myrdal, 1957, 27–28). There is no doubt that this orientation of processes tends to be dominant. This is documented by long-term concentration of population and economic activities in cities and the most developed regions or by the capital accumulation and monopolising tendencies in market economies. It must be noted that we can only speak of a dominant character of the tendencies and not claim an inevitable tendency of these processes toward accumulation and concentration. From a general point of view, we can stress adaptive character of various processes of a re-specialising type that tend to allow poorer units to improve their positions in the whole system. Similarly, a deeper specialisation in division of labour tends to help weaker units to improve their position through their co-operation with leading units. There are to some extend in geographical reality also deconcentration tendencies that are conditioned by ecological circumstances, such deconcentration of activities and population from large cities and industrial concentrations. The deconcentration tendencies usually imply a shift towards higher scales of concentration (see metro- politan regions). Finally, there are special processes having different impacts on the development of societal inequalities, on the one hand, and territorial inequalities, on the other. Migration is an important process of this type. Migration is a basic process leading to increasing geographical concentration of population (i.e. urbanisation), but it also involves relocation of labour force from less progressive and lower wage profes- sions to progressive and higher wage activities. From a social viewpoint, this tends to suppress inequalities. The existence of inequalities intensified by spontaneous processes that cumulatively increase social and territorial inequalities obviously contributes to serious societal tensions and often tends to social destabilisation. In consequence, there emerge social pressures on political elites to accept various regulatory mechanisms and redistribu- tive measures orientated on a mitigating of inequalities. It largely involves inequalities in social and economic standard of people and only exceptionally it involves also dif- ferences in their territorial intensity (see various attempts to regulate growth of large cities). The regulations are usually called social and regional policy, but they are in fact components of macro-economic or macro-social policies. It must be stressed that the function of such policies is in contradiction with the above-discussed spontaneous reactions. In brief, such mitigating policies have the character of negative feedback relationships. There are obviously long-lasting debates concerning these various mechanisms. At a first glance, the debates are dealing with contradictory issues, i.e. social stability and economic performance. Yet, it should be noted that an objective basis of a search for adequate (i.e. compromise-orientated) solutions there are political decisions on levels and types of regulations responding to “internal” social (political) pressures in the system concerned and “external” pressures of economic (eventually political) competition.
- 11. 39systemic geographical approach, methodological plurality The schematic qualitative model in Figure 1 is obviously an attempt to conceptual- ise the societal system of developed democratic states and their societal and territorial evolution in the long period of modernisation. In a sense, the model can be under- stood as a specification of final state of functioning of societal systems. However, we can also make an attempt to indicate possible further direction of elaborations on the model following the dimension of development and the structural dimension. Considering the former dimension there is the theme of specifications of develop- ment stages, for example inspired by theories of stages (see Hampl, 2005). It would be crucial to elaborate on changes in quality and forms of interactions (mechanisms) and on generally increasing role and variety of feedback relationships (especially regula- tory mechanisms). Considering the latter dimension, there is the themes dealing with necessary more detailed specifications of mentioned “aggregate” mechanisms indicat- ing their partial components. Such detailed specifications would contribute to our deeper understanding of the development process of society in environment. Detailed specifications are necessary in order to disaggregate the integral model in a series of submodels that can make needed attempts at quantitative modelling easier. We can mention research themes concerned with migration processes or capital flows. In the line of this discussion we may conclude that there is a need for explicit inter- faces between qualitative and quantitative approaches to modelling of environmental and societal systems and associated processes. It is especially true in respect to these systems, because their structural and development complexity is of the highest level. There is also corresponding high level of “vagueness of risk processes”. This must be understood as a result of insufficient scientific knowledge of complex real systems, but it is also immanent feature of the complex systems that is conditioning their ex- traordinary development dynamics: (i) there are alternative development trajectories and associated choices of more variants, and (ii) there is room for differentiations in interest articulations and activities of par- tial societal subjects, etc. In view of these considerations, we can argue that the qualitative modelling can largely be understood as a result of empirical generalisations and also as an initial foundation for quantitative modelling. This contributes to necessary development of inductive and deductive approaches. 4. Conceptualisations of uncertainty and risk in environmental and (geo)societal systems The debate concerning possibilities and limits in studies of complex systems is only a partial, but autonomous section of scientific cognition in a narrow sense. The dy- namic development of the complex systems and associated uncertainties and risks are understandably an important problematic of societal practice and political and economic decision-making. Due to limited scientific knowledge and scientific founda- tions for solutions of current societal and environmental problems the approaches and
- 12. 40 petr dostál — martin hampl procedures are basically orientated on pragmatic ways and tackling specific narrow issues. The decision-making processes are of key importance and reflect the diversi- fied interest articulation of actors, but also their subjective reflections of “objective” problems. In consequence, the debate is shifted towards the behaviour of key actors (states, transnational companies, etc.). But also their behaviour and decision-making is an important study object of scientific research. Accordingly, we consider these issues in this last section of the chapter. Research considering uncertainties and risks in the development of geographical systems is obviously a very complex matter. Research on global environment indicates that many environmental problems are uncertain. Also, possible effects of various proposed solutions appear to be beset by uncertainties (DeSombre, 2005, 53–70). There is general tendency to confuse the notions of uncertainty and risk. Risk relates to the probability of the occurrence of an undesirable effect. The reason for confus- ing risk and uncertainty stems from the circumstance that there are uncertainties about probabilities for political, economic or environmental events. The likelihood of these sorts of events is uncertain and, consequently, risk specifications are impos- sible. Research specifications of uncertainty and risk are thus separate but interre- lated problems of scientific efforts. Examinations of risks demonstrate that people (whether politicians, entrepreneurs or general public) do not evaluate risk in the same way that risk assessors do. The role of science and scientific research within this field is very complicated, but also essential. Uncertainty is usually referred to as incom- plete information or disagreement between information sources. Policy-makers wish to know what other actors will do in order to be able to decide what decisions they have to make. Considering risks in functioning of complex geographical systems it is clear that political uncertainty is intersecting with scientific uncertainty. It appears that in decisions relating to regulatory processes in geographical systems research- ers are confronted with data which are weak in terms of accuracy and precision. For example, relevant actors require finding out how much they appear to contribute to environmental problems and tend to monitor their behaviour afterwards. There is obviously the precautionary principle suggesting that uncertainty should not prevent regulatory decisions to mitigate environmental problems. This principle suggests not using any new technologies until proven safe for environment and society and has become part of international environmental law and also in environmental law in the European Union Treaty (DeSombre, 2005, 57). Accordingly, we may draw the conclu- sion that the estimation of risk and associated decision-making processes about how best prioritize mitigations of risk appears essential in specifications how to respond to environmental issues locally, regionally, nationally or globally. It is clear that politi- cal perceptions and assessments of risk situations by key decision-makers contribute considerably to regulatory decisions to mitigate environmental problems. It is also apparent that the multi-level character of mechanisms in geographical systems sig- nificantly constraints possibilities to specify risks in exact ways. Burgman (2005, 1) defines risk as “the chance, within a time-frame, of an adverse event with specific consequences”. He gives a definition that is assuming possibilities of calculations of occurrence of adverse events. He is also viewing the notion according
- 13. 41systemic geographical approach, methodological plurality to two dimensions of probability. It is usually understood as the statistical frequency (or relative frequency) with which a certain event is expected to take place. But, it can also be viewed as the degree of belief warranted by evidence. This conception is also concerned with situations in which a probability of an event is unknown or unknow- able. This relates to the idea of “subjective probability” (Burgman, 2005, 7). It has the meaning of a lack of knowledge about a process or it specifies personal degrees of belief of various actors. It is obvious that the concept of subjective probability seems to be most applicable when assessments of risks in complex geographical systems are considered. There is also language problem of words clustering around the notion of probability and risk. The spectrum is wide and ranges from such words as change, belief or tendency to possibility or plausibility to more exact ones as confidence, like- lihood and risk. Language permits borderline case and it results in vagueness. Also due to this linguistic variation, there is a tendency to assess risk in inconsistent ways. It is obvious that risk assessments depend on relevant settings. These considerations show that risk assessment is inherently also a subjective affair which is significantly influenced by societal context. The conclusion can be drawn that “(r)isk assessments are invariably subject to distorting influences, perhaps more so than other types of scientific analysis, because the public setting of many of the problems” Burgman, 2005, 25). We can draw another conclusion saying that risks are largely social constructs and there is no objective method indicating that one risk valuation is better or worse than another. However, it also must be added that better and more complete informa- tion about risk situation is important in giving decision-makers necessary insight to participate more effectively in the political process (DeSombre, 2005, 67). The general point to take from these considerations is that the notions of risk and uncertainty are not simply objective numbers or concepts that can be specified in unequivocal terms by more scientific research. Some authors have argued earlier that risks cannot be defined objectively because what is considered as risk is depending upon social values held by groups and individuals (Douglas and Wildavsky, 1982). Societal priorities for what issues to value and what questions to ask influence how policy-makers think about risks under conditions of uncertainties of the development of geographical systems. The level of uncertainty about risk thus remains high. It becomes clear in our exposition of the geographical system sketched in Figure 1 that geographical research does not provide exact results which enable to fully compre- hend how different subsystems of geographical systems function in their own and how they intersect with political or economic subsystems where major risks assess- ments are to be made. It is therefore worthwhile to consider the concepts of uncertainty and risk also in the view of basic mechanisms that we distinguished in the postulated integral qualita- tive model in Figure 1. Of course, the concept of risk is inseparable from the notions of uncertainty and probability. But, risk is not the same as danger or hazard (Giddens, 2002, 20–35). Because risk refers to hazards which are assessed in terms of future probabilities and thus actively assessed. Assessments of risk are important in societal systems that are actively future-oriented. In brief, risk presumes a societal system that is trying to break away from its current state and this characterises the development
- 14. 42 petr dostál — martin hampl orientation of modern industrial and post-industrial societies. “Redistribution can however also refer to the sharing of risk. It is here that the successes of the welfare state have been most marked” (Giddens, 1994, 149). This short quote indicates the importance of institutionalised regulatory redistributive mechanisms indicated in Figure 1 and also suggests their causal connections with socially and spatially unequal consequences of the market driven mechanisms in the advanced integral societal system. Because in the market economy which is based on a diversified structure of economic property forms in which private sector performs the pivotal role of scarce resources allocation, there is risk which is energizing wealth creation in the subsys- tem of economic mechanisms. Advanced economic system is future-orientated, it is unthinkable without risk and based on calculations of future profit and loss, and thus upon risk assessment. The redistributive mechanisms of welfare state are designed to protect against hazards of social inequality and specifically against job loss, disa- blement, old age, etc. It is also useful to note that insurance against risks is actually redistributing risk in a large set of units (insured subjects). In the advance post-in- dustrial society and economy risk has assumed particular significance. In words of Giddens, “(o)ur very attempts to control future tend to rebound us, forcing us to look for different ways of relating to uncertainty” (2002, 26). He makes a crucial distinction between types of risk. First, there is external risk that is coming from the “fixities” of nature and tradition. Second, there is so-called manufactured risk created by impacts of advancing knowledge upon the entire world system including natural environment (Giddens, 1994, 152–153). In other words, the concept of manufactured risk refers to risk situations with which the mankind has little experience of confronting such as greenhouse effect or collapsing energy networks. It also applies to world-wide capital markets (Bernstein, 1996). These various risks are influenced by intensive globalisa- tion processes. Manufactured risks confront modern societies with new forms of risks that significantly differ from risks that existed in the past. Old risk situations had well- know causes and expectable effects. Manufactured risks appear to be incalculable in terms of relevant factors and causal mechanisms and indeterminate in their societal and territorial impacts. 5. Concluding remarks Considering these new risks the conclusion can be drawn that only wide ranges of competences of territorial governments can provide in the advance society sufficient decision-making capacity which is particularly important given the increasing un- certainty of future developments and spillover effects confronting local and regional government and also national governments in the globalisation era (see further Dostál, 2002). Because manufactured risks are usually diffuse in origin, it is not clear how they could be addressed, and who is responsible for taking decisions to rem- edy them. These risks are often about what the societal developments have done to natural environment. There are almost no aspects of natural environment that are not affected by the societal development. The increasing importance of the notion of
- 15. 43systemic geographical approach, methodological plurality risk has been closely linked up with possibilities of calculation. This is illustrated by advanced forms of insurance, such as against car accidents, fluctuating capital rents or river floods. Real manufactured risks are even more risky, because their levels are not known until it is often too late. However, active taking of risks is a key element of a dynamic economic development and in an integrated geo-societal system necessar- ily orientated on production of innovations and experimentations. In consequence, in the globalisation era this means to live with a diversity of risk situations that neces- sitate continuous adaptations and innovations of institutionalised and spontaneous regulative mechanisms. Such regulatory mechanisms are needed, because they can contribute among other things to risk management (see also Beck, 1992). What post- modern social scientists usually call chaos lacking regularities, Beck conceptualises in terms of risks and uncertainties. Management of risks attempting to control them has become important characteristic of advanced geo-societal systems at different geographical scales. Systematic research concerning development of geo-societal systems at different scales can hopefully contribute to our better understanding and assessment of these uncertain phenomena. However, we must also draw a disturbing conclusion with Bennett and Chorley saying that “we reach the bounds of our inves- tigation of environmental systems from which we must withdraw to more tractable considerations… If environmental systems are as complex and counter-intuitive as has been argued by some commentators, or if the impeding environmental collapse is as imminent as has been argued by others, then the environmental imperative requires changes in social values which are needed now... The true extent of environmental complexity and imperative will determine the degrees of freedom in answering our epistemological questions, but it is unlikely that there will be answer to environmen- tal problems” (1978, 551). References BECK, U. (1992): Risk Society: Towards a New Modernity. London: Sage. BENNETT, R.J., CHORLEY, R.J. (1978): Environmental Systems. Philosophy, analysis and control. London: Methuen. BERNSTEIN, P. L. (1996): Against the Gods: The Remarkable Story of Risk. New York: Wiley. BERTALANFY, L. von (1969): General Systems Theory. New York: Braziller. BURGMAN, M. (2005): Risks and Decisions for Conservation and Environmental Management. Cam- bridge: Cambridge University Press. CHORLEY, R.J., HAGGETT (1967): Models, Paradigms and the New Geography. In R.J. Chorley and P. Haggett (eds.) Models in Geography. London: Methuen, 19–41. CLOKE, P., PHILO, C., SADLER, D. (1991): Approaching Human Geography. An Introduction to Contem- porary Theoretical Debates. London: Paul Chapman Publishing. DESOMBRE, E. (2005): Global Environment and World Politics. London: Continuum. DOSTÁL, P. (2002): Territorial government and flexibility: a critical assessment. In: Belgian Journal of Geography, 3, 227–241. DOSTÁL, P., HAMPL, M. (1995): Geographical organisation and societal development: searching for an integral order. In: Acta Universitatis Carolinae – Geographica, 30, 21–42. DOSTÁL, P., HAMPL, M. (2000): Globalisation: processes of integration or multi-polarization? In: Acta Universitatis Carolinae – Geographica, 35, 5–20.
- 16. 44 petr dostál — martin hampl DOSTÁL, P., HAMPL, M. (2004): Geography of post-communist transformation and general cycle of regional development: experiences of the Czech Republic in a global context. In: European Spatial Research and Policy, 11, no. 1, 7–29. DOUGLAS, M., WILDAVSKY, A. (1982): Risk and Culture: an Essay on the Selection of Technical and Environmental Dangers. Berkeley: University of California Press. DOWNS, A. (1957): An Economic Theory of Democracy. New York: Harper Publishers. GIDDENS, A. (1984): The Constitution of Society. Cambridge: Polity Press. GIDDENS, A. (1994): Beyond Left and Right. The Future of Radical Politics. Cambridge: Politiy Press. GIDDENS, A. (2002): Runaway World. How Globalisation is Reshaping Our Lives. London: Profile Books. HAMPL, M. (1994): Environment, society and geographical organisation. In: GeoJournal, 32, 3, 191–198. HAMPL, M. (2000): Reality, Society and Geographical/Environmental Organization: Searching for an Integral Order. Prague: Chales University, Faculty of Science. HAMPL, M. (2005): Geografická organizace společnosti v České republice: transformační procesy a jejich kontext. Praha: Univerzita Karlova, Přírodovšdecká fakulta. HARRÉ, R. (1979): Social Being. Oxford: Blackwell. JOHNSTON, R.J. et al. (2000): The Dictionary of Human Geography. Fourth edition. Oxford: Blackwell. MYRDAL, G. (1957): Economic Theory and Under-Developed Regions. London: Duckworth. SAYER, A. (1992): Method in Social Science. London: Routledge. SCHARPF, F.W. (1997): Games Real Actors Play. Actor-Centred Institutionalism in Policy Research. Boul- der: Westview Press. SCHUMPETER, J. (1954): Capitalism, Socialism and Democracy. Fourth edition. London: Unwin Univer- sity Books. SHEPHERD, W.G. (1979): The Economics of Industrial Organization. Englewoods Cliffs: Prentice-Hall.