Home Page (MVA) > Integrated Assessment Models (IAMs) and Resources > IAMs Thematic Guide
Thematic Guide to Integrated Assessment Modeling
[HOME] [PREVIOUS] [NEXT] [BOTTOM]
Cultural Theory
Integrated assessment models attempt to account for interactions between social and natural systems. One of the more recent bold initiatives in this regard is the attempt to account for the role that cultural perspectives play in framing outcomes in the TARGETS integrated assessment model (van Asselt et al., 1995; Rotmans, 1994; Janssen and Rotmans, 1995). The TARGETS group is to be commended for their attempt to make explicit the role of cultural perspectives on model outcomes. The TARGETS group base their representations of cultural theory on the work of Schwartz and Thompson (1990) and Thompson et al. (1990), which combines anthropological and ecological insights. This work raises a number of important issues in doing integrated assessments. Among them is the issue of how a hybrid community like that of integrated assessment, that is often isolated from the disciplines that contribute to it, provides quality control over theoretical and analytical perspectives that are imported into the integrated assessment realm for the first time. This problem is especially acute, as in this case, where the field from which the theoretical perspective is imported (cultural anthropology) is relatively distant from the usual areas of expertise of integrated assessment practitioners. There is a distinct danger that ill-formed or untested insights from outside disciplines can too easily become entrenched in integrated assessment models without undergoing critical review. These formulations might then provide the archetype for future representations of that type in integrated assessment models.
Rotmans et al. (1994) note that ``Thompson et al. (1990) claim that their cultural theory is universal. They argue that distinctive sets of values, beliefs and habits (in nations, neighborhoods, tribes and races) are reducible to only a few cultural biases and preferences''. Cultural theory is one of many approaches that have been used to cope with the subjectivity inherent in analyzing long term global change. As we discuss below, claims of universality are highly controversial and difficult to defend; they also provide a primary distinction between Thompson et al. cultural theory and other approaches. Other classification schemes, for example, Ecocentrists vs. Technocentrists (O'Riordan, 1995), and Environmentalists vs. Industrialists (Lave and Dowlatabadi, 1993) make no claims of universality[FN]. In fact, the limitations of their taxonomies are self-evident. Thompson et al. cultural theory on the other hand claims to provide a universal basis for its taxanomy. It is not clear to us that it does. For that matter, it could be argued that cultural theory is not a theory at all; and, like the others, is another interpretive scheme.
We dispute the claim that the Thompson et al. cultural theory represents a realistic model of the richness, diversity, and complexity of universal cultural underpinnings. We also question whether it is appropriate to apply it to biophysical phenomena in integrated assessment models as the TARGETS group does. In other words, we are concerned about what is left out when the Thompson et al. cultural theory is explicitly put in to integrated assessments, and what is arbitrarily introduced in the process. We are also concerned that the integrated assessment community has few mechanisms in place to critique good cultural theory from bad cultural theory, and good implementations of cultural theory in integrated assessment models from bad implementations of cultural theory. These concerns take on added significance in light of the claims of universality. The range of our concerns might best be summarized under the headings robustness, incompleteness, and model bias as follows:
Robustness:The egalitarian, individualist, and hierarchist groupings in the Thompson et al. cultural theory are necessarily idealizations and generalizations, and this is generally understood by the users of Thompson et al. cultural theory. However, this raises the question of the robustness of these cultural classifications. It is unclear whether different people would make the same assignments of a set of cultural characteristics to the different groupings. The users of cultural theory in integrated assessment models make the assumption that because people have different perspectives, the role of these perspectives can therefore be elicited and characterized. It is not clear at all that this can be done, and the great difficulty of doing this contributes to a lack of robustness in the implementation of cultural theory in integrated assessment models.
The problem of robustness is even more acute when the Thompson et al. cultural theory is extended to categorize biophysical phenomena such as CO2 fertilization, soil moisture changes, and aerosol effects in integrated assessment models (van Asselt et al., 1995). The uncertainties in these sorts of parameters are very weak functions of world view, and arguably have more to do with epistemological ignorance about complex properties of the Earth system than they do with cultural characteristics. The assignment of cultural attributes to variables whose cultural associations are very weak or hard to tease out is bound to lead to arbitrary analyses.
Incompleteness (what gets left out) :The cultural theory of Thompson et al. leaves out numerous important facets of culture, though it is not our purpose to provide a complete listing here. Suffice is to say that cultural features such as kinship, esthetics, historical backgrounds and chance, and personality are difficult to capture in the Thompson et al. cultural theory framework. In the implementation of Thompson et al. cultural theory in the TARGETS model, policy options are determined by setting various model parameters according to the preferences classified by the different cultural perspectives. In other words, policies and outcomes are determined by who we are as egalitarians, hierarchists, or individuals. But just who we are changes according to the sorts of choices we make. Any static view of cultural theory fails to capture the dynamic nature of human perspectives which change over time, partly as a result of earlier decisions made according to earlier preferences.
Model Bias (what gets put in) :It is also important to bear in mind that much of what is left out when cultural theory is added to integrated assessment models may not even be amenable to the modeling form. In the words of Shackley and Wynne (1995), ``much of the cultural, political and institutional is highly variable, ever changing and uncertain, or possibly, indeterminate and unpredictable. It does not lend itself to ready quantification in an integrated modeling framework''. Furthermore, use of the model framework selectively employs from the cultural domain only those attributes which can be expressed in parameterized form in the model. IA models with different structure and design will be able to represent different aspects of culture, perhaps spanning much of what we value in this domain, though any single model will always exclude certain facets when a particular model structure is chosen.
The fact that implementation of any particular version of cultural theory in IA models lacks robustness, entails incompleteness, and introduces a model bias in representation of the theory is not reason not to engage in the exercise. However, we have to be aware that selective compromises will be made in the exercise, and the shortcomings of the theory and the biases introduced in model implementation of the theory should be explicitly articulated. As part of this latter process, it will be necessary to initiate studies of cultural theory and global change in non-modelling integrated assessment frameworks as well. Unless we pursue non-modelling IA approaches in tandem with modelling approaches, the insights from integrated assessments will be skewed by incorporating only that which is most easily represented in model frameworks
