Portal de Periódicos da CAPES

Event Ecology, Causal Historical Analysis, and Human–Environment Research

2009; American Association of Geographers; Volume: 99; Issue: 3 Linguagem: Inglês

10.1080/00045600902931827

1467-8306

Bradley B. Walters, Andrew P. Vayda,

Geographies of human-animal interactions

Abstract Research on human–environment interactions is especially challenging given its interdisciplinary character and its need to address complexly interacting causes in time and space. Event ecology has been suggested and illustrated as an approach that can effectively address these challenges. Yet, previous writings on event ecology offer only a limited rationale for the approach. This article attempts to address this shortcoming through a more explicit examination of the underlying logic and practice of event ecology. Event ecology is based on a pragmatic view of research methods and explanation, articulated by such scholars as Peirce, Lewis, and Chamberlin, that has recently resurfaced in scholarly debates. This view places at the center of research inquiry the answering of "why" questions about specific environmental changes of interest, instead of evaluating causal theories, models, or factors that are thought in advance to influence such changes. Explaining environmental change this way involves constructing causal histories of interrelated social and biophysical events through a process of eliminative inference and reasoning from effects to causes, called abduction. Precise questions, concrete event descriptions, and counterfactual analysis are central to this. In practical terms, researchers should strive to be skeptical about what constitutes evidence, yet open-minded and adaptable to unexpected findings, and be willing to employ whatever sound methods and theoretical ideas are best suited to answer the question at hand. Examples from field research experience on people–forest interactions in the Philippines and Saint Lucia are included to illustrate these features of the event ecology approach. La investigación sobre las interacciones hombre–medio ambiente es todo un desafío especial, dado su carácter interdisciplinario y la necesidad de que se consideren causalidades interactivas de gran complejidad, en el tiempo y el espacio. La ecología de eventos (event ecology) ha sido sugerida e ilustrada como un enfoque que pueda atender efectivamente estos retos. Con todo, escritos anteriores sobre este tipo de enfoque ecológico ofrecen apenas un planteamiento limitado del mismo. Este artículo intenta abocar esta limitación a través de un examen más explícito de la lógica y práctica subyacentes de la ecología de eventos. Esta rama de la ecología se basa en una visión pragmática de los métodos de investigación y explicación, articulados por autores como Peirce, Lewis y Chamberlin, visión que ha resurgido en recientes debates académicos. El enfoque ubica como centro de la indagación investigativa la búsqueda de respuesta a preguntas sobre el "qué" acerca de específicos cambios ambientales de interés, en vez de evaluar teorías causales, modelos, o factores que con anticipación se supone influyen tales cambios. El explicar cambios ambientales de esta manera implica construir historias causales de acontecimientos sociales y biofísicos interrelacionados a través de un proceso de inferencia eliminadora y razonamiento de efectos a causas llamado abducción. Preguntas precisas, descripciones concretas de los acontecimientos y análisis de alternativas, son centrales para esto. En la práctica, los investigadores deben preocuparse de ser escépticos acerca de lo que se toma como evidencia, pero también abiertos y adaptables al descubrimiento inesperado; y estar dispuestos a emplear el método sano o ideas teóricas que mejor sirvan para obtener la respuesta a la pregunta en cuestión. Para ilustrar todos estos rasgos del enfoque de la ecología de eventos, se incluyen ejemplos de las experiencias de campo para el estudio de las interacciones gente–bosque en las Filipinas y Santa Lucía. Key Words: abductioncultural ecologyexplanationpolitical ecologyresearch methodology关键词: 外展文化生态学解释政治生态学研究方法Palabras clave: abducciónecología culturalexplicaciónecología políticametodología de investigación Notes 1. We follow many philosophers here in making a conceptual distinction between the tools and techniques of research (methods) and the logic and justifications determining how the tools and techniques are deployed and how research results are interpreted (methodology). 2. We note this recent interest as "renewed" in acknowledgment of the work of such early pioneers as anthropologists Franz Boas and A. L. Kroeber and geographer Carl Sauer, who argued many decades ago for the importance of understanding human–environment interactions in particular historical contexts (CitationLivingstone 1992). 3. Our prescription for usually moving outward in space reflects our own research experience, but it might in some cases be appropriate to move inward in space to gain causal understanding as, for example, when medical researchers move to the genetic level to understand the epidemiology of particular disease outbreaks (see CitationVayda 2008 for a more detailed discussion of moving both outward and inward in space). 4. In the natural and some social sciences, theory-driven research has been typically based on the application of the nomothetic or "covering law" model. We argue that many social scientists, although they might not strictly follow this model to its more positivist extremes, nonetheless are like those who openly subscribe to it in being less intent on showing actual causal histories than in finding that observed cases accord in certain selected respects with their theories or a priori generalizations. 5. Examples of such theory-driven approaches as applied to human–environment research can be found in the work of the human behavioral ecologists or Darwinian ecological anthropologists (e.g., CitationCronk, Chagnon, and Irons 2000; E. A. CitationSmith and Winterhalder 1992; for a critique of this approach, see CitationVayda 1995a, Citation1995b) and in such variants of "political ecology" as those privileging, for the purpose of explanation, any finding of correspondence in particular cases between such generalized factors as capitalist accumulation or inequality and environmental degradation (for critiques of this approach, see CitationMcGuire 1997; CitationVayda and Walters 1999; CitationWalters 2008; also, cf. C. CitationSmith 1984). See also CitationPerz (2007) on the recent, problematic emergence of "grand theories" in land use and land cover studies. 6. For critiques of theory-driven research and the covering law model, see the following books and articles and the references therein: Scriven (Citation1966, Citation1975); CitationMartin (1981); CitationCaldwell (1982); CitationElster (1989); CitationRichards (1992); CitationLittle (1993); Vayda (Citation1995a, Citation1995b, Citation1996); CitationRoth (1999); CitationMorad (2004). 7. This is not to suggest that these subinvestigations would not also yield results of theoretical interest in their own right. For example, Walters's assessments of regeneration in cut and uncut mangrove forests led to general insights about the influence of forest canopy structure on species recruitment and forest regeneration (CitationWalters 2000, Citation2005a). The point is that the pursuit of such theoretical insights was not the priority here. 8. See Freedman (Citation1991, Citation1997), CitationMcKim (1997), and S. CitationTurner (1997) for criticism of regression models as employed by sociologists. It is, of course, possible that much LULCC research is directed toward something other than the kind of causal explanation we advocate as a goal of research. This possibility is suggested, inter alia, by the frequent references in the LULCC literature to "drivers" rather than "causes." 9. It might also be the case that researchers are especially interested in why certain changes have not occurred. For example, it might be interesting to know why deforestation is not taking place in a particular location. This is a valid question but one that relies, nonetheless, on considering and explaining counterfactual cases of deforestation to provide a meaningful answer. Without understanding the causes of deforestation in comparable situations, it is meaningless to posit explanations for the lack of deforestation in a particular place. To illustrate, it makes no sense to claim that the absence of commercial timber market penetration explains why deforestation has not occurred in a particular place unless we can show that the penetration of such a market has caused deforestation in a comparable situation. 10. In a similar vein, CitationWells (1994–1995) deplores the fact that projects on interrelationships of biodiversity conservation and sustainable economic development have proceeded without adequately defining their subject matter or being able to measure changes in it in anything but the crudest way (cited in CitationVayda 1997, 14). 11. As Elster (Citation1989, 3) argues, "explaining events is logically prior to explaining facts." 12. This is in keeping with Hawthorn's (Citation1991, 27) observation that pursuing "possibilities suggested in an explanation can lead one to revise the initial description of what is to be explained and, thus, the explanatory question itself." See also Anscombe (Citation1969, 155) and Emmet (Citation1985, 30) on "explanatory force" or "causal relevance" as a basis for redescription. CitationMcIntyre (2004), a philosopher, has argued that flexibility in redescribing phenomena in the pursuit of explanations is the foundation for the success of the natural sciences and should be adopted by the social sciences. 13. Our concern here is with causal explanation rather than with the nature of causality, a matter that, we think, is best left to metaphysicians. That causal explanation can proceed without our being explicit or definitive about the nature of causality is a point that various philosophers have themselves made—among them John Stuart Mill. Agreeing with them, we are taking a naive, everyday, or "commonsense" concept of causation, as discussed by such philosophers as CitationScriven (1966), CitationAnscombe (1979), CitationHart and Honore (1985), CitationLewis (1986), and CitationNorton (2003), and showing that it can be used advantageously to deal with or resolve issues that might be of interest to human ecologists and other social scientists (see CitationVayda 2008 on this). 14. Criteria that should be followed when constructing counterfactual scenarios include the clear specification of events and their causes; logical, theoretical, and statistical consistency; and the minimal rewriting of history (CitationGriffin 1993; CitationTetlock and Belkin 1996b). Although it is beyond the scope of this article to elaborate these criteria, interested readers are referred to the growing literature on the use of counterfactual reasoning in history, psychology, and the social sciences (e.g., CitationHawthorn 1991; CitationBiersteker 1993; CitationMcGill and Klein 1993; CitationRoese and Olson 1995; CitationFearon 1996; CitationTetlock and Belkin 1996a, Citation1996b; CitationFerguson 1997; see also the earlier discussions in CitationMill [1881] 1936 and CitationWeber 1949, 165ff.). Contrary to CitationKuznar and Long (2008), we do not regard our use of counterfactual reasoning in the analysis of particular cases as committing us to one or another counterfactual theory of causation in general (see CitationCollins, Hall, and Paul 2004 on such theories and the section on "Causes Without Metaphysics" in CitationVayda 2008). 15. Counterfactual "thought" experiments use essentially the same logic as is used in empirical (i.e., scientific) experiments, although the latter rely on actual physical controls as benchmarks from which the influence of events can be assessed. 16. Csordas (Citation2004, 475–76) echoes this point: "the idea of evidence for the validity of a construct (say the habitus) or for the existence of a process (say globalization) may not make sense, since neither the habitus nor globalization are ontic entities but in fact ways of organizing data. So the evidence would be simply that the data can be organized under such rubrics or in a way consistent with them." See also Castree's (2005) epistemological critique of the "neoliberalization" of nature. 17. This is not to suggest that mangrove planting was a purely individualistic activity. To the country, individual decisions to plant and relative success in planting were influenced by the sharing of knowledge and imitation of actions between kin, neighbors, and so on (CitationVayda, Walters, and Setyawati 2004; CitationWalters 2004). 18. Illustrations of this methodologically superfluous exercise can be found in CitationYoung, Berkhout et al. (2006), C. Smith (Citation1984; as cited in CitationVayda, McCay, and Eghenter 1991, 323), and in various works in the University of Michigan Press series, Linking Levels of Analysis, edited by Emilio Moran. See also CitationMarston, Jones, and Woodward (2005) on flat ontology.