World Development Vol. 36, No. 10, pp. 19381955, 2008 2008 Elsevier Ltd.

. All rights reserved 0305-750X/$ - see front matter www.elsevier.com/locate/worlddev

doi:10.1016/j.worlddev.2008.02.008

The Knowledge that Counts: Institutional Identities, Policy Science, and the Conict Over Fire Management in the Gran Sabana, Venezuela
BJRN SLETTO * University of Texas at Austin, TX, USA
Summary. The cultures of environmental planning agencies shape institutional identities and management interventions. Central to such institutional cultures is knowledge production, which is shaped by politicaleconomic processes, dominant narratives and institutional desires to produce conservation landscapes. Through knowledge production, certain scientic knowledge and data are appropriated, while others are excluded. In the case of re management in the Gran Sabana, Venezuela, a project of policy science draws on selected scientic knowledge and emphasizes remote sensing and quantitative analysis at the expense of indigenous knowledge and prescriptive burning practices. This policy science emerges from an institutional culture that favors re suppression as a means to recreate a desired, imaginary forest. 2008 Elsevier Ltd. All rights reserved. Key words Latin America, Venezuela, conservation management, protected areas, knowledge production, indigenous people

1. INTRODUCTION In the Global South, protected area management is intimately linked with national development strategies and often implicated in conict between indigenous peoples, state agencies, and commercial interests. Because of these links between politics, development and the environment, indigenous people and landscapes in the South have become objects of hegemonic environmentalist discourse and action (see e.g., Adger et al., 2001; Crush, 1995; Dalby, 1998; Dryzek, 1997; Escobar, 1995; Ferguson, 1995; Gupta, 1998; Mitchell, 1995). These environmentalist discourses are shaped by hegemonic narratives of landscapes, people, and Nature, which in turn are reproduced through the privileging of some knowledge over others in the development project (see e.g., Agrawal, 1995; Moore, 1996; Nygren, 1999). Here I propose to explore some of the ways in which global environmental discourses articulate with local systems of knowledge production, which in turn shape planning practices that tend to marginalize indigenous knowledge, sometimes at the expense of conservation objec* This article is based on dissertation research conducted in JulyAugust 2000, JulyAugust 2001, FebruaryApril 2002, OctoberDecember 2002, and April 200304 with a Fulbright-Hays dissertation award; a Peace Studies Fellowship from the MacArthur Foundation (administered by the Peace Studies Program, Cornell University); an NSF Geography doctoral dissertation research award (No. 0221324); and funding from the Mario Einaudi Center for International Studies and the Department of City and Regional Planning, Cornell University. Logistical assistance and research permits were provided by Instituto Nacional de Parques (INPARQUES), Electri n del Caron (EDELCA), Corporacio n Venezolcacio ana de Guayana (CVG), the Department of Anthropology at the Instituto Venezolano de Investicas (IVIC); and the community of Kugaciones Cient marakapay, the Gran Sabana. This article was written while a 2005/06 Mellon Graduate Fellow at the Cornell Society for the Humanities. The author wishes to thank Dr. Barbara Lynch, Dr. John Forester, Dr. Stanford Zent, and four anonymous reviewers. All omissions and errors are those of the author alone. Final revision accepted: February 22, 2008. 1938

THE KNOWLEDGE THAT COUNTS

1939

tives. In particular, I seek to examine how development and conservation agencies appropriate and shape scientic and local knowledge in ways that are sometimes contradictory, inconsistent, and ironical, and which leads to everyday, situated practices that are less rational that what is presented in agency representations such as annual reports, Web sites, and so on. 1 I argue that production and representation of knowledge is closely associated with institutional identities, which are shaped by global discourses of environmental degradation and risk, but also, or perhaps even more so, emerge from homespun, institutional narratives and imaginaries of local landscapes and peoples. Ultimately, certain knowledges are privileged in planning institutions because they t the rules of what counts as true knowledge production (Foucault, 1972, 1980). I suggest that the formation of such institutional identities and privileging of certain knowledge are in part shaped by desire: desires to produce certain landscape formations, to create formalized systems of environmental management, and to know the truth. I draw here on Lacanian psychoanalysis (1978, 1995) to suggest that desires are important determinants in institutional processes of place-making, that is, the social and material production of space occurring through the fantasies of planning institutions and the material resources (economic power, exclusive right to violence, and equipment) of state agencies. Hence on a broader level, these desires to know are intimately implicated with relations of domination and resistance, since knowledge is never divorced from the workings of power. Ultimately, the issue here is the ways in which certain knowledges and modes of research lead to practices that, more often than not, derive from institutional narratives and desires rather than the most rational models of development and conservation management (see Gunder, 2003, 2004; Hillier, 2003; Hillier & Gunder, 2003 for other Lacanian interpretations of planning practice). In the following pages, I draw on these critical perspectives on institutional identities, knowledge production and planning practice to examine a conict between state and indigenous re management in the Gran Sabana, a grasslandforest mosaic which extends for 18,000 km2 through southeastern Venezuela and into Brazil and Guyana. I suggest that processes of identity formation and knowledge production have reproduced an unremitting

n conict between the state agency Electricacio (EDELCA) and the indigenous Pedel Caron mon, a Carib Indian group that numbers about 20,000. State re management is characterized by misunderstandings and deprecation of Pemon re management, resulting in practices that favor intensive surveillance techniques, re suppression strategies, and attempts to discourage indigenous prescriptive burning which may, ironically, serve to prevent the occurrence of extensive and destructive res. I begin by positioning my analysis in a broader context of social theory, drawing in part on the Foucaultian concepts of surveillance and will to truth (in the sense that truth is the outcome of contests to dene the rules of proper knowledge production), but modifying this constructivist approach with Lacanian 2 insights on desire and subject formation (Lacan, 1978, 1995). Next I review the links between development, conservation and knowledge production in the Gran Sabana, before I proceed with an examination of EDELCAs policy science. I suggest that EDELCAs approach to data collection and analysis is shaped in part by global discourse formations, in part by politicaleconomic realities, and in part by a dominant body of Venezuelan literature on re ecologybut also by desires and institutional identities that derive from local constructions and narratives. Then I discuss the prescriptive burning practices of the Pemon, drawing parallels with participatory management approaches based on indigenous knowledge in Australia and South Africa. This allows me to explicate some of the ways in which EDELCAs policy science could be reoriented toward more participatory and eective re management. 2. THEORETICAL FRAMEWORK: DESIRE, IDENTITY, AND KNOWLEDGE PRODUCTION To think critically about the politics of environmental planning, I view domination as facilitated through mechanisms that Foucault calls small-scale, regional, dispersed Panopticisms, or modes of surveillance (Foucault, 1980). The hegemonic knowledge production eected through these small-scale modes of surveillancesuch as the local data produced by EDELCAis shaped by institutional identities and narratives of place and people, submerges or appropriates local knowledges, and makes of real places (whether they be

1940

WORLD DEVELOPMENT

tropical forests or city parks) metaphorical terrains of domination and resistance. Local knowledge is thus a mode of learning that is often disempowered in development and conservation projects, simply because such subjugated knowledge loses the battle over the rules according to which the true and the false are separated and specic eects of power attached to the true (Foucault, 1980, p. 132; see also Gutting, 1989). In the case of the Gran Sabana, the unwritten rules that determine what is true about the natural environment and human ecologies emerge from the privileged status of certain scientic representations and data within EDELCA agency culture (see Agrawal, 1995; Mignolo, 2000; Nader, 1996; Purcell, 1998; Sillitoe, 1998 for more on local and scientic knowledge systems). Ultimately, then, at issue here is not what data, what modes of data collection, and what forms of data analysis are more valid from an objective, environmental management perspective. Instead, the question is: Why are certain forms of knowledge production believed to be and represented as more valid within a certain environmental, historical, politicaleconomic and institutional context? I turn here to the psychoanalytic theory inspired by Lacan, where knowledge is understood as discourse constructions that are linked with subject structure and thus serve as foundational elements in processes of identity formation (Alcorn, 1994, p. 35; Gunder, 2004, p. 300). Dierent knowledges draw on iconic master signiers that symbolize what is common-sensically accepted as true and desirable (such as democracy, development, and sustainability), which means knowledges are intimately entangled with subjective understandings of what is good or bad (Fink, 1995, p. 130, 1999; Gunder, 2004, p. 305). I suggest that in the case of the Gran Sabana, forest has assumed the role as a master signier in EDELCAs institutional narratives. Since a master signier secures a whole eld and, by embodying it, eectuates its identity (Zizek, 2002, p. 88), the idea of forest conservation has become a common-sensical, desirable quality. Ultimately, the desire for a forested, scientically controlled landscape shapes EDELCAs project of knowledge production and unwittingly serves to reproduce institutional narratives of the Pemon as destructive and irrational. Since desire is implicated in both wanting to be and also wanting to know (Bracher, 1993, p. 19, 1994), the identities of

EDELCA agents are closely associated with the desire to recreate a mythical forest of the past, which in turn makes it seem sensible and acceptable to use institutionally accepted technologies to collect information. This desire to know not only inuences institutional identities, it also leads to the production of imaginary, social boundaries (Foucault, 1977; Major-Poetzl, 1983; see also Philo, 2000; Pile, 1996). EDELCAs desire for certain data, collected through certain means, unavoidably leads to social processes of boundary-making, which construct the Pemon as subjects that are not rational, not behaving appropriately, not modern, and rhetorically place them outside the socially constructed space of rational environmental planning and decision-making. From a Foucaultian perspective on power/knowledge, the Pemon are symbolically excluded from the conservation space that is the Gran Sabana because their subjugated knowledge has lost the battle for truth and thus has less validity than that of EDELCA. Ironically, however, EDELCA must confront the limits of power and its contradictory consequences: occasional attempts to engage indigenous people in participatory approaches to re management have met with opposition within the agency, and have not been consistent or very long-lived. In part because of the pervasive, institutional narratives of forest destruction, EDELCA has maintained its emphasis on re suppression and environmental education, which in turn has fomented resistance and failed to reduce indigenous re use. Such indigenous resistance to EDELCAs re management is possible because EDELCAs desire for knowledge is built on an unstable, Lacanian discourse of policy science that can be resisted by the Pemon and contradicted within the agency itself (see Alcorn, 1994, p. 27, 30; Fink, 1995, p. 138), and also because EDELCAs surveillance (intended in part to produce the agencys knowledge) has its limits both in space and in time (see Robinson, 2000; Sawicki, 1989, 1994). 3 3. ELECTRICITY, CONSERVATION, AND KNOWLEDGE PRODUCTION IN THE GRAN SABANA The strategic and economic interests of the Venezuelan state, which sees in its southern frontier a vast, empty space with huge

THE KNOWLEDGE THAT COUNTS

1941

development potential, have long inuenced the character of scientic research in the Gran Sabana (see e.g., Coronil, 1997; Miranda et al., 1998; Mujica, 1984). In 1960, the parastatal n Venezolana de Guayana (CVG) Corporacio was formed to spearhead the development of the Guayana, the entire southern half of Venezuela, including the Gran Sabana, and initiated an extensive research program to map the geography, vegetation, fauna, and ethnology of this frontier region. In 1969, the regional develop n para el Desarrollo del ment agency Comisio Sur (CODESUR) was charged with mapping, conducting forest inventories, and promoting development of the Guayana (Mujica, 1984, pp. 37, 81). In 1994, CODESUR was replaced by PRODESSUR, which follows a development plan that calls for a 15% population increase through a network of frontier settlements in unoccupied indigenous territory; accelerated natural resource exploitation, including intensied mining, oil exploration, agriculture, and forestry; and extensive infrastructure development (Miranda et al., 1998). EDELCA was established in 1963 to develop , the the hydroelectric potentials of the Caron largest river downstream from the Gran Sabana. Today, EDELCA produces 75% of the countrys hydroelectric power through a series of dams and power stations, including Maca basin, comgua I (the rst dam in the Caron pleted in 1959), Macagua II, Guri, and rez, 1999, p. Carucahi (EDELCA, 2000; Pe 104). Electricity has thus long represented a country-building and development tool par excellence, both in material and in symbolic terms. The inauguration of Macagua I, which took place only one year after the fall of the last rez Jime nez, was promoted dictator, Marcos Pe as a public spectacle and hailed as the birth of rez, 1999, p. 17; see the new Venezuela (Pe also Coronil, 1997). Electricity has reduced Venezuelas dependence on hydrocarbons (in 2000, the power produced by EDELCAs hydroelectric dams represented a value equal rez, 1999, p. 104)), to 11% of oil exports (Pe and has become a crucial factor in regional integration in northern South America (Guil n, 2002). le Because of the economic importance of the Guri and Macagua dams, environmental pro watershed is tection of the 95,000 km2 Caron crucial for the states project of capital accumu lation. Hydroelectric energy from the Caron is the foundation for the development of the great industrial complex in Guayana, said

the then-director of the National Park Service Rafael Garc a in 1982. (INPARQUES), Jose Partly because of the signicance of Guri as the countrys principal source of hydroenergy and also because of its geopolitically strategic location, near the borders of Guyana and and Gran Sabana were Brazil, the Caron incorporated in Canaima National Park in n Interinstitucional, 1962 (Figure 1) (Comisio 1989, p. 1). In the early 1980s, EDELCA assumed responsibility for re management in watershed, specically to the Upper Caron protect gallery forests in order to avoid erosion a, and sedimentation of Guri (Figure 2) (Garc 1994). These important links between economic development and environmental conservation inform scientic research in the Gran Sabana. This body of literature, in turn, has been selectively appropriated in EDELCAs institutional culture to support narratives of indigenous destructiveness and forest loss. I am referring here to writers like Dezzeo (1994), Dezzeo and n (2005), Ferna ndez (1984, 1987), Fo Chaco lster ndez (1987), Huber (1995, (1986, 1995), Herna chap. 1) and Worbes (1999), who have made seminal contributions to our understandings of re ecology in the Gran Sabana based on extensive and rigorous eld research, but whose work has been used to support a re management model that overlooks situated practices. Specifically, these authors suggested that the Gran Sabana was once more extensively forested and that anthropogenic burning is a primary explanatory factor for the current grasslandforest mosaic. Although these suggestions are contested and expressed with contingencies, they have been transformed into sweeping truths in EDELCA narratives and now form the basis for the agencys policy science. ndez writes that it is For instance, Ferna obvious that the Gran Sabana is experiencing a process of forest degradation because one can see res appearing every year, and also because of the historical evidence of extensive res ndez, 1987). Palyin 1939/40 and 1926 (Ferna nologist Martin Worbes argues that (forest patches) constitute partial remnants of mature forests and severely disturbed forests; both forms of vegetation appear surrounded for the most part by savanna. The presence of carbonized (burned) trees in the area around these forests attests to the existence in the past of a more extensive forest cover (Worbes, 1999, pp. 101 102). Fo lster and Dezzeo argue that it is quite probable that the extensive savannas also

1942

WORLD DEVELOPMENT

Figure 1. Ecological regions and national parks in Venezuela. Cartography: Bjorn Sletto. Source: USGS Global GIS; Instituto Nacional de Parques. Date: June 18, 2005.

resulted from a similar process (of burning). In that case, the small forest islands distributed within these savannas could be considered relicts of a forest cover that originally was more closed (Fo lster & Dezzeo, 1994, p. 147). And n (2005, p. most recently, Dezzeo and Chaco 344) suggested that the Gran Sabana resulted from a conversion of large forested areas . . . associated with forest res without intentional conversion by the sparse human population. Indeed, indigenous elders and other longtime residents agree (and see Kingsbury, 1999, 2001): in areas of increasing populated density, primary forest cover has declined during the last two decades, becoming replaced by secondary forest and to some extent savanna. But data are lacking to support the sweeping contention that forests are disappearing in the Gran Sabana, or to verify the extent of forest degradation caused by the penetration of savanna res into forest patches. Forest cover

has not decreased everywhere in the Gran Sabanain fact, forests appear to be slowly expanding in some areas (Sletto, 2006)and revisionist analyses indicate that the great res of decades past, frequently used as examples of Pemon irrationality, may in fact have been accidentally started by European explorers guez, 2004). Also, indigenous landscape (Rodr conversion needs to be placed in a historical context: penetration of missionaries into Pemon lands, Venezuelan state development projects, and the coming of modernity with the construction of the Pan-American highway and incentives to settle in permanent communities, led to unsustainable population concentrations and thus inevitable overuse of limited resources (see e.g., Butt-Colson, 1985; Comi n Interinstitucional, 1989, p. 22; Kingsbury, sio 1999, 2001). In addition, another body of research suggests that the Gran Sabana has never been

THE KNOWLEDGE THAT COUNTS

1943

river basin. Inset: Canaima National Park. Cartography: Figure 2. EDELCAs operations area in the Upper Caron Bjorn Sletto. Source: Instituto Nacional de Paeques. Date: June 18, 2005.

completely forested. In areas dominated by grasses and herbaceous vegetation, the soils derive from Precambrian sedimentary rocks and are highly acidic, poor in minerals and nutrients, and sandy with low water retention capacity (Huber & Zent, 1995, p. 39). In upland hills with severely eroded ultisols, only grasses or deep-rooted shrubs can hold out against the droughts and occasional heavy rains (Huber, 1995, pp. 1011, chap. 1; Huber & Febres, 2000; Rull, 1992, p. 138; Urbina & Dieter Heinen, 1982). A mosaic of forest patches, grasslands, and shrub lands has dominated the

Gran Sabana since the last Ice Age (Eden, 1974; Rull, 1992; Schubert, 1995, 1986; Van der Hammen, 1974), contradicting the argument that deforestation is exclusively a consequence of anthropogenic burning (Eden, 1974; see also Barse, 1990; Cruxent, 1971, 1972). Palynological research suggests that oristic composition has varied with climate change and not with anthropogenic activity, and that during most of the Holocene the Gran Sabana was characterized by climatic conditions that encouraged savanna development (Rinaldi & Schubert, 1991).

1944

WORLD DEVELOPMENT

Although this research reveals a great deal of uncertainty about the role of anthropogenic burning in the Gran Sabana, this is not reected in prevailing institutional representations. Instead, re managers and seasonal re ghters alike maintain that the forest-grasslands mosaic is unnatural and primarily the result of anthropogenic activity. In the words of Boanerges Ramos Milan, chief of EDELCAs environmental conservation: The Gran Sabana was a forest before. There are many areas in the Gran Sabana right now, where one can see there was forest, that there had been continuous forests (Ramos, 2003, pers. int.). And among re ghters at the local level in San Ignacio, the idea of forest n, loss is repeated as truth. Ambrosio Pinzo a veteran re ghter in his 50s, said once: In 2030 years, this (the Gran Sabana) will be like a desert. And in the words of the younger re ghter Nestor Ayuso: This might become a desert in the future. We wont be calling it the Gran Sabana, but the Gran (Great) Desert. Ultimately, the issue is not whether it is true that the Gran Sabana once was a forest, nor how likely it is that the Gran Sabana will become a desert. Instead, I suggest that selected representations in a specic body of scientic literature have been incorporated into a narrative of indigenous destructiveness and forest loss, in part to simplify a complex situation and to satisfy the desires for order, for protection, and for forest. These narratives, in turn, provide the weight of scientic rationality to EDELCAs policy science and approaches to re management, and in so doing, foreclose alternative modes of knowledge production. 4. STATE FIRE MANAGEMENT, TECHNOLOGY, AND PRIVILEGED WAYS OF KNOWING From a post-structuralist perspective on natural resource management, EDELCAs re management project is situated in a danger discourse of re (see Dalby, 1999) intimately associated with two tropes that shape environmentalism in the Global South: deforestation and desertication. Since re is framed as a driving force of both deforestation and desertication, re management becomes shaped by the common-sensical imperatives of protecting forest and avoiding savan-

nization. This has led to the privileging of management models that emphasize restrictive interventions to limit re use by indigenous peoples, that overlook structural reasons for misuses of re, and that emphasize the primacy of spatial and quantitative technologies (Adger et al., 2001; Harwell, 2000; Pyne, 1991, 1995, 1996). The advent of remote sensing, in particular, has fundamentally inuenced research on re ecologies and allowed for representations that overwhelm with their technological wizardry (Pyne, 1991, 1995, 1996). Dominant models of re control now combine remote sensing imagery with quantitative analysis in ways that may devalue the ecological benets of diverse, situated practices (Bassett & Bi Zueli, 2000; Fairhead & Leach, 1996, 1998; Leach & Fairhead, 2000a, 2000b, 2002; Turner, 2003; Turner & Taylor, 2003). Although the services provided by institutions such as NASAs Fire Information for Resource Management System and the Global Fire Monitoring Centre (GFMC) will facilitate comprehensive monitoring of res, the pervasiveness of such imagery might have the unintended consequence of devaluing the importance of locallevel analyses of the complex causes and consequences of re use. 4 In the case of EDELCA, this merging of new technology with hegemonic narratives of re and risk has led to a privileging of quantitative approaches to re management. Remote sensing analysis is increasingly seen as complementary to such quantitative methods, especially as a means to persuade indigenous people of the need for more rational re management. In the words of Fransisco Zerpa, the chief of the environmental information section, if we had remote sensing images for the past twenty years, they (the Pemon) would see how much forest has been lost. Thus remote sensing is easily incorporated into EDELCAs pervasive, institutional narrative of the need for more science. By more science re managers typically mean more quantitative data, rather than more local knowledge of indigenous re management:
(The Pemon) need to see the scientic data (to be convinced they need to stop burning). This is for sure. We all need more statistics to see what the problem really is. We want to see numbers showing how much forest has been lost. We want to have data to conduct a spatial analysis (of remote sensing images). This is the direction we need to go. We ought to have more exact science (Ramos, 2003, pers. int.).

THE KNOWLEDGE THAT COUNTS

1945

The agency has in the recent years expanded its use of Geographic Information Systems (GIS) with the goal of producing exactly this type of quantitative spatial analysis. Although the agencys GIS system is still primarily used as a data collection, management, and display tool, agency staers envision that this powerful technology will take center stage in a more sophisticated modeling approaches to re management:
(The problem of burning will be resolved by) designing and implementing an information system that serves to characterize the occurrence and severity of the res in the watershed. The system will allow for temporal and spatial statistical evaluation of re events and will constitute a support tool for decision-making at the PCIV (Zerpa, Acevedo, Ablan, & Garc a, 2003, p. 2).

a mythical forest. Because of this, the agency may foreclose possibilities for alternative knowledge production based on indigenous knowledge and ultimately hamper future, more participatory environmental management approaches.

5. EDELCA POLICY SCIENCE, DESIRE, AND INSTITUTIONAL IDENTITIES The heart of EDELCAs operations in the Gran Sabana is the re ghting station in the village of San Ignacio, which is staed by two Venezuelan re managers and 25 re ghters working in the dry season, between January and May. The agencys re ghting operations rely on three watch towers to spot res (see Figure 2), a helicopter to ferry re ghters to the site of res, a small eet of 4 4 Toyotas, and radios and modern re ghting equipment. The agency also pursues an environmental education program to dissuade the Pemon from burning, but its failure to successfully engage with Pemon elders in participatory approaches have led to tense and conictual relations between EDELCA and the Pemon. Despite the 20-year presence of the re suppression program, the Pemon burn as before (albeit a little less frequently in some areas than elsewhere because of resettlement, social change, and the inuences of EDELCA) and re use remains integral to their reproduction of indigenous identity (Sletto, 2006) (Figure 3).

In the following pages, I analyze the data collection strategies and methods that provide the quantitative basis for the agencys GIS analysis. The question here is not whether scientic or local knowledge is better; nor do I seek to downplay the potentially important contributions of GIS and remote sensing for more eective re management in the Gran Sabana. Instead, at issue here is the role of agency culture in shaping knowledge production. Specically, I suggest that EDELCA collects, aggregates, categorizes, and ultimately represents data of re occurrences in ways that reproduce and bolster agency narratives of the destructiveness of indigenous burning, ultimately providing rhetorical support for the desire to protect the Gran Sabana and recreate

1600 1400 1200 No. of fires 1000 800 600 400 200 0

Detected and Combated Fires, Gran Sabana, Venezuela

Detected Fires Combated Fires

00

01

02

92

93

94

95

96

97

98

99

20

20

20

19

19

19

19

19

19

19

Year

Figure 3. Fires detected and combated from 1992 to 2003 in EDELCAs operations area in the Upper Caroni river basin. Source: EDELCA, Puerto Ordaz.

19

20

03

1946

WORLD DEVELOPMENT

EDELCAs and the Pemons re management system dier fundamentally in one, key respect: while the Pemon conduct controlled burning to reduce fuel loads in grasslands, EDELCA focuses on extinguishing grasslands res, especially those considered threatening nto forest patches in priority areas (Ferna dez, 1984, 1987). When a watch tower reports a threatening re, re managers in San Ignacio dispatch re ghters via helicopter to the site of the re. Whether a given re needed to be combated, the re managers in San Ignacio record every helicopter response to re observations as a combated re in the agencys GIS. 5 At the end of the year, the data on res observed and combated are aggregated in a central database and presented in the form of maps and tables. Because of the large number

of small, anthropogenic res set over the course of the entire re season (the great majority of res in the Gran Sabana are anthropogenic in origin and not the result of natural causes such as lightning), these maps represent the Gran Sabana as a landscape under siege from misguided, human activity (Figure 4). However, here I will focus less on questions of representation. Instead, I seek to illuminate how EDELCAs approach to data collection and categorization tends to frame 6 re events in ways that tacitly correspond with institutional culture and desires. A principal rhetorical strategy in this framing process is categorization: res combated are always classied by cause (causa) (Figure 5). These causes are illuminating both in what story they tell and in what information they

n de incendios detectados por sector prioritario. Temporada 2003. From Informe Final del Figure 4. Distribucio n, Temporada 2003. Developed by Yajaira Maita, Edgar E. Rivas, and Programa Control de Incendios de Vegetacio Rafael A. Salas, July 2003, EDELCA, Puerto Ordaz.

THE KNOWLEDGE THAT COUNTS

1947

EDELCA Fire Program Data Table

Classification of Combated Fires Number combated Sub-watershed Time of departure Latitude Year Intensity Time arriving at fire Longitude Fire number Cause Time beginning combat Type of fire Detection number Severity Time finished combat State of fire Date Area total burned Time return to station Priority sector Time receiving call Total time spent

Classification of Detected Fires Number combated Sub-watershed Year State of fire Detection number Date Time of notification Priority sector

Figure 5. Classication system used to categorize res combated and detected in EDELCAs operations area in the upper Caroni river basin. Source: EDELCA, Puerto Ordaz.

obscure. To begin, most res are categorized from a distance by re ghters manning the watch towers and only rarely based on direct observations. Also, the choice of the term cause instead of reason or purpose is rhetorically important. Cause carries a sense of passivity, as if res are due to random and destructive forces. This rhetorically places Pemon burning within a realm of Nature and n) or unreason. A term such as reason (razo sito) for the re, however, purpose (propo would have carried a connotation of Western rationality. It would have implied that res are set purposefully for a reason and that this could potentially be a good reason. In EDELCAs geographic information system, there are seven causes given for res: agriculture, animal husbandry, forest, hunting, ria, tradition, and intentional (agricultura, pecua a, tradicio n, and intencional). forestal, cacer Thus two categories are chosen to represent agriculture-related causesagriculture and animal husbandryand this despite the fact that there are no farms in the Venezuelan sense of the word and very little cattle raising in the Gran Sabana. However, these categories support decades of indigenist policies intended to covert indigenous people to agriculturalists and eect their resettlement in smaller, more easily controlled areas (Galletti, 1980, p. 7; Mujica, 1984). The category hunting, meanwhile, accounts for a large portion of the total res set by the Pemon, and also a signicant proportion of res combated each year. This is also a cause of re that most raises the

ire of re managers and EDELCA scientists, who often commented in interviews conducted in 200204 that hunting is unnecessary in an age when Pemon could move to modern communities and become responsible citizens. By representing hunting as a passive cause, the category is bounded within the realm of the irrational, thus reproducing the boundarymaking that puts the Pemon out-of-place in the Gran Sabana as people who reject the inevitable coming of modern development. The two remaining categoriestradition and intentionalare the most rhetorically powerful signiers because of the ways in which they articulate with narratives of indigeneity. In EDELCAs institutional narratives, Pemon re use is typically referred to as the result of indigenous tradition: In the words of Francisco Zerpa:
According to the statistics from the re control program, the majority of res are due to indigenous tradition. Especially the making of gardens, when they go hunting and shing, they burn to dene their trails and to inform communities where they are. Im not sure about this but (I think) they also burn as a religious act for feast days. . .Im not sure. Also because of the presence of bugs, mosquitoes, one way of getting rid of them is by re. And to clean trails. Those are the principal causes of re in the Gran Sabana (Zerpa, 2003, my emphasis).

In EDELCA narratives, Pemon tradition is not essentialized as a set of primordial beliefs and practices that originated in the Gran Sabana, but instead constructed as something foreign and invented. Although Pemon culture in

1948

WORLD DEVELOPMENT

the Gran Sabana most likely originated before the arrival of Europeans (Cousins, 1991; Mansutti, 1981; Thomas, 1982; Urbina, 1979), agency ocials typically represent the Pemon as recent immigrants who are pursuing landuse practices that originated on the forested slopes of Guyana and which are unsustainable in the sparse savanna environment in the Gran Sabana. This construction allows for the framing of indigenous burning as pre-modern, but also unbecoming the Gran Sabana. What remains is the category intentional, which has the rhetorical consequence of further delegitimizing Pemon burning practices. It was dicult to obtain a clear denition of this category, since of course every re is set intentionally: there is an obvious contradiction between the passivity inherent in cause and the agency of intentional. This category is generally used for res set for reasons that from the perspective of EDELCA cannot be assigned to tradition. Intentional res thus include res set for no other reason than for the sake of burning (quemar por quemar) and res set against EDELCA (quemar contra EDELCA). Both practices are frustrating to re ghters and personnel in Puerto Ordaz, and also to elder indigenous re managers (Sletto, 2006). Partly because of the uncertainty associated with this category, few res are dened as intentional. Ultimately, I suggest that the categories used in the re reporting system serve to bolster the emphasis on culture instead of environment in institutional narratives about Pemon re use. When asked in a survey what they thought were the reasons for the Pemons use of re, 70% of re managers, policy scientists and administration ocials in Puerto Ordaz suggested that Pemon burn for cultural reasons. Meanwhile, only 23.5% of re ghters in San Ignacio listed culture as the reason why Pemon burn. These numbers reect subtle dierences in the attitudes toward indigenous re use. Survey respondents in Puerto Ordaz, located far from everyday re management in the Gran Sabana, may be more likely to support institutional narratives of the Pemon as destructive and recalcitrant. However, these dierences do not matter signicantly in what the survey tells us about institutional culture: not a single respondent in Puerto Ordaz or in San Ignacio suggested that the Pemon burn as a means of managing the environment in the Gran Sabana. Since EDELCA narratives foreclose any possibility for the existence of a rational, indigenous re regime, the agency is also at risk of

eliminating ecologically benecial, indigenous burning practices from its project of knowledge production. 6. INDIGENOUS FIRE MANAGEMENT AND ALTERNATIVE KNOWLEDGE PRODUCTION Pemon burn the savanna for a host of reasons: to communicate (e.g., using smoke signals to announce their arrival in a village), to clean trails (using re to remove tall grasses and snakes, bugs, and scorpions), to hunt (burning grasses to encourage deer to leave their forest hideouts), and to remove re-prone grasses from the outskirts of villages. The Pemon see the mosaic of grasslands and forest patches as the natural state of the Gran Sabana: the grasslands provide opportunities for movement and a sort of canvas to imprint their culture through burning; the forests furnish a source of subsistence through hunting, shing, and gardening. Unlike state agents, Pemon do not imagine forest as desirable per se, nor do they seek to produce a forested landscape. Instead, they wish to protect the existing forest patches for reasons of subsistence. A cause of particular tension between Pemon and EDELCA is indigenous practices of prescriptive burning in grasslandforest boundaries. Pemon follow principles of patch burning; that is, they light frequent, slow-burning surface res, which result in a mosaic of small (between 20 and 50 m2) grass patches in dierent stages of regrowth. When the re reaches a relatively green and damp grass patch, it will stop from lack of fuel. The Pemon say that re (Apok) knows when to stop burn rez, 2002). The key to indigenous preing (Pe ventative re management is thus to prevent annual grasses from accumulating for several generations and reach the stage of auruta, that is, tall and yellowing grasses that burn with high heat and might escape control (see Biddulph & Kellman, 1998 for the potential benets of Pemon prescriptive burning). Such preventative burning is particularly important in savanna edges; that is, what the Pemon call the tureta kata, the last few meters of the grasslands before the beginning of the forestgrasslands ecotone. Although the tureta kata is not considered a separate spatial category in EDELCAs re management system, it is nevertheless a functional landscape unit. The dense vegetation in these grasslands

THE KNOWLEDGE THAT COUNTS

1949

communities may either facilitate or prevent re entry into forest patches: high fuel levels may allow slow-burning surface res to reach dangerous speeds and high temperatures; low fuel levels will cause surface res to stop. The Pemon therefore burn savanna edges in the late rainy season, between August and December, when the forest litter and woody vegetation is too damp to catch re. The principle behind this burning so that res dont enter forests (apok womu namai tureta tak) is thus to reduce the level of combustible grasses near forests and to form a re break (apok wako nin). We burn part by part (yanupu tupata kene) . . . so that the re stops when it reaches an area that has been burned. If we dont burn, the grass would build up and the re might enter into the forest and then damage nature (Romero, 2002). 7 Because of the documented benets of such controlled burning to reduce fuel loads in savanna environments, some natural resource agencies have incorporated indigenous burning strategies into state re management, particularly in Australia and South Africa. The complex pattern of early dry season burning of microhabitats practiced by aboriginal Australian now forms the cornerstone of current re management policies in the Northern Territory (Cooke, 2000, p. 104; Williams, Gill, & Moore, 2000, p 94; Williams et al., 1998), with EuroAustralian landscape managers insisting that re is an essential feature of all savanna environments (Whitehead et al., 2003, p. 415; see also Haynes, 1978, 1985; Lewis, 1986, 1989). Because of local variation in moisture and topography, the result of such early res is a patchy, ne-grained mosaic of burned and unburned grasses (Whitehead et al., 2003, p. 416), which serves to protect certain vegetation patches and to enrich biodiversity (Braithwaite, 1996; Laris, 2002; Mistry, 2005; see also Laris (2002, 2004) on similar patch-mosaic re regimes in Mali). Landscapes that have been prevented from burning suer from invasions of woody species or exotics (Russell-Smith et al., 2000, p. 95) and are more at risk from catastrophic res, such as those that ravaged the Northern Territory in the 1960s following the imposition of bans on burning. In South Africa, landscape managers have also developed approaches to re management similar to the Australian model: patch burning to reduce fuel loads, create mosaic patterns, and facilitate landscape heterogeneity. This is a means to reproduce indigenous patterns of savanna burning, which probably harkens back

thousands of years (Edwards, 1984; Hall, 1984, p. 43; see also Hall, 1984, p. 43 on the !Kung; Hall, 1984, p. 45 on the San and Kull, 2002a, 2002b, 2002c on the Malagasy). In South Africa, land managers maintain that regular burning is not merely acceptable, it is essential to maintain a healthy ecosystem. The accumulation of organic material hampers growth and decreases plant vigor in grasslands ecosystems (Everson, Everson, & Tainton, 1988), while regular patch burning promotes landscape heterogeneity by increasing patch sizes, size variability, and shape complexity and proximity (Hudak, Fairbanks, & Brockett, 2004). This means that the issue in South Africa is not whether res should or should not be allowed. Rather, the challenge is to develop a system of preventative burning that is appropriate given 21st century ecological, cultural, and politicaleconomic realities (Bond & Archibald, 2003). Ultimately, the question is, why does EDELCAs policy science and re management practices divert to such a degree from the models developed in similar savanna environments elsewhere? Part of the explanation lies in the ways in which institutional cultures reect development imperatives, and in how institutional cultures are supported by desires and narratives of people and place. This is to say, EDELCA is not the only re management agency in the Global South that privileges re suppression: while South African agencies are incorporating controlled burning strategies, forest services in Francophone African countries such as Mali and Madagascar tend to represent smallholder re use as a cause of desertication (Kull, 2002b, 2002c; Laris, 2002, 2004). Also, as in the case of other state institutions, re management agencies are not monolithic entities but social spaces characterized by contested priorities and narratives. In the case of EDELCA, some ocials do indeed see environmental benets in Pemon patch burning, but they constitute a small minority in the agency. mez, EDELCAs project direcEduardo Go tor and manager of the re program from 1990 to 1996, once proposed that EDELCA develop a prescribed burn program in collaboration with Pemon, but this was not well received, he explained in an interview. The argument was that it would not be eective, but this is not the case at all, according to Go mez. I think this would be an eective solution. We could co-manage quite well with the

1950

WORLD DEVELOPMENT

Pemon, because they already use this technique mez per(of prescribed burning). From Go spective, it does indeed make sense to reduce fuel loads in order to protect forest patches:
Certainly, when there is a large fuel load, it needs to be reduced. The Pemon have a management vision, and this vision is prescribed burning: to reduce the fuel load so that when a re of great proportion arrives, there is not enough combustible materials. If they didnt burn the Gran Sabana, the res would mez, 2003, pers. int.). be spectacular (Go

mez, then, Pemon knowledge could To Go form the foundation for an alternative project of knowledge production, leading to re management that is more participatory and from his point of view, more eective. His position is supported by practice and theory developed by re ecologists and landscape managers elsewhere; in fact, participatory re management based on principles of controlled burning constitutes a growing body of scientic literature. The question is: given his relatively high position in the agency and his status as a former re manager in the Gran Sabana, why doesnt his perspective carry more weight? Part of the explanation lies in the selective appropriation and representation by EDELCA policy scientists, administrators, and re managers of a certain body of literature on re ecologies in the Gran Sabana. 7. DISCUSSION I have suggested that agency narratives are shaped by a desire for a landscape that is rationally managed, orderly, and free from the supposedly irrational burning practices of the Pemon. These narratives draw on a body of literature that represents the Gran Sabana as a vulnerable landscape threatened by anthropogenic burning, both through its framing of the re problem and through its selection and presentation of data. Although a great deal of uncertainty remains about the role of anthropogenic burning in the Gran Sabana, this landscape has been produced as a lost forest through what Latour (1986) calls a process of constructing order, whereby policy scientists render alternative interpretations of scientic data less plausible. The disorderly understandings of the role of anthropogenic re have been simplied to construct an orderly, winning explanation for the current state of the Gran Sabana: the grasslands have been created by indigenous burning and continued burning

will result in savannization (Latour, 1986, pp. 3339). In a Lacanian sense, forest and desert have become desired and dreaded categories, respectively, which shape institutional identities, forging institutional narratives that exclude the Pemon from the realm of the rational, and serving as powerful motivators for the agencys policy science. Because of these links between desires, institutional narratives and a politicized Science (Latour, 2004), EDELCA has overlooked alternative modes of knowledge production that may facilitate more eective, democratic re management in the Gran Sabana. In particular, Pemon tenets and practices of controlled burning deserve further attention by re ecologists to forge a preventative, rather than reactive approach to re management (see Biddulph & Kellman, 1998). Instead, EDELCA re managers collect, categorize, and interpret data on re occurrences in ways that support the winning explanation for forest loss in the Gran Sabana. This ironical situationthis un-seeing of knowledge and practices that many scientists argue are benecialis due to the institutional narratives of risk which underpin the agencys raison detre, but also the role that EDELCA plays in Venezuelas state-building project in the Guayana. That is, in the interest of furthering state control, EDELCA provides to the state a mass of information which its strategic position can enable it to exploit (Foucault, 1980, p. 75). Thus the combat operations pursued by EDELCA are not simply directed against res but rather against the Pemon; or more specically, against the presumed irrationality and destructiveness of the Pemon, which ostensibly stem from their lack of proper knowledge. Ultimately, the conict that has held the Gran Sabana in its grip for the past two decades is not about re, but about knowledge. What counts in this war for truth (Foucault, 1980) is quantitative data, and those who have the greatest political and economic means to produce numbers and make them seem valuable are in a privileged position to dene truth. The case of the Gran Sabana is not unique, nor is it more important per se than other environmental conicts, which as often as not are fueled by politicaleconomic imperatives and by identity formations shaped by institutional cultures, desires, and powerful narratives of place and people. The identities of environmental planners are shaped in part by their institutional cultures, which just as inevitably are

THE KNOWLEDGE THAT COUNTS

1951

suused with narratives of places and people, memories of what have been and fantasies of what could be, and desires to know what to do to so that landscapes can be fashioned the way they should be. What behooves planning researchers is to carefully unpack such cultural constructions, track the ways in which knowledge is read, appropriated and trans-

formed in agency cultures, and assess how the multiple, local Panopticisms of environmental planning agencies eect relations of power and environmental realities on the ground, in the actual geographies where planners can never avoid but becoming embroiled in contests of place-making.

NOTES
1. Institutional ethnography is a growing tradition in planning research, where the focus is often the disjunctures between broader institutional planning goals and actual practice. One approach is the micropolitics of practice (see e.g., Allmendinger, 1998; Forester, 1989), another is the work of planning theorists who question how planning issues are framed and how planning practices are inuenced by narratives, knowledge systems, politicaleconomic structures, and everyday negotiations. Examples of this approach are Flyvbjergs (1998, 2002) application of Foucaultian insights in his exploration of rationality and power in Danish city planning and Fischlers (1998) use of Foucaultian genealogies in explorations of zoning histories. Other important writers in the related eld of constructivist institutional analysis, which oers important critiques of assumptions of rationality in environmental institutions, are Espeland (1998), Ferguson (1995), Fischer (2000), Hajer (1995). 2. I draw on my interpretation of Lacanian thought primarily on Bracher (1993, 1994), Fink (1995, 1999), Hillier (2003), Hillier and Gunder (2003), Gunder (2003, 2004), Philo (2000) and Zizek (1993, 1999). 3. Although traditional conceptualizations of surveillance see it as a technique of power that works to enforce domination over bodies (McHoul & Grace, 1993, p. 65; Merquior, 1985, p. 113), Robinson (2000) emphasizes that surveillance is embodied; that is, made operational through specic individuals in specic places, and therefore by necessity is fractured and incomplete (Robinson, 2000, pp. 68, 7879; see also Prado, 2000, p. 73). 4. See http://maps.geog.umd.edu/rms/ and http:// www.re.uni-freiburg.de/intro/About.html. 5. To decide whether to send a squad, the re manager in San Ignacio must perform a rapid evaluation of potential costs and risks. A helicopter response is expensive and must be justied to superiors, which puts pressure on the re manager to represent each re as threatening, regardless of any eld observations to the contrary. However, re ghters told me they rarely ght res once they arrive in the eld. Instead, they typically stand by to ensure the res do not spread out of control, or they stop res that, in their estimation, would have died down on their own. These contentions are unsupported by empirical data, of course; the purpose here is not to speculate on the potential destructiveness of res, but rather to note that such alternative observations are not reected in the agency reporting system. From the point of view of re managers, this reporting system is justied because of the implied threat of these res: there is certainly no nefarious intent on the part of re managers to inate the numbers of res. However, the unintended consequence of this strategy of quantication is to provide rhetorical support for continued re suppression and to further solidify institutional narratives of indigenous destructiveness. 6. The notion of framing is borrowed from Erving Go mans social theory, in which he describes frames as principles of organization which dene the meaning and signicance of social events (Goman, 1997, p. xlvi). 7. Such prescriptive burning is also a common indigenous practice in grasslands environments elsewhere in Latin America. In Brazil, the Kayapo burn the cerrado (shrublands) for esthetic reasons, to reduce populations of snakes and scorpions, to make walking easier (Anderson & Posey, 1989), and to modify soil properties and create re breaks (Hecht & Posey, 1989). The Kraho also burn the cerrado to eliminate pests and to create an esthetically clean landscape, but also to create re breaks in order to protect desired vegetation (Mistry, 2005). In Chile, the Araucano used prescribed burning to create and maintain stands of the monkey-puzzle tree (Araucaria araucana) (Aagesen, 2004).

1952

WORLD DEVELOPMENT

REFERENCES
Aagesen, D. (2004). Burning monkey-puzzle: Native re ecology and forest management in northern Patagonia. Agriculture and Human Values, 21, 233242. Adger, W. N. et al. (2001). Advancing a political ecology of global environmental discourses. Development and Change, 32, 681715. Agrawal, A. (1995). Dismantling the divide between indigenous and scientic knowledge. Development and Change, 26, 413439. Alcorn, M. (1994). The subject of discourse: reading Lacan through (and beyond) post-structuralist contexts. In M. Bracher et al. (Eds.), Lacanian theory of discourse (pp. 1945). New York: New York University Press. Allmendinger, P. (1998). Deconstructing communicative rationality: a critique of Habermasian collaborative planning. Environment & Planning A, 30(11), 19751992. Anderson, A. B., & Posey, D. A. (1989). Management of a tropical scrub savanna by the Gorotire Kayapo of Brazil. Advances in Economic Botany, 7, 159173. Barse, W. P. (1990). Preceramic occupations in the Orinoco River valley. Science, 250, 13881390. Bassett, T., & Bi Zueli, K. (2000). Environmental discourses and the Ivorian savanna. Annals of the Association of American Geographers, 90, 6795. Biddulph, J., & Kellman, M. (1998). Fuels and re at savanna-gallery forest boundaries in southeastern Venezuela. Journal of Tropical Ecology, 14, 445461. Bond, W., & Archibald, S. (2003). Confronting complexity: Fire policy choices in South African savanna parks. International Journal of Wildland Fire, 12, 381389. Bracher, M. (1993). Lacan, discourse and social change. Ithaca: Cornell University Press. Bracher, M. (1994). On the psychological and social function of language: Lacans theory of the four discourses. In M. Bracher et al. (Eds.), Lacanian theory of discourse: Subject, structure and society (pp. 107128). New York: New York University Press. Braithwaite, R. W. (1996). Biodiversity and re in the savanna landscape. In O. T. Solbrig, E. Medina, & J. Silva (Eds.), Biodiversity and savanna ecosystem processes. Berlin: Springer. Butt-Colson, A. (1985). Routes of knowledge: An aspect of regional integration in the circum-Roraima area of the Guiana Highlands. Antropologia, 6364, 103149. n Interinstitucional (1989). Parque Nacional Comisio n y desarrollo del Canaima: bases para la ordenacio n sector oriental y plan operativo. Caracas: Comisio Interinstitucional MARNR-INPARQUES-CVGEDELCA-CORPOTURISMO. Cooke, P. (2000). Fire management on Aboriginal Lands in the top end of the Northern Territory, Australia. In J. Russell-Smith et al. (Eds.), Fire and Sustainable Agricultural and Forestry Development in Eastern Indonesia and Northern Australia (pp. 102107). Canberra: Australian Centre for International Agricultural Research. Coronil, F. (1997). The magical state: Nature, money, and modernity in Venezuela. Chicago: University of Chicago Press. tnica Pemon y el Cousins, A. (1991). La Frontera E Impacto Socio-economico de la Mineria de Oro. PhD Dissertation. Caracas: Instituto Venezolano de nes Cient cas. Investigacio Crush, J. (Ed.) (1995). Power of development. New York: Routledge. a venCruxent, J. M. (1971). Apuntes sobre arqueolog rez ezolana. In M. G. Arroyo, J. M. Cruxent, & S. Pe nico de Venezuela Soto de Atencio (Eds.), Arte Prehispa n Eugenio Mendoza. (pp. 1925). Caracas: Fundacio Cruxent, J. M. (1972). El Paleoindio en Venezuela. Venezuela Suya, 1(1), 413. Dalby, S. (1998). Introduction: Environmental geopol . Tuathail, S. Dalby, & P. Routledge itics. In G. O (Eds.), The geopolitics reader (pp. 112). London: Routledge. Dalby, S. (1999). Globalization or global apartheid? Boundaries and knowledge in postmodern times. In D. Newman (Ed.), Boundaries, territory and postmodernity (pp. 132150). Portland: Frank Cass. Dezzeo, N. (1994). Ecology of the Gran Sabana uplands (Venezuelan Guayana). Scientia Guaianae, 4, xvxxxviii. n, N. (2005). Carbon and nutrients Dezzeo, N., & Chaco loos [sic] in aboveground biomass along a re induced forestsavanna gradient in the Gran Sabana, southern Venezuela. Forest Ecology and Management, 209, 343352. Dryzek, J. (1997). The politics of the Earth: Environmental discourses. Oxford: Oxford University Press. EDELCA (2000). EDELCA Cifras. Caracas: CVG n del Caron . Electricacio Eden, M. J. (1974). Palaeoclimatic inuences and the development of savanna in southern Venezuela. Journal of Biogeography, 1(2), 95109. Edwards, P. J. (1984). The use of re as a management tool. In P. de V. Booysen, & N. M. Tainton (Eds.), Ecological eects of re in South African ecosystems (pp. 350362). Berlin: Springer-Verlag. Escobar, A. (1995). Encountering development. Princeton: Princeton University Press. Espeland, W. (1998). The Struggle for Water. Chicago: University of Chicago Press. Everson, C. S., Everson, T. M., & Tainton, N. M. (1988). Eects of intensity and height of shading on the tiller initiation of six grass species from the Highland Sourveld of Natal. South African Journal of Botany, 54, 315318. Fairhead, J., & Leach, M. (1996). Misreading the African landscape: Society and ecology in a forestsavanna mosaic. London: Cambridge University Press. Fairhead, J., & Leach, M. (1998). Reframing deforestation: Global analyses and local realitiesStudies in West Africa. London: Routledge. Ferguson, J. (1995). The anti-politics machine: Development, depoliticization and bureaucratic power in Lesotho. Minneapolis: University of Minnesota Press.

THE KNOWLEDGE THAT COUNTS ndez, L. (1984). Determinacio n de las areas prioFerna o ritarias del sector suroriental de la cuenca alta del R ria descriptive. Tarenken, La Gran Yuruani: memo n del Caroni, DepartaSabana: C.V.G. Electricacio n y Mantenimiento de Cuencas. mento de Proteccio ndez, L. (1987). Inventario biof sico de las subcuFerna encas Yuruani, Alto Kukenan y Arabopo; aspecto n del n. Puerto Ordaz: C.V.G.Electricacio vegetacio Caroni. Fink, B. (1995). The Lacanian subject: Between language and Jouissance. Princeton: Princeton University Press. Fink, B. (1999). The master signier and the four discourses. In D. Nobus (Ed.), Key concepts of Lacanian psychoanalysis (pp. 2947). New York: Other Press. Fischer, F. (2000). Citizens, experts and the environment: the politics of local knowledge. Durham: Duke University Press. Fischler, R. (1998). Toward a genealogy of planning: zoning and the Welfare State. Planning Perspectives, 13, 389410. Flyvbjerg, B. (1998). Rationality and power: Democracy in practice. Chicago: University of Chicago Press. Flyvbjerg, B. (2002). Bringing power to planning research: one researchers praxis story. Journal of Planning Education and Research, 21, 353366. Fo lster, H. (1986). Forestsavanna dynamics and desertication processes in the Gran Sabana. Interciencia, 11(6), 311316. Fo lster, H. (1995). Local population concentrations in the Venezuelan Guayana and the breakdown of traditional swidden agriculture. Scientia Guaianae, 5, 6578. n de la Fo lster, H., & Dezzeo, N. (1994). La degradacio n. Scientia Guaianae, 4, 145186. vegetacio Forester, J. (1989). Planning in the face of power. Berkeley: University of California Press. Foucault, M. (1972). The archaeology of knowledge. Andover: Tavistock Publications. Foucault, M. (1977). Discipline and punish [1975] (A. Sheridan, Trans.). New York: Random House. Foucault, M. (1980). In C. Gordon (Ed.), Power/ knowledge: selected interviews and other writings, 19721977 (C. Gordon, L. Marshall, J. Mepham, K. Soper, Trans.). New York: Pantheon Books. n de incenGalletti, U. A. 1980. Proyecto de prevencio n ecolo gica de la Gran Sabana. dios y proteccio Internal memo, Instituto Nacional de Parques. a, J. R. (1982). Zona neotropical-caso de estudio: Garc Cataratas, energia hidroelectrica y agua para industrias: Parque Nacional Canaima, Venezuela. Paper presented at the World National Park Congress, Bali, Indonesia, 1982. a, J. R. (1994). Prevencio n y combate de incendios Garc en los parques nacionales y monumentos naturals. Memorandum no. 618, addressed to the president of a, October 25, INPARQUES, Francisco Loreto Mej 1994. Goman, E. (1997). The Goman reader. Malden, MA: Blackwell Publishers. mez, E. (2003). Personal interview. Puerto Ordaz, Go Venezuela.

1953

n, I. (2002). La Estructura Regional Venezolana al Guille o Orinoco y el Mercosur. Aldea Mundo, Sur del R 6(11), 6373. Gunder, M. (2003). Passionate planning for the others desire: an agonistic response to the dark side of planning. Progress in Planning, 60, 235319. Gunder, M. (2004). Shaping the Planners ego-ideal: A Lacanian interpretation of planning education. Journal of Planning Education and Research, 23, 299311. Gupta, A. (1998). Postcolonial developments: Agriculture in the making of modern India. Durham: Duke University Press. Gutting, G. (1989). Michel Foucaults archaeology of scientic reason. New York: Cambridge University Press. Hajer, M. (1995). The politics of environmental discourse: ecological modernization and the policy process. Oxford: Clarendon Press. Hall, M. (1984). Mans historical and traditional use of re in Southern Africa. In P. de V. Booysen, & N. M. Tainton (Eds.), Ecological eects of re in South African ecosystems (pp. 4052). Berlin: SpringerVerlag. Harwell, E. (2000). Remote sensibilities: Discourses of technology and the making of Indonesias natural disaster. Development and Change, 31, 307340. Haynes, C. D. (1978). Land, trees and man (gunret, gundulk, dja bining). Commonwealth Forestry Review, 2, 99106. Haynes, C. D. (1985). The pattern and ecology of Munwag: Traditional aboriginal re regimes in North-Central Arnhemland. Proceedings of the Ecological Society of Australia, 13, 203214. Hecht, S., & Posey, D. (1989). Preliminary results on soil Indians. management techniques of the Kayapo Advances in Economic Botany, 7, 174188. n de los bosques de la ndez, L. (1987). Degradacio Herna Gran Sabana. Pantepui, 3, 1125. Hillier, J. (2003). Agonizing over consensus: Why Habermasian ideals cannot be real. Planning Theory, 2(1), 3759. Hillier, J., & Gunder, M. (2003). Planning fantasies? An exploration of a potential Lacanian framework for understanding development assessment planning. Planning Theory, 2(3), 225248. Huber, O. (1995). Geographical and physical features. In P. Berry, B. Holst, & K. Yatskievych (Eds.), Introduction. Flora of the Venezuelan Guayana (Vol. I, pp. 161). St. Louis: Missouri Botanical Gardens. a Ecolo gica de la Huber, O., & Febres, G. (2000). Gu Gran Sabana. Caracas: Ecograph Proyectos Ediciones. Huber, O., & Zent, S. (1995). Indigenous people and vegetation in the Venezuelan Guayana: Some ecological considerations. Scientia Guaianae, 5, 3764. Hudak, A., Fairbanks, D., & Brockett, B. (2004). Trends in re patterns in a southern African savanna under alternative land use practices. Agriculture, Ecosystems and Environment, 101, 307325. Kingsbury, N. (1999). Increasing pressure on declining resources: A case study of Pemon Amerindian shifting cultivation in the Gran Sabana, Venezuela. Ph.D. dissertation. Toronto: York University.

1954

WORLD DEVELOPMENT Merquior, J. G. (1985). Foucault. Berkeley: University of California Press. Mignolo, W. (2000). Local histories/global designs: Coloniality, subaltern knowledges, and border thinking. Princeton: Princeton University Press. Miranda, M. et al. (1998). All that glitters is not gold: Balancing conservation and development in Venezuelas frontier forests. New York: World Resources Institute. Mistry, J. (2005). Indigenous re management in the of cerrado of Brazil: The case of the Kraho ns. Human Ecology, 33(3), 365386. Tocant Mitchell, T. (1995). The object of development: Americas Egypt. In J. Crush (Ed.), Power of development (pp. 129157). London: Routledge. Moore, H. (1996). The changing nature of anthropological knowledge: An introduction. In H. Moore (Ed.), The future of anthropological knowledge (pp. 115). London: Routledge. Mujica, O. (1984). Estudio Socio-Economico de San Francisco de Yuruani, San Ignacio de Yuruani y Santa Cruz de Mapauri a los nes del aprovechamiento o Yuruan, Estudio preliminary del hidroelectrico del R n impacto ambiental. Caracas: C.G.G. Electricacio , Divisio n de Planicacio n de Generacio n del Caron e Informatica. Nader, L. (1996). Introduction: Anthropological inquiry into boundaries, power, and knowledge. In L. Nader (Ed.), Naked science: Anthropological inquiry into boundaries, power, and knowledge (pp. 168). New York: Routledge. Nygren, A. (1999). Local knowledge in the environmentdevelopment discourse. Critique of Anthropology, 19(3), 267288. rez, J. W. (2002). Personal interview. Kumarakapay, Pe Venezuela. rez, L. (1999). Revista Especial Macagua II 23 de Pe blicos. Enero. Puerto Ordaz: Gerencia de Asuntos Pu Philo, C. (2000). Foucaults geography. In M. Crang, & N. Thrift (Eds.), Thinking space (pp. 205238). London: Routledge. Pile, S. (1996). The body and the city: Psychoanalysis, space and subjectivity. New York: Routledge. Prado, C. G. (2000). Starting with foucault: an introduction to Genealogy. Boulder: Westview Press. Purcell, T. (1998). Indigenous knowledge and applied anthropology: Questions of denition and direction. Human Organization, 57(3), 258272. Pyne, S. (1991). Burning bush: A re history of Australia. New York: Henry Holt and Company. Pyne, S. (1995). World re: The culture of re on Earth. Seattle: University of Washington Press. Pyne, S. (1996). Introduction to wildland re. New York: Wiley. Ramos, B. M. (2003). Personal interview. Puerto Ordaz, Venezuela. Rinaldi, M., & Schubert, C. (1991). Aspectos paleo ticos y su possible relacio n con la ocupacio n clima humana en la Gran Sabana, Escudo de Guayana, gica, 7576, Venezuela; informe preliminar. Antropolo 127140. Robinson, J. (2000). Power as friendship: Spatiality, femininity and noisy surveillance. In J. Sharp, P.

Kingsbury, N. (2001). Impacts of land use and cultural change in a fragile environment: Indigenous acculturation and deforestation in Kavanayen, Gran Sabana, Venezuela. Interciencia, 26(8), 327337. Kull, C. (2002a). Empowering pyromaniacs in Madagascar: Ideology and legitimacy in community-based natural resource management. Development and Change, 33, 5778. Kull, C. (2002b). Madagascars burning: The persistent conict over re. Environment, 44, 819. Kull, C. (2002c). Madagascar aame: Landscape burning as peasant protest, resistance, or a resource management tool?. Political Geography, 21, 927953. Lacan, J. (1978). In J.-A. Miller (Ed.), The four fundamental concepts of psycho-analysis (A. Sheridan, Trans.). New York: Norton. Lacan, J. (1995). In R. Feldstein, B. Fink, & M. Jaanus (Eds.), Reading seminar XI: Lacans four fundamental concepts of psychoanalysis. Albany: State University of New York Press. Laris, P. (2002). Burning the seasonal mosaic: Preventative burning strategies in the wooded savanna of southern Mali. Human Ecology, 30(2), 155186. Laris, P. (2004). Grounding environmental narratives: The impact of a century of ghting against re in Mali. In W. Moseley, & B. Ikubolajeh Logan (Eds.), African environment and development: Rhetoric, programs, realities (pp. 6385). London: Ashgate. Latour, B. (1986). Laboratory life: The social construction of scientic facts. London: Sage Publications. Latour, B. (2004). Politics of Nature: How to bring science into democracy (C. Porter, Trans.). Cambridge, MA: Harvard University Press. Leach, M., & Fairhead, J. (2000a). Challenging neoMalthusian deforestation analyses in West Africas dynamic forest landscapes. Population and Development Review, 26(1), 1743. Leach, M., & Fairhead, J. (2000b). Fashioned forests past, occluded histories? International environmental analyses in West African locales. Development and Change, 31, 3559. Leach, M., & Fairhead, J. (2002). Manners of contestation: Citizen science and indigenous knowledge in West Africa and the Caribbean. International Social Science Journal, 173, 299311. Lewis, H. T. (1986). Fire technology and resource management in aboriginal North America and Australia. In N. Williams, & E. Hunn (Eds.), Resource managers: North American and Australian huntergatherers (pp. 4567). Canberra: Australian Institute of Aboriginal Studies. Lewis, H. T. (1989). Ecological and technological knowledge of re: Aborigines versus park rangers in Northern Australia. American Anthropologist, 91, 940961. Major-Poetzl, P. (1983). Michel Foucaults archaeology of western culture: Towards a new science of history. Chapel Hill: The University of North Carolina Press. n y Cambio Social entre Mansutti, A. (1981). Penetracio los Akawaio y Pemon. PhD Dissertation. Caracas: nes Cient cas. Instituto Venezolano de Investigacio McHoul, A., & Grace, W. (1993). A Foucault primer: Discourse, power and the subject. New York: New York University Press.

THE KNOWLEDGE THAT COUNTS Routledge, C. Philo, & R. Paddison (Eds.), Entanglements of power: Geographies of domination/resistance (pp. 6792). New York: Routledge. guez, I. (2004). Conocimiento ind gena vs. cientRodr co: el conicto por el uso del fuego en el Parque Nacional Canaima, Venezuela. Interciencia, 29(3), 121129. Romero, L. (2002). Personal interview. Kumarakapay, Venezuela. Rull, V. (1992). Successional patterns of the Gran Sabana (Southeastern Venezuela) vegetation during the last 500 years and its responses to climatic uctuations and re. Journal of Biogeography, 19, 329338. Russell-Smith, J. et al. (Eds.). (2000). Fire and Sustainable Agricultural and Forestry Development in Eastern Indonesia and Northern Australia. In Proceedings of an international workshop held at Northern Territory University, Darwin, Australia (pp. 1315). Sawicki, J. (1989). Foucault, feminism, and questions of identity. In G. Gutting (Ed.), The Cambridge companion to Foucault (pp. 286313). New York: Cambridge University Press. Sawicki, J. (1994). Feminism and the power of Foucaultian discourse. In J. Arac (Ed.), After Foucault: Humanistic knowledge, postmodern challenges (pp. 161178). New Brunswick: Rutgers University Press. Schubert, C. (1986). Paleoenvironmental studies in the Guayana Region, Southeast Venezuela. Current Research in the Pleistocene, 3, 8890. Schubert, C. (1995). Origin of the Gran Sabana in Southeastern Venezuela: No longer a Lost World. Scientia Guaianae, 5, 147174. Sillitoe, P. (1998). The development of indigenous knowledge: a new applied anthropology. Current Anthropology, 39(2), 223252. Sletto, B. (2006). Burn marks: The becoming and unbecoming of an indigenous landscape. Unpublished PhD Dissertation. Ithaca: Cornell University. Thomas, D. J. (1982). Order without government: The society of the pemon indians of Venezuela. Illinois Studies in Anthropology No. 13. Urbana: University of Illinois Press. Turner, M. (2003). Methodological reections on the use of remote sensing and geographic information

1955

science in human ecological research. Human Ecology, 31(2), 255279. Turner, M., & Taylor, M. (2003). Critical reections on the use of remote sensing and GIS technologies in human ecological research. Human Ecology, 31(2), 177182. n ecolo gico-cultural de los Urbina, L. 1979. Adaptacio n-Arekuna: el caso de Tuauke n. MSc Thesis. Pemo Caracas: Instituto Venezolano de Investigaciones cas. Cient a, Urbina, L., & Dieter Heinen, H. (1982). Ecolog n social y distribucio n especial: estudio de organizacio genas: Pemo n y Warao. caso de dos polaciones ind gica, 57, 2554. Antropolo Van der Hammen, T. (1974). The Pleistocene changes of vegetation and climate in tropical South America. Journal of Biogeography, 1(1), 326. Whitehead, P. et al. (2003). Customary use of re by indigenous peoples in Northern Australia: Its contemporary role in savanna management. International Journal of Wildland Fire, 12, 415425. Williams, R. J., Gill, A. M., & Moore, P. H. R. (1998). Seasonal changes in re behaviour in a tropical savanna in Northern Australia. International Journal of Wildland Fire, 4, 227239. n e histo ria de la Worbes, M. (1999). Degradacio n boscosa de la Gran Sabana. Scientia vegetacio Guaianae, 9, 84148. Zerpa, F. (2003). Personal interview. Puerto Ordaz, Venezuela. a, S. (2003). Zerpa, F., Acevedo, M., Ablan, M., & Garc n espacio-tempoSIPCIV: Un sistema de informacio ral para caracterizar areas afectadas por incendios. Report presented at CBComp 2003, I Workshop de Tecnologia da Informac ao Aplicada ao Meio Ambi cas, UniverenteSistema de Informac o es Geogra sidade do Vale do Itajai, Itajai, Brazil. Online: http:// www.cbcomp.univali.br/anais/pdf/2003/ wma061.pdf, April 30, 2006. Zizek, S. (1993). Tarrying with the negative: kant, hegel and the critique of ideology. Durham: Duke University Press. Zizek, S. (1999). The seven veils of fantasy. In D. Novus (Ed.), Key concepts of Lacanian Psychoanalysis (pp. 190218). New York: Other Press. Zizek, S. (2002). For they know not what they do. London: Verso.

Available online at www.sciencedirect.com