Why is the Amazon scenario so complicated?

Economic Development, Social Identity and Community Empowerment in the Central and Western 

Abstract-- A multi-dimensional and multi-scalar perspective is used to contrast different experiences with community development involving new settlers, the historic riverine peasantry and Amerindians on and beyond the advancing frontier in the Brazilian Amazon. Local development is shown to occur only when a compromise is reached between the economic development of individual farmers, common social background and community political empowerment. However, the local development of communities does not necessarily result in regional development because contradictions present between political actors can cancel out gains in one dimension or scale vis-à-vis others and so frustrate wider processes.

A laboratory of alternative, multi-scalar politics
The complexity of multi-scalar politics in the Amazon is examined with the objective of contributing to the on-going debate concerning the continued relevance of the concept of scale in Geography. We will defend those opposed to jettisoning the concept of scale from the geographical tool bag (Hoefle, 2006a; Jonas, 2006; Leitner and Miller, 2007) against those in favour of a ‘flat ontology’ without the hierarchical notion of scale (Marston et al., 2005; Jones et al., 2007; Escobar, 2007). Nevertheless, with due justice to the latter, the fundamental importance of focusing political research on local people instead of doing only ‘leadership research’ is emphasised here, because leaders and representatives, who as part of governmental and non-governmental power hierarchies, tend to paint a much too rosy picture of their practice in the Amazon. The Brazilian Amazon is one of the world’s great laboratories of alternative politics in which complex political networks arose over the last twenty years involving multi-scalar alliances surrounding regional and global environmental, developmental and ethnic issues. A host of global, national, regional, state-level, municipality-level and community-level actors interact and struggle over the fate the Amazon and the concept of scale is extremely important for understanding what appears to be a political free-for-all. Depending on the place, or even juxtaposed in the same space, four kinds of political networks are encountered: 1) rural clientelist politics merged as part of the elective base; 2) mass-mobilisation politics of large urban areas; 3) alternative environmentalist and ethnic activist non-governmental networks reaching in from different non-local scales to directly, or through partnerships with different levels of government, implement ecological, agrarian reform, ethnic and sustainable development projects; and 4) communitarian bottom-up political mobilisation which tries to tap into the other networks in order to obtain improved social and economic public services, of which the conservation of local resources may or may not play an important part. The latter two political networks arose during the 1990s in the attempt to circumvent the historical gatekeepers (for example, village leaders, county leaders, state and federal congress members, governors and ministers) who control access to resources at different levels of Brazilian politics. We will highlight the limits to the new communtarian politics as well as the scalar contradictions present even in successful Amerindian and farmer movements. This complex network of different kinds of political actors has been researched in the Central and Western Amazon since 1997 in two different phases. During the first four years county and state-level politics were surveyed in seventeen municipalities of Amazonas and Roraima states. At these levels electoral politics were analysed and ‘leaderships’ were interviewed: state and local officials, priests and bishops, military officers, representatives of non-governmental organisations and elderly ‘local historians’. Results from this more regional approach to multi-scale politics as related to social and ideological issues were published in Hoefle (2000; 2006b). Then, from 1999 to 2005, we returned to eleven of these municipalities in Amazonas State and then to two other municipalities in Pará State in 2008 to interview 593 farmer and fisher families in 80 different communities/localities (Figure 2). In this field research, environmental and economic dimensions were added to the social and ideological Decision Making INTERNATIONAL NGOs, President Church Business & Labour NATIONAL Minister MST Governor Senator COIAB STATE State Deputy Federal Deputy Limits of Communitarian Politics County Leader LOCAL Federal Sustainable Amerindian Reservations, Village Leader & Development Councilperson Projects Agrarian Reform Projects Grassroots Local People Alternative / Communitarian Rural / Urban clientelism merged Ecological Reserves & Figure 1 Cross-cutting communitarian, alternative and hierarchical politics in the Brazilian Amazon. 282 Geographical Research • August 2010 • 48(3):281–296 © 2009 The Authors Journal compilation © 2009 Institute of Australian Geographers dimensions of the previous investigation of multi-scalar politics. This enabled ecological and logistical restraints to be related to farming and degree of technical capitalisation and market articulation to the participation of farmers in community politics and to their success or failure in pressuring other levels of government for public services and infrastructure meant to promote local development. Researchers of the Amazon usually do either: 1) exquisitely detailed research of farmers in specific localities, normally along roads in the devastated eastern and southern Amazon, where landless rural worker movements and non-governmental organisations concentrate their activities, and then over-generalise to the regional level; or 2) engage in leadership research, in which key individuals at the municipal and regional levels are interviewed, usually in town, so that trends presented second-hand tend to be idealised. Both kinds of investigation have been undertaken in the present study in an attempt to bridge scales and produce a composite picture of Amazonian politics as seen bottom-up by the local people living in different parts of this diverse region. We will see that some of political experiments evaluated started well but ended poorly; some are still quite promising while others never took hold nor were not even meant to. Economic development and/or community development? In this part we describe the environmental, economic and social details of the farming systems practised in the different study areas which is crucial for evaluating the role of grassroots democracy in sustainable development. All too often, the economic development of small farming in the Amazon is ignored by environmentalist non-governmental organisations or thought to be impossible by regional planners. It is argued that some economic development needs to occur but it must be fairly equitable in order to produce a level political playing field, which in turn permits input of local experience and knowledge into development policy. This is reinforced in the second part of the article where it is shown that developmentalist politicians, eco-centric activists and even some researchers, each in their own way, take a superficial view of the regional diversity present in the Amazon and impatiently try to implement top-down change which only ends up short-circuiting the process. The principal aim of this section is to demonstrate that introducing environmentally benign farming Figure 2 Research sites in the Central and Western Amazon. A.M. de Souza Mello Bicalho and S.W. Hoefle: Economic Development, Social Identity and Community Empowerment 283 © 2009 The Authors Journal compilation © 2009 Institute of Australian Geographers systems or communitarian production schemes come to naught if farm produce cannot be taken to market or if consumers do not exist for their goods. Consequently, the Political Economy of sustainable development in the Amazon must be constructed from the details and not based on vague assumptions and sweeping regional generalities about ‘peoples of the forest’ and ‘subsistence-minded peasants’. Polarised economic development and stymied community empowerment in frontier road settlement Over the last three decades, research on the Amazon has usually been directed towards issues surrounding deforestation and land ownership disputes present along, and increasingly behind, the advancing frontier which is moving north and west along roads deeper into the Amazon basin (Ianni, 1979; Foweraker, 1981; Moran, 1981; Bakkx, 1988; Schmink and Wood, 1992; Oliveira, 1999; Simmons et al., 2007). However, the most interesting political and economic experiments are not found along roads. As the frontier is located far from major consumer markets in terms of absolute and relative distance, due to poor transport systems which are often inoperable during the rainy season, settlement patterns are dispersed and involve interactions amongst a host of strangers from all over Brazil. Attempts to grow commercial crops or even export commodities fail so that socially polarised cattle ranching is often the only viable alternative. First, we review the implications of these trends for political mobilisation and for pressuring local and state government for basic services in southern Amazonas State, which is located far from consumer markets on the greater frontier. Then we turn to more positive economic, though not political, trends in the municipality of Presidente Figueiredo located on a secondary front of road settlement just north of Manaus (the main consumer market of the Central and Western Amazon). Finally, we examine a successful case of political mobilisation without economic development on the distant frontier of Itaituba in western Pará State. The south of Amazonas State is undergoing social and environmental transformation with the arrival of the greater frontier in the municipality of Humaitá and surrounding areas, located at the cross roads of the BR-230 (Transamazonian Highway) and the BR-319 (Porto Velho-Manaus Highway). The arrival of settlers from the developed south and south-east of Brazil, who pass through intermediate stages of migration via the booming Central-West, and from the underdeveloped north-eastern states, who first pass through the eastern Amazon, gives rise to a great diversity of agrarian actors, such as small food and regional fruit producers, middle-scale ranchers and large to agribusiness-scale commodity and beef producers. These recent migrants settle alongside older settlers on the roads as well as the historic peasantry on the Madeira and Purus rivers and their tributaries (treated in the next section). Contrary to typical dualistic approaches to road settlement in the Amazon, pitting subsistence squatters against speculative ranchers, complex trends are at work along the cutting edge of the frontier in southern Amazonas State with the presence of different rural actors with complementary and opposing productive strategies. Typical smallholders plant on average 4.5 ha of basic food crops on the 100 ha lot of public land which they received from the government. Nearly half are semi-proletarians who earn more off the farm in wages (US$1337 per year in 2005) and from government transfer payments (US$591) than from their own farming activities (US$643). Instead of the classic problem of the Brazilian frontier of conflict existing between peasants and ranchers, a relationship of symbiosis prevails because the ranchers need labour to maintain their pastures clear in a context where secondary forest reappears very quickly. Alongside these historic rural actors of the Brazilian frontier are two new groups: 1) a third of the smallholders who capitalise their activities using fertilisers and some pesticides to plant on average 7.6 ha of food and regional fruit crops as well as practising small-scale cattle ranching; and 2) large to agribusiness-scale commodity and beef producers. The latter group uses state-ofthe-art mechanised cropping and improved cattle ranching methods in savannah enclaves, where environmental legislation (at the time of the research) permitted the conversion of 35% of the farm into productive land as opposed to only 20% in forest areas (presently in the process of being increased to 50% for all Amazonian ecosystems). As the small capitalising farmers face diffi- culty in finding markets for their crops, they are increasingly turning to ranching, which demands more land than they have (average property size is 146 ha) so that they may end up selling out to neighbouring ranchers. Frontier ranching 284 Geographical Research • August 2010 • 48(3):281–296 © 2009 The Authors Journal compilation © 2009 Institute of Australian Geographers requires a lot of land to be economically viable. Capitalising smallholders earned US$3180 in 2005 while ranchers earned US$4584 but on average needed 916 ha of land to do this (Table 1). The agribusiness firms are also moving into beef production because they are still located too far from farm input factories and grain terminals to export soybeans profitably. Specialised grain producers have great difficulty turning a profit while in 2004 one diversified agribusiness earned an enormous sum of money by regional standards. Despite losing US$33 169 in export soybean production, a healthy profit was earned from rice (US$80 000) and beef production (US$113 591) for the Manaus market. Over the last eight years, this property increased its size by 50% by buying out smallholders along the asphalted BR-319 south in the direction of Porto Velho. Consequently, the end result in both forested and savannah areas of this distant frontier area is increasing concentration of land ownership. The presence of recent migrants from different developed and under-developed regions of Brazil as well as from nearby riverine areas who live dispersed along roads in a context of great social inequality means that settlers have very little in common. Neighbours live far from one another; they are seldom relatives or even friends and instead of identifying themselves with properly denominated communities, as is common in riverine Amazonia, they simply say they live so many kilometres from the municipal seat, where they have to visit occasionally for public services and to buy provisions. Political organisation is weak and settlers are poorly served by public services in the form of electricity, water treatment, schools, clinics and other community buildings, which are far and few between, a situation aggravated by poorly maintained roads which are not trafficable during the rainy season. Negative Lesson 1: lack of markets and socially polarised development make ecological, economic and political sustainability difficult. The area of road settlement north and east of Manaus (Itacoatiara, Presidente Figueiredo and Rio Preto da Eva municipalities) is exceptional by Amazonian standards. The dynamic industrial city of Manaus provides a large consumer market for farm products, which is accessible on all-weather roads, two crucial features generally absent in most areas of highway settlement in the Amazon. Also, the immediate rural hinterland of Manaus has benefited from newer decentralised and socio-environmental trends in development policy, which stress the search for alternative farming methods for smallholders who supply fruit and vegetables to the metropolitan market. Agro-forestry farming systems have become common in this area. In these systems permanent regional tree crops well adapted to Amazonian environments and poor sandy soils, such as açaí palms (Euterpe oleracea) and cupuaçu fruit trees (Theobroma grandiflorum), are interplanted with avocados, bananas, mangoes and papayas. The results are a vast improvement over typical, frontier open-field cropping of basic foodstuffs and extensive cattle raising. The few interviewed farmers who still plant only basic foodstuffs have an average of 2.3 ha in temporary crops with 17.5 ha (61%) of their land in forest and decreasing, in contrast to agro-forestry farmers who have 8.4 ha in crops and 36.9 ha (79%) in forest there is another 1.4 ha (3%) in fallow. Table 1 Disparity in farm incomes according to type of producer and location vis-à-vis consumer markets (deflated 2005 equivalent Brazilian Real converted to 2005 US dollar). Distance and Mode of Market Access Road Riverine Distant1 Accessible2 Nearby3 Distant4 Accessible5 Nearby6 Non-Capitalised Peasant 360 1 592 1 296 363 1 852 768 Small Capitalised Farmer 3 180 4 098 5 876 4 529 3 874 7 833 Medium Rancher 4 584 n.d. – – – – Medium Capitalised Farmer -219 – 29 294 – – – Large Rancher, Agribusiness 113 591 n.d. 467 136 – 75 521 – 1 Humaitá, 2 Presidente Figueiredo, 3 Rio Preto da Eva, 4 Benjamin Constant, Tabatinga, 5 Manicoré, 6 Careiro da Várzea, Iranduba, Manacapuru. A.M. de Souza Mello Bicalho and S.W. Hoefle: Economic Development, Social Identity and Community Empowerment 285 © 2009 The Authors Journal compilation © 2009 Institute of Australian Geographers Income for farmers planting basic food crops in Presidente Figueiredo municipality was only the 2005 equivalent of US$1592 per year in 1999 while agro-forestry farmers earned US$4098 that year. As more fruit trees came of production age and capitalised small farmers diversified products, income increased to US$5876 by 2002 in Rio Preto da Eva, where medium and large rural enterprises also arose to attend to the expanding demand of Manaus. The social results are promising. While typical migratory frontier farmers live in a small crude wood shack and have only a radio, a gas stove or a television, agro-forestry farmers live in houses three times as large, built of wood and brick and have an array of household appliances. One way of life is adapted to clearing forest for pasture, selling out and moving on, while agro-forestry holds farmers in the same place. Consequently, viable farm activities and transport systems for supplying the Manaus metropolitan market and a better quality of life in agro-forestry systems reduced deforestation, social inequality and outmigration. Typical frontier ranching is only encountered in the far north of Presidente Figueiredo municipality as well as two agribusiness holdings which predate agro-forestry farming there. Positive Lesson 1: socially equitable and environmentally benign development is possible through market access to metropolitan markets. However, the relatively equitable prosperity of smallholders has not been accompanied by political empowerment, which argues against a simplistic view of economic development being associated with political empowerment. As in the southern part of Amazonas State, settlers along roads are a mixture of strangers from different parts of Brazil. Legally constituted communities exist but settlers live dispersed along the roads and often only interact socially on the weekend. As a result, rural areas are poorly served by basic social services, which is ironic given the fact that Presidente Figueiredo municipality is second only to Manaus in government receipts. Unfortunately, kleptocracy reigns, a depressing trend in much of contemporary Brazil. Some of the generous royalty payments from the Balbina hydro-electric dam and from the Pitinga mines are spent in urban embellishment while much is squandered in political corruption and the municipal government is under investigation for embezzlement. Negative Lesson 2: together, dispersed settlement and lack of social ties hinder political empowerment. The mobilisation of small farmers of Itaituba municipality, who are trying to legalise their holdings within and on the edge of the Parque Nacional da Amazônia, shows yet another combination of environmental, economic and social circumstances influencing political empowerment. Itaituba is located along the advancing frontier in western Pará, and as part of this process, squatters settled in the poorly demarcated area of one of the largest forest reserves of the Amazon. This case shows how economic development is not necessary for political empowerment if farmers have the same social background and are threatened by a common problem. Indeed, due to transport difficulties and problems with fungi in rice, the main commercial crop, local economic development is low and farmers have a semi-subsistence focus. However, almost all of the farmers are ex-gold prospectors, who were originally from Maranhão State, which is an unusual situation on the advancing frontier, where settlers come from all over Brazil and have little in common. Add to this, their movement to regulate landownership, united through the local Farm Worker Union, and one can see how these frontier farmers achieved a high degree of political union through which they set up properly constituted communities and pressured for schools, clinics and even received housing grants to build far superior homes than those usually encountered along the frontier. With regard to legalising land ownership, a compromise was reached between the Farm Worker Union, the public land agency INCRA (Instituto Nacional de Colonização e Reforma Agrária) and the Brazilian environmental protection agency IBAMA (Instituto Brasileiro do Meio Ambiente e dos Recursos Naturais Renováveis), whereby the boundary of the park was shifted westward and the farmers beyond that point pulled back to the part conceded for settlement. The local Farm Worker Union was very important in the process. These unions in Brazil usually represent workers and smallholders, while medium and large farmers have separate associations, when they have them at all. In neighbouring Amazonas State, farm worker unions have little political importance, usually limiting their activities to processing requests for rural retirement, while in the land conflicts of Pará State they have an important political role of representing squatters against ranchers trying to evict them from their land (for details see Oliveira, 1999; Simmons et al., 2007). Positive 286 Geographical Research • August 2010 • 48(3):281–296 © 2009 The Authors Journal compilation © 2009 Institute of Australian Geographers lesson 2: common social identity and problems can overcome the centrifugal forces at work along the frontier. Riverine-road farmer interaction and high empowerment in capitalised metropolitan agriculture near Manaus Highly capitalised smallholder farming along roads southwest of Manaus is another rural system in the Central Amazon which involves relatively few farmers but shows how a metropolitan market can sustain a prosperous lifestyle and promote political mobilisation albeit at a cost to the environment. Farms are small with an average of 6 ha in intensive vegetable cropping with the extensive use of irrigation, plasticulture, agro-chemical fertilisers and crop defensives, including herbicide. Half of these farmers are Japanese immigrants and their children or southern Brazilian immigrants, though the rest are from Amazonas State, and all worked their way up from modest origins. Aquarter of the farmers are local people who have never moved while the rest have moved from one to three times during their lifetimes. However, once established locally, they have been living there for decades and they have reason to remain because high incomes underwrite an exceptionally prosperous lifestyle byAmazonian standards. Farmers have large modern houses with numerous sophisticated household appliances and urban style manufactured furniture. Capitalised commercial riverine farmers are located in close proximity to the highly visible capitalised farmers along roads and have historically interacted with them. In the beginning, the riverine farmers helped the new settlers devise appropriate methods for Amazonian conditions, with which they were totally unfamiliar, and later, when the farmers along roads developed capitalised vegetable cropping, they helped the riverine farmers. In Manacapuru, Iranduba, Careira da Várzea and Itacoatiara municipalities, this kind of farming is practised on the extensive flood plains along the Amazon River and is sustained by the demand from metropolitan Manaus and to a lesser extent from the small cities of this area, so that average area of crop land (6.3 ha) is twice that planted in peasant riverine areas (discussed below). A mixture of labour- and capitaldemanding fruit and vegetable crops are grown using an array of chemical and organic fertiliser, insecticide and to a lesser extent herbicide. As defined by local values, the standard of living of capitalised commercial riverine farmers is significantly better than in non-capitalised riverine areas, both for economic and political reasons. Higher monetary income translates into larger, better built wooden houses with greater attention to the use of decorative trim and house paint. A typical family owns a variety of higher quality household appliances, such as a sound system, gas stove, television, refrigerator, sewing machine and occasionally a freezer. Nearly all of the population is of local origin and equally prosperous, which is conducive to political mobilisation. Led by the capitalised small farmers along roads, pro-active political participation emerged in the riverine municipalities situated south and south-west of Manaus. More equitable economic success enhanced political participation, which, in turn, overcame bias in state development policy against small farmers, further fuelling economic success in an upward spiral. Within a national context of political decentralisation with the end of military rule in the 1980s, which devolved considerable development decisionmaking to states and municipalities, farmer-state relationships near Manaus were transformed, giving individuals and organised groups the possibility of directly interacting with government officials, so that local interests could be incorporated into policy from the municipality level up. Greater political participation near Manaus is exceptional because it goes beyond the municipal level. Political participation evolved spontaneously, came to include multiple concerns, with a strong emphasis on local productive and economic interests and not just the provision of social services as occurs elsewhere in Amazonas State. Change was not immediate, nor spatially uniform, varying between and even within municipalities. This can be observed in Iranduba and Manacapuru counties and is related to the economic potential of different farming systems situated on and above the floodplain. At one extreme are the highly capitalised small farmers along roads of Iranduba, who earned the 2005 equivalent of US$39 981 in 2000 and have the most frequent and highest level of government contacts through which they deal with a variety of interests. At the other extreme are noncapitalised riverine farmers, who only earned US$768 in 2002 and are the least organised so that they only have occasional contact with government officials for social services. In between are the capitalised floodplain farmers who earned US$7833 in 2002 and politically interact with the road farmers (Figure 3). A.M. de Souza Mello Bicalho and S.W. Hoefle: Economic Development, Social Identity and Community Empowerment 287 © 2009 The Authors Journal compilation © 2009 Institute of Australian Geographers The case of political participation near Manaus shows that devolving some political power to state and municipal government can permit greater local input into decision-making but this only happens if the local players have the economic power to make themselves heard. However, full political empowerment will not exist in the Central Amazon without the economic empowerment of the majority of farmers of the state and not just a relatively small number of capitalised farmers located near Manaus. Positive Lesson 3: to overcome socio-spatial disparity all farmers need to be pro-active, to participate in local decision-making and to have channels to higher scales of government. However, the intensive farming methods, which enable economic success, cause problems with ecological sustainability. If on one hand, farmers concentrate their efforts in relatively small areas and abandon extensive cattle raising, on the other, conventional modern methods involve the use of chemical fertiliser, pesticides and irrigation in covered planting areas, which can cause contamination of soil and run-off water, lead to soil salinisation, if the land is not uncovered and fallowed, and represent a health risk for farmers and consumers. Negative Lesson 3: economic and political sustainability do not necessarily lead to ecological sustainability. Modest capitalisation, significant market articulation and high empowerment in accessible riverine areas Highly favourable trends for political participation are also present in riverine farming along the Madeira River, which in recent decades has become more commercial in outlook but has experienced limited technical capitalisation. In contrast to many settlers along roads, riverine farmers have long been present in the Amazon, constituting a historical mestizo peasantry, which was the product of cultural fusing between acculturated Amerindians and Brazilian peasants who settled in the areas over the centuries. Consequently, in a basically riverine municipality like Manicoré, located in the middle of the Madeira Valley, there are almost no outsiders and the population has low rates of migratory mobility. Interviewed riverine farmers of the Madeira usually do not use modern agro-chemicals or Mayor Municipal Technicians & Officials Mayor Municipal Technicians & Officials Capitalised Road Farmers Capitalised Riverine Farmers Non-capitalised Riverine Farmers Type of Contact Frequency of Contact technical and economic contact frequent contact social contact occasional contact sporadic contact F Local Representative Mayor Municipal Technicians & Officials State Federal F F F Local Representative Local Representative F F F F F F F F Figure 3 Farmers politically scaling up near Manaus. 288 Geographical Research • August 2010 • 48(3):281–296 © 2009 The Authors Journal compilation © 2009 Institute of Australian Geographers machinery. This is especially true of farming above the floodplain along black water tributaries, where manioc is about the only crop that will grow in the relatively poor soils encountered there. Manioc is highly resilient to the equatorial environmental conditions and the only capitalised input is the rare use of insecticide. While the main product is manioc flour, the traditional basic foodstuff of the Amazon, and cropping areas are limited to 3.3 ha on average, a considerable part of the production is commercialised, so dispelling the notion of subsistence-minded riverine peasants. The Madeira, like the Amazon, is a white water river with rich alluvial floodplain soils, and as such, are the focus of cropping activities. There are two kinds of floodplain farmer: noncapitalised and capitalised. On average, noncapitalised farmers plant 1.5 ha of crops while capitalised farmers plant 4.1 ha. For noncapitalised farmers the main cash crop is bananas but manioc flour is also an important product while capitalised farmers plant the same crops as well as watermelons for the Manaus market and some vegetables for local urban markets. In social terms, monetary income was limited to the 2005 equivalent of US$1852 per annum in 2002 for peasant producers compared with US$3874 for capitalised floodplain farmers. Farmers live in simple houses made of wood, though larger than those of farmers who plant basic foodstuffs along roads. Riverine farmers also have more household appliances such as a radio, a television, a gas stove and a sewing machine. In addition to chicken and pig production, abundant fishing resources are available, so that the diet of riverine farmers is better than that of poor farmers along roads. If limited monetary income does not afford a better lifestyle, political mobilisation does, and this occurs independently of the degree of capitalisation. Indeed, some of the best mobilised communities are located in black-water, off-the- floodplain farming areas, which again cautions against interpreting too linearly the relationship between economic development and political empowerment. Consequently, public services in the form of schools, health clinics, electric generators and community buildings in the rural zone are far better in riverine municipalities than in frontier municipalities. Riverine communities consist of kin and long-time friends rather than immigrant strangers so that community members have a good deal in common, which favours greater political mobilisation. Lifestyles may be simple but have improved significantly over time so that 62% of the farmers stated that they were better off today than 15 years ago and migration trends support their perception. Nearly twothirds of the interviewed farmers of Manicoré have never moved and none five times or more (the common situation along the frontier). As a result, community associations pressure municipal governments successfully to build community centres, to open rural schools and health clinics as well as to provide other amenities such as community electric generators, water towers, public telephones and even all-weather football fields (a basic necessity in Brazil). Municipal-level politicians have adapted to the new communitarian politics by trying to tie furnishing public services and community buildings to votes. State deputies also try to do this by intermediating sustainable development projects. On the positive side, the Amazonas state rural development agency IDAM (Instituto para o Desenvolvimento Agropecuário do Estado do Amazonas) uses community structures to donate farm equipment and provide water and electric services as well as to channel crop loans and technical courses through community associations. This is an interesting way to try to overcome the classic largeholder bias in Brazilian rural credit and farm extension but when there is no viable economic activity available, as in the cases of Silves and Benjamin Constant municipalities discussed below, the result is the movement of people into rural slums where they are easy prey to the new clientelist politics. This has not happened in the Madeira valley, where communities have expanded commercial production and politicians have not been successful in creating dependencies. Politicians must now deal with communities united in associations whose president is usually an elderly relative of all of the inhabitants. This is particularly the case in riverine communities, which grew more from natural demographic increase than from massive in-migration of strangers, as occurs along highways. When confronted by unified communities, politicians now must negotiate the provision of goods and services and not just demand votes in elections, as was the case in previous, top-down patronage networks of the past. Positive Lesson 4: socially homogeneous communities which have experienced more equitable development are better able to pressure municipal government for services. A.M. de Souza Mello Bicalho and S.W. Hoefle: Economic Development, Social Identity and Community Empowerment 289 © 2009 The Authors Journal compilation © 2009 Institute of Australian Geographers Economic stagnation and political clientelism in distant riverine peasant farming In sharp contrast to the area of metropolitan agriculture near Manaus and even to the semicommercial riverine farming of the Madeira, the distant upper Amazon is one of the poorest areas of the Brazilian Amazon and is the object of new state and church development policy emphasising greater scale of production. Over the last ten years, in the face of under-employment caused by the banning of unsustainable logging, the limited size of local urban markets and the great distance from the regional consumer markets, riverine peasants fell back into subsistence farming and fishing thereby putting pressure on local resources. In this immense area there are four small urban areas: Benjamin Constant (14 158 in 2000) and Tabatinga (26 539 in 2000) in Brazil, Leticia (23 883 in 2007) in Colombia and Islandia (2152 in 2005) in Peru. This area is ten to twelve days distance by boat from Manaus, the principal consumer market of the Western Amazon, and as such, farmers cannot send even the least perishable crops like bananas to this market. There are a few capitalised fruit and vegetable producers on islands located close to these small cities but the local consumer market is so limited that they are able to meet the limited local demand. Other farmers located beyond these islands have great difficulty in finding a buyer for their produce and they also run up against problems with the lack of good quality soil. In the upper basin of the Amazon, floodplain lands suitable for agriculture are in short supply, basically being restricted to alluvial islands. Vegetables will only grow on the floodplains and will not grow directly in the highly acid soil away from the floodplain. Regional fruit trees grow off the floodplain but farmers complain that there are no buyers for their produce so they end up planting a large number of different food crops and fruit trees for self-provisioning and only sell manioc flour, which can make the long trip to distant markets. Even fishing is limited due to the presence of Colombian commercial trawlers which ply the main stream of the Amazon and ship fish to Bogotá by airplane. An international compromise was reached whereby commercial trawlers stay out of the lakes which are fished by riverine peasants and Amerindians. However, even here, older fishers complain that after the ban on logging, ex-workers turned to fishing, increasing the competition and reducing stocks. Negative Lesson 4: banning an environmentally damaging activity causes unemployment if alternative activities are not introduced at the same time, thereby putting pressure on other natural resources. Consequently, monetary income and living standards are much lower than elsewhere in the Amazon. Peasant producers earned on average US$363 a year in 2005 and 30% of them had no monetary income from farming at all. This kind of person is easy prey for the classic clientelist politics of the Amazon in which a vote is exchanged for basic farm tools and clothes. In response to this situation, the new Catholic bishop of Tabatinga moved beyond Liberation Theology church policy focused on inducing political mobilisation to that of promoting local development through introducing greater economies of scale in farm production and marketing. At the same time, the state government of Amazonas loudly proclaimed the area to be a priority for improvement in infrastructure and provision of social services. However, the results of these top-down strategies have been modest. Chilling fish and sending it to Manaus has its limits as does increased production of manioc flour. The bureaucratic implementation of new state services has been painfully slow and has run into traditional rivalries between the different municipalities of the area. During fieldwork people complained that, unlike the previous governor, who had distributed tools and clothes in a clientelist way, they had received absolutely nothing from the so-called progressive new governor nor had he even visited the area. Negative lesson 5: collective production schemes and the provision of government services meant to improve human capital are of little benefit if markets for commercial production are not available. In addition to these problems, there is an utter lack of political union among what would appear to be a highly homogeneous riverine population. All interviewed farmers had been born and raised locally but communities are separated or fragmented internally according to ethnic group (Amerindian, mestizo historic peasant), nationality and language (Brazilian Portuguese speakers and Colombian and Peruvian Spanish speakers) and religious affiliation (Catholic, Protestant and Cruzista) (Cruzista is a local dissident, fundamentalist Catholic movement). The result is that few schools are found in the rural zone, though there are some community centres, but people must go to the municipal seat to obtain health 290 Geographical Research • August 2010 • 48(3):281–296 © 2009 The Authors Journal compilation © 2009 Institute of Australian Geographers care. Negative Lesson 6: so-called traditional riverine populations are not always socially homogeneous and inclined to political mobilisation. The contradictions of multi-scalar politics in the Amazon In this section, we will see how difficult it is to conciliate the different dimensions of rural sustainability at different scales of political action in a very diverse region. In other words, ‘synergy’ and ‘local development’ are often easier said than done in the Amazon. Contradictions in religious multi-scalar politics The silent revolution of communitarian politics in riverine Amazonia outlined above was encouraged by the Brazilian Catholic Church as part of the general policy of Liberation Theology, introduced after the Second Ecumenical Council of the Vatican (1962–1965) and the Bishop’s Conference in Medellin, Colombia (1968), which advocated a policy in favour of the poor and disadvantaged. Since the sixteenth and seventeenth centuries, the Church has undertaken missionary activity in the Central and Western Amazon. Today the Franciscans, the Priests of Scarborough and the Silesians work in areas of old riverine settlement and new colonisation along roads. The orders are hierarchically organised and priests are present in all municipalities of the region. Some priests may limit their actions to catechism but most engage in different forms of social assistance, particularly in education and health services provided in the municipal seat and in their rounds through the countryside. Special attention is also given to young people and to adults in their problems with alcoholism and drug addiction, which are increasing in the region and are one of the principal causes of violence. Despite these social services, the Catholic Church is losing members to Pentecostal Protestantism, a process that dates from at least the 1970s when Moran (1981, 176–177) noted the presence of Protestants from the South in the colonisation projects along the Transamazonian Highway. Pentecostal groups, and in particular Jehovah’s Witnesses, are expanding rapidly among both riverine peasants and new settlers, so that Protestants make up 12% of the population of the North, putting it together with the South, as the regions where Protestants are most numerous in Brazil (FIBGE, 2000). As occurs elsewhere in Brazil and Latin America, rite in Pentecostal Protestantism is highly decentralised in both structural and spatial terms. The active participation of lay members of any social standing is encouraged, which attracts a large following among the poor (cf. Martin, 1990; Stoll, 1990; Bastian, 1993). Churches are located where the poor live on the periphery of the towns and in the riverine and highway communities in the countryside. Religious services can be attended on a regular basis in the rural zone where Catholic priests are only able to celebrate mass once or twice a year in their annual round through municipalities whose size ranges from 30 000 to 90 000 km2 . Consequently, Catholic communities are only able to resist Protestant inroads where lay services are well organised and are consistently offered. This notwithstanding, the same characteristics, which highlight the greater ideological sustainability of Pentecostal Protestantism vis-à-vis hierarchical Catholicism, interfere with political sustainability on the Brazilian frontier. The Protestant churches are highly independent of one another so that their political horizon is restricted and common action on a regional basis is circumscribed. Protestants usually hold the view that religion is confined to rite in the church and to personal mores so that they rarely participate actively in changing the world through politics. Specific Protestant congregations may elect a councilperson at the municipal level but rarely does such an elected official pursue social actions with a view to the whole community. Negative Lesson 7: ideological sustainability does not necessarily lead to political sustainability. The Catholic Church, on the other hand, takes an integrated view of religion and community so that most priests do not limit their work to providing only religious services nor stop at undertaking social work but also promote the organisation of communities and their political mobilisation in order to better pressure government for basic services and equitable development, a view reafirmed by the Congregation of Brazilian Bishops in 2008, which devoted the year to the theme of ethics in politics as a way to mark the fortieth anniversary of Medellin. The strongly hierarchical structure of the Church, ironically, has been a great aid in working at the various scales of action in contemporary Brazilian politics. At the same time priests and bishops have been sensitive to deferring to local interests, even when they disagree, such as economic development versus environmental preservation, A.M. de Souza Mello Bicalho and S.W. Hoefle: Economic Development, Social Identity and Community Empowerment 291 © 2009 The Authors Journal compilation © 2009 Institute of Australian Geographers and have been quite perceptive in identifying the moment when they should step aside and let a community movement ‘walk on its own two feet’. However, having said this, the grassroots politics, which emerged at the municipal level in the Amazon, was not just a product of Church Liberation Theology but was also the result of the population coming together in communities with the end of rubber tapping in the 1960s. It would be better to talk of a dovetail fit between formal and popular religion interacting with community building. Contradictions in non-governmental multi-scalar politics Given the size of its territory and the scope of its social and environmental problems, Brazil occupies a prominent place in the activities of home-grown and foreign non-governmental organisations and foundations. However, few NGOs are active in the study area and instead concentrate their efforts behind the frontier in a reactive manner. The MST (Movimento dos Trabalhadores Rurais Sem Terra) is one of the great agrarian movements of the world, meriting the cover plate of a recent collected work on the subject (Moyo and Yeros, 2005), but its activities are practically absent in the study area. In fact, the MST has a relatively low profile even in the violent Amazonian states where colonisation is more consolidated, such as Pará and Rondônia. In 1996 only 8% of the families involved in MST land invasions nationwide were located in the north (Petras, 1998), a figure which declined to 6% in 2003 (MST, 2005). This is, of course, consistent with the MST policy of promoting agrarian reform in long-settled regions of Brazil, rather than merely transferring the problem to the Amazon (Grzybowski, 1987; Gohn, 1997; Hammond, 1999). In the study area, the first land invasion occurred in 2000 in an area of public domain in Rio Preto da Eva, a municipality located near Manaus. However, the municipality already had a tradition of decades of spontaneous land invasions by individual farmers, and even by inhabitants from Manaus seeking to establish weekend homes, so that the MST action had little impact and is not even listed on the movement’s website (www.mst.org.br). Amazonas State still has extensive forests and the Madeira valley in the study area is the next large valley in line to be degraded by the advancing frontier so that one would expect a flurry of environmentalist NGO activity there. Unfortunately, outside perceptions as to placing and timing are out of synch with reality on the ground. Bouclet (2009) shows how NGOs are concentrated further east in Pará State, where he encountered more than 200 organisations, and are not positioned pro-actively beyond the frontier in an attempt to set up sustainable rural livelihoods which could better resist its arrival. In our field research, only one nonAmerindian NGO was encountered in the Amazonas part of the Madeira Valley: the Associação de Serigueiros de Manicoré (ASM), which had existed for decades as a sort of rural worker union but was reorganised in 1992 by a local economist inspired by the example of the riverine peasant movements of Acre State. However, the ASM, unlike the famous rubber tapper associations of Acre and the Silves craft fisher movement described below, was not a political agent nor even an environmental agent, but instead concentrated on improving public social services. By imitating the rubber tapper associations of Acre in a superficial way, the ASM was able to receive funding from the Order of Malta, the Red Rock Dragons Ltd., the Found Bank and the governments of France, Italy and Saudi Arabia. These funds were used to buy a pickup truck for the president to drive around town (there are practically no other roads locally) and to purchase a number of small boats and vaccines for a health programme targeting the riverine population. In addition, the association distributed educational grants, supported sporting events and maintained a communally-owned chicken farm. The emphasis on social services of the type usually employed to garner votes is seen in another light by what provoked the demise of the NGO. When the director was not elected councilperson in 2000 he abandoned the organisation and moved to a neighbouring municipality. The NGO was a conventional political wolf in lamb’s clothes (which has become a major problem in Brazilian politics with NGOs serving as facades to conceal graft and corruption). Furthermore, as the ASM only worked in Maricoré, it displayed the typical NGO problem of limited spatial impact (Friedmann, 1992; Chambers, 1994; Ring et al., 1999). Negative Lesson 8: localised action does not necessarily lead to horizontal political mobilisation among the same social actors. Negative Lesson 9: localised action does not necessarily lead to sustainable development at the regional level. The relationship between local populations and NGOs also highlights the limits of bottom- 292 Geographical Research • August 2010 • 48(3):281–296 © 2009 The Authors Journal compilation © 2009 Institute of Australian Geographers up communitarian politics, which rarely go beyond the municipal level because beyond that point local actors do not possess the cultural repertoire (for example, levels of education, urban customs and knowledge of foreign languages) for dealing with higher scale political players. Consequently, centrifugal forces exist at the municipal level, which can divide local political allies along urban-rural lines and so sabotage even the most successful of social movements like the famous case of the Silves craft fisher movement. This movement arose in the 1980s as a means of resisting the incursion of large commercial fishing boats supplying the insatiable markets of Belém, Manaus and Santarém. Progressive clergy and laypersons of the Catholic Church helped the communities organise a municipal-level association, which effectively pressured local and later state and federal government for protection of their fishing grounds. The fishers were so successful that they attracted a good deal of outside attention, particularly during the UN Conference on Environment and Development held in Rio de Janeiro in 1992. Various international NGOs and foreign governments provided millions of dollars in aid and the movement split in two: a naïve environmental movement controlled by local and outside urban actors, whose objective was to maintain fishers as subsistence producers in order to preserve natural resources, and a farmer movement controlled by local rural people, which sought financing for what were supposed to be environmentally benign, commercial agro-forestry systems. The resulting disarticulation (of scales) was a disaster: the environmental movement turned into a make-work project, which is now going through the throes of going cold turkey financially, and the farmer movement became heavily indebted as agro-forestry systems failed due to over-specialisation in a few products which led to crop disease and produce prices to plummet. In addition to these limits to grassroots communitarian democracy, within the complex political processes of the Amazon, international NGOs often become frustrated with the intermediary scales of power and try to go directly to the aid of local communities. No doubt, like that which happens on historical and contemporary frontiers all over the world (cf. Rigg 1991; Colechester and Lohmann, 1995; Hine and Faragher, 2000), many national, regional and state-level players are committed to productivist and nationalist objectives and so act to stymie the efforts of foreign NGOs. However, simply circumventing these scales of power short-circuits the process because, in an environmentally and culturally complex world, international players often do not have the detailed knowledge to choose the right place, time and amount of funding to give, and end up throwing money at success story communities, which do not always use it appropriately, while the great majority of less visible communities go wanting. Negative Lesson 10: foreign activists do not have the detailed local environmental and cultural knowledge to implement regional policy and are attracted to highly publicised success stories which often do not need additional funding. Contradictions in ethnic-based multi-scalar politics From a non-local vantage point, the regional political movement the COIAB (Coordenação das Organizações Indígenas da Amazônia Brasileira), a consortium of Amerindian social movements of the Brazilian Amazon, has been a resounding success. Led by the acculturated Tukanos and Makuxis of Amazonas and Roraima states respectively and helped by an impressive array of Brazilian and foreign anthropologists, environmentalists and religious organisations, the different Amerindian groups of the Amazon formed the consortium in 1989. Today COIAB includes 75 organisations representing over 430 000 Amerindians, from 163 native peoples of the nine states, which comprise the Brazilian Amazon. COIAB, together with its different political allies situated at all scales of political action, has been highly successful at setting aside enormous areas of the Amazon for reservations and at attracting funding for health and community development programmes. Representatives are regularly called upon to testify before state and federal congressional committees, to participate in ministry commissions and to attend all of the important world environmental and non-governmental association events (COIAB, 1991–1998; 2008; Survival International, 2000). However, COIAB leaders may have become too successful in scaling up. Some leaders have become globetrotters and political allies grumble that the leadership living in town and driving around in expensive utility vehicles may be losing contact with the reservation communities. This could be called the Kayapó Syndrome, named after the Amerindian group of the eastern Amazon whose most famous leader, Raoni, A.M. de Souza Mello Bicalho and S.W. Hoefle: Economic Development, Social Identity and Community Empowerment 293 © 2009 The Authors Journal compilation © 2009 Institute of Australian Geographers toured the world with the rock star Sting in fund raising drives for native people and environmental causes. It is ironic that environmentalists espousing egalitarian politics should help create such ‘chiefs’ among Amerindians for the first time since the Conquest destroyed the chiefdoms of the Amazon centuries ago (cf. Lathrap, 1970; Hornborg, 2005). However, this really only follows a long history of Europeans strengthening political hierarchies because they wanted to deal with (and co-opt) one representative and not become involved with the messy business of tribal consensus politics (Wolf, 1982). The Amerindian-environmentalist alliance is also suffering antagonism on the ecological front. Ecological reserves and Amerindian reservations were once thought to be the same thing (including by one of the present authors who defended such a view with regard to the implementation of a Yanomamö-Ye’cuana reservation in 1975). According to this line of thought, because Amerindians of the Amazon have a tribal culture with an ecologically friendly cosmology and an egalitarian economy, they do not provoke environmental degradation. However, today, few Amazonian Amerindians have such an idealised form of tribal culture. In states further along the process of consolidating settlement, such as Pará and Rondônia, threats and outright violence as well as acculturation and economic incentives have given rise to complaints of Amerindians being forced to permit or becoming directly involved in lumber, diamond and gold extraction on their reservations. The Tikuna Amerindians of the Benjamin Constant-Tabatinga area display such scalar contradictions. These Amerindians still speak their native tongue and learn Portuguese as a second language, retain some social customs and myths but were catechised Catholics long ago. To the outside world they were united in the quest to set aside areas for a reservation and they actively participate in COIAB. However, at the local level, they are splintered into factions representing different communities who practise winnertakes-all, clientelist politics, not much different from their peasant neighbours. Political fragmentation is exacerbated by differing religious affiliations whereby whole communities are Catholic, Pentecostal Protestants or Cruzistas to the point that members of different faiths can be pressured to leave communities. Similarly, Amerindian farming systems also vary, some being similar to their peasant neighbours, who too can have communitarian land tenure, while others have adopted more capitalised systems and have attained a prosperous standard of living far superior to that of non-capitalised Amerindians and peasants. Furthermore, the Tikunas are far from being ecologically noble savages and undertake high levels of deforestation on their land and capitalised Amerindian farmers use agrochemicals. Negative Lesson 11: Amerindian political sustainability does not necessarily lead to local political sustainability or to ecological sustainability. Positive and negative lessons for political empowerment and regional development in the Amazon Many of the positive and negative lessons for community development cited in this article, each taken on its own, may seem obvious, but the complex relationships between different dimensions of local development and scales of political action in the Amazon present in the lessons are not. Successful community development in the Amazon depends on: 1) local ecological potential for commercial and capitalised farming; 2) equitable economic development in order to maintain a level social field which is essential for grassroots democracy as well as to create the visibility necessary for attracting political actors from other scales of action; and 3) long-standing kinship and friendship relations which are conducive to political union. Instead of ‘synergy’ occurring from this, specific combinations of ecological, economic, political and ideological influences can work at cross purposes, annulling the overall process. The same can occur with scales of political action working against one another. Success at the communitarian and municipal level can run up against clientelist politics and productivist development policy at the state and national levels, provoking transnational environmentalist actors to bypass these scales of action and go directly to the aid of highly publicised success story communities in the mistaken assumption that their positive experience will automatically diffuse to other communities. We have shown that environmentally benign and equitable development occurs when longstanding social networks and local knowledge are politically mobilised and when new economic activities with market access are introduced. The so-called ‘peoples of the forest’ are not noble savages or subsistence-minded peasants (also see Stearman, 1994; Nugent, 2002). Limited economic opportunity and shared 294 Geographical Research • August 2010 • 48(3):281–296 © 2009 The Authors Journal compilation © 2009 Institute of Australian Geographers poverty should not be confused with the absence of a market mentality and the lack of consumer aspirations. In fact, economic success and common social identity enhance political participation, which, in turn, overcomes bias in state development policy, further enhancing economic success in an upward spiral. However, on a regional basis, only the Amerindians have managed to effectively scale upward to higher levels of political participation, albeit with a cost to grassroots democracy and with no guarantee of ecological sustainability. Similarly, the rise of enclaves of pro-active farming communities and local development do not necessarily lead to regional economic development and political empowerment. Within municipalities, rural development is uneven and communities are played off against one another by politicians. The same thing happens to municipalities at the state level so that much still needs to be done to break through the stone wall of productivist interest groups and clientelist state and national politics in Brazil. ACKNOWLEDGEMENTS Research funded by the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq – Brazil) and the Institut pour la Recherche de Développement (IRD – France). Over the years logistical support for fieldwork was provided by the Instituto de Desenvolvimento Agropecuário e Florestal Sustentável do Estado do Amazonas (IDAM), the Instituto Nacional de Pesquisas da Amazônia (INPA) and the Exército Brasileiro. REFERENCES Bakkx, K., 1988: From proletarian to peasant: Acre 1870– 1986. Journal of Development Studies 24, 141–160. Bastian, J.-P., 1993: The metamorphosis of Latin American Protestant groups. Latin American Research Review 28, 33–62. Bouclet, B., 2009 (in press): Organizações nãogovernamentais na Amazônia Oriental. In Araújo, R. and Lenas, P. (eds) Alternativas de Desenvolvimento Sustentável e Tendências de Mobilidade Sócio-Espacial na Amazônia. Museu Goeldi, Belém. Chambers, R., 1994: Challenging the Professions. Intermediate Technology, London. COIAB (Coordenação das Organizações Indígenas da Amazônia Brasileira), 1991–1998: Informativo da COIAB 2–10. COIAB (Coordenação das Organizações Indígenas da Amazônia Brasileira), 2008: Quem somos nos. www. coiab.com.br. Colechester, M. and Lohmann, L., (eds), 1995: The Struggle for Land and the Fate of the Forests. World Rainforest Movement/Zed, Penang. Escobar, A., 2007: The ‘ontological turn’ in social theory. Transactions of the Institute of British Geographers 32, 106–111. Foweraker, J., 1981: The Struggle for Land. Cambridge University Press, Cambridge. Friedmann, J., 1992: Empowerment. Blackwell, Oxford. FIBGE (Fundação Instituto Brasileiro de Geografia e Estatística), 2000: Recenseamento Geral do Brasil: Censo Demográfico – 2000. Rio de Janeiro. Gohn, M.G., 1997: Os Sem-terra, ONGs e Cidadania. Cortez, São Paulo. Grzybowski, C., 1987: Caminhos e Descaminhos dos Movimentos Sociais no Campo. Vozes, Petrópolis. Hammond, J.L., 1999: Law and disorder: the Brazilian landless farmworkers’ movement. Bulletin of Latin American Research 18(4), 469–489. Hine, R.V. and Faragher, J.M., 2000: The American West. Yale University Press, New Haven. Hoefle, S.W., 2000: Patronage and empowerment in the Central Amazon. Bulletin of Latin American Research 19, 479–499. Hoefle, S.W., 2006a: Eliminating scale and killing the goose that laid the golden egg? Transactions of the Institute of British Geographers 31, 238–243. Hoefle, S.W., 2006b: L’empowerment politique et la construction de communauté en l’Amazonie Centrale. Cahiers du Brésil Contemporain 61/62, 45–72. Hornborg, A., 2005: Ethnogenesis, regional integration, and ecology in prehistoric Amazonia. Current Anthropology 46, 589–620. Ianni, O., 1979: Colonização e Contra Reforma Agrária na Amazônia. Vozes, Petrópolis. Jonas, A.E.G., 2006: Pro-scale: further reflections on the ‘scale debate’. Transactions of the Institute of British Geographers 31, 399–406. Jones III, J.P., Woodward, K. and Marston, S.A., 2007: Situating flatness. Transactions of the Institute of British Geographers 32, 264–276. Lathrap, D., 1970: The Upper Amazon. Thames and Hudson, London. Leitner, H. and Miller, B., 2007: Scale and the limitations of ontological debate: a commentary on Marston, Jones and Woodward. Transactions of the Institute of British Geographers 32, 116–125. Marston, S.A., Jones III, J.P. and Woodward, K., 2005: Human geography without scale. Transactions of the Institute of British Geographers 30, 416–432. Martin, D., 1990: Tongues of Fire. Blackwell, Oxford. Moran, E.F., 1981: Developing the Amazon. Indiana University Press, Bloomington. Moyo, S. and Yeros, P., (eds), 2005: Reclaiming the Land. Zed Books, London. MST (Movimentos dos Trabalhadores Rurais Sem Terra), 2005: Famílias acampadas. Retrieved 10 April, 2009 from 〈www.mst.com.br〉. Nugent, S., 2002: Whither o campesinato? Historical peasantries of Brazilian Amazonia, Journal of Peasant Studies 29, 162–189. Oliveira, A.U., 1999: A Geografia das Lutas no Campo. Contexto, São Paulo. Petras, J., 1998: The political and social basis of regional variation in land occupations in Brazil. Journal of Peasant Studies 25(4), 124–133. Rigg, J., 1991: Southeast Asia. Routledge, London. Ring, I., Klauer, B. and Wätzold, F., 1999: Towards regional sustainability. In Ring, I., Klauer, B., Wätzold, F. and Mansson, B. (eds) Regional Sustainability. Physica-Verlag, Heidelberg, 3–18. Schmink, M. and Wood, C.H., 1992: Contested Frontiers in Amazonia. Columbia University Press, New York. Simmons, C.S., Walker, R.T., Arima, E.Y., Aldrich, S.P. and Caldas, M.M., 2007: The Amazon land war in the south of A.M. de Souza Mello Bicalho and S.W. Hoefle: Economic Development, Social Identity and Community Empowerment 295 © 2009 The Authors Journal compilation © 2009 Institute of Australian Geographers Pará. Annals of the Association of American Geographers 97, 567–592. Stearman, A.M., 1994: Revisiting the myth of the ecologically noble savage in Amazonia. Culture and Agriculture 49, 2–6. Stoll, D., 1990: Is Latin America turning Protestant? University of California Press, Berkeley. Survival International, 2000: Disinherited. London. Wolf, E., 1982: Europe and the People without History. University of California Press, Berkeley. 296 Geographical Research • August 2010 • 48(3):281–296 © 2009 The Authors Journal compilation © 2009 Institute of Australian Geographers Copyright of Geographical Research is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.