1. Introduction

During the 1990s the environmental movement in Brazil, following an international trend, shifted from a “preservationist” to a “conservationist” approach. The key point of this change is that the conservation framework accepts that human activity and use of natural resources is not in principle incompatible with environmental protection, and might even be important means to promote it. This new framework of environmental protection was coupled with (and sometimes even encompassed by) the also emerging framework of sustainable development, proposed as a means to make socioeconomic development environmentally sound and socially inclusive. This integrative strategy fueled the enthusiasm of environmental NGOs, federal and state governments, and international development funding institutions (the World Bank, the Interamerican Development Bank [IDB], and bilateral agencies like US-AID). A variety of programs and projects emerged at national, regional, or local levels, aimed at promoting conservation and sustainable development in rural areas of the country where vulnerable natural environments and human populations (mostly poor, marginal to mainstream politics, economic development, and culture) co-existed. Although programs and projects varied immensely in the types of conservation and development (hereafter C&D) activities they promoted, they shared similar strategies of social mobilization and organization as the central process of implementation at the local level. The key guiding principles driving this strategy were those of what later was to be called empowerment, where inclusiveness and organized active participation at all levels of social, political, and economic processes, were expected to develop or strengthen a new type of citizenry. Empowerment was seen as a process of individual and collective learning supported by investment in human capital and in organizational capacity building.

This movement toward the integration of conservation and sustainable development was not exclusive to Brazil. Countries throughout the world's tropical belt, where high levels of biodiversity coincide with low socioeconomic development, experienced a similar trend. In the space of half a decade, or even less, C&D programs and projects became the key strategy adopted by environmental NGOs, by a variety of United Nations programs (mostly the United Nations Environment Program and the United Nations Development Program and the Commission for Sustainable Development), national governments (leading to the creation of agencies for environmental protection at various levels of government), and international funding institutions.

Not surprisingly, soon afterward the evaluation of the effectiveness of their results also became a key concern for all those involved, generating a variety of indicators of measurement and methods of monitoring and evaluation. Most evaluations and analyses have focused on the impacts of programs and projects in terms of their intended conservation and development objectives, with impacts in terms of social organization receiving attention in a more mechanistic fashion; that is, usually limited to the verification of quantitative aspects of group formation and participation in various activities (such as workshops, training courses, conferences, etc.) rather than analyzing processual and structural impacts. The lack of critical analysis of impacts on social organization and structure is surprising given that the key strategy that differentiates these C&D programs and projects from previous environmental protection and economic development projects is focused on changing processes of social organization and relations, such as creating and strengthening organizations, promoting participatory procedures of decision-making and conflict resolution, training local population, gaining mandate for local population to manage natural resources, democratizing access to governmental resources locally, among others.

The neglect to investigate C&D projects or programs through the lens of their impact on the structure and dynamics of social organization compromises the potential to understand what is, arguably, the most important overall goal of a C&D project: social change. The challenge to achieve economic viability of productive activities of rural populations in combination with the conservation and often the restoration of the natural environment is the main substantive goal of conservation and sustainable development initiatives. However, from the perspective of the concept of sustainability, it is also as important that these goals are accomplished for an extended time frame (one that accommodates for the “future generations” component of sustainability). In a world of increasing internationalization of markets, and the inherent variability of the natural environment, it is not farfetched to expect that a C&D activity (for instance the community management of forest through a management plan for sustainable logging), successfully carried out for a period of time, becomes either economically and/or environmentally unviable. The sustainability of conservation and development activities in the long run also depends on the ability of a population to withstand the variability of market and environmental conditions while maintaining C&D practices, even if it means a radical change in economic activity. This social sustainability is especially important because unlike their goals, C&D projects and programs are carried out in a limited time frame, after which funding and technical support to alternative activities ends. In other words, the maintenance of sustainable practices will likely depend on the commitment of local populations to the principles of sustainable C&D, as well as their ability to seek out the necessary support and resources to pursue their goals.

In my research I intend to study the impact of NGO-led C&D projects on the structure and processes of social organization in their target regions. More specifically, I want to investigate why these C&D projects have shown different degrees of impact at two levels: first, in establishing durable new forms and processes of social organization; second, in shifting the distribution of power and resources among sectors of their target population, toward engagement and identification with environmentally sound, economically viable, and socially just practices. While there are differences in results among different projects, my research is specially driven by the observation of different impacts within projects. My general expectation is that the degree of success in both cases depends on the extent to which the lead organizations in the project manage to mobilize other social actors, at a variety of organizational levels locally (in the direct area of project implementation) and non-locally, to join them in a collective effort to achieve and maintain the underlying principles of sustainability. This locates my research project within the analytical framework of social movements, even though most of the literature that has socio-political aspects of conservation and development initiatives characterizes them as elite and/or state-led programs designed to dominate rather than empower local poor populations. I will conduct the research through the comparative analysis of social networks in the region of three C&D projects in Brazil. Through this investigation I intend to begin to fill a gap in the understanding of the processes and impacts of an integrative approach to conservation and development, and thus contribute to the larger debate about environmental protection and sustainable development both among scholars and practitioners.

In the next sections I will develop my arguments in the following manner. In section 2 I make an overview of the literature of political ecology and its limitations in the analysis of the impacts of C&D projects in social structure and processes. In section 3 I elaborate on the adequacy of a social movement framework to conduct such analysis. In section 4 I present my research proposal, the cases I will study, and the comparative framework of analysis I intend to use. I conclude with some remarks about the relevance and possible policy implications of my study.

2. The Political Ecology Framework

Social, political, and cultural impacts of conservation and sustainable development programs have been critically studied mostly by anthropologists and geographers from within the field of political ecology. The central thesis of political ecology is that environmental degradation results from socioeconomic and political processes and structures of capitalist development. The origin of political ecology can be arguably attributed to Piers Blaikie's seminal book The Political Economy of Soil Erosion in Developing Countries (Blaikie 1985). Blaikiedemonstrates that soil degradation in rural Nepal, India, and other former colonies are due to development policies of post-colonial states that seek to benefit national and international elites at the expense of increased loss of entitlements (Sen 1981), vulnerability and poverty of peasant populations. For him, the interpretation of environmental change as a problem reflects a social perception of this change as negative to the political economy of capitalism in a global system of states. He argued that state programs to increase agricultural productivity through the integration of small farmers into capitalist agriculture pushes small farmers into a system of dispossession (inability to purchase inputs such as improved seeds, fertilizers, and pesticides, leads to debt and loss of land) and political exclusion (the state is controlled by the interests of national and international elites) that locks them in a cycle of progressive poverty and general marginalization, which in turn leads to degradation of natural resources beyond sustainability. Nowadays this argument might seem obvious to most students of C&D, but Blaikie's piece represented the first significant challenge to what he calls the “colonial” or classical technological interpretation of environmental degradation in favor of a political interpretation. The key consequence of this shift of approaches lies, from a policy standpoint, in the type of solution implied by each: while the technological approach required increased technological advancement and its diffusion to peasants through programs such as the “green revolution”, the political economic approach called for solutions that addressed social factors as much as environmental ones (perhaps even more).

This neo-Marxist approach is common to most of the work that can be subsumed under the label of political ecology until the end of the 1990s (Thompson, Warburton et al. 1986; Yapa 1993; Agrawal 1997; Bryant 1997; Bryant and Bailey c1997). This approach has two key characteristics that determine and limit its analysis of social processes and structures. First, social relations are marked by a dichotomy between state and peasants (or indigenous populations). The inclusion of environmental NGOs in this equation, a result of increased focus on conservation since the mid 1990s, has not substantially changed the polarized view of the relationship between state and civil society. To the extent that NGOs are international or affiliated with international organizations (through alliances or financing), they are considered, along with states, as interventionist outside forces impinging on local populations. Whether promoting industries that require natural resources owned and/or traditionally used by rural populations (such as state programs of agricultural expansion, logging concessions, etc.) or promoting the sustainable natural resource use or management of protected areas (for instance, non-governmental programs of forestry certification, development of management plans for the extraction of non-timber resources in reserves, or ICDPs), states, and (non-grassroots) environmental NGOs are motivated by the sectoral interest of elites and/or urban populations, national or international, in promoting development or appropriating local populations' resources (for instance, protected areas and sustainable natural resource use are necessary to guarantee the reproduction of the future generations of urban middle class, not of local populations living at or near the subsistence level).

The second characteristic emerged as a response to the discourse of development agencies, states, and natural scientists committed to environmental protection in relation to local populations and their knowledge about the natural environment from which they gained their living. According to this discourse, local populations were ignorant about their environment, therefore unable to manage it rationally and sustainably. A number of authors, mostly anthropologists, pointed out that this interpretation was based on an uncritical acceptance of positivism as the source of Truth (Watts 1983, among others; Sachs 1992; Fairhead and Leach 1996; Dove and Kammen 1997; Dove 1998). More than an academic debate over epistemology, political ecologists maintain that privileging positivist science and its technological products fosters the political project of cultural, political, and economic domination through techno-scientific knowledge. Moreover, as James C. Scott argued, the dogmatic acceptance of scientifically generated knowledge as the only basis for the design of sustainable natural resource management technologies is leading to the disappearance of local knowledge developed through numerous years of “practical experimentation”, or “mētis” (Scott 1998). Thompson, Warburton, and Hatley pointed out that researchers of social and natural sciences involved in conservation should seek and investigate uncertainty, rather than try to establish full knowledge about the environment and the relationship of human populations with it, which would no doubt make decision-making on management issues much more certain (Thompson, Warburton et al. 1986).

Political ecology has given an invaluable contribution to the study of conservation and development. It demonstrated that environmental problems are enmeshed in political, social, cultural, and economic processes at different scales (regional, national, international). Even the identification of environmental changes as problems derives from a political process. However, there are critical limitations for the application of either strand of political ecology to the analysis of the structure and processes of social organization. First, political ecology sees the state as an abstraction that contains the project of domination. For political ecology, the state is the archenemy of indigenous cultures and knowledge, and ultimately of real conservation and sustainable development. Second, the state and all other non-local actors are seen as homogeneous entities. Yet, the modern bureaucratic state is highly diverse with representations of multiple interests and decentralized, both geopolitically - national, regional, local levels - and institutionally. Furthermore, the state is made up for the most part of civil servants, people who are embedded in relationships other than those where they are identified as part of the “state”. There are great differences between the interests and loyalties of the head of an institution, elected or appointed by executive officials, based in the capital of a country, and the civil servants, usually local residents, who provide technical assistance to a rural community in an isolated area of the country. Thus, to the extent that state bureaucracies operate locally, there is necessarily an overlap between the state and civil society. Third, despite its focus on demonstrating the value and knowledge of local populations, especially indigenous populations, political ecology holds them in a subject position, without agency. In other words, it accepts the characterization employed by the state and NGOs of “target” populations or program “beneficiaries”. Perhaps more critically, political ecology denies local populations or parts of it the possibility of a legitimate claim to development and environmental protection: it is never the decision or wish of local populations to pursue C&D projects. Finally, political economy critiques the involvement of international agents in local matters through states, multilateral funding institutions, and NGOs, mainly by contrasting their ineffectiveness with the strength, knowledge, and self-reliance of local populations. This has strengthened a dichotomy between local and non-local political actors, even while scholars seek to break away from a “local” geographical bounding of the relationship of human populations with the environment (Bryant 2000).

In sum, political ecology reifies broad macro-level structures of class and power relations. Despite my sympathy toward this framework, motivated by the undeniable role of the state and elites in authoritarily imposing changes on populations with no or very little political and economic leverage, its oversimplifying dichotomous structure does not resonate with the empirical reality of the deployment of many NGO-led C&D projects. In the last years a number of scholars have recognized these discrepancies between the frameworks of political ecology and of poststructural development critiques and the social, political, and cultural processes observed by them. Critiques have focused on two aspects. First, scholars have questioned the “local” as a social unit and as a unit of analysis in the study of C&D. Among anthropologists, the notion of locality as a bounded, isolated geographical and social place is dispelled in favor of contested, shifting, culturally and historically constructed boundaries (Moore 1998; Tsing 1999; Moore 2000). In a more direct critique of much literature in neo-Marxist political ecology, the analytical treatment of populations in rural settings as a “community”, understood “as a small spatial unit, as a homogeneous social structure, and as shared norms” (Agrawal and Gibson 1999: 630) has been pointed out as one of the reasons behind a number of failed community-based natural resource management C&D projects . Rural populations are mobile (with various types of in- and out-migration taking place), which means social, cultural, political, and economic relations are continuously under pressure to change. They are also characterized by a diversity of interests, often conflictual, that are negotiated through some type of stratification (Rahnema 1992; Mosse 1997).

A second focus of critique is the representation of the state and non-local organizations as homogeneous agents of neo-colonial forms of domination (that is, conservation and development). This representation has been pervasive not only among neo-Marxist political ecologists, but also among poststructural critics of development such as Arturo Escobar and James Ferguson (Ferguson 1994; Escobar 1995; Escobar 1998). Despite their defense of local cultures and practices, these representations replicate the conceptualization of local populations as powerless objects of such domination, made homogeneous in their subjection. These representations of local populations devoid of agency in the process of C&D are rejected in recent critiques in favor of a process where projects from the state and other “outside” organizations are transformed, to the limits of their possibility, by rural populations to fit their practices and needs (Conklin 1997; Li 1999; Moore 2000). In turn, the notion of a homogeneous and abstract “state” that forces conservation and development on subject populations gives way to a multiplicity of diverse governmental organizations, that are as engaged in and transformed by the same process in which “local” populations are engaged in and are transformed by (for related arguments, see Brosius 1999; Moore 2000; Zerner 2001).

These recent critiques call for research that seeks to unveil the textures of multiple and trans-local social agents (or multi-sited research see Marcus 1995; Brosius 1999), but also to investigate the relational dynamics among these textured sites as social agents (which by no means implies that there is an equal distribution of power among them, or, in other words, that they are all equally autonomous to determine their choices and decisions). My research proposes to rise up to the challenge posed by these authors to devise a framework capable of investigating social relations among trans-local organizations and groups with different and multiple identities/interests, engaged in a common, though often loosely identified, project of social change. I believe such framework might be located within the field of social movements, to which I turn now.

3. Applying the Social Movement Framework to C&D Projects

Development has always been a state-led program, designed by advanced Western capitalist democracies and promoted by their funding institutions, to turn countries into capitalist economies, democratic political systems, and societies whose degree of standard of living is highly associated with a technological consumerist culture. Despite the principles of empowerment embraced by the concept of sustainable development, the process that led to the elaboration of the concept and its adoption by international funding organizations and national governments (see introduction) does not represent a shift toward more popular participation. In other words, sustainable development projects also appear to have been elaborated for grassroots groups, rather than by them . From the perspective of environmental protection of natural areas it has not been much different. Efforts to protect natural ecosystems, especially in tropical third world countries, have also been characterized by the mobilization of a highly educated middle class in NGOs or research institutes, supported by international pressures from a diversity of interests, leading to the creation by decree of protected areas and regulation of natural resource use by rural local populations. In short, taking a summarized definition of social movements as “sustained challenges against authorities on behalf of excluded groups that involve the use of non-institutionalized tactics at least part of the time” (Debra Minkoff, class handout, Spring 2001) C&D projects are unlikely candidates for a characterization as social movements.

There are three aspects of C&D projects that represent a difficult fit with such conceptualizations of social movements. First, C&D projects usually have the participation of government organizations or officials. These projects are not accurately described either as top-down, external programs or as grassroots bottom-up initiatives. Instead, they are both. The projects are constituted by grassroots groups, non-governmental organizations, governmental institutions, the business sector, local and non-local actors that are engaged in an initiative that mobilizes against the attitudes and practices of an equal variety of actors, both geopolitical and organizational. Second, challenging powerholders in the state or elites is not the ultimate goal of these projects, even though many important intermediate objectives might be focused on changing policy, regulations, legislation, or state structure at some level. A project's main objectives reflect desired changes in attitudes and practices, and its impacts are measured according to impacts on attitudes and practices. Third, projects are not oriented toward disruptive public forms of collective mobilization, even though different project participants might use various forms of protest at different stages, in different places, targeting different institutions or groups. Institutional strengthening of local groups (and often local government institutions), development of local human capital, techno-scientific assistance to conservation and development activities, and environmental education constitute the core of NGO-led C&D projects.

Nevertheless, as a process of social organization and change, C&D projects present similar elements as those analyzed in the study of social movements. The core of social movement literature has been developed in the United States and Western European countries, starting in the 1970s, to explain the emergence, dynamics, and more recently the impacts of social movements in their own societies. In other words, the concept of social movement was developed in order to explain a certain type of collective action that is characteristic of advanced capitalist democracies (Melucci and Lyyra 1998; McAdam 1999; Tarrow 1999). In the process of understanding popular movements of the 1960s and 1970s, scholars in the United States went from the theories of collective action as rational phenomena (Tilly 1978; Gamson 1990) to the framework of resource mobilization (McCarthy and Zald 1977; Freeman 1979; Jenkins 1983; Staggenborg 1991; Clemens 1993), to political process (Kitschelt 1986; Costain 1992; Morris 1992; Meyer and Staggenborg 1996; Tarrow 1996; McAdam 1999; Tarrow 1999), and to framing, identity, and culture (Snow, Rochford et al. 1986; Tarrow 1992; Snow and Oliver 1995; Gamson and Meyer 1996; Gamson 1998). By the late 1990s these various frameworks had been mostly integrated into one single framework, heavily focused on social movements as a political process. Sidney Tarrow locates social movements within a spectrum of forms of contentious politics:

“Not all of these [new forms of contention that have spread from one region of the world to another] warrant the term ‘social movement,' which I reserve for those sequences of contentious politics that are based on underlying social networks and resonant collective action frames, and which develop the capacity to maintain sustained challenges against powerful opponents. But all are part of the broader universe of contentious politics, which can emerge, on the one hand, from within institutions, and can expand, on the other, into revolution.” (1999: 2-3).

This general framework has been applied in the comparative study of social movements in advanced western societies (Mueller 1992; McAdam, McCarthy et al. 1996; Voss 1996; Giugni, McAdam et al. 1998; Giugni, McAdam et al. 1999).

Western European scholars, on the other hand, have provided their greatest contribution to the study of social movements by demonstrating the importance of cultural change as a goal of the movements in the last three decades of the century, which Melucci called “new social movements” (Melucci 1985; Offe 1985; Melucci 1989; Larana, Johnson et al. 1994). Although largely ignored by the literature, Melucci made a particularly important contribution to the analytical framework when he classified the various analytical elements of the study of collective action(Melucci 1996: ch. 1). Melucci argues that phenomena of collective action can be analytically classified along three axes: analysis should distinguish (a) “between a reaction to a crisis and the expression of a conflict” (p. 22); (b) “among different orientations of collective action”, involving solidarity among participants (or mutual “recognition” of a shared identity) or aggregation (p.23); and (c) among the “systems of relationships within which such action takes place and towards which it is directed” (p.25). Of special interest in the context of my research is his classification of four systems, according to the “specific types of relations that characterize them”: the system that comprises “the production, appropriation, and allocation of a society's basic resources”, the political system, the organizational system, and the lifeworld, or the reproductive system (p. 27). Under this analytical framework, we can expand our understanding of social movements to conceive of the existence of movements or at least collective action whose goals are not exclusively changes in policies, regulations, or legislation, that does not use public mobilization and disruption as its main strategy of action, and that does not assume challengers will be politically or socio-economically powerless while the challenged ones will be the state and elites.

The combination of Tarrow's elements of social movements and Melucci's analytical framework, provide, I believe, an adequate framework to study social change in C&D projects, focusing on the process of mobilization (including the creation of collective identities, frames, or solidarity), organizational structuring, and diffusion of both identity and social organization culture. I believe a synthetic definition of social movements was provided by Mario Diani:

Social movements are defined as networks of informal interactions between a plurality of individuals, groups and/or organizations, engaged in political or cultural conflicts, on the basis of shared collective identities. (1992: 13).

Two of these elements are of special importance for my research. As we observed earlier, collective identity has been considered the central issue behind collective action theory, and a key issue in social movement theory since David Snow and Robert Benford showed the importance of framing for movement mobilization and dynamics in the interaction with their “opponents” (Snow and Benford 1988). Following Melluci, collective identity is defined in a looser, perhaps fluid way, so that social agents in different sectors, regions, with sometimes-major disagreements or competing interests in some aspects, can still be linked through a process of mutual recognition, of solidarity. However, the existence of a shared collective identity does not imply or require that all social agents operate the same way in their efforts to achieve their goals. On the contrary, each agent operates according to their interests, characteristics (organizational, institutional, communal, etc.), and social location in the context of the specific movement (for instance, in the environmental movement a grassroots organization will operate in a very different way from a transnational NGO; see (Keck and Sikkink 1998)).

A second element in the analytical framework of collective action and social movements of critical importance for my research is the relational character of the interactions among social agents, individual or organizational through social networks (Friedman and McAdam 1992; Diani 1995; Gould 1995; Giugni 1998; Rucht 1999). Social networks have been studied in social movements research before, in a number of different contexts. Resource mobilization and political process approaches have focused on the role of social networks in recruitment for organizations, in mobilization for protest events, or in lobbying (Snow, Zurcher et al. 1980; Rosenthal, Fingrutd et al. 1985; McAdam 1986; Fernandez and McAdam 1988; Taylor 1989). A second approach, adopted by Diani in the definition above, focuses on the network character of social movements (Diani 1995; Gould 1995). A third application looks into networks as social agents (Keck and Sikkink 1998). Although the difference between these approaches might not be very significant, I believe it is important to point it out because the third approach indicates a reification of networks, where the focus shifts from relational interactions to the construction of a social agent. Indeed, this is not necessarily a misrepresentation, given that the number of formalized coalitions that are being called networks is rapidly increasing, both nationally and internationally (i.e. the Network of NGOs, in Brazil). Analytically, I believe it is important to keep this difference in mind, or we run the risk of treating networks as organizations and the organizations as the social movement . In my research I will follow Diani and focus on the network character of C&D projects.

4. Research Proposal

In this section I present my research proposal, starting with my research questions and main hypotheses, followed by a summary description of the cases I have been studying, and a discussion of the methodology I propose to follow.

Research questions and hypotheses

In my research I intend to study the impact of NGO-led C&D projects on the structure and processes of social organization in their target regions. More specifically, I want to investigate why these C&D projects have shown different degrees of impact at two levels: first, in establishing durable new forms and processes of social organization; second, in shifting the distribution of power and resources among sectors of their target population, toward engagement and identification with “environmentally sound, economically viable, and socially just practices” (that is, abridged version of the principles of sustainable development). While there are differences in results among different projects, my research is specially driven by the observation of different impacts within projects. My general expectation is that the degree of “success” in both cases depends on the extent to which the lead organizations in the project manage to mobilize other social actors, at a variety of organizational levels locally (in the direct area of project implementation) and beyond the region where the bulk of investments are made, to join them in a collective effort to achieve and maintain the underlying principles of sustainability. The investigation will focus on four hypotheses:

Hypothesis (1): To achieve its conservation or development goals, a project must create networks that link various social agents of different spheres, trans-locally, in flows of goods, information, services, and organizational experience.

Hypothesis (2): Viability and durability of transformed practices of natural resource use and production depends on the diffusion of identity from direct participants of the project to others in their other networks.

Hypothesis (3): The project as a process of social change depends on its ability to create new fora and forms of social organization that facilitate the establishment and diffusion of these new identities and practices.

Hypothesis (4): The creation of new fora and forms of social organization depends on the project's ability to mobilize existing identities, forms and processes of social organization to transform them or open space for the emergence of new identities and forms of social organization from within the existing ones.

These hypotheses will be verified through a comparative framework involving C&D projects in three regions of Brazil.

The case studies

I will conduct my research in three different regions of Brazil where NGO-led conservation and development projects have been carried out for over 5 years. Three of these projects are, to a greater or lesser extent, Integrated Conservation and Development Projects (ICDPs), the framework World Wide Fund for Nature (or World Wildlife Fund; or simply WWF) adopted to orient the design and implementation of C&D projects ICDPs are long-term demonstrative projects (that is, projects that seek results that can be generalized to other areas), usually (but not exclusively, as we will see in one of the cases I will study) implemented inside or in the periphery of environmental protected areas (for instance, a national park, a wilderness reserve, etc.) with the overarching goal of generating alternative and/or sustainable productive practices linked to the protection of nature. In ICDPs, non-governmental organizations (NGOs) and/or research institutes work with sectors of local communities and all relevant levels of government to implement conservation activities (i.e. ecological research, management of forests, protected areas, or fisheries), and development activities (i.e. palm heart production, ecotourism, handcrafts, sustainable logging, rubber extraction, etc.). In order to implement these two main lines of activities, ICDPs rely on a variety of cross-cutting activities designed to create the necessary social and political contexts to support the project, that is, increased local institutional and human capacity (i.e. creating and strengthening local groups of natural resource users and/or “managers”, training on new technologies or business practices, etc.), generating collective conservationist identity and behavior. The strategic framework of ICDPs is summarized in the diagram below:

Figure 1: ICDP Components at WWF-Brazil

Principle of Stakeholder Participation

Community Organizing

Monitoring and Evaluation

Training for local Community


Conservation & Development Activities


Environmental Education



Institutional Strengthening

Two of the ICDPs I will study (Silves and Várzea) are in the floodplain of the Amazonas river, called “várzea”, and one is in the cocoa region of the Atlantic Forest, northeastern coast of the country (Jupará). After my second visit for pre-dissertation research (late May to June, 2001), I included a fourth C&D project (IESB), in the same region of the Jupará project, but not a WWF Brazil ICDP. In the two projects in the floodplain the main conservation goal is to protect fisheries through the community management of the various floodplain lakes (not strictly speaking lakes, as during the rainy season they overflow and get connected to each other). Culturally, the population in both areas is called “ribeirinhos”, and constitute what in Brazil is considered a traditional population: they have lived in these areas for a long time (around a century, although there is no legal time requirement), they make their living from the management of natural resources, most of the land is not (or was not) titled, and they have a distinct culture, strongly marked by the environment where they live and their relationship with it. They are, however, located in two different states (respectively Amazonas and Pará) and have experienced different historical processes of social organization, economic and political development. On one hand, Silves is a municipality with a population of around 8,000, rather marginal to the economy and political composition of the Amazonas state. On the other hand, Santarém, the municipality where the Várzea project takes place, has a population of around 250,000 and is the second largest harbor in the state of Pará. It is located close to the western border of the state, at the confluence between the two largest rivers in the Amazon basin (Amazonas and Tapajós). These differences in population and geographic location have a major impact in the subsistence economy of fishing communities in both places (therefore affecting all activities directly related to environmental sustainability and economic viability of fishing). The relationship between the similar ICDP organizational strategy, which directs most investments to the selected target population in their focal area (working with one a small portion of the larger population of, for instance, fishing communities), and the socioeconomic characteristics of each region is a focal point of my interest in these two projects. I expect that trans-locality will have a special impact in project results, especially in the Silves project

As projects they also have considerably different trajectories, both institutionally and in terms of strategies. The Silves project was initiated and is run by a grassroots organization, while the Várzea project was initiated and is run by a research institute. In terms of strategy, the main development goal of the Silves project is to run a grassroots ecotourism operation, centralized on an ecotourism lodge (whose construction took up most of the funding), in order to generate resources for the conservation of the lakes, and consequently the restoration and management of fisheries for local populations. In turn, during its first years, the Várzea project focused on scientific research, environmental education, institutional strengthening of the municipality's fishermen association, and support for the diversification of agriculture. Finally, in Silves the local movement led to the creation of a municipal protected area, or lake reserve, while in region of project Várzea there is no protected area.

The third and fourth projects are located in the southern coast of the state of Bahia, with the largest close-by city being Ilhéus. Aside from the fact that these are also C&D projects, therefore following a similar core strategic framework, these projects could not be more different from the other two. This region was one of the first parts of the country to be colonized, and its history has been marked by very high social, economic, and political inequality along racial divisions (this is the state of Brazil that has the largest black population). Environmentally, this is a highly endangered part of the Atlantic forest, with only small remnants separated by farms. From the second half of the nineteenth century until the 1980s the region was occupied by a number of very large cocoa farms (some with well over 200,000 hectares). At the height of this period, cocoa production was one of the most financially successful businesses in the country, creating enormous wealth concentrated in the hands of very few white families. A contagious cocoa disease that renders trees unproductive (witch's broom disease, a cocoa tree fungus) has led to financial crisis and the efforts of third generation farm owners to sell the properties. For the last two decades this has also been one of the areas of Brazil where former farm workers and small landowners who lost their land have carried out the movement for agrarian reform. Even though cocoa farm owners in the last five years or so have been soliciting INCRA (the national institute responsible for the agrarian reform program in the country) to select their farms for expropriation, the history of this movement has been of violent conflict.

In this region, the research will be carried out with projects of two different lead organizations in the same municipality. Although they follow similar strategies in terms of social organization (even though the methods they use in the mobilization of people is different), their beliefs, conservation objectives, and development initiatives are different. This will allow me to compare the importance, if any, of those differences on the structure and process of social organization in the region. Although the projects do not work with the same communities in the municipality (Una), in both cases the communities involved are located in the area surrounding a biological reserve managed by the federal government. The Jupará project in the region was initiated by a technical grassroots organization, that is, formed by local people, the majority landless farm workers or peasants, who have had professional training in agricultural extension. Their institutional goal is to provide technical assistance and political support to official or still contested settlements to develop sustainable agriculture and agro-forestry. The IESB project in the region is a research institute focused on biological or ecological research and conservation. Their primary goal in the municipality of Una is the effective implementation and long term viability (as a protected area) of the biological reserve (which carries the name of the municipality). Both projects are involved in conservation and development activities.

Despite the differences between the four projects, especially between the northern and northeastern regions, they have all followed very similar strategies of social organization. Aside from being at least 5 years old, the selection of these projects was motivated primarily by a combination of differences/similarities among the four projects, which is conducive to different types of comparisons among them. The two projects in the Amazonian floodplain share a number of similarities (environmental, type of population, life-style, economic activity, historical) but are managed by very different organizations (grassroots vs. scientific institute), are in very different socio-economic situation (see brief explanation above), and focus on different development strategies (ecotourism hotel vs. fishery management, and diversified agriculture). In both cases the progressive Catholic Church has had a very significant social impact, but also a very different one (organization of communities vs. literacy). Although it can be said that the movement to manage fisheries through community fishing agreements and zoning of lakes for preservation, subsistence, or commercial fishing is a grassroots movement, in the Várzea project the managing NGO (non-grassroots) took the lead in the organizing process since the project's inception

In the northeastern projects there is an even tighter control for socio-economic, environmental, and demographic variables, as both projects are carried out in the same municipality. The differences in objectives of the two organizations, as well as the different organizational profiles, have led not only to a different perception of who the preferred target groups for mobilization are (IESB works with both small and large land-owners, while Jupará works strictly with small poor farmers), but also in their strategy of framing (IESB considers its work a techno-scientific conservation project, while Jupará is involved in a political project), and mobilization (IESB identifies communities or farmers to work with according to their proximity and importance to the biological reserve, while Jupará identifies communities mostly by their involvement with the agrarian reform movement or are identified by traditional communities of small farmers). Given that the two projects operate in the same small region, it is not surprising to see that their paths cross. It is however interesting to see that they have not been able to establish a partnership and have had a number of conflicts.

Other factors also influenced the decision to choose these projects: they are located in ecosystems considered priority for conservation; the project participants have different interpretations of the results they achieved or failed to achieve, but in the four cases a key element in the explanations is social organization. Lastly, two factors related to my work with WWF-Brazil: I have good knowledge of three of the projects (in the case of Silves, I was project supervisor for almost 4 years) and throughout my experience with WWF-Brazil I developed good relationships with the project leaders, if not participants. Both these factors have facilitated access to the projects and their participants. Although I also had some degree of knowledge about the fourth project (IESB), my decision to add their work to my comparative framework came during my conversations and interview with a few of their staff members during my second pre-dissertation visit to the region, this summer.

Methodology: Relational data and social network analysis

A relational approach to the identification of the agents of C&D projects allows for the study of social organization from a frame not constrained by traditional boundaries in social, political, and economic relations: boundaries of class distinction, between rural and urban, local and non-local populations, between market and subsistence economies, and, especially important in the case of conservation and development initiatives, between social relations and the environment. This is possible by defining the systems to be studied as the projects from the perspective of their lead organizations. The unit of analysis becomes the social agents that are directly or indirectly involved or affected by the projects. The goal is to understand the structure and process of social interactions among these social agents.

By searching for a method that allows the crossing of boundaries in the study of C&D I do not intend, by any means, to challenge the reality of these boundaries and sometimes rather rigid stratification of individuals and social groups in any of these dimensions. On the contrary, these boundaries are perhaps more dramatically evident in C&D projects than in most other collective action contexts; after all, the challenge of a C&D project lies in the extremes it seeks to - and to different degrees do - pull together: the illiterate local population, the institute for scientific research, the federal agency for environmental protection, the local council for the environment (with mixed participation of government and civil society), the international tour operator, the local fish market, the NGOs and the grassroots groups, the technological and the practical knowledge(that sometimes becomes the encounter between the faith in science and the faith in God), and so many others. In fact, the ways these boundaries are transformed, reproduced, rebuilt, created, or even remain unchanged through the intervention of a C&D project is at the core of my research.

The character of the difference I am trying to highlight between these two ways of approaching the study of social processes is methodological. John Scott observes that there are two main types of data in social science: attribute and relational

Attribute data relate to the attitudes, opinions and behavior of agents, in so far as these are regarded as the properties, qualities or characteristics which belong to them as individuals or groups. (…) The methods appropriate to attribute data are those of variable analysis, whereby attributes are measured as values of particular variables (income, occupation, education, etc.).

Relational data, on the other hand, are the contacts, ties and connections, the group attachments and meetings, which relate one agent to another and so cannot be reduced to the properties of individual agents themselves. Relations are not the properties of agents, but the systems of agents; these relations connect pairs of agents into larger relational systems. The methods appropriate to relational data are those of network analysis, whereby the relations are treated as expressing the linkages which run between agents (Scott 1991: 2-3, emphasis in the original).

I my research I will focus on the study of relational data. The various social agents (individuals, families or households, informal groups, governmental and non-governmental organizations, firms, churches, etc.) that participate in or are affected by a C&D project, as social agents in any circumstance, are part of multiple networks associated with different dimensions of their lives: kinship, economic, political, religious, neighborhood, environmental, special interests, organizational, etc.. These networks overlap to different degrees and through different types of social agents. However, even if there is great overlap among different networks, each one is still characterized by some shared identity that is prevalent above all others, usually derived from the core motivation that led to the constitution of the network. This also means that networks are not by definition closed systems or systems of homogeneous social agents. On the contrary, variations both in the intensity of different links in the network, and in the level of overlapping identities are key elements in the analysis of networks, as Mark Granovetter pointed out in his seminal work on the role of weak ties in networks (Granovetter 1973).

From the perspective of a project that seeks to mobilize collective action for social change, social agent has a variety of identities and participates in a variety of networks. Unless the membership of all networks is completely coincidental, as social agents go from one network to the other, there is a potential process of diffusion of information, experiences, and resources from one to the other. In his study of the French commune of 1871 Roger Gould demonstrated how the mobilization that led to the commune was not that of class, as usually interpreted, but of neighborhood (Gould 1995). He argued that mobilization by organizations is successful when they manage to identify these prevalent identities and mobilize around them, creating a “participation identity” (1995: 13) related to the motivation for the collective action. Although his piece does not center on explaining social change, but rather the emergence and mobilization dynamics of a historical event of collective action, his analysis opens the way for us to understand that through networks and this process of diffusion and mobilization, new collective identities and, I would argue, new forms of organization are created. Carol Muller also points out that new identities do not spring from an organization's ideas, but rather are present, more or less latent, in existing social networks (Mueller 1992: 7).

In sum, network analysis has been applied to the study of collective action in a number of ways that are also applicable to my research: (a) the social group to be studied is defined by the systems of relationships in which one is interested; (b) identity, resources, information, goods, services, etc., are diffused through social networks, including from one type of social network to another; (c) new identities, forms and fora of organization do not emerge from an organization's vision or mission, but must be developed from already existing collective identities; and (d) organizations serve as catalizers of collective mobilization and the transformation of existing identities into new ones. Therefore, I expect social network analysis will provide me with both a method of data collection and analytical tools to address my research questions and hypotheses.

The research design

In each project, the system within which I will map the networks will be the project itself. That is, all those social agents, of different kinds, directly or indirectly involved and/or affected by the project, located at the project main site or not, compose the system. I will map three types of networks in each project (or system):

1. Organizational networks:


Community or neighborhood group: membership is mostly defined by local residence or by family relations; for instance, the local church.

Multi-local group or grassroots organization: membership is not defined by place or family, but by other criteria such as occupation, religion, etc. The key differential characteristic here is that they are multi-local. Examples are fisherman colônia (trade association), membership in the church municipal organization, etc..

Organizations, both governmental and non-governmental: technical or research institutes whose leadership and most of the staff are non-local.

2. Market networks.

3. Natural resource use networks.

I will mostly follow a snowball sample. In this process, one of the key aspects I will concentrate on is the relationship of each social agent (type of tie) has with the project lead organization(s). I intend to define the boundaries of the mapping process according to the conservation and development objectives of the project. In other words, as long as the next level of connection is related to these objectives (therefore influencing their results), the snowball sample will be carried out. The boundary is determined by the relationship to the project objectives, not its lead organizations or activities. This means that, theoretically, I will identify connections that are beyond the scope of work of the project (which I indeed expect to find)

In order to verify my hypotheses I will use a two-level comparative framework. First I intend to compare the part of the networks mapped in each system that is mobilized or targeted by the project lead organizations and/or activities with the networks of those who are mobilized or targeted by the project. I expect that there will be different types of intersection between the networks. By studying (a) the structures of the networks and (b) the flow of organizational, market or natural resource information, products, and services of each network, and comparing the pattern of intersection among them, I expect to gain insight on the degree of insertion of the project in the structure of social organization in the region it targets. At a second level I intend to study the relationship between (a) the networks and their intersections and (b) the project results. I will then compare the results of this study among the different projects. Through this analysis I hope to identify patterns of regularity between a project's impacts on the structure and process of social organization and its reported results in terms of conservation and development.

With regards to identity, since there are many possible identities among the participants of the networks, I will focus on those elements that are promoted by the project through environmental education programs or through capacity building in certain activities (such as the promotion of sustainable forestry or fishing practices, inventories of biodiversity and ecosystem dynamics in protected areas, etc.). I am particularly interested in the impacts of C&D projects in changing attitudes and practices (socio-political, economic, and of natural resource use) in the region where they are implemented.

I will measure diffusion of identity according to the testimony of individuals in different networks, as well as the programs and practices of groups and organizations. I will measure the forms and processes of social organization through a combination of various indicators:

  • Type of organization;
  • Level of formalization;
  • Way of solving internal disagreements and conflict;
  • Way of solving disagreements and conflict with other groups, organizations, and government;
  • Decision-making process (especial attention to who has voice and how, and to the perception of effectiveness of this voice);
  • Way of allocating responsibility over different activities;
  • System of internal and external accountability, and perception of its importance and effectiveness;
  • Procedures and process to define leadership;
  • Role and performance of leadership;
  • Way of allocating and controlling organizational resources.

To measure the impacts in terms of development and conservation, which are the core goals of these C&D projects, I will use relative measures, based on the comparison of project stated goal and the perception of achievement of those involved in the project (the same levels used to map networks), and documentation of the project and other agents (including donor evaluations of development and conservation impacts; these ideally are done by a panel of multi-disciplinary experts, with special expertise on the activities promoted by the project). I will also use data from socioeconomic surveys and scientific research whenever they are available. By adopting this relative approach I avoid the highly controversial issue of what constitutes sustainability of either development or conservation results. This relative measure of success is also compatible with the understanding of C&D as social movements, as the focus of the evaluation shifts away from scientific or otherwise external definitions and standards to the experiences (attitudes and behavior) of those involved in it.

My research questions and hypotheses are clearly focused on understanding processes of change. Therefore, as a colleague has pointed out a number of times, I need data in a time series. Even though I have been gathering data for the last two summers, this is by no means close to a time series research model. To substitute for a time series model, I will use the extensive documentation each project has, starting sometimes at the phase of project design (a phase that can last up to two years).

V. Conclusion

While activists in the north protest against the international trade agreements and send letters to their elected representatives requesting action to protect the environment, in the world's tropical belt, where so much of the world's biodiversity is concentrated, the movement for environmental conservation takes up different goals and different strategies. Nonetheless, however distinct C&D projects might be from what is traditionally seen as a social movement, or from the common set up of players in a social movement (challengers and challenged), they are still an effort of various social agents to promote social change through collective action: new shared identity, new practices of natural resource use, new forms of organizing around common interests and dealing with disagreement. As we observed about the characterization of environmental change as a problem, the way one conceptualizes a phenomenon directly affects how one deals with it. In the case of C&D projects, I believe understanding them as a social movement opens a number of options to scholars, practitioners, and policy makers. It puts the structure and processes of social organization right in the center of C&D projects, stretching our view to include social change as a key objective along with the substantive conservation and development goals (if not, ultimately, the key objective, the sine qua non condition for the all other objectives). From this new outlook, C&D projects should be focusing direct attention on the structure and process of social organization in the region where they are implemented, and on how they might be changed to integrate a variety of social agents, across geographical, social, economic, political, cultural, and environmental boundaries toward the same objectives. This means social agents involved in a C&D project should change the way they design, implement, and evaluate the project, focusing more (at least initially) on social process than C&D material outcomes (even though these are also important, especially the latter, in motivating participation and continuity in the movement). NGOs and funding organizations, which usually define the project's strategy of social organization because of their position of power among the social agents involved, should focus on supporting the transformation of identities and forms of social organization, rather then imposing the creation of new ones. This means that projects are less likely to generate quick, short term quantitative results, such as the number of associations created and the size of their membership, the number of collaborative agreements signed with new grassroots organizations, or the number of people trained in new technologies. It also means that the type of structure and process of social organization that emerges from the movement might not follow a pre-established pattern, increasing the difficulty in measuring their change. Finally, if my hypotheses are supported by my research, I believe social networks and relational data, might be a viable alternative to measure the social change in C&D projects.