Opposition to the social and environmental impacts of large-scale mining has become more visible in Chile since the early 1990s, yet not all mining projects catalyze mobilization. Building on the concept of defensive mobilization, I argue that opposition is more likely when a project is perceived as a threat to some members of a community. Using a data set of all major mining projects submitted for environmental licensing since environmental impact assessments were implemented in Chile, I identify the conditions under which mining projects lead to opposition. The results, based on binary logistic regression analysis, show that projects threatening agrarian and indigenous communities, where threats to existing water and land uses are especially salient, are more likely to be opposed. Community challenges are also more likely for projects majority-owned by international investors. About four out of every ten proposed projects have faced opposition, and only a handful of projects have ever been definitively rejected, even as projects that are found to violate regulations are increasingly fined and challenged in court as well as facing protests and public scrutiny.

INTRODUCTION

On a mild morning in August 2013, a group of activists gathered in front of the regional government office building in Arica, a city in northern Chile, while another group set up roadblocks to protest the approval of Los Pumas, a new manganese mining project (La Estrella de Arica 2013). The open-pit mine and processing plant were to be located near the town of Putre, from where the manganese would then be trucked to the port in Arica and shipped to China, to be used in steel production. Farmers and members of local water users’ associations, Aymara indigenous organizations, and neighborhood groups argued that mining operations and waste disposal from the processing plant threatened livelihoods and land uses, and could worsen water and air quality. Residents disagreed about the development strategy for the area, with some supporting the mine and others contesting the claims that it would bring jobs and development, questioning the promise of new economic gains and whether they would be worth the risks.

Mining projects tend to generate contrasting local responses. For some, mining is entrenched in local and regional economic identities, especially in the north of Chile, where some towns build their entire local economies around the mining industry and where many residents tend to voice strong support for projects that bring employment and income. Mining exports—above all copper, but also iron, silver, gold, molybdenum concentrate, and lithium carbonate—are a core part of the Chilean economy and are recognized as such. For example, 70.5 percent of respondents in a Latinobarómetro (1997) survey said that mining contributes more to the country’s economic development than any other economic activity. Copper has accounted for between 70 and 90 percent of mining exports every year since the 1960s (Banco Central de Chile 2018; Comisión Chilena del Cobre 2015), as well as serving as an important source of jobs and state revenue, mainly through the state-owned companies Corporación Nacional del Cobre de Chile (Codelco) and Empresa Nacional de Minería (Enami). Mining activities account for more than 15 percent of total GDP and more than half of Chilean exports (Ministerio de Relaciones Exteriores 2017), with most shipments going to China, Japan, South Korea, the European Union, the United States, or Brazil. With the mining boom of the 1990s, foreign investment in the mining sector increased dramatically (Lagos 1997). Of the more than 90 countries that received mining-related investments between 1990 and 2001, for example, Chile was the leading destination (Bridge 2004).

Yet during the last 30 years, Chile has also seen the growth of contentious episodes around questions of environmental justice and has created new institutions for environmental licensing and oversight. From early examples of protest and legal mobilization around pollution from mining, such as the case of copper tailing wastes from the Potrerillos and El Salvador mines being deposited into the Salado River and polluting Chañaral Bay (Asenjo 1989; Castilla 1996), to more recent emblematic conflicts such as the construction of the Pascua Lama gold mine on the border of Chile and Argentina (Urkidi 2010), public opposition to mining has become more visible. As Chilean society has become more open to protest and has developed more associational capacity, mining projects have faced increasing scrutiny. In this context, some projects have become flashpoints for mobilization, while others have not. Why are some projects more likely to provoke opposition than others?

In this article, I suggest that opposition to proposed mining projects is a form of defensive mobilization in response to threats from extractive projects. Local opposition to mining is more likely when a project poses a threat—or is perceived as posing one—to some members of a community. Defensive mobilization occurs when threats induce groups to organize to protect themselves (Almeida 2003, 2007; Tilly 1978). In the context of large-scale mining projects, defensive mobilization is collective action that aims to prevent harm to economic livelihoods, dispossession, and the disruption of everyday life. Here I identify the characteristics of localities where defensive mobilization in response to mining is most likely, and the kinds of projects that catalyze it.

I argue that defensive opposition to mining at the local level is more likely in the presence of particular structural factors—specifically, in agrarian communities and communities where a substantial proportion of residents identify as indigenous, where residents may perceive mining as competing with or threatening existing uses and practices associated with water and land. Such threats are especially salient because water and land have important material and cultural dimensions tied to livelihoods, identities, and forms of associational and community activity, making their disruption a significant threat to everyday life. Associational activities around water and land in such localities, such as spaces bringing together water users and agricultural producers, are also likely to provide the underlying social categories and networks from which organized opposition may emerge. Since opposition usually occurs in response to particular projects, a project’s features—such as size, international versus domestic investors, and the type of substance being extracted—are also likely to matter.

To test whether these hypotheses hold for opposition to mining projects in democratic Chile, I use a data set of mining projects based on environmental impact assessment data collected by the Chilean government, as well as other local governmental and non-governmental sources. The data set includes all mining projects for which environmental impact studies were submitted since the creation of a national environmental institutional framework in Chile. My quantitative analysis is informed by an extensive review of qualitative data sources and is part of a broader project involving fieldwork on water politics in Chile.

I make two main contributions to existing scholarship. First, I extend the concepts of threat and defensive mobilization in sociology (Almeida 2003; Goldstone and Tilly 2001; Gramling and Freudenberg 1992; Tilly 1978; Van Dyke and Soule 2002) to make sense of local opposition to extractive industries in Latin America, responding to the call to consider the “character and dynamics of different categories of movements” in the study of opportunity and threat (McAdam and Boudet 2012:97). Second, bringing in insights from the growing literature on recent social mobilization in Chile (Delamaza, Maillet, and Neira 2017; Donoso and Von Bülow 2017; Medel and Somma 2016), I use a new data set to systematically examine patterns of local opposition to mining in Chile since environmental impact assessment and reporting were introduced, an effort that can provide the basis for future research on the dynamics of specific conflicts or comparisons between mining projects and other kinds of projects, such as hydroelectric dams. My aim is also to contribute to scholarship that brings social movement scholars in U.S. sociology in dialogue with extensive work by political scientists, political ecologists, and geographers on extractive industries and their consequences for collective action and inaction (Arce 2014; Bebbington et al. 2008; Haslam and Tanimoune 2016; Hochstetler and Tranjan 2016; Martinez-Alier 2002) so that the insights of both sets of scholars can jointly advance understandings of local opposition to, and other forms of engagement with, extractive development in the global South.

The paper proceeds as follows. In the first section, I outline the concepts of threat and defensive mobilization, discussing how my emphasis on structural conditions here fits with a more complex understanding of threat. I then explain why mining projects in agrarian communities and communities with a substantial indigenous population might lead to increased perceptions of threat and subsequent opposition, emphasizing the importance of water and land. Next, I incorporate explanatory factors that capture how project characteristics such as size and ownership by international versus domestic investors may further contribute to defensive opposition. I also briefly sketch the broad changes in political conditions for mobilization about socio-environmental issues in Chile since the democratic transition. I then outline my methodological approach, present the results, and conclude with reflections on implications and directions for future research.

MINING, THREAT, AND DEFENSIVE MOBILIZATION

Compared to the dominant lines of inquiry in social movement scholarship, and despite a resurgence of interest in recent years, threat typically receives less attention. Tilly (1978) wrote about threat as a determinant of collective action alongside opportunity in the 1970s, pointing out that collective action may, in fact, “mount more rapidly as a function of threat than as a function of opportunity” (135). Yet, more than 15 years later, Tarrow (1994:86) observed that while opportunity was widely studied, much less was known about how “threats to interests, values, and, at times, survival that different groups and individuals experience” influence contention, an observation echoed by McAdam (1999), who pointed out that opportunity had eclipsed threat as a concept in the study of mobilization.

With renewed interest in threat came different approaches to conceptualizing it, the core of which were firmly oriented toward the state. Rather than examining the political opportunities provided by openings within the state, for example, scholars argued that state-based threats such as repression and the erosion of rights also shape patterns of protest (Almeida 2003; Goldstone and Tilly 2001). Scholars of Latin American social movements in particular showed how perceived threats to social citizenship rights, such as access to basic services, and state-attributed economic problems led to defensive collective action by popular sector groups affected by austerity politics (Almeida 2007; Eckstein and Wickham-Crowley 2003). The cultural and emotional turn in social movement scholarship addressed threats from another angle, reviving some aspects of strain theory, arguing that emotional responses such as moral outrage also affect mobilization, and showing how movements emerge as a consequence of not only material issues but also the threatened disruption of everyday life (Jasper 1998, 2008; Snow et al. 1998). Meanwhile, environmental sociologists analyzed the role of ecological threats such as declining bird populations and increasing levels of air pollutants in galvanizing environmentalist organizing in the United States (Johnson and Frickel 2011), as well as incorporating threat into long-standing discussions of environmental and technological risk.

Building on this literature, I show that defensive mobilization also occurs in response to threats from investments in extractive industries—especially threats to water and land. Perceived or actual threats from large-scale extractive projects can catalyze defensive collective action that aims to prevent long-term harm to economic livelihoods, dispossession, and the disruption of everyday life. Perceptions of threat stem from anticipated or actual changes in ecological and social landscapes in communities where projects are proposed and implemented. Beyond posing environmental risks at a particular site, such local changes tend to be connected to more complex regional and global commodity chains, and driven by the extraction of natural resources for their financial value. In this context, threats to water and land activate defensive responses from communities for material and cultural reasons, with their disruptive potential rooted in their capacity to affect material conditions as well as everyday routines. Water and land are also the basis for associational activities, such as water user associations and agricultural cooperatives, which can form the initial networks for emergent social mobilization.

Threats from mining are often material, affecting the physical world and the availability of natural resources for other activities, with consequences for livelihoods and health. For example, mining not only uses copious amounts of freshwater, often in arid regions, but also generates air pollution and creates waste rock, slurry, and effluent that must be disposed of according to a strict set of standards if it is not to contaminate local soil and water sources (Pizarro et al. 2010). In the short term, competition from a mine over land and water can threaten livelihoods in an agropastoralist community, while gradual overuse of water or dispossession can force residents to find other ways to make a living or lead to relocation altogether, as in the case of Quillagua (Morales and Azócar 2016). Mining projects that take the form of mega-development projects can lead to displacement, transforming landscapes as well as communities (Gellert and Lynch 2003). Slowly accumulating pollution or toxic accidents can lead to disproportionate health risks and physical damage, including in already marginalized communities (Carruthers 2008).

Yet threats to resources like land and water are not only about the potential effects on livelihood or health, but also what those resources signify. A number of scholars have discussed the multiple dimensions of land in indigenous political struggles, especially land as both a means of subsistence and a connection to ethnic identity, as a “factor of production,” and as “territory,” which invokes “notions of autonomy and sovereignty” (Lucero 2008:105; see also Bebbington et al. 2008; Yashar 2005). These accounts suggest, then, that threats to land may be perceived as threats to both economic livelihood and identity. Similarly, writing about a different kind of dispossession (water privatization in Cochabamba, Bolivia), Simmons (2016) argues that privatization provoked mobilization not only because of water’s material qualities but also because of its role in reproducing “imagined and quotidian communities” (38). An extractive project may therefore disrupt livelihoods or pose health risks, as well as threatening to disrupt spaces in which identities are lived and practiced, especially if we think of identities as “categories of everyday social experience, developed and deployed by ordinary social actors” (Brubaker and Cooper 2000). Extractive projects may lead to, as Bebbington et al. (2008:2890) put it, the “disarticulation of existing moral economies.”

Responses to threat tend to emerge from preexisting networks and social categories of social actors, where shared identities on the basis of territory, or on the basis of associational activity around the allocation of water or agricultural production, may contribute to the creation of new ties that are mobilized for opposition. While a new mining project may threaten the routines of daily life in a community, those established routines—and the communication networks and shared understandings they imply—can also be used to respond to threat (Tilly 1978:135). The structural conditions I identify here are thus important not only because some members of agrarian and indigenous communities are especially likely to feel vulnerable with respect to mining, but also because they may have forms of social organization—the beginnings of a movement infrastructure (Andrews 2001)—on which oppositional efforts are built.

For example, coordinating committees are often formed to organize opposition to a proposed mining project. In the Chilean context, examples include the Coordinating Committee for the Defense of Water and the Environment in Tierra Amarilla, the Aymara Coordinating Committee for the Defense of the Natural Resources of Arica and Parinacota, the Coordinating Committee for the Defense of the Loa River, and many others.1 While they are often supported by NGOs in Santiago or through transnational advocacy groups, such coordinating committees are typically built on networks of associational activity based on ties among neighbors, water users, cultivators, participants in women’s organizations, or members of indigenous organizations and communities. Preexisting associational activities around water and land, such as around the question of allocating water for irrigation or maintaining a rural water users’ association, may be activated to support mobilization.

Mining and Defensive Mobilization in Agrarian Communities

The majority of mining activities considered in this study—more than 70 percent—take place in the Norte Grande and Norte Chico regions of Chile, which are mostly desert or semiarid.2 Consequently, the agrarian communities3 most likely to clash with the mining projects discussed here correspond predominantly to small-scale subsistence agriculture, rather than the agroindustry present in the major agricultural regions of the country. In the Norte Grande, agropastoralist activities include raising camelids such as vicuñas and guanacos and small-scale fruit and vegetable cultivation, mostly in valleys and oases. In the Norte Chico, fruit and vegetable cultivation, as well as goat husbandry, are common in the transverse valleys that run from east to west, such as the Copiapó Valley and the Huasco Valley. By contrast, most activities involving the large-scale cultivation of land—both for key domestic crops and for export-oriented crops—take place in central and southern Chile, especially between the Angostura de Paine and the Bío Bío River (ODEPA 2012).

Discourses around the defense of agricultural and pasture land, the livelihoods of rural communities, and the protection of the quantity and quality of water available for agriculture and daily life in rural communities can be found across cases of mobilization. Mining projects pose a particularly salient threat to small-scale agriculture because both require water resources as a productive input and because of the potential impacts of establishing mining waste dumps next to areas where food is grown. The perceived threat from competition over water resources or their overuse is an especially contentious point in Chile, where water resources were commodified under Pinochet (Bauer 1998), where development projects are accompanied by the acquisition of indefinite use rights over water, and where some rivers have been declared “used up” in terms of the allocation of water rights (e.g., DGA 2000).

In this context, the allocation of water for mining and small-scale agricultural activities has been a recurring issue of contention in northern Chile. For example, groups tracking socio-environmental conflicts find that since at least the 1990s agrarian communities have been feeling threatened by increasing competition over water resources as well as potential and actual pollution from weakly regulated mining activities (San Martín 1997:35, 49, 59). In some cases, the conflicts are also over how mining projects infringe on crop cultivation on adjacent lands, pressure farmers or other residents to sell their land, or construct mining infrastructure close to places where people live.

Numerous recent examples continue these themes. For instance, in the Los Pumas case described in the introduction, one activist leader argued that by approving the project, regional authorities were “risking the development of our region, of agriculture in our valleys” and remarked that it was “hard to see how this activity will generate progress” given that there would be fewer new mining jobs than existing sources of livelihood (Correa 2013). In another controversial case, the Tres Valles project near Salamanca, the Committee for the Defense of Chuchiñi Valley (Comité de Defensa del Valle de Chuchiñi) was created in response to plans to construct heap leach pads for extracting copper from ore. Residents were concerned that leach pads would be built near a settlement, in an area that was being used as grazing land by residents of Salamanca and Illapel for goats and other livestock. As a result of concerns about the placement of leach pads and the use of chemical leaching processes and sulfuric acid in close proximity, the company proposed resettling a number of families, some of whom had been living in the area for over 50 years (Minería Chilena 2008). In another case, the Candelaria mining company operating in Tierra Amarilla in the Atacama region faced opposition and was eventually fined for environmental violations, which included disposing of liquid industrial waste in an unauthorized location, occupying a larger area than they had promised, locating a mining waste dump next to land used for crop cultivation, and not complying with their commitments to decrease water consumption from the Copiapó River aquifer (SMA 2017).

If water sources used for irrigation are contaminated by mining activity or if the cost of water use rights becomes too high, then the livelihood of agrarian communities can be threatened, especially in an arid landscape. In such contexts, mining and small-scale agriculture represent competing economic strategies. Land and water are also central to everyday activities that are both routine and necessary for the reproduction of life. For example, more than half of rural residents do not get water through a piped connection but rather directly from wells and rivers; about 38 percent of rural residents get their water from a well or noria, and about 19 percent get their water from a river or stream (INE 2002). Those getting their water directly from rivers and wells are more vulnerable to pollution and other changes in water quality and quantity.

People living in rural communities that are reliant on agricultural and pastoralist uses of land and water are therefore especially likely to feel threatened and develop oppositional mobilization. In rural Chile, the productive uses of land and water tend to be organized through associational activity such as farmers’ associations, agricultural cooperatives, and water users’ associations, while potable water is often managed by rural drinking water associations and cooperatives, which can provide networks and interactions that may become the basis of subsequent defensive collective action. These observations echo scholarship that shows that mining clashes with rural livelihoods in the Andean region (Bebbington 2007; Bebbington et al. 2008; Taylor 2011) and that agrarian and agropastoralist communities often form part of coalitions that oppose large extractive development projects in Chile (Cortez and Maillet 2018; Urkidi 2010). I expect that (H1) mining projects in municipalities where a substantial proportion of land is used for agricultural activities are more likely to result in opposition.

Mining and Defensive Mobilization in Indigenous Communities

Research also documents that projects have resulted in conflicts with indigenous communities throughout South America (McDonell 2015; Rice 2012; Sawyer 2004), including in Chile (Delamaza, Maillet, and Neira 2017; Richards 2013). Most mining projects clashing with indigenous communities are in northern Chile, where Aymara, Atacameño, Coya, Diaguita, and Quechua groups represent the majority of indigenous peoples, in contrast to southern Chile, where the Mapuche are the largest indigenous group. Whether opposition occurs primarily on the basis of threat to productive activities and livelihoods, or a broader set of claims about collective rights and indigenous territories, land and water are often at the center.

For rural indigenous communities reliant on subsistence crops and small-scale commerce, mining projects can mean competition over scarce freshwater sources (Castro-Lucic 2002; Eisenberg 2013), similar to the dynamics described above. Despite government efforts since the mid-1990s to regularize indigenous water rights, under the legal framework regulating water allocation, mining companies have been able to purchase customary water rights from indigenous groups (Cuadra 2000). In the Antofagasta region in the Atacama Desert, for example, some water rights sales by Atacameño communities have led to abandonment of traditional agriculture and to relocation (Budds 2010). For indigenous communities living in rural areas, mining may threaten not only water for irrigation but also drinking water. Approximately 32 percent of indigenous households in rural areas get their water directly from a river or stream, and 48 percent from a well, meaning that the vast majority of rural indigenous communities get water from sources that are more vulnerable to contamination (INE 2002).

Primarily agrarian indigenous communities may feel threatened by large-scale mining development projects for economic reasons. But development projects clashing with indigenous territories may also generate opposition because they are perceived to threaten the spaces in which collective identities are practiced. For example, for Aymara agropastoralists, water is considered both a “precious” natural resource and a substance inextricably tied to survival (Eisenberg 2013:71), which is reflected in both economic and cultural practices around water. Indigenous leaders participating in environmental impact assessment processes also often emphasize the multidimensionality of water. For instance, comments on the Cerro Casale mining project from the Colla Pai-Ote indigenous community directly state that the project would be located on indigenous territory where the community’s “productive and economic” as well as “social, cultural, and ritual activities” take place, with “absolute dependence on the resources of the territory” (Araya Altamirano 2011:1, 7). In practice, livelihoods and identities tend to be connected.

Indigenous communities and organizations provide networks and interactions that may serve as the basis for defensive mobilization in response to threats from extractive projects.4 For example, community organizations engaged in agricultural and fishing activities have historically coordinated to defend their heritage “against external threats and internal conflicts” (Gundermann and Vergara 2009:108). Alongside associations forged on the basis of productive or cultural activities, and territorially organized organizations such as neighborhood councils, organizations representing broader ethnic interests (usually based in urban areas) also serve as resources for local opposition (Gundermann and Vergara 2009), as in the case of the Consejo de Pueblos Atacameños (Morales and Azócar 2016:119).

Yet ethnographic studies also reveal the complexities of indigenous communities’ relationships to mining and how groups negotiate the long-standing and powerful presence of the mining industry. For example, Francisco Molina (2012) analyzes the “competing rationalities” used by the Atacameño community in Chiu Chiu and the national company Codelco in their conflict over water from the Loa River, as well as documenting indigenous leaders’ emphasis that their village remains an “indigenous development area” even if it is in the middle of a mining region. Anita Carrasco (2016:131) also shows how Atacameño indigenous people in the Loa River basin use “competing cultural logics” of water, treating “water as a commodity when leasing water to—but not when defending water from—a mining company,” following ritual practices that attest to a moral economy of water that values it as sacred while also participating in market transactions. In addition to purchasing or leasing water rights, mining companies develop programs to establish ties with local communities and offer material benefits, as a way to mitigate conflict and continue extractive activities.

Indigenous mobilization in socio-environmental conflicts often connects claims about indigenous collective rights with demands about the use of natural resources and the environment in indigenous territories, such that threats to the local economy, territorially linked identities, and the environment tend to be closely intertwined, and connected to past stories of dispossession and conflict with the mining industry going back to the early twentieth century. Thus I expect that (H2) projects in municipalities with indigenous communities are more likely to result in opposition.5 

Defensive Mobilization in Response to Project Characteristics

All else equal, mining project characteristics can also affect the likelihood of opposition. A wave of foreign investment in mining has taken place since the 1990s (Lagos 1997), with foreign direct investment accounting for more than half of total investment in the sector, in contrast to agriculture, forestry, and fishing, which tend to be dominated by Chilean companies. It is therefore plausible that a company’s country of origin matters. Distrust of projects run by non-Chilean companies may originate in attitudes of resource nationalism toward resources that have a central connection to Chilean economic identity, combining political economy and cultural politics (Perreault and Valdivia 2010), as well as echoing earlier debates about national and foreign involvement in the extraction of copper, which were highly salient in Chile’s history.

Distrust may also be exacerbated by the impression that Chileans are not benefitting adequately from projects that the public recognizes as posing serious risks. According to Zerene et al. (2018), for example, compared to the 10 largest transnational companies extracting copper in Chile, the state-owned company Codelco pays almost four times more in taxes per ton of copper produced, which suggests that large private mining companies could be paying more. Perceptions of threat may also be situational and rooted in a firm’s “liability of foreignness” (Zaheer 1995), its practices relative to local expectations, and its interactions with community members, which may foster perceptions that a project constitutes a threat rather than bringing positive development. I expect that (H3) projects majority-owned by a non-Chilean company are perceived as more threatening and are more likely to result in opposition.

The investment size of projects may also matter. Though larger projects may be viewed by supporters as a greater investment in regional and national economies, such projects usually entail more intensive extraction processes that last longer and have more environmental and social impacts on the land, water, and local economies and forms of social organization, thereby increasing the perceived environmental risk and social threat. Larger projects are also expected to be in place longer, thus leaving a lasting impact on the community. I expect that (H4) larger projects are more likely to lead to opposition.

The substance being extracted may also matter, in part because of the higher risks of particular extraction processes. For instance, in gold mining, crushed rock is treated with cyanide, and in copper mining, with sulfuric acid, both of which have shown up in local water sources. Sites adjacent to mining projects have been associated with environmental health risks that take time to be detected, caused by “cyanide and sulfuric acid leakage, long-term exposure to acid mine drainage, respiratory problems from dust inhalation, and altered water quality caused by the diversion of existing water supplies” (Helwege 2015). I expect that (H5) mining projects involving copper or gold extraction are more likely to result in opposition than those involving the extraction of other substances.

Putting Defensive Mobilization in Political and Associational Context

Politically, Chilean society was relatively quiescent during the early years of the democratic transition that began in 1989 (Carlin 2006; Delamaza 2015; Hipsher 1996), which is often characterized as an incomplete transition and a transition by agreement (Arcaya 1999; Garretón 1991). Nascent democracy during the 1990s was built on consensus politics (Garretón 2003), during which center-left governments continued Chile’s extractive, export-led development model, while prioritizing democratic stability, rebuilding democratic institutions, and channeling social grievances into institutional spaces. Yet during the last 30 years, new state institutions for environmental licensing and oversight were created, and defensive mobilization around questions of environmental justice has become more common.

The first efforts to put in place an environmental regulatory framework took place at the start of the democratic transition, with the creation of a national environmental commission, CONAMA (Comisión Nacional del Medio Ambiente), and the passage of the 1994 Environmental Law, which was the first comprehensive environmental legislation, and introduced environmental licensing for extractive industries. The next set of major environmental laws originated in Michele Bachelet’s administration, which is usually characterized as having a stronger environmental agenda than the predecessor governments of the Concertación, despite not living up to all of its promises (Latta and Aguayo 2012). The new round of legislation created the Ministry of the Environment, effectively giving CONAMA ministerial status, and transformed the existing environmental impact assessment system into the Environmental Evaluation Service (Servicio de Evaluación Ambiental) as an independent division of the new ministry, with new features for public participation.

The legal reform also created the Ministerial Council for Sustainability and the Superintendency of the Environment (Superintendencia del Medio Ambiente) to provide oversight and issue fines for violations of agreements made during the environmental licensing process.6 During Bachelet’s first term, the Chilean Congress also ratified the Indigenous and Tribal Peoples Convention (ILO Convention 169), after almost two decades of internal debate about whether it was appropriate for Chile. Despite ongoing disagreements about its implementation, this development has been consequential for both indigenous rights and the environmental approvals process. For instance, as a result of the work of activist lawyers filing claims on behalf of affected communities, the courts began to overturn environmental qualification resolutions because they had not carried out prior consultation, as required by ILO 169.

Thus, Chile has experienced variation in government commitments to an environmental agenda and ties with environmental movements (Silva 2018), with stronger commitments and stronger ties during Michele Bachelet’s terms in office. During earlier administrations, perceptions of the fragility of democracy may have influenced how threats from extractive development projects were interpreted and the willingness to engage in protest as a response. Meanwhile, the perception of possibilities for influence in institutional spaces may have made opposition on the basis of existing threats more likely. I expect that (H6) opposition to mining projects is more likely under a left-leaning administration open to socio-environmental agendas.

In parallel to the institutional developments described above, as disillusionment with institutionalized politics and critiques of the dictatorship’s legacy grew during the 1990s, social movements emerged around various demands, such as education reform and new environmental politics, with protest intensifying since the early 2000s (Somma and Medel 2017). In the case of socio-environmental issues, while protest politics have been driven by “new” social movement groups, at times supported by transnational advocacy groups, the core organizing activities to diagnose problems and articulate claims have taken place through movement networks on the basis of existing community organizations such as neighborhood councils (juntas de vecinos) and other grass-roots organizations of residents who are likely to experience threats from new extractive projects firsthand. Such spaces are likely to contribute to the interactions and ties on the basis of which broad-based opposition to threats from new extractive development projects may form. Following established ideas about the potential effects of organizational capacity on mobilization, I expect that (H7) opposition to mining projects is more likely where there is greater local associational capacity.

DATA AND METHODS

I test my arguments using a data set of major mining projects in Chile, which contains all 189 projects for which company representatives submitted a request for authorization via an environmental impact study (EIS) between 1992 and mid-2017. By law, companies are instructed to submit a full EIS if the project is characterized by health risks, relocation or other substantial changes to the ways of life of local communities, proximity to protected areas, or adverse effects on natural resources such as air, water and soil; otherwise, they can submit an environmental impact declaration.7 

Documents associated with the EISs are accessible through the Environmental Impact Evaluation System (SEIA) database, accessible online and with additional resources available at the Documentation Center of the Ministry of the Environment in Santiago, Chile. The database provides descriptive information such as the date the project was submitted for review, decision date, project location, the company submitting the study, and the outcome of the submission process (whether the project was approved, rejected, not admitted, withdrawn by the company, or is still being processed).8 Project files also contain the impact study materials and documents such as letters submitted by various actors consulted in the process—including the relevant municipal and regional organizations—as well as documents providing evidence of public meetings and observations made by groups in affected communities, especially during the last decade. The SEIA data were combined with other sources of publicly available data from the Chilean government, newspapers, and NGOs (Table 1).

TABLE 1.

Summary of variables and data sources

VariableDescriptionSources
Opposition 1 if there is a documented challenge to the mining project, 0 otherwise National Institute of Human Rights, Observatory of Latin American Mining Conflicts, El Mercurio, and Environmental Impact Assessment System 
Agricultural community presence 1 if at least 30% of the project site is agricultural (predios agrícolas), 0 otherwise (see  Appendix for continuous version) National Municipal Information System 
Indigenous community presence 1 if the proportion of population that identifies as indigenous (Alacalufe [Kawaskar], Atacameño, Aymara, Colla, Mapuche, Quechua, Rapanui, and Yámana [Yagán] groups) is higher than the national average (see  Appendix for continuous version) National Statistical Institute of Chile 
Foreign ownership of project 1 if more than 50 percent of shares were owned by a non-Chilean company at the time the EIS was submitted, 0 otherwise Environmental Impact Assessment System, BNamericas, and company websites 
Investment size (USD 100 million) Overall estimated investment in the proposed project, in units of USD 100 million, adjusted for inflation Environmental Impact Assessment System 
Copper or gold mining project 1 if the project involves copper or gold extraction, 0 if it involves neither Environmental Impact Assessment System 
Community organizations (per 1,000) Sum of neighborhood councils, mothers’ centers, community unions, and “other community organizations,” per 1,000 people at project site National Municipal Information System 
Left party government 1 if the outcome of the environmental licensing process was given during a left-leaning presidential administration (Partido Socialista), 0 otherwise Environmental Impact Assessment System 
Percentage living in poverty Percentage of the population of the project site living in poverty, according to CASEN (Encuesta de Caracterización Socioeconómica Nacional) National Municipal Information System 
Natural log of population size Natural log of population size at project site National Municipal Information System 
VariableDescriptionSources
Opposition 1 if there is a documented challenge to the mining project, 0 otherwise National Institute of Human Rights, Observatory of Latin American Mining Conflicts, El Mercurio, and Environmental Impact Assessment System 
Agricultural community presence 1 if at least 30% of the project site is agricultural (predios agrícolas), 0 otherwise (see  Appendix for continuous version) National Municipal Information System 
Indigenous community presence 1 if the proportion of population that identifies as indigenous (Alacalufe [Kawaskar], Atacameño, Aymara, Colla, Mapuche, Quechua, Rapanui, and Yámana [Yagán] groups) is higher than the national average (see  Appendix for continuous version) National Statistical Institute of Chile 
Foreign ownership of project 1 if more than 50 percent of shares were owned by a non-Chilean company at the time the EIS was submitted, 0 otherwise Environmental Impact Assessment System, BNamericas, and company websites 
Investment size (USD 100 million) Overall estimated investment in the proposed project, in units of USD 100 million, adjusted for inflation Environmental Impact Assessment System 
Copper or gold mining project 1 if the project involves copper or gold extraction, 0 if it involves neither Environmental Impact Assessment System 
Community organizations (per 1,000) Sum of neighborhood councils, mothers’ centers, community unions, and “other community organizations,” per 1,000 people at project site National Municipal Information System 
Left party government 1 if the outcome of the environmental licensing process was given during a left-leaning presidential administration (Partido Socialista), 0 otherwise Environmental Impact Assessment System 
Percentage living in poverty Percentage of the population of the project site living in poverty, according to CASEN (Encuesta de Caracterización Socioeconómica Nacional) National Municipal Information System 
Natural log of population size Natural log of population size at project site National Municipal Information System 

Outcome Variable: Opposition to a Proposed Mining Project

The dependent variable captures whether communities expressed opposition to a proposed mining project, coded dichotomously as either 1 (documented challenge) or 0 (no documented challenge). These challenges include a range of practices and forms of opposition, expressed through protest politics such as demonstrations and marches, as well as institutional politics—relying on lawsuits, pressuring elected government officials, and coordinated efforts to provide comments for the EIS process. I identified instances of community challenges by searching for evidence across four sources: the database of socio-environmental conflicts compiled by the Chilean National Human Rights Institute (Instituto Nacional de Derechos Humanos); the database of mining conflicts compiled by the Observatory of Latin American Mining Conflicts (Observatorio de Conflictos Mineros de América Latina); articles in the national Chilean newspapers El Mercurio and La Tercera; and comments provided by organized civil society groups as part of the citizen participation process.

If evidence of community opposition to the project after the submission of the study appeared in any one of these sources, the variable was coded as 1. Early publications by organizations working on documenting cases of socio-environmental conflict in the 1990s were also used to contextualize community challenges (Padilla and San Martín 1995; San Martín 1997). I relied on multiple sources to assess the presence of a challenge because instances of socio-environmental conflicts and community resistance are often underreported in the national newspapers.9 

Main Predictor Variables and Additional Controls

I measured the agrarian variable by calculating the proportion of land used for agriculture at the municipal level, based on data from Chile’s National Municipal Information System (SINIM 2018).10 To calculate the proportion of the municipal population that identifies as indigenous, I used census data (INE 2002).11 The two variables are positively correlated, which is consistent with the fact that the indigenous population in northern Chile tends to be less urbanized than the overall population, especially where mining is concentrated (INE and MIDEPLAN 2002), so the two variables are tested in separate models as well as together.

Data for the project ownership, investment size, and substance type variables come from the SEIA database. The variable for projects with majority non-Chilean ownership was coded 1 if more than 50 percent of shares in the company were owned by a non-Chilean company at the time the EIS was submitted, and 0 otherwise. This information was gathered from ownership data in the SEIA database and supplemented with information from company websites and company profiles compiled by BNamericas. Financial investment in the project, adjusted for inflation and estimated in USD 100 millions, was used to approximate relative project size. Finally, each project was also coded for the types of resources being extracted, including copper, gold, lithium, iron, limestone, and many others. Some projects involved the extraction of more than one resource; for example, gold and silver were often extracted together. This information came from the project description.

To test one aspect of political opportunity, I use the relative openness of the government administration to environmental claims (cf. Olzak and Soule 2009). During the first two post-authoritarian decades in Chile, all presidential administrations were affiliated with the Coalition of Parties for Democracy (Concertación de Partidos por la Democracia), with Aylwin and Frei Ruiz Tagle from the Christian Democrat Party, Lagos from the Party for Democracy (Partido por la Democracia), and Bachelet from the Socialist Party (Partido Socialista de Chile.) In 2010, the dominant coalition shifted from the Coalition of Parties for Democracy to the Coalition for Change (Coalición por el Cambio), led by Piñera and based on an alliance between the conservative National Renewal (Renovación Nacional) and Independent Democratic Union (Unión Demócrata Independiente) parties; since then, Bachelet and Piñera have each been elected for additional terms. None of the parties are especially interested in an environmentalist agenda, but the Socialist Party incorporates some ideas about sustainable development into its platform and of these parties is therefore the most likely to be viewed as open to socio-environmental demands. I created a dummy variable, coded 1 if the president was affiliated with the Socialist Party, which in practice refers to Bachelet’s two terms. For associational capacity, I used SINIM data on community organizations by municipality and transformed the variable to be in terms of organizations per 1,000 people. The organizations include neighborhood associations (juntas de vecinos), mothers’ centers (centros de madres), community unions (uniones comunales), and other community organizations (otras organizaciones comunitarias funcionales).

I included two additional control variables: the socioeconomic level of communities and population size. For the socioeconomic variable, communities with higher poverty rates may be more likely to protest when their sources of livelihood are threatened, though it is also possible that less time and fewer resources to make grievances visible, and other mechanisms, may suppress mobilization. For this variable, I use the proportion of the population living in poverty where the project will be taking place. I also control for the size of the municipality where a project is proposed, using the natural log of municipal population. The data for both variables come from SINIM (2018). Summary statistics for all variables are provided in Table 2.

TABLE 2.

Descriptive statistics

Continuous variableMeanS.D.
Percentage agricultural land 15.46 19.83 
Percentage indigenous population 6.5 11.82 
Investment size (USD 100 million) 12.43 74.15 
Community organizations (per 1,000 people) 5.61 5.27 
Percentage living in poverty 15.02 9.35 
Natural log of population size 10.46 1.63 
Categorical variable Frequency/percentage (0) Frequency/percentage (1) 
Opposition 114 (60.32) 75 (39.68) 
Agricultural community presence (above 30%) 153 (80.95) 36 (19.05) 
Indigenous community presence (above 5%) 140 (74.07) 49 (25.93) 
Foreign ownership of project 96 (50.79) 93 (49.21) 
Copper or gold mining project 53 (28.04) 136 (71.96) 
Leftist party government 129 (68.25) 60 (31.75) 
Continuous variableMeanS.D.
Percentage agricultural land 15.46 19.83 
Percentage indigenous population 6.5 11.82 
Investment size (USD 100 million) 12.43 74.15 
Community organizations (per 1,000 people) 5.61 5.27 
Percentage living in poverty 15.02 9.35 
Natural log of population size 10.46 1.63 
Categorical variable Frequency/percentage (0) Frequency/percentage (1) 
Opposition 114 (60.32) 75 (39.68) 
Agricultural community presence (above 30%) 153 (80.95) 36 (19.05) 
Indigenous community presence (above 5%) 140 (74.07) 49 (25.93) 
Foreign ownership of project 96 (50.79) 93 (49.21) 
Copper or gold mining project 53 (28.04) 136 (71.96) 
Leftist party government 129 (68.25) 60 (31.75) 

FINDINGS AND DISCUSSION

Since the start of environmental licensing in Chile, about four out of every ten proposed mining projects have led to documented public opposition (Figure 1). The projects covered here required a full environmental impact study (as opposed to the hundreds more that required only a declaration) and therefore were a priori the most expected to have serious environmental and social impacts. The finding is generally consistent with what studies have found in other countries and other sectors. For example, Hochstetler and Tranjan (2016:505) found that about 41.5 percent of large hydroelectric plants in Brazil led to community challenges, though small hydro and wind power plants generally had much lower rates of opposition. The finding is also similar to McAdam and Boudet’s (2012) discovery that of the 20 cases of new energy infrastructure projects in their study, based in the United States, 10 experienced oppositional mobilization. Haslam and Tanimoune (2016), in their comprehensive study of mining in Argentina, Brazil, Chile, Mexico, and Peru, find a somewhat lower proportion: only about 26 percent of all the mines in their data set have a known conflict. In other words, most projects, even those we might expect to be controversial, do not provoke organized opposition—although many do, and presumably not all organized opposition is documented.

FIGURE 1.

Map of opposition to mining projects

FIGURE 1.

Map of opposition to mining projects

Given that my outcome of interest is binary, I use binomial logistic regression to analyze the factors that made challenges more likely. Logistic regression models the likelihood that a case will fall into the higher of two categories of the outcome variable—that is, a project will be more likely to experience public opposition—given a certain set of values for the predictor variables. I use three models in this paper. The first model tests how the presence of agricultural land at a proposed project site, as well as project characteristics, political opportunities, and associational capacity, contribute to the likelihood of opposition to a mining project. The second model tests the presence of indigenous communities and leaves out agricultural land, while keeping the other variables in the model the same. The third model includes both agricultural and indigenous presence variables together with other variables.12 

Table 3 shows the standardized coefficients and the odds ratios for all three models. I include both because standardized coefficients give a better sense of the magnitude of the effects of the variables relative to each other, whereas odds ratios offer a more direct and intuitive way of interpreting the effects, especially for dichotomous variables (Menard 2011). Across the models, I find evidence supporting some of my hypotheses. First, as Model 1 shows, projects proposed in agricultural communities are 6.9 times as likely to provoke opposition as projects proposed in communities where agriculture is less central (p < 0.001).13 Furthermore, as the extent to which agrarian livelihoods are central to a locality increases, suggesting that threats to water and land may be especially consequential for continued ways of life in that locality, opposition becomes more likely (see Figure A-1 in the  Appendix).

TABLE 3.

Logistic regression results for public opposition to mining

(1)(2)(3)
βOdds ratioβOdds ratioβOdds ratio
Agricultural community presence 0.404*** 6.940***   0.353** 5.607** 
Indigenous community presence   0.234* 2.527* 0.095 1.517 
Foreign ownership of project 0.236* 2.431* 0.241* 2.309* 0.225* 2.370* 
Investment size (USD 100 million) −0.022 0.999 –0.045 0.999 –0.022 0.999 
Copper or gold mining project 0.227* 2.586* 0.214 2.289 0.227* 2.635* 
Community associations (per 1,000) 0.493** 1.193** 0.547*** 1.198*** 0.498** 1.199** 
Leftist party government 0.310** 3.503** 0.353** 3.724** 0.307** 3.543** 
Percentage living in poverty 0.091 1.019 0.110 1.021 0.097 1.020 
Natural log of population size 0.336** 1.477** 0.245 1.300 0.345** 1.505** 
N 189 189 189 
Prob > chi2 0.0000 0.0000 0.0000 
LR chi2 58.01 50.45 58.79 
Pseudo-R2 0.2285 0.1987 0.2316 
(1)(2)(3)
βOdds ratioβOdds ratioβOdds ratio
Agricultural community presence 0.404*** 6.940***   0.353** 5.607** 
Indigenous community presence   0.234* 2.527* 0.095 1.517 
Foreign ownership of project 0.236* 2.431* 0.241* 2.309* 0.225* 2.370* 
Investment size (USD 100 million) −0.022 0.999 –0.045 0.999 –0.022 0.999 
Copper or gold mining project 0.227* 2.586* 0.214 2.289 0.227* 2.635* 
Community associations (per 1,000) 0.493** 1.193** 0.547*** 1.198*** 0.498** 1.199** 
Leftist party government 0.310** 3.503** 0.353** 3.724** 0.307** 3.543** 
Percentage living in poverty 0.091 1.019 0.110 1.021 0.097 1.020 
Natural log of population size 0.336** 1.477** 0.245 1.300 0.345** 1.505** 
N 189 189 189 
Prob > chi2 0.0000 0.0000 0.0000 
LR chi2 58.01 50.45 58.79 
Pseudo-R2 0.2285 0.1987 0.2316 

*p < 0.05; **p < 0.01; ***p < 0.001 (two-tailed test)

Notes: β represents the standardized coefficient; numbers in brackets are 95% confidence intervals.

I also test the effects of non-Chilean majority-ownership, size, and metal type, and find that some project characteristics do matter for the likelihood of opposition. If a mining project is majority-owned by a non-Chilean company, its odds of facing public opposition are about 2.4 times those in other cases, holding all other variables constant (p < 0.05). Again, this difference is likely to be due to a combination of explanations rooted in how international investors relate to local communities, how they are perceived, and how the perception that the economic gains of extracting natural resources are not staying in the country contribute to perceived threats, expectations about how threat will be managed, and the willingness to tolerate environmental threat. This finding is consistent with what Haslam and Tanimoune (2016) found in their larger sample of mining projects across five countries. The substance being extracted also matters; for mining ventures involving copper or gold extraction, the odds of opposition are 2.6 times the odds for mines involving the extraction of other substances (p < 0.05). The variable capturing investment size is not statistically significant.

The above factors are statistically significant in the presence of the variable capturing shifting political opportunities at the national level by considering leftist party ideology of the presidential administration and general local associational capacity, both of which also have positive effects on the likelihood of mobilization. The odds of opposition are about 3.5 times as high under a leftist government as under a non-leftist government. It is possible that some of this effect is driven by the relative openness of political opportunity structures under such a government, while another part is driven by community challenges becoming a more routine response to projects proposed at later moments during the period of democratic transition and consolidation.14 The odds of opposition also increase by a factor of 1.2 for every unit increase in the number of community associations per 1,000 people (p < 0.01). While the size of the effect is relatively small, it affirms the idea that associational density matters for the likelihood of opposition. In other words, associational spaces such as neighborhood associations can provide some associational resources on the basis of which mobilization emerges.

As Model 2 shows, if a site where a project is introduced has more indigenous residents than the national average, then the odds of opposition increase by a factor of 2.5 (p < 0.05). This finding is consistent with observations in prior descriptive analyses that a significant number of socio-territorial conflicts have taken place in areas of the country with a high indigenous presence (Delamaza, Maillet, and Neira 2017:35–36). In Model 3, I include all explanatory variables together. Because agriculture and indigenous presence are positively correlated, indigenous presence loses statistical significance. But agriculture remains statistically significant: controlling for the presence of indigenous groups, opposition to mining projects is still about 5.6 times more likely at project sites with an agricultural community presence (p < 0.01). The variables capturing foreign ownership of project, copper or gold mining project, associational capacity, and leftist party government also remain positive and statistically significant.15 

Of course, threat does not automatically and inevitably lead to opposition, nor will all community residents experience threat identically. Identifying a set of conditions that contribute to perceptions of threat is intended as a starting point for investigating a more complex process. For example, if a household relies on raising goats and growing crops for their livelihood, a reaction to possible threats to water and land would be rooted in material conditions. Yet the perception of threat and its possible effects on everyday life will also be a social, cognitive, and emotional process that projects the possible impacts of the mining project into the future and weighs them against other considerations, including instrumental concerns. This process will be intersubjective; that is, it will be influenced by others in the community, stories from other parts of the country, the collective memory of other experiences with mines, promises made by mining company representatives, collective-action frames developed by movement leaders and participants, and so on. In this article, then, I consider identifying structural conditions and project characteristics that make opposition more likely as a starting point rather than an endpoint of inquiry.

Furthermore, while here I have focused on the likelihood of opposition, there are also numerous cases for which no documented opposition could be found. While this may be partially attributed to a data issue or the fact that not all socio-environmental conflicts are documented, it may be that some projects are simply less controversial. For some project sites, it is also possible that some mobilization is demobilized or repressed, with potentially disruptive threats from a mining project evaluated against threats originating in fear of repression or the imperative to simply adapt and focus on negotiating the terms of extraction. As scholarship shows, collective inaction is also a response to threat (Auyero and Swistun 2009; Neumann 2016), and there are mechanisms that can work to demobilize even those who feel threatened (Lapegna 2016). Perhaps just as complex are cases where communities adapt to their changing political and economic conditions, such as indigenous communities that employ competing logics, retaining cultural practices relating to water while also shifting to a different set of practices and understandings when leasing water to mining companies (Carrasco 2016).

Finally, it is worth noting that grass-roots opposition seldom prevents such projects from going forward. Research has suggested that environmental licensing is not a mechanism that blocks projects from proceeding, and the virtual absence of variation in the status of proposed projects is consistent with this claim. Since the start of environmental licensing requirements in Chile, most mining projects requiring an EIS have been approved. Only a handful of the proposed mining projects were denied, some in highly controversial decisions (Vargas, Cerda, and Riffo 2018).16 Despite the concerns of pro-business skeptics about environmental regulation, there is little evidence that having institutional mechanisms for public participation means that opposition will be able to block every project.

Environmental institutions are expected to anticipate threats and to help manage conflicts over mining, yet environmental licensing is contested. On the one hand, the process is intended for government agencies and civil society groups to evaluate potential socio-environmental impacts and provide input on proposed development projects, in the context of a government aiming to be more accountable, transparent, and democratic. On the other hand, environmental impact assessments are often criticized for providing a “rubber stamp” for projects rather than real oversight, relying on technical expert knowledge in ways that exclude other voices, and becoming a co-opted or ineffectual space for public participation. This study suggests that at the very least, environmental licensing processes in Chile represent a new site of contestation where claims about potential threats and socio-environmental impacts from new mining projects are made.

CONCLUSION

The exploitation of natural resources, and mining in particular, is a core part of Chile’s economic development strategy. Across the country, mining is intertwined with local economic identities and generates resource nationalism and pride, especially around copper extraction. International investors recognize Chile’s receptivity to mining projects. Despite fluctuations in global prices and demand for copper, gold, and other Chilean exports, this interest is unlikely to fade, though it may shift or expand into new areas, for example, as global demand for lithium rises. To mitigate conflicts, mining companies continue to develop corporate social responsibility programs and are investing in new technologies such as desalination plants to address conflicts over water.

Yet, though Chile is dependent on extractive development, the public has become increasingly vocal about socio-environmental issues. Observers may attribute this to the post-materialist thesis (Inglehart 1997); in other words, as Chile has become wealthier, its public has taken on more post-materialist values. However, it is not only groups with a middle-class base that are mobilizing about social and environmental impacts. Since community opposition to mining is rooted in conflicts and threats of conflicts over water, land use, and environmental health, instances of opposition are often organized along environmental justice lines and increasingly tied to questions about environmental racism, the consequences of newly proposed projects for existing patterns of marginalization and dispossession, and the unequal distribution of environmental hazards relative to the benefits of employment and economic development. The results presented here suggest that in this context, defensive mobilization in response to mining projects is more likely in places where residents may perceive mining as a threat—for example, where agricultural land uses are important and where indigenous people live. Certain kinds of investment projects are also more likely to lead to public opposition, especially those that are majority-owned by international investors. Associational spaces—such as neighborhood associations—also make the emergence of community challenges more likely. As organized civil society has grown more vocal about socio-environmental grievances, the terms of the Chilean economic development are more open for debate.

This article has sketched the contours of defensive mobilization in response to mining projects in post-authoritarian Chile. But the prevalence of local opposition to mining projects—especially projects owned by international companies that are perceived to threaten livelihoods and everyday life in some communities—raises the larger question of what is to be made of this opposition. Local opposition to mining occurs in response to threats to specific communities, places, and ways of life. Yet in the context of a legacy of resource-extractive and export-led development, with ties to an authoritarian past, these instances of opposition cannot be written off as mere sparks of not-in-my-backyard mobilization and facility siting disagreements. Mobilization against mining, while reactive, tends to be intertwined with broader efforts to demand socioeconomic, environmental, and indigenous rights, suggesting that in some contexts, the contrast between “rights struggles” and “siting controversies” is not so clear, with a siting controversy potentially drawing on rights discourse in mobilization efforts, or scaling up to a broader movement around the defense of water (cf. McAdam and Boudet 2012). In the Chilean case, opposition to mining is occurring in a society that has rebuilt democratic institutions over the past three decades—a society experiencing movement cycles unfolding as part of the “third generation” of movements that have revealed “a deeper crisis of democratic governance and political representation” as movements aim to reclaim social citizenship rights and demand decommodification (Roberts 2017:222; see also Silva 2017), following other patterns of defensive mobilization in response to economic liberalization and austerity politics in the region (Almeida 2007; Silva 2009; Simmons 2016). Like the students’ movement and other recent waves of mobilization in Chile, socio-environmental conflicts and opposition to mining raise fundamental questions about the strategies and meanings of development, who extractive-led economic growth benefits, and how different parts of the country envision the future.

More broadly, the findings underscore threat as a core concept that helps make sense of large-scale patterns of public opposition to extractive industries, alongside political opportunities, organizational resources, and other factors that lead opposition groups to mobilize around a set of claims and demands. The findings also suggest questions for future research. First, future work can address the extent to which the actions of experts, government officials, and mining firms influence the perceptions of threat from proposed mining projects (cf. Amengual 2018; Barandiarán 2018). Moreover, while here I have emphasized primarily agrarian communities dependent on water and land resources, future studies can examine how other social actors—such as union leaders, rank-and-file miners, and owners of pro-mining local businesses—in communities where projects are proposed fit into this story, as well as how residents negotiate the terms of extraction rather than directly opposing projects. Finally, a series of questions remain about the dynamics of defensive mobilization, such as the extent to which threats to livelihoods overlap with threats to collective rights and territories, and how local opposition to new mining projects transforms into and scales up to broader movements.

While this study focuses on communities dependent on water and land for socioeconomic and cultural reasons, it also raises questions about how threats from possible future events are construed and socially constructed, and provides a starting point for analyzing community responses to a broader range of socio-environmental events that may destabilize patterns of livelihood and ways of life. Future research can do more to trace how different groups construct and evaluate such threats, as well as how local defensive mobilization impacts policy change and other political responses to changing relationships between society and the natural environment.

Appendix. Logistic Regression Results with Additional Independent Variables

FIGURE A-1.

Predicted probability of a project’s facing opposition at different levels of agricultural presence, comparing projects majority-owned by Chilean and non-Chilean companies, holding other variables at their means.

FIGURE A-1.

Predicted probability of a project’s facing opposition at different levels of agricultural presence, comparing projects majority-owned by Chilean and non-Chilean companies, holding other variables at their means.

FIGURE A-2.

Predicted probability of a project’s facing opposition at different proportions of indigenous population, comparing projects led by Chilean and non-Chilean companies, holding other variables at their means.

FIGURE A-2.

Predicted probability of a project’s facing opposition at different proportions of indigenous population, comparing projects led by Chilean and non-Chilean companies, holding other variables at their means.

TABLE A-1.

Logistic regression results for public opposition to mining (with continuous variables for agricultural and indigenous presence)

(1)(2)
Standardized coefficientOdds ratioStandardized coefficientOdds ratio
Percentage agricultural land 0.553*** 1.057***   
Percentage indigenous population   0.437** 1.076** 
Foreign ownership of project 0.215* 2.355* 0.203* 2.222* 
Investment size (USD 100 million) –0.007 1.000 –0.031 0.999 
Copper or gold mining project 0.266* 3.245* 0.253* 3.022* 
Community associations (per 1,000) 0.356* 1.144* 0.470** 1.192** 
Leftist party government 0.315** 3.851** 0.307** 3.649** 
Percentage living in poverty 0.076 1.016 0.124 1.026 
Natural log of population size 0.392** 1.617** 0.269* 1.385* 
N 189 189 
Prob > chi2 0.0000 0.0000 
LR chi2 65.38 57.57 
Pseudo-R2 0.2575 0.2267 
(1)(2)
Standardized coefficientOdds ratioStandardized coefficientOdds ratio
Percentage agricultural land 0.553*** 1.057***   
Percentage indigenous population   0.437** 1.076** 
Foreign ownership of project 0.215* 2.355* 0.203* 2.222* 
Investment size (USD 100 million) –0.007 1.000 –0.031 0.999 
Copper or gold mining project 0.266* 3.245* 0.253* 3.022* 
Community associations (per 1,000) 0.356* 1.144* 0.470** 1.192** 
Leftist party government 0.315** 3.851** 0.307** 3.649** 
Percentage living in poverty 0.076 1.016 0.124 1.026 
Natural log of population size 0.392** 1.617** 0.269* 1.385* 
N 189 189 
Prob > chi2 0.0000 0.0000 
LR chi2 65.38 57.57 
Pseudo-R2 0.2575 0.2267 

Notes: *p < 0.05; **p < 0.01; ***p < 0.001 (two-tailed test).

TABLE A-2.

Logistic regression results for public opposition to mining (with alternative political opportunity variables)1

(1)(2)(3)
βOdds ratioβOdds ratioβOdds ratio
Agricultural community presence 0.467*** 10.253***   0.407** 7.934** 
Indigenous community presence   0.264* 2.905* 0.108 1.632 
Foreign ownership of project 0.218* 2.351* 0.232* 2.272* 0.210* 2.315* 
Investment size (USD 100 million) –0.098 0.997 –0.148 0.996 –0.102 0.997 
Copper or gold mining project 0.214* 2.545* 0.207 2.260 0.210* 2.548* 
Community associations (per 1,000) 0.454** 1.184** 0.512** 1.188** 0.455** 1.188** 
Democratic transition2 0.377*** 1.107*** 0.420*** 1.108*** 0.382*** 1.111*** 
Leftist party mayor3 0.150 1.829 0.079 1.331 0.140 1.777 
Percentage living in poverty 0.147 1.031 0.170 1.033 0.160 1.035 
Natural log of population size 0.333** 1.495** 0.228 1.283 0.340** 1.520** 
N 189 189 189 
Prob > chi2 0.0000 0.0000 0.0000 
LR chi2 63.53 53.83 64.57 
Pseudo-R2 0.2502 0.2120 0.2543 
(1)(2)(3)
βOdds ratioβOdds ratioβOdds ratio
Agricultural community presence 0.467*** 10.253***   0.407** 7.934** 
Indigenous community presence   0.264* 2.905* 0.108 1.632 
Foreign ownership of project 0.218* 2.351* 0.232* 2.272* 0.210* 2.315* 
Investment size (USD 100 million) –0.098 0.997 –0.148 0.996 –0.102 0.997 
Copper or gold mining project 0.214* 2.545* 0.207 2.260 0.210* 2.548* 
Community associations (per 1,000) 0.454** 1.184** 0.512** 1.188** 0.455** 1.188** 
Democratic transition2 0.377*** 1.107*** 0.420*** 1.108*** 0.382*** 1.111*** 
Leftist party mayor3 0.150 1.829 0.079 1.331 0.140 1.777 
Percentage living in poverty 0.147 1.031 0.170 1.033 0.160 1.035 
Natural log of population size 0.333** 1.495** 0.228 1.283 0.340** 1.520** 
N 189 189 189 
Prob > chi2 0.0000 0.0000 0.0000 
LR chi2 63.53 53.83 64.57 
Pseudo-R2 0.2502 0.2120 0.2543 

Notes: *p < 0.05; **p < 0.01; ***p < 0.001 (two-tailed test).

1

The mayor’s political party was not statistically significant, while with each additional year elapsed since the democratic transition, proposed projects had slightly higher odds of opposition. For researchers interested in political opportunity explanations, the results raise questions for future research about the relative importance of leftist parties, processes of democratic consolidation, and broader protest cycles in explaining opposition to extractive industries. For example, social movement organizations and community associations may go through processes of political learning, acquiring repertoires of opposition as democratic practices become more familiar and the possibility of reverting to authoritarianism diminishes.

2

This variable refers to the years of democratic government between the start of the democratic transition and the time a mining project was submitted for environmental impact assessment.

3

This variable refers to the political party affiliation of the mayor in power at the time the project decision was issued. The data are from the Servicio Electoral de Chile. I created a dummy variable, with leftist parties coded as 1 and other parties coded as 0. Political party affiliations coded as 1 include the Socialist Party (Partido Socialista), Communist Party (Partido Comunista), Green Party (Partido Ecologista), and Humanist Party (Partido Humanista). Centrist parties such as the Christian Democrat Party and right-wing parties such as the Independent Democratic Union (Unión Demócrata Independiente) and National Renewal (Renovación Nacional) were coded as 0. For projects spanning more than one municipality, if at least one participating municipality had a mayor affiliated with a leftist party, the variable was coded 1.

REFERENCES

REFERENCES
Agrawal, Arun, and Clark C. Gibson.
1999
.
“Enchantment and Disenchantment: The Role of Community in Natural Resource Conservation.”
World Development
27
(
4
):
629
49
.
Almeida, Paul D.
2003
.
“Opportunity Organizations and Threat-Induced Contention: Protest Waves in Authoritarian Settings.”
American Journal of Sociology
109
(
2
):
345
400
.
Almeida, Paul D.
2007
.
“Defensive Mobilization: Popular Movements against Economic Adjustment Policies in Latin America.”
Latin American Perspectives
34
(
3
):
123
39
.
Amengual, Matthew.
2018
.
“Buying Stability: The Distributive Outcomes of Private Politics in the Bolivian Mining Industry.”
World Development
104
:
31
45
.
Andrews, Kenneth T.
2001
.
“Social Movements and Policy Implementation: The Mississippi Civil Rights Movement and the War on Poverty, 1965 to 1971.”
American Sociological Review
66
(
1
):
71
95
.
Araya Altamirano, Ercilia.
2011
.
Observaciones de la comunidad Colla Pai-Ote al EIA Optimización Proyecto Minero Cerro Casale
[Observations of the Colla Pai-Ote community for the Cerro Casale mining project EIS].
Copiapó
:
SEA de Atacama
.
Arcaya, Óscar Godoy.
1999
.
“La transición Chilena a la democracia: pactada”
[Chilean democratization: transition by agreement].
Estudios Públicos
79
106
.
Arce, Moises.
2014
.
Resource Extraction and Protest in Peru
.
Pittsburgh, PA
:
University of Pittsburgh Press
.
Asenjo, Rafael.
1989
.
“Innovative Environmental Litigation in Chile: The Case of Chañaral.”
Georgetown International Environmental Law Review
2
:
99
110
.
Auyero, Javier, and Débora Swistun.
2009
.
Flammable: Environmental Suffering in an Argentine Shantytown
.
Oxford
:
Oxford University Press
.
Banco Central de Chile.
2018
.
Base de Datos Estadísticos: Sector Externo, 2003–2018
[Statistical database: external sector, 2003–2018].
Santiago
:
Banco Central
.
Barandiarán, Javiera.
2018
.
Science and Environment in Chile: The Politics of Expert Advice in a Neoliberal Democracy
.
Cambridge, MA
:
MIT Press
.
Bauer, Carl J.
1998
.
Against the Current: Privatization, Water Markets, and the State in Chile
.
New York
:
Springer
.
Bebbington, Anthony (ed.).
2007
.
Minería, movimientos sociales, y respuestas campesinas: Una ecología política de transformaciones territoriales
[Mining, social movements, and peasant responses: a political ecology of territorial transformations].
Lima
:
Instituto de Estudios Peruanos
.
Bebbington, Anthony, Denise H. Bebbington, Jeffrey Bury, Jeannet Lingan, Juan Pablo Muñoz, and Martin Scurrah.
2008
.
“Mining and Social Movements: Struggles over Livelihood and Rural Territorial Development in the Andes.”
World Development
36
(
12
):
2888
2905
.
Bridge, Gavin.
2004
.
“Mapping the Bonanza: Geographies of Mining Investment in an Era of Neoliberal Reform.”
Professional Geographer
56
(
3
):
406
21
.
Brubaker, Rogers, and Frederick Cooper.
2000
.
“Beyond ‘Identity’.”
Theory and Society
29
:
1
47
.
Budds, Jessica.
2010
. “Water Rights, Mining, and Indigenous Groups in Chile’s Atacama.” Pp.
197
212
in
Out of the Mainstream: Water Rights, Politics and Identity
, edited by Rutgerd Boelens, David Getches, and Armando Guevara-Gil.
London
:
Earthscan
.
Carlin, Ryan E.
2006
.
“The Decline of Citizen Participation in Electoral Politics in Post-Authoritarian Chile.”
Democratization
13
(
4
):
632
51
.
Carrasco, Anita.
2016
.
“A Biography of Water in Atacama, Chile: Two Indigenous Community Responses to the Extractive Encroachments of Mining.”
Journal of Latin American and Caribbean Anthropology
21
(
1
):
130
50
.
Carruthers, David (ed.).
2008
.
Environmental Justice in Latin America: Problems, Promise, and Practice
.
Cambridge, MA
:
MIT Press
.
Castilla, Juan C.
1996
.
“Copper Mine Tailing Disposal in Northern Chile Rocky Shores: Enteromorpha Compressa (Chlorophyta) as a Sentinel Species.”
Environmental Monitoring and Assessment
40
:
171
84
.
Castro-Lucic, Milka.
2002
. “Local Norms and Competition for Water in Aymara and Atacama Communities (Northern Chile).” Pp.
187
201
in
Water Rights and Empowerment
, edited by Rutgerd Boelens and Paul Hoogendam.
Assen
:
Koninklijke Van Gorcum
.
Comisiado Presidencial para Asuntos Indígenas
.
2008
.
Informe de la Comisión Verdad Histórica y Nuevo Trato con los Pueblos Indígenas
[Report of the commission on historical truth and a new agreement with indigenous peoples].
Santiago
.
Comisión Chilena del Cobre.
2015
.
Embarques de Exportación de Cobre, Minería, y Productos No Mineros, 1960–2015
[Export shipments of copper, mining, and non-mining products, 1960–2015].
Santiago
:
Cochilco
.
Correa, Paula.
2013
.
“Alcaldes aseguran que proyecto minero Los Pumas implica efectos negativos en Arica”
[Mayors assert that the Los Pumas mining project entails negative effects in Arica].
Diario UChile
,
August
19
.
Cortez, Mauricio, and Antoine Maillet.
2018
.
“Trayectoria multinivel de una coalición promotora e incidencia en la agenda política nacional. El caso del conflicto de Pascua Lama y la ley de glaciares en Chile”
[Multilevel trajectory of an advocacy coalition to influence the national political agenda. The case of the Pascua Lama conflict and the glacier law in Chile].
Colombia Internacional
94
:
3
25
.
Cuadra, Manuel.
2000
.
“Teoría y práctica de los derechos ancestrales de agua de las comunidades atacameñas”
[Theory and practice of ancestral water rights of Atacameño communities].
Estudios Atacameños
19
:
93
112
.
Delamaza, Gonzalo.
2015
.
Enhancing Democracy: Public Policies and Citizen Participation in Chile
.
New York
:
Berghahn
.
Delamaza, Gonzalo, Antoine Maillet, and Christian Martínez Neira.
2017
.
“Socio-Territorial Conflicts in Chile: Configuration and Politicization (2005–2014).”
European Review of Latin American and Caribbean Studies
104
:
23
46
.
DGA
.
2000
.
“Declara el agotamiento del Río Loa y sus afluentes, II Región”
[Declaration of the depletion of the Loa River and its tributaries, Region II]. Edited by
Dirección General de Aguas
,
Santiago
.
Donoso, Sofia, and Marisa Von Bülow (eds.).
2017
.
Social Movements in Chile: Organization, Trajectories, and Political Consequences
.
New York
:
Palgrave Macmillan
.
Eckstein, Susan, and Timothy Wickham-Crowley (eds.).
2003
.
Struggles for Social Rights in Latin America
.
New York
:
Routledge
.
Eisenberg, Amy.
2013
.
Aymara Indian Perspectives on Development in the Andes
.
Tuscaloosa
:
University of Alabama Press
.
Garretón, Manuel Antonio.
1991
.
“La transición Chilena: Una evaluación provisoria”
[The Chilean transition: a provisional assessment].
Estudios Políticos
(
8
):
1
20
.
Garretón, Manuel Antonio.
2003
.
Incomplete Democracy: Political Democratization in Chile and Latin America
.
Chapel Hill
:
University of North Carolina Press
.
Gellert, Paul K., and Barbara D. Lynch.
2003
.
“Mega-Projects as Displacements.”
International Social Science Journal
55
(
175
):
15
25
.
Goldstone, Jack A., and Charles Tilly.
2001
. “Threat (and Opportunity): Popular Action and State Response in the Dynamics of Contentious Action.” Pp.
179
94
in
Silence and Voice in the Study of Contentious Politics
, edited by Ronald R. Aminzade, Jack A. Goldstone, Doug McAdam, Elizabeth J. Perry, Sidney Tarrow, William H. Sewell, and Charles Tilly.
Cambridge
:
Cambridge University Press
.
Gramling, Robert, and William R. Freudenberg.
1992
.
“Opportunity-Threat, Development, and Adaptation: Toward a Comprehensive Framework for Social Impact Assessment.”
Rural Sociology
57
(
2
):
216
34
.
Gundermann, Hans, and Jorge Iván Vergara.
2009
.
“Comunidad, organización y complejidad social andinas en el norte de Chile”
[Andean community, organization, and social complexity in northern Chile].
Revista Estudios Atacameños
38
:
107
26
.
Haslam, Paul A., and Nasser A. Tanimoune.
2016
.
“The Determinants of Social Conflict in the Latin American Mining Sector: New Evidence with Quantitative Data.”
World Development
78
:
401
19
.
Helwege, Ann.
2015
.
“Challenges with Resolving Mining Conflicts in Latin America.”
Extractive Industries and Society
2
(
1
):
73
84
.
Hipsher, Patricia L.
1996
.
“Democratization and the Decline of Urban Social Movements in Chile and Spain.”
Comparative Politics
28
(
3
):
273
97
.
Hochstetler, Kathryn, and Ricardo Tranjan.
2016
.
“Environment and Consultation in the Brazilian Democratic Developmental State.”
Comparative Politics
48
(
4
):
497
516
.
INE
.
2002
.
Chile: Censo de Población y Vivienda 2002
[Chile: population and housing census 2002].
Santiago
:
Instituto Nacional de Estadística de Chile
.
INE and MIDEPLAN
.
2002
.
Estadísticas sociales de los pueblos indígenas en Chile: Censo 2002
[Social statistics about indigenous people in Chile: 2002 census].
Santiago de Chile
:
Instituto Nacional de Estadísticas and Ministerio de Planificación Nacional
.
Inglehart, Ronald.
1997
.
Modernization and Postmodernization: Cultural, Economic, and Political Change in 43 Societies
.
Princeton, NJ
:
Princeton University Press
.
Jasper, James M.
1998
.
“The Emotions of Protest: Affective and Reactive Emotions in and around Social Movements.”
Sociological Forum
13
(
3
):
397
424
.
Jasper, James M.
2008
.
The Art of Moral Protest: Culture, Biography, and Creativity in Social Movements
.
Chicago, IL
:
University of Chicago Press
.
Johnson, Erik W., and Scott Frickel.
2011
.
“Ecological Threat and the Founding of U.S. National Environmental Movement Organizations, 1962–1998.”
Social Problems
58
(
3
):
305
29
.
La Estrella de Arica
.
2013
,
August
19
.
“Cinco personas fueron detenidas tras bloquear la ruta 5 Norte y la que une Arica-La Paz en rechazo a la minera Los Pumas.”
Lagos, Gustavo.
1997
.
“Developing National Mining Policies in Chile: 1974–1996.”
Resources Policy
23
(
1–2
):
51
69
.
Latinobarómetro
.
1997
. Database (http://www.latinobarometro.org/).
Lapegna, Pablo.
2016
.
Soybeans and Power: Genetically Modified Crops, Environmental Politics, and Social Movements in Argentina
.
New York
:
Oxford University Press
.
Latta, Alex, and Beatriz E. Cid Aguayo.
2012
.
“Testing the Limits: Neoliberal Ecologies from Pinochet to Bachelet.”
Latin American Perspectives
39
(
4
):
163
80
.
Lucero, José Antonio.
2008
.
Struggles of Voice: The Politics of Indigenous Representation in the Andes
.
Pittsburgh, PA
:
University of Pittsburgh Press
.
Martinez-Alier, Joan.
2002
.
The Environmentalism of the Poor: A Study of Ecological Conflicts and Valuation
.
Cheltenham
:
Edward Elgar
.
McAdam, Doug.
1999
.
“Revisiting the US Civil Rights Movement: Toward a More Synthetic Understanding of the Origins of Contention.”
Working Paper 588,
Center for Research on Social Organization
,
Ann Arbor, MI
.
McAdam, Doug, and Hilary Schaffer Boudet.
2012
.
Putting Social Movements in their Place: Explaining Opposition to Energy Projects in the United States, 2000–2005
.
New York
:
Cambridge University Press
.
McDonell, Emma.
2015
.
“The Co-constitution of Neoliberalism, Extractive Industries, and Indigeneity: Anti-mining Protests in Puno, Peru.”
Extractive Industries and Society
2
(
1
):
112
23
.
Medel, Rodrigo M., and Nicolás M. Somma.
2016
.
“Marchas, ocupaciones o barricadas? Explorando los determinantes de las tácticas de la protesta en Chile”
[Marches, occupations, or barricades? Exploring the determinants of protest tactics in Chile.]
Política y gobierno
23
(
1
).
Menard, Scott.
2011
.
“Standards for Standardized Logistic Regression Coefficients.”
Social Forces
89
(
4
):
1409
28
.
Minería Chilena.
2008
.
“Crean el Comité de Defensa del Valle de Chuchiñi”
[Committee for the Defense of the Chuchiñi Valley created].
Minería Chilena
,
Santiago
.
Ministerio de Relaciones Exteriores.
2017
.
Comercio Exterior de Chile
[Chilean foreign trade].
Santiago
:
Dirección General de Relaciones Económicas Internacionales
.
Molina Camacho, Francisco.
2012
.
“Competing Rationalities in Water Conflict: Mining and the Indigenous Community in Chiu Chiu, El Loa Province, Northern Chile.”
Singapore Journal of Tropical Geography
33
:
93
107
.
Morales, Héctor, and Rodrigo Azócar.
2016
.
“Minería y Relaciones Interétnicas en Atacama”
[Mining and interethnic relations in Atacama].
Estudios Atacameños: Arqueología y Antropología Surandinas
52
:
113
27
.
Neumann, Pamela.
2016
.
“Toxic Talk and Collective (In)action in a Company Town: The Case of La Oroya, Peru.”
Social Problems
63
(
3
):
431
46
.
ODEPA
.
2012
.
Panorama de la Agricultura Chilena
[Chilean agriculture at a glance].
Santiago
:
Oficina de Estudios y Políticas Agrarias del Ministerio de Agricultura
.
Olzak, Susan, and Sarah A. Soule.
2009
.
“Cross-Cutting Influences of Environmental Protest and Legislation.”
Social Forces
88
(
1
):
201
25
.
Padilla, César, and Pablo San Martín.
1995
.
Conflictos Ambientales: Una Oportunidad para la Democracia
[Environmental conflicts: an opportunity for democracy].
Santiago
:
Observatorio Latinoamericano de Conflictos Ambientales and Instituto de Ecología Política
.
Perreault, Tom, and Gabriela Valdivia.
2010
.
“Hydrocarbons, Popular Protest and National Imaginaries: Ecuador and Bolivia in Comparative Context.”
Geoforum
41
:
689
99
.
Pizarro, Jaime, Pablo M. Vergara, José A. Rodríguez, and Alejandra M. Valenzuela.
2010
.
“Heavy Metals in Northern Chilean Rivers: Spatial Variation and Temporal Trends.”
Journal of Hazardous Materials
181
(
1–3
):
747
54
.
Rice, Roberta.
2012
.
The New Politics of Protest: Indigenous Mobilization in Latin America’s Neoliberal Era
.
Tucson
:
University of Arizona Press
.
Richards, Patricia.
2013
.
Race and the Chilean Miracle: Neoliberalism, Democracy, and Indigenous Rights
.
Pittsburgh, PA
:
University of Pittsburgh Press
.
Roberts, Kenneth M.
2017
. “Chilean Social Movements and Party Politics in Comparative Perspective: Conceptualizing Latin America’s ‘Third Generation’ of Anti-Neoliberal Protest.” Pp.
221
47
in
Social Movements in Chile: Organization, Trajectories, and Political Consequences
, edited by Sofia Donoso and Marisa Von Bülow.
New York
:
Palgrave Macmillan
.
San Martín, Pablo.
1997
.
Conflictos Ambientales en Chile: 1995–1996
[Environmental conflicts in Chile: 1995–1996].
Santiago
:
Observatorio Latinoamericano de Conflictos Ambientales and Instituto de Ecología Política
.
Sawyer, Suzana.
2004
.
Crude Chronicles: Indigenous Politics, Multinational Oil, and Neoliberalism in Ecuador
.
Durham, NC
:
Duke University Press
.
Silva, Eduardo.
2009
.
Challenging Neoliberalism in Latin America
.
New York
:
Cambridge University Press
.
Silva, Eduardo.
2017
. “Post-Transition Social Movements in Chile in Comparative Perspective.” Pp.
249
80
in
Social Movements in Chile: Organization, Trajectories, and Political Consequences
, edited by Sofia Donoso and Marisa Von Bülow.
New York
:
Palgrave Macmillan
.
Silva, Eduardo.
2018
.
“Environmental Movement, Contentious Politics, and Institutional and Policy Change: Chile, 1994–2017.”
Paper presented at the Latin American Studies Association Congress
,
Barcelona, Spain
,
May 23–25, 2018
.
Simmons, Erica.
2016
.
Meaningful Resistance: Market Reforms and the Roots of Social Protest in Latin America
.
Cambridge
:
Cambridge University Press
.
SINIM
.
2018
.
Sistema Nacional de Información Municipal
[National System of Municipal Information].
Santiago
:
Subsecretaría de Desarrollo Regional y Administrativo, Ministerio del Interior
.
SMA
.
2017
.
“Resolución Exenta No. 206”
[Exempt Resolution no. 206].
Procedimiento sancionatorio contra Compañía Minera Candelaria
[Penalty proceedings against Candelaria mining company],
March
22
,
1
7
.
Snow, David A., Daniel M. Cress, Liam Downey, and Andrew W. Jones.
1998
.
“Disrupting the ‘Quotidian’: Reconceptualizing the Relationship between Breakdown and the Emergence of Collective Action.”
Mobilization
3
(
1
):
1
22
.
Somma, Nicolás M., and Rodrigo Medel.
2017
. “Shifting Relationships between Social Movements and Institutional Politics.” Pp.
29
61
in
Social Movements in Chile
, edited by Sofia Donoso and Marisa Von Bülow.
New York
:
Palgrave Macmillan
.
Tarrow, Sidney.
1994
.
Power in Movement: Social Movements, Collective Action and Politics
.
Cambridge
:
Cambridge University Press
.
Taylor, Lewis.
2011
.
“Environmentalism and Social Protest: The Contemporary Anti-mining Mobilization in the Province of San Marcos and the Condebamba Valley, Peru.”
Journal of Agrarian Change
11
(
3
):
420
39
.
Tilly, Charles.
1978
.
From Mobilization to Revolution
.
Reading, MA
:
Addison-Wesley
.
Urkidi, Leire.
2010
.
“A Glocal Environmental Movement against Gold Mining: Pascua-Lama in Chile.”
Ecological Economics
70
:
219
27
.
Van Dyke, Nella, and Sarah A. Soule.
2002
.
“Structural Social Change and the Mobilizing Effect of Threat: Explaining Levels of Patriot and Militia Organizing in the United States.”
Social Problems
49
(
4
):
497
520
.
Vargas, H., C. Cerda, and C. Riffo.
2018
.
“Tribunal ordena nueva votación de Dominga al declarar ilegal actuación del Comité de Ministros de Bachelet”
[Court orders a new vote on Dominga after declaring Bachelet’s committee of ministers has acted illegally].
El Mercurio
,
April
28
.
Yashar, Deborah.
2005
.
Contesting Citizenship in Latin America: The Rise of Indigenous Movements and the Postliberal Challenge
.
Cambridge
:
Cambridge University Press
.
Zaheer, Srilata.
1995
.
“Overcoming the Liability of Foreignness.”
Academy of Management Journal
38
(
2
):
341
63
.
Zerene, Gino Sturla, Ramón E. López, Simón Accorsi O., and Eugenio Figueroa B.
2018
.
“La riqueza regalada a la gran minería del cobre en Chile: nuevas estimaciones, 2005–2014”
[The wealth handed over to large-scale copper mining in Chile: our estimates, 2005–2014].
Revista de la CEPAL
124
:
107
29
.

NOTES

NOTES
I would like to thank the Center for Inter-American Policy & Research at Tulane University, where I conducted this research as a postdoctoral fellow. Earlier versions of the article benefitted from feedback from Elisabeth Clemens, Paul Haslam, Christopher Chambers-Ju, Jessica Price, Héctor Bahamonde, participants at the Latin American Studies Association’s annual congress in May 2018, and anonymous reviewers. Thanks also to August Ritchart, Ariel Lubin, and Matt Pietrus for their assistance with data collection and GIS.
1.
The names in Spanish are: Coordinadora por la Defensa del Agua y del Medio Ambiente de Tierra Amarilla, Coordinadora Aymara de Defensa de los Recursos Naturales de Arica y Parinacota, and Coordinadora Defensa Río Loa.
2.
The Norte Grande corresponds approximately to the administrative regions of Arica y Parinacota, Tarapacá, Antofagasta, and the northern part of Atacama. The Norte Chico corresponds approximately to the southern part of Atacama, Coquimbo, and part of the Valparaíso region.
3.
I use “community” as a general shorthand term to refer to groups of people living near proposed mining project sites. For critiques of the term, see, e.g., Agrawal and Gibson (1999).
4.
However, the organizational landscape is complex and has been influenced by the historical relationships of different indigenous communities to the state (see, e.g., Comisionado Presidencial para Asuntos Indígenas 2008), including formal state efforts to promote indigenous associations within the framework of the national indigenous law (Ley Indígena, Law 19,253).
5.
Indigenous groups in Chile include the Alacalufe (Kawaskar), Atacameño, Aymara, Colla, Mapuche, Quechua, Rapanui, and Yámana (Yagán).
6.
The laws mentioned in this section include Law 19,300, “Sobre Bases Generales del Medio Ambiente”; Law 20,417, “Crea el Ministerio, el Servicio de Evaluación Ambiental, y la Superintendencia del Medio Ambiente”; Law 20,551, “Ley de Cierre de Faenas”; and Law 20,600, “Crea los Tribunales Ambientales.” The Environmental Impact Evaluation System (SEIA) officially went into effect in April 1997.
7.
“Environmental impact studies” are generally more intensive and less common than “environmental impact declarations.” In the mining industry, between 1992 and 2017, only 222 submissions were made for the former, while 2,425 submissions were made for the latter.
8.
In the SEIA database, some projects appear multiple times under separate entries because they were submitted, then withdrawn or not admitted, and then resubmitted. In these cases, earlier duplicates were dropped from the data set and only the project getting a final decision was used in the analysis.
9.
Not all observations requesting changes to a proposed project amounted to community challenges. For example, statements of opposition by private individuals or those requesting minor adjustments to project operations (such as a truck route change) would be coded as 0. Labor strikes by miners were not included if they had only labor-specific demands.
10.
Where projects corresponded to more than one municipality, I used the average value for the set of municipalities involved in the project.
11.
In the 2002 census, the question about indigenous identity was, “Do you belong to any of the following indigenous or First Nations groups?” The answer included nine options: Alacalufe (Kawaskar), Atacameño, Aymara, Colla, Mapuche, Quechua, Rapanui, Yámana (Yagán), and none of the above. By contrast, the 1992 census asked, “If you are Chilean, do you consider yourself as belonging to one of the following cultures?” The answer included four options: Mapuche, Aymara, Rapanui, and none of the above. Though in this paper I treat indigenous communities collectively, the question of which indigenous communities are affected more than others is an important one, since different indigenous groups have distinct historical relationships to the state and the private sector; the variation in defensive responses among indigenous groups is a question for future research.
12.
Agriculture and indigenous presence are positively correlated (0.5). I show their effects separately before including them in the same model because they are analytically and theoretically distinct.
13.
I ran alternative models where I treated the agricultural and indigenous population variables as continuous (see Table A-1 in the Appendix).
14.
I tested two alternative control variables: the political party affiliation of the mayor in power when the project decision was issued, and the years of democratic government elapsed between the start of the democratic transition and the submission of the environmental impact study (see Table A-2 in the Appendix).
15.
I conducted collinearity diagnostics and post-estimation checks, including goodness-of-fit tests and ROC curve.
16.
The denied projects, all of which faced public opposition, included the Pampa Colorada water supply project for the Escondida copper mine, the Los Pumas manganese project near Arica in the far north (initially approved but later reversed), and the Dominga project near La Higuera in the Coquimbo region. Pascua Lama was stopped after initial approvals had already been granted.